This book is a unique compendium and synthesis of the cumulative

knowledge of more than 100 years of discovery and study of North

American Tertiary mammals. A valuable contribution of this book
is the detailed information of the distribution in time and space of
each species at fossil localities, recorded in a uniform scheme, so
that each chapter provides the same level of information.
Thirty-seven chapters are devoted to each particular family or
order, written by a leading North American authority, including
discussion of anatomical features, systematics, and paleobiology.
Four introductory chapters summarize information on the geological
time scale, Tertiary vegetation, and Pleistocene events, and four
final chapters integrate systematic and biogeographic information
for higher tax.a.
This book will serve as a unique database for continuing studies in
faunal diversification and change, and will answer questions such as
how changing biogeography and climates influenced the evolution
of mammalian communities. It will be an invaluable addition to the
libraries of paleontologists and zoologists.
Evolution of Tertiary Mammals
of North America
Evolution of
Tertiary Mammals of
North America

Volume 1: Terrestrial Carnivores, Ungulates,

and Ungulatelike Mammals

EDITED BY

CHRISTINE M. JANIS
Brown University
KATHLEEN M. SCOTT
Rutgers University

LOUIS L. JACOBS
Southern Methodist University and
Shuler Museum of Paleontology

CAMBRIDGE
UNIVERSITY PRESS
PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge CB2 lRP, United Kingdom

CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, United Kingdom
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

©Cambridge University Press 1998

This book is in copyright. Subject to statutory exception and to the provisions

of relevant collective licensing agreements, no reproduction of any part may
take place without the written pennission of Cambridge University Press.

First published 1998

Printed in the United States of America

Typeset in Times Roman 9/12 pt. in IbTJY( 2, [TB]

Library of Congress Cataloging-in-Publication Data

Evolution of Tertiary mammals of North America I edited by Christine

M. Janis, Kathleen M. Scott, Louis L. Jacobs.
p. cm
Includes bibliographical references (p. ).
Contents: v. 1. Terrestrial carnivores, ungulates, and
ungulatelike mammals
ISBN (invalid) 0-521-25519-2 (hardbound)
1. Mammals, Fossil - North America. 2. Paleontology - Tertiary.
3. Animals, Fossil - North America. I. Janis, Christine M.
(Christine Marie), 1950- . II. Scott, Kathleen M. (Kathleen
Marie) III. Jacobs, Louis L.
QE881.E857 1997
569'.097 - dc21 97-5757
CIP

A catalog record for this book is available from the British Library

ISBN0521355192hardback
To the memory Bryan Patterson,
his contributions to the study offossil mammals, and
his dedicated mentoring of students (especially me).
 Contents

List of contributors vii 14 Hyaenidae 243

Preface ix ANNALISA BERTA

0 Introduction Part III: Archaic ungulates and ungulatelike mammals

CHRISTINE M. JANIS
15 Archaic ungulates and ungulatelike mammals 247
Part I: Overview of context for the evolution of CHRISTINE M. JANIS, J. DAVID ARCHIBALD,
North American Tertiary mammals RICHARD L. CIFELLI, SPENCER G. LUCAS,
The chronological, climatic, and paleogeographic background to CHARLES R. SCHAFF, ROBERT M. SCHOCH, and
North American mammalian evolution 9 THOMAS E. WILLIAMSON
DONALD R. PROTHERO 16 Taeniodonta 260
2 Tertiary vegetation of North America as a context for mammalian SPENCER G. LUCAS, ROBERT M. SCHOCH, and
evolution 37 THOMAS E. WILLIAMSON
SCOTT L. WING 17 Tillodontia 268
3 The Pleistocene terrestrial mammal fauna of SPENCER G. LUCAS and ROBERT M. SCHOCH
North America 66 18 Pantodonta 274
RUSSELL W. GRAHAM SPENCER G. LUCAS

Part II: Carnivorous mammals 19 Dinocerata 284

SPENCER G. LUCAS and ROBERT M. SCHOCH
4 Carnivorous mammals 73
CHRISTINE M. JANIS, JON A. BASKIN, ANNALISA BERTA, 20 Archaic ungulates ("Condylarthra") 292
JOHN J. FLYNN, GREGG F. GUNNELL, ROBERT M. HUNT, J. DAVID ARCHIBALD
JR., LARRY D. MARTIN, and KATHLEEN MUNTHE 21 Arctostylopida 332
5 Creodonta 91 RICHARD L. CIFELLI and CHARLES R. SCHAFF
GREGG F. GUNNELL
Part IV: Artiodactyla
6 Early Cenozoic Carnivora ("Miacoidea") 110
JOHN J. FLYNN
22 Artiodactyla 337
CHRISTINE M. JANIS, JAMES A. EFFINGER,
7 Canidae 124 JESSICA A. HARRISON, JAMES G. HONEY,
KATHLEEN MUNTHE
DONALD G. KRON, BRUCE LANDER, EARL MANNING
8 Procyonidae 144 DONALD R. PROTHERO, MARGARET S. STEVENS,
JON A. BASKIN RICHARD K. STUCKY, s. DAVID WEBB, and
9 Mustelidae 152 DAVID B. WRIGHT
JON A. BASKIN 23 Eocene bunodont and bunoselenodont Artiodactyla
IO Ursidae 174 ("dichobunids") 358
ROBERT M. HUNT, JR. RICHARD K. STUCKY

11 Amphicyonidae 196 24 Entelodontidae 375

ROBERT M. HUNT, JR. JAMES A. EFFINGER

12 Nimravidae 228 25 Anthracotheriidae 381

LARRY D. MARTIN DONALD G. KRON and EARL MANNING

13 Felidae 236 26 Tayassuidae 389

LARRY D. MARTIN DAVID B. WRIGHT

vii
 Contents viii

27 Oreodontoidea 402 36 Brontotheriidae 525

BRUCE LANDER BRYN J. MADER

28 Oromerycidae 426 37 Equidae 537

DONALD R. PROTHERO BRUCE J. MACFADDEN

29 Protoceratidae 431 38 Chalicotherioidea 560

DONALD R. PROTHERO MARGERY C. COOMBS

30 Camelidae 439 39 Tapiroidea and other moropomorphs 569

JAMES G. HONEY, JESSICA A. HARRISON, MATTHEW w. COLBERT and ROBERT M. SCHOCH
DONALD R. PROTHERO, and 40 Amynodontidae 583
MARGARET S. STEVENS WILLIAM P. WALL
31 Hornless ruminants 463 41 Hyracodontidae 589
S. DAVID WEBB DONALD R. PROTHERO
32 Dromomerycidae 477 42 Rhinocerotidae 595
CHRISTINE M. JANIS and EARL MANNING DONALD R. PROTHERO
33 Antilocapridae 491 43 Proboscidea 606
CHRISTINE M. JANIS and EARL MANNING w. DAVID LAMBERT and JEHESKEL SHOSHANI
34 Cervidae and Bovidae 508
Part VI: Eutheria incertae sedis
S. DAVID WEBB
44 Eutheria incertae sedis: Mingotherium and ldiogenomys, with
Part V: Perissodactyla and Proboscidea
editors' inclusion of other problematical eutherian taxa 623
35 Perissodactyla and Proboscidea 511 SPENCER G. LUCAS and ROBERT M. SCHOCH
CHRISTINE M. JANIS, MATTHEW W. COLBERT,
MARGERY C. COOMBS, W. DAVID LAMBERT, Appendices
BRUCE J. MACFADDEN, BRYN J. MADER, Appendix I: Tertiary mammal localities 625
DONALD R. PROTHERO, ROBERT M. SCHOCH,
Appendix II: References for locality listings 655
JEHESKEL SHOSHANI, and
WILLIAM P. WALL Appendix ill: Museum acronyms 667
 Contributors

J. David Archibald Christine M. Janis

Department of Biology, San Diego State University San Diego, Department of Ecology and Evolutionary Biology, Brown University,
CA 92182-0057 Providence, RI 02912

Jon A. Baskin Donald G. Kron

Department of Geosciences, Texas A&I University, Kingsville, TX 78363 1566 MacArthur Drive, Boulder, CO 80303

Annalisa Berta W. David Lambert

Department of Biology, San Diego State University, San Diego, CA 92182 Department of Biology, St. Peter's College, Jersey City, NJ 07306

Richard L. Cifelli Bruce Lander

Oklahoma Museum of Natural History and Department of Zoology, Paleo Environmental Associates Inc., 2248 Winrock Avenue, Atladena,
University of Oklahoma, Norman, OK 73019 CA 91001

Matthew W. Colbert Spencer G. Lucas

Department of Geological Sciences, University of Texas at Austin, Austin, New Mexico Museum of Natural History, 1801 Mountain Rd. NW,
TX 78712 Albuquerque, NM 87104

BruceJ.MacFadden
Margery C. Coombs
Florida Museum of Natural History, University of Florida, Gainesville,
Department of Biology, University of Massachusetts, Amherst, MA 01003
FL 32611
James A. Effinger
Bryn J. Mader
10128 Toltec Rd. N.E., Albuquerque, NM 87111
Department ofMammalogy, American Museum of Natural History,
Central Park West at 79th Street, New York, NY 10024
John J. Flynn
Department of Geology, The Field Museum, Roosevelt Road at Lakeshore
Earl Manning
Drive, Chicago, IL 60605 Department of Geology, Tulane University, New Orleans, LA 70118
Russell W. Graham Larry D. Martin
Denver Museum of Natural History, 2001 Colorado Blvd., Denver, Museum of Natural History and Department of Systematics and Ecology,
co 80205 University of Kansas, Lawrence, KS 66045

Gregg F. Gunnell Kathleen Munthe

Museum of Paleontology, University of Michigan, Ann Arbor, P.O. Box 80, Escalante, UT 84726
Ml48109-1079
Donald R. Prothero
Jessica A. Harrison Department of Geology, Occidental College, Los Angeles, CA 90041
1220 E. Hampton, Tucson, AZ 85719
Charles R. Schaff
James G. Honey Museum of Comparative Zoology, Harvard University, Cambridge,
10657 W. 6th place,# IOI, Lakewood, CO 80215 MA02138

Robert M. Hnnt Jr. Robert M. Schoch

Division of Vertebrate Paleontology, W436 Nebraska Hall, University of College of General Studies, Boston University, 871 Commonwealth,
Nebraska, Lincoln, NE 68588-0549 Boston, MA 02215

ix
 Contributors x

Jeheskel Shoshani S. David Webb

The Elephant Research Foundation, 106 E. Hickory Rd., Bloomfield Hills, Florida Museum of Natural History, University of Florida,
MI 48304 and Department of Biological Sciences, Wayne State University, Gainesville, FL 32611
Detroit, MI 48202

Margaret Skeels Stevens Thomas E. Williamson

Department of Business, Science, and Mathematics, Lamar University New Mexico Museum of Natural History, 1801 Mountain Rd. NW,
at Port Arthur Stevens, Department of Mathematics and Sciences, Albuquerque, NM 87104
Lamar University at Orange, Texas.
Mailing address: Department of Geology, Box 10031, Lamar University,
Scott L. Wing
Beaumont, TX 77710
Department of Paleobiology, NHB 121, National Museum
Richard K. Stucky of Natural History, Smithsonian Institution, Washington,
Denver Museum of Natural History, 2001 Colorado Boulevard, Denver, DC 20560
co 80205
William P. Wall David B. Wright
Department of Biology, Georgia College & State University, Milledgeville, Department of Orthodontics, SM-46 and Burke Museum,
GA 31061 University of Washington, Seattle, WA 98195
 Preface

CHRISTINE M. JANIS

This volume was originally conceived as a companion volume to the A final, more personal, desire was to provide good restorations of
book by Maglio and Cooke ( 1978) on the evolution of African mam- extinct taxa for each chapter, and to this end I am exceedingly grate-
mals. North America has a much richer fossil record than Africa, ful to Brian Regal for the drawings that grace many of the chapters.
but, at least in the early 1980s when the idea for this book was first The past couple of decades have seen a revolution in the restoration
conceived, it had received much less attention, at least in terms of of dinosaurs as living animals, but extinct mammals have largely
syntheses of Cenozoic patterns of fauna! evolution. Perhaps this is tended to be portrayed as if they were already stuffed and behind
because we are preoccupied with Africa, rather than North America, museum glass; virtually the only activity seen would be that of a
as the center stage for human evolution. A problem with the Maglio sabertooth killing something! However, Brian has restored many
and Cooke volume is that there is no standardized layout for the extinct mammals in a variety of typically mammalian activities, in-
chapters: Information is presented at different taxonomic levels in cluding scratching and grooming. Note, also that many restorations
different chapters, and the method of presentation of fauna! infor- for many of the chapters are oftaxa that are not normally illustrated:
mation is also idiosyncratic to each author (or locality information For example, Menoceras (rather than Teleoceras) is illustrated for
is absent entirely). Despite the large amount of useful information the Rhinocerotidae, and Leptauchenia (rather than Merycoidodon)
in the volume, it is very difficult to use it to reconstruct consistent provides an example of an oreodont.
fauna! information of "what was where when." This volume still represents only a portion of the fauna of North
The initial aim in designing this volume was to have each chap- America: the paraphyletic groupings of the larger mammals, ter-
ter in a standardized format so every chapter would contain the restrial carnivores and ungulates. However, in terms of the utility
same level of information, and fauna! and biogeographical infor- of mammal taxa in studies of evolutionary patterns, and so on, it
mation could be readily extracted. Another prime motive was to is generally agreed that "macromammals" (i.e., those of greater
present North American Tertiary mammals in a paleobiological con- mass than around a kilogram) provide a more reliable data source
text (hence the section entitled "Biology and Evolutionary Patterns" than smaller mammals (in part because of taphonomic problems of
in each chapter), and to provide locality information on their distri- preservation; see, e.g., Behrensmeyer, Western, and Dechant Boaz,
bution in space and time. This is in contrast to the volume edited by 1979). Certainly, macromammals can provide information on their
Woodburne ( 1987), which highlights the importance of mammals in own, and it has been suggested that macromammals and micromam-
biostratigraphy but does not emphasize systematics or paleobiology. mals should be examined separately in fauna! studies (Fortelius et
The rich mammalian fossil record of North America should prove al., 1996). The only macromammals missing from this volume are
an excellent database for testing evolutionary hypotheses. Indeed, the edentates. North American Tertiary edentates are known only
portions of the locality data gathered for the purposes of this vol- from the latest Miocene and Pliocene, and their fauna! occurrences
ume (Van Valkenburgh & Janis, 1993), or similar databases of North could easily be determined from the information provided in appen-
American localities (Alroy, 1992), have already been used for this dices I and II. The learning experience that we have obtained from
purpose. A recently-published compilation (Prothero and Emry, editing this volume will make the task of producing Volume II much
1996) also provides information on taxonomy and stratigraphic oc- easier.
currence for a number of North American mammal taxa, but from a Thanks are due to numerous people. First, I must thank Kathy
limited portion of the Tertiary record: unfortunately, this book was Scott and Lou Jacobs for the editorial work they put in at the in-
published too late for most of its material to be incorporated into ception of this volume, and Brown University and the University
this volume. of Chicago for providing me with the facilities to edit this volume,

xi
 Preface xii

and for assisting with photocopying and postage. Many thanks also Jack and I were not yet together when I embarked on the production
to Brian Regal, not only for the restorations of fossil mammals, but of this volume: I can only credit his patience and devotion to the
for the many instances when he "picked up the slack" in terms of fact that we are still together at its completion.
other illustrations for contributors.
CHRISTINE M. JANIS
Unpublished (or at least, unpublished at that time) fauna! informa-
tion was kindly provided by Dan Bryant, Robert Emry, Ralph Eshel-
man, Dan Garcia, James Morgan, Alan Tabrum, and Mike Voorhies, REFERENCES
and John Alroy was extreme! y generous in providing access to litera-
ture listings that he had gathered himself. Help with the fauna! local- Alroy, J. (1992). Conjunction among taxonomic distributions and the
Miocene mammalian biochronology of the Great Plains. Paleobi-
ities and their correlations was provided by Dave Archibald, Larry
ology, 18, 326-43.
Barnes, Jon Baskin, Margery Coombs, Daryl Domning, John Flynn, Behrensmeyer, A K., Western, D., & Dechant Boaz, D. E. (1979). New per-
Gregg Gunnell, Brook Ellen Hall, Jim Honey, Donald Kron, Dave spectives in vertebrate paleontology from a recent bone assemblage.
Lambert, Don Prothero, Bill Simpson, John Storer, Alan Tabrum, Paleobiology, 5, 12-21.
and Richard Tedford. Fortelius, M., Werdelin, L., Andrews, P., Bernor, R. L., Gentry, A, Mittman,
W., & Viranta, S. (1996). Provinciality, diversity, turnover and pa-
Students who helped collate fauna! lists, or with editing work, in-
leoecology in land mammal faunas of the later Miocene of western
clude Wendy Derman, John Hunter, Tricia Brady Wilhelm, Jeremy Eurasia. In The Evolution of Western Eurasian Neogene Mammal
Kahn (Brown University), and Dennis Su (University of Chicago), Faunas, eds. R. L. Bernor, V. Fahlbusch, & H.-W. Mittman, pp.
Secretarial assistance was provided by Jean Pasdeloup (University 414-48. New York: Columbia University Press.
of Chicago), Helen Shuman, and Carol Stewart (Brown University). Maglio, V. J., & Cooke, H.B. S. (eds.) (1978). Evolution of African Mam-
mals. Cambridge, Mass.: Harvard University Press.
Of the many contributors to this volume, some special thanks are
Prothero, D. R., & Emry, R. J. (eds.) (1996). The Terrestrial Eocene-
due to certain people: Don Prothero, Gregg Gunnell, and Dave Oligocene Transition in North America. Cambridge: Cambridge
Archibald all provided extensive advice on the fauna! listings; Earl University Press.
Manning valiantly came to my aid with a couple of chapters at a Van Valkenburgh, B., & Janis, C. M. (1993). Historical diversity patterns in
relatively late date, and improved their content immensely; and the North American large herbivores and carnivores. In Species Diversity
in Ecological Communities, eds. R. E. Ricklefs & D. Schluter, pp.
volume would probably not have been completed without continual
330-40. Chicago: University of Chicago Press.
infusions of encouragement from Jon Baskin. Woodbume, M. 0. (ed.) (1987). Cenozoic Mammals of North America.
Finally, I would like to thank my husband, Jack Sepkoski, for his Berkeley: University of California Press.
continual encouragement and support through this editing process.
0 Introduction
CHRISTINE M. JANIS

AIMS OF THE VOLUME institutional abbreviations have also been standardized and are listed
in Appendix III.
This volume provides a compilation of descriptive taxonomy and
paleobiology of North American Tertiary mammals, with special
emphasis on providing a compilation of fauna! locality data for THE STANDARDIZED LAYOUT OF
each taxon arranged in a standardized format. EACH CHAPTER
During the past couple of decades there has been increasing inter-
est in assembling information on fauna! listings of Cenozoic mam- The contributors adhere to a common layout for each chapter in or-
mals. Maglio and Cooke (1978) assembled this type of information der to provide uniform information throughout the book. The Intro-
for African mammals. The volume by Savage and Russell (1983) duction introduces the group and outlines some interesting aspects
encompasses faunas of the entire world and also includes a brief of its paleobiology. The Defining Features section lays out the ba-
account of Mesozoic faunas. The volume edited by Woodbume sic cranial, dental, and postcranial features of the taxon. The term
(1987) deals specifically with Cenowic North American mammals, "defining features" was used rather than the cladistically preferred
although the aims of that book, concerned primarily with biostratig- term "diagnostic features" because this section was intended to be
raphy, are rather different from those of this volume. In addition, a general introduction to the characters of the group as a whole,
fauna! lists have been compiled (or are in the process of being com- plesiomorphic as well as apomorphic. Because of the constraints
piled) for various portions of the Old World, such as the Neogene of production costs, contributors were generally requested to limit
of the Siwaliks (e.g., Barry, Lindsay, and Jacobs, 1982), Australian their illustrations to one taxon for pictures of the skull, dentition, and
faunas (Rich et al., 1991), and the Neogene of Eurasia in general skeleton. However, for very diverse groups (e.g., archaic ungulates),
(Bemor, Fahlbusch, and Mittman, 1996). more illustrations were allowed.
The taxonomic level of interest in this volume is typically the The Systematics section includes a Suprataxon section that deals
genus, but locality information is (usually) provided at the level of with the history of the ideas of the relationships of the taxon in
the species. The fauna! localities have been standardized throughout question among mammals in general, and an Infrataxon section that
the chapters and are listed in Appendix I (see explanation later in this deals with interrelationships within the group, including a clado-
chapter). We decided to exclude Pleistocene mammals because they gram. Rather than have a more general "suprataxon" cladogram in
are covered in the volume by Kurten and Anderson (1980), and the each chapter, a single consensus cladogram is presented in the sum-
large number of Pleistocene taxa and localities would have added mary chapter at the start of each section (Chapters 4, 15, 22, and 35).
considerably to the length of this volume. Although part of the stan- The Included Genera section includes a brief description of each
dardized layout includes a section on systematics and a cladogram genus, including the listing of the type species and type specimen,
for the group in question, the aim of the book is not to be a forum for and a listing of the valid species, including the localities in which
a taxonomic update or systematic debate. (Note, however, that for each species was found. Some contributors gave the type local-
some taxa, such as the Camelidae, the cladogram presented in this ity for each species, although this was not requested. Similarly,
volume represents the only existing cladogram.) Additionally, for some contributors included the authors for each species, and also
the purposes of standardization, and to provide equal quality of in- cited those references: Standardization in this case included ensur-
formation across each chapter, the stratigraphic range charts in the ing only that references were given and cited for the genus and type
chapter are presented according to a standardized format, and the species. Finally, I encouraged contributors to provide me with as
 Christine M. Janis 2

many taxonomic synonyms as possible because the literature is rife THE NUMBERING OF THE LOCALITIES
with alternate names for tax.a. Most contributors provided generic AND THE RATIONALE FOR THE
synonyms, and many also provided species synonyms (including an BIOGEOGRAPHIC REGIONS
entire appendix listed by Gunnell for Chapter 5). I also requested
A locality number encompasses an entire formation. Subdivisions
contributors to provide an average dental length measurement for
within that formation are then numbered A, B, C, and so on, ac-
each genus, m2 if it was available, if not, some other tooth. The
cording to relative age. For the purposes of numbering, as well as
purpose was to provide a size estimate for the taxon because dental
for the creation of the biogeographic range charts, the localities are
length measurements are a good proxy of body mass (see Damuth
grouped into various biogeographic regions (see later). Within each
and MacFadden, 1990), and m2 length is the most reliable measure-
biogeographic region the localities are numbered according to strati-
ment, at least in ungulates (Janis, 1990).
graphic position. The biogeographic regions are themselves ordered
Finally, the Biology and Evolutionary Patterns section provides a
in a general west to east fashion, except for the Pacific Northwest
synopsis of the paleobiology and evolutionary trends of the group.
and northern Great Plains localities, which are listed after those of
This section includes the standardized temporal range chart for each
the central Great Plains (see ordering in the figures in the summary
taxon. The biogeographic range charts (which may combine a num-
chapters). A few localities appear to be slightly out of order; this is
ber oftaxa) are in the summary chapters (see later in this chapter).
because information about the exact age was later revised after the
creation of the list (see also comments later).
The unification of the localities necessitated a certain degree of
THE UNIFIED LOCALITY LISTING grouping of sites. Sometimes this involved grouping of the quar-
ries within a single time horizon in a formation (e.g., the quarries
in the Miocene Valentine Formation, localities CP114A-CP114D).
THE CREATION OF THE LISTING AND THE USE
At other times, localities that were in a similar location at a similar
OF THE APPENDIXES
time were grouped together (e.g., the North Coalinga Local Fauna
It was not my original intent to standardize the locality information and Domengine Creek, in the Temblor Formation of the Miocene
into the system of unified numbers presented in this volume. I de- of California, locality CC23). To list every single fossil-containing
cided to do this after a suggestion Mike Voorhies made, following site as its own separate locality would have increased the number
his review of my chapter on dromomerycids. I think that if I had had of individually listed localities by at least an order of magnitude.
any inkling at the time of the immense amount of work (literally Because references are provided for each locality, it should be pos-
years) it would take to devise a system of unified localities, and to sible in most cases for a concerned researcher to reconstruct finer
check with every author where to place his or her localities (not to detail.
mention double-checking with lists from the literature to ensure that Because the numbering of the localities was accomplished be-
taxa had not been omitted), I never would have embarked on this fore all final contributions were received, revisions had to be made
procedure. However, now that the task has been completed, I hope to the listings that made the final product less tidy than I would have
the listing and unification prove useful. preferred. In the case of new formations (primary locality num-
The original unified listing was created from the lists oflocalities bers), additions were made by creating an intermediate locality be-
supplied by the authors who submitted their chapters early, supple- tween two existing ones, affixing the suffix II to the younger of the
mented with lists derived from Woodburne (1987). Later-arriving two localities (e.g., NP19 and NP19II). In the case of new subdivi-
submissions were fitted into the existing scheme, sometimes in- sions within formations, double letters were created (e.g., NPlOB,
volving the creation of new localities. Normally there were several NPl OBB, NPl OB2). This rather cumbersome mode of renumbering
go-arounds with each author to check that his or her taxa were cor- localities as "work in progress" proved to be more practical than
rectly assigned to the appropriate localities. For the final version renumbering localities throughout, which would have then neces-
of each chapter, I personally double- or triple-checked the locality sitated renumbering the localities for the individual taxa that had
numbers for each taxon on several separate occasions, in addition already been processed (not only of the numbered locality itself,
to all the earlier checking that had been performed by both my- but of all younger localities within the region). In fact, I started out
self and the authors. It is impossible to have complete confidence by trying to renumber throughout each time I received a correction,
that these listings are totally error free, but every attempt has been and swiftly came to realize the error of my ways.
made to minimize errors. However, the ultimate responsibility for Sometimes revisions were necessitated because I received new
correctness is the obligation of each author. stratigraphic information. For example, I was advised by contribu-
The localities in the individual chapters (e.g., CPl, NP5) must tors that the Paleocene Montana locality NP19 was better divided
be looked up in Appendix I. This saves space in the volume, as into two formations, the Lebo Formation and the Melville Forma-
well as providing an overall unification. Appendix II provides the tion. Sometimes revisions reflect new localities supplied by authors
references for the faunal listings from the localities (as referenced after the original numbered lists were created (e.g., locality CPl 16II
in Appendix I). If no reference is given for a particular locality, this in the late Miocene of Nebraska). In some instances locality numbers
locality represents one in which the only faunal information came are missing: For example, in the Pliocene of the California Central
from the contribution of one or more authors. and Coast region, there is a locality CC47 and a locality CC50,
 Introduction 3

but no locality CC48 or CC49. This lack reflects locality informa- eludes Arizona, New Mexico, the Texas Big Bend area, southwest-
tion provided by one author who later withdrew his contribution. ern Colorado (i.e., the Paleocene Animas Formation and the Eocene
The author who took over this chapter was scrupulous about not Huerfano Formation), and northern Mexico. The Texas Big Bend
using information that had been originally provided and could not area includes all the Paleogene Texas localities (except for the Duch-
be obtained from the published literature or from museum records esnean Yegua Formation, grouped with the Gulf Coast); the Miocene
obtained by that author or by myself. Delaho, Rawls, and Banta Shut-In Formations; and the Pliocene
Camp Rice and Love Formations.
The "southern Great Plains" includes the Texas/Oklahoma pan-
BIOGEOGRAPHIC REGIONS handles. The "central Great Plains" includes Nebraska, South Da-
kota, Wyoming, Kansas, most of Utah, and northern/northeastern
I decided to divide localities into biogeographic regions, so that bio-
Colorado (i.e., the Paleocene Denver and Wasatach formations;
geographical variation as well as stratigraphic ranges could be seen
the Eocene Wasatach, DeBeque, Uinta, and Washakie formations;
(Figure I.I). The biogeographic regions were originally based on
and all Oligocene and Neogene sites). Note that Paleogene fau-
those presented by Tedford et al. (1987), with some modifications
nas from western Wyoming, which I have grouped with the central
after personal communication with Tedford. The division of the
Great Plains faunas, might better be described as "Northern Rocky
Mexican localities (as to inclusion with the southern Great Basin,
Mountains." The "northern Great Plains" includes Montana, North
California Coast, or the Central American region) follows distribu-
Dakota, and western Canada (Alberta, Saskatchewan, and British
tional maps in chapters in MacFadden (1984).
Columbia). Also included here, for reasons of convenience rather
The "West Coast" localities include coastal faunas of Washington,
than biogeographical appropriateness, is the early Eocene Ellesmere
Oregon, and California. There is no separate "West Coast" section
Island locality in the Eureka Sound Formation of the Northwest Ter-
on the unified biogeographic charts: Taxa from these localities are
ritories.
instead placed into the "Pacific Northwest" or "California Central
The "Pacific Northwest" includes Washington, Oregon, and
and Coast" sections according to latitude. The "Central America" lo-
Idaho. The "northern East Coast" includes the East Coast north
calities also include Central Mexico. The "Gulf Coast" includes the
of the Carolinas. Faunas from the "East Coast Marine" localities
Texas Gulf Coast, Florida, and the southern East Coast (Louisiana,
are not accorded their own section on the biogeographic charts, but
Mississippi, Georgia, and North and South Carolina) (following in
are included in the "northern East Coast" or "Gulf Coast" sections,
part the Carribean Fauna! Provinces for fish; see Gillette, 1984).
depending on their latitude.
"California Central and Coast" also includes localities in Baja Cal-
The way in which I have grouped localities into biogeographic
ifornia, Mexico.
regions may not be agreeable to everyone. For example, other au-
The "northern Great Basin" includes southeastern California,
thors have devised slightly different schemes of biogeographic re-
Nevada, and southwestern Utah. The "southern Great Basin" in-
gions: for example, Storer (1989) on fossil rodents or Flynn (1986a)
on early Eocene mammals. Some time after the listings had been
completed, and locality numbers applied to taxa, several authors
(David Archibald, Richard Cifelli, Spencer Lucas, and Thomas
Williamson) criticized my use of the term "southern Great Basin"
in referring to Paleogene faunas. I had originally devised this term
based on the distribution of Miocene localities, after consultation
with Richard Tedford. But the Paleogene faunas in the San Juan
basin of southwestern Colorado and northern New Mexico would be
better described as being in the "southern Rocky Mountains." How-
ever, not only would it have been extremely difficult to make changes
throughout the volume once I became aware of the problem, but the
term "southern Rocky Mountains" would not be well applicable
to Neogene localities included within this biogeographical region,
such as those of the Texas Big Bend region and northern Mexico.
It is important to emphasize that because all the individual infor-
mation has been preserved in this volume (in the form of Appendixes
I and II), any scheme that I have used to lump together informa-
tion, for the purposes of diagrams or discussion, has not been lost.
The original information is retained for others to reconstruct their
own biogeographic scheme. I suspect that biogeographic regions
Figure 1.1. Map of North America showing the biogeographic regions em- (as defined by faunal associations) would vary, either subtly or pro-
ployed in this volume. Key: CC = California Coast; CP = Central Great
foundly, at different times in the Tertiary, as a result of the effects
Plains; GC = Gulf Coast; NB = Northern Great Basin; NC = Northern
East Coast; NP = Northern Great Plains; PN = Pacific Northwest; SB = of the Lararnide and Basin range orogenies and changing climatic
Southern Great Basin; SP = Southern Great Plains. conditions. The information provided and organized in this volume
 Christine M. Janis 4

provides a database for testing of that notion, as well as for other boundary where it has been traditionally placed. This decision is
hypotheses of fauna! change and community evolution. apparently supported by the recent paper by Gunnell, Bartels, and
Gingerich (1993), but it is at variance with Figure 1.2 (Prothero,
this volume, Chapter 1), where it is placed within the earliest
THE SUMMARY BIOGEOGRAPHIC CHARTS
Wasatchian.
The biogeographic charts in the summary chapters represent the There have also been a number of minor shifts in the dating of
combined information from the authors' original contributions and, the NALMA boundaries, relating in large part to the changing of the
in the majority of cases, from information added by myself (in con- dating of the Cretaceous/Tertiary boundary (see Swisher, Dingus,
sultation with the original authors) from the published literature and Butler, 1993) and the Eocene/Oligocene boundary. The bound-
and from my records from museum collections. The original inten- ary dates presented in the summary biogeographic and stratigraphic
tion was to have a summary chart for each taxon in each individual charts in this volume are from Prothero (this volume, Chapter 1 and
chapter. It was later decided for several reasons to transfer these pers. comm.) and Bob Hunt (pers. comm.) regarding early Miocene
charts to the summary chapters. First, space was saved by combin- dates. It is inconceivable, with all the current activity in this area,
ing taxa. Second, overall diversity trends were more easily visible that these dates will all still be accurate at the time this volume is
when closely related taxa were grouped together. Finally, the sum- published, or even that those new dates will be the last word on the
mary chapters proved to be a suitable venue in which to discuss subject. Because the organization in this volume links each taxon
suprataxon evolutionary trends, and the biogeographic charts pro- to the particular localities in which it is found, no information is
vide the appropriate illustrations. lost by changing dates, shifting NALMAs or epoch boundaries, or
redating particular localities. Any person using the information in
this volume to build a database is strongly cautioned to keep abreast
THE GEOLOGICAL TIME SCALE AND of current changes in time scales, epoch boundaries, and so on.
NALMA DIVISIONS
DIVISIONS OF LAND MAMMAL "AGES"
When unification work first started on the localities and the sum-
mary stratigraphic and biogeographic diagrams, the time scale and I have chosen to adhere mainly to the Woodburne (1987) NALMA
divisions of North American land mammal "ages" (NALMAs) were scheme. However, I describe some of the controversies here (see also
adopted from Woodburne (1987). This was mainly for the purpose Prothero, this volume, Chapter 1), and discuss reasons for some of
of consistency: In many ways this book should serve as a companion the changes I have incorporated.
volume to Woodburne's, and it made sense to have this volume in The division of Puercan 0 (earliest Paleocene) did not exist in
the same format. However, in the decade since the publication of Woodburne (1987). I added this subdivision to accommodate fau-
Woodburne (1987), a number of changes have occurred. Some of na! locality information provided by David Archibald (this volume,
these changes have been incorporated into this volume; others have Chapter 20) on archaic ungulates (see also Archibald and Lofgren
not, as explained later in this chapter. 1990; Lofgren, 1995).
The division by Archibald et al. (1987) of the Torrejonian (early
Paleocene) into To 1, To2, and To3 has been challenged by Willamson
THE GEOLOGICAL TIME SCALE AND
and Lucas (1993) and Williamson (1993), especially regarding the
EPOCH BOUNDARIES
correlation of the Nacimiento Formation in New Mexico. In unify-
A major change in the past decade in epoch boundaries has been ing the Nacimiento localities (SB23), I impose the Archibald et al.
the recognition that the Chadronian LMA is late Eocene rather than (1987) Torrejonian boundaries onto the stratigraphic diagram pro-
early Oligocene (Swisher and Prothero, 1990). This change has been vided by Williamson and Lucas (1993) to create subdivisions that
incorporated here, as has the subsequent reorganization of the other match the rest of the information elsewhere in the Paleocene. The
Eocene LMAs: The Uintan and the Duchesnean are now considered individual locality numbers of Williamson and Lucas are listed in
middle Eocene rather than late Eocene (see Prothero, this volume, the subdivisions of the locality listings, so if the subdivisions of the
Chapter 1). Torrejonian are completely revised, there should be little difficulty
There has also been considerable debate about the placement of in reassigning individual taxa.
the Paleocene/Eocene boundary, which is traditionally placed bet- There has been a recent tendency to ignore Ti6, the final division
ween the Clarkforkian and the Wasatchian LMAs. However, it has of the Tiffanian recognized by Archibald et al. ( 1987). Prothero (this
been suggested that the boundary should be changed, either upward, volume, Chapter 1), includes Ti5 and Ti6 together as a single unit.
to lie within the early Wasatchian (e.g., Wing, 1984; Lucas, 1989), In contrast, Maas and Krause (1994) lump together taxa from Ti6
or downward, to lie within the Clarkforkian (e.g., Gingerich, 1976; with taxa from the earliest Clarkforkian, Cfl. It is true that there is
Rose, 1980; Archibald et al., 1987). As dissenters from the tradi- sparse fossil evidence available from this time period, but some of
tional position of the boundary appear to be split about evenly in it may yet be important (e.g., the fact that the first large oxyaenid is
either direction, and neither advocated change has yet been widely found in Ti6; see Gunnell, Chapter 5, this volume, and Figure 4.2).
accepted (in contrast to the shift in the Eocene/Oligocene bound- I add an earliest Wasatchian (early Eocene) subdivision, WaO, to
ary), I decided to take the conservative approach and leave the reflect the numerous taxa described by Gingerich in recent years
 Introduction 5

from the very earliest part of this epoch, but I do not follow Gin- but other workers would place this fauna in the early Clarendonian
gerich's further subdivisions of the Wasatchian into Wa0-Wa7 (see (e.g., Webb, 1969; Voorhies, 1990). Voorhies (1990) cautions that
Gingerich, 1983, 1989). WaO is equivalent to my "early early the "late late Barstovian" of Tedford et al. (1987) is broadly equiva-
Wasatchian." The division of the rest of the Wasatchian follows lent to the "early Clarendonian" as used by many other authors and
Krishtalka et al. ( 1987), who recognized early ( = Graybullian), mid- as portrayed in Prothero (this volume, Chapter 1). The "early late
dle ( = Lysitean), and late (= Lostcabinian) portions. The division of Barstovian" of Tedford et al. (1987) also encompasses the "medial"
the Bridgerian (early middle Eocene) also follows Krishtalka et al. and "late" Barstovian of Voorhies ( 1990) and others.
(1987), in which early(= Garnerbuttian), middle(= Blackforkian), Alroy (1992) recognizes a tripartite Barstovian (early, middle,
and late(= Twinbuttean) portions are recognized. and late), based on conjunctions of taxonomic distributions in the
Despite the establishment of a separate, early portion of the Uin- Great Plains regions. His "late Barstovian" includes the Burge fauna,
tan as the Shoshonean (Flynn, 1986b), I follow the more traditional upholding the scheme of Tedford et al. (1987). However, Alroy's
bipartite division into early and late portions, as in Krishtalka et al. (1992) three divisions do not precisely correspond to those of Ted-
(1987). Likewise, the Duchesnean (late middle Eocene) has been ford et al. (1987); for example, the Norden Bridge Quarry in the
retained as a single unit here, following Krishtalka et al. (1987), Valentine Formation (locality CP114A) falls into the "middle Barsto-
despite more recent attempts to subdivide this poorly known LMA vian" of Alroy but into the "late late Barstovian" of Tedford et al.
into early and late sections (see discussion in Lucas, 1992). Further controversy about the Barstovian/Clarendonian boundary
There is also controversy about the Hemingfordian/Barstovian was raised by Whistler and Burbank (1992). Working with the radio-
(early/middle Miocene) boundary. The start of the Barstovian has metric dating of the subdivisions of the Dove Spring Formation in
been defined by the earliest appearance of proboscideans in North Nevada (locality NB7), they proposed a new dating scheme for sub-
America (Evander, 1986; Tedford et al., 1987), which has led to divisions of the later Miocene and made suggestions on correlations
the suggestion that localities in the Central Great Plains which with faunas in the Great Plains region. Whistler and Burbank (1992)
do not contain proboscideans, traditionally regarded as Barstovian may well be correct in this redating, but without companion redat-
(such as the Olcott Formation [CPl 10]), are actually late Hem- ing of the Great Plains late Miocene localities, their information is
ingfordian in age (Evander, 1986). However, the presence of un- difficult to use in isolation. Thus, in using their fauna! lists and lo-
doubted proboscideans (the mammutid Zygolophodon) at two of cality subdivisions, I see no alternative at present but to reassign the
late Hemingfordian western sites (Massacre Lake, Nevada, locality localities to NALMA subdivisions based on their absolute dating,
NB17, and Deep River Local Faunas, Montana, locality NP34D) not on their reassessment of the NALMA subdivision boundaries.
shows not only that proboscideans may not define the Heming- In this fashion, the locality information on taxa from the northern
fordian/Barstovian boundary, but also that proboscideans evidently Great Basin is at least time-equivalent to the information currently
appeared earlier in the western portion of North America before obtainable from localities in other parts of North America. Should
they arrived in the central Great Plains region. Lindsay (1995) also later Miocene NALMAs and NALMA subdivision boundaries sub-
considers that proboscideans are not a suitable marker for the Hem- sequently be shifted to reflect all known localities, the information
ingfordian/Barstovian boundary. presented here will still be of use because the taxa are tied to the
The issue of the Hemingfordian/Barstovian boundary has been es- individual localities rather than to particular dates or ages (although
pecially heated with regard to the Barstow Formation in California. revision may be needed in the summary stratigraphic and biogeo-
I follow Woodburne, Tedford, and Swisher (1990) in placing the graphic charts).
boundary between the Rak Division fauna (locality NB6B) and the Tedford et al.'s (1987) division of the Hemphillian also differs
Green Hills fauna (locality NB6C). Lander (1985) argues, on the ba- from the traditional usage and from that employed by Voorhies
sis of oreodont stratigraphy, that the Green Hills fauna correlated, (1990). Rather than recognize an "early" and "medial" Hemphillian,
at least in part, with late Hemingfordian faunas of the Great Plains, Tedford et al. use the terms "early early" and "late early" for the
shifting the Hemingfordian/Barstovian boundary upward. In con- same stages. Alroy's (1992) division of the Hemphillian does not
trast, Lindsay (1995) argues for a downward shifting of the bound- subdivide the early Hemphillian. His "middle Hemphillian" appears
ary in the Barstow Formation, to a level between the Rak Division to correspond to the "late Hemphillian" of Tedford et al., and his "late
fauna and the Red Division fauna (locality NB6A), based on the Hemphillian" appears to correspond to their "latest Hemphillian."
first appearance of the cricetid rodent Copemys. This would set Finally, my consideration of Pliocene taxa includes only the latest
the Hemingfordian/Barstovian boundary at 16.8 Ma, about a mil- Hemphillian and the Blancan LMAs, although the earliest Irving-
lion years earlier than the current estimation. Lindsay's evidence tonian is now also considered to belong to the Pliocene (Lundelius
is convincing, but it was impossible to include his conclusions in et al., 1987).
this volume. As noted by Prothero (this volume, Chapter 1), there
is clearly a need for a detailed biostratigraphic zonation of the
Miocene. SUMMARY
Tedford et al. (1987) divides the middle Miocene into the early,
early late, and late late Barstovian, as has been followed in this vol- I have presented a clear picture of how this book was organized and
ume. Tedford et al.'s (1987) late late Barstovian encompasses the the judgment calls I had to make concerning changes in dating and
Burge fauna of Nebraska (Valentine Formation, locality CP114D), subdivision of land mammal "ages" that occurred while this book
 Christine M. Janis 6

was in progress. In a rapidly changing field such as North American (1989). New earliest Wasatchian mammalian fauna from the Eocene of
Tertiary chronology and biostratigraphy, it is almost impossible for northwestern Wyoming: composition and diversity in a rarely sam-
pled high-floodplain assemblage. University of Michigan Papers on
a work to remain current between its submission and the time of
Paleontology, 28, 1-97.
publication. The problem is only magnified in a case such as this Gunnell, G. F., Bartels, W. S., & Gingerich, P. D. (1993). Paleocene-
volume, which took many years to compile and includes authors Eocene boundary in continental North America: biostratigraphy and
with different interpretations of time scales and locality informa- geochronology, northern Bighorn Basin, Wyoming. In Vertebrate Pa-
tion. I have stayed abreast of the latest developments and made leontology in New Mexico, eds. S. G. Lucas & J. Zidek. New Mexico
Museum of Natural History and Science, 2, 134-44.
changes where possible to the work in progress. This introduction
Janis, C. M. ( 1990). Correlation of cranial and dental variables with body size
should help readers interpret and understand the modes of unifica- in ungulates and macropodoids. In Body Size in Mammalian Paleobi-
tion of information in this volume and track down some more recent, ology: Estimation and Biological Implications, eds. J. Damuth & B.
alternative schemes of biochronological zonations. J. MacFadden, pp. 255-99. Cambridge: Cambridge University Press.
Krishtalka, L., West, R. M., Black, C. G., Dawson, M. R., Flynn, J. J.,
Turnbull, W. D., StuckY, R. K., McKenna, M. C., Bown, T. M., Golz,
D. J., & Lillegraven, J. A. (1987). Eocene (Wasatchian through
ACKNOWLEDGMENTS
Duchesnean) biochronology of North America. In Cenozoic Mam-
mals ofNorth America: Geochronology and Biostratigraphy, ed. M.
I thank Dave Archibald, Jon Baskin, Rich Cifelli, Gregg Gunnell, 0. Woodburne, pp. 77-117. Berkeley: University of California Press.
Spencer Lucas, Don Prothero, and Tom Williamson for comments Kurten, B., & Anderson, E. (1980). Pleistocene Mammals ofNorthAmerica.
on the original draft of the manuscript. New York: Columbia University Press.
Lander, E. B. (1985). Early and middle Miocene continental vertebrate
assemblages, central Mojave desert, San Bernadino County, Califor-
nia. In Geological Investigations Along Interstate 15, Cajon Pass to
REFERENCES Manix Lake, California, ed. R. E. Reynolds, pp. 127-44. Redlands,
Calif.: San Bernadino County Museum.
Alroy, J. (1992). Conjunction among taxonomic distributions and the Lindsay, E. H. (1995). Copemys and the Barstovian/Hemingfordian
Miocene mammalian biochronology of the Great Plains. Paleobi- boundary. Journal of Vertebrate Paleontology, 15, 357-65.
ology, 18, 326-43. Lofgren, D. L. (1995). The Bug Creek problem and the Cretaceous-Tertiary
Archibald, J. D., Gingerich, P. D., Lindsay, E. H., Clemens, W. A., Krause, transition at McGuire Creek, Montana. University of California
D. W., & Rose, K. D. (1987). First North American land mammals Publications in the Geological Sciences, 140, 1-185.
ages of the Cenozoic Era. In Cenozoic Mammals of North Amer- Lucas, S. G. (1989). Fossil mammals and Paleocene-Eocene boundary in
ica: Geochronology and Biostratigraphy, ed. M. 0. Woodburne, Europe, North America, and Asia. 28th International Geological
pp. 24-76. Berkeley: University of California Press. Congress (Washington, D.C.). Abstracts, vol. 2, p. 335.
Archibald, J. D., & Lofgren, D. L. (1990). Mammalian zonation near the (1992). Redefinition of the Duchesnean land mammal "age," late
Cretaceousffertiary boundary. In Dawn of the Age of Mammals in Eocene of western North America. In Eocene-Oligocene Climatic
the Northern Part of the Rocky Mountain Interior, North America, and Biotic Evolution, eds. D. R. Prothero & W. A. Berggren, pp.
eds. T. M. Bown & K. D. Rose, pp. 31-50. Boulder, Colo.: Geological 88-105. Princeton: Princeton University Press.
Society of America, Special paper 243. Lundelius, E. L., Jr., Churcher, C. S., Downs, T., Harington, C. R., Lindsay,
Barry, J.C., Lindsay, E. H., & Jacobs, L. L. (1982). A biostratigraphic zona- E. H., Schultz, G. E., Semken, H. A., Webb, S. D., & Zakrewski,
tion of the middle and upper Siwaliks of the Pot war Plateau of north- R. J. ( 1987). The North American Quaternary sequence. In Cenozoic
ern Pakistan. Palaeogeography, Palaeoclimatology, Palaeoecology, Mammals of North America, ed. M. 0. Woodburne, pp. 211-35.
37, 95-130. Berkeley: University of California Press.
Bernor, R. L., Fahlbusch, V., & Mittman, H.-W. (1996). The Evolution of Maas, M. C., & Krause, D. W. (1994). Mammalian turnover and community
Western Eurasian Neogene Faunas. New York: Columbia University structure in the Paleocene of North America. Historical Biology, 8,
Press. 91-128.
Damuth, J., & MacFadden, B. J. (1990). Body Size in Mammalian Paleo- MacFadden, B. J. (ed.). (1984). The origin and evolution of the Ceno-
biology: Estimation and Biological Implications. Cambridge: Cam- zoic vertebrate fauna of middle America. Journal of Vertebrate
bridge University Press. Paleontology, 4, 169-283.
Evander, R. L. (1986). Formal redefinition of the Hemingfordian-Barstovian Maglio, V. J., and Cooke, H. B. S. (eds.). (1978). Evolution of African
boundary. Journal of Vertebrate Paleontology, 6, 374-81. Mammals. Cambridge, Mass.: Harvard University Press.
Flynn, J. J. (1986a). Fauna! provinces and the Simpson coefficient. Con- Rich, T. H., Archer, M., Hand, S. J., Godhelp, H., Muirhead, J., Pledge, N.
tributions to Geology, University of ifyoming, Special Paper 3, S., Flannery, T. F., Woodburne, M. 0., Case, J. A., Tedford, R.H.,
317-38. Turnbull, W. D., Lundelius, E. L. Jr., Rich, L. S. V., Whitelaw, M. J.,
(1986b). Correlation and geochronology of middle Eocene strata from Kemp, A., & Rich, P. V. (1991 ). Australian Mesozoic and Tertiary ter-
the western United States. Palaeogeography, Palaeoclimatology, restrial mammal localities. In Vertebrate Palaeontology of Australa-
Palaeoecology, 55, 335-406. sia, eds. P. Vickers-Rich, J.M. Monaghan, R. F. Baird, & T. H. Rich,
Gillette, D. D. (1984). A marine icthyofauna from the Miocene of Panama, pp. 1005-69. Lilydale, Victoria, Australia: Pioneer Design Studio.
and the Tertiary Caribbean fauna! Province. Journal of Vertebrate Rose, K. D. (1980). Clarkforkian land-mammal age: revised definition, zona-
Paleontology, 4, 172-86. tion, and tentative intercontinental correlations. Science, 208, 744-6.
Gingerich, P. D. (1976). Cranial anatomy and evolution of early Tertiary Savage, D. E., & Russell, D. E. (1983). Mammalian Paleofaunas of the
Plesiadapidae (Mammalia, Primates). University ofMichigan Papers World. Reading, Mass.: Addison-Wesley.
on Paleontology, 15, 1-141. Storer, J.E. (1989). Rodent fauna! provinces, Paleocene-Miocene of North
(1983). Paleocene-Eocene fauna! zones and a preliminary analysis of America. In Papers on Fossil Rodents in Honor ofAlbert Elmer Wood,
Laramide structural deformation in the Clarks Fork Basin, Wyoming. eds. C. C. Black & M. R. Dawson, pp. 17-29. No. 33, Science Series.
ifyoming Geological Association Guidebook, 34, 185-95. Los Angeles: Natural History Museum of Los Angeles County.
 Introduction 7

Swisher, C. C., III, Dingus, L., & Butler, R. F. (1993). 40Ar!39Ar dating Whistler, D. P., & Burbank, D. W. (1992). Miocene biostratigraphy and
and magnetostratigraphic correlation of the terrestrial Cretaceous- biochronology of the Dove Spring Formation, Mojave Desert,
Paleogene boundary and the Puercan Mammal Age, Hell Creek- California, and characterization of the Clarendonian mammal age
Tullock formations, eastern Montana. Canadian Journal of Earth (late Miocene) in California. Geological Society ofAmerica Bulletin,
Sciences, 30, 1981-96. 104, 644-58.
Swisher, C. C., III, & Prothero, D.R. (1990). Single crystal 40 Ar/3 9 Ardating Williamson, T. E. (1993). The Beginning of the Age of Mammals in the
of the Eocene-Oligocene transition in North America. Science, 249, San Juan Basin: Biostratigraphy and Evolution of Paleocene
760-2. Mammals of the Nacimiento Fonnation. Unpublished Ph.D. Thesis.
Tedford, R.H., Skinner, M. F., Fields, R. W., Rensberger, J.M., Whistler, D. Albuquerque: University of New Mexico.
P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987). Williamson, T. E., & Lucas, S. G. ( 1993 ). Paleocene vertebrate paleontology
Fauna! succession and biochronology of the Arikareean through of the San Juan Basin, New Mexico. In Vertebrate Paleontology
Hemphillian (late Oligocene through earliest Pliocene epochs) in in New Mexico, eds. S. G. Lucas & J. Zidek. Bulletin of the New
North America. In Cenozoic Mammals of North America, ed. M. 0. Mexico Museum of Natural History and Science, 2, 105-35.
Woodburne, pp. 153-210. Berkeley: University of California Press. Wing, S. L. (1984). A new basis for recognizing the Paleocene/Eocene
Voorhies, M. R. (1990). Vertebrate biostratigraphy of the Ogallala group in boundary in Western Interior North America. Science, 226,
Nebraska. In Geologic Framework and Regional Hydrology: Upper 439-41.
Cenozoic Blackwater Draw and Ogallala Fonnations, Great Plains, Woodburne, M. 0. (ed.). (1987). Cenozoic Mammals of North America.
ed. T. C. Gustavson, pp. 115-51. Austin: Bureau of Economic Berkeley: University of California Press.
Geology, University of Texas. Woodburne, M. 0., Tedford, R. H., & Swisher, C. C., III. (1990). Lithos-
Webb, S. D. (1969). The Burge and Minnechaduza Clarendonian mam- tratigraphy, biostratigraphy, and geochronology of the Barstow
malian faunas of North-central Nebraska. University of California Formation, Mojave Desert, southern California. Geological Society
Publications in Geological Sciences, 78, 1-191. of America Bulletin, 102, 459-77.
 Part I: Overview of context for the evolution of North American
Tertiary mammals

1 The chronological, climatic, and paleogeographic background to North

American mammalian evolution

DONALD R. PROTHERO

INTRODUCTION principles followed by European stratigraphers at the time (Tedford,

1970). But the good beginning established by Osborn and Matthew
This chapter summarizes changes in North American climate and was essentially lost because the next generation of vertebrate pale-
landscape through the Tertiary. Since Webb's (1977) account, there ontologists ignored their pioneering work.
have been significant changes in the interpretation of North Ameri- Unlike most marine invertebrates, fossil mammals typically occur
can paleogeography and paleoclimate, largely due to more detailed in localized fossiliferous horizons or in isolated pockets or quarries
stratigraphic work and an increased understanding of global climatic without apparent stratigraphic superposition. As a result, traditional
history. biostratigraphic methods developed by European invertebrate pale-
Most significantly, our understanding of mammalian evolution is ontologists (based on detailed stratigraphic ranges of fossils in mea-
based on a chronological framework that has been radically changed. sured sections) were not widely followed by North American verte-
Although we still use the relative sequence of North American fau- brate paleontologists. In 1937, the Vertebrate Paleontology Section
nas first codified over fifty-five years ago by the Wood Committee of the Paleontological Society appointed a committee to clarify the
(Wood et al., ,1941), its geochronology and correlation with the confusion over correlation and to adopt a terminology that could be
global time scale has been fundamentally altered by breakthroughs widely used. Known as the Wood Committee (Wood et al., 1941),
in 40Ar/3 9Ar dating and magnetic stratigraphy. Some faunas that were it was chaired by Horace E. Wood II, and included four other ver-
once thought to be Eocene are now considered Paleocene; others tebrate paleontologists (Edwin H. Colbert, John Clark, Glenn L.
long considered early Oligocene are now late Eocene; much of the Jepsen, and Chester Stock), plus paleobotanist Ralph Chaney and
North American "early Miocene" is actually late Oligocene, and invertebrate paleontologist J. B. Reeside.
most of the original "Pliocene" of the Wood Committee is now late The Wood Committee's "Provincial Ages" were a complex hybrid
Miocene. Some of these changes are reflected in the most recent of local rock units and time units delineated by first and last occur-
summary of North American biochronology (Woodburne, 1987), rences of mammalian genera. As Tedford (1970) and Woodburne
but many others have occurred in the last few years. Because much (1977, 1987) have pointed out, these units were not true time-
of this new geochronology is still in progress, I cannot give a com- stratigraphic ages, which must be based on biostratigraphic zones
plete and detailed description here. However, this chapter summa- and stages (according to Western stratigraphic codes such as the
rizes the major changes that have occurred and describes how they 1983 North American Code of Stratigraphic Nomenclature). Be-
affect the correlation of North American fauna and climatic changes cause they were not true stratigraphic ages, the North American
with the rest of the world. land mammal "ages" should properly be put in quotes in all publica-
tions. Instead, the Wood Committee's methods resemble a system
called "biochronology," first proposed by H. S. Williams (1901),
PRINCIPLES OF NORTH AMERICAN
which attempts to reconstruct the sequence of occurrences of taxa
LAND CHRONOLOGY
without necessarily documenting every event in a stratigraphic sec-
In North America, the only practical method of correlating and tion. Demonstration of stratigraphic superposition was noted when
dating most Cenozoic terrestrial deposits has been with land mam- available, but detailed work in the European biostratigraphic tradi-
mals. At the turn of the century, William Diller Matthew and Henry tion was not considered essential to their "provisional" system.
Fairfield Osborn (Matthew, 1899; Osborn and Matthew, 1909; In spite of their loose characterization, the land mammal "ages"
Osborn, 1907, 1910, 1929) attempted to create biostratigraphic of the Wood Committee worked quite well for over forty years,
zonations of the North American terrestrial. sequence based on mostly because mammals do evolve and disperse very rapidly

9
 Donald R. Prothero 10

(Savage, 1977). However, problems eventually arose. For example, The most recent breakthrough has been due to the application
the Chadronian was originally defined both on the co-occurrence of of magnetic stratigraphy to terrestrial sections. Unlike any other
Mesohippus and brontotheres, and also on the limits of the Chadron method, magnetic stratigraphy can supply many globally synchro-
Formation. At the time, the last occurrence of brontotheres was nous, numerically dated time horizons to terrestrial sections
thought to coincide with the top of the Chadron Formation, so (Lindsay et al., 1987; Prothero 1988, 1990; Opdyke, 1990). The
there was no conflict. When Morris Skinner documented brontothere combination of both magnetic stratigraphy and 40Ar/3 9Ar dating has
specimens in rocks correlative with the overlying Orella Member provided much higher temporal resolution and precision than was
of the Brule Formation, however, the difficulty with defining the thought possible just twenty years ago. More importantly, magnetic
Chadronian both biochronologically and lithostratigraphically be- stratigraphy is the only technique that allows direct correlation with
came apparent (Prothero, 1982; Emry, Bjork, and Russell, 1987; the global polarity record, and thus with the marine time scale. This
Evanoff, Prothero, and Lander, 1992). Yet many of Skinner's con- in turn allows us to make direct comparison between global climatic
temporaries could not accept this evidence because the misconcep- changes, diversity fluctuations, and mass extinctions and the North
tion that rock units could be treated as if they were time units was American terrestrial record for the first time.
widespread among paleontologists of that generation.
Since the 1950s, another generation of paleontologists (see
Savage, 1955, 1962, 1977;Tedford, 1970;Woodburne, 1977, 1987) PALEOCENE
has tried to bring vertebrate paleontology back to classical biostrati-
graphic methods. Trained in modern stratigraphic thinking, verte-
CHRONOSTRATIGRAPHY AND
brate biostratigraphers now appreciate the possibility that rock units
BIOSTRATIGRAPHY
can be time transgressive over distance (Shaw, 1964; Prothero,
1990), and so they rarely confuse rock units with time units. In The term "Paleocene" was originally proposed in 1874 by the pa-
addition, recent generations have come to realize that detailed strati- leobotanist W. P. Schimper for strata in Europe with a distinctive
graphic zonations of mammal fossils provide much higher resolution assemblage of fossil plants (Schorn, 1971). Carved out of a se-
of time than do collections whose only stratigraphic information ries of rock units that were part of the original Lyellian Eocene
is the formation from which they came. As discussed by Wood- sequence in France, the Paleocene was difficult to recognize be-
burne (1977), such detailed zonation could potentially subdivide cause its characteristic fossil plants did not occur widely outside
the Cenozoic into increments of time of 300,000 years or less. In the type area, and many of the rock units in Schimper's origi-
recent years, much of the North American continental Tertiary has nal list were clearly Eocene (Berggren, 1971; Savage and Russell,
been formally subdivided by proper biostratigraphic methods. How- 1983). As a result, early twentieth-century vertebrate paleontolo-
ever, the zonation of the entire Cenozoic is still in progress, and gists (e.g., Osborn, 1929) used "lower Eocene" for beds we now
biochronological methods are still widely used. consider Paleocene, and older literature must be read with this in
Although mammalian paleontologists are increasingly moving mind. The U.S. Geological Survey did not adopt the term until 1939
toward classical biostratigraphic procedures, there are still prob- (Russell, 1964).
lems. For example, biostratigraphic "zonations" proposed by Two years later, the Wood Committee (1941) created five land
Gingerich (1980, 1983) and Gunnell (1989) do not meet all the mammal "ages" for an interval they considered Paleocene (Fig-
criteria established by the North American Code of Stratigraphic ure 1.1 ). The first two, the Puercan and Torrejonian, were based on
Nomenclature. Typically, these "zones" do not have type sections faunas from the San Juan Basin in New Mexico. The controversial
(required in Article 54e), and in some cases, the actual local strati- Dragonian "age" was based on the limited Dragon Canyon local
graphic ranges of key taxa are not clearly indicated. In this sense, fauna from the North Horn Formation of central Utah. Although
they are "biochrons" based on the abstract first and last occurrences correlative faunas were found in New Mexico (Tomida, 1981 ), most
of taxa, not true biostratigraphic zones and stages, which must be paleontologists now consider the Dragonian to be the beginning of
based on local ranges of fossils in a particular section. the Torrejonian (Archibald et al., 1987). The Tiffanian was orig-
In addition to higher resolution biostratigraphy, other techniques inally based on faunas from the San Juan Basin in Colorado, but
have come along to improve terrestrial correlations. The original since then the Tiffanian has become much better known from fau-
application of K-Ar dating by Evernden, Savage, Curtis, and James nas in Montana and Wyoming. The Clarkforkian Land Mammal age
( 1964) to the North American terrestrial record provided an indepen- was originally based on faunas from the Clark's Fork Basin on the
dent test of the Wood Committee sequence, and showed it to be sub- Montana Wyoming border. Originally considered latest Paleocene
stantially correct. For over twenty-five years, K-Ar methods con- by the Wood Committee (1941), in recent years it was thought to
tinued to refine the chronology and provided numerical estimates of straddle the Paleocene/Eocene boundary based on correlations of
the age for most of the sequence (Savage, 1977; Savage and Russell, plesiadapids from Europe and North America (Gingerich, 1976;
1983; Woodbume, 1987). However, in the last ten years, the de- Gingerich and Rose, 1977; Rose, 1981). As we see in the next
velopment of 40Ar!3 9Ar dating (McDougall and Harrison, 1988) has section, the Paleocene/Eocene boundary is now thought to occur in
shown that many of the classical K-Ar dates must be recalibrated. the earliest Wasatchian, so not only the Clarkforkian but also the
In some cases, they have radically changed the chronology that was earliest Wasatchian are now considered Paleocene.
accepted for decades (Swisher and Prothero, 1990; Prothero and North American Paleocene chronology was most recently sum-
Swisher, 1992). marized by Archibald et al. (1987). Most of the Paleocene has now
 Background to North American mammalian evolution 11

San Juan Basin, NM

East Fork Basin, WY Figure 1.1. Paleocene-early Eocene
time scale for North America. Global
time scale and magnetic polarity time
scale after Berggren et al. (1995).
North American land-mammal "ages"
!M!JJfi!JJfiiliJ@i@[ii.§jH!it§fiJ]!@j Crazy Mtn. Basin, MT (NALMA) and "subages" after Gun-
nell (1989). Temporal span of im-
Buck Hill-Vieja Groups, TX portant magnetostratigraphically cal-
ibrated sections are shown; sources
lfd!Ji!@J!afJ!lJ:tl&.1J.J@ Dragon Canyon, UT
are given in text. CF. = Clarkforkian;
Puer. = Puercan; Torrejon. = Torre-
Big Bend, TX jonian.

been subdivided into a series ofbiostratigraphic "zones," abbreviated 61.3 Ma), so the three Torrejonian "zones" Tol-To3 range from
"Pul, Pu2, Pu3, Tol" ("Puercan 1, 2, 3, Torrejonian 1") and so on, 0.3-1.2 million years in duration. The TorrejonianfTiffanian bound-
under a scheme originally introduced by Gingerich (1976, 1980, ary occurs early in Chron C26r, about 61 Ma. Based on magnetic
1983) based on plesiadapids and primates (Figure 1.1). As al- stratigraphy from the northern Bighorn Basin, Butler, Gingerich,
ready noted, these "zones" do not meet the criteria of the North and Lindsay (1980, 1984) showed that Ti2 occurs early in Chron
American Code of Stratigraphic Nomenclature because they lack C26r (about 59-60 Ma), and Ti3 in late Chron C26r and C26n
type sections. Although there are some difficulties with this zonation (57.5-58.5 Ma). Ti4 occurs in earliest Chron C25r (57.0-57.3 Ma),
(Schankler, 1980, 1981), the scheme has been modified with addi- and Ti5 at the base of Chron C25n (56.1-56.3 Ma). The Tiffa-
tional mamma,Jian groups and has been widely adopted (Archibald nian/Clarkforkian boundary occurs in Chron C25n, about 56.0 Ma,
et al., 1987). There are three Puercan "zones," three Torrejonian so the five Tiffanian "zones" range from 0.2-1.0 million years in du-
"zones," six Tiffanian "zones," and three Clarkforkian "zones" (plus ration. The Clarkforkian/Wasatchian boundary occurs in the middle
Wasatchian 0, which is also latest Paleocene). of Chron C24r, about 55.5 Ma, so the three Clarkforkian "zones"
Relatively few radiometric dates are available for the Paleocene. are each about 150,000 years in duration.
The most recent dates on the Cretaceousffertiary boundary place
its age at 65 Ma (Berggren et al., 1995). Revisions to the dating
PALEO CLIMATE
of the magnetic polarity time scale and new dates on the Eocene
(Berggren et al., 1992, 1995; Cande and Kent, 1995) place the Paleoclimatic evidence from the Paleocene of North America sug-
Paleocene/Eocene boundary at about 55 Ma. Magnetic polarity gests that the warm, subtropical "greenhouse" conditions that dom-
stratigraphy (Figure 1.1) has been studied in the key sections of the inated in the Cretaceous persisted into the Paleocene (Wing, this
Bighorn Basin of Wyoming and Montana (Butler, Gingerich, and volume, Chapter 2). This is consistent with other data from around
Lindsay, 1980, 1984), the Crazy Mountains in Montana (Butler, the globe that argue for a much warmer, "greenhouse" planet with a
Krause, and Gingerich, 1987), the San Juan Basin in New Mex- reduced pole-to-equator temperature gradient during the Paleocene.
ico (Butler et al., 1977; Lindsay, Jacobs, and Butler, 1978; Taylor However, there was a slight cooling compared to the Late Creta-
and Butler, 1980; Lindsay, Butler, and Johnson, 1981; Butler and ceous, especially in the Tiffanian and Clarkforkian. Oceanic tem-
Lindsay, 1985), Dragon Canyon in Utah (Tomida and Butler, 1980), peratures were quite high, although estimates vary widely. For
and the Big Bend region in Texas (Rapp, MacFadden, and Schiebout, example, Shackleton and Kennett (1975) calculated average sea
1983). surface temperatures of l8°C in the sub-Antarctic, and bottom wa-
Based on magnetic stratigraphy from the San Juan and Crazy ter temperatures about l6°C at 1,000 m depth. However, Miller,
Mountain Basins, Butler and Lindsay (1985) and Butler, Krause, Fairbanks, and Mountain ( 1987) obtain estimates closer to 10°C for
and Gingerich (1987) squeezed the Puercan into Chron C29n and the deep waters of the Atlantic in the Paleocene. Because both South
iatest C29r (64-65 Ma), so Puercan "zones" Pu0-Pu3 are each about America and Australia were still connected to Antarctica, there was
250,000 years in duration. In the San Juan and Crazy Mountain no circum-Antarctic circulation. Today, this current traps polar wa-
Basins, Torrejonian "zone" Tol occurs in Chron C28n (62.5-63.5 ters in a clockwise flow around the Antarctic in the Southern Ocean,
Ma), To2 in C27r (61.3-62.5 Ma), and To3 in Chron C27n (61.0- allowing cold conditions to build up and accumulate a permanent
 Donald R. Prothero 12

Antarctic ice sheet. In the Paleocene and Eocene, polar waters mixed by the Paleocene. In California, the onset of the proto San An-
with equatorial waters, ameliorating temperatures and climate and dreas fault and its associated crustal stresses produced a variety of
promoting a low temperature gradient between poles and equator. early Cenozoic basins, most filled with deep marine turbidites and
In addition, a warm tropical Tethys seaway still stretched from the shales of Paleocene age (Nilsen, 1977, 1987). Major portions of
Mediterranean to Southeast Asia, contributing to the homogeneity coastal California were still hundreds of kilometers south of their
and warmth of the low latitudes. present position and acted as emergent islands, shedding submarine
fan complexes in all directions. The Salinian block, which now un-
derlies most of the California Coast Ranges, has apparently traveled
PALEOGEOGRAPHY about 2,600 km northwesterly since the late Cretaceous (Kanter and
In many ways, the paleogeographic characteristics of the Paleocene Debiche, 1985).
were a continuation of those found in the Cretaceous. The Laramide East of the Rockies, most of the Midwest and Appalachian region
Orogeny, which began in the latest Maastrichtian, uplifted high was emergent and accumulated little or no Paleocene record. Fos-
basement-cored mountain ranges in the Rocky Mountain region; sil mammals are rare in this region, but do occur (Schoch, 1985).
these same areas had been under marine seaways as recently as the The Atlantic and Gulf Coast passive margins continued to sink and
early Maastrichtian (Dickinson et al., 1988). The last of these great accumulate thick floodplain and marine shelf sequences, as they
epeiric seaways, the Cannonball Sea, covered much of Montana had been doing since the North Atlantic began to open in the Juras-
and North Dakota in the early Paleocene; its final retreat marked sic. The Atlantic margin subsided relatively slowly, accumulating
the end of the repeated transgressions of epicontinental seas that only about 2 km of Cenozoic sediment (covering over 10 km of
had lasted throughout the Cretaceous and much of the Paleozoic. Cretaceous deposits). The Gulf margin, by contrast, had a much
As the Laramide ranges rose, deep basins formed throughout the more abundant supply of elastics from the drainages coming off
Rocky Mountains, which filled up with Paleocene and Eocene flu- the Laramide Orogeny, producing over 12 km of Cenozoic sedi-
vial and lacustrine sediments (Dickinson et al., 1988). Some of these ments alone (Frazier and Schwimmer, 1987; Galloway et al., 1991).
basins still contain Paleocene-Eocene deposits as thick as 10,000 m Florida and the Bahamian Platform, isolated from elastic sediments,
(33,000 feet), producing our primary record of the early Cenozoic were carbonate banks and reef complexes, as they had been since
in western North America. the Cretaceous.
Although there were regional differences, most of these Laramide
basins were accumulating thick sequences of swampy floodplain
deposits during the Paleocene (Dickinson et al., 1988). The Fort EOCENE
Union Group fills many of these intermontane basins in Wyoming,
Montana, and North Dakota. In the Powder River Basin of
CHRONOSTRATIGRAPHY
Wyoming and Montana, Fort Union sediments were formed in
broad swampy floodplains that deposited coal seams over 100 m The Eocene was one of the three original Cenozoic epochs estab-
(330 feet) thick in places (Flores, 1981, 1983; Flores and Ethridge, lished by Lyell (1831-1833, p. 55). Lyell's conception of the
1985). The Bighorn Basin Fort Union sediments were formed in a Eocene was based on a "clock" model of turnover of molluscan
similar swampy floodplain environment. In the San Juan Basin of faunas and not well suited to traditional stratigraphic procedures
New Mexico and Colorado, the Paleocene Nacimiento Formation (Rudwick, 1978; Berry, 1987). Although Lyell gave some idea of
(from which the "type" Puercan, Torrejonian, and Tiffanian faunas what he considered "typical" deposits of Eocene age, his list led to
came) was deposited in a swampy floodplain setting, with abundant much confusion. Indeed, some of the units were clearly not Eocene
subtropical plants, fish, crocodiles, and aquatic turtles in addition in the modern sense (Berggren, 1971; Savage and Russell, 1983;
to the mammals (Smith, Lucas, and Elston, 1985). On the edges Berggren, Kent, and Flynn, 1985). Based on studies of marine mi-
of rapidly uplifting Laramide basins were active alluvial fan com- crofossils in the various type areas of the marine Eocene stages in
plexes, which produced thick sequences of Paleocene conglomer- Europe, a consensus has gradually emerged as to what constitutes
ates in Utah, Idaho, and Montana (Ryder, Fouch, and Elison, 1976; "Eocene" and how it correlates with global marine planktonic record
Dickinson et al., 1988). and magnetic polarity time scale (Berggren, 1971; Hardenbol and
Outside the Laramide tract, North America was much less ac- Berggren, 1978; Aubry, 1985; Berggren, Kent, and Flynn, 1985;
tive tectonically. The great Sierra Nevada volcanic arc ceased to Aubry et al., 1988). Recently, however, new radiometric dates have
erupt in the Maastrichtian, probably as a result of shallow subduc- forced a significant recalibration of the Eocene time scale (Berggren
tion, which may have caused the Laramide Orogeny (Dickinson et al., 1992; Cande and Kent, 1992, 1995; Berggren et al., 1995).
and Snyder, 1978; Keith, 1978; Dickinson, 1979). However, the Although the relative sequence of biostratigraphic stages and their
Sierran volcanic chain was still a high drainage divide that caused magnetic polarity correlations are still valid, the numerical age es-
rivers to flow east across Nevada and Idaho into the Laramide la- timates of most Paleogene events are 1-2 million years too old in
custrine basins (Nilsen and McKee, 1979). Most of Oregon and most publications.
Washington were part of a marine basin with an actively subduct- Fortunately, the type areas of many of the European marine stages
ing trench, and large portions of coastal Oregon and Washington interfinger with mammal-bearing beds in both the Paris and London
were exotic terranes that had still not accreted to North America Basins (Savage and Russell, 1983). This allows direct correlation of
 Background to North American mammalian evolution 13

l;.l l;.l l;.l

U\ .j>. l;.l
I

~-
~ I
I
l:WBU1J!SMiJi!@.!fji\!JiiJfilji!jJ San Diego, CA

Washakie Fm., WY

Figure 1.2. Middle Eocene-Oligo-

cene time scale for North Amer-
ica. Global time scale and magnetic
Flagstaff Rim, WY polarity time scale after Berggren
et al. (1995). Temporal span of im-
Douglas, WY portant magnetostratigraphically cal-
ibrated sections are shown; sources
White River Group are given in text. Lt. Olig. = Late
NE-WY-SD-ND Oligocene; Orel!. = Orellan.

the European Eocene mammalian chronology with the global time ocene/Eocene boundary (Berggren, 1993). These correlations were
scale. During intervals of fauna! interchange with North America borne out by the detection of the striking carbon isotopic event near
(such as in the early Eocene, but not the middle or late Eocene), we the Paleocene/Eocene boundary (Rea et al., 1990; Kennett and Stott,
can correlate the North American sequence with the Lyellian epochs 1991) in earliest Wasatchian terrestrial carbon isotopes extracted
based in Europe. We also have magnetic stratigraphy and a large from paleosols and mammalian teeth and bones (Koch, Zachos, and
number of radiometric dates for many areas in the North American Gingerich, 1992).
Eocene. New radiometric dates, however, have greatly changed our Yet some mammalian paleontologists (Gingerich, 1989; Gunnell,
notions of the middle and late Eocene (Swisher and Prothero, 1990; Bartels, and Gingerich, 1993) continue to place the Paleocene/
Prothero and Swisher, 1992). Eocene boundary at the Clarkforkian/Wasatchian boundary. The key
When the Wood Committee subdivided what they considered the to the entire controversy lies in the interpretation of the Sparnacian
Eocene beds of North America, they recognized four land mammal mammalian faunas of Europe. In the past, European mammalian
"ages" (Figures 1.1, 1.2). The Wasatchian was named for the paleontologists had considered the Sparnacian the beginning of the
Wasatch Formation in basins of southern and eastern Wyoming. Eocene because its mammals were more similar to later Eocene
The Bridgerian got its name from the faunas of the Bridger Basin in faunas and differed radically from the underlying upper Paleocene
southwest Wyoming, and the Uintan and Duchesneau were named Thanetian faunas (Savage and Russell, 1983). The discovery of a
for the Uinta and Duchesne River Formations of northeast Utah. new earliest Wasatchian fauna (WaO) in North America with strong
The Wasatchian was considered early Eocene, the Bridgerian mid- similarities to the Sparnacian in Europe supported that correlation
dle Eocene, and the Uintan and Duchesneau were thought to be late (Gingerich, 1989, pp. 83-87). Gunnell, Bartels, and Gingerich
Eocene. The next "age," the Chadronian (named after the Chadron (1993) argued that the beginning of the Clarkforkian/Wasatchian
Formation in the High Plains, especially Nebraska and South transition was the time of greatest fauna! turnover, and therefore it
Dakota) was considered early Oligocene. seemed to be the best place to mark the boundary.
However, a number of recent lines of evidence suggest that the However, the Paleocene/Eocene boundary is defined on the ba-
Paleocene/Eocene boundary actually falls within the earliest sis of marine taxa in European type sections, not on the basis
Wasatchian. The original correlations of Gingerich and Rose were of North American or European mammalian faunas. Where the
based primarily on the interpretation of adapid primates and ple- Clarkforkian/Wasatchian boundary in North America (or the Thane-
siadapids. Other taxa, such as pantodonts (Lucas, 1984, 1989, tian/Sparnacian boundary in Europe) falls in relation to marine stra-
1993; Rea et al., 1990) and omomyid primates (Beard and Tabrum, totypes must be demonstrated, not asserted. One cannot assume
1991) suggested that the boundary occurs in the earliest Wasatchian. the coincidence of mammalian faunal turnover and European epoch
Another datum is the first appearance of Platycarya pollen in the boundaries. For instance, the "Grande Coupure" in Europe was
earliest Wasatchian of the northern Bighorn Basin (Wing, Bown, long thought to represent the Eocene/Oligocene boundary because
and Obradovich, 1991). This palynological datum occurs at the of the great fauna! turnover, but more recent work (Hooker, 1992)
NP9/NP10 nannofossil zone boundary in the Gulf Coast has demonstrated that it actually occurred in the early Oligocene,
(Frederiksen 1980), which is 300,000 years older than the Pale- about 2 My after the Eocene/Oligocene boundary.
 Donald R. Prothero 14

Marine stratigraphers have long ago shown that the European stages (Figures l.l, l.2). However, biostratigraphic zonations for
Sparnacian correlates with late Paleocene nannoplankton zone NP9 the Wasatchian and Bridgerian have been proposed by Savage
(Costa and Downie, 1976; Costa, Denison, and Downie, 1978; (1977), Gingerich (1976, 1980, 1983), Gingerich and Simons
Berggren, Kent, and Flynn, 1985; Berggren et al., 1995; Aubry (1977), Schankler (1980), Stucky (1984), and Gunnell (1989).
et al., 1988). It is at least one or two sequences lower (and about l Some of these biostratigraphic schemes (especially those of Savage,
million years older) than the Paleocene/ Eocene boundary as denoted 1977; Schankler, 1980; and Stucky, 1984) follow the 1983 North
by the base of the London Clay or the Argile d'Ypres (Berggren, American Stratigraphic Code in specifying type sections, but oth-
1993), long recognized as the base of the Ypresian and therefore the ers do not. Currently, the Wasatchian is divided into eight "zones,"
base of the Eocene (Berggren, Kent, and Flynn, 1985; Aubry et al., labeled WaO to Wa7 (Gingerich, 1983, 1989). Clyde, Stamatakos,
1988). and Gingerich (1994) found that WaO to lower Wa5 occurred in
After the Wasatchian, North America was separated from di- Chron C24r (53.5-55 Ma) in the northern Bighorn Basin, and that
rect interchange with Europe, and the correlations become more upper Wa5 and Wa6 correlated with C24n (52.3-53.4 Ma). Wa7
indirect. As a result, the most effective technique has been ra- occurred in C23r (51.6-52.3 Ma). This is good agreement with a
diometrically dated magnetic stratigraphy (Figures l.l, l.2). The new 40Ar!3 9Ar date on Wa6-7 of 52.8 +I- 0.3 Ma.
magnetic stratigraphy of the type Bridgerian in the Bridger Basin Gunnell (1989) also named two "zones" (Uil, the Epihippus as-
of Wyoming has been studied, but not published (Jerskey, 1981). semblage "zone," and Ui2, the camelid-canid appearance "zone")
The magnetic stratigraphy of the classic sections of the Uintan and for the early and late Uintan. However, unlike earlier Eocene zones,
Duchesneau in the Uinta Basin of Utah was summarized by Prothero these are not based on recent detailed biostratigraphic work, but
and Swisher (1992) and Prothero (1996a). Bridgerian and Uintan simply formalize the distinction between the faunas of Uinta For-
beds have been studied in Wyoming by Flynn (1986) and in Texas mation Member "B" (Wagonhound Member of Wood, 1934) and
by Walton (1992) and Prothero (1996b). Uintan-Duchesnean beds Member "C" (Myton Member of Wood, 1934). In addition, both of
in the Sespe Formation of Ventura County, California (Prothero, Gunnell's (1989) names are inappropriate because camelids appear
Howard, and Dozier, I 996), the Poway Group in San Diego County, in the early Uintan in the Washakie Basin (Mccarroll, Flynn, and
California (Flynn, 1986; Walsh, Prothero, and Lundquist, 1996), Turnbull, 1996), and Epihippus ranges throughout the Uintan. More
the Galisteo Formation in central New Mexico (Prothero and Lucas, detailed biostratigraphic work to subdivide the 6 million years of the
l 996), and several units in western Montana (Tabrum, Prothero, and Uintan in the Uinta Basin is presented in Prothero (l996a). How-
Garcia, 1996) have also been sampled. These studies show that the ever, the high degree of endemism of Uintan faunas makes true first
Wasatchian/Bridgerian boundary occurs in Chron C22r, about 50.4 and last occurrences difficult to distinguish from local geographic
Ma. The Bridgerian-Uintan transition occurs late in Chron C20n, effects (Lillegraven, 1979; Flynn, 1986; Krishtalka et al., 1987), so
about 46.5 Ma. The Uintan-Duchesnean transition occurs within a complete zonation for the Uintan is not yet available.
Chron Cl8n, about 40 Ma. The latest stratigraphic correlations The Duchesneau has always been the most controversial land
of the middle Eocene Bartonian and late Eocene Priabonian stages mammal "age" of the Eocene. Originally considered latest Eocene
in Europe (Berggren et al., 1995) place the Bartonian/Priabonian by Wood et al. (1941), opinions have fluctuated from placing it in the
boundary in Chron Cl 7nl, so that the middle Eocene includes not early Oligocene (Scott, 1945) or back in the late Eocene (Simpson,
only the Bridgerian, but also the Uintan and Duchesneau. 1946), and back and forth several times since then. Gazin (1955,
The most surprising conclusion of recent dating, however, con- 1956, 1959) assigned the faunas from the lower Duchesne River For-
cerns the Chadronian Land Mammal age. Considered early Oligo- mation (Randlett and Halfway faunas) to the late Uintan. The poor
cene by the Wood Committee, it had been K-Ar dated by Evernden quality of the overlying Lapoint fauna (the "classic" Duchesnean),
et al. (1964) at between 32 and 36 Ma. Because the Eocene/Oligo- and its similarity to the Chadronian fauna, led several workers to
cene boundary was generally placed around 36.5 Ma (Berggren, either reduce the Duchesneau to a subage of the Chadronian or drop
Kent, and Flynn, 1985), the correlation of the Chadronian with the it altogether (Wilson, 1978, 1984, 1986; Emry, 1981). In recent
early Oligocene seemed secure. However, the Eocene/Oligocene years, opinion has swung back toward recognizing a distinct Duch-
boundary has been recently redated at 33.5 Ma (Berggren et al., esnean, with important correlatives in Saskatchewan, South Dakota,
1995). At the same time, redating of the K-Ar-dated ashes by Wyoming, Montana, Texas, New Mexico, Oregon, and California
40Ar!39Ar methods have shown that the Chadronian spans an interval (Krishtalka et al., 1987; Kelly, 1990; Lucas, 1992).
from 34-37 Ma (Swisher and Prothero, 1990; Prothero and Swisher, Even though the Duchesneau spans nearly 3 million years, efforts
1992). This places the Chadronian at the end of the Eocene, and to subdivide it into biostratigraphic zones have been controversial
the Eocene/Oligocene boundary appears to fall near the Chadro- because the faunas are so sparse and often endemic to the many
nian/Orellan boundary, not at the Duchesnean/Chadronian bound- scattered localities (Lucas, 1992). Several authors have attempted
ary, as long thought. to recognize an early and late Duchesneau in Texas (Wilson, 1984,
1986) and California (Kelly, 1990). Further radiometric dating and
magnetic stratigraphy will probably provide the best test of the age
BIOSTRATIGRAPHY of these faunas and determine whether a zonation can be constructed
The most recent summary of most of the Eocene North for the Duchesneau.
American land mammal "ages" (Krishtalka et al., 1987) did not At the time of the Wood Committee report, the biostratigra-
attempt to divide formally the entire interval into biostratigraphic phy of the Chadronian was very poorly understood. Many of the
 Background to North American mammalian evolution 15

taxa (especially brontotheres and oreodonts) were badly oversplit, trend was much more dramatic. In areas like the Big Badlands,
and the stratigraphic data on the fossils from the classic collec- ancient soil horizons give a vivid picture of the landscape roamed
tions in the Chadron Formation were inadequate (Emry, Bjork, and by the land mammals. According to Greg Retallack (1981, 1983a,
Russell, 1987). Attempts to divide the Chadronian based on lithos- 1983b, 1990, 1992), early late Eocene forests received about 1,000
tratigraphy of the Chadron Formation in Nebraska (Schultz and mm of annual precipitation. By the latest Eocene, the forests had
Stout, l 955) or South Dakota (Clark, 1937, 1954; Clark, Beerbower, changed to a dry woodland with open patches of grasses and herbs.
and Kietzke, 1967) were unsuccessful (Emry, 1973; Emry, Bjork, This allowed archaic browsers, like the brontotheres, to persist.
and Russell, 1987). Since that time, however, Emry (1973, 1992) The more varied, open habitat gave room for a variety of mixed-
has carefully documented the mammalian biostratigraphy in the feeding herbivores, some of which could eat tougher grasses and
thickest and most fossiliferous Chadronian sequence at Flagstaff herbs.
Rim, Wyoming. Based on this work, Emry (1992; Emry, Bjork, and What caused this dramatic cooling around the globe? Although
Russell, 1987) suggested criteria for recognizing early, middle, and there are still major arguments about this "doubt house" world be-
late Chadronian, but no formal biostratigraphic zonation was pro- tween the early Eocene "greenhouse" and the Oligocene "ice house,"
posed. Suggestions of a zonation for the Chadronian are presented evidence is accumulating that there were significant glaciers in
by Prothero and Emry (1996b). Antarctica as early as the middle Eocene, and major pulses in glacia-
tion in the late Eocene (Miller, Wright, and Fairbanks, 1991; Wise
et al., 1992). Evidence of middle Eocene mountain glaciers on the
PALEO CLIMATE
Antarctic Peninsula (Birkenmajer, 1987) and in the Pacific sector
From the slight cooling trend of the Paleocene, the globe underwent of the Southern Ocean (Margolis and Kennett, 1971; Wei, 1989)
a dramatic warming at the beginning of the Eocene. Global tempera- suggests a short-term ice buildup during the middle Eocene. The
tures may have risen by about 5°C (Miller, Fairbanks, and Mountain, ultimate cause of the cooling that led to this glaciation is still very
1987), reaching the warmest state since the mid- controversial. McGowran (1989) attributes some of it to a "reverse
Cretaceous. Changes in the carbon isotopes indicate a dramatic greenhouse," with the excess greenhouse gases in the atmosphere
alteration of deep-water circulation, and mass extinction of benthic returning to the lithosphere in the form of lignites. Bartek et al.
forarniniferans is thought to have been caused by a sudden warm- (1992) point out that increased seafloor spreading between Aus-
ing of deep marine waters, so that the water column was almost tralia and Antarctica produced a broad area of ocean that could have
isothermic from the shallow surface waters to the bottom (Rea et al., supplied moisture for snow and ice to the Antarctic for the first
1990; Kennett and Stott, 1991). The causes of this dramatic change time since the early Mesozoic. If, as they argue, the Antarctic was
are controversial, although an equatorial warm saline deep water already cold enough for ice caps, the triggering factor for middle-
(Kennett and Stott, 1990, 1991) or a sudden increase in greenhouse late Eocene glaciation may have been the supply of moisture in the
gases, such as C02 from increased seafloor spreading (Rea et al., proper location.
1990) or methane from Eocene swamps (Cirbus Sloan et al., 1992),
have been suggested. This dramatic warming was reflected in early
PALEOGEOGRAPHY
Eocene landscapes and biotas all over the world, including North
America (Wing, this volume, Chapter 2). The Laramide Orogeny, which had changed the North
After the early Eocene peak in warmth, global climate deterio- American landscape since the latest Cretaceous, continued into the
rated through the rest of the Cenozoic. The effects are apparent in early Eocene. Many of the Rocky Mountain basins that began to
North America as early as the middle Eocene. Wolfe (1978) and develop and fill in the Paleocene contain even thicker Eocene se-
Wing (this volume, Chapter 2) describe two episodes of cooling in quences (Dickinson et al., 1988). Most of these units, such as the
the Eocene of North America, one at about 50 million years ago Willwood Formation of the Bighorn Basin of Wyoming and Mon-
(Wasatchian-Bridgerian transition) and another at about 38-40 mil- tana, the Wasatch and Wind River Formations in other basins in
lion years ago (approximately the Uintan-Duchesnean transition). Wyoming, and the San Jose Formation in the San Juan Basin of
As discussed by Wing (this volume, Chapter 2), the floral record New Mexico, were deposited on broad swampy floodplains (Smith,
showed changes at the beginning and the end of the middle Eocene Lucas, and Elston, 1985; Wing and Bown, 1985). In the middle
that seem to correspond to the oceanic cooling shown in the oxygen Eocene, thick lacustrine sequences of the Green River Formation
isotopic record. The early-middle Eocene transition is represented accumulated in the Green River and Washakie Basins of Wyoming,
by only slight evidences of cooling and drying. The middle-late the Piceance Basin of Colorado, and the Uinta Basin of Utah. The
Eocene transition, on the other hand, shows significant evidence of overlying fluvial Uinta, Duchesne River, and Washakie Formations
cooling of about 7-I 1°C (l l-l5°F) in northwestern North America show that these lakes had receded by the late middle Eocene and
and a loss of tropical elements in both Europe and North America. late Eocene (Dane, 1955; Bradley, 1964; Ryder, Fouch, and Elison,
Clearly, the tropical rain forests of the early Eocene were on the 1976; Johnson, 1985).
decline around the world. Continued tectonism along the active margin of the Pacific pro-
Yet warm conditions were not gone forever. In some marine duced a wide variety of Eocene basins in California, Oregon, and
isotopic records, there is evidence of a slightly warming trend in Washington (Nilsen, l 987). High sea levels during the Eocene,
the late Eocene, just before the great cooling event in the early however, meant that most of these basins were filled with marine
Oligocene. According to Wolfe (1978), the late Eocene warming sediments, and only a few have nonmarine deposits containing fossil
 Donald R. Prothero 16

mammals. Notable among these were the interfingering of ma- original "type" sections were inadequate, marine stratigraphers have
rine/nonmarine units in the San Diego region (Lohmar and Warme, long sought better sections that preserve the transition without un-
1979; Flynn, 1986; Walsh, Prothero, and Lundquist, 1996) and conformities. Scientists who participated in International Geologi-
the Sespe Formation in Ventura County, California (Dibblee, 1966; cal Correlation Project 174 met in the 1980s to settle the question,
Taylor, 1983; Kelly, 1990; Prothero, Howard, and Dozier, 1996). and their results have been published in several places (Pomerol
Basins in Washington and Oregon were also filled with marine de- and Premoli-Silva, 1986; Premoli-Silva, Coccioni, and Montanari,
posits, including major deltaic complexes in central Oregon. During 1988). From these conferences, a consensus emerged that the
the Eocene, a number of seamounts and other exotic terranes appar- Eocene/Oligocene boundary should be defined at the last appear-
ently collided with the Pacific Northwest, dramatically increasing ance of the spiny planktonic foraminiferan Hantkenina.
the land area (Cady, 1975). Eventually, these seamounts and other At the 24th International Geological Congress in Washington,
exotic fragments clogged up the ancient subduction zone in cen- D.C., in July 1989, the International Subcommission on Paleogene
tral Washington and Oregon, and forced it to jump westward to Stratigraphy met and considered the recommendation of Isabella
near its present position. At the same time, the entire Coast Range Premoli-Silva and her colleagues. After some debate, they voted
province rotated clockwise by about 50° (Simpson and Cox, 1977; to establish the Massignano section, near Ancona on the Adriatic
Hammond, 1979). By the later Eocene, the Laramide magmatic coast of Italy, as the type section for the Eocene/Oligocene bound-
null had ended, and arc volcanism returned all over the Cordillera ary. By vote, the "golden spike" that had been driven at the 19-meter
(Lipman, Prostka, and Christiansen, 1972; Snyder, Dickinson, and mark on the Massignano quarry face was recognized as "the bound-
Silberman, 1976), indicated by the eruption of the Clarno volcanics ary," and further disputes seemed to be over. Even where plank-
in central Oregon, the Challis volcanics in Idaho, and the Absaroka tonic foraminiferans were unavailable, the magnetic stratigraphy
volcanics in Wyoming and Montana. at Massignano and Gubbio showed that Hantkenina last appeared
High eustatic sea levels meant that much of the Atlantic and in the upper third of magnetic Chron C13r. Using this criterion,
Gulf Coastal Plains were inundated during most of the Eocene the Eocene/Oligocene boundary could be recognized in nonmarine
(Olsson et al., 1988; Galloway et al., 1991). Although most of rocks as well.
the Eocene units found in the Coastal Plain were deposited under Just as the issue seemed settled, problems emerged. In type sec-
shallow to deep marine conditions, a few were nearshore enough to tion of the uppermost Eocene Priabonian Stage near Priabona, north-
contain fossil mammals washed out to sea (summarized in Westgate, ern Italy, Brinkhuis (1992) reported that the upper part of the type
1986). Priabonian is early Oligocene, using the Hantkenina datum. The
latest research on the changes in oxygen isotopes and Antarctic ice
volume increases (Miller, 1992), global oceanic circulation (Mc-
OLIGOCENE Gowran et al., 1992), and land plant extinctions (Wolfe, 1992) shows
that the most natural event on which to draw the boundary is about
a million years later than the Hantkenina datum, in what the Paleo-
CHRONOSTRATIGRAPHY gene Subcomrnission designated "early Oligocene" or within Chron
The Oligocene was proposed by H. E. von Beyrich in 1854 for strata Cl3n (Berggren and Prothero, 1992). Future stratigraphic commis-
in Germany and Belgium that had previously been assigned to the sions may decide to change the boundary stratotype, but for now,
upper Eocene or lower Miocene. Von Beyrich 's list of "Oligocene" the boundary remains the Hantkenina datum in the upper third of
rocks was very heterogeneous, including units that are now clearly Chron Cl3r.
assignable to the Eocene or Miocene, and one unit (the bone sand of Most European mammalian paleontologists placed the Eocene/
Eppelsheim) which produced a late Miocene Hipparion fauna. An- Oligocene boundary at the dramatic change in mammals known as
other problem is that the "type" strata of von Beyrich's Oligocene the "Grande Coupure" (Stehlin, 1909; Brunet, 1977; Savage and
in Belgium and Germany do not overlie the "type" strata of the Russell, 1983; Brunet et al., 1987). The Grande Coupure marked
Eocene in the Paris Basin or Italy, so the Eocene/Oligocene bound- the extinction of many archaic, endemic Eocene mammalian groups
ary cannot be recognized in either area. As in the case of the rest of and their replacement by a number of Asian immigrants, especially
the European Cenozoic, the type sections of the stages within the rhinocerotoids (amynodonts, hyracodonts, and rhinocerotids), an-
Oligocene represent only a small portion of the total time (Hardenbol thracotheres, entelodonts, peccaries, several families of advanced
and Berggren, 1978). Although these problems made the Oligocene rodents (including beavers, pocket gophers, theridomyids and
hard to correlate, it has now been tied to the global marine plank- eomyids), hedgehogs, and a variety of camivorans. It was long
ton stratigraphy and magnetic polarity time scale (Berggren, 1971; assumed that this event correlated with the Hantkenina datum
Berggren, Kent, and Flynn, 1985; Aubry et al., 1988; Berggren et (Pomerol and Premoli-Silva, 1986). However, the most recent re-
al., 1995). search (Hooker, 1992) has shown that the Grande Coupure is earliest
The biggest dispute has been over recognition of the Eocene/ Oligocene (using the Hantkenina criterion), but probably coincides
Oligocene boundary and its chronostratigraphy. This argument has with the global events (circulation changes, Antarctic glaciation,
received much attention because its date is important to understand- oxygen isotope event, land floral crisis) found to occur in early
ing the major extinction events that occurred in the later Eocene Oligocene Chron Cl3n. This evidence might also weigh in favor of
(Berggren and Prothero, 1992; Prothero, 1994a). Because the changing the Eocene/Oligocene boundary definition.
 Background to North American mammalian evolution 17

The geochronology of this boundary has also been very con- the years, but now seems stabilized around 23.5 Ma (Berggren et
troversial. For decades, the dates produced by K-Ar methods on al., 1995). Recent 40Ar!3 9Ar dating now places the early Arikareean
volcanics and by various isotopes in marine glauconites produced Gering Fauna between 28 and 30 Ma (Tedford et al., 1996), and
highly contradictory estimates ranging from 32 to 38 Ma (Hardenbol most of the Monroe Creek Fauna also appears to be late Oligocene
and Berggren, 1978; Odin, 1978, 1982; Curry and Odin, 1982; (Tedford et al., 1987, 1996). The Harrison Fauna and the overlying
Glass and Crosbie, 1982; Berggren, Kent, and Flynn, 1985; Odin Agate Springs Quarry (in the Upper Harrison Formation of Peterson,
and Curry, 1985; Berggren, 1986; Glass, Hall, and York, 1986; 1909) may be earliest Miocene.
Aubry et al., 1988; Obradovich, 1988; summarized in Berggren
et al., 1992 and Prothero, 1994a). However, recent 40Ar/39Ar dat-
BIOSTRATIGRAPHY
ing of volcanic ashes in the deep marine sections in Gubbio and
Massignano, Italy, directly tied to marine microfossils and magnetic Unlike the detailed biostratigraphic "zonation" now in place for
stratigraphy (Montanari et al., 1985, 1988; Montanari, 1988, 1990; most of the Paleocene and Eocene (Gingerich, 1983; Archibald
Odin et al., 1988, 1991) and of terrestrial volcanic ashes in North et al., 1987; Krishtalka et al., 1987; Gunnell, 1989), a detailed
America (Swisher and Prothero, 1990; Prothero and Swisher, 1992) Oligocene biostratigraphy is still not fully published. For decades,
have resolved the controversy by eliminating many erroneous age it was impossible to subdivide the Orellan or Whitneyan land mam-
estimates. The emerging consensus places the Eocene/Oligocene mal "ages" because most of the early collections had very poor
boundary at about 33.5 Ma (Berggren et al., 1995). stratigraphic data. The stratigraphic scheme of Schultz and Stout
In addition to redating the Eocene/Oligocene boundary, new 40Ar/ (1955, 1961) was actually based on lithostratigraphic units and did
39
Ar methods provided radiometric dates on the rest of the North not have biostratigraphic resolution within the units. Outlines of
American Oligocene land mammal sequence for the first time the biostratigraphic potential of the Oligocene were presented by
(Figure 1.2). The Wood Committee named three land mammal Emry, Bjork, and Russell (1987), but no detailed zonation was pro-
"ages" that they thought were Oligocene: the Chadronian, the Orel- posed. Thanks to decades of work by Morris Skinner, Bob Emry,
lan (based on the Orella Member of the Brule Formation in Ne- and collectors of the Frick Laboratory, however, there are now large
braska), and the Whitneyan (based on the Whitney Member of the collections of White River mammals with stratigraphic data zoned
Brule Formation in Nebraska). They considered the Chadronian to the nearest foot from volcanic ashes. These collections allowed
Land Mammal age to be early Oligocene, the Orellan to be middle Prothero (1982) to propose a preliminary biostratigraphy that di-
Oligocene, and the Whitneyan to be late Oligocene. The Arika- vided the Orellan into four zones and recognized one zone for the
reean (based on the Arikaree Group in Nebraska) was thought to be early Whitneyan. Korth (1989) also proposed a biostratigraphic
early Miocene. As we have already seen, the Chadronian is now zonation for the Orellan, but it is based on University of Nebraska
latest Eocene. The Orellan/Whitneyan boundary occurs at about collections that do not have the resolution to subdivide the lithos-
32 Ma, within,Chron C12r, and the Whitneyan/Arikareean bound- tratigraphic units. Hence Korth's (1989) "fauna! zones" are really
ary within Chron Cl ln, about 30 Ma (Swisher and Prothero, 1990; based on lithostratigraphic boundaries.
Prothero and Swisher, 1992; Tedford et al., 1996). In this cali- A finely resolved biostratigraphy for the Orellan and Whitneyan
bration, the Orellan, Whitneyan, and earliest Arikareean are early has long been in preparation, but its publication has been delayed by
Oligocene, and most of the Arikareean is late Oligocene. (Because the lack of systematic revisions of many of the key taxa, especially
the European type Oligocene has only two stages, the Rupelian oreodonts, leptomerycids, and ischyromyid rodents. Such revisions
and Chattian, there is no formal "middle" Oligocene.) Magnetic are now published (various papers in Prothero and Emry, 1996a),
stratigraphy (Figure 1.2) has now been completed on virtually all and Prothero and Emry (1996b) suggested criteria for a zonation of
the important Orellan, Whitneyan, and early Arikareean outcrops the Chadronian through early Arikareean. Prothero and Whittlesey
from the White River and Arikaree Groups (Prothero, 1982, 1985a, (1998) proposed a formal biostratigraphic zonation for the Orellan
1985b, 1996c; Prothero, Denham, and Farmer, 1983; Evanoff, and Whitneyan in the White River Group.
Prothero, and Lander, 1992; Prothero and Swisher, 1992; Tabrum, Rensberger(1971, 1973, 1983;FisherandRensberger, 1972)pro-
Prothero, and Garcia, 1996), as well as important localities in Mon- posed a biostratigraphic zonation of rodents from the late Oligocene-
tana (Prothero, 1984; Tabrum, Prothero, and Garcia, 1996) and early Miocene John Day Formation of central Oregon. The mag-
California (Prothero, 1991; Prothero, Howard, and Dozier, 1996). netostratigraphy of these beds has also been published (Prothero
The Oligocene/Miocene boundary is less well constrained. When and Rensberger, 1985), although it will need revision with the new
the Wood Committee correlated the Arikareean with the early 40
Ar!3 9Ar dates recently analyzed by Carl Swisher. Unfortunately,
Miocene, their concept of the Arikareean was based on the Agate some of the key rodent taxa used in Rensberger's zonation do not
Springs fauna, which is latest Arikareean and does have taxa in occur in the classic areas of the Arikareean in the High Plains, so his
common with the early Miocene in Europe. However, all of the biostratigraphic zones have proven useful only in Oregon, Montana,
underlying units in the Arikaree Group have a very different fauna, and South Dakota (Tedford et al., 1987). When the magnetic stratig-
and the entire Arikareean appears to span the interval from 30 Ma raphy of the type Arikareean in Nebraska is published (Hunt and
to 21 Ma, almost 9 million years - by far the longest land mam- MacFadden, 1998), it will be possible to overcome these difficulties
mal "age" (Tedford et al., 1987, 1996). In addition, the definition and establish biostratigraphic zones for the Arikareean throughout
of the Oligocene/Miocene boundary in Europe has fluctuated over the western United States.
 Donald R. Prothero 18

megascopic plant fossils) are not well preserved, there is evidence of

PALEO CLIMATE
vegetational change. In the Big Badlands of South Dakota (which
Ever since the term "Terminal Eocene Event" (TEE) was proposed produces many of the critical mammal fossils), the only plant fossils
(Wolfe, 1978), scientists have focused on events at the end of the are hackberry seeds (which are virtually fossilized with calcite when
Eocene or at the boundary itself (e.g., Pomerol and Premoli-Silva, they are alive) and fossil wood of walnut trees, so the fossil flora
1986). Actually, recent evidence shows that the TEE was insignif- is too small for paleoclimatic reconstruction. However, Retallack
icant (Berggren and Prothero, 1992; Prothero, 1994a), and the im- (1983a, 1983b, 1990, 1992) suggested that paleosols preserved in
portant climatic shift occurred a million years later (about 33 Ma), the late Eocene Chadron Formation were formed under moist closed
in what is now defined as early Oligocene (see earlier discussion). forest canopies, with between 500 and 900 mm (20-35 inches) of
In oxygen isotope records, for example, both benthic and plank.tonic rainfall per year. In the overlying early Oligocene (Orellan) Scenic
foraminiferans show an increase of about 1.3 per mil in the earliest Member of the Brule Formation, paleosols indicated that rainfall
Oligocene (Miller, Fairbanks, and Mountain, 1987; Miller, 1992; was less than 500 mm (20 inches) per year, and the soil structure
Zachos, Breza, and Wise, 1992). Assuming some Oligocene ice, was typical of a more open, dry woodland.
Miller (1992) calculated that 0.3--0.4 per mil of the change is due In eastern Wyoming, Evanoff (in Evanoff, Prothero, and Lander,
to an ice volume increase, which lowered global sea level by 30 m 1992) found that moist Chadronian floodplain deposits abrupt-
(Haq, Hardenbol, and Vail, 1987). The remaining 0.9-1.0 per mil ly shifted to drier, windblown deposits in the early Orellan.
is explained by about a 5-6°C (9-11°F ) cooling, lowering global These same beds preserve an excellent record of climate-sensitive
mean temperature to about 5°C (41°F). It had been as high as l3°C land snails. According to Evanoff, Chadronian land snails are large-
(55°F) in the early Eocene, and 7°C (45°C) in the latest Eocene shelled taxa similar to snails now found in subtropical climates
(Miller, Fairbanks, and Mountain, 1987). This was the most dra- with seasonal precipitation, such as in the southern Rocky Moun-
matic temperature shift during the entire Cenozoic, until a similar tains and the central Mexican Plateau. Based on modem analogues,
increase in the middle Miocene signaled the onset of the modem these snails indicate a mean annual temperature of 16.5°C (62°F)
Antarctic ice cap. and a mean annual precipitation of about 450 mm (18 inches)
In 1987, Ocean Drilling Project (ODP) (successor to the Deep Sea during the late Chadronian in eastern Wyoming. By the Orel-
Drilling Project) Leg 113 drilled the East Antarctic margin, Maud lan, the large-shelled snails had been replaced by drought-tolerant
Rise, and the Weddell Sea in the South Atlantic sector of Antarctica small-shelled taxa indicative of a warm-temperate open woodland
and the Southern Ocean (Kennett and Barker, 1990). These drill habitat with a pronounced dry season. Such snail faunas are to-
cores produced evidence of ice-rafted detritus beginning in the mid- day found in regions like southern California and northern Baja
dle Oligocene. Dramatic confirmation of early Oligocene ice in California.
Antarctica came in 1988, when ODP Leg 119 drilled cores on the The amphibian and reptile fauna shows a similar trend toward
submerged Kerguelen Plateau and in Prydz Bay, both on the In- cooling and drying. According to Hutchison (1982, 1992), the aqua-
dian Ocean side of Antarctica (Barron et al., 1989). Thick glacial tic forms (especially salamanders, freshwater turtles, and crocodil-
deposits found in Prydz Bay and ice-rafted sediments on the south- ians) steadily declined in the late Eocene, and by the Oligocene
ern Kerguelen Plateau showed that a major ice sheet was grounded only terrestrial tortoises were common. This indicates a pronounced
over part of Antarctica by at least the earliest Oligocene. In ad- drying trend during the late Eocene. Crocodiles were gone by the
dition, glacial sediments were found in even older strata, which Chadronian, but more cool-tolerant alligators persisted until the
suggested a middle or late Eocene glaciation near Prydz Bay, al- early Orellan.
though the age of these deposits was not well constrained. The next Although it triggered massive vegetational changes and signif-
leg of the Ocean Drilling Project (Leg 120) drilled the central Ker- icant extinctions and migrations in much of the world, the early
guelen Plateau and provided further evidence of early Oligocene Oligocene glacial event was not very long lived. According to Miller,
glaciers (Breza et al., 1989; Schlich et al., 1989; Zachos, Breza, Wright, and Fairbanks (1991), the early Oligocene pulse of glacia-
and Wise, 1992). The CIROS-1 drill hole in the Ross Sea also pro- tion may have lasted only a million years, concentrated in west-
duced lowermost Oligocene glacial sediments (Barrett et al., 1989). ern Antarctica and the Indian Ocean sector of the Southern Ocean.
Although the size and duration of these early Oligocene glaciers is It apparently had limited effect in East Antarctica and the South
still controversial (Kennett and Barker, 1990), it is clear that there Atlantic sector of the Southern Ocean (Kennett and Barker, 1990).
were significant glaciers on some parts of the Antarctic continent However, there is evidence for a larger and more protracted glacial
about 33 million years ago and short-term glaciation events since event or events starting in the late early Oligocene (sometimes
the middle Eocene. called "middle" Oligocene, although the Oligocene has been for-
One of the most dramatic responses to the early Oligocene cool- mally divided only into early and late stages), in magnetic Chrons
18
ing was demonstrated by North American land plants (see Wing, C 11, about 30 million years ago. Benthic foraminiferan 8 0 values
this volume, Chapter 2). In the middle Eocene, continental climate again exceeded 1.6 per mil, which approximates the threshold of
exaggerated the global temperature change, producing much more renewed glaciation (Miller, Wright, and Fairbanks, 1991; Miller,
extreme chilling on land. The Oligocene deterioration, however, 1992). These glacial values of 8 18 0 persisted for about 4 million
was the most extreme paleobotanical change in all of the Ceno- years, inflicting Antarctica with the longest glaciation (or multiple
zoic. Even in regions where organic material (including pollen or episodes of glaciation) since the Triassic.
 Background to North American mammalian evolution 19

The effects of this great ice sheet are apparent all over the south Kennett et al., 1972, 1975; Kennett, 1977). When evidence for the
polar region. The CIROS-1 drill hole in the Ross Sea region cored beginning of circum-Antarctic circulation first emerged, attention
almost 200 m ( 650 feet) of mid-Oligocene glacial sediments (Barrett focused on the timing of the separation of Antarctica and Australia.
et al., 1989). Mid- or late Oligocene glacial sediments were also The two continents began to separate in the late Cretaceous and
reported on Marie Byrd Land (LeMasurier and Rex, 1982), 200 km Paleocene (Weisse!, Hayes, and Herron, 1977; Mutter et al., 1985;
away in the MSST-1 core in Victoria Land (Barrett et al., 1987), Veevers, 1986), with rifts opening from west to east (McGowran,
and in King George's Island on the Antarctic Peninsula near South 1973; Kennett et al., 1975). By the middle and early late Eocene
America (Birkenmajer, 1987). Even more striking is the evidence there was oceanic spreading between the continents. The microfos-
from voyages of the research vessel RIV Polar Duke in the Ross sils suggest that there was still only a shallow marine gulf between
Sea region. Seismic reflections off the sediments beneath the Ross the continents, apparently blocked from the South Pacific by the
ice cap revealed a major mid-Oligocene unconformity, overlain by South Tasman Rise (Kennett et al., 1975; Kennett, 1977, 1978,
hundreds of meters of late Oligocene glacial sediments (Bartek and 1980; Murphy and Kennett, 1986; Kamp, Waghom, and Nelson,
Anderson, 1990; Bartek et al., 1992). Bartek and colleagues were 1990). In 1973, Deep Sea Drilling Project Leg 29 drilled in the
able to trace this unconformity up onto the Antarctic continent, South Pacific and discovered the first evidence of shallow marine
and over at least 100,000 square kilometers beneath the Ross ice circulation over the Tasman Rise; it apparently occurred in the latest
shelf. Such a large feature could only have been produced by the Eocene (Kennett et al., 1975; Shackleton and Kennett, 1975). This
grounding of a large ice sheet, comparable to the glacial activity in may explain why Antarctica began to cool and deep bottom waters
the Pleistocene. were formed shortly thereafter in the earliest Oligocene.
The most striking global effect of this massive ice sheet was a Continued spreading between Antarctica and Australia enlarged
worldwide drop in sea level. As sea level retreated, the continental the seaway between them. Evidence of the sudden expansion of
shelf was exposed to erosion, forming a deeply incised unconfor- glacial ice and the widespread "middle" Oligocene unconformity
mity. A major mid-Oligocene unconformity is found in many ma- suggests that a new phase of circumpolar circulation had begun
rine sequences around the world, including the Gulf Coast (Keller, at the end of the early Oligocene. Based on the isotopic and mi-
1985a), the Atlantic Coast (Olsson, Miller, and Ungrady, 1980; Poag cropaleontological evidence, several authors (Kennett et al., 1975;
and Schlee, 1984; Poag and Ward, 1987), in Europe (Aubry, 1985), Kennett, 1977, 1978, 1980; Murphy and Kennett, 1986; Kamp,
in Australia (Quilty, 1977; Loutit and Kennett, 1981; McGowran, Waghom, and Nelson, 1990) have attributed this massive cooling
Moss, and Beecroft, 1992), and in southern Africa (Siesser and event to the beginning of deep-water circulation through the gap
Dingle, 1981). Deep-sea cores typically have a major gap repre- between the South Tasman Rise and Antarctica. This is particularly
senting the mid-Oligocene; some were so deeply eroded that the apparent in DSDP Site 277, which lies on the submarine Campbell
early Oligocene record was wiped out as well (Kennett et al., 1972; Plateau (due south of New Zealand); it is right in the path of currents
Keller et al., 1987). Seismic reflections from beneath the continental flowing through the gap. In the early Oligocene, Site 277 was still
shelf off New Jersey and Virginia revealed deeply incised subma- bathed in warm waters flowing down the coast of eastern Australia,
rine canyons, cut by rivers when the retreating ocean left the shelf but as time passed, the differences in isotopic values between shal-
exposed in the mid-Oligocene (Miller, Fairbanks, and Mountain, low and deep-dwelling foraminiferans increased. Just before the
1987). time of the major mid-Oligocene unconformity, there was a sharp
Even though this mid-Oligocene glaciation was much bigger increase in the difference in isotopic values between Site 277 and
than the early Oligocene ice buildup, it had much less of an ef- sites northwest of New Zealand (still bathed in warm currents). Ac-
fect on global climate. Wolfe (1978) indicated no significant mid- cording to Murphy and Kennett (1986) and Kamp, Waghom, and
Oligocene effect on North American land plants after the earliest Nelson (1990), this is evidence that a blast of deep cold water was
Oligocene refrigeration, possibly because late early Oligocene flo- passing south of Tasmania, separating the Antarctic circumpolar
ras were already cold adapted, and tropical taxa had been driven current from currents flowing further north. Once these currents
extinct or to lower latitudes. In the Badlands paleosols, Retallack were decoupled, the beginning of the modem Antarctic polar front
(1983a, 1983b, 1990, 1992) documented a transition from early was established, separating the circum-Antarctic current from the
Oligocene (Orellan) wooded grasslands and gallery woodland to late more northerly currents.
Oligocene (Whitneyan and early Arikareean) open grasslands with Given the scale of the mid-Oligocene glaciation event and sea-
trees only along watercourses. There was also a drying trend, from level drop, it seems that circum-Antarctic circulation should have
average annual rainfall of 500-900 mm (20-35 inches) in the early also occurred between Antarctica and South America through the
Oligocene to 350-450 mm (14-18 inches) in the late Oligocene. Drake Passage. Studies in this region (Barker and Burrell, 1977,
Whitneyan and early Arikareean deposits in the Great Plains were 1982; Sclater et al., 1986) suggest that the Drake Passage did not
predominantly windblown silts and volcanic dust, indicating very open until the late Oligocene. However, the timing on this is not
dry conditions. as well constrained in terms of biostratigraphy or magnetics. Thus
What caused these episodes of glaciation.in the Oligocene? For the opening of the Drake Passage might explain either the long
three decades, oceanic drilling and isotopic studies have suggested duration of the mid-Oligocene cooling event (possibly event Oi2a
that the isolation of the Southern Ocean and the development of of Miller, Wright, and Fairbanks, 1991) or the renewed cooling
circum-Antarctic circulation was the critical factor (Devereux, 1967; and glaciation at the Oligocene/Miocene boundary (event Mil of
 Donald R. Prothero 20

Miller, Wright, and Fairbanks, 1991). If Barker and Burrell (1982) especially with silicic volcanic ash from the explosive arc volcanoes
are correct, circulation through the Drake Passage did not really to the west. Eventually the Laramide basins had filled to the point
commence until the Oligocene/Miocene boundary. where only a low, rolling topography with isolated remnant uplifts
Although the Southern Ocean seems to have been the major "cold was left of the mighty Lararnide orogeny (Lille graven and Ostresh,
spigot" providing deep, cold bottom waters, it was not the only 1988). This is shown by the development of drainages across the
source. A major early Oligocene unconformity in the North Atlantic ancient ranges (Clark, 1975; Seeland, 1985), an Eocene-Oligocene
produces reflection horizons in the seismic records of the Atlantic surface incised into the tops of many of the Rocky Mountain ranges
margin (Miller and Tucholke, 1983; Mountain and Tucholke, 1985). (Epis and Chapin, 1975; Scott, 1975; Bradley, 1987), and isolated
In addition, studies of benthic foraminiferans showed that in the sedimentary remnants of Oligocene and early Miocene age perched
early Oligocene, the North Atlantic bottom waters were richer in on the top of many ranges throughout the Cordillera, such as the
13 C than those of the Pacific (Miller and Fairbanks, 1983; Miller Gravelly Range in Montana and the Absarokas and Bighorns in
and Tucholke, 1983; Miller and Thomas, 1985; Miller, 1992). This Wyoming (McKenna and Love, 1972; McKenna, 1980). However,
suggests that the source of some of the deep waters of the Atlantic Evanoff (1990) has shown that the relief on this surface was not
must have come from the north. The Arctic Ocean had been isolated as gentle as once thought. Some paleovalleys filled with White
from the rest of the world's oceans since the Mesozoic, developing River sediments have Oligocene relief on the order of 500-1,000
its own cold waters with high 13 C ratios. When the Arctic was recon- m, and many of the supposed "Eocene" subsummit surfaces are
nected with the North Atlantic, such cold waters would be analogous actually Miocene in age. Thus the beveling did not develop to its
to the modem North Atlantic Deep Water (NADW). The best can- full extent in all parts of the Rockies at the end of the Eocene, but
didate for this marine passageway was the Norwegian-Greenland was completely developed by the Miocene.
Sea, which apparently opened in the early Oligocene (Talwani and The burial of these ranges and the filling of their surrounding
Eldholm, 1977; Berggren, 1982). Other possibilities include the basins meant that the bulk of sedimentation in the late Eocene and
Faeroe-Shetland channel, or possibly through the Denmark Straits Oligocene shifted to the High Plains further east. The best record
(Miller and Curry, 1982). of this is the White River Group, famous for the spectacular spires
There is no shortage of potential oceanographic sources for the and pinnacles of Badlands National Park in South Dakota (Emry,
cooling and climatic changes in the Oligocene. The well-established Bjork, and Russell, 1987). White River sediments were deposited
shallow-water circum-Antarctic circulation, plus the cold Arctic in forested floodplains in the Chadronian and Orellan (Retallack,
source in the North Atlantic, probably triggered the earliest Oligo- 1983a, 1983b), with the volcaniclastic component increasing toward
cene cooling event that produced the first significant Antarctic the sources in the west (Emry, Bjork, and Russell, 1987). By the
glaciers. The passage of deep water south of Tasmania in the mid- Whitneyan and Arikareean, eolian volcaniclastics dominated the
dle Oligocene further accentuated the cold trend, producing a major section, indicating a semiarid landscape with abundant dunes of
Antarctic ice cap and the largest sea-level drop in the last 100 million volcanic ash and streams with flash flooding (Swinehart et al., 1985;
years. The completion of circum-Antarctic circulation through the Emry, Bjork, and Russell, 1987).
Drake Passage completed the refrigeration, whether it happened in While the Rocky Mountain region quieted down, the West Coast
the late Oligocene or at the Oligocene-Miocene boundary. became more tectonically active in the Oligocene. Sometime around
30 Ma, the spreading ridge and associated transforms between the
Farallon and Pacific Plates collided with the subducting margin of
PALEOGEOGRAPHY North America (Atwater, 1970). When this occurred, the mar-
The Laramide Orogeny, which had dominated western North Amer- gin switched from conventional subduction of the Farallon Plate
ica through most of the Paleocene and Eocene, ceased by the late to a transform margin. This occurred because the motion of the
Eocene. In its place was a return of normal subduction and arc Pacific Plate relative to the North American Plate was sliding along
volcanism (Snyder, Dickinson, and Silberman, 1976; Dickinson, a northwest-trending transform fault, rather than colliding in a sub-
1979), especially in the Buck Hill-Vieja volcanics of Trans-Pecos duction zone. This transform, the San Andreas Fault, dramatically
Texas, the Mogollon-Datil volcanic field of New Mexico and changed the tectonics of the California coast. Exotic terranes were
Arizona, the San Juan volcanics of southwest Colorado, numer- transported north from Mexico and southern California at a rapid
ous volcanic fields in Nevada and Utah, and the ancestral Cascades rate. Numerous pull-apart basins were developed in the Oligocene
of central Oregon and Washington (Armstrong et al., 1969; Lipman, and Miocene between sliding fault blocks, and crustal blocks in the
Prostka, and Christiansen, 1972; Snyder, Dickinson, and Silberman, Transverse Ranges rotated in a clockwise direction by about 90°
1976). Some of these episodes of explosive volcanism were truly (Luyendyk, Kamerling, and Terres, 1980; Luyendyk, 1991). In the
immense, dwarfing events such as the 1980 eruption of Mt. St. Pacific Northwest, conventional subduction continued to generate
Helens. For example, the Needles Range volcanics of eastern forearc basins filled with marine sediments.
Nevada and Utah consist of five flows traceable over 35,000 square Although basins were subsiding rapidly all along the Pacific
kilometers, and in southern Nevada, there are individual ash flows Coast, low eustatic sea levels (especially the 250 m drop in the
containing over 500 cubic miles of volcanic debris (Cook, 1965)! mid-Oligocene - Haq, Hardenbol, and Vail, 1987) meant that most
With the end of Laramide uplift, the Rocky Mountains began to of the region was exposed to erosion during the Oligocene. In a
erode away and their surrounding basins began to fill with debris, few places, basins accumulated nonmarine deposits bearing fossil
 Background to North American mammalian evolution 21

mammals. For example, in southern California, the late Eocene and Because Miocene marine and nonmarine beds frequently interfin-
Oligocene is represented by sandy floodplain-alluvial fan deposits ger in Europe, it is possible to establish correlation of European
of the Sespe Formation (Dibblee, 1966; Taylor, 1983; Kelly, 1990). mammal zones with the marine stratotypes (Fahlbusch, 1976; Mein,
The Sespe produces Uintan-Duchesnean mammals and, after an 1979; Savage and Russell, 1983). However, such a correlation has
8-million-year hiatus possibly related to the mid-Oligocene sea- been more difficult with the North American mammalian record.
level drop, early Arikareean assemblages (Prothero, Howard, and In parts of the coastal regions of New Jersey, Florida, Texas, and
Dozier, 1996). Similar Oligocene continental deposits are known California, marine beds with planktonic foraminiferans, calcareous
elsewhere along the California coast (Bartow, 1978; Nilsen, 1987), nannofossils, or diatoms interfinger with mammal-bearing beds, al-
although they rarely produce identifiable fossil mammals. lowing direct correlation with the global time scale (Tedford and
In addition to fault-bounded basins along the coast, the abundance Hunter, 1984; Tedford et al., 1987). But most of the classic Miocene
of arc volcanics also preserved a record of the Oligocene landscape faunas come from the western interior, with no possibility of marine
in the interior. The widespread, nearly uniform blankets of ash all interfingering. In these regions, land mammal biostratigraphy is
over eastern Oregon, Nevada, and western Utah suggest there was fundamental, greatly aided by radiometric dating and magnetic
very little relief between the Rockies and California. In Nevada, vol- stratigraphy (Tedford et al., 1987).
canic ash frequently filled major lakes that are now preserved by the When the Wood Committee subdivided their North American
Elko and Sheep Pass Formations (Nilsen and McKee, 1979). Cen- "Miocene" in 1941, they recognized only three land mammal
tral Oregon was blanketed by hundreds of meters of volcaniclastics "ages" - the Arikareean, Hemingfordian, and Barstovian (thought
of the late Oligocene-early Miocene John Day Formation (Fisher to be early, middle, and late Miocene, respectively) (Figure 1.3).
and Rensberger, 1972). The landscape was apparently forested and The Hemingfordian was named after the Hemingford Group in
semiarid but not as humid or densely forested as the late Eocene western Nebraska, and the Barstovian received its name from the
environments represented by the Clarno Formation. Barstow Formation in the Mojave Desert of southeastern Califor-
As in the earlier Cenozoic, the Midwest and Appalachian re- nia. Tedford et al. (1987) reviewed the status of the North American
gion continued to erode during the Oligocene, and we have little or Miocene and pointed out that most of the Arikareean is actually late
no record for that epoch. However, low eustatic sea levels (Haq, Oligocene (discussed earlier). The Hemingfordian is early, not mid-
Hardenbol, and Vail, 1987) mean the Atlantic and Gulf Coast were dle Miocene, and the Barstovian is middle, not late Miocene. More
also emergent and eroding during much of the Oligocene as well importantly, revisions of the Miocene/Pliocene boundary in Europe
(Olsson, Miller, and Ungrady, 1980; Galloway et al., 1991). In most (Berggren and Van Couvering, 1974) produced ages that were much
places, there is a widespread Oligocene unconformity; in others, younger than the estimates of the Wood Committee in 1941. This
there are thin nearshore marine units and fiuvial-deltaic units, such pushed the Clarendonian and Hemphillian Land Mammal ages (both
as the Catahoula Formation of Texas. These deposits show that the named after towns in the Texas Panhandle region, and considered
Gulf Coast was' mostly a fiat, swampy floodplain in the Oligocene, early and middle Pliocene by the Wood Committee) into the late
with major rivers and deltas building out into the Gulf (Galloway, Miocene.
Hobday, and Magara, 1982; Galloway et al., 1991). Tedford et al. (1987) reviewed the large number of K-Ar dates on
North American Miocene rocks. Some of these age estimates will
have to be revised when more 40Ar/39Ar dates become available. Un-
MIOCENE fortunately, many of these classic Miocene localities have limited
stratigraphic continuity or fauna! superposition, so the Miocene has
not been as amenable to magnetic stratigraphy as have the Paleogene
CHRONOSTRATIGRAPHY
rocks just discussed (Figure 1.3). The late Arikareean has been
Like the Eocene, the Miocene was another of the three original magnetically sampled (Prothero and Rensberger, 1985; Hunt and
epochs in Lyell's (1831-1833) subdivision of the Cenozoic. It also MacFadden, 1998), but most of the classic early and middle Hem-
had a controversial history because it was based on the "clock" ingfordian localities have not. Tedford et al. (1987) interpolated
model of molluscan turnover (Rudwick, 1978) and included a con- the age of the Arikareean/Hemingfordian boundary at about 20 Ma.
fusing combination of "typical" rock units over a wide area of Eu- On the other hand, the Hemingfordian/Barstovian boundary has
rope (Berggren, 1971; Berggren and Van Couvering, 1974; Savage been precisely dated at 15.9 Ma in Chron C5Br in the Barstow area
and Russell, 1983). The limits of the Miocene have been even of California (MacFadden et al., 1990; Woodbume, Tedford, and
more debated (Berggren, 1971; Berggren and Van Couvering, 197 4; Swisher, 1990). The Barstovian/Clarendonian boundary has been
Berggren, Kent, and Van Couvering, 1985; Berggren et al., 1995), dated at about 12.5 Ma in Chron C5Ar in the Dove Spring Formation
but the present consensus places the base of the Miocene at the base (Ricardo Group) in California (Burbank and Whistler, 1987; Loomis
of the Aquitanian Stage in Europe, now dated around and Burbank, 1988; Whistler, 1991; Whistler and Burbank, 1992).
23.5 Ma. Six marine stages- the Aquitanian, Burdigalian, Langhian, Barghoom ( 1981) reported the magnetic stratigraphy for Barstovian
Serravillian, Tortonian, and Messinian - are .recognized in Europe, and Clarendonian faunas in the Tesuque Formation of the Santa Fe
and their chronostratigraphy and magnetic stratigraphy have been Group of New Mexico, which corroborated these correlations.
much more stable than the Paleogene chronostratigraphy reviewed The Clarendonian/Hemphillian boundary was dated in the Dove
earlier (Berggren, Kent, and Van Couvering, 1985). Spring Formation (Ricardo Group) of California at 8.9 Ma, early in
 Donald R. Prothero 22

-
00 0 - 00 0

i._:t_...L_..r_-1-...r.E_ar~l-y~M::.._ioLc-eLne......l--if-..1:-:-~::-l--L-Lr-L~-:-;-!'--...._...l..-r-;P~l~io~c~en~e::l'"~PI~ei:-:1st EPOCH
Q Q CHRON
tn t:l
POLARITY
NALMA

JOHN DAY FM., OR

BARSTOW FM., CA

fi!llffi§l SANTAFEGROU~NM

1llM¥fHHM:t!!M¥Hl!MlimH!M RICARDO GROUP, CA

VERDE FM., AZ

ANZA-BORREGO, CA

et al. ~.1995!,·. Temporal

Figure 1.3. Miocene-Holocene time scale for North America. Global time scale and magnetic pol~ty ~ime scale.after Berggren
phically calibrated sectio~s. are sho.wn; sources are given 1~ text:, M10cene l~nd-mammal ages m order as
span of some of the important magnetostratigra
and Rancholabrean.
follows: Arikareean, Hemingfordian, Barstovian, Clarendonian, Hemphilhan. Pleistocene land-mammal ages are the Irvmgtoman

Chron 10 (Whistler and Burbank, 1992). Magnetic stratigraphies other Clarendonian localities in California (Savage, 1955). How-
have been published for the late Hemphillian Chamita Formation ever, many critical Miocene mammal localities in North America
of the Santa Fe Group in New Mexico (MacFadden, 1977) and the must still be correlated by biochronology because they often occurin
Verde Formation of Arizona (Bressler and Butler, 1978). Based on thin, isolated river channel deposits or quarries with no stratigraphic
the biostratigraphy, radiometric dating, and magnetic polarity of a continuity or superposition (Tedford et al., 1987). Nevertheless, de-
large number of late Hemphillian and early Blancan localities all tailed biochronology of the North American Miocene, emphasizing
over the western United States, the Hemphillian/Blancan boundary the first and last appearance of key mammalian taxa, yielded a series
is estimated at about 4.5 Ma, midway through the Gilbert magnetic ofbiochrons typically 1-2 million years in duration (Tedford et al.,
chron (Lindsay, Johnson, and Opdyke, 1975; Lindsay, Opdyke, and 1987). This is comparable to the resolution produced by range-zone
Johnson, 1984). Because the Miocene/Pliocene boundary is placed biostratigraphy in the earlier Cenozoic (as discussed earlier). For
at the base of the Gilbert chron (about 5 .3 Ma), the latest Hemphillian consistency, however, it would be advantageous eventually to estab-
and all of the Blancan are Pliocene (Lindsay, Opdyke, and Johnson, lish a true biostratigraphic zonation for the entire Miocene, as has
1984; Tedford et al., 1987). already been achieved for most of the Paleogene.

BIOSTRATIGRAPHY PALEO CLIMATE

Range-zone biostratigraphies have been proposed for the late Arika- After the Oligocene refrigeration, oxygen isotopic evidence in-
reean and early Hemingfordian of Oregon (Fisher and dicates that climates warmed and recovered slightly in the early
Rensberger, 1972; Rensberger, 1971, 1973 ), the late Hemingfordian Miocene (Woodruff, Savin, and Douglas, 1981; Miller, Fairbanks,
and "type" Barstovian in California (Lindsay, 1972; Woodburne, and Mountain, 1987; Miller, Wright, and Fairbanks, 1991), with
Tedford, and Swisher, 1990), the late Barstovian-Clarendonian- the peak of warming occurring about 17 Ma (Woodruff, Savin,
early Hemphillian sequence in the Dove Spring Formation of the and Douglas, 1981). During most of the early and early middle
Ricardo Group in California (Whistler and Burbank, 1992), and Miocene (from about 22 to 15 Ma), Antarctica experienced only
 Background to North American mammalian evolution 23

short episodes of glaciation (Miller, Wright, and Fairbanks, 1991). apparently for grazing. Solounias and Quade (1993) argue that the
Oceanic circulation was relatively sluggish in many parts of the traditional scenario of early middle Miocene savannas is a myth.
world, with distinct provincialism in foraminiferal faunas and rel- Most middle Miocene mammals with hypsodont teeth apparently
atively little depth stratification (Keller and Barron, 1983; Keller, grazed within a mixed forest-C3 grass habitat, and this is concordant
1985b). with the palynological evidence of mixed habitats in the Miocene
In the middle Miocene (about 15 Ma), major ice sheets returned (Leopold and Denton, 1987).
permanently to the Antarctic, and the modern episode of polar However, evidence from paleosols (Retallack, 1983a, 1990) and
glaciation began. Oceanic circulation became much more vigor- pollen (Leopold, Liu, and Clay-Poole, 1992) indicate that some
ous, and depth-stratified foraminiferans indicate the presence of kind of low-biomass vegetation formed open habitats in the late
shallow, intermediate, and deep waters in the tropics (Kennett et al., Oligocene and early Miocene, even if C4 grasses were not yet the
1975; Keller and Barron, 1983; Keller, 1985b). Although the trig- dominant vegetation. A seasonally dry scrubby community, similar
ger for this pulse of glaciation is uncertain, the connection of the to the saltbush communities of the modern Great Basin, has been
Arctic Ocean to the Atlantic over the sunken Iceland Ridge might suggested (Retallack, 1990). Thomasson (1985) argues that the mid-
have allowed the Arctic to become a heat sink for the first time Miocene silicified grass record is too diverse, and must have been
(Tucholke et al., 1979). Its most striking effect is the increase preceded by an Oligocene radiation of unsilicified, unfossilizable
in North Atlantic Deep Water, which pushed warm saline tropical grasses. According to this scenario, silicification then evolved in
Atlantic waters south of the Antarctic Convergence and supplied the the mid-Miocene in response to heavy grazing by herbivorous mam-
moisture to build the Antarctic ice sheet (Schnitker, 1980; Miller mals. By the late Miocene (7-8 Ma), however, all sources agree that
and Fairbanks, 1985). Also, the closure of equatorial circulation true C4 grasslands were found throughout many of the continents.
through the Indonesian Archipelago had a profound effect on Pa- Contrary to general misconceptions, North America was not cov-
cific and Antarctic circulation, and possibly on global temperature ered by a uniform grassland during the middle and late Miocene. In-
balance at this time as well (Kennett et al., 1975). stead, there was great biogeographic diversity, with habitats ranging
Through the middle and late Miocene (15 to 6 Ma), oxygen iso- from High Plains grasslands to a warm swampy Gulf Coastal Plain to
topes indicate many short pulses of cooling and glaciation in the cool temperate mesophytic forests in the Pacific Northwest to fairly
Southern Hemisphere (Woodruff, Savin, and Douglas, 1981). Fi- arid conditions in parts of California and Nevada (Stenzel, Turner,
nally, the Miocene was terminated by a major glacial advance and and Hesse, 1944; Axelrod, 1958; Webb, 1977; Tedford et al., 1987).
global cooling event that helped cut off flow into the Mediterranean
through Gibraltar. This resulted in the great Messinian salinity crisis,
PALEOGEO GRAPHY
with the entire Mediterranean Basin drying up and refilling several
times between 6 and 5 Ma (Adams et al., 1977; Hsii et al., 1977). Relative to the simple history of North America in the Paleogene,
These changes were reflected in the climate of North America. the story from the Miocene onward is very complex (Hamilton and
Marine Miocene records from the Pacific Coast show a general Myers, 1966; Atwater, 1970, 1989; Cole and Armentrout, 1979;
cooling trend after a peak of warmth in the early middle Miocene, Dickinson, 1979; Dickinson and Snyder, 1979; Christiansen and
with several cycles of warming and cooling throughout the Miocene Yeats, 1989; Oldow et al., 1989). Intricate plate tectonic interac-
(Addicott, 1970; Barron, 1973). In land floras (Wing, this volume, tions along the Cordilleran margin caused the whole region to un-
Chapter 2), early Miocene warming was followed by a mid-Miocene dergo major changes in paleogeography, leading to the complicated
(about 13 Ma) cooling trend that corresponds to the growth of the regional geology still found in the West. Before we review these
Antarctic ice sheet. Temperatures in North America warmed again complexities, however, we summarize the relatively "quiet" areas.
about 8 Ma, in the early late Miocene, followed by a final cooling As we have seen throughout the Cenozoic, the Atlantic and Gulf
at the end of the Miocene that corresponds to the great Messinian margins continued to sink passively as the Atlantic widened. Un-
glaciation. like the lowstands of the Oligocene, however, the Miocene was
Along with general cooling in the Miocene was a pronounced characterized by higher sea levels (Haq, Hardenbol, and Vail, 1987),
drying trend, reflected in both flora and fauna. Although woodland so that the coastal plains were repeatedly inundated, and shallow ma-
savannas persisted well into the Miocene, the traditional interpre- rine sequences (such as those exposed in the famous Calvert Cliffs
tation is that the Miocene was characterized by a great increase in of Chesapeake Bay in Maryland) were deposited over a wide area.
grasslands (Webb, 1977; Singh, 1988). The analogy with typical Due to this complex transgressive-regressive history, fossil mam-
savanna grasslands, however, may not be completely apt because mals are known from a variety of nonmarine and nearshore units
the isotopic signature of abundant C4 grasses (warm-season grasses in the coastal plains (Tedford and Hunter, 1984 ), which indicate
typical of plains and savannas) in the soil carbonate does not begin that the coastal plain was again low and forested throughout the
until the late Miocene (around 7-8 Ma) in most parts of the world Miocene. This is particularly apparent in the Gulf Coastal Plain in
(Quade, Ceding, and Bowman, 1989; Ceding, 1992; Ceding, Wang, the early and middle Miocene, which hosted a variety of endemic
and Quade, 1993; Wang, Ceding, and MacFadden, 1994; Quade forest dwellers not found on the High Plains (Patton, 1969; Tedford
and Ceding, 1995). Yet a number of middle Miocene ungulates et al., 1987; Prothero and Sereno, 1982).
(especially horses - MacFadden, 1992, p. 240; Wang, Ceding, The Great Plains and Rocky Mountains produce the major part of
and MacFadden, 1994) had already developed high-crowned teeth, our North American Miocene record, yet the region was relatively
 Donald R. Prothero 24

inactive. The Laramide orogeny, which had shut down in the late Idaho. Today, a hot spot apparently lies under Yellowstone National
Eocene, continued to be eroded away and buried in its own debris. Park in northwestern Wyoming. If these eruptions are all related,
By the late Miocene, most of the Rocky Mountains were buried to the then they probably reflect the eastward apparent migration of the
top, producing many of the flattened summit surfaces discussed pre- hot spot as the North American plate rode westward over it (Suppe,
viously and the erosional remnants left at their crests (Scott, 1975). Powell, and Berry, 1975).
With the filling of the basins and loss of relief, the sediment supply Through the late Miocene, the Basin and Range stretching and the
to the High Plains also began to decrease. Instead of the nearly con- shutoff of the Sierra Nevada volcanic arc continued as the bound-
tinuous White River and Arikaree "blanket," which covered most of ary between the subduction zone and the San Andreas transform
the northern High Plains during the Oligocene, Miocene sedimenta- (known as the Mendocino triple junction) moved north. By the end
tion was much more local and episodic. In Nebraska, for example, of the Miocene (5 million years ago), the San Andreas reached
much of the Miocene is represented by a series of stacked paleo- north of San Francisco. Basin and Range extension had ripped
valley sequences, indicating decreased sediment supply and more Nevada apart, breaking it into its modem geometry of north-south-
rapid uplift (Stanley, 1976; Skinner, Skinner, and Gooris, 1977; trending fault-block mountain ranges. The stretching of Nevada
Swinehart et al., 1985). The abundant volcanic debris and eolian meant that the Sierras and Cascades had to move west to accom-
deposition that were so important during the Oligocene had virtually modate the expansion. Indeed, paleomagnetic studies ofrocks from
vanished. both the Sierras and Cascades show that they have moved west-
The most dramatic changes occurred in the Great Basin and Pa- ward by as much as 270 km (170 miles) since the middle Miocene
cific Coast. A variety of events happened almost simultaneously, (Magill and Cox, 1980). They swung like a door, with the hinge
and most are probably interrelated. One of the most important events located in the Olympic Peninsula of Washington; the Cascades
was the opening of the Basin and Range province to form a series rotated slightly clockwise around it, and the southern Sierras ro-
of north-south-trending horsts and grabens (Nolan, 1943; Stewart, tated the most. It is analogous to opening a paper fan, with the
1971; Proffett, 1977; Eaton, 1980, 1982). On a physiographic map, opening frame represented by the Sierra-Cascade segment and the
they form a long series of parallel ranges that C. P. Dutton compared stretched paper folds symbolizing the stretched crust of the Basin
to "an army of caterpillars marching north." Most of the ranges and Range in Nevada and Arizona. The earliest and most rapid
reached 2,000-3,000 m (6,000-9,000 feet) in elevation, and the stretching occurs near the tip of the fan (Arizona), followed by
basins are filled with as much as 3 km (about 2 miles) of sediments. stretching in the heart of the fan (Nevada), and areas near the
Today, the Basin and Range province extends through Nevada all hinge (in the Olympic Peninsula) stretched the latest and the least.
the way to southern Oregon. The crust of Nevada is now stretched to Recent estimates of extension of about 200 percent in northern
twice its Miocene width, resulting in thin crust only 20-30 km deep Nevada (Proffett, 1977) and up to 500 percent (170-250 km of
(as opposed to 50-120 km in most regions). This is some of the stretching) across the Las Vegas portion of the Basin and Range
thinnest crust on any continent, and the nearby mantle supplies heat are in good agreement with this model (Wernicke, Axen, and Snow,
and magma to the fault zones all through the Basin and Range re- 1988).
gion (Eaton, 1980, 1982). In the middle Miocene, Basin and Range By the late Miocene, the great Columbia River basaltic eruptions
extension had just begun in the south, and a much narrower Nevada had long cooled, but the same mantle hot spot was under southern
was intruded by arc volcanoes. Idaho, erupting the Snake River basalt floodplain. The eastern edge
Another remarkable phenomenon of the middle Miocene were of the Colorado Plateau in New Mexico was marked by another
flood basalt eruptions of eastern Washington and Oregon. Known great rift valley that began opening in the middle Oligocene (Chapin,
as the Columbia River basalts, these eruptions spewed out of crustal 1979). It is called the Rio Grande Rift because the Rio Grande River
fissures from deep sources in the mantle (Waters, 1961). Like the now flows through it. Miocene and Pliocene sediments accumulated
Deccan traps and many of the other great flood basalt eruptions, they and then broke into normal fault blocks, very similar to the Great
produced enormous volumes of very fluid basaltic lava in a short Rift Valley of East Africa or the Triassic rift valleys formed when
period of time. Some flows covered over 40,000 square kilometers the Atlantic first opened. However, it is not clear whether the Rio
in a matter of days, moving at about 5 km/hour in flows 30 m deep Grande Rift is formed by a continent splitting apart or is part of
and IOO km wide at temperatures of I I00°C! In about 3.5 million more complex tectonics.
years, flow after flow erupted from these fissures until they covered The northward expansion of the San Andreas transform meant
300,000 square kilometers, filling former valleys with as much as that many pieces of Mexico slid northward into California (Crowell,
4,000 m of stacked lava flows. Between eruptions, the flows cooled 1974). The Salinian Block had almost reached the central Coast
and then weathered so deeply that forests grew on them. In central Ranges, where it now resides. Other blocks moved northward as
Washington, the famous Gingko Petrified Forest preserves hundreds well. In some places, crustal blocks could not slide smoothly north-
of fossilized trees from these ancient forests. ward along the faults parallel to the San Andreas. Instead, they got
The Columbia River eruptions cannot be explained by simple caught between fault zones and were sheared into a clockwise rota-
interactions of the plates, but were probably the result of a giant tion. This is typical of the Transverse Ranges north of Los Angeles.
"hot spot" in the mantle, similar to those that produced Hawaii and As their name implies, the Transverse Ranges lie in an east"west
the Deccan eruptions. In the late Miocene and Pliocene, similar trend that is almost perpendicular to the northwest-southeast trend
eruptions occurred farther east in the Snake River plain of southern of the San Andreas and the coastal ranges along it. Paleomagnetic
 Background to North American mammalian evolution 25

data shows that the Transverse Ranges have been rotated clock- spreading. Because the Pacific Plate is spreading northwestward
wise about 90° since the middle Miocene (Luyendyk, Kamerling, away from it, there is no problem with the East Pacific rise ending
and Terres, 1980; Luyendyk, 1991). This movement can only be abruptly in a transform fault. Indeed, that geometry is characteristic
explained if these crustal blocks had pivoted like ball bearings be- of nearly every transform fault along the midocean ridges around
tween two sliding surfaces. the world. Some (e.g., Davis, 1980) have postulated that the Basin
As rotation proceeded, deep structural chasms opened up be- and Range is produced by back-arc spreading, as is now happening
tween the pivoting crustal blocks (Crowell, 1974). These became west of Japan. However, the Basin and Range is no longer behind an
very narrow, steep-walled, rapidly subsiding sedimentary basins. arc, but behind a transform margin, which is no longer subducting.
They were so deep that they rapidly filled with thousands of me- Still others have suggested that the Pacific Plate is being subducted
ters of marine shales and turbidites, accumulating thicknesses of as and melting beneath Nevada, but this does not produce Basin and
much as IO km (30,000 feet) of sedimentary fill in the Los Ange- Range extension.
les Basin in less than 5 million years! If you stand in downtown Examining the plate geometries of the last 20 million years, it
Los Angeles, the pre-Miocene bedrock lies more than 6 miles be- is striking that the beginning of the San Andreas transform means
neath your feet, yet all the sedimentary fill is less than 8 million that no plate is going down in that region (Dickinson and Snyder,
years old. The Ventura Basin to the north is filled with almost 6 km 1979; Severinghaus and Atwater, 1990). There are remnants of the
(over 20,000 feet) of Miocene and Pliocene sediments, and there Farallon Plate sinking underneath the Pacific Northwest and under
were many smaller basins with similar histories. The Ridge Basin, Mexico, but the Pacific Plate is not subducting beneath the North
to the north of Los Angeles, contains 13.5 km (44,000 feet, or over American Plate. Instead, the two plates scrape sideways against
8 miles) of fill, yet is only I0-15 km wide! These basins are re- each other, shearing off small fault blocks. Consequently, the North
markable, not only because they are so narrow and yet very deep, American Plate lies directly on the mantle in this region, whereas
but also because they were deformed as they grew, so they have in the subduction zones to the north and south, Farallon Plate lies
steep folds and faults throughout them. Much of the oil wealth of between the deeper mantle and the North American Plate. In other
southern California comes from these Mio-Pliocene basins. Oil mi- words, the region east of the San Andreas transform is a "slab gap"
grated out of the organic-rich Miocene shales to the faults and folds in the Farallon Plate.
along the basin edges, where it accumulated and formed great oil The power of the slab-gap hypothesis is that it predicts so much
fields. ofCordilleran geology. Using known Pacific seafloor spreading ge-
An even more remarkable phenomenon was the "great rip-off" ometries, and assuming a 30° dip of the Farallon Plate, Dickinson
of Baja California. The East Pacific Rise makes a triple junction and Snyder (1979) projected the position of the gap in the Faral-
with the southern San Andreas transform and the Central American lon Plate through the last 20 million years. As the gap expands, it
trench. During the Miocene, this triple junction gradually moved successfully explains multiple events. For example, the Basin and
south along the· southern California coast as the San Andreas ex- Range opened progressively from south to north because the expan-
panded along the former subduction zone (Atwater, 1970, 1989). sion of the gap would generate a northward exposure of the mantle.
About 5 million years ago, the triple junction reached the southern As the northern edge of the gap moves north, the Cascade arc is
tip of Baja California, and the entire peninsula tore away from main- shut off in the south. The current northern edge of the gap would
land Mexico (Crouch, 1979). The East Pacific Rise began to spread lie east of the Mendocino triple junction across the northern part of
open the Gulf of California, placing Baja on the Pacific Plate. Baja California, just south of the southernmost active Cascade volcanoes,
California continues to move north today, bringing parts of Mexico Mt. Lassen and Mt. Shasta. The expansion of the slab gap under
with it. the Colorado Plateau occurred between IO and 5 million years ago,
The unusual plate geometry of a subducting margin turning into when its uplift began. Geophysical data indicate that there is mantle
a transform margin makes Cordilleran geology very complex. For upwelling underneath the Plateau.
a long time, the Cordillera seemed too complicated for a plate tec-
tonics explanation. In 1970, Atwater used the magnetic record of
the Pacific seafloor to predict the behavior of Cordilleran crust. In PLIOCENE
the more than twenty-five years since then, the plate tectonic hy-
potheses just outlined have explained many parts of this complicated
CHRONOSTRATIGRAPHY
story.
Still, problems remain. We can see the coincidence in timing be- The Pliocene was the third part of Lyell's (1831-1833) original
tween the opening of the Basin and Range Province and the growth threefold division of the Tertiary. Lyell divided it into the "Older
of the San Andreas, but what is the connection? A number of hy- Pliocene," which became the Pliocene in the modern sense, and
potheses have been offered. Some have argued that the East Pacific the "Newer Pliocene," which he later renamed the "Pleistocene"
Rise must lie under Nevada, producing the mantle upwelling and in 1839. Lyell based his "Older Pliocene" on the molluscan fau-
crustal extension. However, plate tectonic geometries do not re- nas from the "Subappenine formations" in northern Italy. Like the
quire that this midocean ridge continue indefinitely. It is merely other epochs of the Tertiary, the stratigraphic arguments over the
the expression between two plate boundaries, and there need not definition of the Pliocene were long and convoluted (Berggren and
be any spreading in the mantle where there is no overlying crustal Van Couvering, 1974). Eventually, stratigraphers have come to
 Donald R. Prothero 26

accept the Pliocene as the epoch following the great terminal Miocene by the immigration of the muskrat Pliopotamys at 3. 7 Ma; Blan-
Messinian event in the Mediterranean, whose effects can be recog- can IV spans an interval from 3.2 to 2.6 Ma; Blancan V, marked
nized around the world. The Pliocene-Pleistocene boundary is cur- by the immigration of two subgenera of the lemming Synaptomys,
rently placed at the top of the Olduvai magnetic subchron (Chron runs from 2.6 to 1.9 Ma. It ends with the Blancan/ Irvingtonian
C2n) about 1.8 Ma (Berggren, Kent, and Van Couvering, 1985; boundary, marked by the immigration of Phenacomys, Proneofiber,
Berggren et al., 1995). Microtus, andAllophaiomys.
The upward revision and redating of the Miocene/Pliocene
boundary to the top of the Messinian stage in Europe (Berggren PALEO CLIMATE
and Van Couvering, 1974) caused the recorrelation of the "early
Pliocene" Clarendonian and "middle Pliocene" Hemphillian Land After the great Messinian crisis, the Earth fluctuated in and out
Mammal ages of Wood et al. (1941) with the late Miocene (Tedford of glacial conditions through the Pliocene-Pleistocene. By the mid-
et al., 1987). Of the Wood Committee's original units, only the Pliocene, closure of the Isthmus of Panama and the onset of Northern
latest Hemphillian and the Blancan remain within the Pliocene Hemisphere glaciation brought global climate into its
as presently defined. Most authors (Berggren, Kent, and Van present-day mode of orbitally controlled Ice Age fluctuations
Couvering, 1985; Lundelius et al., 1987) place the Blancan/ (Shackleton and Opdyke, 1973, 1977; Hays, Imbrie, and Shackleton,
Irvingtonian boundary at the base of the Olduvai subchron, about 1976; Keigwin, 1978, 1979; Berger, 1982; Berggren, 1982).
1.88 Ma, slightly older than the European Pliocene-Pleistocene The effect on North America was seen in the dramatic change
boundary. in vegetation, with increased cooling and drying to form a steppe
Blancan localities have been studied for their magnetic stratigra- environment (Gregory, 1971; Webb, 1977, 1983, 1984). This cli-
phy (Figure 1.3) and dated by numerous methods over a wide area matic change at the end of the Miocene (latest Hemphillian) re-
of North America (Johnson, Opdyke, and Lindsay, 1975; Lindsay, sulted in dramatic faunal changes, including extinctions among
Johnson, and Opdyke, 1975; Opdyke et al., 1977; MacFadden, many North American savanna denizens. These included nearly
Johnson, and Opdyke, 1979; Neville et al., 1979; Lindsay, Opdyke, all the rhinocerotids (one tooth scrap is known from the Blancan of
and Johnson, 1984). According to the most recent summaries Texas), most of the horses, camels, antilocaprids, tapirs, peccaries,
(Lindsay, Opdyke, and Johnson, 1984; Lindsay et al., 1987; mastodons, and all of the protoceratids, moschids, dromomerycids,
Lundelius et al., 1987), the Hemphillian/Blancan boundary occurs as well as a variety of rodents, including mylagaulids. In their place
within the upper Gilbert magnetic chron (approximately at the Co- came immigrants from Eurasia, including several types of micro-
chiti event), about 4.0 Ma. Many events within the Blancan have tine and cricetid rodents, cervids, as well as edentates from South
also been precisely dated, using both magnetic stratigraphy and ra- America. Native horses and camels that survived are noticeably
diometric techniques. more hypsodont in response to even grittier diets.

BIOSTRATIGRAPHY PALEOGEOGRAPHY

Unlike some other intervals of the Cenozoic, the Pliocene-Pleisto- Most of the events initiated in the Miocene still continue. The
cene was characterized by rapid faunal turnover and frequent im- Yellowstone hot spot left the Snake River plain and now lies in
migrations of mammals into North America from both Eurasia and its present location, generating geysers and hot springs in a huge
South America. This has allowed fine-scale subdivision of the late caldera formed by a gigantic explosive eruption (Suppe, Powell,
Neogene by numerous faunal events. Lindsay, Opdyke, and John- and Berry, 1975). The Basin and Range continues to rip apart,
son (1984) and Lindsay et al. (1987) recognize at least five separate with the extension just now beginning in southeastern Oregon and
faunal horizons within the Blancan based on lowest and highest Idaho (Dickinson and Snyder, 1979). The Cascades and Sierras have
stratigraphic datum levels. These levels have been precisely dated continued to rotate until they reached their present north-south ori-
in numerous sections with good stratigraphic and faunal data, espe- entation. However, the Cascade arc seems to be less active that
cially those in the Anza-Borrego State Park in California (Opdyke it was in the Miocene (Lipman, Prostka, and Christiansen, 1972;
et al., 1977) and the San Pedro Valley in Arizona (Johnson, Opdyke, Hammond, 1979). Only a few volcanoes, such as Mt. St. Helens,
and Lindsay, 1975). are still erupting in the Washington segment. The northern rem-
Another scheme for subdividing the late Neogene was proposed nant of the Farallon Plate (known as the Juan de Fuca Plate) is still
by Repenning (1987). He used the biochronology of microtine ro- going down the subduction zone, although it is slowing down and
dents, a particularly common and rapidly evolving group found in subduction appears to be shallow. This may explain why there are
nearly all late Cenozoic mammal localities. Microtines were also fewer deep-focus earthquakes in the subduction zone beneath the
frequent migrants to and from the Old World, further enhancing Cascades. The trench off the Oregon-Washington coast is not very
their biochronological potential. Repenning (1987) marked the deep because it is very rapidly filled with sediments eroded from
Hemphillian/Blancan boundary by the immigration of two subgen- the humid, recently glaciated lands drained by a large river system.
era of Mimomys at 4.8 Ma (older than the estimate discussed by Pliocene tectonism extended far beyond the Basin and Range
Lundelius et al., 1987). He subdivided the Blancan into five inter- and Cascades, however. About 5 million years ago, the "four
vals, labeled as Blancan I-V. Blancan I spans the interval from 4.8 comers" region in the comers of Arizona, Utah, Colorado, and
to 4.25 Ma; Blancan II from 4.25 to 3.7 Ma; Blancan III is marked New Mexico was uplifted as a single coherent block known as the
 Background to North American mammalian evolution 27

Colorado Plateau (McKee and McKee, 1972). Unlike the Basin paleotemperatures to a few degrees. More importantly, our im-
and Range, the Colorado Plateau did not break into hundreds of proved methods of correlation and dating have allowed us to tie these
parallel normal faults. Instead, it is only slightly deformed by small North American events to the global history of climate and to de-
normal faults and gentle folds draped over them; most of the rocks termine what sort of causal relationships might exist between them.
remained nearly horizontal, even though they have been uplifted at This connection is very important, for an improved understanding
least 1.5 km (about a mile). Some parts of the Colorado Plateau, such of global climatic effects on North American landscape and vege-
as the Kaibab Plateau on the north rim of the Grand Canyon, reach tation is critical to hypotheses of mammalian community evolution
elevations over 2,700 m (9,000 feet). Geophysical studies show that and interactions. For example, Osborn (1929) attributed the ex-
the crust beneath the Colorado Plateau is not thick enough to explain tinction of brontotheres to non-Darwinian internal forces such as
its uplift by isostasy alone, so mantle upwelling must be occurring, "racial senescence." We now realize that their extinction at the end
as it is in the Basin and Range. Today, the Colorado Plateau is deeply of the Chadronian coincides with a wide variety of climatic changes
eroded due to its gradual uplift, forming the spectacular "red rock" induced by Antarctic glaciation and global cooling, which had dra-
canyons and monuments of many scenic parks: Bryce, Zion, Capitol matic effects on the vegetation and habitat on which they depended
Reef, Canyonlands, Arches, Monument Valley, and especially the (Retallack, 1983a, 1983b; Prothero, 1994a, 1994b; Berggren and
Grand Canyon. In the Grand Canyon and the Goosenecks of the San Prothero, 1992; Evanoff, Prothero, and Lander, 1992).
Juan River, the gently meandering course of the river was deeply The importance of this climatic-evolutionary nexus is now be-
incised into Paleozoic sandstone and limestone bedrock, producing coming widely appreciated as the recent workshops on ancient
meanders that lie in canyons almost a mile deep. This could only terrestrial ecosystems attest (e.g., Behrensmeyer et al., 1992). As
happen if the meanders had been established on a low-elevation, low- further details emerge from the deep-sea floor, we are also reach-
relief surface during the Miocene and then were forced to cut down ing a much greater level of understanding of global climate than
through hard bedrock as the Colorado Plateau rose beneath them. ever before. However, the critical key to all of these hypotheses
A similar phenomenon occurred in the Rocky Mountains and the is chronostratigraphy. Unless the precise temporal correspondence
Sierras. We have seen that the great Laramide uplifts were almost between events can be established, the cause-and-effect connec-
completely buried during the Miocene, and only small local basins tion between climatic change and organismal response is fatally
accumulated sediments in Wyoming, Colorado, and Utah. At the flawed. As the fundamental changes in Eocene-Oligocene chronol-
end of the Miocene, however, long-buried Rockies were resurrected ogy have demonstrated, evolutionary-climatic hypotheses built on
again after nearly disappearing under their own debris (Scott, 1975). a weak chronological foundation are as fragile as a house of cards
As their sedimentary cover was stripped away, the flat Eocene (Prothero and Swisher, 1992).
Rocky Mountain surface was exposed. The basins between the an-
cient Laramide uplifts were even easier to erode because they were
filled with soft Cenozoic sediments, not hard granites and metamor- ACKNOWLEDGMENTS
phic basement. The rivers that once easily crossed the buried moun-
tain ranges were now forced to cut down rapidly as they stripped I thank Christine Janis for inviting me to contribute this chapter,
away the basin fill on either side. Such rapid downcutting produced and David Schwimmer, Scott Wing, and Michael Woodbume for
deep canyons cut through the hard cores of mountains, so that rivers reviewing it.
appear to take the "hard" way through a mountain rather than the
"easy" way around it. The Rocky Mountains are full of examples
of water gaps caused by drainages superimposed on a buried to- REFERENCES
pography. Spectacular canyons such as Royal Gorge in Colorado
and Split Mountain and Flaming Gorge in Utah are well-known Adams, C. G., Benson, R. H., Kidd, R. B., & Wright, R. C. (1977).
examples of superposed drainages. The Messinian salinity crisis and evidence of late Miocene eustatic
Other regions, such as the Rio Grande Rift (Chapin, 1979) and changes in the world ocean. Nature, 269, 383-6.
Addicott, W. 0. (1970). Latitudinal gradients in Tertiary molluscan faunas
Alaska, have long and complex Pliocene-Pleistocene histories, but
of the Pacific coast. Palaeogeography, Palaeoclimatology, Palaeoe-
space limitations prevent a full discussion here. Most of the Pliocene cology, 8, 287-312.
paleogeography of North America is a continuation of Miocene Archibald, J. D., Clemens, W. A., Gingerich, P. D., Krause, D. W., Lindsay,
trends, and still can be seen today (although overprinted by the E. H., & Rose, K. D. (1987). First North American land mammal
effects of Pleistocene glaciation). ages of the Cenozoic Era. In Cenozoic Mammals of North America,
ed. M. 0. Woodburne, pp. 24-76. Berkeley: University of California
Press.
Armstrong, R. L., Ekren, E. B., McKee, E. H., &Noble, D. C. (1969). Space-
CONCLUSION time relations of Cenozoic silicic volcanism in the Great Basin of the
western United States. American Journal of Science, 267, 478-90.
Our understanding of the Cenozoic paleogeography, climate, and Atwater, T. (1970). Implications of plate tectonics for the Cenozoic tectonic
chronostratigraphy of North America is constantly improving. From evolution of western North America. Geological Society of America
Bulletin, 81, 3518-36.
the generalized understanding of climatic and geographic patterns
(1989). Plate tectonic history of the northeast Pacific and western North
of Osborn (1910) and the chronology of Wood etal. (1941), we now America. In The Eastern Pacific Ocean and Hawaii, eds. E. L. Win-
can resolve many events to less than a million years and determine terer, D. M. Hussong, & R. W. Decker, pp. 21-71. The Geology
 Donald R. Prothero 28

of North America. Vol. N. Boulder, Colo.: Geological Society of (1993). NW European and NE Atlantic Paleocene-Eocene boundary in-
America. terval: bio- and sequence stratigraphy and geochronology. Journal
Aubry, M.-P. (1985). Northwestern European Paleogene magnetostratigra- of Vertebrate Paleontology, 13, suppl. to no. 3, p. 36A.
phy, biostratigraphy, and paleogeography: calcareous nannofossil Berggren, W. A., Kent, D. V., Aubry, M.-P., Swisher, C. C., & Miller, K.
evidence. Geology, 13, 198-202. G. (1995). A revised Paleogene geochronology and chronostratigra-
Aubry, M.-P., Berggren, W. A., Kent, D. V., Flynn, J. J., Klitgord, K. D., phy. Society of Economic Paleontologists and Mineralogists Special
Obradovich, J. D., & Prothero, D.R. (1988). Paleogene geochronol- Publications, 54, 130-212.
ogy: an integrated approach. Paleoceanography, 3, 707-42. Berggren, W. A., Kent, D. V., & Flynn, J. J. (1985). Paleogene geochronology
Axelrod, D. I. (1958). Evolution of the Madro-Tertiary geoflora. Botanical and chronostratigraphy. In The Chronology of the Geological Record,
Review, 24, 433-509. ed. N. J. Snelling. Memoir of the Geological Society of London, 10,
Barghoorn, S. ( 1981 ). Magnetic polarity stratigraphy of the Miocene type 141-95.
Tesuque Formation, Santa Fe Group, in the Espanola Valley, New Berggren, W. A., Kent, D. V., Obradovich, J. D., & Swisher, C. C., III. (1992).
Mexico. Geological Society of America Bulletin, 92, 1027-41. Toward a revised Paleogene geochronology. In Eocene-Oligocene
Barker, P. F., & Burrell, J. (1977). The opening of the Drake Passage. Marine Climatic and Biotic Evolution, eds. D. R. Prothero & W. A. Berggren,
Geology, 25, 15-34. pp. 29-45. Princeton: Princeton University Press.
(1982). The influence upon Southern Ocean circulation sedimentation Berggren, W. A., Kent, D. V., & Van Couvering, J. A. (1985). Neogene
and climate of the opening of Drake Passage. In Antarctic Geo- geochronology and chronostratigraphy. In The Chronology of the
science, ed. C. Craddock, pp. 377-85. Madison: University of Wis- Geological Record, ed. N. J. Snelling. Memoir of the Geological
consin Press. Society of London, 10, 211--60.
Barrett, P. J., Elston, D. P., Harwood, D. M., McKelvey, B. C., & Webb, Berggren, W. A., & Prothero, D.R. (1992). Eocene-Oligocene climatic and
P. N. (1987). Mid-Cenozoic record of glaciation and sea-level change biotic evolution: an overview. In Eocene-Oligocene Climatic and
on the margin of Victoria Land Basin, Antarctica. Geology, 14, Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 1-28.
634-7. Princeton: Princeton University Press.
Barrett, P. J., Hambrey, M. J., Harwood, D. M., Pyne, A. R., & Webb, P. N. Berggren, W. A., & Van Couvering, J. A. (1974). The Late Neogene. Ams-
(1989). Synthesis. In Antarctic Cenozoic History from the CIROS-1 terdam: Elsevier.
Drillhole, McMurdo Sound, ed. P. J. Barrett, pp. 241-51. Wellington, Berry, W. B. N. (1987). Growth ofa Prehistoric Time Scale Based on Organic
New Zealand: Science and Information Publishing Centre, Depart- Evolution (2nd ed.). Palo Alto, Calif.: Blackwell Scientific.
ment of Scientific and Industrial Research Bulletin no. 245. Beyrich, H. E. von. (1854). Ober die Stellung die hessischen Tertiar-
Barron, J. (1973). Late Miocene-early Pliocene paleotemperatures for bildungen. Koniglijke. Preussische Akademie Wissenschaften Berlin
California from marine diatom evidence. Palaeogeography, Palaeo- Monatsheft, 1854, 664-6.
climatology, Palaeoecology, 14, 277-91. Birkenmajer, K. (1987). Tertiary glacial and interglacial deposits, South
Barron, J., Larsen, B., & Leg 119 Shipboard Party. (1989). Leg 119 sum- Shetland Islands, Antarctica: geochronology versus biostratigraphy
mary. Proceedings of the Ocean Drilling Program, 119, 1-939. (a progress report). Bulletin of the Polish Academy Science, Earth
Bartek, L. R., & Anderson, J. B. (1990). Neogene stratigraphy of the Ross Science, 36, 133-45.
Sea continental shelf: revelations from Leg 2 of the 1990 Ross Sea Bradley, W. C. (1987). Erosion surfaces of the Colorado Front Range. In
Expedition of the RIV Polar Duke. United States Geological Survey Geomorphic Systems ofNorth America, ed. W. L. Graf, The Geology
Open-File Report, 90-309, 53-62. of North America. Boulder, Colo.: Geological Society of America
Bartek, L. R., Sloan, L. C., Anderson, J.B., & Ross, M. I. (1992). Evidence Centennial Special Vol. 2, 215-20.
from the Antarctic continental margin of late Paleogene ice sheets: Bradley, W. H. (1964). Geology of the Green River Formation and associ-
a manifestation of plate reorganization and synchronous changes ated Eocene rocks in southwestern Wyoming and adjacent parts of
in atmospheric circulation over the emerging Southern Ocean? In Colorado and Utah. United States Geological Survey Professional
Eocene-Oligocene Climatic and Biotic Evolution, eds. D.R. Prothero Paper, 496-A, 1-86.
& W. A. Berggren, pp. 131-59. Princeton: Princeton University Bressler, S. L., & Butler, R. F. (1978). Magnetostratigraphy of the late Ter-
Press. tiary Verde Formation, central Arizona. Earth and Planetary Sci-
Bartow, J. A. (1978). Oligocene continental sedimentation in the Caliente ences Letters, 38, 319-30.
Range, California. Journal of Sedimentary Petrology, 48, 75-98. Breza,J., Wise, S. W.,Zachos,J.C.,and theLeg 120ShipboardParty. (1989).
Beard, K. C., & Tabrum, A. R. (1991). The first early Eocene mammal Lower Oligocene ice-rafted debris at 58°S on the Kerguelen Plateau:
from eastern North America: an omomyid primate from the Bashi the 'Smoking Gun' for the existence of an early Oligocene ice sheet
Formation, Lauderdale County, Mississippi. Mississippi Geology, on East Antarctica. Third International Conference on Paleoceanog-
11, 1-6. raphy. Cambridge: Blackwell Scientific, Abstract Volume, p. 24.
Behrensmeyer, A. K., Damuth, J. D., DiMichele, W. A., Potts, R., Sues, Brinkhuis, H. (1992). Late Paleogene dinoflagellate cysts with special
H.-D., & Wing, S. L. (eds.). (1992). Terrestrial Ecosystems Through reference to the Eocene/Oligocene boundary. In Eocene-Oligocene
Time. Evolutionary Paleoecology of Terrestrial Plants and Animals. Climatic and Biotic Evolution, eds. D. R. Prothero & W. A.
Chicago: University of Chicago Press. Berggren, pp. 327-40. Princeton: Princeton University Press.
Berger, W. (1982). Climate steps in ocean history - lessons from the Pleis- Brunet, M. (1977). Les mammireres et le probleme de la limite Eocene-
tocene. In Climate in Earth History, eds. W. Berger & J.C. Crowell, Oligocene en Europe. Memoires Speciaux Geobios, 1, 11-27.
pp. 43-54. Washington, D.C.: National Academy of Sciences. Brunet, M., Franzen, J. L., Godinot, M., Hooker, J. J., Legendre, S.,
Berggren, W. A. (1971). Tertiary boundaries. In Marine Micropaleontology Schmidt-Kittler, N., & Vianey-Liaud, M., coordinators (1987).
of the Oceans, eds. W. R. Riedel & B. M. Funnell, pp. 693-803. European reference levels and correlation tables. Miinchner
Cambridge: Cambridge University Press. Geowissenschaften Abhandlungen (A), 10, 13-31.
(1982). Role of ocean gateways in climatic change. In Climate in Earth Burbank, D., & Whistler, D. (1987). Temporally constrained tectonic
History, eds. W. Berger & J. C. Crowell, pp. 118-285. Washington, rotations derived from magnetostratigraphic data: implications for
D.C.: National Academy of Sciences. the initiation of the Garlock Fault, California. Geological Society of
(1986). Geochronology of the Eocene/Oligocene boundary. In Terminal America Bulletin, 99, 1172-5.
Eocene Events, eds. C. Pomerol & I. Premoli-Silva, pp. 349-56. Butler, R. F., Gingerich, P. D., & Lindsay, E. H. (1980). Magnetic polarity
Amsterdam: Elsevier. stratigraphy and Paleocene-Eocene biostratigraphy of Polecat Bench,
 Background to North American mammalian evolution 29

Wyoming. University of Michigan Papers in Paleontology, 24, Costa, L. I., & Downie, C. (1976). The distribution of the dinoflagellate Wet-
95-8. zeliella in the Palaeogene of north-western Europe. Palaeontology,
(1984). Magnetic polarity stratigraphy and biostratigraphy of Pale- 19, 591--614.
ocene and lower Eocene continental deposits, Clark's Fork Basin, Crouch, J. K. (1979). Neogene tectonic evolution of the California con-
Wyoming. Journal of Geology, 89, 299-316. tinental borderland and western Transverse Ranges. Geological
Butler, R. F., Krause, D. W., & Gingerich, P. D. (1987). Magnetic polarity Society of America Bulletin, 90, 338-45.
stratigraphy and biostratigraphy of middle-late Paleocene continental Crowell, J.C. (1974). Origin oflate Cenozoic basins in southern California.
deposits of south-central Montana. Journal of Geology, 95, 64 7-57. Society of Economic Paleontologists and Mineralogists Special
Butler, R. F., & Lindsay, E. H. (1985). Mineralogy of magnetic minerals Publication, 22, 190-204.
and revised magnetic polarity stratigraphy of continental sediments, Curry, D., & Odin, G. S. (1982). Dating of the Paleogene. In Numerical
San Juan Basin, New Mexico. Journal of Geology, 93, 535-54. Dating in Stratigraphy, ed. G. S. Odin, pp. 607-30. New York:
Butler, R. F., Lindsay, E. H., Jacobs, L. L., & Johnson, N. M. (1977). Wiley.
Magnetostratigraphy of the Cretaceous-Tertiary boundary in the Dane, C. H. (1955). Stratigraphic and facies relationships of the upper part
San Juan Basin, New Mexico. Nature, 267, 318-23. of the Green River Formation and lower part of the Uinta Formation
Cady, W. M. (1975). TectOnic setting of Tertiary volcanic rocks of the in Duchesne, Uintah, and Wasatch Counties, Utah. American
Olympic Peninsula, Washington. United States Geological Survey Association of Petroleum Geologists Bulletin, 38, 405-25.
Research, 3, 575-82. Davis, G. A. (1980). Problems of intraplate extensional tectonics, western
Cande, S. C., & Kent, D. V. (1992). A new geomagnetic polarity time United States. In Continental Tectonics, Studies in Geophysics,
scale for the late Cretaceous and Cenozoic. Journal of Geophysical pp. 84-95. Washington, D.C.: National Academy of Sciences.
Research, 97, 13917-51. Devereux, I. (1967). Oxygen isotope paleotemperature measurements on
(1995). Revised calibration of the geomagnetic polarity time scale for New Zealand Tertiary fossils. New Zealand Journal of Science, 10,
the late Cretaceous and Cenozoic. Journal of Geophysical Research, 988-1011.
100, 6093-5. Dibblee, T. W. Jr. (1966). Geology of the central Santa Ynez Mountains,
Cerling, T. E. (1992). Development of grasslands and savannas in East Santa Barbara County, California. California Division of Mines and
Africa during the Neogene. Palaeogeography, Palaeoclimatology, Geology Bulletin, 186, 1-99.
Palaeoecology, 97, 241-7. Dickinson, W.R. (1979). Cenozoic plate tectonic setting of the Cordilleran
Cerling, T. E., Wang, Y., & Quade, J. (1993). Expansion of C4 ecosystems region in the United States. In Cenozoic Paleogeography of the
as an indicator of global ecological change in the late Miocene. Western United States, eds. J. M. Armentrout, M. R. Cole, & H.
Nature, 261, 344-5. TerBest. Pacific Section, Society of Economic Paleontologists and
Chapin, C. E. (1979). Evolution of the Rio Grande Rift- a summary. In Rio Mineralogists Symposium, 3, 1-13.
Grande Rift-Tectonics and Magmatism, ed. R. E. Riecker, pp. 1-5. Dickinson, W. R., Klute, M. A., Hayer, M. J., Janecke, S. V., Lundin, E. R.,
Washington, D.C.: American Geophysical Union. McKittrick, M. A., & Olivaresss, M. D. (1988). Paleogeographic
Christiansen, R. L., & Yeats, E. H. (1989). Post-Laramide geology of the and paleotectonic setting of Laramide sedimentary basins in the
Cordilleran region. In The Cordilleran Orogen, eds. B. C. Burchfiel, central Rocky Mountain region. Geological Society of America
P. W. Lipman, & M. L. Zoback. Geology of North America, vol. Bulletin, 100, 1023-39.
G-3. Boulder, Colo.: Geological Society of America. Dickinson, W.R., & Snyder, W. S. (1978). Plate tectonics of the Laramide
Cirbus Sloan, L.; Walker, J. C. G., Moore, T. C., Jr., Rea, D. K., & Zachos, orogeny. Geological Society of America Memoir, 141, 355-66.
J.C. (1992). Possible methane-induced polar warming in the early Dickinson, W. R., & Snyder, W. S. (1979). Geometry of subducted slabs
Eocene. Nature, 357, 320-2. related to the San Andreas transform. Journal of Geology, 887,
Clark, J. (1937). The stratigraphy and paleontology of the Chadron 609-27.
Formation in the Big Badlands of South Dakota. Annals of the Eaton, G. P. ( 1980). Geophysical and geological characteristics of the crust
Carnegie Museum of Natural History, 25, 261-350. of the Basin and Range Province. In Continental Tectonics, Studies
(1954). Geographic designation of the members of the Chadron in Geophysics, pp. 96-113. Washington, D.C.: National Academy
Formation in South Dakota. Annals of the Carnegie Museum of of Sciences.
Natural History, 33, 197-8. (1982). The Basin and Range province: origin and tectonic significance.
(1975). Controls of sedimentation and provenance of sediments in the Annual Reviews of Earth and Planetary Science, 10, 409-40.
Oligocene of the central Rocky Mountains. Geological Society of Emry, R. J. (1973). Stratigraphy and preliminary biostratigraphy of the
America Memoir, 144, 95-118. Flagstaff Rim area, Natrona County, Wyoming. Smithsonian
Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene Contributions to Paleobiology, 18, 1-42.
sedimentation, stratigraphy, paleoecology, and paleoclimatology in ( 1981 ). Additions to the mammalian fauna of the type Duchesnean, with
the Big Badlands of South Dakota. Fieldiana Geology Memoir, 5, comments on the status of the Duchesnean. Journal of Paleontology,
1-158. 55, 563-70.
Clyde, W. C., Stamatakos, J., & Gingerich, P. D. ( 1994). Chronology of the (1992). Mammalian range zones in the Chadronian White River For-
Wasatchian land-mammal age (early Eocene): magnetostratigraphic mation at Flagstaff Rim, Wyoming. In Eocene-Oligocene Climatic
results from the McCullough Peaks section, northern Bighorn Basin, and Biotic Evolution, eds. D. R. Prothero & W. A. Berggren, pp.
Wyoming. Journal of Geology, 102, 367-77. 106-15. Princeton: Princeton University Press.
Cole, M. R., & Armentrout, J. M. (1979). Neogene paleogeography of Emry, R. J., Bjork, P.R., & Russell, L. S. (1987). The Chadronian, Orellan,
the western United States. In Cenozoic Paleogeography of the and Whitneyan land mammal ages. In Cenozoic Mammals of North
Western United States, eds. J. M. Armentrout, M. R. Cole, & H. America, ed. M. 0. Woodburne, pp. 118-52. Berkeley: University
TerBest. Pacific Section, Society of Economic Paleontologists and of California Press.
Mineralogists Symposium, 3, 297-324. Epis, R. C., & Chapin, C. E. (1975). Geomorphic and tectonic implications of
Cook, E. F. (1965). Stratigraphy of Tertiary volcanic rocks in eastern the post-Laramide, late Eocene erosion surface in the Southern Rocky
Nevada. Nevada Bureau of Mines Report: 11, 1-61. Mountains. Geological Society of America Memoir, 144, 45-74.
Costa, L. I., Denison, C., & Downie, C. (1978). The Palaeocene/Eocene Evanoff, E. (1990). Early Oligocene paleovalleys in southern and cen-
boundary in the Anglo-Paris Basin. Journal of the Geological tral Wyoming: evidence of light local relief on the late Eocene
Society of London, 135, 261-4. unconformity. Geology, 18, 443-6.
 Donald R. Prothero 30

Evanoff, E., Prothero, D. R., & Lander, R. H. (1992). Eocene-Oligocene Fork mammal fauna, and its correlation with similar faunas in
climatic change in North America: the White River Formation near Europe and Asia. Geobios Memoir Special, 1, 39-45.
Douglas, east-central Wyoming. In Eocene-Oligocene Climatic and Gingerich, P. D., & Simons, E. L. (1977). Systematics, phylogeny, and
Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 116-30. evolution of early Eocene Adapidae (Mammalia, Primates) in
Princeton: Princeton University Press. North America. Contributions from the Museum of Paleontology,
Evernden, J. F., Savage, D. E., Curtis, G. H., & James, G. T. (1964). University of Michigan, 24, 245-79.
Potassium-argon dates and the Cenozoic mammalian chronology of Glass, B. P., & Crosbie, J. R. (1982). Age of the Eocene/Oligocene boundary
North America. American Journal of Science, 262, 145-98. based on extrapolation from North American microtektite layer.
Fahlbusch, V. (1976). Report on the International Symposium on mam- Bulletin of the American Association of Petroleum Geologists, 66,
malian stratigraphy of the European Tertiary. Newsletters in 471-76.
Stratigraphy, 5, 160-7. Glass, B. P., Hall, C. D., & York, D. (1986). 40Ar!3 9Ar laser-probe dating
Fisher, R. V., & Rensberger, J. M. (1972). Physical stratigraphy of the of North American tektite fragments from Barbados and the age
John Day Formation, central Oregon. University of California of the Eocene-Oligocene boundary. Chemical Geology (Isotope
Publications in Geological Sciences, 101, 1-33. Geoscience Section), 59, 181-6.
Flores, R. M. (1981). Coal deposition in fluvial paleoenvironments of the Gregory, J. T. (1971). Speculations of the significance of fossil vertebrates
Paleocene Tongue River Member of the Fort Union Formation, for the antiquity of the High Plains in North America. Abhand-
Power River area, Powder River Basin, Wyoming and Montana. lung Hessisches Landesamtes far Bodenforschung (Heinz Tobien
Society of Economic Mineralogists and Paleontologists Special Festschrift), pp. 64-72.
Publication, 31, 169-90. Gunnell, G. F. (1989). Evolutionary history of Microsyopoidea (Mam-
(1983). Bains facies analysis of coal-rich Tertiary fluvial deposits, malia, ?Primates) and the relationship between Plesiadapiformes and
northern Powder River Basin, Montana and Wyoming. International Primates. University ofMichigan Papers in Paleontology, 27, 1-155.
Association ofSedimentologists Special Publication, 6, 501-15. Gunnell, G. F., Bartels, W. S., & Gingerich, P. D. (1993). Paleocene-Eocene
Flores, R. M., & Ethridge, F. G. (1985). Evolution of intermontane fluvial boundary in continental North America: biostratigraphy and
systems of Tertiary Powder River Basin, Montana and Wyoming. geochronology, northern Bighorn Basin, Wyoming. New Mexico
In Cenozoic Paleogeography of the West Central United States, eds. Museum of Natural History Bulletin, 2, 137-44.
R. M. Flores & S. Kaplan, pp. 107-27. Special Publication of the Hamilton, W., & Myers, W. B. (1966). Cenozoic tectonics of the western
Rocky Mountain Section, Society of Economic Paleontologists and United States. Reviews of Geophysics, 4, 509-47.
Mineralogists. Hammond, P. E. (1979). A tectonic model for the Cascade Range. In
Flynn, J. J. (1986). Correlation and geochronology of middle Eocene strata Cenozoic Paleogeography of the Western United States, eds. J. M.
from the western United States. Palaeogeography, Palaeoclimatol- Armentrout, M. R. Cole, & H. TerBest. Pacific Section, Society of
ogy, Palaeoecology, 55, 335-406. Economic Paleontologists and Mineralogists Symposium, 3, 219-38.
Frazier, W. J., & Schwimmer, D. R. (1987). Regional Stratigraphy of North Haq, B. U., Hardenbol, J., & Vail, P. R. (1987). The chronology of
America. New York: Plenum. fluctuating sea level since the Triassic. Science, 235, 1156-67.
Frederiksen, N. 0. (1980). Paleogene sporomorphs from South Carolina and Hardenbol, J., & Berggren, W. A. (1978). A new Paleogene numerical
quantitative correlation with the Gulf Coast. Palynology, 4, 125-79. time scale. American Association of Petroleum Geologists Studies
Galloway, W. E., Bebout, D. G., Fisher, W. L., Dunlap, J. B., Jr., Cabrera- in Geology, 6, 213-34.
Castro, R., Lugo-Rivera, J.E., & Scott, T. M. (1991). Cenozoic. In Hays, J. D., Imbrie, J., & Shackleton, N. J. (1976). Variations in Earth's
The Gulf of Mexico Basin, ed. A. Salvador, The Geology of North orbit: pacemaker of the ice ages. Science, 194, 1121-32.
America, vol. J. Boulder, Colo.: Geological Society of America. Hooker, J. J. (1992). British mammalian paleocommunities across the
Galloway, W. E., Hobday, D. K., & Magara, K. (1982). Frio Formation Eocene-Oligocene transition and their environmental implications.
of the Texas Gulf Coastal Plain: depositional systems, structural In Eocene-Oligocene Climatic and Biotic Evolution, eds. D. R.
framework, and hydrocarbon distribution. American Association of Prothero & W. A. Berggren, pp. 494-515. Princeton: Princeton
Petroleum Geologists Bulletin, 66, 649-88. University Press.
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North Hsti, K. J., Montaderi, L., Bernoulli, D., Bianca Cita, M., Erickson, A., Garri-
America. Smithsonian Miscellaneous Collections, 128, 1-96. son, R. E., Kidd, R. B., Meiliers, F., Mi.iller, C., & Wright, R. ( 1977).
(1956). The geology and vertebrate paleontology of upper Eocene History of the Mediterranean salinity crisis. Nature, 267, 399-403.
strata in the northeastern part of the Wind River Basin, Wyoming, Hunt, R. M., & MacFadden, B. J. (1998). Magnetostratigraphy of the type
Part 2. The mammalian fauna of the Badwater area. Smithsonian Arikaree Group, western Nebraska. Geological Society of America
Miscellaneous Collections, 131, 1-35. Special Paper (in press).
(1959). Paleontological exploration and dating of the early Tertiary Hutchison, J. H. (1982). Turtle, crocodilian and champsosaur diversity
deposits adjacent to the Uinta Mountains. IntermountainAssociation changes in the Cenozoic of the north-central region of the western
of Petroleum Geologists Guidebook, 10, 131-8. United States. Palaeogeography, Palaeoclimatology, Palaeoecology,
Gingerich, P. D. (1976). Cranial anatomy and evolution of early Tertiary 37, 149-64.
Plesiadapidae (Mammalia, Primates). University of Michigan (1992). Western North American reptile and amphibian record across
Papers on Paleontology, 15, 1-141. the Eocene/Oligocene boundary and its climatic implications.
(1980). Evolutionary patterns in early Cenozoic mammals. Annual In Eocene-Oligocene Climatic and Biotic Evolution, eds. D. R.
Review of Earth and Planetary Sciences, 8, 407-24. Prothero & W. A. Berggren, pp. 451-63. Princeton: Princeton
(1983). Paleocene-Eocene fauna! zones and a preliminary analysis of University Press.
Laramide structural deformation in the Clarks Fork Basin, Wyoming. Jerskey, R. G. (1981). A Paleomagnetic study of the Bridger Formation,
"1yoming Geological Association Guidebook, 34, 185-95. Southern Green River Basin, Wyoming. Unpublished M.S. Thesis.
(1989). New earliest Wasatchian mammalian fauna from the Eocene Milwaukee: University of Wisconsin.
of northwestern Wyoming: composition and diversity in a rarely Johnson, N. M., Opdyke, N. D., & Lindsay, E. H. (1975). Magnetic
sampled high-floodplain assemblage. University ofMichigan Papers polarity stratigraphy of Pliocene-Pleistocene terrestrial deposits and
on Paleontology, 28, 1-97. vertebrate faunas, San Pedro Valley, Arizona. Geological Society of
Gingerich, P. D., & Rose, K. D. (1977). Preliminary report on the Clark America Bulletin, 86, 5-12.
 Background to North American mammalian evolution 31

Johnson, R. C. (1985). Early Cenozoic history of the Uinta and Piceance ODP Leg 113. Proceedings of the Ocean Drilling Program,
Creek Basins, Utah and Colorado, with special reference to the Scientific Results, 113, 865-80.
development of Eocene Lake Uinta. In Cenozoic Paleogeography ( 1991 ). Abrupt deep-sea warming, paleoceanographic changes and
of the West Central United States, eds. R. M. Flores & S. Kaplan, benthic extinctions at the end of the Palaeocene. Nature, 353, 225-9.
pp. 247-77. Special Publication of the Rocky Mountain Section, Koch, P. L., Zachos, J.C., & Gingerich, P. D. (1992). Correlation between
Society of Economic Paleontologists and Mineralogists. isotope records in marine and continental carbon reservoirs near the
Kamp, P. J. J., Waghorn, D. B., & Nelson, C. S. (1990). Late Eocene-early Palaeocene/Eocene boundary. Nature, 358, 319-22.
Oligocene integrated isotope stratigraphy and biostratigraphy for Korth, W.W. (1989). Stratigraphic occurrence of rodents and lagomorphs in
paleoshelf sequences in southern Australia: paleoceanographic the Orella Member, Brule Formation (Oligocene), northwestern Ne-
implications. Palaeogeography, Palaeoclimatology, Palaeoecology, braska. Contributions to Geology, University of"1yoming, 27, 15-20.
80, 311-23. Krishtalka, L., Stucky, R. K., West, R. M., McKenna, M. C., Black,
Kanter, L. R., & Debiche, M. (1985). Modeling the motion histories of C. C., Bown, T. M., Dawson, M. R., Golz, D. J., Flynn, J. J.,
the Point Arena and Central Salinia terranes. In Tectonostrati- Lillegraven, J. A., & Turnbull, W. A. (1987). Eocene (Wasatchian
graphic Terranes of the Circum-Pacijic Region, eds. D. G. Howell, through Duchesnean) biochronology of North America. In Cenozoic
pp. 227-38. Houston: Circum-Pacific Council for Energy and Mammals of North America, ed. M. 0. Woodburne, pp. 77-117.
Mines, Earth Science Series I. Berkeley: University of California Press.
Keigwin, L. D., Jr. (1978). Pliocene closing of the Isthmus of Panama, LeMasurier, W. E., & Rex, D. C. (1982). Volcanic record of Cenozoic
based on biostratigraphic evidence from nearby Pacific Ocean and glacial history in Marie Byrd Land and western Ellsworth Land:
Caribbean Sea cores. Geology, 6, 630-4. revised chronology and evaluation of tectonic factors. In Antarctic
(1979). Late Cenozoic stable isotope stratigraphy and paleoceanography Geoscience, ed. C. Craddock, pp. 725-34. Madison: University of
of Deep Sea Drilling Project sites form the East Equatorial and Wisconsin Press.
Central North Pacific Oceans. Earth and Planetary Sciences Letters, Leopold, E. B., & Denton, M .F. (1987). Comparative age of grassland and
45, 361-82. steppe east and west of the northern Rocky Mountains. Annals of
Keith, S. B. (1978). Paleosubduction geometries inferred from Cretaceous the Missouri Botanical Garden, 74, 841--67.
and Tertiary magmatic patterns in southwestern North America. Leopold, E. B., Liu G., & Clay-Poole, S. (1992). Low-biomass vegetation in
Geology, 6, 516-21. the Oligocene? In Eocene-Oligocene Climatic and Biotic Evolution,
Keller, G. (1985a). Eocene and Oligocene stratigraphy and erosional eds. D. R. Prothero & W. A. Berggren, pp. 399-420. Princeton:
unconformities in the Gulf of Mexico and the Gulf Coast. Journal Princeton University Press.
of Paleontology, 59, 882-903. Lillegraven, J. A. (1979). A biogeographical problem involving compar-
(1985b). Depth stratification ofplanktonic foraminifers in the Miocene isons of late Eocene terrestrial vertebrate faunas of western North
ocean. Geological Society of America Memoirs, 163, 177-93. America. In Historical Biogeography, Plate Tectonics, and the
Keller, G., & Barron, J. A. (1983). Paleoclimatic implications of Miocene Changing Environment, eds. J. Gray & A. J. Boucot, pp. 333-47.
deep-sea hiatuses. Geological Society of America Bulletin, 94, Corvallis: Oregon State University Press.
590-613. Lillegraven, J. A., & Ostresh, L. M., Jr. (1988). Evolution of Wyoming's
Keller, G., Herbert, T., Dorsey, R., D'Hondt, S., Johnsson, M., & Chi, W. early Cenozoic topography and drainage patterns. National
R. (1987). Global distribution of late Paleogene hiatuses. Geology, Geographic Research, 4, 303-27.
15, 199-203. Lindsay, E. H. (1972). Small mammal fossils from the Barstow Formation,
Kelly, T. S. (1990). Biostratigraphy ofUintan and Duchesnean land mammal California. University of California Publications in Geological
assemblages from the middle member of the Sespe Formation, Simi Sciences, 93, 1-104.
Valley, California. Contributions to Science of the Natural History Lindsay, E. H., Butler, R. F., & Johnson, N. M. (1981). Magnetic polarity
Museum of Los Angeles County, 419, 1-42. zonation and biostratigraphy of Late Cretaceous and Paleocene
Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the continental deposits, San Juan Basin, New Mexico. American
Circum-Antarctic Ocean, and their impact on global paleoceanog- Journal of Science, 281, 390-435.
raphy. Journal of Geophysical Research, 82, 3843-60. Lindsay, E. H., Jacobs, L. L., & Butler, R. F. (1978). Biostratigraphy and
(1978). The development of planktonic biogeography in the Southern magnetic polarity stratigraphy of upper Cretaceous and Paleocene
Ocean during the Cenozoic. Marine Micropaleontology, 3, 301-45. terrestrial deposits, San Juan Basin, New Mexico. Geology, 6,
( 1980). Paleoceanographic and biogeographic evolution of the 425-9.
Southern Ocean during the Cenozoic, and Cenozoic microfossil Lindsay, E. H., Johnson, N. M., & Opdyke, N. D. (1975). Preliminary
datums. Palaeogeography, Palaeoclimatology, Palaeoecology, 31, correlation of North American land mammal ages and geomagnetic
123-52. chronology. University of Michigan Papers in Paleontology, 12,
Kennett, J. P., & Barker, P. F. ( 1990). Latest Cretaceous to Cenozoic climate 111-9.
and oceanographic developments in the Weddell Sea, Antarctica: Lindsay, E. H., Opdyke, N. D., & Johnson, N. M. (1984). Blancan-
an ocean drilling perspective. Proceedings of the Ocean Drilling Hemphillian land mammal ages and late Cenozoic mammalian
Program, 113, Part B, 937--60. dispersal events. Annual Reviews of Earth and Planetary Science,
Kennett, J. P., Houtz, R. E., Andrews, P. B., Edwards, A. R., Goslin, V. 12, 445-88.
A., Hahos, M., Hampton, M.A., Jenkins, D. G., Margolis, S. V., Lindsay, E. H., Opdyke, N. D., Johnson, N. M., & Butler, R. F. (1987).
Ovenshine, A. T., & Perch-Nielsen, K. (1975). Cenozoic paleo- Mammalian chronology and the magnetic polarity time scale. In
ceanography in the southwest Pacific Ocean, Antarctic glaciation Cenozoic Mammals of North America, ed. M. 0. Woodburne,
and the development of the circum-Antarctic current. Initial Reports pp. 269-84. Berkeley: University of California Press.
of the Deep Sea Drilling Project, 29, 1155-69. Lipman, P. W., Prostka, H. J., & Christiansen, R. L. (1972). Cenozoic
Kennett, J. P., & the Leg 21 Shipboard Party. (1972). Australian-Antarctic volcanism and plate tectonic evolution of the western United States.
continental drift, paleocirculation changes, and Oligocene deep-sea I. Early and middle Cenozoic. Philosophical Transactions of the
erosion. Nature (Physical Sciences), 239, 51-5. Royal Society of London, Series A, 271, 217-48.
Kennett, J.P., & Stott, L. D. (1990). Proteus and Proto-Oceanus: ancestral Lohmar, J. M., & Warme, J. E. (1979). An Eocene shelf margin: San
Paleogene oceans as revealed from Antarctic stable isotopic results, Diego County, California. In Cenozoic Paleogeography of the
 Donald R. Prothero 32

Western United States, eds. J. M. Armentrout, M. R. Cole, & H. McDougall, I., & Harrison, T. M. (1988). Geochronology and Ther-
TerBest. Paci.fie Section, Society of Economic Paleontologists and mochronology by the 40 Arl3 9Ar Method. New York: Oxford
Mineralogists Symposium, 3, 165-76. University Press.
Loomis, D. P., & Burbank, D. W. (1988). The stratigraphic evolution of the McGowran, B. (1973). Observation Borehole No. 2, Gambier Embayment
El Paso Basin, southern California: implications for the Miocene of the Otway Basin: Tertiary micropaleontology and stratigraphy.
development of the Garlock Fault and the uplift of the Sierra Nevada. South Australian Department of Mines and Mineral Resources
Geological Society of America Bulletin, 100, 12-28. Review, 135, 43-55.
Loutit, T. S., & Kennett, J. P. (1981). New Zealand and Australian Ceno- (1989). Silica burp in the Eocene ocean. Geology, 17, 857-60.
zoic sedimentary cycles and global sea level changes. American McGowran, B., Moss, G., & Beecroft, A. (1992). Late Eocene and
Association of Petroleum Geologists Bulletin, 65, 1586-1601. early Oligocene in southern Australia: local neritic signals of
Lucas, S. G. (1984). Systematics, Biostratigraphy, and Evolution of Early global oceanic changes. In Eocene-Oligocene Climatic and Biotic
Cenozoic Coryphodon (Mammalia, Pantodonta). Unpublished Evolution, eds. D. R. Prothero & W. A. Berggren, pp. 178-201.
Ph.D. Thesis. New Haven: Yale University. Princeton: Princeton University Press.
(1989). Fossil mammals and the Paleocene-Eocene boundary in Europe, McKee, E. D., & McKee, E. H. (1972). Pliocene uplift of the Grand
North America, and Asia. Abstracts, 28th International Geological Canyon region: time of drainage adjustments. Geological Society
Congress, 2, 355. of America Bulletin, 83, 1923-32.
(1992). Redefinition of the Duchesnean land mammal "age," late McKenna, M. C. (1980). Remaining evidence of Oligocene sedimentary
Eocene of western North America. In Eocene-Oligocene Climatic rocks previously present across the Bighorn Basin, Wyoming.
and Biotic Evolution, eds. D. R. Prothero & W. A. Berggren, pp. University of Michigan Papers on Paleontology, 24, 143--6.
88-105. Princeton: Princeton University Press. McKenna, M. C., & Love, J. D. (1972). High-level strata containing early
(1993). Mammalian biochronology of the Paleocene-Eocene bound- Miocene mammals on the Bighorn Mountains, Wyoming. American
ary in North America, Europe, and Asia. Journal of Vertebrate Museum Novitates, 2490, 1-31.
Paleontology, 13, suppl. to no. 3, p. 47A. Mein, P. (1979). Rapport d'activite du groupe de travail vertebres mise a
Lundelius, E. L., Jr., Churcher, C. S., Downs, T., Harington, C.R., Lindsay, jour de la biostratigraphie du Neogene basee sur Jes mammireres.
E. H., Schultz, G. E., Semken, H. A., Webb, S. D., & Zakrzewski, R. Anna/es Geologiques du Pays Hellenique, 1979 pp. 1367-72.
J. (1987). The North American Quaternary sequence. In Cenozoic Miller, K. G. (1992). Middle Eocene to Oligocene stable isotopes, climate
Mammals of North America, ed. M. 0. Woodburne, pp. 211-35. and deep-water history: the Terminal Eocene Event? In Eocene-
Berkeley: University of California Press. Oligocene Climatic and Biotic Evolution, eds. D. R. Prothero & W.
Luyendyk, B. P. (1991 ). A model for Neogene crustal rotations, transtension, A. Berggren, pp. 160-77. Princeton: Princeton University Press.
and transpression in southern California. Geological Society of Miller, K. G., & Curry, W. B. (1982). Eocene to Oligocene benthic
America Bulletin, 103, 1528-36. foraminiferal isotopic record of the Bay of Biscay. Nature, 296,
Luyendyk, B. P., Kamerling, M. J., & Terres, R.R. (1980). Geometric model 347-50.
for Neogene crustal rotations in southern California. Geological Miller, K. G., & Fairbanks, R. G. (1983). Evidence for Oligocene-Middle
Society of America Bulletin, 91, 211-17. Miocene abyssal circulation changes in the western North Atlantic.
Lyell, C. (1831-1833). Principles of Geology (lst ed.). 3 vols. London: Nature, 306, 250-3.
John Murray. (1985). Oligocene to Miocene global carbon isotope cycles and abyssal
MacFadden, B. J. (1977). Magnetic polarity stratigraphy of the Chamita circulation changes. In The Carbon Cycle and Atmospheric C02:
Formation stratotype (Mio-Pliocene) of north-central New Mexico. Natural Variations Archean to Present, eds. E. T. Sundquist & W. S.
American Journal of Science, 277, 769-800. Broecker. American Geophysical Union, Geophysical Monograph,
(1992). Fossil Horses: Systematics, Paleobiology and Evolution of the 32, 469-86.
Family Equidae. Cambridge: Cambridge University Press. Miller, K. G., Fairbanks, R. G., & Mountain, G. S. (1987). Tertiary oxygen
MacFadden, B. J., Johnson, N. M., & Opdyke, N. D. (1979). Magnetic isotope synthesis, sea level history, and continental margin erosion.
polarity stratigraphy of the Mio-Pliocene mammal-bearing Big Paleoceanography, 2, 1-19.
Sandy Formation of western Arizona. Earth and Planetary Sciences Miller, K. G., & Thomas, E. (1985). Late Eocene to Oligocene benthic
Letters, 44, 349--64. foraminiferal isotopic record, Site 574, equatorial Pacific. Initial
MacFadden, B. J., Swisher, C. C., ill, Opdyke, N. D., & Woodburne, M. Reports of the Deep Sea Drilling Project, 85, 771-7.
0. (1990). Paleomagnetism, geochronology, and possible tectonic Miller, K. G., & Tucholke, B. E. (1983). Development of Cenozoic abyssal
rotation of the middle Miocene Barstow Formation, Mojave Desert, circulation south of the Greenland-Scotland Ridge. In Structure
southern California. Geological Society of America Bulletin, 102, and Development of the Greenland-Scotland Ridge, eds. M. H. P.
478-93. Bott, S. Saxov, M. Talwani, & J. Thiede, pp. 549-89. New York:
Magill, J., & Cox, A. V. (1980). Tectonic rotation of the Oregon western Plenum.
Cascades. Oregon Department of Geology and Mineral Industries Miller, K. G., Wright, J. D., & Fairbanks, R. G. (1991). Unlocking the Ice
Special Paper, 10, 1-67. House: Oligocene-Miocene oxygen isotopes, eustasy, and margin
Margolis, S. V., & Kennett, J. P. (1971). Cenozoic paleoglacial history erosion. Journal of Geophysical Research, 96, 6829-48.
of Antarctica recorded in subantarctic deep-sea cores. American Montanari, A. (1988). Geochemical characterization of volcanic biotites
Journal of Science, 271, 1-36. from the upper Eocene - upper Miocene pelagic sequence of the
Matthew, W. D. (1899). A provisional classification of the fresh-water northeastern Apennines. In The Eocene - Oligocene Boundary in the
Tertiary of the West. Bulletin of the American Museum of Natural Marche-Umbria Basin (Italy), eds. I. Premoli-Silva, R. Coccioni,
History, 12, 19-75. & A. Montanari, pp. 209-27. International Subcommission' on
McCarroll, S. M., Flynn, J. J., & Turnbull, W. D. (1996). A review Paleogene Stratigraphy, Eocene/Oligocene Boundary Meeting,
of the biostratigraphic and magnetic polarity correlations of the Ancona, Oct. 1987, Special Publication.
Washakie Formation, Washakie Basin, Wyoming. In The Terrestrial (1990). Geochronology of the terminal Eocene impacts; an update.
Eocene-Oligocene Transition in North America, eds. D.R. Prothero Geological Society of America Special Paper, 247, 607-16.
& R. J. Emry, pp. 25-39. Cambridge: Cambridge University Montanari, A., Deino, A. L., Drake, R. E., Turrin, B. D., DePaolo, D.
Press. J., Odin, G. S., Curtis, G. H., Alvarez, W., & Bice, D. (1988).
 Background to North American mammalian evolution 33

Radioisotopic dating of the Eocene-Oligocene boundary in Chemical Geology (Isotope Geoscience Section), 86, 203-24.
the pelagic sequences of the northeastern Apennines. In The Oldow, J. S., Bally, A. W., Ave Lallemant, H. G., & Leeman, W. P. (1989).
Eocene/Oligocene Boundary in the Marche-Umbria Basin (Italy), Phanerozoic evolution of the North American Cordillera: United
eds. I. Premoli-Silva, R. Coccioni, & A. Montanari, pp. 195- States and Canada. In The Geology of North America-An Overview,
208. International Subcommission on Paleogene Stratigraphy, eds. A. W. Bally & A. R. Palmer, pp. 139-232. Boulder, Colo.:
Eocene/Oligocene Boundary Meeting, Ancona, Oct. 1987, Special Geological Society of America.
Publication. Olsson, R. K., Gibson, T. G., Hansen, H.J., & Owens, J.P. (1988). Geology
Montanari, A., Drake, R., Bice, D. M., Alvarez, W., Curtis, G. H., Turrin, of the northern Atlantic coastal plain: Long Island to Virginia. In
B., & DePaolo, D. J. (1985). Radiometric time scale for the upper The Atlantic Continental Margin, eds. R. E. Sheridan & J. A. Grow,
Eocene and Oligocene based on Kl Ar and Rb/Sr dating of volcanic The Geology of North America, pp. 87-106. Vol. I-2. Boulder,
biotites. Geology, 13, 596-9. Colo.: Geological Society of America.
Mountain, G. S., & Tucholke, B. E. (1985). Mesozoic and Cenozoic geology Olsson, R. K., Miller, K. G., & Ungrady, T. E. (1980). Late Oligocene
of the United States Atlantic continental slope and rise. In Geologic transgression of middle Atlantic coastal plain. Geology, 8, 549-54.
Evolution of the United States Atlantic Margin, ed. C. W. Poag, Opdyke, N. D. (1990). Magnetic stratigraphy of Cenozoic terrestrial sedi-
pp. 292-341. New York: Van Nostrand Reinhold. ments and mammalian dispersal. Journal of Geology, 98, 621-37.
Murphy, M. G., & Kennett, J.P. (1986). Development of latitudinal thermal Opdyke, N. D., Lindsay, E. H., Johnson, N. M., & Downs, T. (1977).
gradients during the Oligocene: oxygen-isotope evidence from the The paleomagnetism and magnetic polarity stratigraphy of the
southwest Pacific. Initial Reports of the Deep Sea Drilling Project, mammal-bearing section of Anza-Borrego State Park, California.
90, 1347-60. Quaternary Research, 7, 316-29.
Mutter, J. C., Hegarty, K. A., Cande, S. C., & Weisse!, J. K. (1985). Breakup Osborn, H.F. (1907). Tertiary mammal horizons of North America. Bulletin
between Australia and Antarctica: a brief review in light of new of the American Museum of Natural History, 23, 237-54.
data. Tectonophysics, 114, 255-79. ( 1910). The Age of Mammals in Europe, Asia, and North America. New
Neville, C., Opdyke, N. D., Lindsay, E. H., & Johnson, N. M. (1979). York: Macmillan.
Magnetic stratigraphy of Pliocene deposits of the Glenns Ferry For- (1929). The titanotheres of ancient Wyoming, Dakota, and Nebraska.
mation, Idaho, and its implications for North American mammalian United States Geological Survey Monograph, 55, 1-953.
biostratigraphy. American Journal of Science, 279, 503-26. Osborn, H. F., & Matthew, W. D. (1909). Cenozoic mammal horizons of
Nilsen, T. (1977). Early Tertiary tectonics and sedimentation in California. western North America. United States Geological Survey Bulletin,
In Late Mesozoic and Cenozoic Sedimentation and Tectonics in 361, 1-138.
California, ed. T. H. Nilsen, pp. 86-98. Bakersfield, Calif.: San Patton, T. H. (1969). Miocene and Pliocene artiodactyls, Texas Gulf Coastal
Joaquin Valley Geological Society Short Course. Plain. Florida State Museum Bulletin, Biological Sciences, 14,
(1987). Paleogene tectonics and sedimentation of coastal California. 115-226.
In Cenozoic Basin Development of Coastal California, eds. R. Peterson, 0. A. (1909). A revision of the Entelodontidae. Memoirs of the
V. Ingersoll & W. G. Ernst, pp. 81-123. Englewood Cliffs, N.J.: Carnegie Museum, 9, 41-158.
Prentice Hall. Poag, C. W., & Schlee, J. S. (1984). Depositional sequences and strati-
Nilsen, T., & McKee, E. H. (1979). Paleogene paleogeography of the graphic gaps on submerged United States Atlantic margin. American
western United States. In Cenozoic Paleogeography of the Western Association of Petroleum Geologists Memoir, 36, 165-82.
United States, eds. J.M. Armentrout, M. R. Cole, & H. TerBest. Pa- Poag, C. W., & Ward, L. W. (1987). Cenozoic unconformities and de-
cific Section, Society of Economic Paleontologists and Mineralogists positional supersequences of North Atlantic continental margins:
Symposium, 3, 257-76. testing the Vail model. Geology, 15, 159-62.
Nolan, T. B. (1943). The Basin and Range province in Utah, Nevada, and Pomerol, C. & Premoli-Silva, I. (eds.). (1986). Terminal Eocene Events.
California. United States Geological Survey Professional Paper, Amsterdam: Elsevier.
197, 141-93. Premoli-Silva, I., Coccioni, R., & Montanari, A. (eds.). (1988). The
North American Commission on Stratigraphic Nomenclature. ( 1983). North Eocene/Oligocene Boundary in the Marche-Umbria Basin
American stratigraphic code. American Association of Petroleum (Italy). International Subcommission on Paleogene Stratigraphy,
Geologists Bulletin, 67, 841-75. Eocene/Oligocene Boundary Meeting, Ancona, Oct. 1987, Special
Obradovich, J. D. (1988). A different perspective on glauconite as a Publication.
chronometer for geologic time scale studies. Paleoceanography, 3, Proffett, J. M., Jr. (1977). Cenozoic geology of the Yerington District,
757-70. Nevada, and implications for the nature and origin of Basin and
Odin, G. S. (1978). Isotopic dates for the Paleogene time scale. Ameri- Range faulting. Geological Society of America Bulletin, 88, 247-66.
can Association of Petroleum Geologists Studies in Geology, 6, Prothero, D.R. (1982). Medial Oligocene Magnetostratigraphy and Mam-
247-57. malian Biostratigraphy: Testing the Isochroneity of Mammalian
(1982). Numerical Dating in Stratigraphy. New York: Wiley. Biostratigraphic Events. Unpublished Ph.D. Thesis. New York:
Odin, G. S., & Curry, D. (1985). The Paleogene time-scale: radiometric Columbia University.
dating versus magnetostratigraphic approach. Journal of the (1984). Magnetostratigraphy of the early Oligocene Pipestone Springs
Geological Society of London, 142, 1179-88. locality, Jefferson County, Montana. Contributions to Geology,
Odin, G., Guise, P., Rex, D. C., & Kreuzer, H. (1988). K-Ar and 39Arf40Ar University of Wyoming, 23, 33-6.
geochronology of late Eocene biotites from the northeastern (1985a). Chadronian (early Oligocene) magnetostratigraphy of eastern
Apennines. In The Eocene/Oligocene Boundary in the Marche- Wyoming: implications for the Eocene-Oligocene boundary.
Umbria Basin (Italy), eds. I. Premoli-Silva, R. Coccioni, & A. Journal of Geology, 93, 555-65.
Montanari, pp. 239-45. International Subcommission on Paleogene (l 985b). Correlation of the White River Group by magnetostratigraphy.
Stratigraphy, Eocene/Oligocene Boundary Meeting, Ancona, Oct. In Fossiliferous Cenozoic Deposits of Western South Dakota and
I987, Special Publication. Northwestern Nebraska, ed. J. E. Martin. Dakoterra, Museum of
Odin, G. S., Montanari, A., Deino, A., Drake, R., Guise, P. G., Kreuzer, Geology, South Dakota School of Mines, 2, 265-76.
H., & Rex, D. C. (1991). Reliability of volcano-sedimentary biotite (1988). Mammals and magnetostratigraphy. Journal of Geological
ages across the Eocene-Oligocene boundary (Apennines, Italy). Education, 34, 227-36.
 Donald R. Prothero 34

(1990). Interpreting the Stratigraphic Record. New York: Freeman. Geology, 5, 336-40.
(1991). Magnetic stratigraphy of Eocene-Oligocene mammal localities Rapp, S. D., MacFadden, B. J., & Schiebout, J. A. (1983). Magnetic polarity
in southern San Diego County. In Eocene Geologic History, San stratigraphy of the early Tertiary Blacks Peak Formation, Big Bend
Diego Region, eds. P. L. Abbott & J. A. May, pp. 125-30. Pacific National Park, Texas. Journal of Geology, 91, 555-72.
Section, Society of Economic Paleontologists and Mineralogists Rea, D. K., Zachos, J. C., Owen, R. M., & Gingerich, P. D. (1990).
Guidebook, 68. Global change at the Paleocene-Eocene boundary: climatic and
(1994a). The Eocene-Oligocene Transition: Paradise Lost. New York: evolutionary consequences of tectonic events. Palaeogeography,
Columbia University Press. Palaeoclimatology, Palaeoecology, 79, 117-28.
(1994b). The late Eocene-Oligocene extinctions. Annual Review of Rensberger, J.M. ( 1971 ). Entoptychine pocket gophers (Mammalia, Geomy-
Earth and Planetary Sciences, 22, 145-65. oidea) of the early Miocene John Day Formation, Oregon. University
(1996a). Magnetostratigraphy and biostratigraphy of the middle Eocene of California Publications in Geological Sciences, 90, 1-163.
Uinta Formation, Uinta Basin, Utah. In The Terrestrial Eocene- (1973). Pleurolicine rodents (Geomyoidea) of the John Day Formation,
Oligocene Transition in North America, eds. D.R. Prothero & R. J. Oregon, and their relationships to taxa from the early and middle
Emry, pp. 3-24. Cambridge: Cambridge University Press. Miocene of South Dakota. University of California Publications in
( l 996b ). Magnetostratigraphy of Eocene-Oligocene transition in Geological Sciences, 102, 1-95.
Trans-Pecos Texas. In The Terrestrial Eocene-Oligocene Tran- (1983). Successions of meniscomyine and allomyine rodents (Aplodon-
sition in North America, eds. D. R. Prothero & R. J. Emry, tidae) in the Oligo-Miocene John Day Formation, Oregon. University
pp. 189-98. Cambridge: Cambridge University Press. of California Publications in Geological Sciences, 124, 1-157.
(1996c). Magnetic stratigraphy of the White River Group, High Repenning, C. A. (1987). Biochronology of the microtine rodents of the
Plains of North America. In The Terrestrial Eocene-Oligocene United States. In Cenozoic Mammals of North America, ed. M.
Transition in North America, eds. D. R. Prothero & R. J. Emry, 0. Woodbume, pp. 235-68. Berkeley: University of California
pp. 262-77. Cambridge: Cambridge University Press. Press.
Prothero, D. R., Denham, C. R., & Farmer, H. G. (1983). Magne- Retallack, G. J. (1981). Preliminary observations on fossil soils in the
tostratigraphy of the White River Group and its implications for Clarno Formation (Eocene to early Oligocene), near Clarno, Oregon.
Oligocene geochronology. Palaeogeography, Palaeoclimatology, Oregon Geology, 43, 147-50.
Palaeoecology, 2, 151-66. (1983a). Late Eocene and Oligocene paleosols from Badlands National
Prothero, D. R., & Emry, R. J. (eds.). (1996a). The Terrestrial Eocene- Park, South Dakota. Geological Society of America Special Paper,
Oligocene Transition in North America. Cambridge: Cambridge 193, 1-82.
University Press. (1983b). A paleopedological approach to the interpretation of terrestrial
(1996b). Summary. In The Terrestrial Eocene-Oligocene Transition sedimentary rocks: the mid-Tertiary fossil soils of Badlands Na-
in North America, eds. D. R. Prothero & R. J. Emry, pp. 664-82. tional Park, South Dakota. Geological Society of America Bulletin,
Cambridge: Cambridge University Press. 94, 823-40.
Prothero, D. R., Howard, J., & Dozier, T. H. H. (1996). Stratigraphy and (1990). Soils of the Past: An Introduction to Paleopedology. London:
paleomagnetism of the upper middle Eocene to lower Miocene Unwin-Hyman.
(Uintan-Arikareean) Sespe Formation, Ventura County, California. (1992). Paleosols and changes in climate and vegetation across the
In The Terrestrial Eocene-Oligocene Transition in North Amer- Eocene/Oligocene boundary. In Eocene-Oligocene Climatic and
ica, eds. D. R. Prothero & R. J. Emry, pp. 171-88. Cambridge: Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 382-98.
Cambridge University Press. Princeton: Princeton University Press.
Prothero, D. R., & Lucas, S. G. (1996). Magnetostratigraphy of the Rose, K. D. (1981). The Clarkforkian land-mammal age and mam-
Duchesnean portion of the Galisteo Formation, New Mexico. In The malian fauna! composition across the Paleocene-Eocene boundary.
Terrestrial Eocene-Oligocene Transition in North America, eds. D. University of Michigan Papers on Paleontology, 26, 1-197.
R. Prothero & R. J. Emry, pp. 199-205. Cambridge: Cambridge Rudwick, M. J. S. ( 1978). Charles Lyell's dream ofa statistical paleontology.
University Press. Palaeontology, 21, 225-44.
Prothero, D. R., & Rensberger, J. M. (1985). Magnetostratigraphy of the Russell, D. E. (1964). Les mammireres Paleocene d'Europe. Musee
John Day Formation, Oregon, and the North American Oligocene- National d'Histoire Nature/le Memoires, 13, 1-324.
Miocene boundary. Newsletters in Stratigraphy, 15, 59-70. Ryder, R. T., Fouch, T. D., & Elison, J. H. (1976). Early Tertiary sedimen-
Prothero, D. R., & Sereno, P. C. (1982). Allometry and paleoecology of tation in the western Uinta Basin. Geological Society of America
medial Miocene dwarf rhinoceroses from the Texas Gulf Coastal Bulletin, 87, 496-512.
Plain. Paleobiology, 8, 16-30. Savage, D. E. (1955). Nonmarine lower Pliocene sediments in California,
Prothero, D. R., & Swisher, C. C., III. (1992). Magnetostratigraphy and geochronologic-stratigraphic classification. University of California
geochronology of the terrestrial Eocene-Oligocene transition in Publications in Geological Sciences, 31, 1-26.
North America. In Eocene-Oligocene Climatic and Biotic Evolu- (1962). Cenozoic geochronology of the fossil mammals of the Western
tion, eds. D.R. Prothero & W. A. Berggren, pp. 46-73. Princeton: Hemisphere. Revista Museo Argentina Ciencias Natura/es, 8,
Princeton University Press. 53-67.
Prothero, D. R., & Whittlesey, K. E. (1998). Magnetic stratigraphy and (1977). Aspects of vertebrate paleontological stratigraphy and
biostratigraphy of the Orellan and Whitneyan land mammal "ages" geochronology. In Concepts and Methods in Biostratigraphy, eds.
in the White River Group. Geological Society of America Special E. G. Kaufman & J. E. Hazel, pp. 427-42. Stroudsburg, Penn.:
Paper (in press). Dowden, Hutchinson, & Ross.
Quade, J., & Cerling, T. E. (1995). Expansion of C4 grasslands in the late Savage, D. E., & Russell, D. E. (1983). Mammalian Paleofaunas of the
Miocene of northern Pakistan: evidence from stable isotopes in World. Reading, Mass.: Addison Wesley.
paleosols. Palaeogeography, Palaeoclimatology, Palaeoecology, Schankler, D. M. (1980). Fauna! zonation of the Willwood Formation in the
115, 91-116. central Bighorn Basin, Wyoming. University of Michigan Papers on
Quade, J., Ceding, T. E., & Bowman, J. R. (1989). Development of Asian Paleontology, 24, 99-114.
monsoon revealed by marked ecological shift during the latest (1981). Local extinction and ecological re-entry of early Eocene
Miocene in northern Pakistan. Nature, 342, 163-5. mammals. Nature, 293, 135-8.
Quilty, P. G. (1977). Cenozoic sedimentation cycles in western Australia. Schlich, R., & the Leg 120 Shipboard Party. (1989). Proceedings of the
 Background to North American mammalian evolution 35

Ocean Drilling Project, Initial Reports, 120. America Bulletin, 87, 297-309.
Schnitker, D. (1980). Quaternary deep-sea benthic foraminifers and Stehlin, H. G. (1909). Remarques sur !es faunules de mammireres des
bottom-water masses. Annual Reviews of Earth and Planetary couches eocenes et oligocenes du Bassin de Paris. Bulletin Societe
Sciences, 8, 343-70. Geologique de France, 9, 488-520.
Schoch, R. M. (1985). Preliminary description of a new late Paleocene Stenzel, H.B., Turner, F. E., & Hesse, C. J. ( 1944). Brackish and non-marine
land-mammal fauna from South Carolina, U.S.A. Postilla, Yale Miocene in southeastern Texas. American Association of Petroleum
Peabody Museum, 196, 1-13. Geologists Bulletin, 28, 977-1011.
Schorn, H. (1971 ). What is type Paleocene? American Journal of Science, Stewart, J. H. (1971). Basin and Range structure - a system of horsts and
271, 402-9. grabens produced by deep-seated extension. Geological Society of
Schultz, C. B., & Stout, T. M. (1955). Classification of the Oligocene America Bulletin, 82, 1019-44.
sediments in Nebraska. Bulletin of the University of Nebraska State Stucky, R. K. (1984). The Wasatchian-Bridgerian land mammal age
Museum, 4, 17-52. boundary (early to middle Eocene) in western North America.
(1961 ). Field conference on the Tertiary and Pleistocene of western Annals of the Carnegie Museum, 53, 347-82.
Nebraska. Special Publication of the University of Nebraska State Suppe, J., Powell, C., & Berry, R. (1975). Regional topography, seismicity,
Museum, 2, 1-54. Quaternary volcanism, and the present-day tectonics of the western
Sclater, J. G., Meinke, L., Bennett, A., & Murphy, C. (1986). The depth United States. American Journal of Science, 275A, 397-436.
of the ocean through the Neogene. Geological Society of America Swinehart, J. B., Souders, V. L., Degraw, H. M., & Diffendal, R. F., Jr.
Memoir, 163, 1-19. (1985). Cenozoic paleogeography of western Nebraska. In Cenozoic
Scott, G. R. (1975). Cenozoic surfaces and deposits of the southern Rocky Paleogeography of the West Central United States, eds. R. M. Flores
Mountains. Geological Society of America Memoir, 144, 227-48. & S. Kaplan, pp. 209-29. Special Publication of the Rocky Mountain
Scott, W. B. (1945). The Mammalia of the Duchesne River Oligocene. Section, Society of Economic Paleontologists and Mineralogists.
Transactions of the American Philosophical Society, 34, 209-53. Swisher, C. C., III, & Prothero, D.R. (1990). Single-crystal 40Art39Ar dating
Seeland, D. (1985). Oligocene paleogeography of the northern Great Plains of the Eocene-Oligocene transition in North America. Science, 249,
and adjacent mountains. In Cenozoic Paleogeography of the West 760-2.
Central United States, eds. R. M. Flores & S. Kaplan, pp. 187-206. Tabrum, A. R., Prothero, D.R., & Garcia, D. (1996). Magnetostratigraphy
Special Publication of the Rocky Mountain Section, Society of and biostratigraphy of the Eocene-Oligocene transition in western
Economic Paleontologists and Mineralogists. Montana. In The Terrestrial Eocene-Oligocene Transition in North
Severinghaus, J., & Atwater, T. (1990). Cenozoic geometry and thermal America, eds. D.R. Prothero& R. J. Emry, pp. 278-311. Cambridge:
state of the subducting slabs beneath western North America. Cambridge University Press.
Geological Society of America Memoir, 176, 1-22. Talwani, M., & Eldholm, 0. (1977). Evolution of the Norwegian-Greenland
Shackleton, N. J., & Kennett, J.P. (1975). Paleotemperature history of the Sea. Geological Society of America Bulletin, 88, 969-99.
Cenozoic and initiation of Antarctic glaciation: oxygen and carbon Taylor, G. E. (1983). Braided-river and flood-related deposits of the
isotopic analyses in DSDP Sites 277, 279, and 281. Initial Reports Sespe Formation, northern Simi Valley, California. Pacific Section
of the Deep Sea Drilling Project, 29, 743-55. Society of Economic Paleontologists and Mineralogists Guidebook,
Shackleton, N. J., & Opdyke, N. D. (1973). Oxygen isotope and paleo- pp. 129-40.
magnetic stratigraphy of Equatorial Pacific core V28-238: oxygen Taylor, L. H., & Butler, R. F. (1980). Magnetic polarity stratigraphy of
isotope temperatures and ice volumes on a 105 year and 106 year Torrejonian sediments, Nacimiento Formation, San Juan Basin,
scale. Quaternary Research, 3, 39-55. New Mexico. American Journal of Science, 280, 97-115.
( 1977). Oxygen isotope and palaeomagnetic evidence for early Northern Tedford, R. H. (1970). Principles and practices of mammalian geochronol-
Hemisphere glaciation. Nature, 270, 216-19. ogy in North America. Proceedings of the North American
Shaw, A. B. (1964). Time in Stratigraphy. New York: McGraw-Hill. Paleontological Convention, 2F, 666-703.
Siesser, W. G., & Dingle, R. V. (1981). Tertiary sea-level movements around Tedford, R. H., Galusha, T., Skinner, M. F., Taylor, B. E., Fields, R. W.,
southern Africa. Journal of Geology, 89, 83-96. Macdonald, J. R., Rensberger, J. M., Webb, S. D., & Whistler, D.
Simpson, G. G. (1946). The Duchesnean fauna and the Eocene-Oligocene P. (1987). Fauna! succession and biochronology of the Arikareean
boundary. American Journal of Science, 244, 52-7. through Hemphillian interval (late Oligocene through earliest
Simpson, R. W., & Cox, A. V. (1977). Paleomagnetic evidence for tectonic Pliocene Epochs) in North America. In Cenozoic Mammals of North
rotation of the Oregon Coast Range. Geology, 5, 585-9. America, ed. M. 0. Woodbume, pp. 152-210. Berkeley: University
Singh, G. (1988). History of aridland vegetation and climate: a global of California Press.
perspective. Biological Reviews, 63, 156-96. Tedford, R. H., & Hunter, M. E. (1984). Miocene marine-nonmarine
Skinner, M. F., Skinner, S. M., & Gooris, R. J. (1977). Stratigraphy and correlations. Palaeogeography, Palaeoclimatology, Palaeoecology,
biostratigraphy of late Cenozoic deposits in central Sioux County, 47, 129-51.
western Nebraska. Bulletin of the American Museum of Natural Tedford, R. H., Swinehart, J., Prothero, D. R., Swisher, C. C., King, S.
History, 158, 263-370. A., & Tierney, T. E. (1996). The Whitneyan-Arikareean transition
Smith, L. N., Lucas, S. G., & Elston, W. E. (1985). Paleocene stratigraphy, in the High Plains. In The Terrestrial Eocene-Oligocene Transition
sedimentation, and volcanism of New Mexico. In Cenozoic Paleo- in North America, eds. D. R. Prothero & R. J. Emry, pp. 312-34.
geography of the West Central United States, eds. R. M. Flores & Cambridge: Cambridge University Press.
S. Kaplan, pp. 293-316. Special Publication of the Rocky Mountain Thomasson, J. R. (1985). Miocene fossil grasses: possible adaptation in
Section, Society of Economic Paleontologists and Mineralo- reproduction bracts (lemma and palea). Annals of the Missouri
gists. Botanical Garden, 72, 843-51.
Snyder, W. S., Dickinson, W.R., & Silberman, M. L. (1976). Tectonic impli- Tomida, Y. (1981). "Dragonian" fossils from the San Juan Basin and the
cations of space-time patterns of Cenozoic magmatism in the western status of the "Dragonian" land mammal "age." In Advances in
United States. Earth and Planetary Sciences Letters, 32, 91-106. San Juan Basin Paleontology, eds. S. G. Lucas, J. K. Rigby, Jr., &
Solounias, N., & Quade, J. (1993). The savanna.myth of the Miocene. B. S. Kues, pp. 222-41. Albuquerque: University of New Mexico
Journal of Vertebrate Paleontology, 13, suppl. to no. 3, p. 58A. Press.
Stanley, K. 0. (1976). Sandstone petrofacies in the Cenozoic High Plains Tomida, Y., & Butler, R. F. (1980). Dragonian mammals and Paleocene
sequence, eastern Wyoming and Nebraska. Geological Society of magnetic polarity stratigraphy, North Hom Formation, central Utah.
 Donald R. Prothero 36

American Journal of Science, 280, 787-811. Williams, H. S. (1901). The discrimination of time-values in geology.
Tucholke, B., Vogt, P., & the Leg 43 Shipboard Party. (I 979). Initial Reports Journal of Geology, 9, 570-85.
of the Deep Sea Drilling Project, Leg 43. Wilson, J. A. (1978). Stratigraphic occurrence and correlation of early
Veevers, J. J. (1986). Breakup of Australia and Antarctica estimated as Tertiary vertebrate faunas, Trans-Pecos Texas, Part 1: Vieja area.
mid-Cretaceous (95 ± 5 Ma) from magnetic and seismic data at the Texas Memorial Museum Bulletin, 25, 1-42.
continental margin. Earth and Planetary Science Letters, 77, 91-9. (1984). Vertebrate fossil faunas 49 to 36 million years ago and additions
Walsh, S., Prothero, D. R. & Lundquist, S. ( 1996). Stratigraphy and paleo- to the species of Leptoreodon found in Texas. Journal of Vertebrate
magnetism of the middle Eocene Friars Formation and Poway Group, Paleontology, 4, 199-207.
southwestern San Diego County, California. In The Terrestrial (1986). Stratigraphic occurrence and correlation of early Tertiary
Eocene-Oligocene Transition in North America, eds. D.R. Prothero vertebrate faunas, Trans-Pecos Texas: Agua Fria-Green Valley
& R. J. Emry, pp. 120-54. Cambridge: Cambridge University Press. areas. Journal of Vertebrate Paleontology, 6, 350-73.
Walton, A. H. (1992). Magnetostratigraphy of the lower and middle Wing, S. L., & Bown, T. M. (1985). Fine-scale reconstruction of late
members of the Devil's Graveyard Formation (middle Eocene), Paleocene-early Eocene paleogeography in the Bighorn Basin of
Trans-Pecos Texas. In Eocene-Oligocene Climatic and Biotic northern Wyoming. In Cenozoic Paleogeography of the West Central
Evolution, eds. D. R. Prothero & W. A. Berggren, pp. 74-87. United States, eds. R. M. Flores & S. Kaplan, pp. 93-107. Special
Princeton: Princeton University Press. Publication of the Rocky Mountain Section, Society of Economic
Wang, Y., Ceding, T., & MacFadden, B. J. ( 1994 ). Fossil horses and carbon Paleontologists and Mineralogists.
isotopes: new evidence for Cenozoic dietary, habitat, and ecosystem Wing, S. L., Bown, T. M., & Obradovich, J. D. (1991). Early Eocene biotic
changes in North America. Palaeogeography, Palaeoclimatology, and climatic change in interior western North America. Geology,
Palaeoecology, 107, 269-79. 19, 1189-92.
Waters, A. (1961). Stratigraphic and lithologic variations in the Columbia Wise, S. W., Jr., Breza, J. R., Harwood, D. M., Wei, W., & Zachos, J. C.
River basalts. American Journal of Science, 259, 583-611. (1992). Paleogene glacial history of Antarctica in light of Leg 120
Webb, S. D. (1977). A history of savanna vertebrates in the New World, drilling results. Proceedings of the Ocean Drilling Program, 120,
Part I: North America. Annual Review of Ecology and Systematics, 1001-28.
8, 355-80. Wolfe, J. A. (1978). A paleobotanical interpretation of Tertiary climates in
(1983). The rise and fall of the late Miocene ungulate fauna in North the Northern Hemisphere. American Scientist, 66, 694-703.
America. In Coevolution, ed. M. D. Nitecki, pp. 267-306. Chicago: (1992). Climatic floristic and vegetational change near the
University of <'.:hi ca go Press. Eocene/Oligocene boundary in North America. In Eocene-
(1984). Ten million years of mammal extinctions in North America. Oligocene Climatic and Biotic Evolution, eds. D.R. Prothero & W.
In Quaternary Extinctions, eds. P. S. Martin & R. G. Klein, pp. A. Berggrenn, pp. 421-36. Princeton: Princeton University Press.
189-210. Tucson: University of Arizona Press. Wood, H. E. (1934 ). Revision of the Hyrachyidae. Bulletin of the American
Wei, W. (1989). Reevaluation of the Eocene ice-rafting record from Museum of Natural History, 67, 182-295.
subantarctic cores. Antarctic Journal of the United States, 1989 Wood, H. E., II, Chaney, R. W., Jr., Clark, J., Colbert, E. H., Jepsen, G. L.,
108-9. Reeside, J. B., & Stock, C. (1941). Nomenclature and correlation
Weisse!, J. K., Hayes, D. E., & Herron, E. M. (1977). Plate tectonics of the North American continental Tertiary. Geological Society of
synthesis: the displacements between Australia, New Zealand, and America Bulletin, 52, 1-48.
Antarctica since the late Cretaceous. Marine Geology, 25, 231-77. Woodburne, M. 0. (1977). Definition and characterization in mammalian
Wernicke, B., Axen, G. J., & Snow, J. K. (1988). Basin and range extensional chronostratigraphy. Journal of Paleontology, 5 I, 220-34.
tectonics at the latitude of Las Vegas, Nevada. Geological Society (ed.) (1987). Cenozoic Mammals of North America. Berkeley:
of America Bulletin, 100, I 738-57. University of California Press.
Westgate, J. W. (1986). Late Eocene land mammals from the Gulf Coastal Woodburne, M. 0., Tedford, R.H., & Swisher, C. C., III. (1990). Lithos-
Plain. In Montgomery Landing Site, Marine Eocene (Jackson) of tratigraphy, biostratigraphy, and geochronology of the Barstow
Central Louisiana, eds. J. A. Schiebout & W. Van Den Bold, pp. Formation, Mojave Desert, southern California. Geological Society
223-9. Gulf Coast Association of Geological Societies Symposium of America Bulletin, 102, 459-77.
Proceedings. Woodruff, F., Savin, S. M., & Douglas, R .G. (1981). Miocene stable isotope
Whistler, D. P. (1991). Geologic history of the EI Paso Mountains region. record: a detailed deep Pacific ocean study and its paleoclimatic
San Bernardino County Museum Association Quarterly, 38, 108-14. implications. Science, 212, 665-8.
Whistler, D. P., & Burbank, D. W. (1992). Miocene biostratigraphy and Zachos, J.C., Breza, J. R., & Wise, S. W. (1992). Early Oligocene ice-sheet
biochronology of the Dove Spring Formation, Mojave Desert, expansion on Antarctica: stable isotope and sedimentological
California, and the characterization of the Clarendonian land evidence from Kerguelen Plateau, southern Indian Ocean. Geology,
mammal age (late Miocene) in California. Geological Society of 20, 569-73.
America Bulletin, 104, 644-58.
2 Tertiary vegetation of North America as a context
for mammalian evolution

SCOTT L. WING

INTRODUCTION Where floristic data are the only type available for a particular time
and region I have included common names for the nonbotanical
This chapter is an attempt to summarize the Tertiary history of the reader.
vegetation of North America as a context for mammalian evolution. The problems with floristic analogy have been detailed by numer-
From the point of view of a mammalian paleontologist, and presum- ous recent authors, most persuasively by Wolfe ( 1979). These are the
ably from that of the other mammals they study, plants are important main points: (1) many taxonomic identifications of fossils are incor-
chiefly as sources of food and as partial creators of the habitats in rect, (2) evolution within lineages may have resulted in extinct and
which mammals live. Ideally, for any given time and place we would living species having different climatic tolerances, growth forms,
like to be able to describe the quantity and variety of food resources etc., and (3) extant taxa do not necessarily occupy the full range
provided by plants and also the aspects of vegetational structure that of climatic or vegetational situations they can tolerate because of
correspond to mammalian habitat types. What were the mechani- Pleistocene glaciation, anthropogenic disturbance, or other factors.
cal and chemical properties of food resources such as leaves and Any or all of these factors could result in a misinterpretation, and
fruits? How important were seasonal fluctuations in their availabil- points 1 and 2 apply with greater force to older fossil assemblages.
ity? What was the height of the vegetation? How dense was it? How The reliability of paleoclimatic inference based on floristic analogy
many species were in it? How many types of vegetation were in is thought to be increased by considering many species (e.g., Hickey,
the regional mosaic? How big were the patches? At a larger spatial 1977; Hickey et al., 1988) and by restricting analogies to taxa that
scale, to what degree was North American vegetation divided into are currently diverse, widespread, easily recognized by traits pre-
discrete vegetation types and floral provinces? Answering most of served in fossils, and relatively uniform in their climatic tolerances
these questions is difficult for most times and places in the Tertiary (Wing and Greenwood, 1993).
of North America because of the nature of the paleobotanical record. In response to the difficulties with floristic analogy, some workers
The Tertiary paleobotanical record consists largely of dissoci- have turned to physiognomic aspects of fossil assemblages to infer
ated leaves, fruits and seeds, and pollen. There is very limited direct vegetation and climate. These characteristics include the percent-
evidence for the shape and habit of the plants that produced these age of species with entire-margined leaves (Wolfe, 1979), as well
parts, their potential as mammalian food, or their seasonal pattern as the size of leaves and their apical, basal, and overall shape. In
of flowering, fruiting, and leafing. Much of what we know (or think living floras these features are highly correlated with temperature
we know) about Tertiary vegetation is based on analogy with living and rainfall (Wing and Greenwood, 1993; Wolfe, 1993; Gregory,
relatives. That is, species identified in the fossil record are assumed 1994), and thus with vegetation type. Although inferences about
to be similar to their closest living relatives in size and growth habit, paleoclimate and vegetation based on foliar physiognomy are gen-
and to have preferred similar climate and vegetation types. This is erally consistent with other lines of evidence, there are questions
why most descriptions of the past vegetation of North America are about the precision of the technique and its validity with older flo-
filled with the generic names of plants. To the extent possible I avoid ras. The physiological bases for correlations between leaf form and
floristic analogies in this summary, both because the paleozoolog- climate are poorly understood, as are the possible confounding ef-
ical audience presumably will not find these analogies particularly fects of soil type and phylogeny. An additional problem from the
meaningful and because there are persistent questions about the ac- point of view of the paleomammalogist is that foliar physiognomic
curacy of this method of vegetational and climatic reconstruction correlations only give information about climate. The habitats and
(see later). However, much information is lost if such information is resources available to mammals have to be inferred through an addi-
excluded entirely, especially when considering Neogene vegetation. tional step which requires us to assume that ancient vegetation was

37
 Scott L. Wing 38

similar in its structure and trophic resources to extant vegetation forests, but in the long term these deposits would likely be destroyed
growing in a similar climate. This assumption is tenuous for early by erosion. Even in known volcanic areas (e.g., Yellowstone, Challis
Tertiary vegetation that contained many extinct lineages, and also Field) most plant fossils appear to be derived from lowland vegeta-
for vegetation types without close living analogues, such as the mid- tion, rather than from the slopes, and this is particularly true of au-
latitude, subtropical forests of the Eocene. tochthonous assemblages (Fritz, 1986; Yuretich, 1984). At least one
An additional source of information about Tertiary vegetation major Tertiary assemblage of fruits and seeds, the middle Eocene
lies in the local distribution of fossil plants in the sediments. Un- Clarno flora of central Oregon, represents concentration and depo-
like vertebrate remains, most plant parts are quickly degraded if sition of plant debris in a volcanic mudflow and may contain a sub-
they are not buried by sediment. Plant fossils, particularly the com- stantial component of allochthonous remains (Manchester, 1994).
pressed leaves that make up the bulk of the Tertiary record in North Studies of recent volcanic eruptions have shown that in areas some
America, are not easily transported long distances, and consequently distance from the source, ash falls may preserve an accurate record of
much of the fossil record of plants is composed of samples that each pre-eruption forest vegetation in nonswamp settings (Burnham and
represent a small area and a short interval of time (e.g., Gastaldo, Spicer, 1986). At least theoretically, plant assemblages preserved
1988; Burnham, 1993a; Johnson, 1993). Although such samples by ash falls could retain physiognomic, floristic, and positional in-
represent a tiny proportion of the regional flora, they can preserve formation about nonswamp vegetation, and thus hold the greatest
some aspects of the local mosaic of source vegetation. Samples of hope for making reliable, detailed inferences about the vegetational
semiautochthonous to autochthonous plant assemblages should pro- context of fossil mammals. To date no studies have demonstrated
vide reasonably good estimates of the local diversity, species compo- this potential, although the Miocene ash deposits of the Great Plains
sition, and relative abundances of species in the source vegetation might provide a test case.
(Burnham, 1989a; Burnham, Wing, and Parker, 1992; Burnham, Fossil pollen provides another major source of data on the Tertiary
1993b). vegetation of North America. The great strength of the palynolog-
Closely spaced quarry sites in the same horizon can provide irtfor- ical record is its wide distribution and high stratigraphic density.
mation on small-scale heterogeneity in the composition and richness Palynological assemblages vary in the degree to which they repre-
of wet floodplain paleovegetation (e.g., Hickey, 1980, Wing, Alroy, sent regional as opposed to local vegetation. In general, pollen and
and Hickey, 1995), and may provide the basis for inferences about spores are derived from a wider area than megafloras from the same
the canopy size of some taxa (Burnham, l 993b ), although detailed depositional environment, but under forest canopies even pollen as-
lateral studies of this type are still rare. However, the autochthonous semblages are dominated by grains derived from vegetation no more
nature of these assemblages precludes their revealing much about than hundreds to thousands of meters away (Farley, 1989, 1990).
the vegetation of drier or better drained areas that were presumably The greatest weakness of the palynological record is that pollen and
the habitat of most mammals. The same constraints apply to plant spore taxa generally correspond to genera or families of plants, so
fossils derived from peats or coals, except that the vegetation of such changes in vegetational composition across space or through time
edaphically unusual habitats may be even less representative of the are blurred. The interpretation of palynological assemblages relies
region as a whole. Ironically, the sedimentary environments that almost entirely on floristic analogy at this high taxonomic level, there
preserve the most information about the distribution and abundance being no equivalent to leaf physiognomy for pollen. Nevertheless,
of plant species on ancient floodplains may record a habitat that was pollen assemblages are important for reconstructing the history of
little used by mammals and not similar to terra firma vegetation. North American vegetation because for many times and regions they
Plant fossil assemblages preserved in large lakes are another ma- are the only available source of information. Palynological data also
jor component of the Tertiary record of North American vegetation. are the main tool for resolving short-term or rapid changes in floras
Lakes receive some plant debris from inflowing streams as well as because megafloras seldom occur in sufficient stratigraphic density
from the lake margins, so they have the potential to represent sev- to resolve events that are less than tens or hundreds of thousands of
eral types of vegetation, although taphonomic studies show that few years apart.
plant parts are transported long distances, even in areas with steep
topography (Drake and Burrows, 1980; Spicer, 1981; Spicer and
Wolfe, 1987). Lacustrine assemblages may yield a fairly complete TERTIARY VEGETATIONAL HISTORY OF
list of the woody plant species occupying a local lake basin, but NORTH AMERICA
because all of the fossils have been transported to the bottom of the
lake, they preserve little information on the spatial mosaic of the The vegetation of North America has undergone dramatic changes
original vegetation. Inferences about the source vegetation for la- during the Tertiary in response to physiographic modification of
custrine assemblages come largely from physiognomic and floristic the continent, global climatic variation, and plant and animal evo-
analyses. lution. The most notable changes were the shift from broad-leaved
Volcanic source material increases the potential for silicification, evergreen to broad-leaved semideciduous or deciduous vegetation
but a great many volcaniclastic assemblages are still derived from during the middle Tertiary, the development of grasslands in the
wet fluvial or lacustrine settings. Air fall tuffs can preserve plant late Tertiary, and the increasing importance of evergreen conifer
material in situ, either in the form of forest floor litter or as standing forests, especially in northern and montane regions. Global cli-
trunk fields (e.g., Dorf, 1951, 1964; Burnham and Spicer, 1986). Ash matic cooling and widening of the North Atlantic during the
falls may entomb upland forests or the litter layer underneath such Tertiary also had a profound effect on floral composition by
 Tertiary vegetation of North America 39

reducing interchange across Beringia and the North Atlantic and of data points between regions. Nevertheless, some visual represen-
splitting the early Tertiary Boreotropical Flora into fairly distinct tation of vegetational change through time is desirable, and I hope
North American and Eurasian provinces (Wolfe, 1975; Tiffney, the maps will serve as a starting point for further refinements. Thero-
1985a, 1985b). tation of present coordinates to Tertiary positions was accomplished
No part of the continent escaped substantial modification of its with PGIS-Mac, which was also the source for paleogeography.
vegetational cover during the Tertiary, but the southeastern United
States, parts of the Pacific Coast, and some upland areas of Mexico
have been cited as areas with climates and floras somewhat similar PALEOCENE (SEE FIG. 2.1)
to those that were widespread in the earlier Tertiary (MacGinitie, East Coast
1969). Other areas, notably the Rocky Mountains and northern The only published Paleocene megafossil reports from the mid-
midcontinent, have been subject to the greatest climatic and/or phys- Atlantic Coast are of pinaceous seed cones from the nearshore
iographic change, and have also experienced the greatest vegeta- marine Aquia Formation of Virginia (Miller, 1977). Palynofloras
tional change (e.g., Leopold and Denton, 1987; Leopold, Liu, and derived from core and outcrop samples in Virginia (Frederiksen,
Clay-Poole, 1992). Thus the history of vegetational change in North l 979a; Gibson et al., 1980) and South Carolina (Frederiksen, l 980a)
America is in fact a collection of semi-independent, although linked, have been studied, although generally from a biostratigraphic rather
histories of separate areas. than floristic or paleoecological perspective. Lithologies and mi-
Our ability to reconstruct regional floras and vegetation is in di- crofossils of most samples suggest open shelf to nearshore marine
rect proportion to the number, geographic spread, and environmental deposition. Broadly speaking, these palynofloras are similar in com-
range of well-dated paleobotanical localities from a given time in- position to those of the Gulf Coast.
terval, and therefore poorly sampled intervals may falsely appear to
be times during which provincial differences were reduced. Because Mississippi Embayment and Gulf Coast
of the importance of semi-independent changes in different regions, Most of the Paleocene megafloral assemblages of the Mississippi
I have divided the discussion of each Tertiary epoch into sections Embayment and Gulf Coast area are from coastal or fluvio-deltaic
treating five regions: the East Coast, the Mississippi Embayment environments including lagoons, oxbow fills, and channel sands.
and Gulf Coast, the Great Plains and eastern Rocky Mountains, the These rocks are part of the Midway and Wilcox Groups (e.g., Berry,
western Rocky Mountains and Pacific Coast, and the Arctic region 1916a), which have more recently been divided into a series of for-
including Alaska. The quality and quantity of fossil assemblages mations, including the Naheola, Nanafalia, Tuscahoma, Naborton,
vary greatly among these regions. Porters Creek Clay, Rockdale, Seguin, and Clayton Formations
The East Coast of North America has a very poor Tertiary (Fairchild and Elsik, 1969; Tschudy, 1973). In many areas the fos-
megafossil record, although a number of palynofloras have been sil plants are found in clay lenses associated with lignite deposits
described. The Mississippi Embayment and Gulf Coast hold many that presumably formed in lowland swamp settings. Several assem-
Paleocene through Oligocene megafloral assemblages preserved in blages from near Naborton, Louisiana, are probably late Paleocene
clays and lignites, but these local deposits have proved difficult to (Tschudy, 1973, 1975). Wolfe (1978) interpreted these as being de-
correlate with one another, and the Neogene is poorly known. Pa- rived from tropical forest vegetation based on leaf size and percent-
leogene palynofloras of this area have also received considerable age of entire-margined forms. However, the assemblages are small
study. Like the Mississippi Embayment, the Great Plains and east- and poorly preserved, leaving some question about this vegetational
ern Rockies have produced many Paleocene and Eocene megafossil reconstruction.
assemblages. These derive from a fairly wide array of depositional There are a number of reliably dated Paleocene palynofloras
settings, and generally can be correlated through Ii tho- or biostratig- from the Mississippi Embayment, most of which appear to be from
raphy. In the later Eocene and Oligocene volcanic settings were im- the middle or latter part of the epoch (e.g., Elsik, 1968a, 1968b;
portant in preserving plant megafossils, but the region has far fewer Srivastava, 1972; Christopher et al., 1980). Pollen types similar to
Neogene assemblages. The most complete set of Tertiary megaflo- that of living chestnuts are common in many of these palynofloras,
ras in North America comes from the Pacific Northwest area of and diversity is generally lower than in subsequent Eocene paly-
the United States, where deposition in coastal and volcanic settings nofloras (Frederiksen, 1988, 1994). Diversity of Gulf Coast paly-
has preserved numerous floras of every Tertiary epoch except the nofloras increased slowly during the first half of the Paleocene,
Paleocene. Paleobotanical assemblages in Nevada and California then decreased sharply at the very end of the epoch (Frederiksen,
also are sparse in the Paleocene, but more common in the Eocene 1994).
through Pliocene. Floras from Arctic North America also span the
whole Tertiary, but with the bulk of the localities being from south- Great Plains and eastern Rocky Mountains
ern Alaska. The Fort Union Formation has produced megafloral and palynoflo-
The floras on which this summary is based are listed in the ral assemblages from hundreds oflocalities distributed across North
chapter appendix. The paleogeographic positions of the floras, and Dakota, Montana, and Wyoming (e.g., Dorf, 1940; Brown, 1962;
the inferred vegetational boundaries, are presented in reconstruc- Tidwell, Ash, and Parker, 1981; Hickey, 1977, 1980; Pocknall,
tions for each epoch (Figures 2.1-2.4 ). These paleovegetational 1987; McClammer and Crabtree, 1989; Crane, Manchester, and
maps are highly generalized, each being based on floras spanning Dilcher, 1990; Johnson and Hickey, 1990; Johnson, 1992). Pale-
millions of years, and with great variability in the quality and number ocene fossil localities in this region span the whole epoch, and
Scott L. Wing 40

r- -
-cP
r- oo a
cP
"'a

....
..,0
"':::>
UJ
j;;

n
 Tertiary vegetation of North America 41

Figure 2.1. Paleocene vegetation of North America. Numbers indicate the positions of Paleocene floras. Name, age, present coordinates, literature citation,
and inferred paleovegetation are given for each number in the chapter appendix. See text for details.

also provide the most detailed record of change in plant life across Western Rocky Mountains and Pacific Coast
the K!f boundary and during the earliest Paleocene. Paleocene for- Two major floras of the Pacific Northwest that were once thought
mations with broadly similar lithologies have also produced numer- to be Paleocene (Swauk and Chuckanut) more recently have been
ous assemblages in Alberta and Saskatchewan (Berry, 1930, 1935; assigned to the Eocene (Newman, 1981; Johnson, 1984), leaving
Bell, 1965; Christophel, 1976; Mclver and Basinger, 1993), and in little or no information about Paleocene vegetation in this area. In
Colorado and New Mexico (Knowlton, 1922, 1930; Brown, 1962; southern California, a latest Paleocene palynoflora and megaflora
Ash and Tidwell, 1976; Tidwell, Thayn, and Roth, 1976; Wolfe and have been recovered from the Silverado Formation (Gaponoff, 1984;
Upchurch, 1987). During the Paleocene much of the interior area Wing, unpublished data). The Paleocene latitude of this area may
east of the Rockies was occupied by wide basinal or marginal ma- have been as far south as southern Baja California. The stream-
rine lowlands; topography was low, drainage was poor, and swamps side elements of the flora are similar at a high taxonomic level
were widespread (Flores, 1981, 1983). The relatively homogeneous to roughly coeval floras from the Fort Union Formation (members
features of the preserved landscape may account in part for the ho- of the birch and witch hazel families predominate); taxodiaceous
mogeneity and low diversity of the floras. conifers are rare in the palynoflora and absent from the megaflora.
In the northern Rockies floral assemblages derived from carbona- The palynoflora is similar in composition to late Paleocene floras
ceous shales or elastic rocks associated with coals generally are dom- from the Gulf Coast area, containing a variety of forms belonging
inated by the deciduous bald-cypress family conifers (Taxodiaceae) to higher taxa that currently have temperate to tropical distributions
Glyptostrobus and Metasequoia. Swamp vegetation of this type ev- (Gaponoff, 1984). The Elsinore flora, also from the Silverado For-
idently existed almost to the shores of the Arctic Ocean (Sweet mation, but on the east side of the Santa Ana Mountains, has been
and Braman, 1992). Dicotyledons frequently were important though interpreted to represent a fairly moist savanna forest growing under
subdominant elements in such settings, and presumably formed an seasonal precipitation on a well-watered floodplain (Axelrod, 1979).
understory element in the forest. As far north as the U.S./Canadian This conclusion is based on the presence of fems, palms, notophyl-
border some of these di cot tree/shrubs, for instance members of the lous to mesophyllous dicot leaves, and clay minerals deemed to be
tea and laurel families, were probably evergreen, but deciduous di- the product of weathering under a seasonally dry, warm climate.
cots in the birch and witch hazel families were more common. The
herbaceous layer was composed largely of fems, large horsetails, The Arctic and Alaska
and ginger!ike monocots. Vegetation of alluvial ridges had more Late Paleocene megafloras have been described from the Chick-
species of dicqtyledonous trees and shrubs than did that of the back- aloon and Tolstoi formations of southern Alaska by Wolfe (1966,
swamps, many belonging to deciduous temperate zone families of 1977). The leaf physiognomy of these assemblages is similar to that
today such as the walnuts, birches, elms, sycamores, witch hazels, seen in the vegetation of some upland subtropical areas and north-
oaks, dogwoods, and katsura tree. In spite of some familial-level ern Florida, and probably reflects a mixed evergreen and deciduous
similarity between Paleocene and Recent floras it should be empha- broad-leaved forest. Two species of palm were recovered (Wolfe,
sized that most Paleocene species do not belong to extant genera, 1972). Based on the dominance of non-entire-margined leaves, mid-
and many may represent extinct families. Palms were rare through to late Paleocene floras from Greenland (Koch, 1963) may represent
most of the Paleocene in areas north of southern Wyoming and somewhat cooler climate and more deciduous forests than those of
were absent north of southern Montana (Brown, 1962). Megafloral southern Alaska, although this may reflect cooler global climate dur-
assemblages from southern Canada are generally similar to those ing the mid-Paleocene rather than a geographic difference between
from the northern United States, showing dominance by conifers coeval floras.
and a low diversity of deciduous dicotyledons (Christophel, 1976; Megafloral collections from the North Slope of Alaska (Spicer,
Mclver and Basinger, 1993). Wolfe, and Nichols, 1987) and Ellesmere Island (Hickey et al., 1983)
Recent work has pointed out the dissimilarity between floras of demonstrate that forests grew well within the Arctic Circle in the
the northern and southern Rockies in the Paleocene. Swamp assem- late Paleocene. The species present were entirely deciduous, and the
blages from northern New Mexico and southern Colorado generally diversity of the assemblages is low. These have been termed Polar
lacked taxodiaceous conifers; alluvial ridge vegetation was more Broad-leaved Deciduous forests by Wolfe (1985). Palynological and
diverse than in the north and possessed more species with entire- geochemical analysis of the extensive late Paleocene coals in the
margined leaves belonging to living families with subtropical to Stenkul Fiord area of Ellesmere Island have demonstrated that the
tropical distributions such as laurels (Wolfe and Upchurch, 1986). peat swamp vegetation was dominated by taxodiaceous conifers
Palynofloras derived from coals in the Raton Basin are 75-80 percent (Mcintyre et al., 1994), particularly in the latest Paleocene when
angiosperm pollen, including palms and diverse di cots, with conifers temperatures were rising in middle and high latitudes.
only accounting for 1-2 percent of pollen grains (Fleming and
Pierce, 1994). Fossil wood suggests di cots were important elements Summary
in the vegetation of the Big Bend region in Texas during the Pale- There is now abundant evidence that bolide impact(s) at the end of
ocene (Wheeler, 1991). the Cretaceous had a major effect on terrestrial vegetation in North
 Scott L. Wing 42

America. Palynofloras derived from the first few centimeters above in North America represents a long "echo" of K!f boundary distur-
the Iridium-abundance horizon at sites in the eastern Rockies/Great bance, a period of cool climate, or a combination of the two effects.
Plains region tend to show unusual domination by fems, although Whatever the explanation, Paleocene vegetation of the north tem-
there is local and regional variation in the pattern (Tschudy et al., perate latitudes had low diversity compared with the Late Creta-
1984; Nichols et al., 1986; Lerbekmo, Sweet, and St. Louis, 1987; ceous or early Eocene, and many of the genera were widespread
Johnson et al., 1989; Sweet and Braman, 1992). Fern abundance and abundant in assemblages from Alaska to the northern Rockies
is consistent with vegetational devastation followed by regrowth of to Greenland and Europe. The low diveysity, floristic homogeneity,
early successional species because ferns are frequently early col- and high percentage of non-entire-margined species in the early and
onizers in subtropical and tropical environments. In megafloras as mid-Paleocene (Puercan-early Tiffanian) is consistent with broad-
in palynofloras, there are local differences in the pattern of extinc- leaved deciduous vegetation with only a small admixture of ever-
tion and recovery. Fem abundance peaks and a pattern described as green broad-leaved trees. Swamp vegetation of the time was dom-
"quasisuccession" have been described based on megafloral and cu- inated by taxodiaceous conifers in northern regions, but by palms
ticular remains recovered from several sections in the Raton Basin and other angiosperms in the southern part of North America.
of southern Colorado and northern New Mexico (Wolfe and Up- Global and North American climate warmed near the end of the
church, 1986, 1987). Quasisuccession refers to a pattern of gradu- Paleocene (Wolfe, 1978; Miller, Fairbanks, and Mountain, 1987;
ally increasing abundance and diversity of flowering plants during Wing, Bown, and Obradovich, 1991), moving vegetational zones
the first several hundred thousand years of the Paleocene. northward and thus increasing the diversity and probably the struc-
Johnson (1992) and Johnson and Hickey (1990) have demon- tural complexity of forests over much of North America during
strated 70 percent extinction in leaf megafloras across the K!f the early Eocene. Recent studies have shown that a pronounced
boundary in western North Dakota; these high levels of extinction but short-lived period (100,000 years?) of extreme warming took
are based on samples that are tightly controlled in terms of stratigra- place at the very end of the Paleocene, coincident with the earli-
phy, depositional environment, and sample size. However, similarly est Wasatchian faunas (Kennett and Stott, 1991; Koch, Zachos, and
detailed studies have not been completed in other areas, making it Gingerich, 1992). This Terminal Paleocene Event is associated with
difficult to distinguish between local and regional extinction. Di- a brief decrease in floral diversity in both the Gulf Coastal Plain and
versity and angiosperm dominance increase much more quickly in the northern Rockies (Frederiksen, 1994; Wing, Alroy, and Hickey,
North Dakota than in the Raton Basin (within a few meters of the 1995), which may have been the result of midlatitude extinctions of
K!f boundary), but it is not clear whether this reflects real differ- temperate deciduous lineages (Wing, 1996). In general, however, the
ences in the vegetational histories of the two regions or differences increase in temperature and decrease in temperature seasonality that
in sampling regimes and depositional rates. took place in the later Paleocene should have resulted in less annual
The long-term effects of the K!f bolide impact(s) on the terres- fluctuation in leaf and fruit production, and hence a less seasonal
trial vegetation of North America also remain highly controversial. supply of resources for small mammals. In some areas, decreasing
Wolfe (1987a) suggested that vegetation of the northern Rockies temperature seasonality might have been offset by increasing sea-
was "anomalously deciduous" throughout the Paleocene - that is, sonality in precipitation (Wing and Bown, 1985). Global warming
the preferential extinction of evergreen lineages at the Kif bound- in the late Paleocene and early Eocene also permitted increased ex-
ary left a vegetation that had more deciduous species than would be change of species across the North Atlantic, and perhaps Beringia
expected for vegetation growing under climatic conditions which as well (Tiffney, 1985a, 1985b; Frederiksen, 1994).
are thought to have been warm temperate to subtropical. The tradi-
tional explanation for the relatively low diversity of many Paleocene
EOCENE (SEE FIG. 2.2)
assemblages, and the dominance of inferred deciduous species, has
been that climates in the interior of North America were season- East Coast
ally cool (e.g., Hickey, 1980). Studies in both the Gulf Coastal Few Eocene plant megafossils have been recovered from the East
Plain and the northern Rockies demonstrate that floral diversity in- Coast. Fruits similar to the mangrove palm Nypa, and of a probable
creased very slowly through the first half of the Paleocene, a period mangrove belonging to the poinsettia family, have been described
of roughly 5 million years (Frederiksen, 1994; Wing, Alroy, and from the N anjemoy Formation, a set of early Eocene nearshore ma-
Hickey, 1995). Oxygen isotope data from DSDP cores indicate that rine sediments in Virginia (Mazer and Tiffney, 1982). These fossils
Paleocene oceans were cool compared to those of the Cretaceous closely resemble roughly coeval species from Egypt and Britain,
and Eocene (e.g., Miller, Fairbanks, and Mountain, 1987), and al- demonstrating that some plant genera were distributed across the
though continental indicators of paleoclimate agree with this trend, North Atlantic (Tiffney, 1985a, 1985b). Eocene palynofloras have
faunas and floras from the northern United States do contain some been described from Virginia and South Carolina, and also demon-
frost-sensitive forms through most of the Paleocene. At present it strate the similarity of floras across the North Atlantic (Frederiksen,
is difficult to determine if the low diversity of Paleocene vegetation 1979a, 1980a).

Figure 2.2. Eocene vegetation of North America. Numbers indicate the positions of Eocene floras. Name, age, present coordinates, literature citation, and
inferred paleovegetation are given for each number in the chapter appendix. See text for details.
Tertiary vegetation of North America 43

n =
,,,
 Scott L. Wing 44

Mississippi Embayment and Gulf Coast Paleocene "holdovers" became extinct and xerophytes, shrubs,
Some of the most diverse and best preserved Eocene megafloras herbs, and grasses became more diverse and abundant along with an
in North America come from the Wilcox, Claiborne, and Jackson overall decline in diversity. These changes in composition have been
Groups. The sediments in these formations were mostly deposited interpreted as the result of a decrease in rainfall or an increase in the
in fluvio-deltaic or coastal plain settings such as oxbow fills, lagoons seasonality of rainfall, and have been compared in their severity to
and coastal marshes, and swamps (Frederiksen, 1981). Diverse as- the floral changes associated with the earliest Oligocene temperature
semblages with many notophyllous entire-margined species indicate decline (Frederiksen, 1988).
somewhat dry, warm-temperate to subtropical vegetation (Dilcher,
1973) or dry tropical vegetation (Wolfe, 1978, 1985), depending on Great Plains and eastern Rocky Mountains
interpretation. There were also substantial climate-induced fluctua- The Eocene vegetational history of the Rocky Mountain area has
tions in vegetational physiognomy and diversity through the Eocene been the subject of several reviews that are more detailed than the
(Wolfe and Poore, 1982; Frederiksen, 1980b, 1980c, 1988). one given here (Leopold and MacGinitie, 1972; Axelrod and Raven,
Generally speaking, taxodiaceous conifers are not important in 1985; Wing, 1987). Eocene megaftoral assemblages from the region
Mississippi Embayment palynofloras or megafloras, even in have been recovered from a wide variety of depositional settings
swampy or lignitic facies (Frederiksen, 1981). Some conifers that including fluvial swamps, alluvial ridges, small and large lakes, and
are presently typical of temperate seasonal climates are known in ash falls. Palynofloras are known from all of these environments,
palynofloras (e.g., hemlock, fir) and megafloras (podocarps) (Gray, as well as from coals (e.g., Pocknall, 1987; Farley, 1989, 1990).
1960; Dilcher, 1969). These have been construed by some to be Generally, assemblages from fluvial settings predominate in rocks
remains transported from Appalachian or Ozark uplands with a from the early Eocene, whereas volcanic and lacustrine assemblages
cooler climate (Frederiksen, 1985), but recent studies of pollen are better known from later in the Eocene, particular! yin the volcanic
and leaf transport (e.g., Spicer and Wolfe, 1987; Farley, 1989) im- areas of western Wyoming and Idaho (e.g., Thunder Mountain and
ply that such long-distance transport is unlikely to be an impor- Salmon floras of Idaho).
tant source of fossils, particularly megafossils. If these "temperate" Broad-leaved evergreen vegetation with some admixture of de-
conifers were a part of local vegetation on Eocene floodplains in ciduous species was the dominant forest type over the entire region
the Mississippi Embayment, there are two probable explanations: during the Wasatchian and Bridgerian except in areas with cool
These genera evolved greater cold tolerance during the Tertiary (i.e., montane or seasonally dry to arid climates. Although palms and
they did not prefer cool climates in the Eocene), and/or climates in other subtropical to tropical plant groups were common in many
the area were equable enough to permit the co-occurrence of taxa places, the diversity and physiognomy of most assemblages from
requiring frost-free conditions with those preferring cooler temper- the northern Rockies is not consistent with full tropical vegetation.
atures. During the earliest Eocene, floodplain swamps were still dominated
The physiognomy of the dicot leaf assemblages, the presence of by deciduous taxodiaceous conifers, but conifers appear to have
palms, and the overall floristic composition place fairly high limits been replaced by dicots and palms in most of the region during the
on winter low temperatures, and strong or prolonged winter freezing later early and middle Eocene (e.g., Nichols, 1987).
is very unlikely (Wing and Greenwood, 1993). A middle Eocene Throughout the Eocene deciduous broad-leaved species belong-
flora from eastern Texas that occurs in the same sediments with ing to modem temperate groups (e.g., sycamores, poplars, birches)
Bridgerian mammals documents the presence of a mangrove flora, were locally important in some fluvial depositional environments,
including the highly frost-intolerant mangrove palm, Nypa, that is especially in sediments deposited in or near channels. In the inter-
now restricted to southeast Asia (Westgate and Gee, 1990). montane basins of the northern Rockies, the continued importance of
Although megafloras from the Eocene of the Mississippi Embay- deciduous dicots during the thermal maximum of the Tertiary prob-
ment are typically dominated by dicot leaves probably derived from ably reflects seasonal dryness induced by orogeny and volcanism to
trees, it is noteworthy that these deposits have also produced the the west, as well as seasonal fluctuation in light and temperature.
oldest grass megafossils. Tens of specimens from the early Eocene This seasonal aridity began as early as the Clarkforkian in some
Buchanan Clay Pit in western Tennessee represent a morphologi- areas (Wing and Bown, 1985), and is well documented by the late
cally advanced grass, supporting the hypothesis, based on dispersed early and middle Eocene floras of the Green River Formation as
pollen records, that the grass family originated no later than the well as in floras from other parts of Wyoming (MacGinitie, 1969;
Maastrichtian (Crepet and Feldman, 1991). Leopold and MacGinitie, 1972; Wing, 1987). MacGinitie (1969)
Palynofloras from the Mississippi Embayment have also been referred to this subtropical dry vegetation as savanna woodland,
studied extensively from both biostratigraphic and paleoecologi- although he specified that grasses did not play an important role.
cal perspectives (Frederiksen, 1988, and references therein). Gen- Presumably such forests were somewhat open, but the composition
erally these palynofloral analyses have supported reconstruction of the herbaceous vegetation is unknown. Fems may have been an
of the vegetation as seasonally dry and subtropical or paratropi- important element, although pollen attributable to herbaceous di-
cal rather than full tropical. Palynofloral diversity increased very cots and grasses also occurs in rocks of this age in North America
rapidly during the early Eocene, surpassing the levels achieved dur- (Frederiksen, 1981). Woody xerophytes may also have been signifi-
ing the Paleocene (Frederiksen, 1994). Major shifts in palynoflo- cant elements in low-stature vegetation of drier areas (Leopold, Liu,
ral composition occurred during the middle Eocene, when mesic and Clay-Poole, 1992).
 Tertiary vegetation of North America 45

Early and middle Eocene forests of the volcanic uplands in west- (Leopold and MacGinitie, 1972; Leopold, Liu, and Clay-Poole,
ernmost Wyoming, Idaho, and Nevada were apparently quite distinct 1992). Unlike modem dry-climate vegetation in North America,
from those in the intermontane basins to the east (Axelrod, 1966, pollen of grasses and herbs is very rare in Chadronian palynological
1968). The Yellowstone assemblage, for instance, contains a di- assemblages (Leopold, Liu, and Clay-Poole, 1992). Seasonal aridity
verse mixture of evergreen and deciduous broad-leaved taxa along during the Chadronian is further supported by fossil wood from an
with pinaceous and taxodiaceous conifers. Leaf size and margin area adjacent to the Chadronia Pocket vertebrate locality in western
features, and the presence of fossil wood with subdued or absent Nebraska, which shows strongly seasonal growth patterns. Several
growth rings, are consistent with less seasonal precipitation than of the taxa present have living relatives in seasonally dry climates
prevailed in the Bighorn Basin just to the east (Wheeler, Scott, and (Wheelerand Landon, 1992). Paleo sols in the Big Badlands of South
Barghoom, 1977, 1978; Wing, 1987), and the same can be said in Dakota also offer strong support of declining precipitation beginning
comparing the Wind River Flora in the southern Absarokas with in the Chadronian (500-900 mm precipitation/year) and continuing
samples from the eastern Wind River Basin. Some sort of vege- through the Oligocene to less than 500 mm/year by the Arikareean
tational boundary probably extended roughly north-south through (Retallack, 1992).
western Montana, western Wyoming, and perhaps into Utah, sep- The best known of the Chadronian floras is Florissant, an assem-
arating wetter montane forests from the developing dry deciduous blage derived from volcanic lake sediments that has been collected
forest of the continental interior. However, late Eocene palynofloras for over a century. MacGinitie (1953) inferred Florissant megafos-
from British Columbia are similar to roughly coeval Jackson Group sils to represent two types of vegetation: a broad-leaved deciduous
palynofloras from the Mississippi Embayment (36.5 to 37.7 Ma), forest confined to stream course areas and a scrub forest/grass veg-
and are said to indicate a humid subtropical climate (Rouse and etation that occupied steeper topography and interfluves. Lakeside
Mathews, 1987), implying that some floristic similarity was main- vegetation included large taxodiaceous trees of the genus Sequoia.
tained between the two regions, at least at the high taxonomic level Recent reexamination of Florissant megafossils has resulted in a
resolved in palynofloras. substantial revision of the botanical affinities of the assemblage
The presence of sharp floristic and probably vegetational bound- and reduced the apparent relationship between Florissant and liv-
aries in the Rocky Mountain region was not recognized clearly be- ing floras of northern Mexico and Central America (Manchester,
fore the 1980s. Some of the Eocene floral "phases" described by 1994). Florissant fossils include fragments of what may be Stipa
Leopold and MacGinitie (1972) may relate more to paleogeography (MacGinitie, 1953; Thomasson, 1986), a tropical to temperate grass
than they do to temporal change. For example, sites documenting commonly occurring in moderately dry climates.
the earlier phases of more mesic vegetation were located in west- Paleotemperature estimates for Florissant have also served as a
ern areas that probably had a higher paleoelevation and were under starting point for inferring the late Eocene elevation of the Rockies
the climatic influence of the Pacific Ocean (Wing, 1987). Although in southern Colorado. Paleoelevation is determined from fossil flo-
Eocene drying was regional or subcontinental in scope, local varia- ras by comparing mean annual temperatures estimated from coastal
tion in severity and time of onset make it difficult to define phases (presumably sea-level) and inland floras. An elevation for the inland
of floral development that are applicable across the whole Rocky flora is calculated by dividing the difference in temperature estimates
Mountain region. by the lapse rate of temperature with elevation. Paleoelevational es-
Late Eocene Chadronian megafloras are known from several parts timates for Florissant have varied from 1 to 4 km depending on
of the Rocky Mountains, including southwestern Montana (Becker, the lapse rate and temperature estimates used in the calculation, but
1969), northern Utah (Axelrod, 1966), central Wyoming (Leopold more recent work has favored elevations in the 2-3 km range, sug-
and MacGinitie, 1972; Wing, 1987), southern Colorado gesting that by the Chadronian, southern Colorado had already at-
(MacGinitie, 1953; Manchester, 1994; Nichols, 1987), and New tained an elevation close to the modem level (MacGinitie, 1953; Gre-
Mexico (Meyer, 1992). Most of the Chadronian floras indicate some gory and Chase, 1992; Meyer, 1992; Gregory, 1994; Wolfe, 1994a).
variety of subhumid or montane vegetation, probably depending as Several small megafloral collections from New Mexico (Red
much on elevation as on age. The Montana collections represent Rock Ranch, Hermosa, Hillsboro) appear to be latest Eocene in age
mixed coniferous and broad-leaved deciduous forest with some and may represent dry subalpine vegetation (Axelrod and Bailey,
possibly subhumid taxa like Mahonia (Oregon grape holly) and 1976; Meyer, 1986). Small sample size and low diversity make veg-
juniper (Becker, 1969). The floras from Utah and New Mexico re- etational interpretation of the southernmost late Eocene floras of the
flect varieties of conifer-dominated montane forest with a low di- region difficult.
versity of mostly deciduous dicots (Axelrod, 1966; Meyer, 1986).
The Chadronian megaflora from central Wyoming is derived from Western Rocky Mountains and Pacific Coast
fissure-fill deposits in the White River Group, so that comparisons Megafloral assemblages from the Pacific Coast area are derived
with other floras are made difficult by differences in taphonomic from deltaic, fluvio-deltaic, and lacustrine environments (Burnham,
setting. However, leaves from the fissure-fill are all very small, 1989b, 1994). Coals are common in some units, such as the Puget
and may represent subhumid vegetation. Palynological assemblages Group. Inland floras are most often derived from volcanic lacus-
from the same region are relatively low in diversity and commonly trine beds. Important Eocene megafossil assemblages are known
contain pollen of pine and deciduous broad-leaved trees and shrubs, from British Columbia (e.g., Princeton, Joseph Creek), Washington
sometimes including woody xerophytes like mormon tea (Ephedra) State (Puget Group, Swauk, Chuckanut), Oregon (Clarno localities,
 Scott L. Wing 46

Alvord Creek), and California (Montgomery Creek, Susanville, is consistent with somewhat seasonal precipitation, there is nothing
Chalk Bluffs); some of these were summarized by Wolfe (1987b). in the megaflora to indicate real aridity at this time (Myers, 1991).
Many Pacific Coast Eocene floras have leaf physiognomic char- This megaflora has been interpreted as paratropical broad-leaved
acteristics and levels of diversity similar to those of living wet evergreen forest (Myers, 1991).
tropical forests. In general, assemblages are dominated by entire- Southern California evidently was floristically distinct from the
margined, presumably evergreen dicots. Leaf sizes are notophyllous, Gulf Coast region during the middle Eocene (Frederiksen, 1982,
mesophyllous, and in some place even megaphyllous. Palms are a 1989). Eocene paleoclimatic reconstructions for southern California
common element in many assemblages. Diversity can be remarkably have ranged from wet to dry paratropical to tropical (Novacek and
high (>40 species from a single small floodplain site in the Puget Lillegraven, 1979; Peterson and Abbott, 1979; Frederiksen, 1991).
Group [Burnham, 1994]). Several hundred miles inland occurs one Frederiksen (1991) reports a decline in the diversity of southern
of the most diverse North American Tertiary fossil assemblages - California palynofloras that began slightly later than the middle
the silicified fruits, seeds, and leaves from the Clarno Nut Beds Eocene diversity decline in the Gulf Coast region, but attributes it to
(Manchester, 1981, 1994). This Uintan (43--44 Ma) flora contains the same cause - climatic drying. A palynoflora from northern Baja
over 170 species, and the living relatives of Clarno taxa belong to California that co-occurs with the equid Hyracotherium may provide
families with everything from temperate to tropical distributions, the oldest record of pollen of the sunflower and aster family, along
for example, the banana family, chocolate family, sycamore fam- with a variety of wind-pollinated tree types of temperate deciduous
ily, and grape family (Manchester, 1994). Clarno shares a number groups (Cross and Martinez-Hernandez, 1980).
of genera with the Eocene London Clay flora of England (Tiffney, Eocene Pacific Coast megafloras, particularly those of Washington
1985a, 1985b). and Oregon, present the most wet-tropical appearance of any known
In some inland areas, altitudinal effects are more noticeable. Flo- Eocene North American assemblages. This is probably a result of
ras from south-central British Columbia, eastern Washington, Idaho, high rainfall, low temperature range, and low altitudes on the West
and Nevada show cooler climate vegetation dominated by pinaceous Coast of the continent, in conjunction with the globally warm tem-
and cupressaceous conifers, although evergreen and deciduous di- peratures of the Eocene. Evidence exists for a gradient of increasing
cots were diverse at most well-sampled sites except those inferred to seasonal dryness from north to south along the West Coast, but more
represent the highest paleoaltitudes (Axelrod, 1966, 1968; Wolfe, megafossil evidence from southern California and/or Mexico would
1987b; Wolfe and Wehr, 1987). Attempts to reconstruct the late be helpful in confirming this.
Eocene paleoaltitude of northeastern Nevada based on megafloras
have yielded estimates with a large degree of error, but centered The Arctic and Alaska
around elevations close to those seen today (Povey, Spicer, and Alaskan Eocene megafossils have for the most part been collected
England, 1994). High late Eocene paleoelevations in Nevada (and from coal-bearing, fluvio-deltaic sequences in southern Alaska (e.g.,
presumably Idaho) are part of the orographic explanation of the Kulthieth Formation and Kushtaka Formation of Wolfe, 1977). Di-
dry or seasonally dry climates inferred from Chadronian floras and verse middle Eocene (early Ravenian of Wolfe) assemblages from
paleosols in the eastern Rockies and western Great Plains. this area have many large, probably evergreen, leaves, and a fairly
In northern California, megafossil assemblages also indicate di- high proportion of entire margins and drip tips. The floras also con-
verse broad-leaved evergreen vegetation that was probably analo- tain palms, and at some localities there are leaves thought to be
gous in many ways to some extant tropical forests. The early Eocene derived from woody lianas. These characteristics, in addition to the
Chalk Bluffs flora has a number of species with relatives in season- tropical to subtropical Asian affinities of many species, led Wolfe
ally dry parts of Mexico and Central America (MacGinitie, 1941), (1972) to reconstruct the vegetation as near tropical to tropical. Al-
which could imply somewhat drier conditions than prevailed in though more recently Wolfe (pers. comm., 1990) has interpreted
coastal Washington. However, the middle to late Eocene Susanville these floras to represent subtropical vegetation, this reconstruction
assemblage in northern California is notable for extremely large still places broad-leaved evergreen forests as far as 60°N latitude
leaves, many of them with drip tips, suggesting a wet climate. The along the West Coast of North America during the early and middle
early Eocene Montgomery Creek assemblage, also in northern Cal- Eocene.
ifornia, demonstrates that some near-channel fluvial vegetation in The correlation of middle and late Eocene floras of Alaska with
this region was low in diversity and dominated by deciduous mem- those of the continental United States is somewhat problematic,
bers of the birch family. which introduces considerable uncertainty in trying to infer the
Recent work on middle Eocene (Lutetian or early Bartonian latitudinal distribution of paleovegetation. Parts of coastal Alaska
equivalent) palynofloras from southern California (e.g., Ardath have been rifted northward since the early Eocene (Bruns, 1983),
Shale, Mission Valley Formation) has revealed many taxa related but the magnitude of the displacement has been disputed (Plafker,
to the plant families that are today most speciose in seasonally 1984; Wolfe and McCoy, 1984). The resolution of the disagreement
dry climates, but fem spores and conifer pollen are also diverse may lie in a Tertiary displacement of smaller magnitude, one that
(Frederiksen et al., 1983; Frederiksen, 1989). An early middle still indicates much warmer climate along the Northwest Coast of
Eocene megaflora from the Torrey Formation north of San Diego North America than exists today. Later Eocene floras in southern
has some taxa in common with coeval floras in northern California Alaska contain higher proportions of nonentire, presumably decidu-
and the Rocky Mountains, and although the composition of the flora ous, leaves, and have been interpreted as indicating warm, although
 Tertiary vegetation of North America 47

perhaps not entirely frostless, climates (Wolfe, 1972, 1992). Wolfe across Beringia into East Asia. Palynofloras suggest that such vege-
(1992) reconstructs the late Eocene Katalla flora (Poul Creek For- tation also occurred up the East Coast of North America and across
mation) as having been at the latitude of the northern part of south- North Atlantic coastlines to Europe (Frederiksen, 1994). Differences
east Alaska. This flora contains diverse broad-leaved evergreen lau- in the composition of Eocene megafloras from the Mississippi Em-
rels and deciduous broad-leaved taxa related to oaks and hickories bayment and the eastern Rocky Mountains suggest that drying in the
(Wolfe, 1992). Even in the Alaskan interior, late Eocene floras re- interior of North America decreased interchange between the west-
flect forest vegetation growing under a relatively warm climate. The ern and eastern areas of mesic forest. Some of the forest taxa that are
probably late Eocene (36-33 Ma) Rex Creek flora from the central now disjunct between East Asia and eastern North America origi-
Alaska Range contains a diverse group of woody, broad-leaved taxa nally may have separated at this time. Although global and North
represented by both pollen and megafossils, and some leaves that American temperatures decreased in a fluctuating manner through
are likely to be evergreen broad-leaved taxa, such as holly (Wolfe, most of the Eocene (Wolfe, 1978; Miller, Fairbanks, and Moun-
1992). tain, 1987), latitudinal temperature gradients were still low enough
Eocene megafloras from Axel Heiberg and Ellesmere Islands in that precipitation was the preeminent force in creating vegetational
the Canadian Arctic are dominated by conifers, especially the de- differences at the continental scale. Highly seasonal light regimes
ciduous Metasequoia, but also pine-family conifers such as larch, may also have been an important factor in limiting the northward
fir, and spruce (Basinger, Mciver, and LePage, 1988; Greenwood distribution of many evergreen plant lineages (Wolfe, 1985).
and Basinger, 1994). The diversity of these Arctic assemblages is
low, the most common dicots belong to the birch and alder fami-
OLIGOCENE (SEE FIG. 2.3)
lies, and all dicots were deciduous. Physiognomic analysis of the
Axel Heiberg assemblage implies a mild but cool climate, and there East Coast
are no frost-intolerant taxa in the flora (Greenwood and Basinger, Well-dated early Oligocene megafloras from eastern North America
1994). Dicot leaf fossils from Ellesmere Island are extremely large, are not known, but the May's Landing core from southeastern New
perhaps a response to the effects of the polar light regime (Hickey, Jersey has produced palynofloras that record the Eocene-Oligocene
pers. comm.). Although the Arctic Eocene assemblages resemble transition in shallow marine sediments that also contain marine mi-
Paleocene floras of the Rocky Mountain region in the abundance of crofossils (Owens et al., 1988). A minor extinction, and an increase
Metasequoia and birch family species, the presence of oaklike dicots in the dominance of oaklike pollen, appear to correlate with the early
and pine-family conifers is more consistent with an Eocene age. Oligocene (33 Ma) climatic cooling event that is also seen in flo-
ras from the Pacific Northwest and possibly the Gulf Coastal Plain
Summary (Wolfe, 1978; Frederiksen, 1980b; Owens et al., 1988; Wolfe, 1992).
Global climate began to warm during the last 1-2 million years of the There is only one East Coast megafossil assemblage of possi-
Paleocene, and this trend continued into the early Eocene, reaching ble Oligocene age, the Brandon Lignite of Vermont. Although this
a Cenozoic peak during the late early Eocene (Wolfe, 1978; Wolfe deposit produces diverse fruits, seeds, wood, and pollen, its age
and Poore, 1982; Miller, Fairbanks, and Mountain, 1987). During is poorly constrained, most probably being late Oligocene or early
this time, broad-leaved evergreen forests occurred as far north as 60° Miocene (Tiffney, l 985b, 1994). The composition of the flora is con-
(Wolfe, 1985), and forests of bald-cypress relatives and deciduous sistent with a subtropical to warm-temperate forest that had a mix-
broad-leaved trees existed on Ellesmere and Axel Heiberg Islands ture of evergreen and deciduous species, and at least five species of
well within the Arctic Circle (Hickey et al., 1983; Greenwood and vines. Based on comparisons of fossils with living relatives, Tiffney
Basinger, 1994). The early Eocene was the time of maximal inter- (1994) argues that evergreen species were dominant and the frosts
change between floras of North America, Europe, and Asia; even would have been rare. Similar forests occur today along the Gulf
frost-sensitive plant lineages were able to expand their ranges across Coast of the United States and in the highlands of northeastern
the North Atlantic and Beringia (Tiffney, 1985a, 1985b). Mexico (Tiffney, 1994).
However, even as warm climates at middle and high latitudes
permitted maximum spread of broad-leaved evergreen forests and Mississippi Embayment and Gulf Coast
frost-sensitive lineages, volcanic activity and uplift in the Rocky The Mississippi Embayment shrank in size during the Tertiary, but
Mountains were creating montane regions with cooler climates early Oligocene floras are still derived largely from low-energy
(Axelrod, 1966, 1968; Wolfe and Wehr, 1987; Gregory, 1994) and fluvio-deltaic settings, coastal marshes and swamps, or in the case
casting local or regional rain shadows to the east (Leopold and of some palynofloras, from nearshore marine environments
MacGinitie, 1972; Wing, 1987). The fluctuating withdrawal of the (Frederiksen, 1981). Little has been said about the vegetational im-
Mississippi Embayment from the northern midwest also may have plications of Oligocene megafloras of this region, but early Oligo-
contributed to increasing seasonal dryness, or even aridity, in the cene palynofloras have been considered from a paleoecological per-
continental interior during the middle and late Eocene. spective (Frederiksen, 1980b, 1981, 1985, 1988). Early Oligocene
The slow drying trend in the interior of the continent began to di- (lower Vicksburgian) palynofloras are distinct from those of the late
vide North America into separate floristic and vegetational regions. Eocene in the abundance ofoak-family pollen and the greater impor-
Mesic, broad-leaved evergreen and semideciduous forests remained tance of fem spores and pollen of bald-cypress and related families
the dominant vegetation type along the Pacific Coast, and probably (Taxodiaceae-Cupressaceae-Taxaceae [Frederiksen, 1981 ]). This is
Scott L. Wing 48

=
,,.,
 Tertiary vegetation of North America 49

Figure 2.3. Oligocene vegetation of North America. Numbers indicate the positions of Oligocene floras. Name, age, present coordinates, literature citation,
and inferred paleovegetation are given for each number in the chapter appendix. See text for details.

consistent with a significant change in vegetation, and depending on although higher elevation sites in the same region might have sup-
how the strata are correlated with Pacific Coast floras and the ma- ported forests more like those at Creede.
rine record, could reflect the early Oligocene (33 Ma) cooling event
described by Wolfe (1978, 1992). Palm fossils remain abundant at Western Rocky Mountains and Pacific Coast
some Oligocene sites, indicating an absence of severe frost. There are a great many assemblages from Oregon, for example,
Lyons (Meyer, 1973), Rujada (Lakhanpal, 1958), John Day of Fos-
Great Plains and eastern Rocky Mountains sil (Manchester and Meyer, 1987), and many assemblages lumped
There are few megafloral assemblages of Oligocene age in the Rocky under the name Bridge Creek (Chaney, 1927). These floras derive
Mountain region (Wing, 1987). A series of early Oligocene (Orellan) from volcaniclastic sediments associated with the Cascade Range,
floras from the "Medicine Lodge Beds" of western Montana re- and most are dated radiometrically in the range of 30-32 Ma, just
ported by Becker (1969) do not appear to be substantially different following the early Oligocene climatic cooling event (Wolfe, 1978,
from Eocene floras of the same region. (The Oligocene age of these 1992). The vegetational and paleoclimatic implications of these flo-
beds was proposed by Fields et al. [1985), which generally has ras have been reviewed by Wolfe (1981) and Manchester and Meyer
not been noted or accepted by paleobotanists.) Wolfe (1992) argues (1987). Vegetation of this region during the early Oligocene was a
that the Orellan Antero flora from central Colorado represents a cool forest composed of a mixture of deciduous and broad-leaved ever-
conifer-dominated forest, but notes that the small size of the dicot greens and conifers. Differences in the relative dominance and di-
leaves is also seen in woodland and chaparral vegetation. Although versity of evergreen versus deciduous taxa probably reflect a fairly
it is only about 1 million years younger than the Florissant flora, strong coastal effect because more easterly floras tend to be less di-
the Antero flora already shows similarity to the well-known late verse with a higher proportion of toothed, deciduous, small-leafed
Oligocene flora from Creede (Wolfe, 1992). taxa (Wolfe, 1981). Manchester and Meyer (1987) inferred that the
The late Oligocene (approximately 26 Ma) Creede flora is derived vegetation producing the Fossil, Oregon, assemblage was most sim-
from volcanic caldera lake beds in southern Colorado (Axelrod, ilar to the living Mixed Northern Hardwood Forest of Asia, growing
1987). Pinaceous conifers are abundant in the Creede flora, and under a mean annual temperature (MAT) of perhaps 5°C, and pos-
dicot leaves are typically small and represent deciduous species. sibly with a cold month mean below -2°C. Wolfe ( 1981) suggested
Based on Creede, Axelrod and Raven (1985) and Axelrod (1987) the more coastal vegetation was a mixed broad-leaved evergreen and
argued that many elements of the extant vegetation of the south- coniferous forest of moderate diversity (MAT 12-13°C). In both ar-
western United States were already native to the area by the end eas of Oregon several lineages persisted that were extinct in the
of the Oligocene. The Creede flora records both subalpine conifer- midcontinent region during or by the end of the Eocene (e.g., dawn
ous forest and a woodland/scrub vegetation (Axelrod, 1987). Wolfe redwood [Metasequoia], katsura tree [Cercidiphyllum], and a walnut
and Schorn (1990) substantially revised the taxonomy of the flora, relative now restricted to East Asia, Engelhardia). Wolfe (1987b)
but reached broadly similar conclusions about its inferred vegeta- described late Oligocene vegetation of the Columbia Plateau re-
tion (Wolfe and Schorn, 1989). Wolfe and Schorn suggested that gion as Mixed Mesophytic Forest - a rich mixture of broad-leaved
moisture was the primary environmental gradient influencing the evergreen and deciduous species with conifers.
composition of the vegetation around the Creede caldera, with fir- Farther south in California, Oligocene floras are less abundant.
spruce and pine-fir forest occupying the wettest sites, juniper-pine Leaves from the early Oligocene Laporte Flora of the western Sierra
woodland growing on sites with intermediate moisture levels, and Nevada appear to represent a mixed evergreen and deciduous broad-
mountain mahogany chaparral covering the driest areas. Estimates leaved forest growing in a climate with a MAT estimated at about
of the paleoelevation of the Creede flora have ranged from 1 to more 20°C (Potbury, 1935; Wolfe, 1992). Axelrod (1979) suggested that
than 2 km (Meyer, 1986; Axelrod, 1987; Wolfe and Schorn, 1989). seasonally dry forest was characteristic of much of southern Cali-
A palynoflora approximately the same age as the Creede mega- fornia and the southwestern United States, as well as northwestern
flora (26-28 Ma) has been described from the Sevier Desert region Mexico during the Oligocene, although there are few floras of this
of western Utah (Lindsey et al., 1980). This low-diversity paly- age from the region.
noflora was derived from a sequence of interbedded volcanic sedi-
ments and thick evaporites, and is dominated by xeric-adapted plants The Arctic and Alaska
like Ephedra (Mormon tea) and pollen that was probably derived A number of early (Kushtaka, Katalla, and Kuthieth Formations)
from members of the herbaceous families Chenopodiaceae, Elaeag- to late (Healy Creek, Stepovak, Tsadaka, Kootznahoo, and lower
naceae, and Compositae. Pine pollen is rare and may have been Kenai Formations) Oligocene assemblages are known from coastal
transported from a distant upland source. Based on both palynoflo- southern Alaska (Wolfe, 1977). The early Oligocene (Kummerian
ral and sedimentological evidence, the late Oligocene paleoclimate Stage of Wolfe, 1977) vegetation of southern Alaska has been in-
of the region was inferred to have been very dry. The data are con- terpreted as a variety of subtropical forest with strong floristic rela-
sistent with a low-diversity, largely herbaceous local vegetation, tionships to roughly coeval forests of the Pacific Northwest (Wolfe,
 Scott L. Wing 50

1977). Thick-textured, possibly evergreen, broad-leaved species are Although paleobotanical and isotopic data support an ameliora-
a major component of these floras. By the late Oligocene (Angoo- tion of climate during the later Oligocene in coastal regions, it ap-
nian Stage of Wolfe, 1977) low-diversity forests were dominated pears that low-biomass vegetation continued to develop in parts of
by broad-leaved deciduous species to the exclusion of broad-leaved the western interior and southwestern United States. The enhanced
evergreens (Wolfe, 1972). Presumably emplacemen t of the Yaku- vegetational differences between coastal and interior areas are prob-
tat Block was complete by late Oligocene time, although displace- ably attributable to increased continentality (which may have been
ment of blocks from the Pacific Northwest region to southeastern influenced by the regression of the Mississippi Embayment ) and to
Alaska has been proposed for times more recent than 40 Ma (Cowan, continued orographic influences on rainfall.
1982).

MIOCENE AND PLIOCEN E (SEE FIG. 2.4)

Summary
Sharp cooling in the early Oligocene shown by floras from the Pa- East Coast
cific Coast of North America (Wolfe, 1978), and more gradual cool- Miocene and Pliocene plant-bearing deposits on the East Coast are
ing during the late Eocene shown by European floras (Collinson, rare and consist mostly of isolated lignitic layers, clay lenses, and
Fowler, and Boulter, 1981 ), generally coincide with temperature es- nearshore marine deposits. The general impression of paleovegeta-
timates for ocean bottom water derived from benthic foraminifera tion that emerges from these deposits is of dominantly broad-leaved
(Miller, Fairbanks, and Mountain, 1987; Miller, 1992). This implies deciduous forest, perhaps more diverse, but otherwise not too dis-
a global cause for the cooling, probably related to enhanced bipolar similar from that occupying the southeastern part of North America
production of cold deep-ocean currents, and possibly the growth of today. Miocene forests of the East still contained some lineages
continental ice sheets in Antarctica (Miller, 1992). that are presently confined to East Asia and/or Mexico and Central
Although the cause of the climatic cooling may have been global, America. The presence of these warm-climate genera in Miocene
the effects on North American vegetation do not appear to have been floras of the Calvert, Bridgeton, and Pensauken Formations was
uniform. Wolfe (1978, 1992) has argued for dramatic shifts in the noted during the late nineteenth and early twentieth centuries, al-
vegetation of the Pacific Northwest and Alaska, from dominantly though these identifications are not wholly reliable (Berry, 1916b,
broad-leaved evergreen vegetation typical of subtropical or warmer 1937; Hollick, 1882). However, recent workers have confirmed the
climates to forests dominated by deciduous broad-leaved taxa and presence of these "exotic" elements. For instance, an early mid-
conifers. This rapid (less than a million years?) vegetational shift is dle Miocene (middle Barstovian to Clarendonian) palynoftora from
thought to reflect a major increase in temperature seasonality and to Massachusetts contains the bald-cypress-family conifer Sciadopi-
have coincided with the continental extinction of a number of warm- tys and the East Asian walnut relative Pterocarya (Frederiksen,
climate lineages (Wolfe, 1992). Data from eastern North America 1979b). The Cohansey Formation of southern New Jersey, about
are sparse, and although they imply changes in vegetational im- 11 Ma, produces a palynoflora of which four or five genera are now
portance of some taxa, they do not demonstrate major extinctions confined to regions with warmer winters, including Oreomunnea, a
(Frederiksen, 1980b; Owens et al., 1988). Early Oligocene floras walnut relative now restricted to Central America, and Podocarpus,
from the interior of North America are also few in number, but a warm-climate conifer of a largely Southern Hemisphere family
seem to show less dramatic change than those from coastal areas (Rachele, 1976; Greller and Rachele, 1983). Even a late Miocene
(Wing, 1987). There are at least four possible explanations for this to early Pliocene flora from Brandywine , Maryland, still contains
difference in response: (1) Continental areas were subject to more a number of lines now extinct in North America (Tiffney et al.,
seasonal and diurnal temperature fluctuation throughout the Eocene 1989; McCartan et al., 1990). In all probability these genera were
and thus the flora had already lost tax a sensitive to these conditions, eliminated from the East Coast during the later Pliocene or early
(2) seasonality of rainfall during the Eocene had already favored Pleistocene by increasing winter cold. The degree of grassland de-
deciduous taxa, (3) local environmental variations (e.g., paleoalti- velopment in eastern North America during the Miocene is poorly
tudinal differences between floras, rain shadow effects) make the understood, but grass pollen is uncommon in palynofloras. (The
global climatic signal harder to pick out, and (4) climatic change meaning of this negative evidence is not clear because grass pollen
driven by shifts in ocean circulation was not as severe in continental is not always abundant in western Miocene palynofloras that are
interiors. Interpretation of admittedly scantpaleob otanical data from believed to come from areas that had substantial grass cover.)
around North America suggests that even a major climate change
"forced" by shifts in global ocean circulation produced effects of Mississippi Embaymen t and Gulf Coast
different magnitude in different parts of the continent. Continen- In contrast to the abundant Paleocene through Oligocene megafto-
tal ecosystems and climates are so heterogeneous that it is overly ras from this region, Miocene and Pliocene assemblages are few. A
simplistic to expect them to react in a uniform manner to even the small flora reported from Alum Bluff in northern Florida is probably
largest environmental perturbations. early Barstovian, and it is dominated by palms, laurels, and other

Figure 2.4. Miocene-Pliocene vegetation of North America. Numbers indicate the positions of Miocene
and Pliocene floras. Name, age, present coordinates,
literature citation, and inferred paleovegetation are given for each number in the chapter appendix. See
text for details.
Tertiary vegetation of North America 51

5
~
w
1ij

~
0
8
N
l
 Scott L. Wing 52

plants expected in a frostless climate (Berry, 1916c). One Pliocene dry-adapted grasses. The sequence of Miocene floras also is thought
assemblage from the region is the Citronelle flora from Alabama, to reflect the spread of grassland at the expense of woodland and
which strongly resembles extant floras in the same region (Berry, riparian forest, although the development of extensive prairie may
1916d). not have occurred until as late as the Pleistocene (Axelrod, 1985).
Axelrod ( 1985) argued that frequent fire is a prerequisite for exclud-
Great Plains and eastern Rocky Mountains ing trees from grassland vegetation, and the vast prairies observed
Several reviews of Neogene vegetational change in this region cover by European colonists of the Great Plains may have been in part the
the topic in greater detail than is given here (Stebbins, 1981; Ax- result of anthropogenic fire. Stebbins (1981) also argued for a recent
elrod, 1985; Leopold and Denton, 1987; Thompson, 1991). Much origin of prairie vegetation and floras, suggesting that the present
of the available paleobotanical information for the Miocene and communities did not arise until the Pleistocene.
Pliocene is from pollen floras (e.g., Leopold and MacGinitie, 1972; Studies of the stable carbon isotope values of soil nodules and
Bamosky, 1984; Thompson, 1991), although abundant silicified mammalian tooth enamel for the Miocene to Pliocene interval in-
fruits and seeds of herbaceous plants derived from volcanic sedi- dicate a shift from C3- to C4-dominated vegetation at about 7 Ma
ments have been known for some time (Elias, 1932, 1935; Chaney (Cerling, Wang, and Quade, 1993; Wang, Cerling, and MacFadden,
and Elias, 1936) and are currently under study (e.g., Thomasson, 1994). This probably reflects the spread of grasses that grow during
1980, 1982; Thomasson, Nelson, and Zakrzewski, 1986; Thomas- the hot summer months at the expense of cool season grasses and
son et al., 1990). all shrubs and trees. In modem vegetation the relative importance
The oldest Miocene floras from the region are Hemingfordian of C4 grasses decreases with increasing latitude, and C4 grasses
equivalents from the Troublesome Formation, Colorado, and the are not important north of about the U.S./Canadian border. Most of
Split Rock Formation, Wyoming (Leopold and Denton, 1987). Next the carbon isotope data for the Mio-Pliocene are from the southern
in age are the early B arstovian Sheep Creek, Nebraska, flora and the Great Plains and indicate C4-dominated vegetation; carbon isotope
one major Neogene compression flora from the Great Plains, the values from the northern United States and southern Canada are
Kilgore of Nebraska (MacGinitie, 1962), which is mid-Barstovian. consistent with mixed C3/C4 vegetation after about 7 Ma (Wang,
The Saratoga Valley palynoflora from southeastern Wyoming is Cerling, and MacFadden, 1994).
roughly the same age as Kilgore (Leopold and Denton, 1987). Although the bulkofthe paleobotanical and palynological sources
Younger Miocene assemblages include the mid-Clarendonian Ash cited here do not offer evidence for grassland vegetation, the 7 Ma
Hollow flora, the late Clarendonian (9 Ma) Teewinot palynoflora (later early Hemphillian) Minium's Quarry Flora from Kansas
from western Wyoming (Leopold and MacGinitie, 1972; Bamosky, clearly represents both grassland and riparian woodland (Thomasson
1984; Leopold and Denton, 1987), and the Barstovian or younger et al., 1990), and fragments of grass demonstrating C4 anatomy
Moonstone palynoflora (Leopold and Denton, 1987). Most or all of have been recovered from the Miocene of Kansas (Thomasson,
these palynofloras are similar in that they are dominated by pine Nelson, and Zakrzewski, 1986). The carbon isotopic evidence for
pollen. Pollen of other pinaceous conifers (spruce, fir) and junipers widespread C4 plants contrasts with rare fossils of C4 plants. The
is also common, as is that of a few deciduous hardwoods (wal- discrepancy probably reflects the tendency for fossil floras of all
nut family, oaks, elms, birch, willow) and halophytic or xerophytic types (even palynofloras) to be dominated by local wetland plants,
shrubs such as Mormon tea (Ephedra), greasewood (Sarcobatus), whereas the soil nodule data are constrained to come from well-
and herbs of the family Chenopodiaceae. Pollen of sagebrush and drained soils with strong development of a calcic B horizon (Cerling,
other composites is also moderately abundant, but grass pollen gen- Wang, and Quade, 1993). If both sets of data reflect local conditions
erally is rare or absent (Leopold and Denton, 1987). accurately, the implication is a regional vegetation with a strongly
The Kilgore megaflora records a low-diversity deciduous hard- mosaic pattern of domination by C3 herbs, shrubs, or trees in wetter
wood riparian forest (MacGinitie, 1962). Based on this record areas, and open C4 grasslands on the drier interfluves.
Leopold and Denton (1987) inferred two vegetation types: an open Pliocene palynofloras from the Western Interior are relatively
basin or lakeside shrub vegetation dominated by xerophytic or halo- rare, but 2-million-year-old assemblages from the Jackson Hole
phytic shrubs, and montane conifer forest, although Leopold, Liu, area in Wyoming suggest essentially modern vegetation (Barnosky,
and Clay-Poole (1992) suggest that this palynoflora indicates the de- 1984; Thompson, 1991). The distribution of large, terrestrial tor-
velopment of grassy savanna. The deciduous hardwoods may well toises and other temperature-sensitive animals implies that Pliocene
have been restricted to stream courses. Probable Pliocene (>2 Ma) winters in the Western Interior were much warmer than at present
palynofloras from western Wyoming have abundant pine and sage- even though rainfall regimes clearly favored grassland and other
brush, along with other indicators of steppe and montane conifer low-biomass vegetation (evidence summarized in Thompson, 1991 ).
forest, indicating that vegetation similar to that of the present day ex-
isted in the area at this time (Richmond, Mull enders, and Coremans, Western Rocky Mountains and Pacific Coast
1978). A very large number of Neogene megafloras and palynofloras have
Summarizing Hemingfordian through Hemphillian floras from been collected and described from the Columbia Plateau region.
Nebraska, Kansas, Oklahoma, and Texas described by Elias (1932, Many of these are derived from volcaniclastic lacustrine or fluvial
1935), Axelrod (1985) inferred a steady drying trend from about settings. The climatic and vegetational history of the region dur-
16 Ma through the end of the Miocene, with the driest climate period ing the Neogene have been discussed by authors including Wolfe
in the early Pliocene providing a major spur to the diversification of (1981, 1987b), Leopold and Denton (1987), Smiley (1985),
 Tertiary vegetation of North America 53

Taggart, Cross, and Satchell (1982), and Thompson (1991), who deciduous, broad-leaved evergreen and coniferous forest, growing
have provided lists of known floras as well as more detailed interpre- under a mild moist climate (e.g., Axelrod, 1979). Furthermore, mid-
tations of the important assemblages. The plant record is relatively Miocene floras as far east as Nevada are compositionally similar to
sparse during much of late Arikareean and Hemingfordian time, but those of the Pacific Northwest. For instance, Wolfe (1964) placed the
quite dense throughout the Barstovian and Clarendonian and into late Hemingfordian flora from Fingerrock, in southwestern Nevada,
the Pliocene (e.g., Mascall, Clarkia, Succor Creek, Trout Creek, in the same province as the Columbia Plateau assemblages. All of
Payette, Ellensburg, The Dalles, Troutdale, Poison Creek). Wolfe the pre-15 Ma floras from Nevada contain many genera that are
(1981) interpreted many of the earlier Miocene Columbia Plateau now found in summer-wet climates in eastern North America and
floras as representing Mixed Mesophytic forest, a diverse mixture East Asia, but later in the Miocene (Barstovian-Clarendonian) floras
of broad-leaved deciduous, broad-leaved evergreen, and coniferous from this area indicate drier conditions (Axelrod and Schorn, 1994).
species that lived under a mild, summer-wet climate. He argued The floral shift was apparently caused by decreasing summer pre-
that this forest type gave way in some areas to a less diverse Mixed cipitation, which was in turn a consequence of increased upwelling
Coniferous forest during the late Barstovian (13-14 Ma), possibly of cold water along the Pacific Coast. Although the decrease in
as a result of drier summers. summer precipitation is thought to have been time transgressive
Leopold and Denton (1987) place the shift to drier climates (15 Ma in Nevada to as late as 11 Ma in the Pacific Northwest), it
in post-Clarendonian time on the basis of the composition of the appears to have been very rapid in certain local areas (Axelrod and
Ellensburg and Hanford floras in eastern Washington State. For in- Schorn, 1994 ).
stance, swamp settings in eastern Washington continued to harbor Pliocene pollen assemblages from cores and outcrops in Cal-
Taxodium (bald cypress) and Metasequoia (dawn redwood) until ifornia, Oregon, Washington, and Idaho provide a fairly detailed
12-8 Ma. The Deschutes flora (about 4.5 Ma) has been interpreted record of climatic and vegetational fluctuations from 5 to 2 Ma
as riparian vegetation in a region where interfluve areas had few (Thompson, 1991). During the earlier part of this interval, mixed
trees, and taken as an indication of decreased summer precipitation conifer forest predominated in northern California, and climates
(Leopold and Denton, 1987). However, Leopold and Denton (1987) were probably warmer and more equable than they are today along
place the spread of steppe vegetation on the Columbia Plateau as much of the Pacific Coast and even in areas far inland. Summer
a post-Hemphillian phenomenon, and state that steppe and palouse drought remained a feature of climate along the Pacific slope as it had
grassland did not come to dominate the Snake River Plain until the been in the late Miocene. On the Snake River Plain in southeastern
Quaternary. Idaho, pollen assemblages from about 3 Ma have high proportions
Taggart, Cross, and Satchell (1982) argued for a much earlier, al- of sagebrush and other steppe taxa (Thompson, 1991). Between
though areally and successionally limited occurrence of xerophytic 2.4 and 2 Ma, climates in western North America generally be-
vegetation. Many palynofloras from the Succor Creek area (late came colder and drier, followed by a moister phase between 2
Barstovian?, ,about 14 Ma) document fairly diverse, angiosperm- and 1.8 Ma (Thompson, 1991). A core from Tule Lake in north"
dominated or mixed conifer-angiosperm vegetation, but samples em California provides evidence for Milankovitch-scale fluctua-
from intervals immediately following volcanic episodes show much tions in climate, but these were of lower amplitude than Pleistocene
less diversity and are dominated by pine pollen and other paly- glacial/interglacial cycles (Adam et al., 1990; Thompson, 1991).
nomorphs thought to have been derived from forbs and grasses The Tehachapi Flora of southern California (late Hemingfordian,
(Taggart, Cross, and Satchell, 1982). The frequency, intensity, and about 17 Ma) is thought to represent dry woodland, or even semi-
regional extent of volcanic events may have created a major habi- desert, although in a climate without freezing temperatures (Axelrod,
tat for more xeric, open vegetation, even under climatic conditions 1939, 1979). A vegetation of dry woodland or scrubland has also
that tended to produce a diverse forest "climax" vegetation (Taggart, been inferred for the Mint Canyon flora ( 12-13 Ma), although the as-
Cross, and Satchell, 1982). These authors argued that vertebrate fau- semblage is quite diverse (Axelrod, 1940, 1979). The oldest known
nas occupying this open, successional vegetation would have had fossil grass that demonstrates the anatomical and carbon isotope
a better chance of preservation than those of the "climax" forests characteristics of the C4 photosynthetic pathway was recovered
because of the deposition and redeposition of large quantities of ash from roughly 12-million-year-old sediments in California, although
in the intervals immediately following major volcanic eruptions. In it was originally described as Pliocene (Nambudiri et al., 1978;
contrast, megafloral assemblages are mostly found in lake basins and Thomasson, 1986). The undescribed Topanga and Modelo floras
represent more mature forest vegetation because herbaceous plants were cited by Axelrod (1979) as evidence for a wetter coastal strip
are seldom transported to or preserved in such settings. This would of vegetation during the Miocene in southern California. Axelrod
explain why vertebrate faunas containing mammals with grazing (1979) states that in southern California, as in the Pacific Northwest
adaptations would be approximately coeval with floras indicating and the Western Interior, the early Pliocene was the driest phase of
diverse, closed-forest vegetation. the Cenozoic.
Many Miocene floras also are known from California and Nevada,
although as in the Pacific Northwest, the Barstovian and later inter- The Arctic and Alaska
vals are better represented than the earlier Miocene. Generally, early An early Miocene flora from the Houghton Astrobleme (Devon Is-
to mid-Miocene floras from northern California are similar in com- land, Canada, at about 75°N latitude) shows cool-temperate, decid-
position to those of the Columbia Plateau and are thought to be uous, broad-leaved, and coniferous forest at about 22 Ma (Hickey,
derived from a similar type of vegetation, a diverse broad-leaved Johnson, and Dawson, 1988). The low diversity of the megaflora
 Scott L. Wing 54

may in part reflect its recovery from a single locality in a restricted riod, high-latitude climates warmed again rapidly, and boreal conif-
lake basin, but the associated palynoflora indicates low regional erous forest/tundra vegetation is recorded at the northernmost tip
diversity as well. Based on overlap in the climatic preferences of of Greenland about 2 million years ago (Kap Kobenhavn flora;
living relatives of the common genera in the flora (e.g., pine, spruce, Bennike, 1989). Following this shorter period of warmth, Arctic
larch, birch, and pollen of the heath family), these authors inferred climates began to oscillate between colder and warmer periods be-
a mean annual temperature of 8 to 12°C with a cold month mean tween 1. 7 and 1.2 Ma, finally sliding into full glacial/interglacial
of - 7 to -4°C. The closest modern analogue of the vegetation is cycles in the last million years (Repenning and Brouwers, 1992).
the mixed conifer and northern hardwood forest that today occurs
south of boreal forests; grass pollen is absent from the assemblage Summary
(Hickey, Johnson, and Dawson, 1988). Long-term decreases in temperature and precipitation caused
The Seldovia Point Flora of southern Alaska is a late early to changes in the vegetation of North America during the Miocene.
middle Miocene assemblage from southern Alaska that contains The early part of the epoch was relatively warm, but was followed
40-50 species of plants, and is thought to have been derived from by a mid- to late Miocene cooling that was probably one cause
a diverse, but totally deciduous, broad-leaved vegetation growing of a shift to more deciduous broad-leaved forests in Alaska and
under a mean annual temperature of about 12°C and a warm month the Pacific Northwest, and perhaps a greater importance of mon-
mean temperature around 20°C (Wolfe and Tanai, 1980; Wolfe, tane coniferous forest and steppe vegetation in the northern Rocky
l 994b ). The latitude of this area during the Miocene was presumably Mountains. It is probable that increasing seasonal dryness, in part
close to its present-day position of 60°N. Similar assemblages have due to orographic and edaphic effects caused by Cascade volcanism,
been recovered from Nenana Coal Fields south of Fairbanks and was also a major factor in the spread of the steppes and grasslands
suggest the presence of deciduous forest well inland (Wolfe and in the Pacific Northwest. More open, dry-tolerant vegetation was
Tanai, 1980). On the basis of palynological and some supporting widespread by Barstovian time, but probably was much more ex-
megafossil data, however, Wolfe and Tanai (1980, Figure 7) inferred tensive in the Western Interior, the eastern Rockies, and the South-
the presence of dominantly coniferous vegetation along a coastal west than it was in other parts of the continent. However, even in
strip in southern Alaska extending from Kodiak Island at least as areas that today have an extreme continental climate (e.g., northern
far as the present border with Canada. Mid-Miocene palynofloras Nebraska), fossil floras indicate the existence of riparian forest with
from the Seward Peninsula and the North Slope are also dominated eastern North American affinities growing under a warm climate
by conifers (Wolfe and Tanai, 1980). (MacGinitie, 1962). This may indicate that fingers of broad-leaved
Later Miocene floras from southern Alaska are the basis of the forest extended along river courses much farther into the continent
Homerian Stage (roughly 8-13 Ma; Wolfe et al., 1966, Wolfe, from the East Coast than presently. Floristic and vegetational con-
1994b), the upper part of which is now thought to be approx- nections between eastern and central North America may have been
imately correlative with the Clarendonian or early Hemphillian stronger than between the interior and the West Coast, based both
Land Mammal ages (8-11 Ma; Wolfe, 1994b). Generally, Home- on the composition of Miocene floras and on the floristic affini-
rian floras are much less diverse than the earlier Miocene floras ties of modem vegetation. The steppe and montane conifer forest
such as Seldovia Point and are more heavily dominated by pines of the Rockies evidently formed a major barrier to floral migra-
and hardy deciduous groups like the birch family (Wolfe, Hopkins, tion between developing grasslands on either side of the mountains
and Leopold, 1966). These floral changes have been attributed to an (Leopold and Denton, 1987). The Rockies may have formed a bar-
approximately 7°C decline in mean annual temperature and a 9°C rier to fauna! interchange as well. For instance, differences between
decrease in the mean temperature of the warmest month from the the (coevolved?) herbivore faunas of the Great Plains as opposed
Seldovian to the Homerian (Wolfe, 1972; Wolfe and Tanai, 1980; to the western palouse prairie are suggested by the greater toler-
Wolfe, 1994b). A number of genera of broad-leaved deciduous trees ance of eastern grasses to intense grazing (Mack and Thompson,
have their last Alaskan records during the middle and late Homerian 1982).
(Wolfe, 1994b). Even as late as the Pliocene, when forests in eastern North Amer-
Pliocene floras from southern Alaska are even less diverse than ica were compositionally modern, except for a few "exotic" genera
those of the late Miocene, being dominated largely by pines, willows, now confined to East Asia, the range of forest was probably much
and birches (Wolfe, 1972). Palynofloras of this age demonstrate the greater than it is now, both to the north and perhaps into the conti-
increasing abundance and diversity of herbaceous or shrubby an- nental interior. Although high-latitude vegetation generally became
giosperms such as composites, grasses, and sedges, but there is less diverse, lower statured, and more cold tolerant during the late
no evidence for tundra vegetation in southern Alaska during the Miocene and Pliocene, the shift took place through a series of fluc-
Pliocene (Wolfe, 1972). tuations rather than as a unidirectional trend. Boreal forest existed
Repenning and Brouwers (1992) identified two major cycles of on the shores of the Arctic Ocean as recently as about 3 Ma, in-
climatic cooling between 3 and 0.85 Ma in the area bordering the dicating that boreal forest taxa could have moved across Beringia
Arctic Ocean. About 3 million years ago mixed coniferous forests as late as the Pliocene. Interpretations of late Pliocene vegetation
with a small component of deciduous broad-leaved trees extended in the continental interior of North America are based on slim evi-
essentially to the shores of the Arctic Ocean (80°N), but during dence, but in the early Pliocene it appears that climate was warmer
the succeeding 600,000 years the vegetation shifted toward tundra and forest or woodland vegetation was substantially more prevalent
as permafrost developed at high latitudes. Following this cold pe- than in Holocene times.
 Tertiary vegetation of North America 55

ANIMAL-PLANT INTERACTIONS ciations of plants and animals and may retard climatically induced
change. Although the ultimate importance of climate in controlling
This has been a climate-centered description of vegetational his- plant distributions is not disputed, it is possible that many of the
tory, which is appropriate because climate plays the primary role second-order patterns of floral and fauna! change and stasis during
in dictating vegetational structure and floristic composition. the Tertiary are mediated through the reciprocal effect of plants and
However, evolutionary changes in plant lineages and changes in mammalian herbivores on one another.
the ecological interactions of animals and plants should not be dis-
regarded, even when dealing with only the last 65 Ma. The most
obvious ecological relationship between mammals and plants is her- REFERENCES
bivory. Most considerations of the evolutionary interactions between
mammals and plants implicitly assume that the major direction of Adam, D. P., Bradbur, J. P., Rieck, H. J., & Sarna-Wojcicjk, M. (1990).
influence is from producer to consumer: Changes in plants ("fod- Environmental changes in the Tule Lake Basin, Siskiyou and Modoc
der") require changes in herbivores. No doubt this does occur, but Counties, California, from 3 to 2 million years before present. United
States Geological Survey Bulletin, 1933, 1-13.
the numerous mechanisms in plants that deter herbivores testify
Ash, S. R., & Tidwell, W. D. (1976). Upper Cretaceous and Paleocene flo-
that the interaction is bidirectional, raising the possibility of a series ras of the Raton Basin, Colorado and New Mexico. New Mexico
of responses and counter-responses. Seed dispersal and pollination Geological Society Guidebook, 27th Field Conference, p. 197.
are also significant ecological interactions between mammals and Axelrod, D. I. (1939). A Miocene flora from the western border of the
plants, and vegetation is also important as the substrate for mam- Mohave desert. Carnegie Institute of Washington Publication, 516,
1-128.
malian locomotion. In each of these interactions there is the po-
(1940). The Mint Canyon Flora of southern California: a preliminary
tential for reciprocal effects both on ecological and evolutionary statement. American Journal of Science, 238, 545-51.
time scales. ( 1966). The Eocene Copper Basin flora of northeastern Nevada. Univer-
Although climate, particularly rainfall, is the "first-order" pre- sity of California Publications in Geological Sciences, 59, 1-125.
dictor of vegetational characteristics and distributions, the effect (1968). Tertiary floras and topographic history of the Snake River Basin,
Idaho. Geological Society of America Bulletin, 79, 713-34.
of mammalian herbivores should not be discounted. The effect of
( 1979). Age and origin of Sonoran Desert vegetation. Occasional Papers
large herbivores on vegetation may be particularly important in re- of the California Academy of Sciences, 132, 1-74.
gions where cold or drought reduce levels of primary productivity (1985). Rise of the grassland biome, central North America. Botanical
(Oksanen, 1988), conditions that became increasingly common in Review, 51, 164-201.
the interior of North America during the Neogene and Pleistocene. (1987). The Late Oligocene Creede Flora, Colorado. University of Cal-
ifornia Publications in Geological Sciences, 130, 1-235.
The short-term effect of large grazers and browsers on vegeta-
Axelrod, D. I., & Bailey, H. P. (1976). Tertiary vegetation, climate, and
tional structure is well documented: Generally, they tend to reduce altitude of the Rio Grande depression, New Mexico-Colorado. Pale-
the cover of woody plants and maintain a physically more open habi- obiology, 2, 235-54.
tat (e.g., Laws, 1970; Merton, Bourn, and Hnatiuk, 1976). Several Axelrod, D. I., & Raven, P. H. (1985). Origins of the Cordilleran flora.
authors have suggested that these effects are also important at longer Journal of Biogeography, 12, 21--47.
Axelrod, D. I., & Schorn, H. E. (1994). The 15 Ma floristic crisis at
time scales (e.g., Stebbins, 1981; Janzen and Martin, 1982; Mack
Gillam Spring, Washoe County, northwestern Nevada. PaleoBios, 16,
and Thompson, 1982; Janzen, 1986; Owen-Smith, 1987). If mam- 1-10.
malian herbivores do have long-term effects on vegetation, then the Barnosky, C. W. (1984). Late Miocene vegetational and climatic variations
possibility arises that they may have played an important role in the inferred from a pollen record in northwest Wyoming. Science, 223,
success of certain vegetation types. 49-51.
Basinger, J. F., Mciver, E. E., & LePage, B. A. (1988). The fossil forests of
Considering the entire Tertiary in North America, the overall pat-
Axel Heiberg Island. Musk-Ox, 36, 50-5.
tern has been the restriction of forests and the concomitant spread Becker, H. F. (1969). Fossil plants of the Tertiary Beaverhead Basins in
of open, herbaceous vegetation. It may well be that the limited areas southwestern Montana. Palaeontographica Abt. B, 127, 1-142.
of open vegetation during the early Tertiary precluded the radiation Bell, W. A. (1965). Upper Cretaceous and Paleocene plants of western
of large browsing or grazing herbivores and favored the evolution Canada. Geological Survey of Canada Publication, 65-35, 1--46.
Bennike, 0. (1989). The Kap Kobenhavn Formation: stratigraphy and pale-
of small, frugivorous, terrestrial, arboreal, and volant types. These
obotany of a Plio-Pleistocene sequence in Peary Land, North Green-
small frugivores may in turn have favored the success of the animal- land. Meddelelser om Gronland Geoscience, 23, 3-85.
dispersed plants characteristic of closed forest. This kind of ecosys- Berry, E. W. (1916a). The lower Eocene floras of southeastern North
tem appears to have been dominant over much of the continent America. United States Geological Survey Professional Paper, 91,
during the first 10-20 Ma of the Tertiary. Under the combined influ- 5--481.
(1916b). The physical conditions indicated by the flora of the Calvert
ence of drying and cooling in the mid-Tertiary and Neogene, these
Formation. United States Geological Survey Professional Paper, 98F,
closed-forest ecosystems were fragmented. Plants with shorter life 61-73.
cycles and less woody tissue diversified in this climatic regime, and (1916c). The physical conditions and age indicated by the flora of the
larger grazing and browsing herbivores evo_lved to take advantage Alum Bluff Formation. United States Geological Survey Professio-
of this expanded resource. These large herbivores in turn probably nal Paper, 98-E, 41-59.
(1916d). The flora of the Citronelle Formation. United States Geological
favored the success of such herbaceous or grassland vegetation by
Survey Professional Paper, 98-L, 193-208.
physically disturbing it through feeding. Positive feedback loops like (1930). Fossil plants from the Cypress Hills of Alberta and
these may play a role in the long-term persistence of particular as so- Saskatchewan. National Museum of Canada Bulletin, 63, 15-28.
 Scott L. Wing 56

(1935). A preliminary contribution to the floras of the Whitemud and Dorf, E. (1940). Relationship between floras of type Lance and Fort Union
Ravenscrag Formations. Geological Survey of Canada Memoir, 182, Formations. Bulletin of the Geological Society of America, 51,
1-107. 213-36.
(1937). Tertiary floras of eastern North America. Botanical Reviews, 3, (1951). Lithologic and floral facies in the Paricutin ash deposits, Mexico.
31--46. Transactions of the New York Academy of Sciences, 13, 317-20.
Brown, R. W. (1962). Paleocene flora of the Rocky Mountains and Great (1964). The Petrified Forests of Yellowstone National Park. Washington,
Plains. United States Geological Survey Professional Paper, 375, D.C.: U.S. Government Printing Office.
1-119. Drake, H., & Burrows, C. J. (1980). The influx of potential macrofossils into
Bruns, T. R. (1983). Model for the origin of the Yakutat block, an accreting Lady Lake, north Westland, New Zealand. New Zealand Journal of
terrane in the northern Gulf of Alaska. Geology, 11, 718-21. Botany, 18, 257-74.
Burnham, R. J. (1989a). Relationships between standing vegetation and leaf Elias, M. K. (1932). Grasses and other plants from the Tertiary rocks of
litter in a paratropical forest: implications for paleobotany. Review Kansas and Colorado. The University of Kansas Science Bulletin,
of Palaeobotany and Palynology, 58, 5-32. 33, 333-67.
(1989b). Some Late Eocene depositional environments of the coal- (1935). Tertiary grasses and other prairie vegetation from high plains of
bearing Puget Group of western Washington State, U.S.A. Inter- North America. American Journal of Science, 29, 24-33.
national Journal of Coal Geology, 15, 27-51. Elsik, W. C. (l 968a). Palynology of the Rockdale Lignite, Milan Co., Texas.
( 1993a). Time resolution in terrestrial macrofloras: guidelines from mod- I. Pollen et Spores, 10, 263-314.
ern accumulations. In Taphonomic Approaches to Time Resolution ( l 968b). Palynology of the Rockdale Lignite, Milan Co., Texas. II. Pollen
in Fossil Assemblages, eds. S. M. Kidwell & A. K. Behrensmeyer, et Spores, 10, 599-664.
pp. 57-78. Short Courses in Paleontology, no. 6. Knoxville, Tenn.: Fairchild, W. W., & Elsik, W. C. (1969). Characteristic palynomorphs of
The Paleontological Society. the lower Tertiary in the Gulf Coast. Palaeontographica Abt. B, 128,
(1993b). Reconstructing richness in the plant fossil record. Palaios, 8, 81-89.
376-84. Farley, M. B. (1989). Palynological facies fossils in nonmarine environments
(1994). Paleoecological and floristic heterogeneity in the plant-fossil in the Paleogene of the Bighorn Basin. Palaios, 4, 565-73.
record - an analysis based on the Eocene of Washington. United (1990). Vegetation distribution across the Early Eocene depositional
States Geological Survey Bulletin, 2085B, 1-36. landscape from palynological analysis. Palaeogeography, Palaeo-
Burnham, R. J., & Spicer, R. A. (1986). Forest litter preserved by volcanic climatology, Palaeoecology, ~9, 11-27.
activity at el Chichon, Mexico: a potentially accurate record of the Fields, R. W., Rasmussen, D. L., Tabrum, A. R., & Nichols R. (1985). Ceno-
pre-eruption vegetation. Palaios, 1, 158-61. zoic rocks of the intermontane basins of western Montana and east-
Burnham, R. J., Wing, S. L., & Parker, G. G. (1992). Reflection of temperate ern Idaho. In Cenozoic Paleogeography of the West-central United
forest composition and structure in the litter: implications for the States, eds. M. Flores & S. S. Kaplan, pp. 9-36. Denver: Rocky
fossil record. Paleobiology, 18, 34-53. Mountain Section, Society of Economic Paleontologists and Miner-
Cerling, T. E., Wang, Y., & Quade,J. (1993). ExpansionofC 4 ecosystems as alogists.
an indicator of global ecological change in the late Miocene. Nature, Fleming, R. F., & Pierce, B. S. (1994). Paleocene coals from the Raton For-
341, 344-5. mation, Colorado and New Mexico - palynological and petrographic
Chaney, R. W. (1927). Geology and paleontology of the Crooked River characteristics. Geological Society of America Abstracts with Pro-
Basin, with special reference to the Bridge Creek Flora. Carnegie gram, p. A-31.
Institution of Washington Publication, 346, 45-138. Flores, R. M. (1981). Coal deposition in flu vial paleoenvironments of the Pa-
Chaney, R. W., & Elias, M. K. (1936). Late Tertiary floras from the high leocene Tongue River Member of the Fort Union Formation, Powder
plains. Carnegie Institution of Washington Publication, 4 76, 1-72. River Area, Powder River Basin, Wyoming and Montana. Society
Christophel, D. C. (1976). Fossil floras of the Smokey Tower locality, Al- of Economic Paleontologists and Mineralogists Special Publication,
berta, Canada. Palaeontographica Abt. B, 157, 1--43. 31, 169-90.
Christopher, R. A., Prowell, D. C., Reinhardt, J., & Markewich, H. W. ( 1980). (1983). Basin facies analysis of coal-rich Tertiary fluvial deposits, north-
The stratigraphic and structural significance of Paleocene pollen from ern Powder River Basin, Montana and Wyoming. Special Publica-
Warm Springs, Georgia. Palynology, 4, 105-24. tions of the International Association ofSedimentologists, 6, 501-15.
Collinson, M. E., Fowler, K., & Boulter, M. C. (1981). Floristic changes Frederiksen, N. 0. (l 979a). Paleogene sporomorph biostratigraphy, north-
indicate a cooling climate in the Eocene of southern England. Nature, eastern Virginia. Palynology, 3, 129-67.
291, 315-17. (1979b). Miocene sporomorphs from Massachusetts. Geological Survey
Cowan. D.S. (1982). Geological evidence for post-40 m.y. B. P. large scale Professional Paper, 1150, 215-16.
orthwestward displacement of part of southeastern Alaska. Geol- (1980a). Paleogene sporomorphs from South Carolina and quantitative
ogy, 10, 309-13. correlations with the Gulf Coast. Palynology, 4, 125-79.
Crane, P. R., Manchester, S. R., & Dilcher, D. L. (1990). A preliminary (1980b). Sporomorphs from the Jackson Group (upper Eocene) and ad-
survey of fossil leaves and well-preserved reproductive structures jacent strata of Mississippi and western Alabama. United States Ge-
from the Sentinel Butte Formation (Paleocene) near Almont, North ological Survey Professional Paper, 1084, 1-75.
Dakota. Fieldiana, 20, 1-64. (1980c). Mid-Tertiary climate of southeastern United States: the sporo-
Crepet, W. L., & Feldman, G.D. (1991). The earliest remains of grasses in morph evidence. Journal of Paleontology, 54, 728-39.
the fossil record. American Journal of Botany, 78, 1010-14. (1981 ). Middle Eocene to Early Oligocene plant communities of the Gulf
Cross, A. T., & Martinez-Hernandez, E. (1980). Compositae pollen in early Coast. In Communities of the Past, eds. J. Gray, A. J. Boucot, & W.
Tertiary rocks, Baja California, Mexico. 5th International Palyno- B. N. Berry, pp. 493-550. Stroudsburg, Pa.: Hutchinson Ross.
logical Conference Abstracts, p. 97. (1982). Differences between Middle Eocene plant communities of south-
Dilcher, D. L. (1969). Podocarpus from the Eocene of North America. Sci- ern California and those of the Atlantic and Gulf Coastal Plains.
ence, 164,299-301. Abstracts of the II/rd North American Paleontological Convention,
( 1973). A paleoclimatic interpretation of the Eocene floras of south- p. 10.
eastern North America. In Vegetation and Vegetational History of (1985). Review of early Tertiary sporomorph paleoecology. American
Northern Latin America, ed. A. Graham, pp. 39-59. Amsterdam: Association of Stratigraphic Palynologists Contributions Series, 15,
Elsevier. 1-92.
 Tertiary vegetation of North America 57

(1988). Sporomorph biostratigraphy, floral changes and paleoclimatol- 210-27. Short Courses in Paleontology, no. 6. Knoxville, Tenn.: The
ogy, Eocene and earliest Oligocene of the eastern Gulf Coast. United Paleontological Society.
States Geological Survey Professional Paper, 1448, 1-68. Johnson, K. R., & Hickey, L. J. (1990). Megafloral change across the
(1989). Eocene Sporomorph biostratigraphy of southern California. Cretaceous-Tertiary boundary in the northern Great Plains and Rocky
Palaeontographica Abt. B, 211, 135-79. Mountains, USA. In Global Catastrophes in Earth History: An Inter-
(1991). Pulses of Middle Eocene to earliest Oligocene climatic de- disciplinary Conference on Impacts, Volcanism, and Mass Mortality,
terioration in southern California and the Gulf Coast. Palaios, 6, eds. V. L. Sharpton & P. D. Ward. Geological Society of America
564-71. Special Paper, 247, 433--44.
( 1994 ). Paleocene floral diversities and turnover events in eastern North Johnson, K. R., Nichols, D. J., Attrep, M., Jr., & Orth, C. J. (1989.) High-
America and their relation to diversity models. Review of Palaeob- resolution leaf-fossil record spanning the Cretaceous-Tertiary bound-
otany and Palynology, 82, 225-38. ary. Nature, 340, 708-11.
Frederiksen, N. 0., Carr, D. R., Lowe, G.D., & Wosika, E. P. (1983). Mid- Johnson, S. Y. (1984). Stratigraphy, age and paleogeography of the Eocene
dle Eocene palynomorphs from San Diego, California. American Chuckanut Formation, northwest Washington. Canadian Journal of
Association of Stratigraphic Palynologists Contributions Series, 12, Earth Sciences, 21, 92-106.
1-155.
Kennett, J.P., & Stott, L. D. (1991 ). Abrupt deep-sea warming, palaeoccano-
Fritz, W. J. (1986). Plant taphonomy in areas of explosive volcanism. In graphic changes and benthic extinctions at the end of the Palaeocene'.
Land Plants: Notes for a Short Course, ed. R. A. Gastaldo, pp. 1-9. Nature, 353, 225-9.
Knoxville, Tenn.: The Paleontological Society. Knowlton, F. H. ( 1922). The Laramie flora of the Denver Basin. United ·
Gaponoff, F, S. L. (1984). Palynology of the Silverado Formation (late States Geological Survey Professional Paper, 130, 1-175.
Paleocene), Riverside and Orange Counties, California. Palynology, (1930). The flora of the Denver and associated formations of Colorado.
8, 71-106.
United States Geological Survey Professional Paper, 155, 1-142.
Gastaldo, R. A. (1988). Conspectus of phytotaphonomy. In Methods and Koch, B. E. (1963). Fossil plants from the lower Paleocene of the Agat-
Applications of Plant Paleoecology, eds. W. A. DiMichele & S. L. dalen (Angmartussut) area, central Nugssuaq Peninsula, northwest
Wing. The Paleontological Society Special Publication, 3, 14-28. Greenland. Meddelelser om Gron/and udgivne af Kommissionen for
Gibson, T. G., Andrews, G. W., Bybell, L. M., Frederiksen, N. 0., Hansen, Videnskabelige Undersogelser I Gron/and, 172, 9-120.
T., Hazel, J. E., McLean, J. D., Witmer, M., & Van Nieuwenhuise, Koch, P. L., Zachos, J. C., & Gingerich, P. D. (1992). Correlation between
D.S. (1980). Biostratigraphy of the Tertiary strata of the core. Geol- isotope records in marine and continental carbon reservoirs near the
ogy of the Oak Grove Core. Virginia Division of Mineral Resources Palaeocene/Eocene boundary. Nature, 358, 319-22.
Publication, 20, 1-15. Lakhanpal, R. N. ( 1958). The Rujada flora of west central Oregon. University
Gray, J. (1960). Temperate pollen genera in the Eocene (Claiborne) flora, of California Publications in Geological Sciences, 35, 1-66.
Alabama. Science, 132, 808-10. Laws, R. M. (1970). Elephants as agents of habitat and landscape change in
Greenwood, D.R., & Basinger, J. F. (1994). The paleoecology of high- east Africa. Oikos, 21, 1-15.
latitude Eocene swamp forests from Axel Heiberg Island, Canadian Leopold, E. B., & Denton, M. F. (1987). Comparative age of grassland and
High Arctic. Review of Paleobotany and Palynology, 81, 83-97. steppe east and west of the northern Rocky Mountains. Annals of the
Gregory, K. M. (1994). Palaeoclimate and palaeoelevation of the 35 Ma Missouri Botanical Garden, 74, 841-67.
Florissant flora, Front Range, Colorado. Palaeoclimates, !, 23-57. Leopold, E. B., Liu, G., & Clay-Poole, S. (1992). Low-biomass vegetation in
Gregory, K. M'., & Chase, C. G. (1992). Tectonic significance of paleob- the Oligocene? In Eocene-Oligocene Climatic and Biotic Evolution,
otanically estimated climate and altitude of the Late Eocene erosion eds. D. R. Prothero & W. A. Berggren, pp. 399--420. Princeton:
surface, Colorado. Geology, 20, 581-5. Princeton University Press.
Greller, A. M., & Rachele, L. D. (1983). Climatic limits of exotic genera Leopold, E. B., & MacGinitie, H. D. (1972). Development and affinities
in the Legler palynoflora, Miocene, New Jersey, U.S.A. Review of of Tertiary floras in the Rocky Mountains. In Floristics and Pale-
Palaeobotany and Palynology, 40, 149-63. ojloristics of Asia and Eastern North America, ed. A. Graham, pp.
Hickey, L. J. (1977). Stratigraphy and paleobotany of the Golden Valley 147-200. Amsterdam: Elsevier.
Formation (early Tertiary) of western North Dakota. The Geological Lerbekmo, J. F., Sweet, A. R., & St. Louis, R. M. (1987). The relation-
Society of America Memoir, 150, 1-181. ship between the iridium anomaly and palynological floral events
(1980). Paleocene stratigraphy and flora of the Clark's Fork Basin. Uni- at three Cretaceous-Tertiary boundary localities in western Canada.
versity of Michigan Papers on Paleontology, 24, 33--49. Geological Society of America Bulletin, 99, 325-30.
Hickey, L. J., Johnson, K. R., & Dawson, M. R. (1988). The stratigraphy, Lindsey, D. A., Glanzman, R. K., Naeser, C. W., & Nichols, D. J. (1980).
sedimentology, and fossils of the Haughton Formation: a post-impact Upper Oligocene evaporites in basin fill of Sevier Desert region,
crater-fill, Devon Island, N.W.T., Canada. Meteoritics, 23, 221-31. western Utah. The Mountain Geologist, 17, 251-60.
Hickey, L. J., West, R. M., Dawson, M. R., & Choi, D. K. (1983). Arc- MacGinitie, H. D. (1941). A middle Eocene flora from the central Sierra
tic terrestrial biota: paleomagnetic evidence of age disparity with Nevada. Carnegie Institute of Washington Publication, 534, 1-178.
mid-northern latitudes during the late Cretaceous and Early Tertiary. (1953). Fossil plants of the Florissant Beds, Colorado. Carnegie Institu-
Science, 221, 1153-6. tion of Washington Publication, 599, 1-198.
Hollick, A. (1882). Palaeobotany of the Yellow Gravel at Bridgeton, N.J. (1962). The Kilgore flora. A late Miocene flora from northern Nebraska.
Bulletin of the Torrey Botanical Club, 19, 330-3. University of California Publications in Geological Sciences, 35,
Janzen, D. H. (1986). Chihuahuan desert nopaleras: defaunated big mammal 67-158.
vegetation. Annual Review of Ecology and Systematics, 17, 595-636. (1969). The Eocene Green River flora of northwestern Colorado and
Janzen, D. H., & Martin, P. S. (1982). Neotropical anachronisms: the fruits northeastern Utah. University of California Publications in Geolog-
the gomphotheres ate. Science, 215, 19-27. ical Sciences, 83, 1-202.
Johnson, K. R. ( 1992). Leaf-fossil evidence for extensive floral extinction at Mack, R. N., & Thompson, J. N. (1982). Evolution in steppe with a few
the Cretaceous-Tertiary boundary, North.Dakota, USA. Cretaceous large, hooved mammals. American Naturalist, 119, 757-73.
Research, 13, 91-117. Manchester, S. R. (1981). Fossil plants of the Eocene Clarno nut beds. Ore-
(1993). Time resolution and the study of late Cretaceous and early Ter- gon Geology, 43, 75-81.
tiary megafloras. In Taphonomic Approaches to Time Resolution in (1994). Fruits and seeds of the middle Eocene nut beds flora, Clarno
Fossil Assemblages, eds. S. M. Kidwell & A. K. Behrensmeyer, pp. Formation, Oregon. Palaeontographica Americana, 58, 1-205.
 Scott L. Wing 58

Manchester, S. R., & Meyer, H. W. (1987). Oligocene fossil plants of the Oksanen, L. (1988). Ecosystem organization: mutualism and cybernetics or
John Day Formation, Fossil, Oregon. Oregon Geology, 49, 115-27. plain Darwinian struggle for existence? The American Naturalist,
Mazer, S. J., & Tiffney, B. H. (1982). Fruits of Wetherellia and Palae- 131,42~4.
owetherellia (? Euphorbiaceae) from Eocene sediments in Virginia Owens, J. P., Bybel, L. M., Paulachok, G., Ager, T. A., Gonzalez, V. M., &
and Maryland. Brittonia, 34, 300-33. Sugarman, P. J. (1988). Stratigraphy of the Tertiary sediments in a
Mccartan, L., Tiffney, B. H., Wolfe, J. A., Ager, T. A., Wing, S. L., Sirkin, 945-foot-deep corehole near Mays Landing in the southeastern New
L. A., Ward, L. A., & Brooks, J. (1990). Late Tertiary floral assem- Jersey Coastal Plain. United States Geological Survey Professional
blage from upland gravel deposits of the southern Maryland coastal Paper, 1484, 1-39.
plain. Geology, 18, 311-14. Owen-Smith, N. (1987). Pleistocene extinctions: the pivotal role of mega-
McClammer, J. U., & Crabtree, D.R. (1989). Post-Barremian (Early Creta- herbivores. Paleobiology, 13, 351-62.
ceous) to Paleocene paleobotanical collections in the western inte- Peterson, G. L., & Abbott, P. L. (1979). Mid-Eocene climatic change, south-
rior of North America. Review of Paleobotany and Palynology, 57, western California and northwestern Baja California. Palaeogeogra-
221-32. phy, Palaeoclimatology, Palaeoecology, 26, 73-87.
Mcintyre, D. J., Fowler, M. G., Kalkreuth, W. D., Richardson, R. J. H., Plafker, G. (1984). Comment on: "Model for the origin of the Yakutat block,
& Riediger, C. L. (1994). Petrological, palynological and geochem- an accreting terrane in the northern Gulf of Alaska." Geology, 12,
ical characteristics of Eureka Sound Group coals, Stenkul Fiord, 563.
Ellesmere Island, Arctic Canada. Geological Society of America Pocknall, D. T. ( 1987). Paleoenvironments and age of the Wasatch Formation
Abstracts with Program, p. A-32. (Eocene), Powder River Basin, Wyoming. Palaios, 2, 368-76.
Mciver, E., & Basinger, J. F. (1993). Flora of the Ravenscrag Formation Potbury, S. S. (1935). The La Porte flora of Plumas County, California.
(Paleocene), southwestern Saskatchewan, Canada. Palaeontograph- Carnegie Institute of Washington Publication, 465, 29-82.
ica Canadiana, IO, 1-168. Povey, D. A. R., Spicer, R. A., & England, P. C. (1994). Palaeobotanical in-
Merton, L. F. H., Bourn, D. M., & Hnatiuk, R. J. (1976). Giant tortoise and vestigation of early Tertiary palaeoelevations in northeastern Nevada:
vegetation interactions on Aldabra Atoll. I. Biological Conservation, initial results. Review of Palaeobotany and Palynology, 81, 1-10.
9, 293-304. Rachele, L. D. (1976). Palynology of the Legler lignite: a deposit in the
Meyer, H. (1973). The Oligocene Lyons flora of northwestern Oregon. The Tertiary Cohansey Formation of New Jersey, USA. Review of Pale-
Ore Bin, 35, 37-51. obotany and Palynology, 22, 225-52.
Meyer, H. W. (1986). An Evaluation of the Methods for Estimating Pale- Repenning, C. A., & Brouwers, E. M. (1992). Late Pliocene-early Pleis-
oaltitudes Using Tertiary Floras from the Rio Grande Rift Vicinity, tocene ecologic changes in the Arctic borderland. United States Ge-
New Mexico and Colorado. Unpublished Ph.D. Thesis. Berkeley: ological Survey Bulletin, 2036, 1-37.
University of California. Retallack, G. J. (1992). Paleosols and changes in climate and vegetation
(1992). Lapse rates and other variables applied to estimating paleoalti- across the Eocene/Oligocene boundary. In Eocene-Oligocene Cli-
tudes from fossil floras. Palaeoclimatology, Palaeoecology, Palaeo- matic and Biotic Evolution, eds. D. R. Prothero & W. A. Berggren,
geography, 99, 71-99. pp. 382-98. Princeton: Princeton University Press.
Miller, C. N., Jr. (1977). Pityostrobus lynni (Berry) comb. nov., a pinaceous Richmond, G. M., Mullenders, W., & Coremans, M. (1978). Climatic impli-
seed cone from the Paleocene of Virginia. Bulletin of the Torrey cations of two pollen analyses from newly recognized rocks of latest
Botanical Club, 104, 5-9. Pliocene age in the Washburn Range, Yellowstone National Park,
Miller, K. G. (1992). Middle Eocene to Oligocene stable isotopes, climate, Wyoming. Geological Survey Bulletin, 1455, 1-13.
and deep-water history: the Terminal Eocene Event? In Eocene- Rouse, G. E., & Mathews, W. H. (1987). Palynology and geochronology of
Oligocene Climatic and Biotic Evolution, eds. D.R. Prothero & W. Eocene beds from Cheslatta Falls and Nazko areas, central British
A. Berggren, pp. 160-77. Princeton: Princeton University Press. Columbia. Canadian Journal of Earth Sciences, 25, 1268-76.
Miller, K. G., Fairbanks, R. G., & Mountain, G. S. (1987). Tertiary oxygen Smiley, C. J. (ed). (1985). Late Cenozoic History of the Pacific Northwest.
isotope synthesis, sea level history, and continental margin erosion. San Francisco: Pacific Division American Association for the Ad-
Paleoceanography, 2, 1-19. vancement of Science, pp. 9--417.
Myers, J. A. (1991). The Early Middle Eocene Torrey flora, Delmar, CA. Spicer, R. A. (1981). The sorting and deposition of allochthonous plant
In Eocene Geologic History, San Diego Region, eds. P. L. Abbott material in a modern environment at Silwood Lake, Silwood Park,
& J. A. May, pp. 201-16. Los Angeles: Pacific Section Society of Berkshire, England. United States Geological Survey Professional
Economic Paleontologists and Mineralogists. Paper, 1143, 1-77.
Nambudiri, E. M., Tidwell, W. D., Smith, B. N., & Hebbert, N. P. (1978). A Spicer, R. A., & Wolfe, J. A. (1987). Plant taphonomy of late Holocene
C4 plant from the Pliocene. Nature, 276, 816-17. deposits in Trinity (Clair Engle) Lake, northern California. Paleobi-
Newman, K. R. (1981). Palynologic biostratigraphy of some early Tertiary ology, 13, 227--45.
nonmarine formations in central and western Washington. Geologi- Spicer, R. A., Wolfe, J. A., & Nichols, D. J. (1987). Alaskan Cretaceous-
cal Society of America Special Paper, 184, 49-65. Tertiary floras and Arctic origins. Paleobiology, 13, 73-83.
Nichols, D. J. (1987). Palynology of the Vermilion Creek coal bed and Srivastava, S. K. (1972). Some spores and pollen from the Paleocene Oak
associated strata. In Geological Investigations of the Vermilion Creek Hill Member of the Naheola Formation, Alabama (U.S.A.). Review
Coal Bed in the Eocene Niland Tongue of the Wasatch Formation, of Paleobotany and Palynology, 14, 217-85.
Sweetwater County, Wyoming, eds. H. W. Roehler & P. L. Martin. Stebbins, G. L. (1981). Coevolution of grasses and herbivores. Annals of the
United States Geological Survey Professional Paper, 1314, D49- Missouri Botanical Garden, 68, 5-86.
D65. Sweet, A. R., & Braman, D.R. (1992). The KIT boundary and contiguous
Nichols, D. J., Jarzen, D. M., Orth, C. J., & Oliver, P. Q. (1986). Paly- strata in western Canada: interactions between paleoenvironments
nological and iridium anomalies at Cretaceous-Tertiary boundary, and palynological assemblages. Cretaceous Research, 13, 31-79.
south-central Saskatchewan. Science, 231, 714-17. Taggart, R. E., Cross, A. T., & Satchell, L. (1982). Effects of periodic vol-
Novacek, M. J., & Lillegraven, J. A. (1979). Terrestrial vertebrates from canism on Miocene vegetation distribution in eastern Oregon and
the later Eocene of San Diego County, California: a conspectus. western Idaho. Proceedings of the II/rd North American Paleonto-
In Eocene Depositional Systems, San Diego, California, ed. P. L. logical Convention, 2, 535--40.
Abbott, pp. 69-79. Los Angeles: Pacific Section Society of Eco- Thomasson, J. R. (1980). Paleoagrostology: a historical review. Iowa State
nomic Paleontologists and Mineralogists. Journal of Research, 54, 301-17.
 Tertiary vegetation of North America 59

(1982). Fossil grass anthoecia and other plant fossils from arthropod bur- (1996). Global warming and plant species richness - a case study of the
rows in the Miocene of western Nebraska. Journal of Paleontology, Paleocene/Eocene boundary. In Biodiversity //: Understanding and
56, 1011-17. Protecting Our Natural Resourses, eds. M. Reaka-Kudla, D. Wilson,
(1986). Fossil grasses: 1820-1926 and beyond. In Grass Systematics & E. 0. Wilson, pp. 163-85. Washington, D.C.: National Academy
and Evolution, eds. T. R. Soderstron, K. W. Hilu, C. S. Campbell, Press.
& M. E. Barkworth, pp. 159-167. Washington, D.C.: Smithsonian Wing, S. L., Alroy, J., & Hickey, L. J. (1995). Plant and mammal diversity
Institution Press. in the Paleocene to early Eocene of the Bighorn Basin. Palaeogeog-
Thomasson, J. R., Nelson, M. E., & Zakrzewski, R. J. ( 1986). A fossil grass raphy, Palaeoclimatology, Palaeoecology, 115, 117-56.
(Gramineae: Chloridoideae) from the Miocene with Kranz anatomy. Wing, S. L., & Bown, T. M. (1985). Fine scale reconstruction of late
Science, 233, 876-8. Paleocene-early Eocene paleogeography in the Bighorn Basin of
Thomasson, J. R., Zakrzewski, R. J., Lagarry, H. E., & Mergen, D. E. ( 1990). northern Wyoming. In Cenozoic Paleogeography of West-Central
A late Miocene (late early Hemphillian) biota from northwestern United States, eds. R. M. Flores & S. S. Kaplan, pp. 93-105. Den-
Kansas. National Geographic Research, 6, 231--44. ver: Rocky Mountain Section, Society of Economic Paleontologists
Tidwell, W. D., Ash, S. R., & Parker, L. R. (1981). Cretaceous and Tertiary and Mineralogists.
floras of the San Juan Basin. In Advances in San Juan Basin Paleon- Wing, S. L., Bown, T. M., & Obradovich, J. D. (1991). Early Eocene biotic
tology, eds. S. G. Lucas, J. K. Rigby, Jr., & B. S. Kues, pp. 307-32. and climatic change in interior western North America. Geology, 19,
Albuquerque: University of New Mexico Press. 1189-92.
Tidwell, W. D., Thayn, G. F., & Roth, J. L. (1976). Cretaceous and early Wing, S. L., & Greenwood, D.R. (1993). Fossils and fossil climate: the case
Tertiary floras of the intermountain area. Brigham Young University for equable continental interiors in the Eocene. In Palaeoclimates
Geology Studies, 22, 77-98. and Their Modeling with Special Reference to the Mesozoic Era,
Tiffney, B. H. (1985a). Perspectives on the origin of the floristic similarity eds. J. R. L. Allen, B. J. Hoskins, B. W. Sellwood, & R. A. Spicer.
between eastern Asia and eastern North America. Journal of the Philosophical Transactions of the Royal Society, London, Series B.,
Arnold Arboretum, 66, 73-94. Biological Sciences, 341, 243-52.
(1985b ). The Eocene North Atlantic land bridge: its importance in Ter- Wolfe, J. A. (1964). Miocene floras from Fingerrock Wash southwestern
tiary and modern phytogeography of the northern hemisphere. Jour- Nevada. United States Geological Survey Professional Paper, 454-N,
nal of the Arnold Arboretum, 66, 243-73. 1-33.
(1994). Re-evaluation of the age of the Brandon Lignite (Vermont, USA) (1966). Tertiary plants from the Cook Inlet Region, Alaska. United States
based on plant megafossils. Review of Paleobotany and Palynology, Geological Survey Professional Paper, 398-B, 1-32.
82, 299-315. ( 1972). An interpretation of Alaskan Tertiary floras. In Floristics and
Tiffney, B. H., Wolfe, J. A., Ager, T. A., & Wing, S. L. (1989). Late Miocene Paleofloristics of Asia and Eastern North America, ed. A. Graham,
(?) flora of the Brandywine Clay, Maryland. American Journal of pp. 201-33. Amsterdam: Elsevier.
Botany, 76, 234. ( 1975). Some aspects of plant geography of the northern hemisphere dur-
Thompson, R. S. (1991). Pliocene environments and climates in the western ing the Late Cretaceous and Tertiary. Annals of the Missouri Botan-
United States. Quaternary Science Reviews, 10, 115-32. ical Garden, 62, 264-79.
Tschudy, R.H. ( 1973). Stratigraphic distribution of significant Eocene paly- (1977). Paleogene floras from the Gulf of Alaska region. United States
nomorphs of the Mississippi embayment. United States Geological Geological Survey Professional Paper, 997, 1-108.
Survey Professional Paper, 743-B, 19-21. (1978). A paleobotanical interpretation of Tertiary climates in the north-
(1975). Normapolles pollen from the Mississippi Embayment. United ern hemisphere. American Scientist, 666, 94-703.
States Geological Survey Professional Paper, 865, 1--42. (1979). Temperature parameters of humid to mesic forests of eastern Asia
Tschudy, R. H., Pillmore, C. L., Orth, C. J., Gilmore, J. S., & Knight, J. and relation to forests of other regions of the northern hemisphere and
D. (1984). Disruption of the terrestrial plant ecosystem at the Cre- Australasia. United States Geological Survey Professional Paper,
taceous-Tertiary boundary, western Interior. Science, 225, 1030-2. 1106, 1-37.
Wang, Y., Cerling, T. E., & MacFadden, B. J. (1994). Fossil horses and (1981). Paleoclimatic significance of the Oligocene and Neogene floras
carbon isotopes: new evidence for Cenozoic dietary, habitat, and of the northwestern United States. In Paleobotany, Paleoecology, and
ecosystem changes in North America. Palaeogeography, Palaeocli- Evolution, ed. K. J. Niklas, pp. 79-102. New York: Praeger.
matology, Palaeoecology, 107, 269-79. (1985). Distribution of major vegetational types during the Tertiary. Geo-
Westgate, J. W., & Gee, C. T. (1990). Paleoecology of a middle Eocene physical Monograph, 32, 357-75.
mangrove biota (vertebrates, plants, and invertebrates) from south- (1987a). Late Cretaceous-Cenozoic history of deciduousness and the
west Texas. Palaeogeography, Palaeoclimatology, Palaeoecology, terminal Cretaceous event. Paleobiology, 13, 215-26.
78, 163-77. (1987b). An overview of the origins of the modern vegetation and flora
Wheeler, E. A. (1991). Paleocene dicotyledonous trees from the Big Bend of the northern Rocky Mountains. Annals of the Missouri Botanical
National Park, Texas: variability in wood types common in the Late Garden, 74, 785-803.
Cretaceous and Early Tertiary, and ecological inferences. American ( 1992). Climatic, floristic, and vegetational changes near the Eoc
Journal of Botany, 78, 658-71. ene/Oligocene boundary in North America. In Eocene-Oligocene
Wheeler, E. A., & Landon, J. (1992). Late Eocene (Chadronian) dicotyle- Climatic and Biotic Evolution, eds. D.R. Prothero & W. A. Berggren,
donous woods from Nebraska: evolutionary and ecological signifi- pp. 421-36. Princeton: Princeton University Press.
cance. Review of Paleobotany and Palynology, 74, 267-82. (1993). A method of obtaining climatic parameters from leaf assem-
Wheeler, E. F., Scott, R. A., & Barghoorn, E. S. (1977). Fossil dicotyledonous blages. United States Geological Survey Bulletin, 2040, 1-71.
woods from Yellowstone National Park. Journal of the Arnold Ar- ( l 994a). Tertiary climatic changes at middle latitudes of western North
boretum, 58, 280-302. America. Palaeogeography, Palaeoclimatology, Palaeoecology, 108,
Wheeler, E. F., Scott, R. A., & Barghoorn, E. S. (1978). Fossil dicotyledonous 195-205.
woods from Yellowstone National Park, II. Journal of the Arnold (1994b ). An analysis of Neogene climates in Beringia. Palaeogeography,
Arboretum, 59, 1-26. Palaeoclimatology, Palaeoecology, 108, 207-16.
Wing, S. L. (1987). Eocene and Oligocene floras and vegetation of the north- Wolfe, J. A., Hopkins, D. M., & Leopold, E. B. (1966). Tertiary stratigraphy
ern Rocky Mountains. Annals of the Missouri Botanical Garden, 74, and paleobotany of the Cook Inlet region, Alaska. United States
748-84. Geological Survey Professional Paper, 398-A, 1-29.
 ~ ~ ~ ~
~
~
,- '°
- ~
,- ~
,- ~
,-
~~Q9cl'~ ~~~~C)~~ en
n
~
1'"9..>~ = ?>8 l>l"?'-
.?>::::" g
~=·:::r-o..---="
(1)
~ s·
z. .~
~ s· .
...,.
s· : ~ ~ ~ ~ r-'
21~::<..=Pi'o,s.::i"t:l'.:"r;r~
§t0Qg~r:;~§Q8r::;On
_8. -~ P' 2. '.:" ~- 5 t; §" jil ~ ~ Q ~
:::>
~ OO()-CJ Q""1• ~°:< (JQ
;_ I g .., Y1 =s. P 0 ~l:::i~Q\
· t ::s r.n
t;~s:-~ c;;~~·z· s·N~~;_
oo · (1) oo· o ~ · n -
a o· -b ~ !:ti ~- ~t11 --
te @ ~
n=s~r
s·
D,_gs.?S S·;fj
;l Oo o-"'::i::~ C1Q
~ -. ~ ~ ;: ~· ~
~;. =r',.::Snes'rD
"~ a:~3Q
g· S" "·~~a. §3
. en 0 "t:I" . e: " "
c::: >i:l g. E. 3 ~ "" -· g
16" ;:;<ll o"«" '3 =o
en ~·a @~~E:-:r:; g.::s
~ ~ ::s 8 ~ ~ s· (1) ::
Vi 0
"g: " - " "' =
< § -g_ Q ~- ~ -·
~ § g ~8.
;;;· ~g. ~p ~o.. g~
~ C'lo ~>i:i ""5 ao-
s· 8..., ::ti*' z= as.
::ti (1)
o.sIS" 5' g g
p
2. "
;::;· g
=: . ~. ;::l. ';' 0
~::ti ri· o n 3 >i:i 16 ..., ::i.
e.. : : -: 0 ~ :::i (') ;:t> cts.
0 V:i CJQ - -· - ::::-: - ::s
3 ~g ~p ~~ ~a.
~ ~ " o. '° I 5l i'I" ~
g ~ ?S p 8. ~ g- p:: g

~ ~ ~ 0
°'
&.
n ,
~ -;:::
::!.P""t i::ti'""" r:'.'.l'-"O l :-<'.l::l"'
'° ,~
~~~rD;:t>~~';-'~-~ -§(1)
~-~]_;--(1) C'J!""""~~
· en ~ g: i<:o ;:; n ~ g · . . . -·
~~ ~ "w ::4.
s· :p· a Jg ~ ~ Q
-ooi::::i:: ""'"'"
c-.o;:; S::r- "
n;:t> "c::oo
oo'"C:IUli: ::
o%~-g 1
_ ..,.....:Jg 3 ~g.--?
'.:"·
~::r::;::--i~" S'"l>
(1)
N ~ ~ ~ ;_ . ~· ri ~ [
- o:-cor - cl'~ n?'"· ~
-~-..,~
Uloot.Jl::: J.....:J 5""=s @a P
~::S ..-+-.
fs:s~-=-- o%g ~~ ~
~ ~ .._ ~ ~ ~ 3 g~ ~-
b' I -· o. °" Ej "' ..., I:>.
~. 0~ ~ ~ ~- ~ -::: ~
-
O' 0
?om ;::;-n :4'°
~ ::::-: ; · 00
-~
(il c:::tl ~g ~~ C'l
~ ~· ~ 5"" (t Ci
~ ~
:; ~&. ~: g~ ~
(1) V:i g Q. ::s [ ~ -.
~ ~ Q §' ~ e; ~- §..
< ~ (i" i::::i (1) ':< g V:i
a: C'l8° §.~ '.<:' :<:
('i) ~::s c::is i::::ic.
::s c::i 0
~
OQ .-+ - ~· ~
8 ~ ~ ~ g· ~ ~ ~
-· --0 2' 0.. " ::t. :.-c
;.oi e..§ s~ id~ ~
E; V:i ~ ~ 5· __.+;;.. ::s ~
a ~ a ~ ~ ~ ~ §.
5· ~ 3 ~- g f 9 e..
 Appendix Partial list of Tertiary plant fossil localities in North America

# Locality/Flora m/p* NALMA** Epoch or Ma Inferred vegetation Paleoclimate Lat Long Reference
1 POLICE IS p ~64 conifer swamp forest warm temperate 65
2 CASTLE RIV p -125.2 Sweet & Braman, 1992
~64 conifer swamp forest
3 JUDY CRK NORTH warm temperate 49.5 -114 Sweet & Braman, 1992
p ~64 conifer swamp forest
4 LUSCA R STERCO warm temperate 54.5 -115.4 Sweet & Braman, 1992
p ~64 conifer swamp forest
5 RED DEER V KNUDS N warm temperate 53 -116.7 Sweet & Braman, 1992
p ~64 conifer swamp forest
6 CYPRESS MAP FV warm temperate 51.9 -113 Sweet & Braman, 1992
p ~64 conifer swamp forest
7 WOOD MTNCR K warm temperate 49.4 -108.4 Sweet & Braman, 1992
p ~64 conifer swamp forest
8 ROCK CRK EAST warm temperate 49.4 -106.3 Sweet & Braman, 1992
p ~64 conifer swamp forest
9 MARMATH warm temperate 49 -105.5 Sweet & Braman, 1992
m ~64 mixed conifer/broadleaved subtropical 46.5 -103.8 Johnson, 1992
deciduous forest
10 SMOKY TOWER m ~63 conifer swamp forest warm temperate 54.3 -118.7
11 UPPER ATANIKERDLUK m Christophel, 1976
PAL broadleaved deciduous forest warm temperate
12 CHICKALOON 70.3 -53 Koch, 1963
m LPAL broadleaved deciduous forest
13 ELSINO RE warm temperate 61.8 -148.8 Wolfe et al., 1966
m L_PAL savanna forest
14 STENKUL FIORD tropical 33.7 -117.3 Axelrod, 1979
m/p LPAL polar broadleaved deciduous temperate 79 -82.5 Hickey et al., 1983;
& conifer swamp forest
15 SILVERADO p/m Mcintyre et al., 1994
LPAL mixed evergreen & deciduous subtropical 33.5 -117.6 Gaponoff, 1984
broadleaved forest
16 GENESEE m PAL conifer swamp forest warm temperate
17 BEARD EN m 53.4 -114.4 Chandrasekharam, 1974
CF LPAL mixed evergreen & deciduous warm temperate 47.3 -102.1 Hickey, 1977
broadleaved forest
18 ALMON T m CF L_PAL broadleaved deciduous forest warm temperate 46.9 -101.4
19 MONTGOMERY CRK m CF-WA Crane et al., 1990
PAL-EO mixed evergreen & deciduous 40.8 -121.9
broadleaved forest
20 LOWER WILLWOOD m WA E_EO conifer swamp forest subtropical 44.3 -108.1 Bown et al., 1994;
21 MIDDL E WILLWOOD m WA E_EO Wing et al., 1995
mixed evergreen & deciduous subtropical 44.3 -108.4 Bown et al., 1994;
broadleaved forest
22 UPPER WILLWOOD m WA ~52.8
Wing et al., 1995
mixed evergreen & deciduous subtropical 44.2 -108.5 Bown et al., 1994;
broadleaved forest
23 WIND RNER FLORA m WA E_EO Wing et al., 1995
broadleaved evergreen forest paratropical 43.7 -110
24 CAMELS BUITE m/p WA E_EO
MacGinitie, 1969
mixed evergreen & deciduous subtropical 46.8 -103 Hickey, 1977; Bebout, 1978
broadleaved forest
25 YELLOWSTONE m/p WA E_EO mixed evergreen & deciduous subtropical 44.9 -110.5 Knowlton, 1899;
broadleaved forest
26 MEGAN OS"C" p WA E_EO
Wheeler et al., 1977, 1978
mixed evergreen & deciduous paratropical 37.9 -121.8 Lucas-Clark & Lampley, 1988
broadleaved forest
27 CHALK BLUFFS m WA E_EO mixed evergreen & deciduous paratropical 39.17 -120.85 MacGinitie, 1941
broadleaved forest

(cont.)
Appendix (cont.)

# Locality/Flora m/p* NALMA** Epoch or Ma Inferred vegetation Paleoclimate Lat Long Reference

28 FRANKLIN BLUFFS p WA E_EO broadleaved deciduous forest warm temperate 69.8 -148.7
29 SEWARD PEN DV p WA E_EO broadleaved deciduous forest warm temperate 65.1 -162.3 - Ager & Kauffman, 1987;
Dickinson et al., 1987
30 NANJEMOY p/m WA E_EO mangrove subtropical 38.4 -77 Mazer & Tiffney, 1982
31 TUFF LAKE BEDS m WA-BR E/M_EO mixed evergreen & deciduous subtropical 44 -109
broadleaved forest
32 COAL MINE CANY p WA-BR E/M_EO broadleaved deciduous forest warm temperate 41 -115.5 Wingate, 1983
33 BOYSEN m BR M_EO palm swamp forest paratropical 43.4 -108.1 Wing, 1987
34 TORREY p BR M_EO mixed evergreen & deciduous paratropical 33 -117.3 Myers, 1991
broadleaved forest
35 KISINGER LAKES m/p BR M_EO mixed evergreen & deciduous subtropical 43.8 -110 MacGinitie, 1974
broadleaved forest
36 REPUBLIC m BR ~49 mixed conifer/broadleaved temperate 48.71 -118.91 Wolfe & Wehr, 1987
deciduous forest
37 BEARPAW MTNS m BR M_EO mixed evergreen & deciduous subtropical 48.5 -109.9 Brown & Pecora, 1949
broadleaved forest
38 SMITHERS BR-UI M_EO 53.66 -125.27 Wolfe, 1987
39 CHUCHUACRK m BR-UI M_EO 51.22 -119.79 Wolfe, 1987
40 KAMLOOPS m BR-UI M_EO 50.51 -119.93 Wolfe, 1987
41 PRINCETON m BR-UI M_EO 49.21 -120.07 Wolfe, 1987
42 PURYEAR m BR-UI M_EO mixed evergreen & deciduous subtropical 36.4 -88.4 Dilcher, 1973; Crepet
broadleaved forest & Taylor, 1986
43 WARDELL RANCH m/p BR-UI M_EO woodland subtropical 39.8 -107.9 MacGinitie, 1969
44 AXEL HEIBERG m/p BR-UI M_EO mixed conifer/broadleaved temperate 80 -88 Basinger et al., 1988;
deciduous forest Mcintyre, 1990
45 CLARNO NUT BEDS m UI ~43.7 broadleaved evergreen forest subtropical 44.8 -120.4 Manchester, 1990
46 THUNDER MT DEWEY m UI ~45 evergreen coniferous forest cold temperate 44.44 -114.32 Axelrod, 1990
47 THUNDER MT ROAD m UI ~45 mixed conifer/broadleaved temperate 44.44 -114.32 Axelrod, 1990
deciduous forest
48 RATE HOMESTEAD m UI M_EO mixed evergreen & deciduous subtropical 43.8 -107.5 MacGinitie, 1969
broadleaved forest
49 CHUMSTICK I m UI ~43 mixed evergreen & deciduous subtropical 47.6 -120.5 Evans, 1991
broadleaved forest
50 COPPER BASIN m UI ~41 mixed conifer/broadleaved cold temperate 41.8 -114.1 Axelrod, 1966
deciduous forest
51 CHUMSTICK II m UI ~41 broadleaved deciduous forest warm temperate 47.6 -120.5 Evans, 1991
52 SILICA BAR m DU ~40 mixed evergreen & deciduous 47.7 -122 Burnham, 1990
broadleaved forest
53 BARILLA m DU ~38.2 mixed evergreen & deciduous subtropical 30.7 -103.8 Berry, 1919; Axelrod
broadleaved forest & Bailey, 1976
54 SHEEP ROCK CRK m DU-CH M-LEO 43.91 -120.59
55 ALVORDCRK m DU-CH M-LEO 42.47 -118.46 Axelrod, 1944
56 ELKO m/p DU-CH M-LEO 40.76 -115.77
57 WARMAN m DU-CH LEO mixed evergreen & deciduous subtropical 36.3 -88.5 Dilcher, 1973; Crepet
broadleaved forest & Taylor, 1986
58 UPPER BULL RUN m CH ~36 evergreen coniferous forest cold temperate 41.8 -116.2 Axelrod, 1966
59 FLORISSANT m/p CH ~35 mixed evergreen & deciduous warm temperate 39.2 -104.4 MacGinitie, 1953
broadleaved forest
60 BEAVERCRK CH LEO 46.03 -112.88
61 DRUMMOND CH LEO 46.41 -113.09 Pearson, 1972
62 SALMON m CH LEO 45.32 -113.96 Brown, 1937
63 COWCRK m CH LEO 44.91 -113.58
64 BEAVERHEAD m ? 45.15 -113.35
65 METZEL RANCH m ? 44.65 -112.8
66 SUSANVILLE m EO broadleaved evergreen forest paratropical 40.3 -120.6 Wolfe-USGS Collections
67 ELLESMERE m/p EO polar broadleaved deciduous forest warm temperate 79 -85 Hickey et al., 1983
68 GOSHEN EO 43.9 -123
69 COMSTOCK EO 43.8 -123.1
70 LWR CEDARVILLE EO 41.4 -120.2
71 RUJADA m OLI 43.67 -122.56 Lakhanpal, 1958
72 LYONS m OLI 43.82 -121.43 Meyer, 1973
73 YAQUINA m OLI 43.44 -124.18 McClammer, 1978
74 LAPORTE m OR-WH E_OLI mixed evergreen & deciduous paratropical 39.4 -120 Potbury, 1935
broadleaved forest
75 UPPER RUBY RIVER m WH OLI mixed conifer/broadleaved temperate 45.4 -112.1 Becker, 1961
deciduous forest
76 CREEDE 5-MILE m AR ~27.2 woodland cold temperate 36.9 -106.9 Wolfe & Schorn, 1990
77 CREEDE WASON m AR ~27.2 chaparral cold temperate 36.9 -106.9 Wolfe & Schorn, 1990
78 CREEDE BIRDSEY m AR ~27.2 coniferous forest cold temperate 36.9 -106.9 Wolfe & Schorn, 1990
79 CHEENEETNUK R m OLI 63.15 -151.61 Wolfe-USGS Collections
80 CHEENEETNUK R m OLI 63.39 -151.35 Wolfe-USGS Collections
81 COLORADO CRK OLI 61.91 -150.19 Wolfe-USGS Collections
82 EAGLE RIVER OLI 61.12 -149.83 Wolfe et al., 1966
(cont.)
Appendix (cont.)

# Locality/Flora m/p* NALMA** Epoch or Ma Inferred vegetation Paleoclimate Lat Long Reference

83 HARRIET POINT OU 59.94 -152.74 Wolfe-USGS Collections

84 HARRIETCRK OU 60.06 -152.61 Wolfe-USGS Collections
85 REDOUBT PT OU 60.38 -152.32 Wolfe-USGS Collections
86 DOUGLAS MTN OU 58.88 -153.39 Wolfe-USGS Collections
87 SITKlNAKIS OU 56.55 -154.l Wolfe-USGS Collections
88 KUKAKBAY OU 58.11 -154.62 Knowlton, 1904
89 KUKAKBAY OU 58.88 -153.39 Knowlton, 1904
90 KOOTZNAHOO OU 57.94 -134.71 Lathram et al., 1965
91 SOOKE OU 48.42 -123.53 Lamotte, 1935
92 GUMBOOT MTN OU 45.82 -122.3
93 CASCADE LOCKS OU 45.59 -121.95
94 SANDSTONE CRK OU 45.44 -122.43
95 CASCADIA m OU 44.09 -122.33
96 WILLAMETTE OU 44.2 -123.11
97 SHALE CITY OU 42.35 -122.46
98 BRIDGECRK m OU 45.35 -120.39 Chaney, 1927
99 BRIDGECRK m OU 45.35 -120.2 Chaney, 1927
100 ALVORDCRK m AR ~23.8 42.6 -118.6 Axelrod, 1944; Barnett
& Fisk, 1984
101 HAUGHTON ASTROBLEME m/p AR ~22.4 mixed conifer/broadleaved cold temperate 75.4 -89.8 Hickey et al., 1988
deciduous forest
102 SELDOVIA POINT m AR-HE E_MIO broadleaved deciduous forest temperate 59.5 -151.8 Wolfe & Tanai, 1980
103 EASTGATE m HE ~18 woodland temperate 39.4 -117.8 Axelrod, 1989
104 MIDDLEGATE m HE ~18 mixed conifer/broadleaved temperate 39 -118.3 Axelrod, 1989
deciduous forest
105 BUFFALO CANYON HE ~18 mixed conifer/broadleaved temperate 39.2 -118.1 Axelrod, 1991
deciduous forest
106 TEHACHAPI m HE ~17 woodland subtropical 35.3 -118.7 Axelrod, 1939
107 GRAND COULEE m M_MIO mixed evergreen & deciduous warm temperate 48.2 -120.1 Wolfe & Tanai, 1980
broadleaved forest
108 FINGERROCK m HE-BA ~16.5 38.7 -117.9
109 CLARKIA m/p BA ~15.5 47.2 -116.3
110 MASCALL m BA ~15 conifer swamp forest temperate 44.7 -119.7 Chaney, 1959
111 STEWART SPRING m BA ~14 38.5 -117.9
112 SUCCORCRK m/p BA ~13.5 43.5 -117.2
113 TROUTCRK m BA ~13.5 42.4 -118.4
 114 VANTAGE m BA ~13 47 -120.3

115 TEMBLOR BA ~13

36.5 -120.8
116 ALDRICH STATION m BA-CL ~12
38.5 -119 Axelrod, 1989
117 ESMERALDA m BA-CL ~12
38.2 -117.8 Axelrod, 1989
118 STINKING WATER m CL ~11 broadleaved deciduous forest temperate 43.7 -118.2 Chaney, 1959
119 TRAPPERCRK CL ~11 broadleaved deciduous forest temperate 42.4 -113.3 Axelrod, 1964
120 ELLENSBURG m CL ~11
47 -120.6
121 SANPABLO/NEROLY m CL ~10
38.4 -122.8
122 MINIUM QUARRY m HM ~7 grassland/riparian trees warm temperate 39 -100 Thomasson et al., 1990
123 WEAVERVILLE HF MIO 40.5 -123.2 MacGinitie, 1937
124 WEAVERVILLE RC MIO 44.7 -122.9 MacGinitie, 1937
125 BLUEMTNS MIO mixed conifer/broadleaved temperate 44.4 -119 Chaney, 1959
deciduous forest
126 GOOSECRK MIO 42.3 -113.7
127 CHLOROPAGUS m MIO 39.8 -118.9
128 VERDI m MIO 39.4 -119.8
129 FALLON MIO 39.4 -119
130 KAP KOBENHAVN m BL ~2.25 tundra/boreal coniferous forest cold temperate 82.5 -21.5 Bennike, 1989

*m = megaftora, p = palynoftora.
**NA~MA = North American land mammal "age"; CF - Clarkforkian, WA - Wasatchian, BR- Bridgerian, UI - Uintan, DU - Duchesnean, CH - Chadronian, OR_ Orellan, WH
- Wh1tneyan, AR - Arikareean, HE - Hemphillian, BA - Barstovian, CL - Clarendonian, HM - Hemingfordian, BL - Blancan.
 3 The Pleistocene terrestrial mammal fauna of North America

RUSSELL W. GRAHAM

INTRODUCTION PLIO-PLEIS TOCENE TRANSITION AND

FAUNALINTERCHANGE
The Pleistocene was characterized by repeated and rapid climatic
fluctuations that caused the growth and retreat of massive continental The beginning of the Pleistocene is stratigraphically fixed as the base
glaciers in the Northern Hemisphere. Long cold intervals (glacials) of the Calabrian marine stage in Italy (Kurten and Anderson, 1980).
were punctuated by shorter warm periods (interglacials). However, The Pleistocene has been provisionally defined as beginning about
glaciation was not restricted to the Pleistocene but extended well into 1.64 Ma (million years ago) (Aguirre and Pasini, 1985) and ending
the Tertiary. Richmond and Fullerton (1986, p. 7) state, "The pre- around 10 Ka (thousand years ago) (Hopkins, 1975). For North
Illinoian time interval now is known to be represented by [terrestrial] America, the Plio-Pleistocene boundary falls near the boundary be-
deposits of at least five Pleistocene glaciations and one Pliocene tween the Blancan (Pliocene) and Irvingtonian (early Pleistocene)
glaciation in Iowa and Nebraska and by at least seven Pleistocene Land Mammal ages (LMA) (see Figure 3.1). Conventionally, the
glaciations and at least three Pliocene glaciations in the country Blancan/Irvingtonian boundary was defined by Savage (1951) as the
[United States] as a whole." Oxygen isotope studies offoraminifera first North American appearance of the genus Mammuthus (mam-
from deep-sea cores (Shackleton and Opdyke, 1973; Shackleton, moth) and the absence of Bison, which immigrated to North Amer-
1987) suggest that Pleistocene climatic fluctuations were even more ica during the later Rancholabrean LMA (late Pleistocene). Even
frequent and dynamic. However, Pleistocene climates were merely though Mammuthus is an index genus for the Irvingtonian, it is
an extension oflong-term climatic trends that began in the Tertiary. apparently rare or absent at the beginning of the Irvingtonian but
These climatic fluctuations had direct and indirect effects on the became widespread and abundant later (Lindsay, 1978) (but see
terrestrial biota and thus they caused significant changes in the evo- Lambert and Shoshani, this volume, Chapter 43). One of the earli-
lution and distributions of species. Many species that evolved in est records for Mammuthus in North America appears to be from the
the Pleistocene, especially those from the Northern Hemisphere, Leisey Shell Pit local fauna, Hillsborough County, Florida (Webb
were adapted to cold environments (e.g., woolly mammoth [Mam- and Dudley, 1995), which may date between 1.55 and 1.1 Ma (latest
muthus primigenius], woolly rhinoceros [Coelodonta antiquitatis], early Irvingtonian) (Morgan and Hulbert, 1995).
peccaries [Mylohyus spp. and Platygonus spp.], musk oxen [Ovibos The disappearance ofB!ancan taxa and the appearance oflrving-
moschatus, Bootherium bombifrons, Euceratherium collinum], tonian taxa was gradual (Lundelius et al., 1987). It has been difficult,
etc.). They frequently had vicars in warmer climates, especially therefore, to precisely place the boundary between the Blancan and
in the Southern Hemisphere, which survived into the Holocene Irvingtonian (Figure 3.2). However, the fauna! distinctness of the
(e.g., elephants [Loxodonta spp.], rhinoceros [Diceros and Cera- Blancan/Irvingtonian boundary is greater than either the preceding
totherium], javelinas [peccaries], [Tayassu spp.], and cattle [Bos Hemphillian/Blancan or following Blancan/Rancholabrean fauna!
spp.], etc.). In some cases, conspecific taxa occupied both the colder changes. Lundelius et al. (1987) succinctly summarize some of the
and warmer climates (e.g., Eurasian cave lions, American lions, fauna! changes:
and African lions [Panthera lea], spotted hyaena [Crocuta c rocuta],
hippopotamus [Hippopotamus amphibius], etc.). One species in Those [genera] that became extinct at or before the end of
particular, Homo sapiens, evolved in warmer climates during the the Gauss Chron include Dipoides, Ogmodontomys, Nebras-
Pleistocene but rapidly spread to colder areas at more northern lat- komys, Pliolemmus, Pliopotamys, and Pratilepus. Those
itudes and eventually became the dominant species throughout the that disappeared during the early Matuyama Chron at or be-
world. fore the Olduvai magnetic subchron include Paenemarmota,

66
 The Pleistocene terrestrial mammal fauna of North America 67

~ Time
·;::
Land Morine
0 Scale Time Divisions Oxygen Relative
0 Age Select Mammal (USA)
Epoch Chrons a... ( M.Y.) Subchrons Isotope Po Ieat emperotures
Ages (xlO~ears Stages cold
worm

Q)
c Roncholobreon Q) Late
Q) c:

-(.)

Ul
Q)
0

a... c
0.73
0 · 88
0.94 Jaramillo
Irvington ion
4

:::c
Q)
0
0
0
Middle
(H~psi-
t ermol)
E Early

1.72 Late
Olduvai Glacial
1.88 14
Full
2.47 Glacial 2
20 c:
32 c
c:
Q) Bloncon (/)
c c:
Q)
0 c: 3
(.) 3.40 u c
0 64 (/) c:
a... ~
(/)
c: 4
3.97 0
75 u
-t
.c
4.39
~
(/)
a
b
>-
....
5 c
c
w
d
122
5.26
Hemphilion
Q)
Sangamon e
c
Q)
132
(.)
0 6
~
c: Late
c
Figure 3.1. Geomagnetic time scale and land-mammal ages for the late 0
Cenozoic. Modified after Lindsay et al. (1987, Figures 9.1 and 9.8,
225 c: 7
pp. 270, 279). Cross-hatched areas represent transitional boundaries be- -
tween land-mammal ages. Absolute dates for Hemphillian/Blancan and
Blancan/Irvingtonian transitional boundaries are from Lundelius et al.
(1987, pp. 216, 221).
Pre- lllinoian
Procastoroides, Prodipodomys, Ophiomys, Mictomys (Meta- Figure 3.2. Climatic fluctuations (cold/warm), North American glacial/
xyomys), Hypolagus, Borophagus, Rhynchotherium, Equus interglacial stages, land-mammal ages, magnetochrons, and marine oxygen
(Dolichohippus), and Nannippus (p. 219), ... while charac- isotope stages for the last 300 Ka.

teristic Irvingtonian genera such as Mammuthus, Eucerathe-

rium, Soergelia, Tetrameryx, Smilodon, and Lepus began to Webb (1969) believes that small herbivores, especially microtine
appear in North American faunas. (p. 220) rodents, may have replaced many of the larger ones. Although the
Plio-Pleistocene changes were gradual, the rate of climatic and en-
The gradual nature of the fauna! changes at the Plio-Pleistocene vironmental changes were accelerated throughout the Pliocene and
boundary reinforce the view that Pleistocene climatic and environ- Pleistocene, especially the last 400 Ka.
mental changes are rooted in the Tertiary. The continued demise of As in the Tertiary, the Pleistocene mammalian fauna of North
the widespread Miocene savanna fauna of the midcontinent (Webb, America was affected by fauna! exchanges with other continental ar-
1985a) is part of this process. These environmental changes re- eas. These interchanges occurred between both Eurasia (Beringian
sulted in reducing the diversity of such groups as horses, camels, Interchange) and South America (Great American Interchange).
and gomphotheres in the late part of the Tertiary, and consequently They have been extensively documented by other authors (Repen-
only limited numbers of these taxa survived into the Pleistocene. ning, 1967; Stehli and Webb, 1985; Kontrimavichus, 1986), and
 Russell W. Graham 68

only a few salient features are mentioned here. First, the numbers Anderson, 1980; Lundelius et al., 1983, FAUNMAP Working
of taxa interchanged and the directions of immigration were not Group, 1994). Fine-scale temporal resolution allows for detailed
uniform for these three continental areas (Vrba, 1992). As summa- paleoecological and paleoenvironmental reconstructions for thin
rized by Webb (1985b, p. 211), of the 120 known genera of North slices of time.
American mammals during the Irvingtonian, about one third are The terrestrial biota of North America could have responded
new appearances, and of those about half were immigrants (twelve to fluctuating environmental conditions of the Pleistocene in two
from Eurasia and eight from South America). Similarly, during distinctly different ways. One hypothesis, a community unit model,
the Rancholabrean, North America received sixteen genera from proposes that large groups of species (i.e., "communities") shifted
Asia and nine genera from South America. It is quite possible that as tightly linked and highly coevolved assemblages. An individu-
fewer genera arrived from South America because several of the alistic model, another hypothesis, suggests that individual species
nine South American genera are only known as living taxa. Webb responded to these changes in accordance with their own tolerance
(1985b, p. 211) believes these taxa may have actually immigrated limits. The terms deterministic and stochastic have also been used
in either the Blancan and/or Irvingtonian. to describe these two patterns (see Blonde!, 1987). I have not used
Some of the important immigrants to arrive in North America these terms because the individualistic response may not be com-
from South America during the Pleistocene are two sloth gen- pletely stochastic and deterministic factors (directional climate
era (Nothrotheriops and Eremotherium), the opossum (Didelphis), changes) may drive the individualistic process.
anteaters (Myrmecophaga), and a variety of hystricognath rodents The community unit model predicts that communities are sta-
(e.g., Erethizon). Mammoths (Mammuthus), bison (Bison), saiga ble, tightly linked, highly coevolved assemblages of species that
(Saiga), muskoxen (Euceratherium and Soergelia), and humans have maintained equilibrium for long periods of time. Furthermore,
(Homo sapiens) were some of the Eurasian immigrants that en- it suggests that modern communities can be used as direct ana-
tered North America during the Pleistocene. Of particular interest, logues in paleoenvironmental reconstructions. The individualistic
because of their paleoclimatic, paleoecologic, and biochronologic model suggests that communities are loosely organized collections
importance, are the microtine rodents. Repenning (1987) believes of species whose coexistence depends on their tolerance limits and
there were at least eight separate microtine rodent invasion events subsequent distribution along environmental gradients. Therefore,
during the last 6 million years and four of these events occurred they are not highly coevolved systems and their association may
within the last 2 million years. These invasions are interpreted by appear ephemeral in geologic time. Modern communities may not
Repenning ( 1987, p. 240) as representing "southward dispersals dur- serve as direct analogues for past environments.
ing cooler periods occurring with the expansion of the arctic steppe Palynological studies clearly demonstrate that late Quaternary cli-
from a single Siberian-Beringian arctic fauna that encircled more matic changes did not cause simple latitudinal or altitudinal shifts in
than half of the Holarctic world" (Repenning, 1984 ). In each case, vegetational zones (Davis, 1976, 1981; Colinvaux, 1987; Jacobson,
more advanced forms invaded North America from Asia, resulting Webb, and Grimm, 1987; Webb, Hartlein, and Kutzbach, 1987).
in partial extinction of the then endemic microtine taxa (Repenning, Instead, individual species responded to environmental changes.
1987, p. 239). As previously indicated, Webb (1969) suggested that Mammals have also shown an individualistic response to late Qua-
microtine rodents may have filled the niches of many of the large ternary climatic changes (Graham, 1986b; Graham and Grimm,
herbivores of the Tertiary. 1990). As a result, Pleistocene plant (Overpeck, Webb, and Webb,
1992) and animal communities (Lundelius et al., 1983) frequently
contain species whose ranges do not overlap today (Figure 3.3),
LATE RANCHOLABREAN FAUNAS and therefore these Pleistocene communities do not have mod-
ern analogues. These nonanalogue communities frequently contain
The Rancholabrean LMA, named by Savage (1951) for the fauna combinations of species that today inhabit arctic tundra, grass-
from the Rancho La Brea tar pits of California, is characterized by land, and forest (both coniferous and deciduous) environments.
the presence of Bison. The boundary between the Rancholabrean Nonanalogue associations are characteristic of both small-mammal
and Irvingtonian is extremely diffuse, and, as a result, it may be (Graham, 1986b) and large-mammal, for example, arctic steppe
arbitrarily drawn at different times based on the preferences of dif- (Guthrie, 1982), guilds.
ferent investigators (see discussions by Lundelius et al. [1987] and Because seasonal extremes, and not annual means, are the im-
Repenning [1987]). In Figure 3.1, I have arbitrarily placed the portant climatic variables limiting the distributions of organisms,
boundary of the Rancholabrean at the base of the Brunhes Chron coexistence of species in nonanalogue Pleistocene communities
in accordance with Lindsay et al. (1987, Figure 9.8, p. 279). In may have resulted from more equable climates than those of to-
any case, the rate of climatic change increased significantly dur- day (Hibbard, 1960; Axelrod, 1967; Graham and Mead, 1987).
ing the late Pleistocene (Figure 3.2). Changes in the late Ran- Equability refers to the reduction in differences between seasonal
cholabrean mammalian fauna, especially for the last 40 Ka, are extremes, especially for winter and summer temperatures (Graham
well known because of the fine-scale resolution of the radiocar- and Mead, 1987, p. 372; Figure 2). Specifically, it appears that for
bon chronology and because of the vast numbers and geographic most of North America during the Pleistocene, summers were cooler
distribution of the faunas known from this time period (Kurten and than today but winters may not have been any colder. However,
 The Pleistocene terrestrial mammal fauna of North America 69

80 80 60
+
-+ +
+
+
+
(J i-

!
;

50
50
x

" i
I
\

40
40

i-

·,
I

30
30 !-

-1-
~
Figure 3.3. Modem distributions of
three North American small mam-
··1. 20 mals: smoky shrew (Sorex fumeus),
20 .\
thirteen-line ground squirrel (Sper-
+
+ .. mophillus tridecemlineatus), and col-
lared lemming (Dicrostonyx spp.), and
'\
which were found together in de-

ID 0 K)O 500 IOOO Mites

~ Sf1.1me1.1s
posits in a fossil site in western
Pennsy 1vania. The differences in the
k> (j Str1decemlineat1.1s modern distributions illustrate the in-
dividualistic response to changing cli-
~ D. h1.1dsoni1.1s matic conditions at the end of the
120 100
Pleistocene. (Modified after Graham,
1986a)

the equable climates of the Pleistocene were decidedly different Climatic warming during the late glacial (14-10 Ka) caused in-
from warm equable climates of the Eocene and Miocene, but Pleis- dividual species distributions to change along environmental gradi-
tocene climates may share some similarities with the late Oligocene ents in different directions, at different rates, and during different
(Markwick, 1994). times (Lundelius et al., 1983; Graham, 1986b; Graham and Mead,
Although seasonal extremes were reduced, environmental 1987). The individualistic response of the biota resulted in the emer-
gradients did exist. For instance, the full glacial (ca. 18 Ka) gence of new community patterns. Changes in the distributions of
mammalian faunas of the northern United States do indicate that individual taxa continued during the Holocene (the last 10 Ka), but
an environmental gradient extended along the ice front from east to the magnitude of these changes was not the same as those of the
west (Graham and Mead, 1987). In the northeastern United States, late glacial (Semken, 1983). Most modern biomes and ecosystems
tundra mammals were primarily intermixed with a few grassland in North America are probably less than 10,000 years old.
forms. From the upper Midwest to the northern plains, the number
of deciduous forest species decreased and the number of grassland
species increased significantly. These regional differences in the TERMINAL PLEISTOCENE EXTINCTION
temperate tundralike environments were also reflected in paleoveg-
etational patterns. Jacobson, Webb, and Grimm (1987) consider At the end of the Pleistocene, about 10 Ka, there was an extinction
the western region of the ice front as forb-dominated tundra and the event that decimated the large terrestrial mammalian herbivore and
eastern region as sedge-dominated tundra. carnivore guilds, especially in North America, South America, and
 Russell W. Graham 70

Australia. This extinction event appears to have happened over a REFERENCES

short interval of time, perhaps 2,000 years (Martin, 1984 ). In gen-
eral, there are two primary schools of thought with regard to the Aguirre, E., & Pasini, G. (1985). The Pliocene-Pleistocene boundary.
"cause(s)" of the extinction (Martin and Klein, 1984). Some scien- Episodes, 8, 116-20.
Axelrod, D. (1967). Quaternary extinctions oflarge mammals. University of
tists believe these extinctions are the direct result of overexploitation
California Publications in Geological Sciences, 74, 1-42.
oflarge mammal herbivores for food by Ice Age hunters. This model
Blonde!, J. (1987). From biogeography to life history theory: a multithematic
has been referred to as "overkill" or "blitzkrieg" (Martin, 1984). approach illustrated by the biogeography of vertebrates. Journal of
Others, including myself, attribute the demise of the Pleistocene Biogeography, 14, 405-22.
mammals to changing climatic and environmental conditions. Colinvaux, P. ( 1987). Amazon diversity in light of the paleoecological record.
It has been difficult for climatic models of extinction to isolate a Quaternary Science Reviews, 6, 93-114.
Davis, M. B. (1976). Pleistocene biogeography of the temperate deciduous
single cause. Climatically driven changes in floral communities and
forests. Geoscience and Man, 13, 13-26.
the consequences for plant-animal interactions may provide insight Davis, M. B. (1981). Quaternary history and the stability of forest com-
into this problem (Graham, l 986a). Vegetational changes may have munities. In Forest Succession, Concepts, and Application, eds. D.
altered herbivore diets and caused heightened periods of competition C. West, H. H. Shugart, & D. B. Botkin, pp. 132-53. New York:
(Graham and Lundelius, 1984). Also, vegetational reassortment at Springer-Verlag.
FAUNMAP Working Group. ( 1994). FAUNMAP- a database documenting
the end of the Pleistocene may have resulted in lowering the effective
late Quaternary distributions of mammal species in the United States.
nutritional value of the vegetational mosaic, which reduced carrying Illinois State Museum Scientific Papers, XXV (nos. I & 2).
capacity and decreased the variety of palatable plants available for Graham, R. W. (1986a). Plant-animal interactions and Pleistocene extinc-
herbivores, especially hind-gut fermenters (Guthrie, 1982, 1984). tions. In Dynamics ofExtinctions, ed. D. K. Elliott, pp. 131-54. New
Reduction in growing season at the end of the Pleistocene would York: Wiley.
(1986b). Response of mammalian communities to environmental
have also created environmental stress for large herbivores (Guthrie,
changes during the late Quaternary. In Community Ecology, eds.
1982, 1984). J. Diamond & T. J. Case, pp. 300-13. New York: Harper &
Although the large herbivores may have been able to adapt to any Row.
one of these environmental perturbations, the combination of all of Graham, R. W., & Grimm, E. C. (1990). Effects of global climate change on
them at the same time (10-12 Ka) may have been devastating for the patterns of terrestrial biological communities. Trends in Evolution
and Ecology, 5, 289-92.
the megaherbivore guild. Undoubtedly, pressures exerted by human
Graham, R. W., & Lundelius, E. L., Jr. (1984). Coevolutionary disequi-
predation may have also added to the environmental stresses, but librium and Pleistocene extinctions. In Quaternary Extinctions: A
I believe they were less significant than the environmental fluctua- Prehistoric Revolution, eds. P. S. Martin & R. G. Klein, pp. 223-49.
tions. Owen-Smith (1987, 1988) has proposed the "keystone herbi- Tucson: University of Arizona Press.
vore" hypothesis of extinction, linking both environmental and hu- Graham, R. W., & Mead, J. I. (1987). Environmental fluctuations and evo-
lution of mammalian faunas during the last deglaciation in North
man factors. This hypothesis assumes that human predation causes
America. In North America and Adjacent Oceans During the Last
the extinction of the largest herbivores (e.g., proboscideans) and as Deglaciation, eds. W. F. Ruddiman & H. E. Wright, Jr., pp. 371-402.
a consequence of eliminating these keystone species, vegetational Boulder, Colo.: Geological Society of America.
communities are significantly altered, which creates a cascading ef- Grayson, D. K. (1991). Late Pleistocene mammalian extinctions in North
fect of extinction for other smaller species. The problem with this America: taxonomy, chronology, and explanations. Journal ofWorld
Prehistory, 5, 193-231.
hypothesis is that the proboscideans, the largest Pleistocene her-
Guthrie, R. D. (1982). Mammals of the mammoth steppe as paleoenviron-
bivores in North and South America, are one of the last species, mental indicators. In Paleoecology of Beringia, eds. D. M. Hopkins,
instead of the first as required by the hypothesis, to become extinct J. V. Matthews, Jr., C. E. Schweger, & S. B. Young, pp. 307-28. New
(Grayson, 1991). Both the climatic and human models suggest York: Academic Press.
that extinction of the large carnivores is a result of the demise of (1984). Mosaics, allelochemics and nutrients. In Quaternary Extinc-
tions: A Prehistoric Revolution, eds. P. S. Martin & R. G. Klein, pp.
their primary prey, the megaherbivores. Whatever the cause of the
259-98. Tucson: University of Arizona Press.
extinction, the Holocene megaherbivore guilds are depauperate in Hibbard, C. W. (1960). Pliocene and Pleistocene climates in North Amer-
comparison to those of the Pleistocene or Tertiary, and biodiver- ica. Annual Report of the Michigan Academy of Science, Arts, and
sity of the modern mammalian fauna has been shaped by long-term Letters, 62, 5-30.
environmental changes of the past. Hopkins, D. M. (1975). Time-stratigraphic nomenclature for the Holocene
epoch. Geology, 3, 10.
Jacobson, G. L., Jr., Webb, T., III, & Grimm, E. C. (1987). Patterns and
rates of vegetation change during the deglaciation of eastern North
ACKNOWLEDGMENTS America. In North America and Adjacent Oceans During the Last
Deglaciation, eds. W. F. Ruddiman & H. E. Wright, Jr., pp. 277-88.
I wish to thank S. D. Webb, C. A. Repenning, and E. L. Lundelius, Jr. Boulder, Colo.: Geological Society of America.
for their valuable comments on a previous draft of this manuscript. I Kontrimavichus, V. L. (1986). Beringia in the Cenozoic Era. Russian Trans-
thank Julianne Snider for drafting the figures as well as Mary Ann lations Series 28. Rotterdam: A. A. Balkema.
Kurten, B., & Anderson, E. (1980). Pleistocene Mammals ofNorth America.
Graham and Malinda Aeilou for assistance in manuscript prepara-
New York: Columbia University Press.
tion. I also appreciate editorial comments by Christine Janis and Lindsay, E. (1978). Late Cenozoic vertebrate faunas, southeastern Arizona.
the patience of both Christine Janis and Louis Jacobs during the New Mexico Geological Society Guidebook, 29th Field Conference,
preparation of this manuscript. Land of Cochise: 269-75.
 The Pleistocene terrestrial mammal fauna of North America 71

Lindsay, E. H., Johnson, N. M., Opdyke, N. D., & Butler, R. F. (1987).

pp. 236-68. Berkeley: University of California Press.
Mammalian chronology and the magnetic polarity time scale. In
Richmond, G. M., & Fullerton, D. G. (1986). Introduction to Quaternary
Cenozoic Mammals of North America, ed. M. 0. Woodbume, pp.
glaciations in the United States of America. In Quaternary Glacia-
269-84. Berkeley: University of California Press.
tions in the Northern Hemisphere, eds. V. Sibrava, D. Q. Bowen, &
Lundelius, E. L., Jr., Downs, T., Lindsay, E. H., Semken, H. A., Zakrzewski,
G. M. Richmond, pp. 3-10. Oxford: Pergamon Press.
R., Churcher, C. S., Harington, C. R., Schultz, C. E., & Webb, S. Savage, D. E. (1951). Late Cenozoic vertebrates of the San Francisco Bay
D. (1987). The North American Quaternary sequence. In Cenozoic
region, California. University of California Publications Bulletin De-
Mammals of North America, ed. M. 0. Woodburne, pp. 211-35. partment of Geological Sciences, 28, 215-314.
Berkeley: University of California Press.
Semken, H. A., Jr. (1983). Holocene mammalian biogeography and climatic
Lundelius, E. L., Jr., Graham, R. W., Anderson, E., Guilday, J., Holman, J.
change in the eastern and central United States. In Late Quaternary
A., Steadman, D. W., & Webb, S. D. (1983). Terrestrial vertebrate Environments of the United States, Vol. 2: The Holocene, ed. H.
faunas. In Late Quaternary Environments of the United States, Vol. E. Wright, Jr., pp. 182-207. Minneapolis: University of Minnesota
I: The Late Pleistocene, ed. S. C. Porter, pp. 311-53. Minneapolis: Press.
University of Minnesota Press.
Shackleton, N. J. (1987). Oxygen isotopes, ice volume and sea level. Qua-
Markwick, P. J. (1994). "Equability," continentality, and Tertiary "climate": ternary Science Reviews, 6, 183-90.
the crocodilian perspective. Geology, 22, 613-16.
Shackleton, N. J., & Opdyke, N. D. (1973). Oxygen isotope and paleo-
Martin, P. S. (1984). Prehistoric overkill: the global model. In Quaternary
magnetic stratigraphy of equatorial Pacific core V28-238; oxygen
Extinctions: A Prehistoric Revolution, eds. P. S. Martin & R. G. isotope temperatures and ice volumes on a 105 and 106 year scale.
Klein, pp. 354-403. Tucson: University of Arizona Press. Quaternary Research, 3, 39-55.
Martin, P. S., & Klein, R. G. (eds.). (1984). Quaternary Extinctions: A Stehli, F. G., & Webb, S. D. (eds.). (1985). The Great American Biotic
Prehistoric Revolution. Tucson: University of Arizona Press. Interchange. New York: Plenum.
Morgan, G. S., & Hulbert, H. C., Jr. (1995). Overview of the geology and Vrba, E. (1992). Mammals as a key to evolutionary theory. Journal ofMam-
vertebrate biochronology of the Leisey Shell Pit local fauna, Hillsbor- malogy, 73, 1-28.
ough County, Florida. Bulletin Florida Museum of Natural History, Webb, S. D. (1969). Extinction-origination equilibria in late Cenozoic land
37, Part I, 1-92.
mammals of North America. Evolution, 23, 688-702.
Overpeck, J. T., Webb, R. S., & Webb, T., III (1992). Mapping eastern North
( l 985a). Main pathways of mammalian diversification in North America.
American vegetation change of the past 18 ka: no-analogs and the
In The Great American Biotic Interchange, eds. F. G. Stehli & S. D.
future. Geology, 20, 1071-4.
Webb, pp. 201-17. New York: Plenum.
Owen-Smith, N. (1987). Pleistocene extinctions: the pivotal role of mega- ( 1985b). Late Cenozoic mammal dispersals between the Americas. In
herbivores. Paleobiology, 13, 351-62.
The Great American Biotic Interchange, eds. F. G. Stehli & S. D.
(1988). Megaherbivores: The Influence of Very Large Body Size on Webb, pp. 357-86. New York: Plenum Press.
Ecology. Cambridge: Cambridge University Press. Webb, S. D., & Dudley, J.P. (1995). Proboscidea from the Leisey Shell Pits,
Repenning, C. A. (1967). Palearctic-Nearctic mammalian dispersal in the Hillsborough County, Florida. Bulletin Florida Museum of Natural
late Cenozoic. In The Bering Land Bridge, ed. D. M. Hopkins, pp. History, 37, Part II, 645-60.
288-311. Stanford: Stanford University Press.
Webb, T., III, Bartlein, P. J., & Kutzbach, J. E. (1987). Climatic change in
(1984). Quaternary rodent biochronology and its correlation with cli- eastern North America during the past 18,000 years; comparisons
matic and' magnetic stratigraphies. In Quaternary Chronologies, ed. of pollen data with model results. In North America and Adjacent
W. C. Mahaney, pp. 105-18. Norwich: Geo.
Oceans During the Last Deglaciation, eds. W. F. Ruddiman & H.
(1987). Biochronology of the microtine rodents of the United States. E. Wright, Jr., pp. 447-62. Boulder, Colo.: Geological Society of
In Cenozoic Mammals of North America, ed. M. 0. Woodburne, America.
 Part II: Carnivorous mammals

4 Carnivorous mammals

CHRISTINE M. JANIS, JON A. BASKIN, ANNALISA BERTA, JOHN J. FLYNN, GREGG F. GUNNELL,
ROBERT M. HUNT, JR., LARRY D. MARTIN, and KATHLEEN MUNTHE

INTRODUCTION Janis, 1995). All known families of creodonts and carnivorans, with
the exception of the carnivoran families Viverridae (civets, etc.) and
Carnivorous mammals are usually assumed to be those within the Herpestidae (mongooses), have been found in the Tertiary of North
orders Carnivora and Creodonta. Of course, not all members of America, although the Hyaenidae are represented only by a sin-
these orders have a strictly carnivorous diet: Most members of the gle Pliocene genus. Oxyaenids (Creodonta) and canids (Carnivora)
Procyonidae and Ursidae are omnivorous rather than carnivorous, probably originated in North America: Their earliest records and
and one ursid (the giant panda) is a strict herbivore. Among ex- main centers of radiation and diversification are in North America.
tant carnivores, the Felidae is the only family whose members are Ursids also probably originated in North America, although this was
all strictly carnivorous (hypercarnivorous sensu Van Valkenburgh, not their main center of radiation (Baskin and Tedford, 1996). Al-
1991). Likewise, mammals in other orders may take meat in their though procyonids (Carnivora) have an Old World origin and do not
diet. Among extant mammals, the marsupial superfamily Dasy- appear in North America until the Miocene, today they are known
uroidea includes species that are almost exclusively carnivorous (or almost exclusively from the New World.
insectivorous a~ smaller body sizes). Among other eutherians, many
species may include meat in their diet: Occasional meat eating has
even been observed in duikers (bovids in the tribe Cephalophini) SYSTEMATICS
(Estes, 1990).
Large, specialized meat-eating carnivores did not exist in the ear-
INTERRELATIONSHIPS OF CREODONTA
liest Tertiary of North America. Although viverravid miacoids are
AND CARNIVORA
known from the early Paleocene (Torrejonian), they appear to have
been small, generalized carnivores, rather like the present-day viver- The relationship between the Creodonta and the Carnivora has been
rids and mustelids. Larger, obviously predatory, creodonts or car- subject to much debate historically. Cope (1875) intially coined the
nivorans are not known until the late Paleocene, with the radiation term "Creodonta" to encompass certain taxa originally thought to
of oxyaenids including larger (puma-sized) taxa such as Dipsali- be members of the Carnivora, but seen by him as possessing a dis-
dictis (first known from the latest Tiffanian) and Dipsalodon (al- tinctive tarsal morphology. Cope's (1875) "Creodonta" included the
though the latter taxon may have been more of a bone-cracking oxyaenids Oxyaena and Ambloctonus and the miacoid Didymictis.
scavenger than a true predator). During most of the Paleocene, car- However, a close relationship between "miacids" and Carnivora was
nivorous mammals were probably represented by various archaic recognized at this time (e.g., Cope, 1884, Schlosser, 1888-1890).
ungulates, such as arctocyonid and mesonychid "condylarths" (see Taxa that were later referred to the Creodonta included a wide array
Chapters 15 and 20, although these Paleocene forms do not ap- of taxa now placed in other orders, including primitive carnivorous
pear to have been highly specialized predators, either). Although archaic ungulates such as arctocyonids and mesonychids, and even
we are used to thinking of present-day mammalian communities in- the taeniodont Ectoganus. Note, however, that Cope originally did
variably containing large specialized predators, this apparently was not consider the Hyaenodontidae to be creodonts. Cope (1875) ini-
not an aspect of North American Tertiary mammal communities tially placed the creodonts as a derived group within the Insectivora,
until the later Paleocene (see also Maas and Krause, 1994). Interest- but later (Cope, 1884) considered them a basal group of placentals
ingly enough, the early Eocene radiation of larger, more specialized from which the Carnivora and the Insectivora were both derived.
predatory mammals in North America was accompanied by the ap- Matthew (1909) recognized what would now be termed a
pearance of large, carnivorous, terrestrial diatrymiform birds (see sister-group relationship between the Creodonta (including the

73
 Christine M. Janis et al. 74

Hyaenodontidae, but excluding taxa now recognized as leptictids odontids would also accord with current biogeographical evidence
and pantolestids) and the Camivora. He proposed creodonts as a because oxyaenids are first known from the early Paleocene of North
subgroup of Camivora, and he retained the rniacoids within the America, whereas hyaenodontids are first known from the latest Pa-
creodonts because of Cope's (1875) use of Didymictis as a type leocene of North Africa (Cappetta et al., 1987).
genus, although recognizing the close relationship of miacids to In addition, no well-defined synapmorphy links Creodonta with
modem Camivora. Gregory and Hellman (1939) removed miacids Carnivora. Polly (1994) points out that current evidence may link
from the creodonts, a view supported by Simpson (1945), although Carnivora with Insectivora (in its current usage as synonymous
Simpson still retained Creodonta as a suborder of the Carnivora. with Lipotyphla) (e.g., MacPhee and Novacek, 1993; Polly, 1993;
Van Valen (1966) further removed the arctocyonids and mesony- Wyss and Flynn, 1993), leaving creodonts as indeterminate primi-
chids from the Creodonta, but united Hyaenodontidae, Oxyaenidae, tive eutherians. Both carnivores and creodonts may be derived from
and Palaeoryctidae (including the Deltatheridiidae) in a new group, the "proteutherian" palaeoryctids (Van Valen, 1966; Lillegraven,
Deltatheridia, which he considered to be the sister group to other 1969; Sloan, 1969): Polly (pers. comm.) suggests that carnivores
eutherians. Deltatheridiidae were later removed from the Euthe- and lipotyphlans may be closely related through a derivation from a
ria by Butler and Kielan-Jaworowska (1973), and the majority of Paleoryctes-like ancestry, whereas creodonts may be derived from
later workers have viewed the Creodonta as well nested within the more distantly related paleoryctids such as Cimolestes. Lilligraven
(1969) has argued that both carnivores and creodonts can be de-
Eutheria.
In recent years the Creodonta has been considered to be com- rived from different species of Cimolestes. Alternatively Fox and
prised of oxyaenids and hyaenodontids and generally accepted as Youzwyshyn (1994) concluded that the early Paleogene camivo-
the sister group to the Camivora. McKenna (1975) formalized this rans Pristinictis and Ravenictis are dentally more primitive than
arrangement by uniting the two orders in his grandorder Ferae, al- either creodonts or palaeoryctids, and propose an as yet unknown,
though only a few workers have attempted to define characters that more primitive, eutherian ancestry for the Camivora.
unite the two orders. Wyss and Flynn (1993), for example, cite
the restriction of a carnassial shear to a discrete locus within the INTERRELATIONSHIPS WITHIN
postcanine dentition (but this does not exclude palaeoryictid "pro- THE CARNIVORA
teutherians" or lipotyphlan insectivorans) and the presence of an
intracranial osseous lamina (but this does not exclude pholidotans). Defining features of the Carnivora and their major divisions
Polly (1994) has pointed out that there is no satisfactory def- Two recent papers have discussed the phylogeny of the Camivora
inition of the order Creodonta: The only possible synapomorphy (Flynn, Neff, and Tedford, 1988; Wyss and Flynn, 1993). The reader
of creodonts is a large metastylar blade on Ml, which is proba- is referred to those papers for details of the summary of camivoran
bly a primitive character for eutherians. Taxa are assigned to this phylogeny presented here. Additional details are also presented in
order by virtue of their membership in the Hyaenodontidae or the the individual chapters in this section. A current consensus hypoth-
Oxyaenidae. Systematic separation of the oxyaenids from the hyaen- esis of relationships within the order is presented in Figure 4.1.

CARNIVORA

FELIFORMIA CANIFORMIA

FELOIDEA ARCTOIDEA

"MUSTELOIDEA" URSOIDEA

Figure 4.1. Interrelationships among

carnivorous mammals.* indicates taxa
not found in the North American
Tertiary. The position of "miacoids"
is not indicated.
 Carnivorous mammals 75

6 iii iii ii:' ii:' ii:'

~ !!!. !!!. 2. ~ z 6
<('

-g.g2..,
~

~0 .....
iij
6
2.
...
c
.,
c 'iii
c
c·-
UI

.,
...c.!:
UI
c
'ia E·~
UI
.. ...,
!:. ~
UI

c... -0..
=
--
iij U UI
iii c ., G; ID 111- 111-
e _2! .s::. .s::. 0.. .s::. ;;::: ~
- UI ID 0.. 0

MA ~ e 111 c Ill ...

0
:i
Ill 0
oo t::
0
z ...
'Iii
Ill
:i 1a
0 Ill
:I .,
0 Ill
111 'tij
I!!
t::-
0 .,
z I!!
·u .s::.
., t::
0
ui
~:ll~ 111
0
E
c( Cl (j oil Cl
rn ...
Cl
rn ...
Cl
0
Cl Cl
0.. 0
z z
17.5
I / I
w
zw / I
(.) 18.8
0 /
SE /
19.2
/ Figure 4.2A. Biogeographic ranges
/ of creodonts. A "box" (for a par-
23.0 ticular time period in a particular
biogeographic region) with a cross
through it means no fossil locali-
ties are known for that time period
27.7
-E" c:
from that area; a single dashed line
w
zw il3...::: ~ !:!y through the box means only scant fos-
u.j ~ ~ sil information is available (usually
g 29.4 ' c:
only a single, small, locality). Key:
(!I ."!: <ll Oxyaenidae are in boldface. Tytthaen-
!:!y
:i
0
:s:.s::. >.
Ql
c: inae and Palaeonictinae are in reg-
31.9 ular boldface: Ab = Ambloctonus
c: Dp = Dipsalodon, Pa = Palaeonic~
.!ll !:!y !:!y
~ tis, Ty= Tytthaena. Oxyaeninae are
0 in boldface underlined: DI = Dipsa-
33.4
' c:
lidictis, Ox = Oxyaena, Pf = Pa·
Ql al .!!! !:!y !:!y triofelis, Pr = Protopsazf;. O? =
cu .!::: c:
....J (.) e oxyaenid indet. Machaeroidinae are in
34.5 boldface italics: Ap = Apataelurus,
Ql ' c:
~al.!!! Ma = Machaeroides. Hyaenodonti-
·- ~ c !:!y Hy
::;: (.) e dae are in roman type. Proviverrinae
~ 35.5 and incertae sedis are in regular ro-
man type: Ac = Acarictis, Af = Alfia,
w
(.) ~~-~
<ll .s::. c: tip, !:!y Ga = Galecyon, Is = /schnognathus,
0 w (.) e Pt = Prototomus, Pv = Proviver-
w 37.1
.n roides, Py = Pyrocyon, Si = Sinopa,
~~
g Ql
!:!y, !:!y Tr = Tritemnodon. Hyaenodontinae
Cl c: ?Lm are in roman type underlined: Hp
39.5 = Hemipsalodon, Hy = Hyaenodon.
c:
Ql .l9 H? = hyaenodontid indet. Limnocy-
-<ll ·-
....J
c:
::>
onidae are in italics: Lm = Limnocyon,
Od = Oxyaenodon, Pl = Prolimno-
41.3
cyon, Th = Thinocyon.

The order Camivora can be united by the following dental synapo- (although
. . "miaco·d
. 1 s" 1ac k an ossified . bulla) and details of the cra-
morphies (Flynn, Neff, and Tedford, 1988): P4 and ml modified as mal artenal cuculation (e.g., Matthew, 1909; Wible, . 1986). Carnivo-
the camassial teeth in adults; P4 protocone located far anterolin- rans have .been. traditi II d' .d .
. ona Y 1v1 ed mto the terrestrial Fissipeda and
gual of the paracone; elongate, well-developed P4 metastyle wing the aquatic Pmmped'rn (e.g., s·1mpson, 1945), terms based on gen-
and blade with a deep camassial notch; elongate p4 talonid with eral
. appearance
. and eco1ogy rather than designed . to reflect cladis-
at least two posterior accessory cusps on the posterior flank of the tJc. groupmgs. Whereas the F1ss1peda. . are clearly paraphyletic pin-
main cusp; ml much larger than m2-3, progressive decrease in size mpeds are . .now con s1·dered to be a monophyletlc . group originating
'
from ml-3. Wyss and Flynn (1993) proposed that the term "Car- from w1thm the arct01'd carnivores . (see Wyss and Flynn 1993·
nivora" should apply only to the crown group concept of the clade, Tedford, Barnes • and Ray, 1994). Pmmpeds . . ' in this
'
are not included
including living carnivorans and their descendants: They propose volume.. Camivorans .
. are now grouped mto two main radiations:
the term "Camivoramorpha" for the broader group including the the. Camformia (incl ud'mg the extant fam11Ies .. Canidae, Mustelidae
.
fossil "miacoids." A1lundae, Procyon·d . . .
. . 1 ae, and Urs1dae, the pmmpeds, and the extinct'
Interrelationships within the order have relied heavily on Amph1cyomdae) and th F .,, . . .
. . e e111ormia (mcludmg the extant families
characters of the auditory bulla (e.g., Flower, 1869; Hunt, 1974) V1vemdae, Herpe st'd 1 ae, Fehdae,
. and Hyaenidae, and the extinct
 Christine M. Janis et al. 76

w / I
zw eJ. I
0
0 I
w /
/

/ I
/
I

w
zw
0
0
w
..J
<C
Cl.

Figure 4.2B. (Cont.)

Nimravidae) (Tedford, 1976; Flynn, Neff, and Tedford, 1988). The places miacids as the primitive sister taxon to all other carnivorans,
early Tertiary "miacoids" may (Flynn, Neff, and Tedford, 1988) or and viverravids as the sister taxon to the Feliformia plus Caniformia,
may not (Wyss and Flynn, 1993) be included within these two clades although they note that dental evidence might still support their al-
(see later). location to Caniformia and Feliformia, respectively.

Caniformia. The Caniformia can be subdivided into the Cynoidea

Carnivore phylogeny
"Miacoids". Early Tertiary camivorans were originally consid- (canids) and the Arctoidea. There is little debate that canids are
ered a distinct family, the Miacidae, ancestral to later forms (Cope, distinct from other caniforms based on auditory bulla characters;
1880a). Flynn and Galiano (1982) demonstrated the paraphyletic canids have a low septum in the anterior part of the tympanic
nature of this group, recognizing the miacid "miacoids" as stem cavity, whereas arctoids retain the primitive condition of the lack
caniforms and the viverravid "miacoids" as stem feliforms, thus tak- of a longitudinal septum within the bulla (Flower, 1869). Relation-
ing the split between major living clades of carnivorans back into ships within the arctoids, however, have been the subject of consid-
the early Paleocene. Cladistic analysis by Wyss and Flynn (1993) erable debate. Mustelids and procyonids have often been regarded
 Carnivorous mammals 77

Figure 4.2C. (Cont.)

as sister taxa, forming the superfamily Musteloidea (as in Flynn, (1993) also suggest that the ailurine procyonids should be consid-
Neff, and Tedford, 1988). More recent analysis of morphological ered a separate family, Ailuridae, closer to ursoids than to other
characters places mustelids as the sister group to other arctoids procyonids, a view also supported by molecular evidence (Vrana
(Wyss and Flynn, 1993), a view also supported by molecular evi- et al., 1994; see also discussion in Chapter 8, where Baskin tenta-
dence (Wayne et al., 1989; Vrana et al., 1994). Baskin (this volume, tively places the extinct simocyonines with the ailurines in a family
Chapter 8) defends the notion of the Musteloidea from synapo- Ailuridae).
morphies of the basicranium and dentition described by Tedford Members of the extinct family Amphicyonidae were originally
(1976, = Mustelida in Tedford, Barnes, and Ray, 1994), but ac- included within the Canidae, but they are now usually considered to
knowledges that the group may not be monophyletic. Note that be arctoids. Hunt (1977) determined that amphicyonids possessed
suggestions from molecular data indicate the Mustelidae may be the "ursid loop" feature of the basioccipital, which accommodates
paraphyletic, with skunks more primitive than the other members the inferior petrosal venous sinus and a posterior loop of the carotid
of the family: Wayne et al. (1989) and Vrana et al. (1994) place artery. This character, plus features of the scapula (Davis, 1949),
skunks as the sister taxon of all other arctoids. Wyss and Flynn unite amphicyonids with ursids. However, Tedford, Barnes, and Ray
 Christine M. Janis et al. 78

iii' iii' CL CL CL
0
!:!.. 0 ~ !!?.. !!?.. 2. ~ z 0
~

ii
~

-!·~.
2. .
'Iii
0
.. -1n .....
_o
J=
cQ) ..
c
c"iii
_!m
c
c·-
; ~
.s::. m
E·i"'
Q)-
.s::. 11.
UI
c
'!!.. :!
11.
E·~"'
Gl-
.s::. 11. ~i
e:.
..
'Iii
0
0 cQ) .. t:: 1ii
0
E 0 t:: .. 'S-;
1! e ... c111 ECll = Oo '50 1U "Olt::
.s::.
..
MA :::J (j oll ~ Cle! Q)
Cl) ...
0 I!! 0 I!! ~! 11. 0
ui
z
!!~~ 0 <C Cl Cl Cl) Cl Cl Cl z
35.5
~~.~
m.c: c:
WO e
37.1 c
.c:'
"c
"'
Q)
Mc
Di,
::J"'
Cl Q)
39.4
Vi,
~~ Pd, Ta,
j~ Mc
41.3
c
/
~~
w::>"'·- /
45.9 ~
Q) /
Q) .g / I
.§ &j .ffi I
46.7
.!2 ~ / I
~ :g c
::;;m.!!! I
49.0 ~ / Figure 4.3A. Biogeographic ranges
Q)
>-"' of "miacoid" carnivorans. A "box"
w -c: "C
co ·c: c
zw wm.!!! (for a particular time period in a par-
0 50.4 .c:
ticular biogeographic region) with a
0
w "
1;j cross through it means no fossil lo-
Q) "' calities are known for that time pe-
§~.~ riod from that area; a single dashed
53.5 .c:

-°' B"'
line through the box means only scant
"C "' fossil information is available (usu-
54.0
~~.ffi ally only a single, small, locality).
.s::.
B Key: Viverravinae are in roman type:
>-"'
~~c Br = Bryanictis, Di = Didymictis,
55.0
w s: .!!! In = Intyrictis, Pi = Protictis (Pro-
>-~ tictis), Pt = Protictis (Protictoides),
-c:B
"'"'s:
w"' Ra = Raphictis, Si = Simpsonictis,
"'c
ui .!!! Vi = Viverravus. Miacinae are in bold-
55.5
c face: Mi = Miacis, Oo = Oodectes,
_,:,,"' Pa = Palaearctonyx, Pc = Procyn-
""32"'
c3 .E odictis, Pd = Prodaphaenus, Ta =
55.6 Tapocyon, Ui = Uintacyon, Va= Vas-
w c
z ~.!!!"' sacyon, Vu= Vulpavus. Taxa incertae
C3"'~
w .E sedis and indeterminate are in italics:
0
055.8 Jc= /ctidopappus, Mc= Miocyon (M.
w
-' ~-~~ magnus only), Rv = Ravenictis, Pr=
<C
a. 56.0 C3"'~
Q Pristinictis, M? = miacoid indet.

(1994) consider amphicyonids to be the sister taxon to other arctoids, to understand why they were considered closely related (Cope,
based on the lack of a suprameatal fossa and the retention of M3. 1880b), or even contained within the family Felidae and ancestral
Note that there is some controversy as to the systematic placement to both conical-toothed and saber-toothed felids (Matthew, 1910).
of taxa in the basal ursid subfamily Amphicynodontinae: Tedford, However, their very primitive and uniquely specialized basicranial
Barnes, and Ray (1994) consider them stem pinnipeds, but Baskin anatomy clearly removes them from the Feloidea (Baskin, 1981;
and Tedford (1996) include them within the Ursidae. Hunt, 1987). In recent years they have been placed as the sis-
ter taxon to the living Caniformia (Flynn and Galiano, 1982) or
Feliformia. Within the Feliformia, the extant families are united in as the sister taxon to all living carnivores (Neff, 1983). Flynn,
the Feloidea (= Aeluroidea of Flower, 1869) by a double-chambered Neff, and Tedford (1988) suggested that nimravids should be placed
auditory bulla with a bilaminar septum. The systematic position as incertae sedis within the Carnivora. Baskin (1981) and Bryant
of the extinct Nimravidae has long been problematical. These (1991) tentatively support a sister-group relationship between
hypercarnivous camivorans look so similar to felids that it is easy nimravids and feloids, a view upheld (but with equal caution) by
 Carnivorous mammals 79

0 iii' iii' CL CL CL
~ !:!.. ~ ~ !!?.. !!?.. 2. ~ z
e:.
0
~
i1i -B
2. ..
Ui
0
'iii
b"(;;
c ..
2. ...c
_!m
.c
'(ij
c
c·-
ml
.cm
c.5
...
111-
"'
ca '! s
"'c E·~
Cll-
UI

" 'Iii
lE ~
"!ii
ca
"tJ E ! ·;: 0 Cll 0
m
.s::,11.
:; m
.. 11.
c .. €~ 8
MA c: e lb c:~
!!~~ ~ <C
=
:::J
Cl
Oo
(j oll
t::-
~ Cle!
'5
0
Cl)
I!!
Cl
0
Cl)
I!!
Cl
~ lll
Ci
o ca
z I!!
Cl
" .s::.
.. t::
11. 0
z
56.0

56.2
~l!)
:i:: c
i= .!!!
57.0
~ ...i-
I§ .ffi Pi
57.3
~C")
Di, Pi Ra, Pi
1§ .ffi
58.5
~C\I
:i:: c /
i= .!!! /
59.2
C:
.,~

Br, ?Si Pi, Si,

~.ffi Pr
60.9
C:
w 0
z -~(") Br, Pi, Si,
w
0 61.3
~.~
0 C:
w 0
-' ·~C\I
<C 0 c
Q.
62.5 I-.!!!
C:
0
·~~

63.8 .£ .~
c
~
tE ,Q) " '

64.0 c
"'
~
Q) " '
Rv
:::J
a.
64.5
c
"'
~
Q) ~ /
:::J
a.
65.0
<ll
~
Q) 0
::J
65.5 a.
Figure 4.3B. (Cont.)

Wyss and Flynn (1993). There is still debate over the inclusion of sister-group relationship between felids and hyaenids, with her-
the Neogene barbourofelines with the Paleogene nimravines (see pestids as the most primitive feloids. Hunt (1987; see also Hunt
Neff, 1983). and Tedford, 1993) supports sister-group relationships between fe-
Relationships within the Feloidea have also been of considerable lids and viverrids on the one hand and hyaenids and herpestids on
debate. Although viverrids are strictly an Old World group, and the other. Wayne et al. (1989) support a sister-group relationship
hence not especially relevant for discussion in this chapter, their between viverrids and herpestids, with a trichotomy between this
interrelationships are detailed because their systematic position af- group and other feloids; Wyss and Flynn (1993) support a sister-
fects the relationship of the felids to the hyaenids. Herpestids, orig- group relationship between hyaenids and felids, and a trichotomy
inally perceived as a subfamily within the Viverridae (e.g. Simpson, between this group and viverrids and herpestids. As felids are the
1945), more recently often have been considered as a separate family only feloids that play an important role in the fossil record of the
Herpestidae (see discussion in Wozencraft, 1989), resulting in four North American Tertiary, we consider this debate beyond the scope
feloid families and a great variety of possible interrelationships of this chapter and present an unresolved tetrachotomy among fe-
(Flynn, Neff, and Tedford, 1988). Wozencraft (1989) supports a lids, hyaenids, herpestids, and viverrids.
 Christine M. Janis et al. 80

CL CL CL
0 iii'
~
iii'
ec e ~ ~ z~ 0
~
<(' e. iii~
0 c c·- c .5
UI
c
UI
c.5
UI

'lii 'lii
~ 'lii c'iii .. Ill ~:!
.. Ill Ill
iii Ill .::: 'lii .. Ill
;=
:; ...
GI - GI-
.s::. 0..
()
!E 3:
GI 0

~~
.s::. ID .s::.
e -I!!·;:ti 0
0 c Ill _!ID 0..
t:: 1U () .s::. 0
~i..
GI 0
MA ~ e cu CGI EGI :sCl Oo t:: 1U
z0 ..GI om..
rn rn o~ z0 ..GI Ill
0.. 0
t:: u.i
z
.5~ ~ 0 c( ~oll Cl Cl Cl Cl Cl z
1.8

IX
I
c: Qrr, !Jr, Qn, Cn, Cn, I
Cn, Ce,/
Q) "'c:
0
I
Bo /Bo
Qrr, !Jr
Bo Bo Bo Bo I

x xx
"'.!ll

x rx
w
_J [IJ I /
2.4
zw c: I Cn, Qy, Qrr, !Jr, Qrr, Ur, Cn,'\l!!,
0 >.o
~~
"' I
Cn,
Bo
Bo
Bo '&, Bo Bo Bo Figure 4.4A. Biographic ranges of

x
0 WCD 11
::i 4.5 canids and problematical arctoids. A

IX
Cl.
- 0. c:
I
Cn, Vu, Cn, Cn, Vu, / "box" (for a particular time period in a
_2l E .!!! Bo, Cc
I ?Os Os Os ,, / particular biogeographic region) with
::!~:2
5.2 I a cross through it means no fossil lo-
/

~
/

~t~
:c .c:
I
a. c: I en, Vu, Bo,/
Os /Os
Cn, Vu,
Os
?Ce,
Cn, Vu,
Cn, Vu,
Os /
/
Cn,
Vu,
Os x calities are known for that time pe-
riod from that area; a single dashed
line through the box means only scant

x
_J
I/ Os
6.0 Cn, Le, Cn, fossil information is available (usu-
.<?:-' 'l\!,
.. 0. c:
~ E.!!!
.J~'.2 D<
?YJJ.,
Os
Vu, Cn, Vu,
Cn,
Ae,Os Os, To Ep,Os
Ep,Os Vu, Ep,
Os, To
'\l!J,
Os ,X ally only a single, small, locality).
Key: Hesperocyoninae are in roman
type: Cd = Cynodesmus, Co = Cae-

x
7.0 I I / I
Le, '{y,
-.>.: : a. c: / Le,/ Ae,
Ep, / / Ae, docyon, Ee= Ectopocynus, En= En-
~ E .!!! Os Ep,Os ?yti. Ep,Os Ep,Os, / Os hydrocyon, He= Hesperocyon, Me=
Q) = / /os
ui :c :E I/ I/ Ep I/ To / Mesocyon, Ob = Osbornodon, Pa =
8.8 !,&, Ae, /
Paraenhydrocyon, Px = Philotrox,
~a lg, '{y, / 'iY.1
I ?Le, lg, Vu, Le Cn, Vu, /

.IX
2 (ij'C Ae,Ep, / Ca, Cc, I
Su = Sunkahetanka. Caninae are in
Ol- 0 I Ep Ep Ae /
w _J ( ) -0 , To I/ Ep, To I/ I/ Ep roman type underlined: Ce = Cer-
iii 9.5
lg, lg, Vu, Ae,Ca, !,&, Ae, / Ep/ I
docyon, Cn = Canis, Cy= Chrysa-

D<
0 C: ffi Le, '&, / cyon, Le= Leptocyon, Ur= Urocyon,
0 ~~-c Ae,Cc, Cc, Ep, Ca, Cc,
I I
:e 11.0 Ol- 0 Ae Ep Ae,Ep To
/ Vu= Vulpes. Borophaginae proper are

x
w () -0 Ep, To Ep,To v I/ II
/ / in boldface: Ae = Aelurodon, Bo =

~
I I
Ae, lg, Ae,
,S? .9 / Ep / Cc/ /
C?1 ' Borophagus, Ca = Cynarctus, Cc =
j~@ Ae, To
/ / /Ep
Ca, Cc, ?Eu, / I I Carpocyon, Cm= Cormocyon, Ep =
_j [IJ ·:;;
I/
Ep, St, To I/ 11 II
I/ I/ Epicyon, Eu= Euoplocyon, Os= Os-
12.5 /

~
!,&, Ae Ae Le, I teoborus, Pl = Phlaocyon, St = Stro-
as .s!!! /
Q) '
Ae,Ca, Eu,
To / I bodon, To= Tomarctus (senso latu),
_J c:
Ep,To Cc, Ep : I x Ep, To To
ui ~ -~ To, St /
II B? = borophagine indet. Borophagi-
14.0 To; nae (?) of the Cynarctoides group are
£ I Ob,Ae, Ob, Ob,
Ae,
~~ffi I
~&l-:;; ,
To To Ca, Eu, Ae,Ca
To, Ct Cc, To ·X Ae,Ca,
Eu, To
To
To
I
in italics: Al= Aletocyon, Ba = Bas-
sariscops, Ct = Cynarctoides, Na =

x
15.8 /
"Nothocyon" annectans, Ox= Oxeto-
bi c:

x
I
/ Ob, I&, C?, I cyon. C? = canid indet. Problematical
°' .E 'Ec: "'
/
To To To
To, Ct I I arctoids are in boldface italics: No =
j~.Q 11 I/ I
17.5 Nothocyon.

EVOLUTIONARY AND the Torrejonian and continued into the Tiffanian (late Paleocene),
BIOGEOGRAPHICA L PATTERNS represented by small, generalized carnivores similar to present-
day civets rather than specialized predators. There is little evi-
dence of biogeographical variation: lntyrictis appears to be a pri-
PALEOCENE FAUNA marily New Mexican taxon; Bryanictis and Raphictis are known
As previously noted, the early Paleocene had only a limited radi- only from the northern Great Plains in the later Tiffanian (see
ation of carnivorous mammals (or, at least, of creodonts and car- Figure 4.3).
nivorans: See further discussion in Chapter 15 about the possible There is only a single questionable record of a creodont from
biology of certain archaic ungulates). The only records from the the early Paleocene, and while the earliest oxyaenid, Tytthaena, is
Puercan are of carnivorans of questionable taxonomic position: a known from Tiffanian 3, oxyaenids did not diversify until the latest
single occurrence of Ravenictis and a questionable occurrence of Paleocene (Clarkforkian), when larger, more predatory forms such
Jctidopappus. A modest radiation of viverravines occumed during as Dipsalidictis and Palaeonictis appeared (see Figure 4.2).
 Carnivorous mammals 81

0 iii' iii CL CL CL
<(' e. 0 ~ ec e ~ ~ z~ 0
~
~ iii~ c c·-
UI UI
c
UI
...
~
1:1 e cu
-f! ·cti
'lii
0
Ill
0
....
.::: 'lii
c Ill
GI 0
s·i
_!ID
t:: 1U
._ UI
GI Ill
.s::. ID
:; 1U
c;5
.. Ill
GI-
.s::.
'S 1iS
0.. .
c -0..
iii 'iii
c .5
.. Ill
GI -
.s::.
t:: 1U
0..
() xi
=3:
() .s::.
"iii
Ill
0
0
C GI oo GI °lU Ill t:: u.i
MA
la~ GI E :i
~oll z0 ..GI 0 GI
rn .. 0 GI
rn .. o~ z0 ..GI 0.. 0

17.5
-.I "I:
Cl
0 c( Cl Cl Cl Cl Cl Cl z z
/
g> c: I Eu, To, B? / To, C?
/ I
->.'- "'
w
z ca E·-
Q) "E / / Ct I
w :c .Q
~ 18.8
0
:e -Q)

j""
c:
"' / / C?
*
Q)

~ c:
/ / /
_j
19.2 Q)

cu
...J ~
I m
Q)
/

23.0
w< cu /

.<?:- c: Cd,
~~m ?En,
_j ~ ~
Cm Cm
27.7
~ ffi
w
z ~~~
gw 29.4 < uj

c:
(ij

.,!. Cd, Ee, ?En,

CJ
::i ·-"'
.c:
~~
>. He, Me, Ob, He,
0 Pa, Ox Me
31.9
c:
.!ll He,
He
~ Ob
0
33.4
' c:
ca ..c:~ .!!!ec:
Q)

_Jo
He He He
34.5
Q) ' c:
~~.!!!

~35.5
·- ()
::;: ..c: ..
5 He He
I
I

w ' c:

0
0 "'·-
>.-0 "'
-;::::
w e
"' ..c c:
o
He He
w37.1
.n
Q) c:
-u
6
m
c:
He

39.5
c:
Q) 19
-
_J "'·- c:
::>
41.3 Figure 4.4B. (Cont.)

EARLY TO MIDDLE EOCENE FAUNAS the earlier Eocene, peaking around the start of the middle Eocene
(early Bridgerian) (see Figures 4.2, 4.3). This marks the time of
The start of the early Eocene is marked by dramatic change in the the maximum thermal warming in the Northern Hemisphere, with
carnivore fauna, with the immigration of hyaenodontid creodonts temperatures cooling throughout the later Eocene (see Wing, this
from the Old World and the radiation of miacines (which first ap- volume, Chapter 2, and Janis, 1993, and references therein for re-
peared in the latest Paleocene). Viverravines were not a significant view). The predaceous diatrymiform birds that first appeared in
component of the Eocene faunas, consisting only of the long-lived the early Eocene did not survive into the middle Eocene (Anders,
genus Didymictis, a single occurrence of the poorly known Pro- 1992).
tictis (Protictoides), and the more specialized (more carnivorous) Both oxyaenids and hyaenodontids diversified in the early Eocene
genus Viverravus. On the other hand there _was a significant radi- and were individually more abundant than the "miacoid" carnivorans
ation of miacines, now including more specialized, more curso- (Savage, 1977). Oxyaenids included larger, more specialized forms,
rial taxa (although still of small body size) such as Miacis, Vul- such as the predatory Oxyaena and the possibly bone-crackingAm-
pavus, and Vassacyon. The greatest diversity of "miacoids" was in bloctonus. The predominant hyaenodontids were the rather foxlike
 Christine M. Janis et al. 82

iii' iii' CL CL CL 0
0
0 ~ !!?.. !!?.. 2. ~ ze;. ~
<
2.
~
... iii 2.
c
c ·-
c
c.:
UI UI
c c.5
UI
u;
~¥
c "iii
'iii Ill
UI
Ill
0
~ u;
c Ill
... Ill
_2! ;=
.s::. ID
... Ill
111-
.s::.
!:!
... 0..
... Ill
111-
.s::.
0..
Ill
0
e ID 0.. 0 Figure 4.SA. Biogeographic ranges
"!·~ 0 Ill 0
~'la '5 'la '5 10 c ... ~'la u .s::.
~ e 111 ... Ill
= Oo 0 Ill 0 Ill 0 Ill
Ill Ill
0 !! ~ !! Ill~ u.i of mustelids, procyonids, and Car-
MA ~ E z ... rn ... rn ... 0.. 0
z
~~~ ~oil
:I
Cl Cl Cl Cl Cl z nivora incertae sedis. A "box" (for

x
0 c( Cl ~

1.8 a particular time period in a particu-

c: /
Bu, Ca,
Bu, Mu, Sa, Tr Tr/
Sa, Tr, ?Tx, lar biogeographic region) with a cross
Q) "'c:
1ii ..5!!
0
I
Pc
/
Ba, Pc
Mu, Sp,
Tx
Tr, Tx,
Sp,
Sp, Tr, '----
Bu, Fe, I through it means no fossil localities

x iX
I/ Ba
w 2.4 ....J [IJ I Pc Mu, Sa,
'--- are known for that time period from
Sp
z I Sm,Sp,

LX
c: I Bu, Bg, Bg, Bu,
w Mu/ that area; a single dashed line through
0 "C
"'ffi
>."
I En
/
Tr,
Ba
Mu, Tr,
Sp,
Me, Mu, Sa,
Sp, Tr, Tx,
Tr, Tx,
Ba, Pu, the box means only scant fossil in-
"'-
IXx
0 ,

x x
::i 4.5 UJ [IJ ~ Ba, Na Ba, Pc ?Pc formation is available (usually only a
Cl.
- a. c: I En, Po, ?Bg, Bu,?Ma,
/ single, small, locality). Key: Mustel-
Bu, Sa,
2~~ I Ar, ?Na,
Tx
Tx
Mg, ?Mn, Mu,
Po, Px, Sm,Tr / idae are in roman type. Paleomustelids
j :r: '.E Pc , are in italics: Al = Aelurocyon, Bp
5.2 I

x = Brachypsalis, Mv = Mustelavus,
/ /

[X
I ?Ce, Ma, Me?, Mg, ?Po/
6. c: Ce, En,
/ Po, Px,
Ma, Po,
Ob = Oligobunis, Pb = Paroligobu-
Pg, Po, Po, Px,
2~~ I Po, Px, Px,
j:r:;E ,, ?Sm, Pc , / Px, Ba Ar Ar, Ba /
Ba nis, Pm = Promartes, Ps = Plesictis,
1 (/
Pt = Potamotherium, Zo = Zodi-

[X x
6.0 I
-2' ' Po, Lu, olestes. Neomustelids are in roman
~ a. c: I En, Eo,
Eo, Lu, Px, / Ma, Px,
~ ~:lll I
En ?Mn,
Pg,Px /Ba
Eo, ?Le
?St
?Ba,
Si
type: Bg = Brachyopsigale, Bu =
.J:r::cl1 ?Pr (/ Buisnictis, Ca = Canimartes, Ce =

x
7.0 -; I
~.
/ I
Cernictis, Cr = Craterogale, Di =
"' a. c: / / / Le, ?Mu, /
w E .!.1! ?Ar Px Dinogale, En = Enhydritherium, Eo
ui~:E I/
/
,,. / ,, / Ba, ?Pa
I/
/
= Eomellivora, Fe = Ferinestrix, Ho
8.8 Le, Ho, / / = Hoplicitis, Le = Leptarctus, Li
C: c: Le, Ho, Le, Ho, Le, I

iX
I ?Li, Me? ?Bu,/
Q) ~ .!Q St, Ma, St, Me?, Mn, Pg,
/ Ho, = Limnonyx, Lu = Lutravus, Ma =
-Ol-"' c: I Ac, Ba, /

x
w 0
....J ()"CJ , Ar, Pa Ba /Ma Ba Pp /St Martes, Mg = Martinogale, Me =
? Pr 1,. I/
ffi 9.5
C: c: Bp,
/

I
Mephitis, Mm = Miomustela, Mn =
0 Ma, St, Mg, Bp,
Le, Ma,Mn, I / Mionictis, Mu = Mustela, Po = Ple-
0 ~ ~ .!Q M?
?Pn Le, Mn, I siogulo, Pg= Pliogale, Pn = Plionic-
:e 11.0 ca .m 5 Pn, St, /
,,. /

x
UJ ()"CJ Ba ?St Ar II It tis, Px = Pliotaxidea, Sa= Satherium,
I I I Bp, Pt, /
Sm = Sminthosinis, Sp = Spilogale,
,x I
Q) 6 Bp,
/ /
ca~ c: Le / / Le, Mm, Mn, ?Mu, St = Sthenictis, Tx = Taxidea, Tr =
/ /
,,. / Pn, Pg, St, / ,, I

x
....J
_j [IJ ·::;'
"'"
Ba Trigonictis. M? = mustelid indet.,
(/ 1,. Ar, Ba, Pr (/
12.5 / Me? = mephitine indet. Procyonidae
6 Le, Ma, I

xi/
Q) / Bp,
1ii- Bp,
Mm, Pn, are in boldface: Ac = Actiocyon, Am
!!! c: Mg / I
....J
UJ. "'"'
[IJ ·::; Ba / St, Ba
?Pn
,, = Amphictis, Ar= Arctonasua, Ba=
14.0 Bassiriscus, Ed = Edaphocyon, Na
Bp,
6 I
x
I M?, Le, = Nasua, Pa = Paranasua, Pc =
>.- Bp, Le, Ma, Mn, ?Le
-c: !!! c: Pr
I Procyon, Pp = Protoprocyon, Pr =
"'"'"'
UJ [IJ ·::;
I
I Ar, Ed
Ma Pn, St,
Ar, Pr
?Ba
,, Probassiriscus, Pu = Parailurus, Si
, ,
x
15.8

~
Cl c: Bp, / = Simocyon. Carnivora incertae sedis
.S.!.1! / Le M? Di, Le, Mm, M? I
°'
1ii ~ .£2
E 'E
,
/ Mn, Pn, St ,
/
11
are in roman type underlined: fl. =
17.5 Palaeogale, Sg =cf. Stenogale.

proviverrines. The diversity of the creodonts paralleled that of the more tropical form to a more temperate form, and the diversity of
carnivorans, peaking in the early middle Eocene. The middle Eocene reptiles and amphibians also declined. There were also changes in
creodont fauna is rather distinct from that of the early Eocene. Pa- the mammalian fauna, including the virtual extinction of the carniv-
triofelis replaced Oxyaena as the large, catlike predator and was the orous mammals of the earlier Eocene. The picture is complicated by
only surviving oxyaenid, with the exception of the strange, saber- an influx of immigrants from Asia at around this time, which may
toothed machaerodines of the middle Eocene that may not be true have resulted in competition with the endemic taxa. More derived
oxyaenids. Diversity decreased among the proviviverrine hyaen- carnivorans first appeared in the late middle Eocene fauna! turnovers
odontids and increased among the limnocyonids (see Figure 4.2). and include the amphicyonid Daphoenus at the end of the Uintan
There are no obvious biogeographical patterns among carnivorous and the canid Hesperocyon in the Duchesnean (see Figures 4.4, 4.6).
mammals in the Eocene: Although some taxa appear to be limited to However, these taxa represent small, generalist carnivores: It is hard
the central Great Plains, this probably represents a sampling artifact. to imagine that they competed directly with the diverse "miacoids"
By the late middle Eocene the cooling changes had become more and creodonts. Instead, it seems more probable that the reduction in
profound: There was a general change in the vegetation from a diversity of the early Paleogene carnivores represented the demise
 Carnivorous mammals 83

0 iii' iii' CL CL CL
~ !!?.. !!?.. 2. e ze;. 0
~ 0
< ... Ill 0
- c c UI Ill Ill
... ~

'iii
2. UI
Ill ......
._. c "iii
... Ill
c·-
lii lG
c.5
... Ill
e:!!
c cc
t:i .sdl
u;
Ill

,, ee 111 iii.~
111-
~ lG
- c
0 _2! .s::. ID 0
...... 0 ID .s::.
0..
0.. .s::.
0..
~'la
!t: 3: 0
~m...
... Ill 08 ~~ '5 'laIll 'la
Ill u .s::.
Ill t: u.i
MA la~ '<:(0) 0~ c(E :i
~oil ~ !! 0
rn ... rn 0 !! 0 Ill
z ... 0.. 0

17.5
.... Cl Cl Cl Cl Cl Cl z z
I ?Ob,
/
I M?/
Le, Mm,
/ I I

/ Pm/
/
19.2r::----c:-f---'ll----
111 -¥-----,1L--.....,.i.---3-~1------1L----J
j~m
uj~ ~

~ 35.5 1---f---*-----,-lf--~~-~~-~:..._+-----l-----l
w >.-6 ~
0 ~~·c:
0 woe
w 37.1 ~--.¥.----'!

Figure 4.58. (Cont.)

of an earlier, more tropically adapted fauna that could not survive this volume). The limnocyonid creodont Oxyaenodon was also a
the changing climatic conditions of the later Eocene. larger, more specialized carnivorous. Uintan taxon that paralleled
Although isolated records of "miacoids" occur into the late the "experiment" of the miacines.
Eocene (early Chadronian) (see Figure 4.3), these archaic carnivo-
rans became an insignificant component of the fauna after the end
WHITE RIVER CHRONOFAUNA
of the Uintan. However, there are some interesting developments
among Uintan miacines, with the appearance of several larger, more The carnivore fauna that arose in the Duchesnean persisted into
specialized carnivorous taxa, such as Tapocyon, Procynodictis, and the late Oligocene. This carnivore assemblage is also recognized
Prodaphaenus (the latter obviously so named because it was ini- as a component of the "White River chronofauna" of Emry (1981)
tially mistaken for an amphicyonid). The evolution of these rather and as the "hyaenodontid-nimravid carnivoran association" of Hunt
doglike forms could be interpreted as an "experiment" in adapting to and Tedford (1993). Although the earlier types of creodonts were
a changing climate, perhaps reflecting changes in the ungulate fau- extinct by the end of the Uintan, hyaenodontine hyaenodontids
nas that accompanied changing vegetation (see Chapters 22 and 35, first appear in the Duchesnean (latest middle Eocene, with possible
 Christine M. Janis et al. 84

iii iii ii" ii" ii"

0
!:!. 0 ~ !:!?. !:!?. ~ ~ z
!:..
0
~
~ c c UI UI UI

"!~
c
~ 0ca c'iii
... ca
c·- c.5
,_ca
!1
c.5
,_ca 0Ill i
l ca t = 111- 111- u 0

MA

1.8
~ e - Ill
.5~ ~ 05i c(E
111
"!·~
-
0
0
:;
Cl
-c UI
ca
Ill 0
oo
5o11
_2!
t:: 1ii
ID

z0 ...Ill
Cl
.s::. ID
'5 1a
0 Ill
rn ...
Cl
.s::.O..
'5 1a
0 Ill
rn ...
Cl
-o..
i 'S
0 I!!
Cl
.s::.
~"lii
z I!!
0..

Cl
!E :t
u .s::. 0
cat::
0.. 0
z
ui
z

~
I
c: ' I
Q) "'
iii.!!!
0
c: I /
/
Ir !!r Ir I
w 2.4 -1 CD I [/
zw
0
Q
>. 0
"C
"'-
c:
"'ffi I
I

~ Pn Pn

~ Ur
x
D<rx
.... W CD 11

x
4.5 Figure 4.6A. Biogeographic ranges

IX
Cl. j
- 6. c: 89. / of ursids and amphicyonids. A "box"
2~~ 891 Pn
89 89 89 / (for a particular time period in a par-
j :c '.§
5.2 I '· ticular biogeographic region) with a

6.0
~~§ I
_J
6.
:c ..c:
c: I
[X 89,
Pn
89 /

11
/
/
ilil ilil
Ag,
Pn
/
/
/
/

x cross through it means no fossil lo-

calities are known for that time pe-
riod from that area; a single dashed

~
line through the box means only scant

~
~.
... a. c:
i.\'l ~~
.J :c :c
ID. ID.

-,
lrl ID. ID. ID.,
Pn
.x fossil information is available (usu-
ally only a single, small, locality) .

x
7.0 / / Key: Ursidae are in roman type. Am-
~.
(ij c.. c:
' phicynodontinae are in italics: Al =
/ / / /
w E.!!! / / Allocyon, Ko = Kolponomos, Pa =
/ /
ui~:E

x
I/ 11 l.1 ,/ Parictis. Hemicyoninae are in roman
8.8 / / '
C: c:
I Hn,
I
type: Ce = Cephalogale, Hn = new
Q) ~-~ / Is Is,
/
hemicyonine genus, Ph = Phobero-
w -Ol-
"' c:
0 I / / I
_J () "C
I/
Ps
I/ 11 cyan, Pl = Plithocyon. Ursinae are
ffi 9.5
I

/ underlined: Ursavinae are in roman

~
0 C: c: Hn, / I
0 ~~-~ Is Is Is,
/ type: Ag = Agriotherium, I!! = Jn-
I
:e 11.0 ...
Ol-
WO"C
"' 0c:
Ps
I/
/
11
I
11
darctoS:Uv = Ursavus; Tremarctini
are in italics: Pn = Plionarctos, Ii. =
7 /

~
I

x I
~6 !Jy / Pl / Pl.UY,
cu 1i) c: / / / I Tremarctos; Ursini are double under-
...J (ij cu / / / Is, Ps / I I lined: Ur = Ursus. Amphicyonidae
.J CD·::; 11
l.1 11 11 I/ 11 are in boldface. Daphoeninae are in
12.5 /

"'"' ~ x I
Q) boldface italics: Br = Brachyrhyn-
ca .9!!!'

c:
Pl, UY.
Ps ls,P~ / !Jy !Jy
chocyon, Dd = Daphoenodon, Di =
_J
Is
I
W. CD·::; ,
/
11 Daphoenictis, Dn = Daphoenus,
14.0 Pd = Paradaphoenus. Temnocyoni-

15.8
~~c
6
ca CD·::;
W "' "' 11
I
I
?Pc
Am,
Cy,
Pl,
Am,
Pc
Pl,
Am,
Cy x Am, Cy,
Pc,Ps
Am
I
I
nae are in boldface underlined: Ma=
Mammacyon, Te = Temnocyon. Am-
phicyoninae are in roman boldface:

~ x
I
6i c: I
/ Uy
/ Am = Amphicyon, Cy = Cynelos,
.5 ·"'
°' EQ) °E0
iii /
Am, Cy, Pc
/ I Is = /schyrocyon, Pc = Pliocyon,
-l:C- I/ 11 11
17.5 Ps = Pseudocyon, Ys = Ysengrinia.

earlier records in the Uintan). Nimravids first appeared in the early Wilhelm, 1993). In contrast, nimravids appeared to be specialized
Chadronian. ambush predators, taking the ecomorphological role of present-day
Hyaenodontids were the larger, more cursorial predators. They felids.
probably played a role in the White River chronofauna analogous Other carnivorans at this time were smaller generalists and in-
to predaceous canids in present-day faunas, although they were cluded the daphoenine amphicyonids, the hesperocyonine canids,
more carnivorous than the majority of extant canids. Hyaenodontids the amphicynodontine ursid Parictis, the paleomustelid Mustelavus,
ranged in size from about that of a fox to that of a lion, and seem and the enigmatic, possibly viverravid, taxon Palaeogale. Hespe-
to represent a modernized radiation of more cursorial creodonts rocyonines were especially diverse at the generic level in the mid
that survived until the end of the Oligocene in North America and Oligocene (see Figure 4.4 ). The late Eocene diversity and abundance
well into the Neogene in the Old World. However, they were not as of daphoenine amphicyonids was reduced in the Oligocene, and
cursorial as present-day pursuit predators, and it is unlikely that they the more derived and more cursorial temnocyonine amphicyonids
pursued their prey over long distances or hunted in packs (Janis and first appeared by the early early Arikareean (see Figure 4.6).
 Carnivorous mammals 85

0 iii iii ii" ii" ii"

< !:!. 0 ~ !:!?. !:!?. ~ ~ z 0
~
iii
~ 0
ca
_o
,__
ca -
c
c'iii
... ca
c
c·-
... lG
c.5
...
111-
ca
UI

e:i!!
UI
c
E·i
UI

U -
!:..
UI
0ca

-
111-
e -f! ·~ca 0
0
c UI
- ca
Ill 0
_2! ID _2! ID .s::. 0.. .... 0..
c_ .s::. 0.. !E ; 0
0
~ e 111 -111 Oo t:: 1ii '5 1a '5 10 t:: 1ii u .s::.
MA
.5 ~ ~ 0:ii c(E
:;
~oll z0 Cl...Ill 0 Ill
rn ... 0 Ill
rn ... ~m ~I!!
ca t::
0.. 0
u.i
17.5
Cl Cl Cl Cl Cl z z
/ I
w
z / I
~ 18.8
0 2 Ce, Dd,
:e j~
.J ~
~. /
/
Cy, Vs
19.2
Dd, /
Ma, Tu /
23.0

Ma

27.7
>. c:
"C:
w
zw ~~
ui~
"' Q)
~
Ds,Pd,
Ma
g 29.4 "'
..!. c:
CJ
:::i ·-"'
..c: >.
s:: Q)
?Pa,
Ds,Pd
0 c:
31.9
c: Pa,
.!!!
~ Ds,Pd Ds
0
33.4
' c:
Q) al .!!! Pa, Pa,
jBe ?Br
34.5
Q) ' c: Pa,
:gal.!!! Pa,
~6.e Br, Di
~35.5
w ' c:
Pa,
0 >. "C
"i: "'
cu·- Br, Di,
0 woec:
"' ..c:
Ds
w 37.1
ch
Q) c:
-5:i "' Q)
Ds Ds
Cl c:
39.5
c:
Q) ~
ca ::>
_J
.s:
41.3 Figure 4.6B. (Cont.)

Analysis of the locomotor morphology of early Oligocene (see Figures 4.2, 4.4, 4.7), and toward the end of the Oligocene there
(Orellan) carnivorous mammals by Van Valkenburgh (1985) indi- was a reduction in morphospace diversity with more generalized,
cates an emphasis on climbing (among nimravids and daphoenine omnivorous taxa replacing more specialized, carnivorous ones (Van
amphicyonids) and a relative paucity of terrestrial, more cursorial Valkenburgh, 1994 ).
taxa (two species of Hyaenodon). She interprets the habitat contain- The last records of hyaenodontids are in the early early
ing this guild of predators as a mixture of forest and open areas. Arikareean. Although the nimravids persisted into the late early
The dental morphology of these same taxa suggests a focus on meat Arikareean, and possibly into the early late Arikareean (earliest
eating, and an absence of bone crackers (Van Valkenburgh, 1988). Miocene), by the early early Arikareean their diversity and abun-
Throughout this time period seve~al distinct carnivore dance were greatly reduced (see Figure 4.7). Changes also occurred
"paleoguilds" can be distinguished. In general, hyaenodontid diver- within the other carnivore families, including the immigration of the
sity declined while canid (and initially nimravid) diversity increased first oligobunine mustlid Promartes.
 Christine M. Janis et al. 86

iii' ii:' ii:' ii:'

6
6 ~
iii'
!!!. !!!. 2 ~ ze:. 6
~
~ ~ Ill Ill Ill
2 7ii .....
fti2 c
c... 'iii
ftl
c
c·-
... Ill
&:
E·~ f... ia.
c c.5
... ftl
Cll- () i 7ii
Ill

-g ~e
ftl Cll- 0
-3 0 - Ill j!m j! .c II. .c II. lE ::: (.)
5i !!I c .. t::- () .c
zom
I! ·c
MA Cll Cll
-c E
(.)
= <.>u t::-
0 ftl
z I!!
gi
I/I ...
:; ts
0 I!! ~! ...
ftl
II. 0
t:: u.i
!I~~ ~<(
:I
<-' ~oil <-' <-' I/I<-' <-' <-' z z
1.8

~
/
c: Ly,Mi,Pu I
I ' Hm.~ Pu/
ftl
Pu, Pu,
0 Me Hm, .ill,Hm, !:Im.~
Q) c:
Ch / Hm I
j~ I Ch Ch I

x
w I/
z 2.4

~ ~
c: I Ly, Pu, Ly, Pu,
w >-0
ftl Ly, Pu, Ly, Pu
() I 1::1.m Hm, Me, 1::1.m.
0 ~~ Ch Me

xl.X
Ch

x
WCll I/
:i 4.5

l.X
0.
en Nv,?Ps,
.&~:m
a.. c:
I
' Ly, Ad,
Ly, Ly, /

j :i:: '.2 Me i?Mf!, Me Ma Ma ,, /

5.2 I
/

~
/ Ly, Ps,Ly, Nv,
a. c: I Pr,

;X
Ps, Ps,/ /
Ad. Ps,
2 ~g I Ma &!,Ma /
Mfl
j J:E 11 /Ma Ma, Me Mfl
/ - /
6.0

~ ~
.;::. ' c: Ps, Nv, Nv, Ps, Ps,
.... a. I Ps, Ma. Figure 4.7A. Biogeographic ranges
Ma,
~~~ I Mfl Mfl.
?N? of feloids. A "box" (for a particu-
_j J: :c I/
?N? Ba Ba Ba Jar time period in a particular biogeo-

x
7.0 / / / I
~. Nv, Ps, Ps, Nv, Ps, graphic region) with a cross through it
ftl a. c: / Ps / / / Ps,
w E.!!! Ba / / / Ba Ba / ?N? means no fossil localities are known
ui~'.2 for that time period from that area;
I/ '/ I/ I/
.
x
8.8 /
c c: I Nv, Nv, Ps, F? or ?Ba/
/
?Ps, Nv, Ps, / Ps / a single dashed line through the box
Q) ~ -~ means that only scant fossil informa-
- c:
-« !ftl 0 I Ba Ba N? / ?Ba Ba / I

x
w ...Ju 'O i/ tion is available (usually only a sin-
z 9.5 I I/ I/
gle, small, locality). Key: Felidae are
w /
() c c:
~~.~
Ps,
?Ba
Nv, Ps, Nv, Ps, /, Ps I I
in plain type face. Felinae are in ro-
0 .... <ll c:
Ps Ba I
Ba Ba / I man type: Ly = Lynx, Mi = Miraci-

x
:§ «!- 0
w Uu
11.0 / / /
I/
/
I/
.I/ nonyx, Nv = Nimravides, Pr= Prat-
.2? .9 I I ifelis, Pu = Puma, Ps = Pseudailu-
/ / Ps / Nv,Ps /
j~~ I
12.5
_j Cll ·s: I/
/
I/
/
I/
/
IX /
I/
/
11
/
II
rus. Machairodontinae are in roman
type underlined: Ad =Adelphailurus,

14.0
Q)
1ii .8
.-I~ c
w
'

• ftl ftl
Cll ">
[X Ps Ps Ps
IXV /
/
/
Ps Ps
II
I
I Df = Dinofelis, Hm = Homotherium,
Ma = Machairodus, Me = Megan-
teron. F? = Felid indet. Nimravidae
are in boldface: Ba = Barbourofe-

15.8
~~c
6
ftl ftl ftl
w Cll "> ,
I
I
Ps

"
F? Ps
x Ps Ps
~
I
I
!is, DI= Dinaelurus, Dn = Dinictis,
Eu= Eusmilus, Hp= Hoplophoneus,
Na = Nanosmilus, Nr = Nimravus,

~
bi c: I

17.5
Q)·§~
3~.e /
/
/
F? Ps
x Ps
/
/
/

lj
I
Po = Pogonodon. N? = nimravid in-
det. Hyaenidae are in italics: Ch =
Chasmaporthetes.

RUNNINGWATER CHRONOFAUNA Aelurocyon, which was of similar size and proportions to a jaguar
(see Figure 9.1). Hunt and Tedford (1993) term this interval the
The earliest part of the Miocene in North America, from the early- "temnocyonine-amphicyonid-entelodont carnivoran association."
late Arikareean through the early Hemingfordian, has been termed Entelodont artiodactyls may have filled the role of bone-cracking
the "cat gap" (Van Valkenburgh, 1991) or the "Runningwater chrono- scavengers (Joeckel, 1990), a role that was not occupied by a car-
fauna" (Webb and Opdyke, 1995): Nimravids were extinct, and nivoran until later in the Miocene.
felids did not immigrate from the Old World until the late Hem- Canids were diverse in the early Miocene. Hesperocyonines per-
ingfordian. In the absence of catlike predators, a number of taxa sisted, although the Miocene taxa were mainly specialized, more
developed dental parallels for hypercarnivory (Van Valkenburgh, hypercarnivorous forms (Parenhydrocyon, Ectopocynus, Enhydro-
1991). These included hesperocyonine canids (see later), a boro- cyon, and [to a lesser extent] Osbornodon). The earliest borophflgine
phagine canid (Euoplocyon), a temnocyonine amphicyonid canid, Cormocyon, first appeared in the early early Arikareean, but
(Temnocyon), and several mustelids (e.g., Paroligobunis). Perhaps by the earliest Miocene there was now a diversity of borophagines
the most remarkable cat mimicker of this time was the paleomustelid (which later evolved into specialized bone crackers), including the
 Carnivorous mammals 87

6 iii' iii' ii:' ii:' ii:'

~ 6 ~ !!!. !!!. 2 ~ z 6
~
2 7ii
_(.) c c
c·- c.5
Ill Ill
c
Ill e:. ~

~~II. 7ii
I!!::' c"iii
'iii ftl ~ :a ... ftl
~:! () 7ii
--
... ftl ftl
-3 .cGI -
:e() .c~
-g ee Cll -I!! ·cCll
0
(.)
-c ftl
Ill ~m .cm II. ... a. .c 0
~8 c ... (.)
'5 1U t::-
MA = ~~ :I ftl
0 I!! ~m... 0 ftl t:: u.i
!I~~ (.)5i <(E
0 GI ftl
:I
~oil I/I ... z I!! II. 0
17.5
<-' z<-' <-' I/I<-' <-' <-' z z
I I
w /
z / I
w
() 18.8
0
:§ /
/
19.2
?N? /
/
23.0

Nr Po
27.7
.;::. c:
w lii ftl
Eu, Nr,
w ..s::
z . ·c:
Q)
~ Po
w w <( <ll
g29.4
• c:
C1 ""ftl
.s::. >. Dn, Eu, Hp,
:i $: Q)
c: Nr, Po
Nr
0
31.9
c:
..!!! Dn, Hp,
Dn
~ Na, Po
0
33.4
' c:
Q) ~ .!!!
3 Be Dn, Hp Dn
34.5
Q) ' c:
'6 'O <ll Dn,
'O ftl ·-
·- ..c: c
Dn,Hp
::!:u,'2. Hp
~35.5
w >- 'O c:
' <ll
«S·-
() '1::
ftl .s::. c: Dn,Hp Dn
0 WU '2.
w 37.1
ch
~g Q)ffi
0 c:
39.5
c:
.!!!
Q)
c:
-<ll ·-
...J::>
41.3 Figure 4.7B. (Cont.)

procyonidlike, presumably omnivorous, members of the Cynarc- There is some evidence ofbiogeographical differences during late
toides group. The earliest canine canid, Leptocyon, also made its Oligocene/early Miocene times. The amphicyonines were restricted
first appearance at this time (see Figure 4.4). to the central Great Plains in the Arikareean; other amphicyonids
Mustelid diversity also increased in the earliest Miocene, with the were present in Florida. The hesperocyonine canid Osbornodon was
immigration and radiation of a diversity of additional paleomustelids the only canid present in Florida after the early late Arikareean;
and the appearance of neomustelids in the early Hemingfordian (see the canine canid Leptocyon was present only in the central Great
Figure 4.5). Among amphicyonids, the only surviving daphoenine Plains.
was the larger, specialized Daphoenodon, which may have been This "cat-gap" carnivore fauna of the later Oligocene and early
a wolflike predator, and the more derived, probably bone-cracking, Miocene clearly replaced the earlier late Eocene/earlier Oligocene
amphicyonines first appeared in the late late Arikareean. The earliest fauna dominated by hyaenodontids and nimravids. But was the
hemicyonine ursid (or "dog bear"), Cephalogale, also first appeared demise of the more archaic predators due to competition with the
as an immigrant from the Old World in the late late Arikareean (see newer kinds? A similar pattern, that of an increased abundance and
Figure 4.6). diversity in the Chadronian followed by a decline or extinction by
 Christine M. Janis et al. 88

the end of the Oligocene, is seen in a number of ungulate taxa diversified. First appearances in the late Hemingfordian include
(see Chapters 22 and 35), including hyracodontine rhinoceratoids, the lutrine (otters, etc.) Mionictis and three genera of mustelines
agriochoerid oreodontoids, the Mesohippus!Miohippus complex of (weasels, etc.) (see Figure 4.5). Baskin (this volume, Chapter 9)
anchitheriine equids, and basal camelids (such as Poebrotherium). presents an account of the complex pattern of immigrations of
Other groups show a modification of this pattern, where the pre- neomustelids through the Neogene. Finally, procyonine procyo-
dominant radiation is over by the end of the Oligocene, but one nids made their first appearance at the start of the middle
taxon survives and flourishes through the early Hemingfordian. Miocene.
These include hypertragulids (with survival of Nanotragulus), an- Among amphicyonids, the amphicyonines retained their earlier
thracotheres (with the survival of Arretotherium), and entelodonts diversity, with the addition of more derived taxa such as Pliocyon,
(with the survival of Dinohyus). Clearly there was a turnover in the with Pseudocyon and lschyrocyon appearing by the middle Miocene;
entire large mammal fauna in the late Oligocene, affecting a vari- Daphoenodon and the temnocyonines were extinct, however.
ety of taxa, not just carnivores. Because hyaenodontids and nim- Among ursids, the ursine Ursavus, which first appeared in the late
ravids survived for much longer in the Old World (as did ungulates early Arikareean, persisted through the middle Miocene; the more
such as anthracotheres and traguloids), and there was no influx of primitive hemicyonines, Cephalogale and Phoberocyon, were ex-
immigrants to North America during the Oligocene, it seems likely tinct, but the larger and more cursorial Plithocyon appeared in
that the demise of these two carnivore groups was the result of the middle Miocene, replaced by a new hemicyonine genus in the
climatic and habitat changes that affected a particular sector of Clarendonian (see Figure 4.6).
the late Eocene/early Oligocene fauna (see also discussion in Certain changes occurred in the carnivoran composition dur-
Van Valkenburgh, 1994). ing the duration of this fauna. Both amphicyonids and hemicy-
onine ursids declined in abundance: By the Clarendonian there
remained only the new hemicyonine genus and large, specialized
CLARENDONIAN CHRONOFAUNA
amphicyonids such as Pseudocyon and Ischyrocyon (see Figure
The mid-Miocene camivoran fauna that commenced in the latest 4.6). Canine canids marginally increased their diversity with the
early Miocene (late Hemingfordian) and persisted through the early- addition of the extant genus Vulpus in the early late Miocene (see
late Miocene (Clarendonian) represents the predatory component Figure 4.4 ). Among feloids, the larger, puma-sized felid Nimravides,
of Webb's stable "Clarendonian chronofauna" (Webb, 1989), also with moderate saber-toothed modifications, appeared at the end of
termed the "amphicyonine-amphicyonid carnivoran association" by the middle Miocene, and at the start of the late Miocene the lion-
Hunt and Tedford ( 1993 ). Webb and Opdyke ( 1995) perceive the late sized, saber-toothed nimravid Barbourofelis immigrated from the
Hemingfordian component to be a separate "Sheep Creek chrono- Old World (see Figure 4.7). Among mustelids, mephitines (skunks),
fauna": However, the only distinctive difference between the late ischyrictines (perhaps ancestral to wolverines), and taxidiines (bad-
Hemingfordian and early Barstovian carnivore faunas is the greater gers) made their first appearance as immigrants from the Old World
diversity of borophagine canids and the immigration of the procy- (see Figure 4.5).
onids. The Clarendonian chronofauna has often been compared with There is some evidence of biogeographic variation: Amphicy-
that of the present-day East African savannas (e.g., Webb, 1983): onids are excluded from the Gulf Coast regions after the early mid-
Note, however, that in comparison with that present-day fauna, dle Miocene, and the canine canid Leptocyon was for the most part
the diversity of camivorans was lower (Van Valkenburgh, 1985), restricted to the more central regions of the continent until the start
and there was an apparent absence of pursuit predators (Janis and of the late Miocene. The demise of the Clarendonian chronofauna
Wilhelm, 1993). appears to be related to climatic and vegetational changes, with the
The late Hemingfordian (latest early Miocene) marked not only loss of the more mesic type of savanna habitat (Webb, 1989; Janis,
the immigration of felids to North America, with the first appearance 1993). Many browsing ungulates went extinct or were reduced in
of Pseudailurus, but also many other turnovers in the diversity and diversity at this time (see Chapters 22 and 35).
abundance of carnivores. Many of the groups that had been an initial
component of the "cat-gap" fauna, as well as survivors from the
MIO-PLIOCENE CHRONOFAUNA
earlier fauna, were now extinct, while groups first appearing in the
earliest Miocene diversified. The final Tertiary carnivore fauna originated in the Hemphillian
Among canids, there was only a single survivor past the early (late late Miocene) and persisted through the Blancan (Pliocene),
Hemingfordian of the both hesperocyonines and the procyonidlike equivalent to the "felid-ursid-canid carnivoran association" of Hunt
Cynarctoides group of borophagines, represented by Osbornodon and Tedford (1993) (although note that Webb and Opdyke [1995]
and Cynarctoides, respectively: Both taxa survived into the early do not recognize this chronofauna as a discrete unit). This time
Barstovian. Other borophagines flourished, with addition of the also marked a period of renewed faunal interchange with the Old
genera Aelurodon and Epicyon (both probably predaceous and/or World, and also, increasingly, with South America. The diversity of
bone crackers), and Carpocyon and Cynarctus (both probably omni- both carnivores and herbivores was lower than in present-day East
vorous) in the middle Miocene. Canine canids were still only rep- African savanna habitats; but, as in present-day East Africa, and
resented by the foxlike Leptocyon (see Figure 4.4). unlike the condition today in North America, a diversity of bone-
Paleomustelids were extinct, with the exception of Brachypsalis, cracking predators was present, probably reflecting some type of
which persisted into the early late Miocene, and neomustelids open savanna environment (Van Valkenburgh, 1988).
 Carnivorous mammals 89

Among the canids, some borophagines were now specialized The Clarendonian chronofauna commenced at the end of the
bone crackers, represented by larger species of Osteoborus in the early Miocene in the late Hemingfordian, although a separate late
late Miocene and by Borophagus in the Pliocene. Canine canids Hemingfordian "Sheep Creek chronofauna" may perhaps be recog-
increased in diversity, with the addition of the extant genus Canis nized. This fauna saw the addition by immigration of procyonids
in the Miocene and the extant South American genera Cerdocyon, and felids, diversification among hemicyonine ursids (dog bears),
Urocyon, and Chrysocyon in the Pliocene. Larger, wolflike canids radiation of amphicynodontine amphicyonids, and continued radi-
did not occur until the Pliocene (see Figure 4.4). ation of mustelids and borophagine canids. Finally, this fauna was
This period was also the time of the greatest Tertiary diversity of replaced at the start of the Hemphillian by a Mio-Pliocene chrono-
mustelids, with the immigration of the galictines (grissons), muste- fauna, characterized by the extinction of the amphicyonids and ra-
Iines (weasels, etc.), and modern types of mephitines (skunks). diation of canine canids and large, specialized borophagine canids.
Among the procyonids, certain components of the Clarendonian Both conical-toothed and saber-toothed felids, ursine ursids, and
chronofauna had become extinct, but the genus Bassariscus (extant present-day subfamilies of mustelids all arrived by immigration dur-
today) persisted from this fauna, joined in the Pliocene by two other ing this time period.
extant genera, Nasua and Procyon (see Figure 4.5). Amphicyonids
were extinct, and ursids were now represented only by ursine ursids.
Agriotherium, Plionarctos, and Indarctos were the late Miocene ACKNOWLEDGMENTS
genera: The former two genera persisted into the Pliocene, but the
latter genus was only known during the late early Hemphillian. In We are most grateful to David Polly for providing an expanded ver-
the Pliocene there were further immigrations of the extant genera sion of his 1994 abstract, from which the discussion of the position
Ursus and Tremarctos (see Figure 4.6). of the Creodonta among the Eutheria is largely derived, and for his
There was considerable diversification among the feloids. Bar- comments on the manuscript. Thanks for comments are also due to
bourofelis became extinct before the end of the Miocene, but it Leigh Van Valen and Blaire Van Valkenburgh.
was contemporaneous for a while with an immigrant saber-toothed
felid, Machairodus. Both saber-toothed and conical-toothed felids
were diverse in the Pliocene, with the presence of the living gen- REFERENCES
era Lynx and Puma and the immigration of derived saber-toothed
felids such as Homotherium and Dinofelis (see Figure 4.7). The Anders, A. V. (1992). Reappraisal of the Eocene Groundbird Diatryma
(Aves: Anserimorphae). In Papers in Avian Paleontology Honoring
cheetahlike hyaenid Chasmaporthetes immigrated to North Amer-
Pierce Brodkorb, ed. K. E. Campbell, Jr., pp. 109-26. Los Angeles:
ican in the Blancan, probably representing the first occurrence on No. 36, Science Series, Natural History Museum of Los Angeles
the North American continent of a true pursuit predator (Janis and County.
Wilhelm, 1993). Baskin, J. A. ( 1981 ). Barbourofelis (Nimravidae) and Nimravides (Felidae ),
with a description of two new species from the late Miocene of
Florida. Journal of Mammalogy, 62, 122-39.
SUMMARY Baskin, J. A., & Tedford, R. H. (1996). Small arctoid and feliform
carnivorans. In The Terrestrial Eocene-Oligocene Transition in North
Several distinct fauna! groupings of carnivores existed over the America, eds. D.R. Prothero & R. J. Emry, pp. 486-497. Cambridge:
course of the Tertiary: first, a Paleocene fauna with a limited ra- Cambridge University Press.
diation of viverravine "miacoids," perhaps distinguishable from the Butler, P. M., & Kielan-Jaworowska, Z. (1973). Is Deltatheridium a marsu-
fauna of the later Paleocene that also included oxyaenid creodonts; pial? Nature, 245, 105-6.
Bryant, H. N. (1991). Phylogenetic relationships and systematics of the
second, an early to middle Eocene fauna of oxyaenid, limnocyonid,
Nimravidae (Carnivora). Journal of Mammalogy, 72, 56-78.
and proviverrine hyaenodontid creodonts, and miacine "miacoids." Cappetta, H., Jaeger, J.-J., Sige B., Sudre, J., & Vianey-Liaud, M.
The earlier middle Eocene fauna is characterized by the decline of (1987). Complements et precesions biostratigraphiques sur la fauna
the oxyaenids and the appearance of doglike ecomorphs among the Paleocene a mammireres et selaciens du Bassin d'Ouarzazate
miacines. A distinct turnover occurred in the later middle Eocene (Maroc). Tertiary Research, 8, 147-57.
Cope, E. D. (1875). On the supposed Carnivora of the Eocene of the
(Duchesnean), with the resurgence of the doglike hyaenodontine
Rocky Mountains. Proceedings of the Academy of Natural Science
creodonts and the late Eocene immigration of the nimravids to of Philadelphia, Nov. 1875, 444-8.
fill a catlike role, representing the commencement of the White (l 880a). On the genera of the Creodonta. Proceedings of the American
River chronofauna. Early members of other more modem groups of Philosophical Society, 1882, 76-82.
carnivorans, the amphicyonids, canids, ursids, and mustelids, were (1880b ). On the extinct cats of North America. American Naturalist, 14,
833-58.
present as smaller, generalized forms. This fauna persisted until near
(1884). The Vertebrata of the Tertiary Formations of the West.
to the end of the Oligocene when, following the extinction of the Washington, D.C.: U.S. Government Printing Office.
hyaenodontids and nimravids, it was replaced by an Oligo-Miocene Davis, D. D. (1949). The shoulder architecture of bears and other carnivores.
Runningwater or "cat-gap" carnivore fauna, Principal larger preda- Fieldiana, Zoology, 31, 285-305.
tors were now borophagine canids and temnocyonine amphicyonids, Emry, R. J. (1981). Additions to the mammalian fauna of the type Duches-
nean, with comments on the status of the Duchesnean "Age." Journal
with mustelids as the smaller forms. A number of taxa developed
of Paleontology, 55, 563-70.
hypercamivory in the absence of catlike predators, including some Estes, R. D. (1990). The Behavior Guide to African Mammals. Berkeley:
of the surviving hesperocyonine canids. University of California Press.
 Christine M. Janis et al. 90

Flower, W. H. (1869). On the value of the characters of the base of the cra- tion of Systematics in Evolutionary Biology. Unpublished Ph.D. dis-
nium in the classification of the Order Carnivora and the systematic sertation. Berkeley: University of California.
position of Bassaris and other disputed forms. Proceedings of the ( 1994 ). What, if anything, is a creodont? Journal of Vertebrate Paleon-
Zoological Society of London, 1869, 4-37. tology, 4 (suppl. to no. 3), p. 42A.
Fox, R. C., & Youzwyshyn, G. P. (1994). New primitive carnivorans Savage, R. J. G. (1977). Evolution in carnivorous mammals. Palaeontology,
(Mammalia) from the Paleocene of western Canada, and their bearing 20, 237-44.
on the relationships of that order. Journal of Vertebrate Paleontology, Schlosser, M. (1888-1890). Die affen, lemuren, chiropteren, insectivoren,
14, 382-404. marsupialier, creodonten und carnivoran des Europaischen Tertiars
Flynn, J. J., & Galiano, H. (1982). Phylogeny of Early Tertiary Carnivora, und deren beziehungen zu ihren lebenden und fossilen Aussereu-
with a description of a new species of Protictis from the Middle ropaischen verwandten. Beitriige zur Paliiontolgie Oesterreich-
Eocene of northwestern Wyoming. American Museum Novitates, Ungarns, 6.
2725, 1-64. Simpson, G. G. ( 1945). The principles of classification and a classification
Flynn, J. J., Neff, N. A., & Tedford, R.H. (1988). Phylogeny of the Car- of mammals. Bulletin of the American Museum of Natural History,
nivora. In The Phylogeny and Classification of the Tetrapods, Vol. 2: 85, 1-350.
Mammals, ed. M. J. Benton, pp. 73-116. Systematics Association Sloan, R. E. ( 1969). Cretaceous and Paleocene terrestrial communities of
Special Volume no. 35B. Oxford: Clarendon. western North America. Proceedings of the North American Paleon-
Gregory, W. K., & Hellman, M. (1939). On the evolution and major classi- tological Convention, I, 427-53.
fication of civets (Viverridae) and allied fossil and recent Carnivora: Tedford, R. H. (1976). Relationship of pinnipeds to other carnivores (Mam-
a phylogenetic study of the skull and dentition. Proceedings of the malia). Systematic Zoology, 25, 363-74.
American Philosophical Society, Philadelphia, 81, 309-92. Tedford, R. H., Barnes, L. B., & Ray, C. E. (1994). The early Miocene
Hunt, R. M., Jr. (1974). The auditory bulla in Carnivora: an anatomical basis littoral ursoid carnivoran Kolponomos: systematics and mode of life.
for reappraisal of carnivore evolution. Journal of Morphology, 143, In Contributions in Marine Mammal Paleontology Honoring Frank
21-76. C. Whitmore, Jr., eds. A. Berta & T. A. Demere. Proceedings of the
(1977). Basicranial anatomy of Cynelos Jourdan (Mammalia: Car- San Diego Society of Natural History, 29, 11-32.
nivora), an Aquitanian amphicyonid from the Allier Basin, France. Van Valen, L. (1966). Deltatheridia, a new order of mammals. Bulletin of
Journal of Paleontology, 51, 826-43. the American Museum of Natural History, 132, 1-126.
(1987). Evolution of the aeluroid Carnivora: significance of auditory Van Valkenburgh, B. (1985). Locomotor diversity within past and present
structure in the nimravid cat Dinictis. American Museum Novitates, guilds of large predatory mammals. Paleobiology, 11, 406-28.
2886, 1-74. (1988). Trophic diversity within past and present guilds oflarge predatory
Hunt, R. M., Jr., & Tedford, R.H. (1993). Phylogenetic relationships within mammals. Paleobiology, 14, 155-73.
the aeluroid Carnivora and implications of their temporal and geo- ( 1991 ). Iterative evolution of hypercarnivory in canids (Mammalia: Car-
graphic distribution. In Mammal Phylogeny: Placentals, eds. F. S. nivora): evolutionary interactions among sympatric predators. Pale-
Szalay, M. J. Novacek, & M. C. McKenna, pp. 53-73. New York: obiology, 17, 340-62.
Springer-Verlag. (1994). Extinction and replacement among predatory mammals in the
Janis, C. M. (1993). Tertiary mammal evolution in the context of changing North American Late Eocene and Oligocene: tracking a paleoguild
climates, vegetation, and tectonic events. Annual Review of Ecology over twelve million years. Historical Biology, 8, 129-50.
and Systematics, 24, 467-500. Vrana, P. B., Milinkovitch, M. C., Powell, J. R., & Wheeler, W. C. (1994).
(1995). Patterns of taxonomic diversity in Paleogene carnivorous mam- Higher level relationships of the Arctoid Carnivora based on se-
mals: competition or ecological opportunism? Journal of Vertebrate quence data and "total evidence." Molecular Phy lo genetics and Evo-
Paleontology, 15 (suppl. to no. 3), p. 37A. lution, 3, 47-58.
Janis, C. M., & Wilhelm, P. B. (1993). Were there mammalian pursuit preda- Wayne, R. K., Benveniste, R. E., Janczewske, D. N., & O'Brien, S. J. (1989).
tors in the Tertiary? Dances with wolf avatars. Journal ofMammalian Molecular and biochemical evolution of the Carnivora. In Carnivore
Evolution, 1, 103-26. Behavior, Ecology, and Evolution, ed. J. L. Gittleman, pp. 465-94.
Joeckel, R. M. (1990). A functional explanation of the masticatory system Ithaca: Cornell University Press.
and paleoecology of entelodonts. Paleobiology, 16, 459-82. Webb, S. D. (1983). The rise and fall of the late Miocene ungulate fauna
Lillegraven, J. A. (1969). Latest Cretaceous mammals of upper part of Ed- in North America. In Coevolution, ed. M. H. Nitecki, pp. 267-306.
monton Formation of Alberta, Canada, and review of marsupial- Chicago: University of Chicago Press.
placental dichotomy in mammalian evolution. University of Kansas (1989). The fourth dimension in North American terrestrial mammalian
Paleontological Contributions, 50, 1-122. communities. In Patterns in the Structure of Mammalian Commu-
Maas, M. C., & Krause, D. W. (1994). Mammalian turnover and community nities, eds. D. W. Morris, Z. Abramsky, B. J. Fox, & M. R. Willig,
structure in the Paleocene of North America. Historical Biology, 8, pp. 181-203. Lubbock: Special Publication of the Museum, Texas
91-128. Tech University, no. 28.
MacPhee, R. D. E., & Novacek, M. J. (1993). Definition and relationships of Webb, S. D., & Opdyke, N. D. (1995). Global climatic influence on Cenozoic
Lipotyphla. In Mammal Phylogeny: Placentals, eds. F. S. Szalay, M. J. land mammal faunas. In Studies in Geophysics: Effects ofPast Global
Novacek, & M. C. McKenna, pp. 12-31. New York: Springer-Verlag. Change on Life, pp. 184-208. Board on Earth Sciences and Resources
McKenna, M. C. (1975). Toward a phylogenetic classification of the Mam- Commission on Geosciences, Environment, and Resources National
malia. In Phylogeny of the Primates, eds. W. P. Luckett & F. S. Szalay, Research Council. Washington, D.C.: National Academy Press.
pp. 21-46. New York: Plenum. Wible, J. R. (1986). Transformations in the extracranial course of the inter-
Matthew, W. D. ( 1909). The Carnivora and Insectivora of the Bridger Basin, nal carotid artery in mammalian phylogeny. Journal of Vertebrate
Middle Eocene. Memoirs of the American Museum of Natural His- Paleontology, 6, 313-25.
tory, 9, 291-567. Wozencraft, W. C. (1989). The phylogeny of the Recent Carnivora. In Car-
(1910). The phylogeny of the Felidae. Bulletin of the American Museum nivore Behavior, Ecology, and Evolution, ed. J. L. Gittleman, pp.
of Natural History, 28, 289-316. 536-68. Ithaca: Cornell University Press.
Neff, N. A. ( 1983 ). The Basicranial Anatomy of the Nimravidae (Mammalia: Wyss, A. R., & Flynn, J. J. (1993). A phylogenetic analysis and definition of
Carnivora): Character Analyses and Phylogenetic Inferences. Un- the Carnivora. In Mammal Phylogeny: Placentals, eds. F. S. Szalay,
published Ph.D. dissertation. New York: City University. M. J. Novacek, & M. C. McKenna, pp. 32-52. New York: Springer-
Polly, P. D. (1993). Hyaenodontidae (Creodonta, Mammalia) and the Posi- Verlag.
5 Creodonta

GREGG F. GUNNELL

INTRODUCTION

The Creodonta was a group of archaic, predatory, carnivorous mam-

mals that ranged widely, both temporally and geographically, during
the Cenozoic. Creodonts are known from the late Paleocene (per-
haps earlier in the Paleocene as well) through late Oligocene of
North America, from early Eocene through late Oligocene in Eu-
rope, from early Eocene through late Miocene in Asia, and from the
Paleocene through late Miocene in Africa.
Two distinct families of creodonts are generally recognized,
Oxyaenidae and Hyaenodontidae. A third group, limnocyonids, may
be properly recognized as a distinct family (Gazin, 1946, 1952,
1962), although they have often been included in oxyaenids (Wort-
man, 1902; Matthew, 1909) or hyaenodontids (Denison, 1938; Van
Valen, 1966). Oxyaenids are first recognized from late Paleocene
deposits in North America, and the principal radiation and diversi-
fication of this family occurred in North America. Oxyaenids dis-
appeared after the middle Eocene in North America and by the late
Eocene in Europe and Asia. No oxyaenids are known from Africa.
Hyaenodontids first appeared in North America and Europe in the
early Eocene and may have been present in the late Paleocene of
North Africa (Gheerbrant, 1990). They were present through the
Oligocene in North America and Europe and disappeared in the
Miocene of Asia and Africa. Hyaenodontids were represented by
diverse endemic radiations in North America, Europe, and Africa.
Creodonts ranged in body size from that of a small cat or fox to
hyaena or lion-sized. Most were probably active predators, although
some species were likely scavengers and, perhaps, carrion feeders.
Creodont diversity was greatest during the early and middle Eocene.
Their decline and subsequent extinction may have been the result
of competition with true Carnivora in combination with climatic Figure 5.1. Restoration of A. Machaeroides eothen and B. Arfia opistho-
changes after the middle Eocene. toma, by Brian Regal.

91
 Creodonta 92

DEFINING FEATURES OF ORDER CREODONTA Creodonts had either two or three pairs of carnassial teeth, but the
cutting function was concentrated in only one pair (either Ml/m2
or M2/m3) in adulthood (less so in some proviverrines). Among
CRANIAL
dental features, enlarged lateral incisors, reduced metaconids, open
Small brains, placed low in skull; large sagittal crests and broad tem- trigonids, reduced or absent talonids, and elongate postmetacristae
poral fossae in most taxa; auditory bullae unossified primitively; are all derived for the order (see Figure 5.4).
mastoid processes normally enlarged and broad; postorbital con-
striction distinct, separating the skull into well-defined splanch-
nocranium and neurocranium (see Figures 5.2 and 5.3). Of these
cranial features, large sagittal crests and large, broad mastoids prob-
ably represent derived characters for the order.

DENTAL
Upper and lower canines enlarged and pointed; lateral incisors en-
larged, medial incisors small in most species; premolars usually
simple and pointed; P4/p4 often more complex, even semimolari-
form; lower molars with high trigonids and long, relatively narrow
talonids; trigonids dominated by high, well-developed protoconid,
with paraconid and metaconid (if present) lower, but often distinct
(particularly the paraconid); more derived forms with open trigonids
that form long, tall cutting blades; metaconid same size or smaller
than paraconid in most primitive taxa and often completely lost
in derived taxa; talonids reduced or lost in derived species; upper
molars (and P4) with extended postmetacristae forming elongate
cutting surfaces.
Figure 5.3. Skull of Limnocyon verus (AMNH 12155): A. Lateral view. B.
A Ventral view (after Matthew, 1909). Note long, narrow palate and relatively
narrow skull.

Figure 5.4. Creodont dentitions. A. Tritemnodon strenuus (AMNH 16214),

right dentary in occlusal and lateral views. Note presence of three molars,
laterally compressed premolars, and well-developed metaconid on m2-3. B.
Figure 5.2. Skull of Patriofelis ulta (University of Utah# 850): A. Lateral Oxyaena gulo (AMNH 15199), right dentary in occlusal and lateral view.
view. B. Ventral view (after Gazin, 1957). Note short, broad palate and Note presence of only two molars and well-developed metaconid on ml-2.
wide skull. Figures redrawn from Matthew (1915).
 Gregg F. Gunnell 93

POST CRANIAL SYSTEMATICS

Generalized postcranial skeleton; limbs mesaxonic, ranging from
plantigrade to digitigrade; terminal phalanges are fissured claws; SUPRAORDINAL
unfused scaphoid, lunate, and centrale; astragalartrochlea smooth to
The origin and relationships of Creodonta are still elusive. The oldest
grooved; astragalar head flattened and often obliquely angled from
known representatives are found in North America
body; femur with distinct third trochanter; humerus with shallow
(Gingerich, 1980, 1986, 1989; Fox, 1990). Tytthaena, one of the
olecranon fossa and deep supratrochlear fossa (see Figure 5.5). Of
most primitive creodonts (Gingerich, 1980), shows strong simi-
these characters, fissured claws and flattened astragalar heads can
larities to early palaeoryctid insectivorans, and it is probable that
both be viewed as derived for the order.
creodonts arose from such a group (Van Valen, 1966; Lillegraven,
1969; Sloan, 1969; Gunnell and Gingerich, 1991). The palaeoryctid
insectivoran Cimolestes is a reasonable ancestral morphotype for
creodonts and, perhaps, true carnivorans, as well (Flynn, Neff, and
Tedford, 1988; see Fox and Youzwyshyn, 1994, for a differing view
of carnivoran relationships).
The relationships between oxyaenids and hyaenodontids are also
not well understood. Although these groups share many features in
common, there are many differences as well. Oxyaenids are well es-
tablished in North America when hyaenodontids suddenly appear in
the fossil record at the beginning of the Eocene. Hyaenodontids are
already a diverse group when they appear, suggesting that they were
immigrants (the earliest perissodactyls, artiodactyls, and primates
also appear at the same time, indicating a major immigration into
North America at the beginning of the Eocene). Gingerich (1980)
has suggested that hyaenodontids may have had a separate origin
in the Old World because of the large radiation of hyaenodontids
in Europe, Africa, and Asia. If this is true, and if the North African
hyaenodontid is truly oflate Paleocene age, then the order Creodonta
Figure 5.5. Reconstructed skeletons: A. Sinopa rapax (adapted from is diphyletic (Flynn, Neff, and Tedford, 1988). Figures 5.6 and 5.7
Matthew, 1906). B. Patriofelisferox (adapted from Osborn, 1900). show the interrelationships within each family.

g: Figure 5.6. Oxyaenidae Relation-

c:: ~ c:: ~ ships. Key to characters at nodes:
.2u .!,!
c::
.g cu
~ (1) Broad, relatively short skull; pre-
Q ~ 53cu ::::
~ Cll cu .r::: cu glenoid process present; M 1/m2 de-
.Q Ill Ill
e .!!! .e. .e.
~ ~ veloped as carnassials; M3/m3 ab-

,,
~ Cl Cl
sent; M2 transverse, reduced; P4

, , and Ml with relatively short post-

, , ,
metacristae; ml-2 of similar size;
limbs relatively short. (2) pl sin-
, ,
,, ,,
gle rooted. (3) m2 smaller than

, , ml. (4) PALAEONICTIDAE: In-

creasing size; robust, blunt-cusped

, ,, teeth; accessory cuspules on p4.

(5) M2 vestigial or absent; m2 much
,
,,
smaller than ml; p4 with large ante-
rior basal cusp. (6) OXYAENIDAE:
, , P3 with protocone reduced or absent;

, ,, distal tibia angled; radial head ellip-

tical. (7) P4 and Ml with elongate
,
,,
postmetacristae; lower molar trigo-
nids longer than wide; m2 larger than
, ml; distal tibia flattened, robust; radial
head rectangular, expanded; M2 meta-
cone absent. (8) Ml postmetacrista
very elongate; m2 with reduced or absent metaconid and talonid; m2 much larger than ml; M2 reduced or absent; limbs short and very robust.
(9) MACHAEROIDINAE: Upper canine enlarged, elongate, saberlike; lower canine reduced; preglenoid process absent; mandible with anteromedial
flange; mandible laterally flared anteriorly; P3-4 protocones absent; m2 talonid tiny or absent; scaphoid, lunate, and centrale fused.
 Creodonta 94

Figure 5.7. Hyaenodontidae Re-

lationships. Key to characters at
nodes: (I) Skull relatively nar-
row; astragalar-cuboid articulation
small or absent; lower molar trigo-
nids longer than talonids; limbs
relatively slender, lightly built.
(2) Frontal bone concave; M2/m3
developed as carnassials. (3) M3/m3
reduced. (4) LIMNOCYONIDAE:
P2-3 lack protocones; Ml/m2 devel-
oped as carnassials; M2-3 reduced.
(5) M3/m3 absent. (6) ml smaller
than m2. (7) PROVIVERRINAE:
Lower premolars elongate, laterally
compressed; mandibles shallow;
lower molar paraconids and meta-
conids smaller than protoconids;
astragalus with spiral facet; pl
separated from p2 by a diastema.
(8) pl-3 separated by diastemata;
mandibles elongate; p 1 single rooted.
(9) Increased body size; upper molars with paracone and metacone separated; molar talonids larger than trigonids. (10) Crenulated
enamel. (11) HYAENODONTINAE: Molars sectorial; lower molar metaconids reduced or absent; lower molar talonids reduced or
absent; premolars short and high; very large body size.

INCLUDED NORTH AMERICAN GENERA IN TYTTHAENINAE

THE ORDER CREODONTA
Characteristics: small size; lack well-developed postmetacristae on
P4 and M 1; m2 smaller than ml; narrow talonids on lower molars.
Due to the complex nature of synonymy among creodont taxa, pos-
sible synonyms for each taxon are listed in the appendix at the end Tytthaena Gingerich, 1980
of this chapter, rather than discussed within the text. Type species: Tytthaena parrisi Gingerich, 19SO.
The locality numbers listed for each genus refer to the list of uni- Type specimen: YPM-PU 22353.
fied localities in Appendix I. The acronyms for museum collections Characteristics: As for subfamily (Rose, 19Sl; Gunnell and
are listed in Appendix III. Gingerich, 1991).
The locality numbers may be listed in a couple of alternative ways. Average length of ml: 7.0 mm.
Parentheses around the locality (e.g., [CP101]) mean the taxon in
Included species: T. parrisi (known from localities CP13E,
question at that locality is cited as an "aff." or "cf." the taxon in CP24A); T. lichna (locality CPI 7A).
question. Parentheses are usually used for individual species, thus
implying the genus is firmly known from the locality, but the actual
species identification may be questionable. Question marks in front OXYAENINAE
of the locality (e.g., ?CPlOl) mean the taxon is questionably known Characteristics: P3-4 with protocone; protocone reduced on P2
from that locality, thus implying some doubt that the tax on is actually (sometimes on P3 as well); medium to large sized, with short ros-
present at that locality, either at the genus or the species level. trum; reduction of premolar number; reduction or loss of M2; sec-
torial dentitions.
OXYAENIDAE
Dipsalidictis Matthew, 1915
Defining features: small, wide braincase placed low in skull; occiput Type species: Dipsalidictis platypus Matthew, 1915.
wide at base, narrowing dorsally; tympanic bullae unossified; nasals Type specimen: AMNH 15S75.
long and often expanded posteriorly; lacrimal forms semicircular ex- Characteristics: Dental formula I3/2, Cl/1, P4/4, M2/2; P3
pansion on face; lacrimal foramen contained within orbit; mandibles simple with small protocone (sometimes lost); P4 wide
heavy with a stout symphysis (see Figure 5.2); Ml and m2 carnas- transversely with large protocone lobe; Ml wide trans-
sials; M3/m3 absent; Ml with connate paracones and metacones; versely; M2 transverse and somewhat reduced; p4 with tiny
M2 transverse and reduced or absent (see Figure 5.4B); manus and to small anterior basal cusp; ml-2 subequal in size; lower
pes mesaxonic, pentadactyl, and plantigrade or subplantigrade; the molar trigonids wider than long, with large metaconids
fibula articulates with calcaneum; the astragalus articulates with and well-developed talonids (Jepsen, 1930a; Bown, 1979;
cuboid; astragalar trochlea flattened and smooth or only slightly Mc Kenna, 19SO; Rose, 19S l; Gingerich, 19S9; Gunnell
grooved; scaphoid, lunar, and centrale distinct and separate; pha- and Gingerich, 1991).
langes compressed and fissured at tip (see Figure 5.5B). Average length of ml: S.3-13.2 mm.
 Gregg F. Gunnell 95

Included species: D. platypus (known from localities CPI 7B, Characteristics: P3 with large protocone; Ml with small pro-
CPlSB, CP19, CP26C); D. aequidens (localities SB20C, tocone; ml reduced with metaconid appressed to proto-
CP17B, CPISB); D. transiens (localities CP19, CP20A, conid; m2 with small talonid, vestigial metaconid, enlarged
[CP27A], CP63); D. krausei (localities CP17A, CP17B, paraconid and protoconid; mandible deep (Gazin, 1957;
CPlSB, CP26C). Guthrie, 1971 ).
Dipsalidictis sp. is also known from localities CPI SA, Average length of m2: 25.0 mm.
CC50. Included species: P. tigrinus only, known from locality
CP27D only.
OxyaenaCope,1874
Type species: Oxyaena lupina Cope, 1S74.
PALAEONICTINAE
Type specimen: USNM 1049.
Characteristics: Dental formula I3/2, Cl/l, P4/4, M2/2; Il- Characteristics: P4 and Ml with short postmetacristae; accessory
2 small, I3 larger; Pl/pl single rooted; Ml with strong cusps on p4; m2 smaller than ml; angle of jaw bluntly rounded;
protocone; M2 transverse; well-developed postmetacristae teeth robust and bluntly pointed.
on P4 and Ml; ml-2 with small talonids, relatively well
developed to small metaconid internal to protoconid; ros- Palaeonictis Blainville, 1842
trum deep and heavy; mandible deep; occiput wide, mas- Type species: Palaeonictis gigantea Blainville, 1S42.
toid moderately stout; limbs and feet of moderate length Type specimen: Paris, MNHN unnumbered specimen (fig-
(Cope, 1SS4a, 1SS4b; Osborn, 1S97, 1900; Kelley and ured by Blainville, l S42).
Wood, 1954; Robinson, 1966; West, 1973; Kihm, 19S4; Characteristics: Dental formula I3/3, Cl/l, P4/4, M2/2; Ml
Davidson, 19S7; Gunnell and Gingerich, 1991). with well-separated paracone and metacone, large proto-
Average length of ml: 6.0--15.0 mm. cone and small conules; M2 vestigial; ml-2 with large and
Included species: 0. lupina (known from localities SB22B, well-separated protoconid and paraconid, well-developed
SB24, [CP25C], [CP25G], [CP25H], [CP64B]); 0. gulo metaconid, and large, basined talonids; m2 smaller than
(localities CP19, CP20A, CP26D, [CP64A]); 0. interme- ml (Osborn and Wortman, 1S92; Rose, 19Sl; Kihm, 19S4;
dia (localities CP19, CP20A); 0. forcipata (localities Gunnell and Gingerich, 1991).
SB24, CP19, CP20B, CP20C, [CP25H], CP27B, CP64B); Average length of ml: 12.5-16.9 mm.
0. pardalis (localities CP20D, [CP64C]); 0. simpsoni (lo- Included North American species: P. peloria (known from
cality SB24). locality CPI SB); P. occidentalis (localities CP19, CP20A).
Oxyaena sp. is also known from locality CP25F. Palaeonictis sp. is also known from locality CP62C.

Patriofelis Leidy, 1870 Ambloctonus Cope, 1875

Type species: Patriofelis ulta Leidy, l S70. Type species: Ambloctonus sinosus Cope, 1S75.
Type specimen: USNM 105. Type specimen: USNM 2329.
Characteristics: Dental formulaI?/3, Cl/l, P3/3, Ml/2; Pl/pl Characteristics: Dental formula I?/?, C 1/1, P4/4, Ml-2/2; M2
absent; M2 absent; Ml protocone small or absent; well- vestigial or absent; p4 with large anterior basal cusp; ml
developed postmetacristae on P4 and Ml; ml with small with elongated trigonid cusps, large, basined talonid; m2
talonid and posteriorly placed metaconid; m2 large, com- with small or absent talonid; m2 smaller than m 1 (Matthew,
pletely sectorial with vestigial metaconid and talonid; ros- 1915; Denison, 193S; Gazin, 1952; Lucas et al., 19Sl).
trum short and heavy; mandibles massive; zygomatic Average length of ml: 16.0--19.0 mm.
arches wide and deep; occiput broad, sagittal crest very Included species: A. sinosus (known from localities SB24,
heavy; limbs and feet short and heavy (Wortman, 1S94; CP20B); A. major (localities CP20C, CP25F, CP25G); A.
Adams, 1S96; Osborn, 1900; Matthew, 1909; Thorpe, priscus (localities CPI 9, CP20A); A. hyaenoides (locality
1923; Denison, 193S; Gazin, 1957; Robinson, 1966; SB24).
Gunnell and Bartels, 1994).
Average length of ml: 16.0--19.0 mm. Dipsalodon Jepsen, 1930b
Included species: P. ulta (known from localities [SB22C], Type species: Dipsalodon matthewi Jepsen, 1930b.
CP34C); P.ferox (localities CP34D, CP3SB); P. coloraden- Type specimen: YPM-PU 13152.
sis (locality SB22C). Characteristics: Dental formula I?/?, Cl/1, P?/4, M?/2; cl and
Patriofelis sp. is also known from localities CP27E, p 1-4 robust; p4 with large anterior basal cusp; ml slightly
CP31E, CP34A, PN5A. larger than m2; ml-2 with unreduced metaconid, no car-
nassial development, large, basined talonids; hypoconid,
Protopsalis Cope, 1880 entoconid, and hypoconulid distinct on ml; mandible deep
Type species: Protopsalis tigrinus Cope, ISSO. and robust (Rose, 19Sl; Gunnell and Gingerich, 1991).
Type specimen: AMNH 4S05. Average length of ml: 12.3-14.2 mm.
 Creodonta 96

Included species: D. matthewi (known from localities CPI 7B, articulates with calcaneum; astragalar-cuboid articulation reduced
?CP26C, CP62B); D. churchillorum (localities CPl3G, or absent; terminal phalanges compressed and fissured at tip; cen-
CP26C). trale, scaphoid, and lunar unfused (except perhaps for Pterodon,
Cf. Dipsalodon sp. is also known from locality CPI 7B. Polly, pers. comm., 1995).

INDETERMINATE OXYAENIDS PROVIVERRINAE

Fragmentary remains assigned as "creodont indet." (presumably Characteristics: Ml-3 tritubercular; ml-3 tuberculo-sec-
oxyaenids because of their Paleocene occurrence) have been re- torial; ml-3 usually with distinct metaconids and moderate shearing
ported from localities NPIC, ?NP2, ?NP3A. A questionable specializations; premolars elongate and often laterally compressed;
oxyaenid has been reported from locality CP34A. feet and limbs may be specialized for climbing or cursoriality.

OXYAENIDAE?: MACHAEROIDINAE Prototomus Cope, 1874

Type species: Prototomus viverrinus Cope, 187 4.
Characteristics: Ml and m2 carnassials; M3/m3 absent; Cl en- Type specimen: USNM 1022.
larged, saberlike; cl reduced; mandible with laterally flaring ventral Characteristics: Dental formula I3/3, C Ill, P4/4, M3/3; M 1-2
flange. with paracone and metacone moderately well separated,
postmetacristae expanded into shearing blades; M3 trian-
Machaeroides Matthew, 1909
gular, transverse, and unreduced with distinct metacone
Type species: Machaeroides eothen Matthew, 1909.
and strong parastyle; premolars elongate, laterally com-
Type specimen: AMNH 12644.
pressed and moderately high; Pl double rooted; pl sin-
Characteristics: Dental formula 13/2-3, Cl/I, P4/4, M2/2;
gle rooted; ml-3 with moderate, relatively narrow talonid
Cl enlarged, saberlike; i3 large; i2 small; ii small or ab-
basins and well-developed metaconids; m2-3 subequal in
sent?; cl small; moderate cl-pl diastema; pl-2 double
size, ml smaller; ml-2 talonids nearly as long as trigo-
rooted; p2-4 elongate and transversely compressed; ml-2
nids; mandibles long and shallow with anterior premolars
with protoconid-paraconid elongate and parallel with jaw,
separated by diastemata; zygomatic arches slender; tym-
metaconid reduced; m2 talonid very small; mandible thin
panic bullae unossified; trunk slender with long, heavy tail;
and moderately deep, flaring laterally at canine; symphysis
astragalar trochlea slightly grooved; small astragalar fora-
deep and vertical (Dawson et al., 1986).
men present; calcaneum with distinct fibular articular facet
Average length of m I: 9 .4 mm.
(Robinson, 1960; Delson, 1971; Guthrie, 1971; Bown,
Included species: M. eothen (known from localities CP34C,
1979; Davidson, 1987; Gingerich and Deutsch, 1989).
CP34D); M. simpsoni (localities CP27D, CP27E).
Average length of ml: 4.4-7.5 mm.
Included species: P. viverrinus (known from localities SB24,
Apataelurus Scott, 1938 CP20B, CP20C, CP20D, CP25G, CP63); P. secundarius
Type species: Apataelurus kayi Scott, 1938. (localities SB24, CP20A, CP20B, CP20C, CP25C, CP25D,
Type specimen: CM 11920. CP25E, CP27D); P. deimos (localities CP19, CP20A); P.
Characteristics: Lower dental formula i ?2, c 1, p4, m2; c 1 very phobos (localities CPI 9, CP20A, [CP27 A]); P. martis (lo-
small; long cl-pl diastema; pl single rooted; p4 slender calities CP19, CP20A, [CP64B]); P. robustus (localities
and transversely compressed with large anterior basal cusp; CP19, CP20A).
ml reduced; m2 large with strong shearing blade, vestigial Prototomus sp. is also known from localities GC21II,
talonid, and no metaconid; mandible with low condyle, CP25A, NP49.
deep anteriorly with ventral flange.
Average length of ml: 12.0 mm. Gazinocyon Polly, 1996
Included species: A. kayi only, known from locality CP6A Type species: Sinopa vulpecula (Matthew, 1915).
only. Type specimen: AMNH 15606.
Characteristics: Ml-2 with relatively long metastylar blades;
Ml-2 with anteriorly placed protocones and reduced para-
HYAENODONTIDAE
styles; lower molars with anteriorly placed metaconids and
Defining features: elongate, narrow skulls with narrow basicrania relatively closed molar trigonids; lower molar metaconids
and high, narrow occiput; frontals concave between orbital regions; large, especially on m3 (Matthew, 1915; Gingerich and
tympanic bullae unossified or ossified; M3 present in most taxa; m3 Deutsch, 1989; Polly, 1996).
always present; M2 and m3 camassials; lower molars range from Average length of ml: 6.25 mm.
tuberculo-sectorial to completely sectorial (see Figure 5.4A); manus Included species: G. vulpeculus only (known from localities
and pes mesaxonic, ranging from plantigrade to digitigrade; fibula CP20D, CP25G, CP25H, CP27B).
 Gregg F. Gunnell 97

Aifia Van Valen, 1965 Tritemnodon Matthew, 1906

Type species: Arfia opisthotoma (Matthew, 1901). Type species: Tritemnodon agilis (Marsh, 1872).
Type specimen: AMNH 99. Type specimen: YPM 11877.
Characteristics: P4 metacrista reduced; M 1-2 with well-sepa- Characteristics: Ml-3 with paracone and metacone closely
rated paracone and metacone and arcuate postmetacristae; appressed, protocone small; ml-3 with reduced metaconid
p3 moderately wide and high; ml-2 unspecialized; m3 and small, trenchant talonids; premolars laterally com-
metaconid moderately reduced, paraconid shifted anteri- pressed; skull slender and elongate (Cope, 1881, 1882;
orly; m3 enlarged; talonids increase in relative size from Robinson, 1966; Guthrie, 1967a, 1971; Krishtalka and
ml to m3 and are relatively broad; diastemata slight or Stucky, 1984; Gingerich and Deutsch, 1989; Gunnell and
absent; enamel crenulated (Delson, 1971; Bown, 1979; Bartels, 1994).
Kihm, 1984; Gingerich, 1989; Gingerich and Deutsch, Average length of ml: 8.5-10.0 mm.
1989). Included species: T. agilis (known from locality CP34C); ? T.
Average length of ml: 5.9-8.5 mm. strenuus (localities SB22C, SB24, [CP5A], CPl9, CP20A,
Included species: A. opisthotoma (known from localities CP20B,CP20C,CP20D,CP25D,CP 25G,CP27B,CP27C,
CPI 9, CP20A, [CP64B ]); A. shoshoniensis (localities CP27D, CP27E, CP64B).
CPl9, CP20A);A. zele (localities CP19, CP25B);A.junnei Specimens questionably assigned to Tritemnodon sp. are
(locality CPI 9). also known from localities CP25H, CP34A.
Arfia sp. is also known from locality CP63.
Comments: As with Sinopa, the taxonomic problems involv-
ing Tritemnodon will require a reassessment of all relevant
Sinopa Leidy, 1871 specimens (see chapter appendix).
Type species: Sinopa rapax Leidy, 1871.
Type specimen: USNM specimen missing (figured in Leidy,
1873). Proviverroides Bown, 1982
Characteristics: Dental formula I3/3, Cl/!, P4/4, M3/3; dif- Type species: Proviverroides piercei Bown, 1982.
fers from Prototomus (of which Sinopa is otherwise very Type specimen: USGS 1984.
similar) in general by being larger; in having Ml-2 with Characteristics: Ml-3 paracones well separated from meta-
paracone and metacone more completely separated; in hav- cones; p3-4 paraconids large, talonids developed; p4 broad
ing M3 with a reduced metacone; in having lower molar and robust; ml-3 robust with well-developed talonid
talonids longer than trigonids; in having m3 the same size basins, trigonids short and broad.
or sm,aller than m2; and in having a vestigial to absent astra- Average length of ml: 8.9 mm.
galar foramen (Cope, 1872; Scott, 1892; Matthew, 1901, Included species: P. piercei only, known from locality
1909; Thorpe, 1923; Kay, 1957; McGrew and Sullivan, CP31E only.
1970; West, 1973; Kihm, 1984; Krishtalka and Stucky, Proviverroides sp. is also questionably reported from
1984; Emry, 1990; Gunnell and Bartels, 1994). locality CP5A.
Average length ofml: 6.7-10.0 mm.
Included North American species: S. rapax (known from Pyrocyon Gingerich and Deutsch, 1989
localities CP34A, CP34C, CP34D, [CP38B]); S. major Type species: Pyrocyon dioctetus Gingerich and Deutsch,
(localities SB43A, CP34C, CP34D); S. grangeri (local- 1989.
ities CP5A, CP34C); S. minor (localities NB14, CP5A, Type specimen: UM 94757.
CP34C); S. pungens (localities CP34C, CP34D). Characteristics: pl double rooted; p3 lower crowned than
Specimens questionably assigned to Sinopa sp. are also p2; ml-3 with anteriorly placed paraconids and reduced
known from localities SB22C, CP34B. metaconids; trigonid more open than Prototomus; premo-
lars crowded anteroposteriorly; enamel lightly crenulated.
Comments: The systematic problems presented by Sinopa Average length of ml: 6.3 mm.
are enormous and will not be solved here (see chapter ap- Included species: P. dioctetus only, known from locality CPI 9
pendix). Of the eight genera of proviverrines recognized only.
in this chapter, only three (Proviverroides, Pyrocyon, and
Acarictis) have not been referred to Sinopa in the past. The Acarictis Gingerich and Deutsch, 1989
species of Sinopa recognized here include at least nine pre- Type species: Acarictis ryani Gingerich and Deutsch, 1989.
viously named species and six genera. The difficult process Type specimen: UM 79081.
of sorting out these taxonomic problems will require a re- Characteristics: Small size; primitive lower molar trigonids
assessment of all early and middle-Eocene hyaenodontids, with well-developed paraconid and metaconid equidistant
not only from North America, but from Europe, Asia, and from protoconid; P3 relatively long and narrow with dis-
Africa as well. tinct buccal cusp and low lingual cusp.
 Creodonta 98

Average length of ml: 3.9 mm. CP42A,CP68C,CP83A,CP83B,CP83C,CP98C,CP99A,

Included species: A. ryani only, known from locality CP19 NPlOA, NPlOB, NP24C, NP27D, [NP32B], NP50B);
only. H. venturae (locality CC9BB); H. microdon (localities
CP39E, NP24C, NP27C); H. crucians (localities SB44B,
SB44D, CP39C, CP39E, CP39G, CP39H, CP39IIC,
SUBFAMILY UNCERTAIN
CP40A,CP40B,CP41A,CP41B,CP68C,CP83C,CP98C,
Galecyon Gingerich and Deutsch, 1989 CP99A, NPlOB, NP24C, NP27C, NP27D, NP50B); H.
Type species: Galecyon mordax (Matthew, 1915). brevirostris (localities CP84B, CP84C, CP85C);
Type specimen: AMNH 16157. H. mustelinus (localities CP39C, CP68C, NPlOB);
Characteristics: Robust canines; p 1 single rooted; ml-2 with H. raineyi (locality SB44D).
short, broad talonids; m3 reduced; short, deep dentary (Van Hyaenodon sp. is also known from localities CC9B,
Valen, 1965, 1966; Delson, 1971). CC12, ?SB26B, CP45, CP84A, NP25B, NP25C.
Average length of ml: 6.5 mm.
Included species: G. mordax only (known from localities Hemipsalodon Cope, 1885
CP19, CP25B). Type species: Hemipsalodon grandis Cope, 1885.
Type specimen: NMC 6497.
Comments: Galecyon differs from proviverrines in having a Characteristics: Dental formula 13/3, Cl/l, P4/4, M3/3; very
single-rooted pl and a reduced m3. These characteristics large; lacks well-developed shearing crests on Ml-2 and
are similar to those found in Prolimnocyon. It differs from m2-3; M2 with paracone and metacone well separated;
Prolimnocyon by having a short, deep mandible and short M3 large; lingual cingula present on molars (Schlaikjer,
lower molars that have broad talonid basins. It is possi- 1935a, 1935b; Russell, 1938; Mellett, 1969; Lucas et al.,
ble that Galecyon may be more closely related to liinno- 1981; Lucas, 1982; Gustafson, 1986).
cyonines than to any other hyaenodontid group. Further Average length of ml: 21.0 mm.
evidence, especially from the upper dentition, is needed to Included species: H. grandis (known from localities CP39C,
determine the phyletic relationship of Galecyon. CP42A, NPlOB, PN5B); H. viejaensis (locality SB44B).
Hemipsalodon sp. is also known from localities CP39B,
HYAENODONTINAE ?SB25B.

Characteristics: Molars sectorial, metaconids absent; carnassial spe-

cialization well developed; M3 transverse or absent; premolars short INDETERMINATE HYAENODONTIDS
and high; body compact; limbs and feet often cursorial.
Ischnognathus Stovall, 1948
Type species: lschnognathus savagei Stovall, 1948.
Hyaenodon Laizer and Parieu, 1838 Type specimen: OUSM 32-4-S 1.
Type species: Hyaenodon leptorhynchus Laizer and Parieu, Characteristics: Mandible thin and high at symphysis; sym-
1838. physis compressed between ventral and lingual margins;
Type specimen: Lyon, unnumbered specimen in the collec- symphysis deeply grooved; mandibles flanged; cl large;
tion of the Departement des Sciences de la Terre (figured incisors small; pl long and procumbent; no apparent di-
by Lange-Badre, 1979). astemata (Van Valen, 1966; Gustafson, 1986).
Characteristics: Dental formula 13/3, Cl/l, P4/4, M2/3; M3 Average length of ml: No teeth known; similar in size to
absent; carnassial shearing extremely well developed, with Hyaenodon horridus.
shearing (in adults) occurring in Ml-2 and m2-3; brain Included species:/. savagei only, known from locality SB44B
small; feet digitigrade (Leidy, 1853; Scott, 1887, 1895; Os- only.
born and Wortman, 1894; Douglass, 1901; Thorpe, 1922;
Peterson, 1931; Stock, 1933; Scott and Jepsen, 1936; Clark, Comments: This taxon is represented by a single, edentulous
1937; Galbreath, 1953; Clark, Beerbower, and Kietzke, jaw fragment. It may represent a hyaenodontid, but given
1967; Macdonald, 1970; Golz andLillegraven, 1977; Mel- the nature of the specimen, this is far from certain.
lett, 1977; Gustafson, 1986; Demee, 1988).
Average length of ml: 7.3-15.0 mm. Questionable hyaenodontids are known also from localities CC9A
Included North American species: H. vetus (known from and SB52.
localities CC9BB, [SB43C], CP7C, CP39B, CP39E); H.
megaloides (localities CP39C, CP39G); H. montanus (lo-
LIMNOCYONIDAE
calities SB44B, CP39C, CP39F, CP39G, [CP83A],
[CP83C], CP98B, NP23C, NP24C, NP24D); H. horridus Characteristics: No protocone on P2-3; M3/m3 reduced or ab-
(localities SB27II, SB44C, CP39C, CP39E, CP39G, sent; small- to medium-sized; elongate rostrum; unreduced teeth;
CP39H, CP39IIC, CP40A, CP40B, CP41A, CP41B, tuberculo-sectorial molars (see Figure 5.4).
 Gregg F. Gunnell 99

Limnocyon Marsh, 1872 P. elisabethae (localities CP20D, CP25G, CP27B); P. iudei

Type species: Limnocyon verus Marsh, 1872. (locality CP27B).
Type specimen: YPM Marsh Collection (figured by Wort- Prolimnocyon sp. is also known from localities CC50,
man, 1902). ?CP26C, NP14.
Characteristics: Dental formula I2-3/3, Cl/1, P4/4, M2/2;
M3/m3 absent; Pl/pl double rooted; P3 double rooted,
Oxyaenodon Wortman, 1899
lacking protocone; Ml protocone slightly reduced; M2
Type species: Oxyaenodon dysodus Wortman, 1899.
transverse with vestigial metacone and elongate parastyle;
Type specimen: AMNH 2515.
ml smaller than m2; m2 trigonid longer than wide; mandi-
Characteristics: Dental formula 13/3, Cl/l, P4/4, M2/2; Il-3
ble deep with robust, gently rounded symphysis; skull
small and equal sized; premolars reduced, crowded, and
broad with strong sagittal and occipital crests; limbs mod-
simple; Pl/pl double rooted; P3 double rooted, with small
erately short and robust; feet plantigrade (Osborn, 1895;
protocone shelf; Ml broader than long; M2 transverse and
Wortman, 1902; Matthew, 1909; Peterson, 1919; Kay,
reduced; ml-2 with reduced metaconid; mandible thick
1957; Gunnell and Bartels, 1994).
and shallow; skull short and broad (Hay, 1902; Peterson,
Average length of ml: 8.5-12.3 mm.
1919; Kay, 1957; Van Valen, 1966).
Included species: L. verus (known from localities [CP5A],
Average length of ml: 11.0 mm.
CP34C, CP34D); L. potens (localities CP6A, CP38C).
Included species: 0. dysodus only (known from localities
Limnocyon sp. is also known from localities CP27E,
CP6A, CP6B).
?CP29D, CP34A.
Comments: The history of Oxyaenodon is particularly con-
Thinocyon Marsh, 1872 fusing, resulting from three different species having been
Type species: Thinocyon velox Marsh, 1872. proposed for two specimens. Matthew ( 1899) first used the
Type specimen: YPM 11797. name Oxyaenodon dysodus (in a footnote Matthew noted
Characteristics: Dental formula I?/3, Cl/l, P4/4, M2/2; shal- that the name was unpublished). Matthew referred an eden-
low mandible; small sagittal and occipital crests; elon- tulous jaw (AMNH 1893) figured by Osborn (1895) to this
gate and slender limbs; otherwise similar to Limnocyon species. Two months later, Wortman (1899) formally pro-
(Matthew, 1909; Thorpe, 1923; McGrew, 1959; McKenna, posed Oxyaenodon dysodus based on a nearly complete
Robinson, and Taylor, 1962; West, 1973; Gunnell and Bar- skull, providing a figure and a diagnosis.
tels, 1994). Hay (1902) felt that Matthew's priority was valid making
Average length of ml: 5.0-6.0 mm. AMNH 1893 the type of 0. dysodus. Hay renamed Wort-
Included species: T. velox (known from localities CP34B, man's species 0. dysclerus noting that it was distinct from
CP34C); T. cledensis (locality CP38B); T. medius (local- AMNH 1893. Van Valen (1966) questioned Matthew's pri-
ity CP34D); T. mustelinus (locality CP34C); T. comptus ority and instead credited Wortman with naming Oxyaen-
(locality CP34D). odon dysodus, making the skull and jaws the type of that
Thinocyon sp. is also known from locality CP34A. species and Hay's 0. dysclerus a junior synonym. Van
Valen then proposed the name ?Oxyaenodon wortmani for
Prolimnocyon Matthew, 1915 AMNH 1893.
Type species: Prolimnocyon atavus Matthew, 1915. Clearly the problem lies in the identity of AMNH 1893.
Type specimen: AMNH 16816. I believe it is best interpreted as a specimen of Limnocyon
Characteristics: Dental formula I?/3, Cl/l, P?-4/4, M3/3; P3 (both Peterson, 1919, and Van Valen, 1966, noted this pos-
triple rooted; Ml wide transversely, protocone unreduced; sibility) and refer it to Limnocyon sp. nr. L. potens. The
M2 reduced with distinct metacone and moderately short dentition contains four premolars and two molars, it has a
parastyle; M3/m3 small; pl single or double rooted; ml double-rooted p 1, and it has a rounded, stout symphysis, all
slightly smaller or subequal to m2; m2 trigonid as wide typical of Limnocyon. The premolar through molar length
as long; m3 single or double rooted; mandible long and of Limnocyon potens is 53 mm (Matthew, 1909), compar-
shallow; skull narrow with long rostrum (Kelley and Wood, ing favorably with AMNH 1893 (54 mm). AMNH 1893
1954; Guthrie, 1967a, 1967b; Delson, 1971; Bown, 1979; is from the "Telmatotherium comutum" level in the Uinta
Macintyre and Guthrie, 1979; Gingerich, 1989; Gingerich Basin. The type of Limnocyon potens Matthew (1909)
and Deutsch, 1989). comes from the "upper Washakie beds" in the Washakie
Average length of ml: 4.4-7.0 mm. Basin. These sequences are likely contemporaneous (Black
Included species: P. atavus (known from localities SB24, and Dawson, 1966a; Krishtalka et al., 1987), supporting
CP19, CP20A, CP20B, CP20C, [CP27A],CP63, [CP64A], the inclusion of AMNH 1893 in Limnocyon sp. nr. potens.
[CP64B], [CP64C]); P. antiquus (localities CP20D, Recognizing AMNH 1893 as Limnocyon leaves Oxyaen-
[CP25H], CP27C, CP27D, CP27E); P. haematus (local- odon dysodus Wortman as the only species of the
ities CP19, CP20A, CP25B); P. eerius (locality CP19); genus.
 Creodonta 100

BIOLOGY AND EVOLUTIONARY PATTERNS may have been scansorial (Rose and Gebo, 1992; Gebo and Rose,
1993). Van Valkenburgh (1987) suggested that frequent digging
Paleobiological considerations concerning Creodonta can be use- may have been a characteristic behavior of Hyaenodon. Janis and
fully divided into dental or feeding adaptations and postcranial or Wilhelm (1993) have shown that Hyaenodon differed from extant
locomotor adaptations. pursuit predators based on multivariate analysis of limb propor-
Most early members of the hyaenodontid and oxyaenid lineages tion, and instead was more similar to living ambush or generalized
had tuberculo-sectorial lower molars and tribosphenic upper mo- predators.
lars with moderate development of shearing crests. Smaller genera Other behavioral patterns of creodonts are subject to greater spec-
such as Prolimnocyon, Tritemnodon, Thinocyon, and Dipsalidic- ulation. Some evidence suggests that Hyaenodon was moderately
tis were probably generalized carnivores that fed on birds, small sexually dimorphic in canine size (Mellett, 1977). Mellett (1977)
mammals, eggs, and perhaps some insects as well. They may have believed that, although Hyaenodon exhibited canine dimorphism, it
been similar to living viverrids (Denison, 1938). Larger genera, such did not live in packs like many living dimorphic carnivorans, but was
as Oxyaena, Aifta, Prototomus, and Limnocyon, had heavier jaws more solitary. Mellett cited taphonomic evidence from Oligocene
and often possessed teeth with better developed shearing capabili- sediments where large carnivoran coprolite concentrations are found
ties manifest in enlarged postmetacristae and expanded prevallids. near many mammal skeletons. Mellett noted that defecating on food
These forms were adapted for meat slicing and shearing and prob- is a method that many solitary carnivores use to mark food concen-
ably concentrated on small- to medium-sized mammals, birds, and trations. Pack carnivorans tend to devour food shortly after a kill.
reptiles for the bulk of their diet. They may have been similar to Oxyaenids underwent a moderate radiation in the late Paleocene
foxes, wolves, and medium-sized felids. Palaeonictines were char- (Clarkforkian) of North America. They survived into the early late
acterized by much heavier jaws and blunt cusped teeth. Shearing Eocene of North America, but their diversity was reduced (Matthew,
adaptations were not well developed in palaeonictines. These forms 1909, 1915; Denison, 1938; Gunnell and Gingerich, 1991). Oxyae-
appear to have been adapted for heavy bone crushing. nids were widely distributed geographically in western North
Later species of oxyaenids and hyaenodontids developed the most America (see Figure 4.2).
exaggerated dental shearing systems of the order. Both Patriofelis Hyaenodontids remain unknown from sediments earlier than the
and Hyaenodon had sectorial upper and lower molars. Both were Wasatchian in North America. Specimens ofhyaenodontids reported
large predators with highly developed slicing teeth and were prob- from earlier in North America (Mc Kenna, 1980; Rigby, 1980) proba-
ably similar to large felids. Patriofelis differed from Hyaenodon in bly represent palaeoryctid or pantolestid insectivorans. Hyaenodon-
having heavier, blunter premolars that may have served as bone- tids may be present in the Adrar Mgorn local fauna from the Pa-
crushing teeth. leocene of North Africa (Cappetta et al., 1978), and it has been
Other hyaenodontids show adaptations similar to those of Pa- suggested that hyaenodontids dispersed from Africa near the Pa-
triofelis and Hyaenodon. Oxyaenodon had a heavy jaw with well- leocene/Eocene boundary (Gheerbrant, 1990). When hyaenodon-
developed shearing teeth and may have been similar to the living tids appeared in the Wasatchian of North America, they were al-
Tasmanian Devil, Sarcophilus (Denison, 1938). Machaeroides and ready widely dispersed geographically and were represented by a
Apataelurus had well-developed shearing teeth and long slicing ca- broad, diverse radiation through the middle Eocene. Late Eocene and
nines (saberteeth) and may have been similar to small- or medium- Oligocene hyaenodontid diversity was lower (Black and Dawson,
sized species of extant Felis (Denison, 1938; Dawson et al., 1986). 1966a, 1966b; Clark, Beerbower, and Kietzke, 1967; Lillegraven,
Creodonts can be divided into two broad locomotor categories: 1970; Emry, 1973, 1975; Love, McKenna, and Dawson, 1976;
ambulatory and cursorial. Ambulatory forms were generalized ter- Mellett, 1977; Storer, 1984; Prothero, l 985a, l 985b; Emry, Russell,
restrial quadrupeds characterized by short, spreading, plantigrade and Bjork, 1987). Although hyaenodontids disappeared shortly be-
feet, short, heavy limbs, and robust bodies. These forms had rel- fore the end of the Oligocene in North America and Europe, they
atively large supinator and deltoid crests on the humerus, limited survived through the late Miocene in Africa and Asia. Geograph-
pronation and supination capabilities, and smooth astragalar troch- ically, hyaenodontids were present from a wider range in North
lea. Ambulatory genera include Oxyaena, Patriofelis, Palaeonictis, America than were oxyaenids (see Figure 4.2).
Ambloctonus, and Limnocyon. Thinocyon and Dipsalidictis show Limnocyonids first appeared in the earliest Wasatchian (Gin-
evidence of more mobility in the ankle and may have been capable gerich, 1989, although McKenna, 1980, lists a specimen of cf.
of climbing trees (Gunnell, 1988; Gebo and Rose, 1993). Prolimnocyon atavus from the Clarkforkian of the Togwotee Pass
Cursorial creodonts can be characterized as follows: elongate area) and maintained low, but relatively constant, diversities through
limbs; narrow scapulae; small supinator and deltoid crests on the the early and middle Eocene. No limnocyonids are known from
humerus; very limited supination and pronation capabilities; a broad sediments later than the early Uintan. Geographically, limnocy-
carpus; an expanded ilium; a slightly grooved astragalar trochlea; onids were mostly restricted to the Rocky Mountain corridor, from
and compressed, digitigrade feet (Matthew, 1906). Cursorial forms Wyoming, Utah, Colorado, and New Mexico (see Figure 4.2). Pro-
may have included Aifta, Prototomus, and Hyaenodon (Matthew, limnocyon has been reported from Ellesmere Island (West and Daw-
1906; Mellett, 1977; Gingerich and Deutsch, 1989). Aifta appears son, 1978) and from Baja California (Novacek et al., 1987).
to have had the capacity for hindfoot reversal suggesting that it was Creodonts were last represented in North American by Hyaenodon
capable of climbing (Gingerich and Gunnell, 1989). Prolimnocyon brevirostris. Hyaenodon was a large, cursorial, active predator. There
 Gregg F. Gunnell 101

w
zw L L.BI.
1.8
2.4
E.BI
0
0 E Lot.Hp,
4.5 -
:::i L.Hp.
5.2
a. L.E.Hp.
5,0
7.0
E.E.Hp.
L 8.8
L.CI.
9.5

E.CI.
w 11.0 -
z
w
L.L.Ba.
12.5

0 E.L.Ba.
M
0 14.0
:§ E.Ba.

L.Hm.
5.8 -
17.5
E.Hm.

E L.L.Ar.
18.8 -
19.2

E.L.Ar.
23.0

w L.E.Ar.
zw
L
0
0
'ZT.1 - - - - - - - - - - - - - - - - - - - - - -
CJ E.E.Ar.
:::i 29.4
0
Wt.
E 31.9 -
Or.
r:: ~
LCh.
33.4
_r:: .g-;:
34.5
_.g
~f r::0~:
L M.Ch.

E.Ch.
35.5 - 0
-ii .e.
II)

Du
37.1

39.6 -
~ e
::c: ~ ~
ilJ -r-
T
I r::
.g

~
L.Un. ?I ?I
_'.L
w
-2[
41.3 -II)
z
w
0
E.Un.

45.9 _Q
....~ II)
0 t ~ ~
w L.Br.
111 QI
M .~
-'fi .s-
g. - ·s·~1
fr-
46.7

l!
~ oq:~
-
111
M.Br. ,0 .~ ~ r:: r:: ~
49.0 s ~
- - - .s: -~ 0 o-

-ri
E.Br.
~ u £;- r::
21?-

~I-
(/)
50.4
.... Q. -g g :§

--l!r . e .... o
r::
L.Wa.
Q.
2
~
~ .g r::
r::
0 ::t_t:_~
53,5 111 -E r:: -o
0 .~ -:::..-
E M.wa. ii .~ .s .g0 .s e ::ii,. -
u .2 .!!!
u
-.S0 .! -o ..,; -111
~
54.2 - -r:: u
0
0
-Cll .Q
-9 ibII) -Q: iEoq: ~ -t
!;;. ~
oq: -
CJ
e
w
zw =a:.
-
.!!
-~-:_'ic:(
-
·--1-=
c:t.
..9
- -
?C
--- --- --- ---
L
0
0 58.6
w
....!
c( To.3
.... , ,Oxyaeninae IIPalaeonictinae
59.2

61.3
I~
Proviverrinae Hyaenodontinae
a. To.2
62.5
Tytthaeninae Machaeroidinae
To.1
E Pu.3
63.8
Pu.2
54.0 - _.J
54.5
Pu.1
Pu.o
66.0
55.1 OXYAENIDAE ? HYAENODONTIDAE ?LIMNOCYONIDAE
Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerlan, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti.= Tiffanian, To.= Torrejonian, Pu. = Puercan.
Figure 5.8. Temporal ranges of North American creodont genera.
 Creodonta 102

were some relatively large camivorans present in the late Eocene- Geological Explorations and Surveys West of the JOOth Meridian, Lt.
early Oligocene, such as the nimravids Hoplophoenus and Dinic- George M. Wheeler in charge. Washington, D.C.: U.S. Government
tis, but these were saber-toothed forms with less heavy jaws than Printing Office, 4, 1-370.
(1880). The Bad Lands of the Wind River and their fauna. American
Hyaenodon, indicating different adaptations. Hyaenodon paralleled
Naturalist, October, 745-8.
the radiation of true carnivorans in its large size and cursorial adap- ( 1881 ). On the Vertebrata of the Wind River Eocene beds of Wyoming.
tations. Canids such as Enhydrocyon, nimravids such as Nimravus, Bulletin of the United States Geological Survey of the Territories, 6,
and hemicyonine ursids may have eventually contributed to the dis- 183-202.
appearance of North American hyaenodontids by the late Oligocene. (1882). The fauna of the Wasatch beds of the Basin of the Big Horn River.
Proceedings of the American Philosophical Society, 20, 139-97.
(1884a). The Vertebrata of the tertiary formations of the west, Book 1:
Report of the Geological Survey of the Territories, 3, 1-1009.
ACKNOWLEDGMENTS (1884b). The Creodonta. American Naturalist, 18, 255-67, 344-53,
478-85.
I would like to thank P. D. Gingerich, P. D. Polly, and (1885). The White River beds of Swift Current River, Northwest Terri-
J. G. M. Thewissen for their comments on the manuscript. C. M. tory. American Naturalist, 19, 163.
Davidson, J. R. (1987). Geology and mammalian paleontology of the Wind
Janis, B. Augusta, and two anonymous reviewers greatly aided in River Formation, Laramie Basin, southeastern Wyoming. Contribu-
improving the final product. Thewissen provided aid in prepara- tions to Geology, University of Wyoming, 25, 103-32.
tion of cladograms. Shane Davidson and Bonnie Miljour aided in Dawson, M. R., Stucky, R. K., Krishtalka, L., & Black, C. C. (1986).
preparation of figures. Machaeroides simpsoni, new species, oldest known sabertooth cre-
odont (Mammalia), of the Lost Cabin Eocene. Contributions to Ge-
ology, University of Wyoming, Special Paper, 3, 177-82.
Delson, E. ( 1971 ). Fossil mammals of the early Wasatchian Powder River Lo-
REFERENCES cal Fauna, Eocene of Northeast Wyoming. Bulletin of the American
Museum of Natural History, 146, 307-64.
Adams, G. I. (1896). The extinct Felidae of North America. American Jour- Demere, T. A. (1988). Early Arikareean (late Oligocene) vertebrate fossils
nal of Science, I, 419-44. and biostratigraphic correlation of the Otay Formation at Eastlake,
Black, C. C., & Dawson, M. R. (l 966a). A review of late Eocene mam- San Diego County, California. In Paleogene Stratigraphy, West Coast
malian faunas from North America. American Journal of Science, of North America, eds. M. V. Filewicz & R. L. Squires, Pacific Sec-
264, 321-49. tion, SEPM Paleogene Symposium, 58, 35-43.
( l 966b ). Paleontology and geology of the Bad water Creek area, central Denison, R.H. (1938). The broad-skulled Pseudocreodi. Annals of the New
Wyoming, Part I: History of field work and geological setting. Annals York Academy of Sciences, 37, 163-256.
of the Carnegie Museum, 38, 297-307. Douglass, E. ( 1901 ). Fossil Mammalia of the White River beds of Montana.
Blainville, H. M. D. de (1842). Osteographie ou description iconographique Transactions of the American Philosophical Society, 20, 237-78.
comparee de squelette et du systeme dentaires des Mammireres Emry, R. J. (1973). Stratigraphy and preliminary biostratigraphy of the
recents et fossiles, 2: Secundates. Des Viverras, Paris: J.B. Bailliere Flagstaff Rim area, Natrona County, Wyoming. Smithsonian Con-
et Fils, pp. 1-100. tributions to Paleobiology, 18, 1-43.
Bown, T. M. (1979). Geology and mammalian paleontology of the (1975). Revised tertiary stratigraphy and paleontology of the western
Sand Creek Facies, Lower Willwood Formation (Lower Eocene), Beaver Divide, Fremont County, Wyoming. Smithsonian Contribu-
Washakie County, Wyoming. Geological Survey of Wyoming Mem- tions to Paleobiology, 25, 1-20.
oir, 2, 1-151. (1990). Mammals of the Bridgerian (middle Eocene) Elderberry Canyon
(1982). Geology, paleontology, and correlation of Eocene volcaniclastic Local Fauna of eastern Nevada. In Dawn of the Age ofMammals, eds.
rocks, southeast Absaroka Range, Hot Springs County, Wyoming. T. M. Bown & K. D. Rose, Geological Society of America, Special
United States Geological Survey Professional Paper, 1201-A, 1-75. Paper, 243, 187-210.
Cappetta, H., Jaeger, J.-J., Sabatier, M., Sige, B., Sudre, J., & Vianey- Emry, R. J., Russell, L. R., & Bjork, P.R. (1987). The Chadronian, Orellan,
Liuad, M. (1978). Decouverte dans le Paleocene du Maroc des and Whitneyan North American Land Mammal Ages. In Cenozoic
plus anciens Mammireres eutheriens d' Afrique. Geobios, 11, Mammals of North America, ed. M. 0. Woodburne, pp. 118-52.
257-63. Berkeley: University of California Press.
Clark, J. (1937). The stratigraphy and paleontology of the Chadron Forma- Flynn, J. J., Neff, N. A., & Tedford, R. H. (1988). Phylogeny of the
tion in the Big Badlands of South Dakota. Annals of the Carnegie Carnivora. In The Phylogeny and Classification of the Tetrapods,
Museum, 25, 261-350. Vol. 2: Mammals, ed. M. J. Benton, pp. 73-115. Oxford: Clarendon
Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedi- Press.
mentation, stratigraphy, paleoecology and paleoclimatology in the Fox, R. C. ( 1990). The succession of Paleocene mammals in western Canada.
Big Badlands of South Dakota. Fieldiana: Geology Memoirs, 5, In Dawn of the Age of Mammals, eds. T. M. Bown & K. D. Rose,
5-158. Geological Society of America, Special Paper, 243, 51-70.
Cope, E. D. (1872). Second account of new Vertebrata from the Bridger Fox, R. C., & Youzwyshyn, G. P. (1994). New primitive carnivorans (Mam-
Eocene. Paleontological Bulletin, 2, 1-3. malia) from the Paleocene of western Canada, and their bearing on
(1874). Report upon vertebrate fossils discovered in New Mexico, with relationships of theorder. Journal of Vertebrate Paleontology, 14,
description of new species.Appendix FF3, Annual Report ofthe Chief 382-404.
of Engineers, 1874. Washington, D.C.: U.S. Government Printing Galbreath, E. C. (1953). A contribution to the Tertiary geology and paleontol-
Office, pp. 588-606. ogy of northeastern Colorado. University of Kansas Paleontological
(1875). On the supposed Carnivora of the Eocene of the Rocky Moun- Contributions, Vertebrata, 4, 1-120.
tains. Paleontological Bulletin, 20, 1-4. Gazin, C. L. ( 1946). Machaeroides eothen Matthew, the saber-tooth creodont
(1877). Report upon the extinct Vertebrata obtained in New Mexico by of the Bridger Eocene. Proceedings of the United States National
parties of the expedition of 1874. In Report of the Geographical and Museum, 96, 335-47.
 Gregg F. Gunnell 103

( 1952). The lower Eocene Knight Formation of western Wyoming and Kay, J. L. ( 1957). The Eocene vertebrates of the Uinta Basin, Utah. Inter-
its mammalian faunas. Smithsonian Miscellaneous Collections, 117, Mountain Association of Petroleum Geologists, Eighth Annual Field
1-82. Conference, 1957, 110-14.
( 1957). A skull of the Bridger Middle Eocene creodont Patriofelis ulta Kelley, D. R., & Wood, A. E. (1954). The Eocene mammals from the
Leidy. Smithsonian Miscellaneous Collections, 134, 1-20. Lysite Member, Wind River Formation of Wyoming. Journal of
( 1962). A further study of the lower Eocene mammalian faunas of south- Paleontology, 28, 337-66.
western Wyoming. Smithsonian Miscellaneous Collections, 144, Kihm, A. J. (1984). Early Eocene Mammalian Faunas of the Piceance Creek
1-98. Basin, Northwestern Colorado. Unpublished Ph.D. Thesis. Boulder:
Gebo, D. L., & Rose, K. D. (1993). Skeletal morphology and locomotor The University of Colorado.
adaptation in Prolimnocyon atavus, an early Eocene hyaenodontid Krishtalka, L., & Stucky, R. K. (1984). Middle Eocene marsupials (Mam-
creodont. Journal of Vertebrate Paleontology, 13, 125-44. malia) from northeastern Utah and the mammalian fauna from Pow-
Gheerbrant, E. (1990). On the early biogeographical history of the African der Wash. Annals of the Carnegie Museum, 53, 31-45.
placentals. Historical Biology, 4, 107-16. Krishtalka, L., Stucky, R. K., West, R. M., McKenna, M. C., Black, C. C.,
Gingerich, P. D. (1980). Tytthaena parrisi, oldest known oxyaenid (Mam- Bown, T. M., Dawson, M. R., Golz, D. J., Flynn, J. J., Lillegraven,
malia, Creodonta) from the later Paleocene of western North Amer- J. A., & Turnbull, W. D. (1987). Eocene (Wasatchian through Duch-
ica. Journal of Paleontology, 54, 570-6. esnean) biochronology of North America. In Cenozoic Mammals of
(1986). Early Eocene Cantius torresi - oldest primate of modem aspect North America, ed. M. 0. Woodbume, pp. 77-117. Berkeley: Uni-
from North America. Nature, 320, 319-21. versity of California Press.
(1989). New earliest Wasatchian mammalian fauna from the Eocene of Laizer, L., & Parieu, D. (1838). Description et datermination d'une milchoire
northwestern Wyoming: composition and diversity in a rarely sam- fossile appartenant aun mammifere jusqu 'a present inconnu, Hyaen-
pled high-floodplain assemblage. University of Michigan, Papers on odon leptorhynchus. Comptes Rendus des Seances de L'Academie
Paleontology, 28, 1-97. des Sciences, Paris, 7, 1-442.
Gingerich, P. D., & Deutsch, H. A. (1989). Systematics and evolution of early Lange-Badre, B. (1979). Les Creodontes (Mammalia) d'Europe occidentale
Eocene Hyaenodontidae (Mammalia, Creodonta) in the Clarks Fork de !'Eocene superieura !'Oligocene superieur. Memoires du Museum
Basin, Wyoming. Contributions from the Museum of Paleontology, National d'Histoire Naturelle, 42, 1-249.
University of Michigan, 27, 327-91. Leidy, J. (1853). Remarks on a collection of fossil Mammalia from Nebraska.
Gingerich, P. D., & Gunnell, G. F. (1989). Hind foot reversal in early Eocene Proceedings of the Academy of Natural Sciences, Philadelphia, 6,
Arfia (Mammalia, Creodonta). Journal of Vertebrate Paleontology, 392-4.
9 (suppl. to no. 3), p. 22A. (1870). Untitled: [Patriofelis ulta proposed during proceedings of the
Golz, D. J., & Lillegraven, J. A. (1977). Summary of known occurrences March 1 meeting of the Academy of Natural Sciences]. Proceedings
of terrestrial vertebrates from Eocene strata of southern California. of the Academy of Natural Sciences, Philadelphia, pp. 9-11.
Contributions to Geology, University of Wyoming, 15, 43-64. (1871). Remains of extinct mammals from Wyoming. Proceedings of
Gunnell, G. F. (1988). Locomotor adaptations in early Eocene Oxyaenidae the Academy of Natural Sciences, Philadelphia, pp. 113-16.
(Mammalia, Creodonta). Journal of Vertebrate Paleontology, 8 (1873). Contributions to the extinct vertebrate fauna of the western ter-
(suppl. to no. 3), p. 17A. ritories, Part I: Report of the United States Geological Survey of the
Gunnell, G. F., & Bartels, W. S. (1994 ). Early Bridgerian (middle Eocene) Territories, F. V. Hayden, Geologist in charge. Washington, D.C.:
vertebrate paleontology and paleoecology of the southern Green U.S. Government Printing Office, pp. 14-358.
River Basin, Wyoming. Contributions to Geology, University of Lillegraven, J. A. ( 1969). Latest Cretaceous mammals of upper part of Ed-
Wyoming, 30, 57-70. monton Formation of Alberta, Canada, and review of marsupial-
Gunnell, G. F., & Gingerich, P. D. (1991). Systematics and evolution oflate placental dichotomy in mammalian evolution. University of Kansas
Paleocene and early Eocene Oxyaenidae (Mammalia, Creodonta) in Paleontological Contributions, 50, 1-122.
the Clarks Fork Basin, Wyoming. Contributions from the Museum of (1970). Stratigraphy, structure, and vertebrate fossils of the Oligocene
Paleontology, University of Michigan, 28, 141-80. Brule Formation, Slim Buttes, northwestern South Dakota. Geolog-
Gustafson, E. P. ( 1986). Carnivorous mammals of the late Eocene and early ical Survey of America Bulletin, 81, 831-50.
Oligocene of Trans-Pecos Texas. Texas Memorial Museum Bulletin, Love, J. D., McKenna, M. C., & Dawson, M. R. (1976). Eocene, Oligocene,
33, 1-66. and Miocene rocks and vertebrate fossils at the Emerald Lake Local-
Guthrie, D. A. (1967a). The mammalian fauna of the Lysite Member, Wind ity, 3 miles south of Yellowstone National Park, Wyoming. United
River Formation (early Eocene) of Wyoming. Memoir, Southern States Geological Survey, Professional Paper, 932-A, Al-A28.
California Academy of Science, 5, 1-52. Lucas, S. G. (1982). Vertebrate paleontology, stratigraphy, and biostratig-
(l 967b ). Paeneprolimnocyon, a new genus of early Eocene Limnocyonid raphy of Eocene Galisteo Formation, north-central New Mexico.
(Mammalia, Deltatheridia). Journal of Paleontology, 41, 1285-7. New Mexico Bureau of Mines & Mineral Resources Circular, 186,
(1971 ). The mammalian fauna of the Lost Cabin Member, Wind River 7-34.
Formation (lower Eocene) of Wyoming. Annals of the Carnegie Mu- Lucas, S. G., Schoch, R. M., Manning, E., & Tsentas, C. (1981). The
seum, 43, 47-113. Eocene biostratigraphy of New Mexico. Bulletin, Geological Society
Hay, 0. P. (1902). Bibliography and catalog of the fossil Vertebrata of of America, 92, 951-67.
North America. Bulletin of the United States Geological Survey, 179, Macdonald, J. R. (1970). Review of the Miocene Wounded Knee faunas
9-868. of southwestern South Dakota. Bulletin of the Los Angeles County
Janis, C. M., & Wilhelm, P. B. (1993). Were there mammalian pursuit preda- Museum of Natural History, 8, 1-82.
tors in the Tertiary? Dances with wolf avatars. Journal ofMammalian Macintyre, G. T., & Guthrie, D. A. (1979). Paeneprolimnocyon Guthrie,
Evolution, 1, 103-25. 1967, a junior synonym of Oodectes Wortman, 1901. Journal of
Jepsen, G. L. (1930a). New vertebrate fossils from the lower Eocene of the Paleontology, 53, 1034-6.
Bighorn Basin, Wyoming. Proceedings of.the American Philosoph- Marsh, 0. C. (1872). Preliminary description of new Tertiary mammals,
ical Society, 69, 117-31. Parts I-IV. American Journal of Science and Arts, 4, 122-8, 202-24,
( 1930b). Stratigraphy and paleontology of the Paleocene of northwestern 504.
Park County, Wyoming. Proceedings of the American Philosophical (1899). Note on aBridger Eocene carnivore. American Journal ofScience
Society, 69, 463-528. and Arts, 7, 397.
 Creodonta 104

Matthew, W. D. ( 1899). A provisional classification of the freshwater Tertiary Prothero, D. R. (1985a). Chadronian (early Oligocene) magnetostratigra-
of the west. Bulletin of the American Museum of Natural History, phy of eastern Wyoming: implications for the age of the Eocene-
12, 19-75. Oligocene boundary. Journal of Geology, 93, 555-65.
(1901). Additional observations on the Creodonta. Bulletin of the Amer- (1985b). North American mammalian diversity and Eocene-Oligocene
ican Museum of Natural History, 14, 1-38. extinctions. Paleobiology, 11, 389-405.
(1903). The fauna of the Titanotherium beds at Pipestone Springs, Mon- Rigby, J. K. (1980). Swain Quarry of the Fort Union Formation, middle
tana. Bulletin of the American Museum of Natural History, 19, Paleocene (Torrejonian), Carbon County, Wyoming: geologic setting
197-226. and mammalian fauna. Evolutionary Monographs, 3, 1-178.
(1906). The osteology of Sinopa, a creodont mammal of the middle Robinson, P. (1960). Sinopa from the Cuchara Formation of Colorado. Pos-
Eocene. Proceedings of the U.S. National Museum, 30, 203-33. tilla, 44, 1-4.
(1909). The Carnivora and Insectivora of the Bridger Basin, middle Robinson, P. ( 1966). Fossil Mammalia of the Huerfano Formation, Eocene,
Eocene. Memoirs of the American Museum of Natural History, 9, of Colorado. Bulletin, Yale Peabody Museum, 21, 1-85.
289-567. Rose, K. D. (1981). The Clarkforkian Land-Mammal Age and mammalian
(1915). A revision of the lower Eocene Wasatch and Wind River faunas, fauna! composition across the Paleocene-Eocene boundary. Univer-
Part I: Order Ferae (Carnivora), suborder Creodonta. Bulletin of the sity of Michigan, Papers on Paleontology, 26, 1-197.
American Museum of Natural History, 34, 4-103. Rose, K. D., & Gebo, D L. (1992). Skeleton and locomotor adaptation in the
McGrew, P. 0. (1959). The geology and paleontology of the Elk Moun- early Eocene creodont Prolimnocyon atavus. Journal of Vertebrate
tain and Tabernacle Butte area, Wyoming. Bulletin of the American Paleontology, 12 (suppl. to no. 3), p. 49A.
Museum of Natural History, 117, 121-76. Russell, L. S. (1938). The skull of Hemipsalodon grandis, a giant Oligocene
McGrew, P. 0., & Sullivan, R. (1970). The stratigraphy and paleontology creodont. Transactions of the Royal Society of Canada, 32, 61-6.
of Bridger A. Contributions to Geology, University of Wyoming, 9, (1972). Tertiary mammals of Saskatchewan, Part II: The Oligocene
66-85. fauna, non-ungulate orders. Royal Ontario Museum, Life Sciences
McKenna, M. C. (1960). Fossil Mammalia from the early Wasatchian Contribution, 84, 1-97.
Four Mile Fauna, Eocene of northwestern Colorado. University of Schlaikjer, E. M. (1935a). Contributions to the stratigraphy and palaeontol-
California Publications in Geological Sciences, 37, 1-130. ogy of the Goshen Hole area, Wyoming, III: A new basal Oligocene
( 1980). Late Cretaceous and early Tertiary vertebrate paleontological formation. Bulletin of the Museum of Comparative Zoology, 76,
reconnaissance, Togwotee Pass area, northwestern Wyoming. In As- 71-93.
pects of Vertebrate History, Essays in Honor of Edwin Harris Col- (l 935b ). Contributions to the stratigraphy and palaeontology of the
bert, ed. L. L. Jacobs, pp. 321-43. Flagstaff: Museum of Northern Goshen Hole area, Wyoming, IV: New vertebrates and the stratigra-
Arizona Press. phy of the Oligocene and early Miocene. Bulletin of the Museum of
McKenna, M. C., Robinson, P., & Taylor, D. W. (1962). Notes on the Comparative Zoology, 76, 97-189.
Eocene Mammalia and Mollusca from Tabernacle Butte, Wyoming. Scott, W. B. (1887). On some new and little known creodonts. Journal of
American Museum Novitates, 2102, 1-33. the Academy of Natural Sciences, Philadelphia, 9, 155-85.
Mellett, J. S. (1969). A skull of Hemipsalodon (Mammalia, Deltatheridia) (1892). A revision of the North American Creodonta with notes on some
from the Clarno Formation of Oregon. American Museum Novitates, genera which have been referred to this group. Proceedings of the
2387, 1-19. Academy of Natural Sciences, Philadelphia, 1892, 291-323.
(1977). Paleobiology of North American Hyaenodon (Mammalia, Cre- (1895). The osteology of Hyaenodon. Journal of the Academy ofNatural
odonta). In Contributions to Vertebrate Evolution, eds. M. K. Hecht Sciences, Philadelphia, 9, 499-535.
& F. S. Szalay, 1, 1-134. (1938). A problematic cat-like mandible from the Uinta Eocene,
Novacek, M. J., Flynn, J. J., Ferrusquia-Villafranca, I., & Cipolletti, R. M. Apataelurus kayi, Scott. Annals of the Carnegie Museum, 27,
(1987). An early Eocene (Wasatchian) mammal fauna from Baja 113-20.
California. National Geographic Research, 3, 376-88. Scott, W. B., & Jepsen, G. L. (1936). The mammalian fauna of the White
Osborn, H.F. (1895). Fossil mammals of the Uinta Basin, expedition of 1894. River Oligocene, Part I: Insectivora and Carnivora. Transactions of
Bulletin of the American Museum of Natural History, 7, 71-105. the American Philosophical Society, 28, 1-153.
( 1897). The Huerfano Lake Basin, southern Colorado, and its Wind River Sloan, R. E. ( 1969). Cretaceous and Paleocene terrestrial communities of
and Bridger fauna. Bulletin of the American Museum of Natural western North America. Proceedings of the North American Paleon-
History, 9, 247-58. tological Convention, I, 427-53.
(1900). Oxyaena and Patriofelis restudied as terrestrial creodonts. Bul- Stock, C. (1933). Hyaenodontidae of the upper Eocene of California. Pro-
letin of the American Museum of Natural History, 8, 269-79. ceedings of the National Academy of Science, USA, 19, 434-40.
( 1910). The Age of Mammals in Europe, Asia, and North America. New Storer, J. E. (1984). Fossil mammals of the Southfork Local Fauna (early
York: Macmillan. Chadronian) of Saskatchewan. Canadian Journal of Earth Science,
Osborn, H. F., & Wortman, J. L. (1892). Fossil mammals of the Wasatch 21, 1400-5.
and Wind River Beds: collection of 1891. Bulletin of the American Stovall, J. W. (1948). Chadronian vertebrate fossils from below the Rim
Museum of Natural History, 4, 81-147. Rock of Presidio County, Texas. American Journal of Science, 246,
Osborn, H. F., & Wortman, J. L. (1894). Fossil mammals of the lower 78-95.
Miocene White River Beds: collection of 1892. Bulletin of the Amer- Stucky,R. K. (1984). The Wasatchian-Bridgerian Land Mammal Age bound-
ican Museum of Natural History, 6, 199-228. ary (early to middle Eocene) in western North America. Annals of
Peterson, 0. A. (1919). Report upon the material discovered in the upper the Carnegie Museum, 53, 347-82.
Eocene of the Uinta Basin by Earl Douglas in the years 1908-1909, Thorpe, M. R. (1922). A new genus of Oligocene Hyaenodontidae. American
and by 0. A. Peterson in 1912. Annals of the Carnegie Museum, 12, Journal of Science, 3, 277-87.
40--169. (1923). Notes on the Bridger (Eocene) Carnivora. American Journal of
(1931). New species from the Oligocene of the Uinta. Annals of the Science, 5, 23-39.
Carnegie Museum, 21, 61-78. Van Valen, L. (1965). Some European Proviverrini (Mammalia, Delta-
Polly, P. D. (1996). The skeleton of Gazinocyon vulpeculus gen. et comb. theridia). Palaeontology, 8, 638-65.
nov. and the cladistic relationships of Hyaenodontidae (Eutheria, (1966). Deltatheridia, a new order of mammals. Bulletin of the American
Mammalia). Journal of Vertebrate Paleontology, 16, 303-19. Museum of Natural History, 132, 3-126.
 Gregg F. Gunnell 105

Van Valkenburgh, B. (1987). Skeletal indicators of locomotor behavior in Wilson, J. A. (1986). Stratigraphic occurrence and correlation of early Ter-
living and extinct carnivores. Journal of Vertebrate Paleontology, 7, tiary vertebrate faunas, Trans-Pecos Texas: Agua Fria-Green Valley
162-82. areas. Journal of Vertebrate Paleontology, 6, 350-73.
West, R. M. (1973). Geology and mammalian paleontology of the New Fork- Wortman, J. L. (1894). Osteology of Patriofelis, a middle Eocene creodont.
Big Sandy Area, Sublette County, Wyoming. Fieldiana, Geology, 29, Bulletin of the American Museum of Natural History, 6, 129-64.
1-193. ( 1899). Restoration of Oxyaena lupina Cope, with descriptions of certain
West, R. M., & Dawson, M. R. (1978). Vertebrate paleontology and the new species of Eocene creodonts. Bulletin of the American Museum
Cenozoic history of the North Atlantic region. Polarforschung, 48, of Natural History, 12, 139-48.
103-19. (1902). Studies of Eocene Mammalia in the Marsh Collection, Peabody
White, T. E. (1952). Preliminary analysis of the vertebrate fossil fauna of the Museum, Part I: Carnivora. American Journal ofScience, 11, 333-48,
Boysen Reservoir area. Proceedings of the United States National 437-50; 12, 143-54, 193-206,281-96,377-82,421-32; 13,39-46,
Museum, 102, 185-207. 115-28, 197-206,433-48; 14, 17-23.
 Appendix: Synonyms/valid speci es
(numbered taxa are considered valid)

OXYAENIDAE 9. Oxyaena gulo Matthew, 1915

Probable synonyms: Oxyaena sp., McKenna, 1960;
Davidson, 1987.
TYTTH AENINA E
10. Oxyaenapardalis Matthew, 1915
1. Tytthaena parrisi Gingerich, 1980 11. Oxyaena intermedia Denison, 1938
2. Tytthaena lichna Gunnell and Gingerich, 1991 Synonyms: Oxyaenaforcipata, Kihm, 1984.
Synonyms: Oxyaena? lichna Rose, 1981. 12. Oxyaena simpsoni Van Valen, 1966
13. Patriofelis ulta Leidy, 1870
Synonyms: Patriofelis compressa Denison, 1938. Patriofelis
OXYAENINAE compressa, Gazin, 1957. Patriofelis cf. P. ulta, Robinson,
1966.
3. Dipsalidictis platypus Matthew, 1915
Probable synonyms: Patriofelis coloradensis Matthew, 1910
Synonyms: Oxyaena platypus, Denison, 1938; McKenna,
(in Osborn, 1910). Patriofelis coloradensis, Matthew, 1909;
1980; Rose, 1981. Denison, 1938. Ambloctonus coloradensis, Matthew, 1915.
4. Dipsalidictis aequidens (Matthew, 1915)
?Patriofelis coloradensis, Gazin, 1957.
Synonyms: Oxyaena aequidens Matthew, 1915. Oxyaena ae-
14. Patriofelisferox (Marsh, 1872)
quidens, Denison, 1938; Jepsen, 1930b; Van Valen, 1966;
Synonyms: Limnofelis ferox Marsh, 1872. Limnofelis lati-
Rose, 1981. dens Marsh, 1872. Oreocyon latidens Marsh, 1872.?Patri-
5. Dipsalidictis transiens (Matthew, 1915)
ofelis leidyanus Osborn and Wortman, 1892. Aelurotherium
Synonyms: Oxyaena transiens Matthew, 1915. Dipsalidic-
leidyanum Adams, 1896. Patriofelis (Limnofelis) latidens,
tis amplus Jepsen, 1930a. Dipsalidictides amplus Denison,
Matthew, 1899. Patriofelis (Oracodon) latidens, Matthew,
1938. Oxyaena transiens, Denison, 1938; McKenna, 1980.
1899. Aelurotherium (Patriofelis) leidyanum, Matthew, 1899.
Dipsalidictides amplus, Van Valen, 1966; Bown, 1979.
Aelurotherium latidens, Wortman, 1902. Aelurotherium bi-
Oxyaena sp. near 0. transiens, Bown, 1979. Oxyaena tran-
cuspis, Wortman, 1902. Patriofelis latidens, Thorpe, 1923.
siens, Rose, 1981 (in part).
15. Protopsalis tigrinus Cope, 1880
6. Dipsalidictis krausei Gunnell and Gingerich, 1991
Synonyms: Patriofelis tigrinus, Matthew, 1899, 1909, 1915;
Synonyms: Oxyaena transiens, Rose, 1981 (in part).
Guthrie, 1971. Patriofelis (Protopsalis) tigrinus, Gazin, 1957.
7. Oxyaena lupina Cope, 1874
Synonyms: Oxyaena morsitans Cope, 1874. Oxyaena mor-
PALAEONICTINAE
sitans, Cope, 1875, 1884a; Matthew, 1899. Oxyaena huerfa-
nensis Osborn, 1897. Oxyaena huerfanensis, Matthew, 1899. 16. Palaeonictis occidentalis Osborn and Wortman, 1892
Oxyaena cf. sp. 0. lupina, Robinson, 1966. Oxyaena sp., Probable synonyms: Palaeonictis sp., Kihm, 1984.
Gazin, 1962; West, 1973. 17. Palaeonictis peloria Rose, 1981
8. Oxyaenaforcipata Cope, 1874 Probable synonyms:?Oxyaena sp. innom., Matthew, 1915.
Synonyms: Oxyaena ultima Denison, 1938. Oxyaena ultima, 18. Ambloctonus sinosus Cope, 1875
Kelley and Wood, 1954. Probable synonyms: Ambloctonus hyaenoides Matthew,
Probable synonyms: Oxyaena lupina, Wortman, 1899; 1915. Ambloctonus hyaenoides, Denison, 1938; Lucas et al.,
Osborn, 1900. Oxyaena sp. cf. 0. lupina, Kihm, 1984. 1981.
106
 Synonyms/valid species 107

19. Ambloctonus priscus Matthew, 1915 9. Arfia shoshoniensis (Matthew, 1915)

20. Ambloctonus major Denison, 1938
Synonyms: Sinopa shoshoniensis Matthew, 1915. Arfia sho-
Probable synonyms: Ambloctonus cf. A. major, Gazin, -1952, shoniensis, Delson, 1971 (in part). Arfia opisthotoma, Bown,
1962. 1979 (in part).
21. Dipsalodon matthewi Jepsen, 1930b 10. Arfia zele Gingerich and Deutsch, 1989
22. Dipsalodon churchillorum Rose, 1981 Synonyms: Arfia shoshoniensis, Delson, 1971 (in part). Arfia
Synonyms: Dipsalodon, new species, McKenna, 1980. opisthotoma, Bown, 1979 (in part).
11. Arfia junnei Gingerich, 1989
MACHA EROIDI NAE (OXYAENIDAE?) 12. Sinopa rapax Leidy, 1871
Probable synonyms: Triacodonfallax Marsh, 1872. Stypolo-
23. Machaeroides eothen Matthew, 1909
phus pungens Cope, 1872. Stypolophus aculeatus Cope, 1872
24. Machaeroides simpsoni Dawson et al., 1986
(not Cope, 1881, 1884a). Triacodon aculeatus Cope, 1872.
25. Apataelurus kayi Scott, 1938
Sinopa pungens, Scott, 1892; Matthew, 1901, 1909. Sinopa
(Stypolophus) pungens, Matthew, 1899. Sinopa minor Wort-
man, 1902. Limnocyon velox, Wortman, 1902 (in part).
HYAENODONTIDAE Sinopa rapax, mutation lania, Matthew, 1909. Sinopa mi-
nor, Matthew, 1909; Thorpe, 1923; Kay, 1957; Krishtalka and
PROVIVERRINAE Stucky, 1984; Emry, 1990. "Stypolophus" aculeatus,
Matthew, 1909. Proviverra rapax, Van Valen, 1965. Pro-
I. Prototomus viverrinus Cope, 1874 viverrapungens, Van Valen, 1965. ?Proviverraminor(Wort-
Synonyms: Stypolophus, Osborn and Wortman, 1892 (in man), Van Valen, 1965. ?Proviverra, sp., West, 1973. Arfia,
part). Sinopa viverrina, Matthew, 1899, 1901, 1915. Sinopa sp. cf. A. opisthotoma, Kihm, 1984 (in part).
cf viverrina, Matthew, 1899. Sinopa, possibly S. viverrina, 13. Sinopa major Wortman, 1902
Gazin, 1962. Probable synonyms: Sinopa grangeri Matthew, 1906. Sinopa
2. Prototomus secundarius Cope, 1875 grangeri, Matthew, 1909; Krishtalka and Stucky, 1984. Pro-
Probable synonyms: Prototomus multicuspisCope, 1875. Sinopa viverra grangeri, Van Valen, 1965. ?Proviverra major
(Stypolophus) multicuspis, Matthew, 1899. Sinopa (Stypolo- (Wortman), Van Valen, 1965.?Proviverra major, Gustafson,
phus) secundaria, Matthew, 1899. Sinopa multicuspis, Matthew, 1986; Wilson, 1986.
1901, 1915. Sinopa secundaria, Matthew, 1901, 1915; Van 14. Tritemnodon agilis (Marsh, 1872)
Valen, 1966. Sinopa cf multicuspis, Gazin, 1962. Prototomus Synonyms: Limnocyon agilis Marsh, 1872. Stypolophus bre-
multicuspis, Van Valen, 1965; Guthrie, 1971. ?Prototomus vicalcaratus Cope, 1872. Sinopa (Limnocyon) agilis,
secundaria, Van Valen, 1965. Matthew, 1899. Stypolophus brevicalcaratus, Matthew, 1899.
3. Prototomus robustus (Matthew, 1915) Sinopa agilis, Matthew, 1901; Wortman, 1902.
Synonyms: Prolimnocyon robustus Matthew, 1915. Prolim- Probable synonyms: Tritemnodon sp. cf. T. whitiae, Krish-
nocyon robustus, Denison, 1938; Van Valen, 1966.?Paene- talka and Stucky, 1984.
prolimnocyon mordax, Delson, 1971 (in part). 15. ?Tritemnodon strenuus (Cope, 1875)
4. Prototomus deimos Gingerich and Deutsch, 1989 Synonyms: Prototomus strenuus Cope, 1875. Stypolophus
Synonyms: Cf. Tritemnodon, Bown, 1979 (in part). Prolim- strenuus, Cope, 1880, 1881. Sinopa strenua, Matthew, 1899,
nocyon atavus, Bown, 1979 (in part). 1901, 1915; White, 1952. Tritemnodon strenua, Van Valen,
5. Prototomus phobos Gingerich and Deutsch, 1989 1965; Guthrie, 1967a, 1971; Stucky, 1984. Tritemnodon sp.
Synonyms: Sinopa sp., McKenna, 1960. Prototomus sp., cf. T. strenua, Kihm, 1984.
Bown, 1979 (in part). Tritemnodon sp., Bown, 1979 (in part). Probable synonyms: Stypolophus hians Cope, 1877. Stypolo-
6. Prototomus martis Gingerich and Deutsch, 1989 phus whitiae Cope, 1881. Stypolophus aculeatus Cope, 1881
Synonyms: Prolimnocyon atavus, Bown, 1979 (in part). (not Cope, 1872). Stypolophus whitiae, Cope, 1882, 1884a.
Tritemnodon sp., Bown, 1979 (in part). Stypolophus aculeatus, Cope, 1884a (not Cope, 1872). Sty-
Probable synonyms: Prototomus sp., Kihm, 1984; Davidson, polophus, Osborn and Wortman, 1892 (in part). Sinopa hians,
1987. Matthew, 1899, 1901, 1915. Sinopa (Stypolophus) whitiae,
7. Gazinocyon vulpeculus (Matthew, 1915) Matthew, 1899. Sinopa (Stypolophus) aculeata, Matthew,
Synonyms: Sinopa vulpecula Matthew, 1915. Sinopa vulpec- 1899. Sinopa whitiae, Matthew, 1901. Sinopa cf. strenua,
ula, Gazin, 1952. Sinopa cf. vulpecula, Gazin, 1962; Gazin, 1952, 1962. Tritemnodon whitiae, Van Valen, 1965.
Robinson, 1960. Prototomus? vulpeculus, Gingerich and Tritemnodo.n hians, Van Valen, 1965. Sinopa cf. S. strenua,
Deutsch, 1989. Robinson, 1966.
8. Arfia opisthotoma (Matthew, 1901) 16. Proviverroides piercei Bown, 1982
Synonyms: Sinopa opisthotoma Matthew, 1901. Sinopa Probable synonyms: ?Proviverroides sp., Krishtalka and
opisthotoma, Matthew, 1915. Arfia sp. cf. A. opisthotoma, Stucky, 1984.
Kihm, 1984 (in part). 17. Pyrocyon dioctetus Gingerich and Deutsch, 1989
 Synonyms/valid species 108

18. Acarictis ryani Gingerich and Deutsch, 1989 31. Hemipsalodon viejaensis Gustafson, 1986
19. Galecyon mordax (Matthew, 1915)
Synonyms: Sinopa mordax Matthew, 1915. Prototomus mor- INDETERMINATE HYAENODONTID?
dax, Van Valen, 1965. Sinopa mordax, Van Valen, 1966.
?Paeneprolimnocyon mordax, Delson, 1971 (in part). 32. lschnognathus savagei Stovall, 1948
20. Hyaenodon horridus Leidy, 1853
Synonyms: Hyaenodon cruentus Leidy, 1853. Hyaenodon
cruentus, Leidy, 1870; Matthew, 1899; Thorpe, 1922; Clark, LIMNOCYONIDAE
1937; Scott and Jepsen, 1936; Russell, 1972. Neohyaenodon
1. Limnocyon verus Marsh, 1872
horridus Thorpe, 1922. Hyaenodon cf. H. cruentus, Gal-
Synonyms: Limnocyon riparius Marsh, 1872. Telmatocyon
breath, 1953. Hyaenodon cf. cruentus, Clark, Beerbower,
riparius Marsh, 1899. Sinopa (Limnocyon) vera Matthew,
and Kietzke, 1967. Hyaenodon cf. horridus, Clark, Beer-
1899. Sinopa vera, Matthew, 1901.
bower, and Kietzke, 1967. Hyaenodon ?horridus, Russell,
Probable synonyms: Triacodon grandis Marsh, 1872.
1972. Hyaenodon (Neohyaenodon) horridus, Mellett, 1977.
2. Limnocyon potens Matthew, 1909
Hyaenodon (N.) cf. H. horridus, Gustafson, 1986.
Synonyms: Limnocyon douglassi Peterson, 1919. Limnocyon
21. Hyaenodon crucians Leidy, 1853
douglassi, Kay, 1957. Limnocyon protens, Kay, 1957.
Synonyms: Hyaenodon leptocephalus Scott, 1887. Hyaen-
Probable synonyms: ?Hyaenodon, Osborn, 1895. Oxyaen-
odon paucidens Osborn and Wortman, 1894. Hyaenodon lep-
odon dysodus, Matthew, 1899; Peterson, 1919; Denison,
tocephalus, Matthew, 1899; Thorpe, 1922. Hyaenodon pau-
1938. ?Oxyaenodon wortmani Van Valen, 1966.
cidens, Matthew, 1899; Scott and Jepsen, 1936. Hyaenodon
3. Thinocyon velox Marsh, 1872
minutus Douglass, 1901. Cf. Hyaenodon ?minutus, Russell,
Synonyms: Limnocyon velox Wortman, 1902 (in part).
1972. Hyaenodon (Protohyaenodon) crucians, Mellett, 1977;
Probable synonyms: Thinocyon mustelinus Matthew, 1909.
Gustafson, 1986.
Thinocyonmustelinus, Thorpe, 1923; Denison, 1938. Thino-
22. Hyaenodon montanus Douglass, 1901
cyon sp., McGrew, 1959.
Synonyms: Hyaenodon cf. montanus, Clark, Beerbower, and
4. Thinocyon medius (Wortman, 1902)
Kietzke, 1967. Hyaenodon (Neohyaenodon) montanus, Mel-
Synonyms: Limnocyon medius Wortman, 1902.
lett, 1977; Gustafson, 1986.
5. Thinocyon cledensis Matthew, 1909
23. Hyaenodon mustelinus Scott, 1895
Synonyms: Hyaenodon (Protohyaenodon) mustelinus, Mel- 6. Thinocyon comptus (Marsh, 1872)
lett, 1977. Synonyms: Entomodon comptus Marsh, 1872. Thinocyon
24. Hyaenodon vetus Stock, 1933 comptus, McKenna, Robinson, and Taylor, 1962.
Synonyms: Pterodon califomicus Stock, 1933. Hyaenodon 7. Prolimnocyon atavus Matthew, 1915
sp., Peterson, 1931. Pterodon californicus, Golz and Lille- Synonyms: Prolimnocyon atavus, Delson, 1971 (in part).
graven, 1977. Hyaenodon (Neohyaenodon) vetus, Mellett, Probable synonyms: Prototomus viverrinus, McKenna, 1960.
1977. Hyaenodon (Neohyaenodon) cf. H. vetus, Gustafson, Prolimnocyon sp., Kihm, 1984.
1986. 8. Prolimnocyon antiquus Matthew, 1915
25. Hyaenodon brevirostris Macdonald, 1970 Probable synonyms: Prolimnocyon elisabethae Gazin, 1952.
Synonyms: Hyaenodon (Protohyaenodon) brevirostris, Mel- Prolimnocyon elisabethae, Kelley and Wood, 1954; Gazin,
lett, 1977. 1962; Van Valen, 1966; Guthrie, 1967a. Prolimnocyon cf. P.
26. Hyaenodon megaloides (Mellett, 1977) antiquus, Gazin, 1962.
Synonyms: Hyaenodon (Neohyaenodon) megaloides Mellett, 9. Prolimnocyon eerius Gingerich, 1989
1977. 10. Prolimnocyon haematus Gingerich and Deutsch, 1989
27. Hyaenodon microdon (Mellett, 1977) Synonyms: Prolimnocyon atavus, Delson, 1971 (in part).
Synonyms: ?Pseudopterodon minutus Matthew, 1903. Prolimnocyon atavus, Bown, 1979 (in part).
Hyaenodon (Protohyaenodon) microdon Mellett, 1977. 11. Oxyaenodon dysodus Wortman, 1899
28. Hyaenodon venturae (Mellett, 1977) Synonyms: Limnocyon dysodus Wortman, 1902. Oxyaen-
Synonyms: Hyaenodon exiguus Stock, 1933. Hyaenodon ex- odon dysclerus Hay, 1902. Oxyaenodon dysclerus, Peterson,
iguus, Golz and Lillegraven, 1977. Hyaenodon (Protohyaen- 1919; Denison, 1938; Kay, 1957.
odon) venturae Mellett, 1977.
29. Hyaenodon raineyi (Gustafson, 1986) NON-CREODONT TAXA (DESCRIBED
Synonyms: Hyaenodon (Protohyaenodon) raineyi, AS CREDONTS)
Gustafson, 1986.
30. Hemipsalodon grandis Cope, 1885 Triacodon nanus Marsh, 1872. Probable miacid or viverravid
Synonyms: Hemipsalodon cookei Schlaikjer, 1935a. carnivore; specimens now lost.
Probable synonyms: Pterodon sp., Lucas et al., 1981; Lucas, Stypolophus insectivorus Cope, 1872. Probable rniacid or
1982. viverravid carnivore; specimens now lost.
Synonyms/valid species 109

Sinopa eximia Leidy, 1873. Specimen of adapid primate

?Sinopa inse~tivorus, Matthew, 1901. See above (Stypolo-
Notharctus; probably Notharctus tenebrosus. phus amencana).
Prototom~s insidiosus Cope, 1874. Type species of hyop-
"Sty~olophus" insectivorus, Matthew, 1909. See above.
sodont1d condylarth Apheliscus.
Prol'.m~ocyon. iudei Guthrie, l 967a. Probable synonym of
Prototomus jarrovii Cope, 1874. Type species of adapid pri- miac1d carnivore Oodectes jepseni.
mate Pelycodus.
Paeneprolimnocyon amissadomus Guthrie, 1971. Probable
Limnocyon protenus Cope, 1874. Type species of viverravid
synonym of miacid carnivore Oodectes herpestoides.
carnivore Didymictis.
?Prolir:inocyon macfaddeni Rigby, 1980. Probable palaeo-
sry_polo~hus bicuspis Cope, 1880. Type specimen ofleptictid ryctJd or pantolestid insectivore.
insectivore Palaeictops bicuspis.
Tritemnodon sp. cf. T. gracilis, Krishtalka and Stucky,
Prov'.verra americana Scott, 1892. Probable artiodactyl.
1984. Probable synonym of rniacid carnivore Miacis
Prov1verra americana, Matthew, 1899. See above. gracilis.
6 Early Cenozoic Carnivora ("Miacoidea")

JOHN J. FLYNN

INTRODUCTION

Early Cenozoic representatives of the Order Carnivora were tax-

onomically and ecologically diverse. The early Cenozoic
Carnivora ("Miacoidea") generally were of small body size, typ-
ically ranging from weasel-to-fisher sized, with a maximum size
about that of a coyote. They were small brained, relative to living
carnivorans, but were comparable in brain size to contemporary
"carnivores" such as the Creodonta and Mesonychidae. Although
they generally are considered to possess relatively primitive skele-
tons, arboreal, scansorial, ambulatory, and cursorial locomotion
styles all have been proposed for various early Cenozoic Camivora
(Fig. 6.1).
Carnivorans are first known from the Puercan (early Paleocene)
of North America. They reach a peak of diversity in the Eocene
of North America, with lower diversity in Europe (late Paleocene-
late Eocene) and Asia (sporadic occurrences of early or middle
Paleocene; ?late Paleocene; late Eocene; and ?early Oligocene).
Their known distribution is exclusively Northern Hemisphere, with
representatives reported from North America, Europe, and Asia.
Early Cenozoic Carnivora generally have been considered a basal
stock of primitive Carnivora, which were ancestral to the later Ceno-
zoic Carnivora. As a group, they were assigned variably to "prim-
itive" taxa termed the "Miacidae" (Cope, 1880), "Viverravidae"
(Wortman and Matthew, 1899), or "Miacoidea" (Teilhard de Chardin,
1915). Animals within these groups have been allied primarily by
Figure 6.1. Restoration of an early Cenozoic carnivoran, Vulpavus profec-
the lack of features diagnostic of the later, more derived Camivora.
tus, based primarily on skeletal anatomy of V. profectus (AMNH 12626),
Because representatives of the living families of Camivora first ap- with additional information from V. ovatus (AMNH 11498). Coat pattern-
pear in late Eocene or younger strata, the archaic "Miacoidea" have ing speculative; inferred by comparison to coat patterns within other Cani-
been considered to range only into the late Eocene (through early formia and within the outgroup Feliformia. Postcranial anatomy indicates a
Chadronian). varied locomotor style, including both scansorial and terrestrial locomotion
(Jenkins and Camazine, 1977). Vegetation (e.g., hickory tree, Carya; fern,
The closest relatives to the Camivora appear to be the Creodonta
Osmunda; horse-tail, Equisetum; herbaceous ground-cover, Lilliales) recon-
(together grouped as Ferae). Morphologic similarities and temporal structed using known plant occurrences in coeval, nearby middle Eocene
and geographic overlap between the two groups has led to sugges- deposits (MacGinitie, Rohrer, and Leopold, 1974); note especially absence
tions of ecological competition between them. Among Cretaceous of grasses. Restoration by Marlene (Hill-Werner) Donnelly.

110
 Early Cenozoic Carnivora ("Miacoidea") 111

Figure 6.2. Comparison of skulls

of early Cenozoic Carnivora; dorsal
views of a representative taxon from
each of the two major clades (Fe-
liformia and Caniformia). A. Viver-
ravus (based on V. minutus, AMNH
12621); B. Vulpavus (based on V.
profectus, AMNH 12626). Graphical
methods used for reconstructions, to
remove distortion caused by crushing;

T
anterior part of rostrum of Viverravus
reconstructed by comparison with all
other available early Cenozoic Car-
nivora skulls preserving this region.
A B Illustration by Clara Simpson.

animals, the "palaeoryctoids" have been considered the closest rel- Viverravinae (shared with Feliformia): molars reduced to 2/2
atives of the Ferae, although there are no uniquely derived features (derived); larger anteroexternal cusp (parastyle) on P4 (derived);
shared by the Ferae and "palaeoryctoids." posterior accessory cusp on p4 (derived, probably present primi-
tively in all Carnivora); elongate skull (character polarity uncer-
tain); presence of calcaneal fibular facet (primitive). Additionally,
DEFINING FEATURES OF EARLY CARNIVORANS there are a few other postcranial features observable in only one
or a few specimens, and of uncertain polarity: digitigrady; humerus
The "Miacoidea" appear to be a paraphyletic group of taxa. As with flattened deltoid crest, high greater tuberosity; femur with small
such, their "defining features" consist of a combination of the pre- prominent third trochanter, long narrow patellar trochlea; astragalus
sence of features diagnostic of all Camivora, and the lack of derived with narrow oblique hear, trochlea more or less excavated, inner and
features characteristic of monophyletic groups within Carnivora. outer keels well developed.
Figures 6.2-6.4 illustrate some aspects of craniodental anatomy Miacinae (shared with Caniformia): full molar complement of
of these primitive camivorans. For instance, the extensive ~naly­ 3/3 (primitive); anteroexternal cusp (parastyle) on P4 rudimentary
sis and diagnosis of the "Miacidae" by Matthew (1909a) included or absent (not true of all "Miacinae"); skull length moderate or short
(1) Carnivora synapomorphies -characters shared by all camivoran (polarity uncertain); absence of calcaneal fibular facet (derived). Ad-
lineages (P4/ml carnassials; postcarnassial teeth generally tuber- ditionally there are a few other postcranial features observable only
cular; metacone smaller than protocone; ungual phalanges com- in one or a few specimens, and of uncertain polarity: plantigrady;
pressed and unfissured at tip; paraxonic digit symmetry), and (2) humerus with high, ridged deltoid crest, small greater tuberosity;
symplesiomorphies - characters shared with Creodonta (and many femur with low broad third trochanter, wide short patellar trochlea;
other primitive Eutheria), but modified in other Camivora (scaphoid, astragalus with wide fiat head, trochlea fiat without inner keel.
lunar, and centrale of carpus generally distinct; third trochanter
present on thefemur; entepicondylar foramen present on the humerus;
small brain; unossified tympanic bulla; five toes present on both SYSTEMATICS
manus and pes). Matthew (1909a) further elaborated characters
taken to be diagnostic of the subfamilies Viverravinae and Miacinae.
These features actually are shared with other Feliformia (Viverrav- SUPRAFAMILY
inae) or Caniforrnia (Miacinae), either as syriapomorphies or sym- Since Cope (1880) first erected the Miacidae, the systematics of the
plesiomorphies for those groups. Matthew's (1909a) diagnoses are early Cenozoic Camivora have been confusing and controversial
as follows: (see summary in Flynn and Galiano, 1982). To emphasize their
 John J. Flynn 112

a Protictis Oi:idectes

b
~~~
Bryanictis Miacis

C lntyrictis
k Vassacyon

d Didymictis Uintacyon

e m
~~l\<CJC>
Prodaphaenus

f Simpsonictis n

Q Raphictis 0 Procynodictis

h
~~i)
lctidopappus " p Tapocyon

Figure 6.3. Comparison of lower dentitions of early Cenozoic Camivora. Portions modified from Flynn and Galiano, 1982, Figure 2, represen-
tative species used, some drawings composites, some specimen orientations reversed for consistent orientation. The first molar of each taxon is
drawn to approximately the same size, and the dentitions are aligned vertically along the anterior edge of the first molar. A. Protictis. B. Bryanic-
tis. C. lntyrictis. D. Didymictis. E. Viverravus. F. Simpsonictis. G. Raphictis. H. lctidopappus. I. Oodectes. J. Miacis. K. Vassacyon. L. Uintacyon.
M. Prodaphaenus. N. Vulpavus. 0. Procynodictis. P. Tapocyon.
 Early Cenozoic Carnivora ("Miacoidea") 113

<J@WJW~E3
h Oiidectes

b
®f!j~~
Bryanictis
Miacis

~~~
C
~VJ~~
Didymictis ~
Vassacyon

d
~
Viverravus

k Uintacyon

9 Simp,sonictis
~
Prodaphaenus

f lctidopappus

m Vulpavus

Q Palaearctonyx
n Procynodictis

0 Tapocyon

Figure .6.4. Comparison o'. upper dentitions of. early Cenozoic Camivora. Portions modified from Flynn and Galiano, 1982, Figure 3; taxa and
conventions the ~~me. as Figur~ ~.3, but exclude lntyric~is a~d .Raphictis (known only from lower dentitions) and add Palaearctonyx (known only
fro~ u~per dentit10n). A. Protictts. B. Bryanzctis. C. D1dym1ctzs. D. Viverravus. E. Simpsonictis. F. Ictidopappus. G. Palaearctonyx. H. Oodectes.
I. Mzacis. J. Vassacyon. K. Uintacyon. L. Prodaphaenus. M. Vulpavus. N. Procynodictis. O. Tapocyon.
 John J. Flynn 114

relatively primitive morphology, early workers (e.g., Cope, 1880; Baskin, this volume, Chapter 9); and the monophyletic Superfamily
Trouessart, 1885; Matthew, 1909a) even included them within the Aeluroidea. Distribution: ?early Paleocene to Recent. Diagnosis:
Creodonta (whose members were all extinct) rather than within a Carnivorans differing from all other Carnivora in the following
group with living representatives - the Carnivora (or approximately derived dental features: P4 parastyle enlarged; deep, narrow
equivalent groups, the Fissipeda or Camassidentia). Since carnassial notch on P4; M3/m3 lost in all taxa (also true of some Can-
Simpson's (1945) classic classification of the Mammalia, most work- iformia); M2 reduced in size and more simple in morphology; m2
ers have recognized the relationship of "miacids" to true carnivo- talonid extremely elongate, entire tooth with elongate oval outline;
rans, and classified them together. Simpson (1945) included three m2 hypoconid larger than, or equal to, hypoconid size (especially
superfamilies (Miacoidea, Canoidea, and Feloidea) within the sub- height).
order Fissipeda, Order Carnivora (the only order within the Super-
order Ferae). More recently, other workers (e.g., Flynn and Galiano, Suborder Caniformia Kretzoi, 1943. Included taxa: the early
1982) have challenged the assumption that the "Miacoidea" were Cenozoic "miacoid" genera Miacis, Oodectes, Palaearctonyx,
a closely related, "primitive" group that formed, in its entirety, the Paramiacis, Paroodectes, Procynodictis, Prodaphaenus, Quercy-
ancestral stock for all later, more "advanced" Camivora. Their al- gale, Tapocyon, Uintacyon, Vassacyon, Vulpavus, and possibly
ternative proposition is that the Carnivora split into two distinct Xinyuictis (synonymized with Miacis by Gingerich, 1983); and the
lineages early in their evolutionary history, and these two lineages monophyletic superfamilies Arctoidea (including Pinnipedia) and
(Caniformia and Feliformia) can be recognized in both living car- Cynoidea. Distribution: early Eocene to Recent. Diagnosis: Car-
nivorans and their early Cenozoic relatives. This debate has yet to nivorans differing from all other Camivora in the following derived
be fully resolved (see Flynn, Neff, and Tedford, 1988), but in this features: loss of the calcaneal fibular facet in the pes (also in Nim-
chapter I accept the assignment of various genera of "Miacoidea" ravidae and some Aeluroidea); great reduction or absence of the P4
to either the Feliformia or Caniformia. parastyle; reduction in size of the P4 protocone; lingual cingulum
on all upper molars complete around base of protocone; large teres
major process on the scapula (also in Nimravidae); lack of Cowper's
INFRAFAMILY glands (not observable in "miacoids"); maxilloturbinals large and
I present diagnoses here for the Order Camivora and its constituent branching, excluding nasoturbinals from narial opening (condition
suborders Feliformia and Caniformia. These diagnoses follow Flynn not yet known for "miacoids"; ?secondarily reduced in lutrines).
and Galiano (1982) and Flynn, Neff, and Tedford (1988); see those
papers for further discussion. Genera of "miacoids" are considered
as members of one of the carnivoran suborders. Figure 6.5 and its INCLUDED GENERA OF NORTH AMERICAN
legend provide a hypothesis of interrelationships among the gen- EARLY CARNIVORANS
era of early carnivorans, excluding the problematical taxa Miocyon,
Ravenictis, and Pristinictis (see discussion later in text). The locality numbers listed for each genus refer to the list of unified
localities in Appendix I. The acronyms for museum collections are
Order Camivora Bowdich, 1821 listed in Appendix III.
Included taxa: Feliformia and Caniformia (Nimravidae incertae The locality numbers may be listed in a couple of alternative ways.
sedis within Carnivora). Distribution: ?early Paleocene to Recent. Parentheses around the locality (e.g., [CP101]) mean the taxon in
Diagnosis: These characters appear to be derived and diagnostic question at that locality is cited as an "aff." or "cf." the taxon in
for the Carnivora, distinguishing them from other Eutheria: P4/ml question. Parentheses are usually used for individual species, thus
modified as the only carnassial teeth in adults, lacking a migratory implying the genus is firmly known from the locality, but the actual
locus for the carnassial as found in the Creodonta; P4 protocone lo- species identification may be questionable. Question marks in front
cated anterolingually, far forward of the paracone; extremely elon- of the locality (e.g., ?CP101) mean the taxon is questionably known
gate P4 metastyle wing with a well-developed metastyle blade and from that locality, thus implying some doubt that the tax on is actually
carnassial notch; well-developed, elongate p4 talonid with at least present at that locality, either at the genus or the species level.
two posterior accessory cusps on the posterior flank of the main
cusp; Ml/ml much larger than M2-3/m2-3, progressive decrease
FELIFORMIA
in size from the first to last molar; distinct rostral entotympanic
element associated with anterior-most portion of extracranial path Bryanictis Macintyre, 1966
of internal carotid artery (uncertain for "miacoids," as no ossified Type species: Bryanictis microlestes (Simpson, 1935)
ectotympanic or entotympanic elements known); lack of a stapedial (= Didymictis microlestes).
artery (uncertain for "miacoids"); retractile claws. Type specimen: USNM 9301, left lower jaw with p4-m2.
Characteristics: Very prominent lingual cingulum on Ml;
Suborder Feliformia Kretzoi, 1945. Included taxa: The early ml-2 trigonid short, talonid broad and basined; p4 cusps
Cenozoic "miacoid" genera Didymictis, Protictis, Bryanictis, In- all strongly laterally compressed (but not as greatly as
tyrictis, Raphictis (all probably within Family Didymictidae ), Simp- in Bryanictis) (Flynn and Galiano, 1982; Gingerich and
sonictis, V!verravus (both within Family Viverravidae), and Icti- Winkler, 1985).
dopappus and Pappictidops (both incertae sedis); Palaeogale (see Average length of ml: 4.66 mm.
 Early Cenozoic Carnivora ("Miacoidea") 115

~ ~
~ ~ o<:: (/)<::
~ -~ .Cf ~ 9.Q. o<:: o<:: 'lt
-~ Cf ~ <fl ~ Cf ~
if Cf ~
~ 0,0
~
~
'lt
OQ.
,'
Cf .l( ,§ ~ ~ ~ ~
~
<:.' ~ -~Q. ~ -~ c1 -~ 9.'lt 0
4' ~ b'lt 0
-~ ~
~ "
,i::: <$' ~ :f Cf
"
-~
--~
60 .:,;~ Q.'lt .:,;~ ;§ ~
Q. ~
'"'
5

11
6: Viverravidae

Feliformia: 2

7: Caniformia

1: Carnivora

Figure 6.5. Tentative hypothesis of phylogenetic relationships for early Cenozoic Carnivora previously referred to as "Miacoidea." Living taxa and
non-"miacoid" fossil taxa not included. Characters hypothesized to be derived for each clade given here; numbers refer to the nodes identified on the
figure. See Flynn and Galiano (1982) and Flynn, Neff, and Tedford (1988) for further discussion of characters defining many of these clades. Generic
characteristics listed in text. Key to characters at the nodes: (I) CARNIVORA: (8 characters, listed in "Systematics" section of text). (2) FELIFORMIA:
(6 characters, listed in "Systematics" section of text). (3) DIDYMICTIDA (DIDYMICTIDAE): m2 paraconid extremely, and metaconid moderately re-
duced; paraconid much smaller than or subequal to metaconid size; well-developed notch between anterior accessory cusp and main cusp on p4; p4
talonid broad; canines multigrooved. (4) UNNAMED TAXON (Protictis, lntyrictis, Bryanictis, and questionably Raphictis): anterior accessory cusp on p4
enlarged and very high; lingual margin of p4 anteroposteriorly straight; P3 develops a weak, but distinct, protocone cusp anterior to the paracone apex;
main cusp on p4 leaf shaped in lateral outline. (5) UNNAMED TAXON (lntyrictis and Bryanictis): p4 anterior accessory cusp tremendously enlarged
and modified for shearing; expansion and very deep basining of lingual portion of p4 talonid; large, well-developed P3 protocone; all p4 cusps in a
straight, anteroposterior line on buccal tooth margin; buccal tooth margin anteroposteriorly straight. (6) VIVERRAVIDAE: Characters shared with other
Aeluroidea: conules absent on upper molars; ml paraconid and metaconid far separated, resulting in an open V between paralophid and protolophid on the
trigonid; loss of second posterior accessory cusp on p4; reduction of molar cingulae. Characters shared with Palaeogale: paraoccipital process extremely
reduced; supraoccipital process extremely reduced. Other characters: postglenoid foramen located far posteroextemally on the postglenoid process of the
squamosal; p4 anterior accessory cusp far anterolingual on crown; m2 talonid extremely elongate; hypoconulid large and bulbous with anterior occlusal
facet; anterobuccal basal trigonid cingulum on ml weak; ml hypoconulid strong, posterocentrally located, and projecting posteriorly from talonid margin;
anterior trigonid notch on m2 reduced, resulting in a continuous, low paralophid. (7) CANIFORMIA: (7 characters, listed in "Systematics" section of text).
(8) UNNAMED TAXON ("Miacis" and questionably Procynodictis): mandible elongate, slender; premolars enlarged; M2-3/m2-3 reduced and lat-
erally compressed; long, slender limbs. (9) UNNAMED TAXON (Oodectes, Vulpavus, and Palaearctonyx): P4 parastyle wing relatively short;
Ml buccal border anteroposteriorly straight, metastyle wing approximately equal in size to parastyle wing; ml trigonid height reduced (barely
higher than, or equal to, premolar height); lower molar talonids relatively broad, basined (especially ml); shear reduced on all molar crests.
(10) UNNAMED TAXON (Vulpavus and Palaearctonyx): hypocone present on Ml; P4 metastyle very short; P4 protocone broad anteroposteriorly,
located relatively far posterior on crown; upper molars quadrate; molar cusps low. (11) UNNAMED TAXON (Vassacyon, Uintacyon, Prodaphaenus, and
Tapocyon): molar talonids subtrenchant; canines compressed; anterior premolars reduced; relatively short mandibles and cranial facial region; mandible
deep in symphyseal region. (12) UNNAMED TAXON (Uintacyon, Prodaphaenus, and Tapocyon). Molar talonids trenchant; sharp angle at anterior border
of mandible; M 1 parastyle broad and projects far buccally.

Included species: B. microlestes only (known from localities Type specimen: USNM 1092, lower jaws from the New Mex-
CP13B, [CP15A], NP19C, [NP20E], [NP47C]). ican Wasatch (Matthew and Granger, 1915).
Characteristics: Relatively primitive feliform, in both den-
Didymictis Cope, 1875 tition and skeleton; p3-4 accessory cusps well developed
Type species: Didymictis protenus (Cope, 1874) (=Limno- (primitive?); upper and lower molar cusps and crests
cyon protenus). robust, rounded, and blunt (modified for "crushing,"
 John J. Flynn 116

according to Matthew, 1937, p. 101); cingulae well de- [NP20E]); P. agastor (locality CP13E); P. (Protictoides)
veloped on all teeth; ml talonid very broad and deeply aprophatos (locality CP36B); P. dellensis (localities
basined; m2 trigonid very low (much lower than in other CP13G, CP18B, CP22B, [NP7D]); P. laytoni (localities
early Cenozoic feliforms); mandible relatively heavy and CP13G, [NP3F]); P. (Protictis) paralus (localities CP13E,
deep; Ml talon and protocone broad anteroposteriorly; NP3C, NP47B); P. schaffi (locality CP13G); P. n. sp. (lo-
M2 ovoid in outline, metacone very small (Matthew and cality SB23E); P. n. sp. A (locality SB23F); P. n. sp. B
Granger, 1915; Matthew, 1937). (locality SB23F).
Average length of ml: (D. protenus): 7.83 mm. Protictis sp. is also known from localities CP13F,
Included species: D. protenus (known from localities [SB21], CP21B, CP24A, ?CP24B, NPlB, NP19B, NP19IIA,
[SB22A], SB24, CP13H, CP17B, CP18B, CP19, CP20A, NP19IIC, NP20E.
CP20B, CP20C, CP25B, CP25C, CP25D, [CP25F],
CP26B, CP26C, [CP27 A], CP27B, CP63, CP64A, Raphictis Gingerich and Winkler, 1985
CP64B); D. altidens (localities SB22B, CP20A, CP20B, Type species: Raphictis gausion Gingerich and Winkler,
CP20C,CP25E,CP25G,CP25H,CP 25J,CP27C,CP27D, 1985.
CP27E, CP64C); D. vancleveae (locality SB22C). Type specimen: PU 21244, left dentary with p3-m2.
Didymictis sp. is also known from localities ?CPlC, Characteristics: Relatively narrow lower cheek teeth; high,
CP13G, CP14E, CP17A, CP17B, CP26A, (CP28C), needlelike ml trigonid cusps; ml talonid set at angle rel-
CP32A, (CP36B), CP62B, CP62C, NP20E (2 spp.), ative to trigonid and long axis of dentary (Gingerich and
NP25B, NP49. Winkler, 1985).
Average length of ml: 3.4 mm.
Included species: R. gausion only (known from localities
lntyrictis Gingerich and Winkler, 1985 CP13E, [NP3B], [NP3C]. NP7D]).
Type species: lntyrictis vanvaleni (Macintyre, 1966) (=Pro- Raphictis sp. is also known from localities NP3B, NP7D.
tictis [Bryanictis] vanvaleni).
'fype specimen: AMNH 16031, left dentary with p4 and par- Simpsonictis Macintyre, 1962
tial ml. Type species: Simpsonictis tenuis (Simpson, 1935) (= Didy-
Characteristics: Relatively low-crowned p4 and relatively mictis tenuis).
low ml trigonid (in comparison to other early Cenozoic Type specimen: USNM 9297, fragment of left dentary with
feliforms) (Gingerich and Winkler, 1985). partial p4 (identification and association uncertain; see
Average length of ml: 4.43 mm. Simpson, 1937, pp. 212-13) and complete ml.
Included species: /. vanvaleni only (known from localities Characteristics: High, sharp piercing cusps on all cheek teeth;
SB23H, CP14A). triangular p4, narrow anteriorly and broader posteriorly in
occlusal view; p4 anterior accessory cusp large (but not
Protictis (Matthew, 1937) (= Didymictis [Protictis]) as large or laterally compressed [bladelike] as in Protic-
Type species: Protictis haydenianus (Cope, 1882) (= Didy- tis; and lacks deep notch between anterior accessory cusp
mictis haydenianus). and main cusp found in Didymictidae); reduced ml para-
Type specimen: AMNH 3368, left dentary with p4-m2, left conid; relatively large m2 with very high trigonid (Flynn
maxilla with P3-M2. and Galiano, 1982; Gingerich and Winkler, 1985).
Characteristics: Moderately wide p4, with anterior acces- Average length of ml (S. tenuis): 2.85 mm.
sory cusp large, high, and laterally compressed (blade- Included species: S. tenuis (known from localities CP13B,
like ); p4 posterior accessory cusp very high on posterior CP16A, [NPlC], NP19C); S. jaynanneae (locality
flank of main cusp; m2 talonid relatively much narrower CP14A); S. pegus (localities CP13B, NPlC).
than other early Tertiary Carnivora; P4 parastyle lobe elon- Simpsonictis sp. is also known from localities ?CP15A,
gate oval in outline (very narrow transversely, elongated NP19IIA.
in anterobuccal/posterolingual direction); posterior margin
of P4 paracone extremely straight and vertically oriented; Viverravus Marsh, 1872
P4 paracone anterior slope long and extends far posteri- Type species: Viverravus gracilis (Marsh, 1872) (= Triacodon
orly, apex relatively far posterior, paracone appears to "tilt" fallax).
posteriorly; upper molars much broader than long, meta- Type specimen: YPM 11836, pair of lower jaws (right jaw
conid small, metaconule and lingual cingulum very small with p3-m2) and left P4.
or absent (Macintyre, 1966; Flynn and Galiano, 1982; Gin- Characteristics: Mandible long and slender; skull elongate;
gerich and Winkler, 1985). p3 higher than p4, p3 almost as long as p4; ml-2 talonids
Average length of ml: (P. haydenianus): 6.96 mm. trenchant-cristid obliqua aligned more directly anteropos-
Included species: P. (Protictis) haydenianus (known from lo- teriorly (parallel to tooth axis) than in other early Ceno-
calities SB23C, SB23E, SB23F, SB23G, SB23GG, zoic feliforms, sulcus obliquus transversely narrow and
SB23H, CPlC, CP13B, CP14A, [CP14D], NP19C, deep; ml shear emphasized on paralophid and reduced on
 Early Cenozoic Carnivora ("Miacoidea") 117

posterior face of trigonid (relative to other early Ceno- [CP27E], CP63, CP64A); M. gracilis (locality CP6B); M.
zoic feliforms), via metaconid reduction and paraconid hargeri (localities CP34C, CP34D); M. (?) hookwayi (lo-
enlargement; ml metaconid height lower than paraconid, cality CC9A); M. jepseni (locality CP27B); M. latidens
and metaconid directly lateral to protoconid, rather than (localities [CP25G]. [CP25H], CP27B, [CP27C], [CP27D],
posterolingual, thus shear on protolophid oriented more [CP27E], CP34B, CP64B); M. medius (locality CP38B);
transversely; sectorial shearing better developed than in M. petilus (localities CP20B, CP20C, CP27B, CP27D,
any other early Cenozoic camivoran (Matthew, 1909a, p. CP27E); M. sylvestris (localities CP34C, CP34D); M.
356); some advanced species develop hypocone on upper uintensis 1 (localities [CP6AJ, CP6B); M. washakius (lo-
molars (Flynn and Galiano, 1982). cality CP38B); M. winkleri (localities CP19, [CP20A]).
Average length of ml CV. vulpinus): 5.44 mm. Miacis sp. is also known from localities CC7C, CC9AA,
Included species: V. gracilis (known from localities SB22B, ?SB22C, SB44D, CP34A, CP38C, NP25B, NP49 (see also
SB22C,CP5A,CP20A,CP20B,CP2 0C,CP27B,CP27D, note 1).
CP27E, CP31E, CP34B, CP34C, CP34D); V. acutus (lo-
calities [CP14E], CPI 7A, CPI 7B, CP18B, CP19, CP20A, Oodectes Wortman, 1901 (synonym: Paeneprolimnocyon)
CP20B, CP20C, [CP25C], [CP25F], [CP26C], [CP28C], Type species: Oodectes herpestoides Wortman, 1901.
CP63); V. bowni (localities CP17B, CP18B, CP19, Type specimen: YPM 11861, type illustrated in Wortman
CP20A); V. lutosus (localities CP20A, [CP20B], [CP20C], (1901, p. 151, Figure 24) as right upper jaw with Cl, P2-
CP25G, CP25H, CP25J, CP27B, CP27D, CP27E, M2; associated dental and skeletal material shown as co-
[CP31B], CP64B); V. minutus (localities NB14, CP5A, types (some type material also included under YPM 12845,
CP34C, CP65); V. politus (localities CPI 7A, CPI 7B, according to Thorpe, 1923).
CP18B, CP19, CP20A, CP26C, [CP64A]); V. sicarius (lo- Characteristics: pl single rooted; lower premolars lack an-
calities SB22C, CP5A, [CP32B], CP34C, [CP36A]). terior and posterior accessory cusps; p2--4 relatively short
Viverravus sp. is also known from localities ?CP24B, anteroposteriorly, high crowned relative to molar trigo-
?CP24C, CP31E, CP34A, CP34B, ?CP38B, CP65, NP8, nid heights; m2-3 larger (in trigonid development and
NP14 (see also note 1). crown length; relative to ml) than in other early Cenozoic
caniforms; P4 parastyle larger than in any other early
Comment: Triacodon Marsh has priority, but suppressed by
Cenozoic caniform (primitive?); Ml metacone and
Commission on Zoological Nomenclature, 1988.
1 Note added following completion of manuscript. Recently, Bryant (1992)
and Fox and Youzwyshyn (1994) erected new early Cenozoic carnivo-
CANIFORMIA ran taxa and proposed significant taxonomic changes or assignments that
' could not be fully evaluated and incorporated into the text and figures
Miacis Cope, 1872 of this chapter. However, for completeness, they are briefly considered
Type species: Miacis parvivorus Cope, 1872. here.
Bryant (1992) reevaluated the taxonomy of Prodaphaenus and con-
Type specimen: AMNH 5019, lower jaw with m2 and alveoli
cluded that the type species of the genus is "Miacis" uintensis (based on
for ml and m3 (Matthew [1909a], Figure 15 incorrectly Matthew, 1899; following Hay, 1902, and Matthew, 1909a) rather than
lists AMNH 11496 as the type). "P." scotti as has been suggested by most later authors, including this
Characteristics: Generally primitive caniform; probably para- chapter. Bryant (1992) raised Miocyon (originally proposed by Matthew,
phyletic; recognized primarily by absence of derived char- 1909, as a subgenus of Uintacyon, with U. (Miocyon) scotti as the type
species) to the generic rank, including three species - M. scotti (Uintan),
acters diagnostic of other early Cenozoic caniforms, as
M. major (previously" Uintacyon" major; ?Wasatchian-Bridgerian, contra
listed by Gingerich (1983, p. 205; after Matthew, 1909a): the apparent Uintan or younger range indicated in the generic distribution
"Camassial and tubercular dentition well differentiated, by Bryant 1992), and M. magnus (new species in Bryant 1992; Uintan to
carnassials and premolars moderate to large, tubercular Duchesnean, described from locality NP9A, and material referred from
dentition small, lower molars with basined heels [deriv- localities CP7C, CP37B, NP8, and from the late Uintan of Texas). He
considered Miocyon a "miacid," indicating assignment to the Caniformia
ed?], Ml in early forms with small hypocone [derived?],
as used in this chapter.
slender or moderately deep jaw with the symphyseal region Fox and Youzwyshyn (1994) describe exciting new carnivoran ma-
comparatively shallow and loosely sutured, and typically terial and two new taxa from the Paleocene of Canada. This material,
a full eutherian dentition, with M3 vestigial or absent in which I temporarily consider incertae sedis within Carnivora, has impor-
some species." tant implications for understanding the phylogeny of the Carnivora and
character transformation within the clade. Pristinictis connata from the
Average length of ml (M. exiguus): 5.45 mm. early Tiffanian of Alberta (locality NPI C) was considered a viverravid
Included species: M. parvivorus (known from localities (or Feliformia as used in this chapter); Ravenictis krausei from the early
[SB22C], [SB24], [CP5A], [CP27D], [CP27E], CP34C, Puercan of Saskatchewan (locality NP7B) was considered incertae sedis
CP35); M. australis (locality SB44B); M. cognitus (local- within the Carnivora.
ity SB44C); M. deutschi (localities CPI 9, CP20A, [CP63]); Additionally, Beard, Dawson, and Tabrum (1995) report possible spec-
imens of Miacis and Viverravus from the Red Hot Local Fauna (locality
M. exiguus (restricted by Gingerich 1983 to early-middle
GC2111) of the early Wasatchian of Mississippi, and Standhardt (1995)
Graybullian only) (localities CP19, CP20A, [CP20B], reports a possible specimen of Protictis from the Dogie Locality (locality
[CP20C], [CP25G], [CP25HJ, CP27 A, [CP27B], [CP27D], SB3911), of the Puercan (Puercan 2 or 3) of Texas.
 John J. Flynn 118

metastyle large and metacone/metastyle wing projects Prodaphaenus Wortman and Matthew, 1899
posterobuccally, subequal in size to paracone/parastyle Type species: Prodaphaenus scotti Wortman and Matthew,
wing (Wortman, 1901; Matthew, 1909a). 1899. 1
Average length of ml: 4.66 mm. Type specimen: ("Princeton collection") PU 11,238, left
Included species: 0. herpestoides only (includes "O. prox- P3-M2 (with alveolus for M3?).
imus") (known from localities SB22C, CP20A, CP20C, Characteristics: Mandible short, deep (including the symph-
CP34C). ysis), and robust, sharp angle ("chin") at the anterior border
Oodectes sp. is also known from localities ?SB22B, (similar to Tapocyon and Uintacyon); extremely deep mas-
SB24, ?CP5A, CP27D, CP27E. seteric fossa; premolars very small; pl single rooted, p2
possibly single rooted; p3 much smaller than p4; P3 small,
Palaearctonyx Matthew, 1909a simple; upper molars with broad stylar shelf; Ml parastyle
Type species: Palaearctonyx meadi Matthew, 1909a. broad, projects buccally, formed by two distinct cusps on
Type specimen: AMNH 12158, fragmentary skeleton includ- type specimen; M2-3 low crowned, almost completely
ing right maxilla (P3-M2). rimmed by a broad cingulum; M2 elongated anteroposteri-
Characteristics: Known only from an upper dentition and orly, relative to other early Cenozoic caniforms; M3 small
partial skeleton; molars quadrate in outline, lingual mar- and ovoid, paracone is only distinct cusp (Wortman and
gin almost as broad as labial margin; strong lingual cin- Matthew, 1899; Gustafson, 1986 ["Uintacyon" scotti]).
gulum on molars; premolars reduced in size relative to Average length of ml: 11.0 mm.
molars; molars low crowned, cusps somewhat bunodont; Included species: P. scotti only (includes "P. vulpinus")
P4 protocone very broad anteroposteriorly, positioned (known from localities SB43A, SB43B, CP6A, CP6B,
farther posteriorly (medial to paracone) than in any other [NP8]).
early Cenozoic camivoran (Vulpavus and Prodaphaenus
are most similar); P4 with broad postcingulum; P4 Tapocyon Stock, 1934
metastyle wing very short; Ml parastyle wing small; scap- Type species: Tapocyon occidentalis Stock, 1934.
hoid and centrale united with lunar separate (Matthew, Type specimen: CIT 1654, right P4 and Ml (paratype, CIT
1909a). 1655, fragmentary left ramus with cl, p2-3, ml-2).
Average length of ml: Unavailable. Average length of P4: Characteristics: Mandible robust, short, sharp angle ("chin")
6.0mm. at anterior border, and deeper at symphysis than in other
Included species: P. meadi only, known from locality early Cenozoic caniforms (Uintacyon and Prodaphaenus
CP34C only. most similar); lower premolars small, p2 much smaller
than p3; ml talonid short, much narrower than trigonid,
Procynodictis Wortman and Matthew, 1899 situated on lingual border of tooth, trenchant ridge con-
(synonym: Plesiomiacis) nects posterior base of trigonid with a single posterome-
Type species: Procynodictis vulpiceps Wortman and dial talonid cusp; ml paralophid oriented more anteropos-
Matthew, 1899. teriorly than in most other early Cenozoic caniforms; m2
Type specimen: AMNH 2514, left and right dentaries (left small; mandible robust anteriorly, with relatively straight
dentary with p2-m3) and left maxilla (with Pl-M2). ventral border; Ml parastyle extremely large, parastyle
Characteristics: Lower premolars separated by diastemata; wing projects far buccally as a distinct lobe, making the
anterior ends of p2-4 inflected lingually; p 1 single rooted; anterior border of Ml much broader transversely than the
lower molar talonids trenchant, with moderate lingual posterior border; Ml transversely broad, narrow antero-
basin; Ml parastyle wing long and narrow, paracone on posteriorly (Stock, 1934; Dawson, 1980).
anterior-most edge of tooth, paracone widely separated Average length of ml: Unavailable. Average length of P4
from metacone, metacone approximately equal in size to (T. occidentalis): 16.5 mm.
paracone, hypocone moderately well developed, all cin- Included species: T. occidentalis (known from locality CC9A,
gulae reduced (especially precingulum), paraconule and CC9AA); T. robustus (localities CP6B, [CP29C], NP22]).
metaconule cristae reduced; M3/3 very reduced or absent; Tapocyon sp. is also questionably known from locality
scaphoid and lunar fused into a single scapholunar bone CC7B.
(Wortman and Matthew, 1899; Stock, 1935; Flynn and
Galiano, 1982). Uintacyon Leidy, 1872
Average length of ml: 18.5 mm. Type species: Uintacyon edax Leidy, 1872.
Included species: P. vulpiceps (known from localities Type specimen: unnumbered (Matthew [1909a], p. 374 states
[SB42B], CP6A, CP6B, CP29B); P. progressus (localities original type specimen lost, but cast present in YPM col-
CC9A, CC9AA). lection).
Procynodictis sp. is also questionably known from lo- Characteristics: Mandible short, with robust, relatively deep
cality CP37B. symphyseal region (but less deep than in Tapocyon);
 Early Cenozoic Camivora ("Miacoidea") 119

premolars small, without accessory cusps; p2 = p3 size; Vulpavus sp. is also known from localities CP19,
trenchant talonids on all lower molars; upper molars lack (CP25C), CP27B, CP34B, ?CP38B, (NP14-couldbe Uin-
hypocones (primitive?), parastyle extends anteroextemally tacyon).
with distinct parastyle crest, metaconule weak; M3 small;
m3 small, single rooted, fiat crowned (Matthew, 1909a).
CARNIVORA INCERTAE SEDIS
Average length of ml ( U. massetericus): 7.1 mm.
Included species: U. edax (known from localities CP34C, lctidopappus Simpson, 1935
[CP64D]); U. acutus (locality CP6A); U. asodes (localities Type species: lctidopappus mustelinus Simpson, 1935.
[SB22C], [CP20A], [CP20B], [CP20C], CP25G, CP25H, Type specimen: USNM 9296, right dentary with p3-ml and
CP27D, CP27E, [CP64C]); U. bathygnathus (localities talonid of m2.
?CP34C, ?CP34D); U. jugulans (localities CP34C, Characteristics: lower premolars reduced, with simple trian-
?CP34D); U. major' (localities [CP5A], [CP32B], CP34C, gular crowns; p4 very distinctive: (1) short and triangu-
?CP34D); U. massetericus (Gingerich, 1983, considered lar in outline, (2) anterior accessory cusp small, high on
this species to be restricted to middle-late Wasatchian) (lo- tooth crown, far lingual, connected to main cusp by strong
calities SB24, [CP19? = U. rudis], [CPiOA? = U. rudis], ridge, (3) slight swelling of enamel lingual to main cusp
CP27B, [CP64A], [CP64B]); U. pugnax (locality CP38B); apex (?incipient "metaconid"), (4) talonid not basined,
U. rudis (Gingerich, 1983, considered all Clarkforkian- very short, wide, weakly cuspidate, with strong, trans-
earliest Wasatchian Uintacyon to be this species) (locali- verse, posterior cingulum; more open trigonid on ml than
ties CP17B, CP18B, CP19, CP20A, [CP25B]); U. vorax in other early Cenozoic feliforms; P4 highly modified for
(localities [CP18B], [CP34A], CP34C, CP65). shearing: (1) parastyle very large, high and integral part of
Uintacyon sp. is also known from localities SB25B, long shearing ridge, (2) protocone lobe relatively far pos-
[SB42B], ?CP20A, CP27B, CP31E, CP64C, [NP14 - terior, anteroposteriorly narrow, (3) camassial (metastyle)
could be Vulpavus]. blade sharp, but shallow and broad (only known P4 broken
at center of metastyle blade, so it is impossible to deter-
Vassacyon Matthew, 1909b mine ifthere is a vertical, slitlike notch as in all other early
Type species: Vassacyon promicrodon (Wortman and Cenozoic Camivora); small P3 protocone present, far an-
Matthew, 1899) (= Uintacyon promicrodon). terolingual to paracone apex (Simpson, 1935; Macintyre,
Type specimen: AMNH 83, dentary with p4-ml, partial p3 1966; Flynn and Galiano, 1982; Gingerich and Winkler,
and m2. 1985).
Characteristics: Upper molars with short parastyle, well- Average length of ml: 4.1 mm.
developed hypocone; ml with large, broadly basined talo- Included species: /. mustelinus only, known from locality
nid; m2 large; m2 talonid trenchant (otherwise similar to NP19C only.
Uintacyon; Matthew, 1909a, 1915). lctidopappus sp. is also questionably known from local-
Average length of ml: 7.93 mm. ity SB23A.
Included species: V. promicrodon only (known from locali-
ties CP20A, CP20B, CP20C, CP25D).
INDETERMINATE EARLY CARNIVORANS
Vassacyon sp. is also questionably known from localities
CP27D, CP27E. Fragmentary remains assigned to "miacoid indet." have been re-
ported from localities CC4, CC5, CC6B, CC8, ?CC50, CP20E,
Vulpavus Marsh, 1871 CP25K.
Type species: Vulpavus palustris Marsh, 1871.
Type specimen: YPM 11841, isolated right Ml.
Characteristics: Upper molars rounded, quadrate in outline, BIOLOGY AND EVOLUTIONARY PATTERNS
cusps low (but less than in Palaearctonyx); shearing re-
duced or absent on upper molars; "P4 not camassiform" Little work has been done on the locomotory specializations of
(Matthew, 1909a, p. 379); lower molar trigonids low, talo- early Cenozoic Camivora, primarily because of a paucity of well-
nid basin as broad or broader than trigonids (Matthew, preserved, complete postcrania. There is evidence in femur anatomy
1909a; Gingerich, 1983). indicating ambulatory locomotion in one Paleocene form (Protictis
Average length of ml (V. australis): 5.7 mm. haydenianus), and more cursorial specializations in some later taxa
Included species: V. palustris (known from locality CP34C); (Vulpavus palustris, Miacis parvivorus, Vassacyon promicrodon,
V. australis (localities SB24, CP3, [CPI 9), CP20A, CP20B, Jenkins and Camazine, 1977; see also Savage, 1977, Table 5 and
CP20C, [CP25E], CP27C, [CP64C]); V. canavus (local- p. 250). Scansorial or arboreal locomotion has been suggested for
ities SB22C, CP20A, CP20C, CP25F, CP25G, CP25H, some other early Cenozoic Camivora (e.g., Springhom, 1982, 1985;
[CP27B], CP27D, CP27E, CP64C); V. ovatus (localities possibly in Oodectes herpestoides, Jenkins and Camazine, 1977).
[CP34A], CP34C); V. profectus (locality CP34C). Savage (1977) suggested that skeletons in carnivorous mammals
 John J. Flynn 120

w
z L L.BI.
1.8
- -
w E.81
2.4~
(.)
4.6 ..._
0 E Lst. Hp.
52
:::i ~ 6.0 1--
Q. L.E.Hp.
7.0 t--
E.E.Hp.
8.8 1--
L L.CI.
9.5 ~

E.CI.
11.0 1--
w
z LL.Ba.
w E.L.Ba.
12.5 .._

(.) M
0 14.0I--

:ii E.Ba.

15.8 1--
L.Hm.
17.5 ~
E.Hm.
18.BI--
E L.L.Ar.
19.2 1--

w
-E.L.Ar.

L.E.Ar.
23.0

z
W L
(.) 27.7 I - - - - - - -
0
CJ E.E.Ar.

:::i 29.4
0 "MIACIDAE" CARNIVORA
Wt.
"VIVERRAVIDAE"
E 31.9 (Feliformia) (Caniformia) INCERTAE SEDIS
Or.
,___ 33.4
L.Ch.
34.5 ,__
M.Ch.
35.5
,__ c: -

=~
E.Ch.
L 37.1 - - -
Du

39.5 ' - - iii'

i'li-
L,Un.
,__ -
w
z
w E.Un.
41.3

:s
~
(.) 45,9 ,__ ...... -
0 ....0
w M
L.Br.
46.7 ,__
,_ e:. - - - -
M.Br.
~
49.0
- - ~ - ?l
- -
E.Br.

- - ~

~~ ..~ ~O~ i ~ ~~
50,4

·- ,.,_ -~~ ___.

L.Wa,

e.wa.
E.E.Wa.
54.2 . _
55
·0 -
......
,__
-
-
-

I.;, - •!!! Q)
- ~ jl
1-__:..;:~L..;::__ _ _ _ _ _ _ _ _~-=-r-----'------::--;;;-j
~~:: "'-.~..._ -e:...-----(3--:.~ - - ~
w g~2, ~:-~ ~ ='tl :=.!!!-=,Ill ~ ·~ -:S .~
Z Tl.6 ss:2 •!:: -·--~ ..... -.;:: .~ Cl) C:
c:111 _-fS--.~-.U1- -S:I ~ !g :;::
'.!!!
e-
W
o
ow
..J
L

...
n.s
Tl.4

i::~
Tl.1
-
57.o
57.3

58.5
59 2
· EQ 1-Q
::.... ..... I.; - .::
~ -l! :: ~c: -
-
I - -
..-
0..-
e:
_

::...._ •(/)-. I= a. _
111
Q. - ':;::
l!;! ,-.~ ~I
--c:-
;:. I

- ? I- .I- - ,_
60,9-- - Q
ct To.3 61.3 -
Q. To2 62.5 - - • - - - 'tJ ,- ~ -
E To.1 63 _8 _ _ L _ ? ':;:: - Ill -
~ ?0:: L
Pu.3 6 4_0 _ _ _ _ • .L
1

~~~ =~~=====
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphilllan, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
wt. =Whitneyan, or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 6.6. Temporal ranges of North American early carnivoran genera.

 Early Cenozoic Camivora ("Miacoidea") 121

should be strong, flexible, and adaptable because of variability in icant shearing is focused on the ml trigonid and the P4-Ml embra-
locomotion during prey location, capture, and kill. sure (posterior margin of P4, anterior margin of Ml), leaving the
Radinsky (1971, 1977, 1978) analyzed brain size and morphol- remainder of the cheek tooth dentition free for a variety of modifi-
ogy in "carnivores" from several perspectives. Living carnivorans all cations. Living Carnivora exhibit a wide spectrum of dental adap-
possess a cruciate sulcus of the cerebral cortex, which is not present tations, ranging from hypercarnivory (e.g., Felidae, Mustelinae), in
in the early Cenozoic "Miacoidea." However, Radinsky (1971) used which carnassial shear is emphasized and the postcarnassial molars
the fossil record of carnivoran brain evolution to show that this are reduced, to hypocarnivory (e.g., Ursidae, Procyonidae) in which
feature evolved independently at least five to seven times. Parallel the postcamassial molars are enlarged and shear on the carnassials is
development of the cruciate sulcus may have been a result of fold- reduced. Many taxa (e.g., Canidae) retain both the shearing function
ing necessitated by general trends of increase in cerebral neocor- of the carnassials and the crushing function of the postcarnassial mo-
tex (particularly motor and somatosensory cortex) throughout the lars. Still others (e.g., Hyaenidae) expand carnassial shear, but also
Carnivora (including trends within "miacoids"). Radinsky (1977) enlarge the premolars for heavy-duty bone crushing. Early Cenozoic
refuted the common belief that larger brains and greater intelligence Carnivora generally retained both carnassial shear and molar crush-
accounted for the evolutionary success of"miacoids" (and their "de- ing, but a diversity of specializations are apparent even within that
scendants"), and lower diversity and extinction of other contempo- group (e.g., compare dentitions of Viverravus, Miacis, and Vulpavus
rary carnivorous mammals (e.g., Creodonta, Mesonychidae). "Mia- in Figures 6.3 and 6.4). Van Valen (1969, p. 127) noted, "Recent
coids" and contemporary carnivorous mammals had similar relative Carnivora with teeth more or less similar to those of miacids gener-
brain sizes, and both groups showed trends of increase in brain size ally triturate their food after slicing it, and undoubtedly most mi acids
and complexity through time (although the pattern of neocortex ex- did the same."
pansion was different between the groups). Radinsky (1978) also Much discussion has been generated regarding competition be-
presented evidence contradicting previous suggestions that fossil tween Carnivora and other contemporary carnivorous mammals (fo-
and living "carnivores" had relatively larger brains than contempo- cusing on Creodonta and Mesonychidae ). Such hypotheses are so
rary ungulates, and that brain size in both increased through time as tempting because of the inverse relationship in diversity between
a feedback selection response between predator and prey. the two groups (e.g., steady decline in creodont diversity, and ulti-
Living carnivorans exhibit wide variation in skull shape. Radinsky mate extinction, through the Cenozoic, with increased diversity of
(1981a, 198lb, 1982) analyzed skull shape in living and fossil carnivorans during the early Cenozoic and rapid radiation of
Carnivora. Initial indications of tight clustering in skull shape "modem" carnivorans during the middle Cenozoic). Savage (1977,
within living families (Canidae, Viverridae, Mustelidae, Felidae; Table 4) showed that Creodonta were much more diverse than Car-
Radinsky, 198la) and significant differences between families de- nivora in North America during the Paleocene (18.4 vs. 2.6 per-
teriorated with broader sampling of families (Ursidae, Procyonidae) cent of taxa in North American faunas) and Eocene (14.0 vs. 5.3
and aberrant tax.a within the original groupings (Radinsky, 198lb). percent). This pattern was reversed during the Oligocene (2.6 vs.
Addition of fossil representatives of the living families further indi- 13.2 percent), and Creodonta went extinct in North America during
cated that early members of these families were much more similar the Miocene while Carnivora flourished. Although one cannot ex-
to each other in skull shape than are the living representatives. clude models of competitive replacement between these groups, no
Analysis of "functionally significant aspects of cranial morphol- independent evidence suggests that early Carnivora were compet-
ogy" in fossil Carnivora (Radinsky, 1982) showed no "key inno- itively "superior" to the Creodonta. Both groups were comparable
vations" in the origin and radiation of modern carnivoran families in sensory apparatus (with creodonts probably possessing superior
from the early Cenozoic "miacoids." Although adequate cranial ma- stereoscopic vision), relative brain size, skeletal morphology, loco-
terial was available only for early caniforms (Vulpavus, Oodectes, motion, and dental specialization for carnivory (although, as dis-
Uintacyon, and Procynodictis), Radinsky (1982) proposed the fol- cussed earlier, "miacoids" had begun to diversify their dentitions
lowing aspects of skull shape were primitive for Carnivora: ( 1) rela- into hypercarnivorous, hypocamivorous, and intermediate mixed
tively great zygomatic arch width, (2) relatively great temporal fossa shearing/crushing types).
width, (3) relatively long tooth rows, (4) relatively long jaw length, Further, these models do not adequately explain the rapid radia-
(5) relatively forwardly placed carnassials, and probably related to tion of "modern" Carnivora at least 20 million years after the origin
this, relatively high moment arm of the masseter, (6) relatively small of the order. Radinsky (1982) could not identify any obvious "key
brain size, and (7) remaining features probably similar to those in innovations" of the cranium, skeleton, soft anatomy, or presumed
modern viverrids, except bullar volume (unossified in "miacoids," ecology that might explain the radiation of the modern carnivoran
and thus not measurable), and moment arm of temporalis and jaw families. He suggested that body size differences and partitioning
cross-section area (too diverse within "miacoids" to be considered of prey resources might have been a factor in the diversification of
characteristic of the entire group of taxa). Savage ( 1977) indicated modern carnivorans. Interestingly, "miacoids" generally were sig-
that "Miacidae" had much lower degrees of stereoscopic vision than nificantly smaller than creodonts. Both creodonts and carnivorans
do living Carnivora. showed sharp drops in diversity by the Uintan, with representatives
Dentitions in all Carnivora are highly specialized. Perhaps the ofliving camivoran families appearing only later in the fossil record.
most significant dental modification has been the evolution of P4/ml The Bridgerian and Uintan began a period of sharp climatic change
as the only carnassial pair (Flynn, Neff, and Tedford, 1988). Signif- in North America, and globally, that peaked dramatically during
 John J. Flynn 122

the Oligocene. These climatic changes were accompanied by major Fox, R. C., & Youzwyshyn, G. P. (1994). New primitive carnivorans (Mam-
drops in global sea level and interconnection of geographic areas malia) from the Paleocene of Western Canada, and their bearing on
relationships of the order. Journal of Vertebrate Paleontology, 14,
that had been separated previously (e.g., disappearance of the Turgai
382-404.
Straits led to the connection of Europe and Asia). A complex pat- Gingerich, P. D. (1983). Systematics of early Eocene Miacidae (Mammalia,
tern of origination of new taxa in areas of endemism isolated by the Carnivora) in the Clark's Fork Basin, Wyoming. Contributions from
climate/sea-level change, regional extinction, and significant fauna! the Museum of Paleontology, University of Michigan, 26, 197-225.
immigration and interchange may explain equally well the patterns Gingerich, P. D., & Winkler, D. A. (1985). Systematics of Paleocene Viver-
ravidae (Mammalia, Carnivora) in the Bighorn Basin and Clark's
of diversification and extinction in Carnivora and Creodonta.
Fork Basin, Wyoming. Contributions from the Museum of Paleon-
tology, University of Michigan, 27, 87-128.
Gustafson, E. P. ( 1986). Carnivorous mammals of the late Eocene and early
Oligocene of Trans-Pecos, Texas. Texas Memorial Museum Bulletin,
ACKNOWLEDGMENTS
33, 1-66.
Hay, 0. P ( 1902). Bibliography and catalogue of the fossil Vertebrata of
Thanks are due to Henry Galiano and Richard Tedford for their early North America. Bulletin of the U.S. Geological Survey, 179, 1-868.
encouragement to work on this fascinating group of mammals, and Jenkins, F. A., Jr., & Camazine, S. M. (1977). Hip structure and locomotion
to Christine Janis for her invitation to contribute to this volume in ambulatory and cursorial carnivores. Journal of Zoology, London,
181, 351-70.
and her incredible persistence in bringing it to completion. Marlene
K.retzoi, M. (1943). Kochitis centenii n.g. n.sp., ein altertumlicher Creodonte
Donnelly and Clara Simpson (The Field Museum) provided the
aus dem Oberoligozan Siebenburgens. Foldtany Kozlony, 52, 10-7,
wonderful illustrations. 190--5.
(1945). Bemerkungen uber das Raubtiersystem. Anna/es Historico-
Naturales Musel Nationalis Hungarici, Budapest, 38, 59-83.
Leidy, J. (1872). Remarks on fossils from Wyoming. Proceedings of the
REFERENCES Academy of Natural Sciences of Philadelphia, 1872, 277.
MacGinitie, H. D., Rohrer, W. L., & Leopold, E. B. (1974). An early mid-
Beard, C., Dawson, M. R., & Tabrum, A. R. (1995). First diverse land mam- dle Eocene flora from the Yellowstone-Absaroka volcanic province,
mal fauna from the early Cenozoic of the southeastern United States: northwestern Wind River Basin, Wyoming. University of California
the early Wasatchian Red Hot Local Fauna, Lauderdale County, Mis- Publications in Geological Sciences, 108, 1-103.
sissippi. Journal of Vertebrate Paleontology, 15 (suppl. to no. 3), Macintyre, G. T. (1962). Simpsonictis, a new genus of viverravine miacid
p. 18A. (Mammalia, Carnivora). American Museum Novitates, 2118, 1-4.
Bowdich, T. E. (1821 ). An Analysis of Natural Classification of Mammalia (1966). The Miacidae (Mammalia, Carnivora), Part 1: The systematics
for the Use of Students and Travellers. Paris: J. Smith. of Ictidopappus and Protictis. Bulletin of the American Museum of
Bryant, H. N. (1992). The Camivora of the Lac Pelletier Lower Fauna Natural History, 131, 115-210.
(Eocene: Duchesnean), Cypress Hills Formation, Saskatchewan. Marsh, 0. C. (1871). Notice of some new fossil mammals and birds from
Journal of Paleontology, 66, 847-55. the Tertiary formations of the West. American Journal of Science,
Commission on Zoological Nomenclature. (1988). Opinion 1517, Viver- series 3, 2, 120-7.
ravus gracilis Marsh, 1872 (Mammalia, Carnivora): generic and spe- (1872). Preliminary description of new Tertiary mammals, Part I. Amer-
cific names conserved. Bulletin of Zoological Nomenclature, 45(3), ican Journal of Science, series 3, 4, 122-8.
248. Matthew, W. D. (1899). A provisional classification of the freshwater Tertiary
Cope, E. D. (1872). Third account of new Vertebrata from the Bridger Eocene of the west. Bulletin of the American Museum of Natural History,
of the Wyoming Territory. Proceedings of the American Philosoph- 12, 19-75.
ical Society, 12, 469-72. (1909a). The Carnivora and Insectivora of the Bridger Basin, middle
(1874). Report upon vertebrate fossils discovered in New Mexico, with Eocene. Memoirs of the American Museum of Natural History, 9,
descriptions of new species. Annual Report ofthe Chief ofEngineers, 291-567.
1874, Geographical Explorations and Surveys West of the One Hun- (1909b). Fauna! lists of the Tertiary Mammalia of the West. Bulletin U.S.
dredth Meridian, Extract from Appendix FF, 4-18. Geological Survey, 361, 91-138.
( 1875). Systematic catalogue of Vertebrata of the Eocene of New Mexico, (1915). A revision of the lower Eocene Wasatch and Wind River faunas,
collected in 1874. Report to the Engineer Department, U.S. Army Part 1: Order Ferae (Carnivora), suborder Creodonta. Bulletin of the
Geographic Exploration Survey of the Territories, 1-37. American Museum of Natural History, 34, 4-103.
(1880). On the genera ofCreodonta. Proceedings of the American Philo- (1937). Paleocene faunas of the San Juan Basin, New Mexico. Transac-
sophical Society, 19, 76--82. tions of the American Philosophical Society, new series, 30, 1-510.
(1882). Synopsis of the Vertebrata of the Puerco Eocene Epoch. Pro- Matthew, W. D., & Granger, W. (1915). A revision of the lower Eocene
ceedings of the American Philosophical Society, 20, 461-71. Wasatch and Wind River faunas. Bulletin of the American Museum
Dawson, M. R. (1980). Geology and paleontology of the Badwater Creek of Natural History, 34, 1-103.
area, central Wyoming. Part 20: the late Eocene Creodonta and Car- Radinsky, L. (1971). An example of parallelism in carnivore brain evolution.
nivora. Annals of the Carnegie Museum, 49, 79-91. Evolution, 25, 518-22.
Flynn, J. J., & Galiano, H. (1982). Phylogeny of Early Tertiary Carnivora, (1977). Brains of early carnivores. Paleobiology, 3, 333-49.
with a description of a new species of Protictis from the Middle (1978). Evolution of brain size in carnivores and ungulates. The American
Eocene of northwestern Wyoming. American Museum Novitates, Naturalist, 112, 815-31.
2725, 1-64. (198 la). Evolution of skull shape in carnivores, 1: Representative modern
Flynn, J. J., Neff, N. A., & Tedford, R.H. (1988). Phylogeny oftheCarnivora. carnivores. Biological Journal of the Linnean Society, 15, 369-88.
In The Phylogeny and Classification of the Tetrapods, Vol. 2: Mam- (1981 b). Evolution of skull shape in carnivores, 2: Additional moder.1
mals, ed. M. J. Benton, Systematics Association Special Volume No. carnivores. Biological Journal of the Linnean Society, 16, 337-55.
35B, pp. 73-116. Oxford: Clarendon Press. (1982). Evolution of skull shape in carnivores, 3: The origin and
 Early Cenozoic Camivora ("Miacoidea") 123

early radiation of the modern carnivore families. Paleobiology, 8, Stock, C. (1934). New Creodonta from the Sespe upper Eocene, California.
177-95. Proceedings of the National Academy of Sciences, 20, 423-7.
Savage, R. J. G. (1977). Evolution in carnivorous mammals. Palaeonto/ogy, (1935). Plesiomiacis, a new Creodont from the Sespe Upper Eocene,
20, 237-71. California. Proceedings of the National Academy of Sciences, 21,
Simpson, G. G. (1935). New Paleocene mammals from the Fort Union of 119-22.
Montana. Proceedings of the U.S. National Museum, 83, 221-44. Teilhard de Chardin, P. (1915). Les carnassiers des Phosphorites de Quercy.
(1937). The Fort Union of the Crazy Mountain Field, Montana, Anna/es Paleontologie (Paris), 9, 103-92.
and its mammalian faunas. U.S. National Museum Bulletin, 169, Thorpe, M. R. (1923). Notes on the Bridger (Eocene) Carnivora. American
1-287. Journal of Science, 5, 23-39.
(1945). The principles of classification and a classification of mammals. Trouessart, E. L. (1885). Catalogue des Mammiferes vivants et fossiles,
Bulletin of the American Museum of Natural History, 85, 1-350. Fascicule IV: Carnivores (Carnivora). Bulletin de la Societe Etudes
Springhorn, R. (1982). Neue Raubtiere (Mammalia: Creodonta et Car- Sciences Angers, 15, 1-108.
nivora) aus dem Lutetium der Grube Messel (Deutschland). Van Valen, L. (1969). The multiple origins of the placental carnivores. Evo-
Palaeontographica Abteilung A: Paliiozoologie-Stratigraphie, 179, lution, 23, 118-30.
105-41. Wortman, J. L. (1901). Studies of Eocene Mammalia in the Marsh collec-
Standhardt, B. R. (1995). Early Paleocene (Puercan) vertebrates of the Do- tion, Peabody Museum. American Journal of Science, series 4, 11,
gie Locality, Big Bend National Park, Texas. In National Park Ser- 333-48,437-50; 12, 143-54, 193-206,281-96,377-82,421-32.
vice Paleontological Research, eds. V. L. Santucci & L. McClelland, Wortman, J. L., & Matthew, W. D. (1899). The ancestry of certain mem-
46-48. National Park Service Technical Report NPS/NRPOINRTR- bers of the Canidae, the Viverridae, and Procyonidae. Bulletin of the
95/16. American Museum of Natural History, 12, 109-39.
7 Canidae
KATHLEEN MUNTHE

INTRODUCTION

The Canidae is a widespread and fairly diverse family of carni-

vores, Recent members of which include foxes, jackals, wolves, and
their close relatives. Recent canids inhabit every continent except
Antarctica and occupy habitats as different as the Sahara Desert
and the Amazonian rain forest. They vary greatly in size, dietary
habit, and social organization. The current wide geographical dis-
tribution enjoyed by canids is a relatively recent phenomenon. Prior
to the latest Miocene (late Hemphillian), the family was restricted
to North America, where its members were more diverse in size
and probable habits than are extant canids. Canids first appeared
in the late middle Eocene (Duchesnean) of North America. Their
subsequent evolution included three major successive and partly Figure 7.1. Restoration of Epicyon saevus, by Brian Regal.
overlapping adaptive radiations, a pattern reflected in a tripartite
subdivision of canids into three subfamilies: Hesperocyoninae,
Caninae, and Borophaginae (see Wang, 1994). Of the numerous bones. The stapedial branch of the internal carotid artery has been
genera of Tertiary canids described in the literature, approximately lost; the promontory branch has shifted medially and is extrabullar
three dozen are valid. Another dozen or so are as yet undescribed. in position, imbedded between the entotympanic and the petrosal.
In early canids a suprameatal fossa, smaller than in contemporary
arctoids, is present anterior to the mastoid process. In many canids
DEFINING FEATURES OF THE FAMILY CANIDAE the base of the paroccipital process is elongate and buttresses the
caudal entotympanic. Finally, the canid malleus is distinctive in size
and in the anterior curvature of its manubrium.
CRANIAL A number of other characters, taken together, are descriptive of
Several cranial characters can be reliably used to separate canids the canid skull, although none is particularly unique to the family.
from other carnivoran clades (see Figure 4.1). Most are characters In general shape, the skull (Figures 7.2A, B) is rather long, with a
of the basicranium, particularly of the auditory region, as identified tapering snout and usually strong zygomatic arches. Large temporal
by Hunt (1987), Flynn, Neff, and Tedford (1988), Wang and Ted- fossae are bordered dorsally by a sagittal crest or, in a few species,
ford (1994), and Wang (1994), and as listed later. The auditory bulla by paired temporal lines. In larger species, a prominent external
is fully ossified in canids and is often rather inflated. It is formed occipital protuberance forms the apex of the triangular occiput. The
primarily by the entotympanic, but with some contribution by the mastoid process is small. Premaxillae, maxillae, and palatine bones
ectotympanic. Except in the earliest known canid, Prohesperocyon, all contribute to the hard palate, which terminates at about the level
the canid bulla is unique in having a partial, anteromedial, intrabullar of the last upper molars. The orbit is open laterally and bordered
septum that is formed entirely by the inflected dorsal edge of the by a moderately developed postorbital process. Three foramina are
caudal entotympanic. The petrosal is flattened and expanded medi- aligned anteroposteriorly in the medial wall of the orbit: the optic
ally and is in solid contact with the basioccipital and basisphenoid canal, the orbital fissure (anterior lacerate foramen), and the anterior

124
 Canidae 125

Figure 7.2. Skulls of selected Tertiary canids: A. Hesperocyon gregar-

ius, scale bar = 10 mm (by Brian Regal, after Scott and Jepsen, 1936).
B. Tomarctus rurestris, scale bar= 20 mm (from Merriam, 1913).

opening of the alisphenoid canal. The foramen rotundum opens into

the latter. The posterior opening of the alisphenoid canal is in a fossa
with, or is adjacent to, the foramen ovale, medial to the glenoid fossa.
The anterior carotid foramen lies adjacent to the median lacerate
foramen at the anteromedial edge of the bulla; the posterior carotid
foramen lies iµ a fissure in common with the posterior lacerate
foramen, along the posteromedial edge of the bulla. A prominent Figure 7.3. Dentition of Osteoborus cyonoides: A. Upper dentition in lat-
postglenoid foramen is present, and there is a distinct condyloid eral (top) and occlusal (bottom) views (from VanderHoof and Gregory,
foramen posterior to the auditory bulla. 1940). B. Lower dentition in lateral (top) and occlusal (bottom) views (from
VanderHoof and Gregory, 1940). All scale bars= 10 mm.
The dentary in canids has a prominent coronoid process, a trans-
versely elongate condyloid process, and a hooked angular process.
In a few species a subangular process is additionally present. In most canids, the incisors have small posterior heels and, in
some taxa, I3 is enlarged. Several species possess complex incisors,
with lateral and medial accessory cusps. As in most carnivorans,
DENTAL
moderately long to long canine teeth are present. The anterior pre-
The primitive dental formula in canids is 13/3, Cl/l, P4/4, M2/3 = molars are usually trenchant (bladelike), but their size relative to the
42 (see Figure 7.3). Cuon (Asian dhole), Lycaon (African hunting molars, their degree of crowding, and their complexity vary consid-
dog), Speothos (South American bush dog), and some fossil species erably within the family.
have fewer molars and reduced dental formulae; Otocyon (African The upper carnassial tooth, P4, is an elongate blade in most
bat-eared fox) has more molars. taxa and, in some, bears an anterior accessory cusp, the parastyle.
Canid dental characters are of less diagnostic value at the family In general, the elongate lower carnassial, ml, bears a prominent
level than is the basicranium, but Wang and Tedford (1994) iden- protoconid, less prominent but usually well-developed paraconid,
tified several derived dental characters for the Canidae: Posterior and small metaconid. In many canids, the talonid is bicuspid, with
accessory cusps are present on both P3 and p3 in canids, and the hypoconid and entoconid subequal, but a trenchant talonid seems to
cingular cusps on these teeth are better developed than in rniacids; be primitive for the family and appears within several lineages.
the parastyle on Ml is reduced, although variable in size and shape; The upper molars of canids are generally similar to those of many
M3 is lost; the metaconid of m2 in canids is. enlarged and subequal other caniform carnivores and have led to some taxonomic confu-
in size to the protoconid. Beyond this, dental morphologies within sion historically. They are wider than long and bear three prominent
the Canidae are quite diverse and are useful primarily in assigning cusps; conules are variably developed and, in some species, acces-
taxa at the subfamily level and below. sory cusps are present. Although many caniform carnivores increase
 Kathleen Munthe 126

the size of their upper molars, canids apparently do so in a unique levels. In general, the vertebral column is rather long and flexible
way (Flynn, Neff, and Tedford, 1988). The posterior portion of the and includes a long tail. The clavicle is small, oval, and at least
internal cingulum is enlarged to form a hypocone, which occludes partly cartilaginous; it does not articulate with either the scapula
with a prominent entoconid on the lower molars. or the sternum. Limb bones are generally long and slender, but
Most canids are mesocarnivorous (their diets consist of both ver- proportions and robustness vary within the family. Except in the
tebrate flesh and other materials), but hypercarnivores (those whose earliest canids, the feet are digitigrade, with blunt, nonretractile
diets consist almost entirely of vertebrate flesh) have evolved conver- claws. The metapodials are elongate and closely appressed, and the
gently several times within the family. Hypercarnivorous canids ex- scaphoid and lunate are solidly fused. Digit I (pollex) of the manus
hibit modified dentitions (Flynn, Neff, and Tedford, 1988; Stevens, is greatly reduced (it is lost in Lycaon, the African hunting dog) and
1991; Van Valkenburgh, 1991) similar to those of other hypercar- digit I (hallux) of the pes is lost. The baculum is well developed and
nivores, which forms the basis for some taxonomic confusion. No- grooved. The earliest canids are primitive with regard to several of
tably, carnassial blade length is increased and the metaconid of m 1 is these characters.
reduced or lost as transverse shear functions are reduced. The inter-
nal cingulum and hypocone on the upper molars are suppressed, and
a monocuspid, trenchant talonid is produced on the lower molars SYSTEMATICS
through reduction of the entoconid and enlargement of the more me-
dially placed and bladelike hypoconid. Other modifications include
SUPRAFAMILY
smaller and/or fewer postcarnassial molars and the simplification of
their crowns. The Order Carnivora has been recognized as a natural grouping
Hypocarnivorous canids (those whose diets include a low per- since at the least the time of Linnaeus (1758). By the mid-1800s,
centage of vertebrate flesh) also have modified dentitions (Wang the Canidae (Gray, 1821) and all other extant families of carnivores
and Tedford, 1992). The blade of P4 is shorter, and Ml is larger had been placed within the order, but there was little agreement as
relative to the size of that tooth. The upper molars are quadrate, to how these families were interrelated. A breakthrough came when
often bearing prominent conules and accessory cusps. On the lower Turner (1848) pointed out that parallelisms existed among the dental
molars, the trigonid is lower and the talonid is broader than in meso- and postcranial characters used in classification schemes up to that
carnivorous canids. time. He advocated use of what he termed "essential characters,"
such as basicranial foramina, in determining relationships. Turner
presented a classification of the Carnivora in which terrestrially
POSTCRANIAL adapted families (pinnipeds will be ignored henceforth in this dis-
The canid postcranial skeleton (Figure 7.4) is of little diagnostic cussion) were divided into three subgroups: ( 1) felids, hyaenids, and
value at the family level but can be useful at subfamily and generic viverrids; (2) ursids, procyonids, and mustelids; and (3) canids.
Flower (1869) further studied the basicranium, particularly the
auditory region, of living carnivores and proposed a tripartite clas-
sification like that of Turner (1848). Flower's Aeluroidea (felids,
hyaenids, and viverrids) was defined by presence of a septum that
divided the auditory bulla into anterior and posterior chambers.
Members of the Arctoidea (ursids, procyonids, and mustelids) were
defined by absence of a septum dividing the bulla, whereas mem-
bers of the Cynoidea (canids) possessed a low, partial bullar sep-
tum. Flower also cited characters of the paroccipital process and
basicranial foramina to support his separation of canids from other
carnivores. Later studies, such as Mivart's (1890) monograph on the
Canidae and Segall's (1943) review of the auditory region in certain
carnivores, lent credence to Flower's scheme. Segall, in fact, stated
that, "The Canidae form a more compact group with reference to
the structure of the auditory region than does any other family."
Nonetheless, there was a subsequent tendency to group canids with
the Arctoidea.
Neontologists quickly accepted the basicranium of carnivores as
a reliable phylogenetic tool, but paleontologists were slow to fol-
low suit. Even when adequate materials were available, most fossil
carnivores were assigned to families on the basis of dental
characters. This was especially true of numerous carnivores with
Figure 7.4. Skeletons of selected canids: A Hesperocyon gregarius (from
Scott and Jepsen, 1936); scale bar = approximately 50 mm. B. Canis
canidlike dentitions. It was not until relatively recently that the bas-
(medium-sized domestic dog) (from Evans and deLahunta, 1971); scale bar icranium of fossil carnivores received much attention (see references
=approximately 100 mm. in Hunt, 1974).
 Canidae 127

When Simpson (1945) reviewed the classification of the Car- small, primitive arctoid carnivores (e.g., Parictis, Amphicynodon,
nivora, he noted that the canid-ursid-procyonid and viverrid-hyaenid Pachycynodon), abundant in the Quercy deposits of France but rare
groupings were best substantiated, and that felids and mustelids in the Oligocene of North America. Clark and Guensburg (1972)
seemed to hold somewhat more isolated positions. He followed evaluated the dental, basicranial, and postcranial characters of these
then current thought, however, in utilizing the term "Canoidea" genera and pronounced them caniform. Baskin and Hunt (this vol-
for the combined Cynoidea and Arctoidea of Flower (1869), and ume, Chapters 8 and 10, respectively) view amphicynodontines as
the term "Feloidea" for Flower's Aeluroidea. Some authors, for primitive ursoids. Tedford, Barnes, and Ray (1994) tentatively raised
example, Romer (1966), later reverted to use of the terms "Arc- the group to family status and included within it the apparently
toidea" and "Aeluroidea," for Feloidea and Canoidea, respectively. marine carnivorans, Allocyon and Kolponomos. Allocyon was orig-
Simpson's (1945) choice of the term "Canoidea" reflected his idea inally described as an aberrant canid by Merriam (1930).
that canids were ancestral to ursids and procyonids. In fact, he in- Amphicyonines (e.g., Amphicyon) were long considered bear-
cluded within the Canidae some small doglike carnivores (e.g., Cyn- like canids. Members of this group show convergences toward both
odictis, Cephalogale) from the early Tertiary of Europe and several canids and ursids, but belong to neither family. The Amphicyonidae
bearlike taxa (e.g., Hemicyon, Dinocyon), which he considered in- is discussed as a separate family of caniform carnivores by Hunt
termediate between canids and ursids. (this volume, Chapter 11).
Simpson's (1945) canoid-feloid dichotomy has much to recom- Following Matthew (1924, 1930), Simpson (1945) separated
mend it, for it separates families with primitive, essentially single- "typical" canids into two subfamilies. Those with a trenchant talonid
chambered bullae (Canoidea) from families with derived bullae that on ml were classified within the Simocyoninae; all other gen-
were already two chambered at their first appearance in the early era were grouped within the Caninae. Simpson included three ex-
Oligocene (Feloidea). Other basicranial characters diagnostic of the tant genera within the Simocyoninae: Cuon (Asian dhole), Lycaon
various carnivore families were also present in their Oligocene rep- (African hunting dog), and Speothos (South American bush dog),
resentatives, but the striking differences among adult members of thought to be relicts of a group widespread in the Tertiary. Mor-
living families, on which Flower (1869) based his tripartite classi- phological evidence (Berta, 1988) and genetic studies (Wayne and
fication, are the result of post-Oligocene evolution. O'Brien, 1987), however, indicate a close relationship among these
Relationships among the Carnivora were examined more recently three genera and other extant canids and support their classification
by Flynn, Neff, and Tedford (1988), who evaluated numerous char- within the subfamily Caninae. A trenchant talonid has evolved con-
acter states within the order and subjected them to cladistic analysis. vergently within several lineages of carnivores, including canids, as
Initially, this analysis was confined to extant families, and the resul- a mechanism to process meat more efficiently (Van Valkenburgh,
tant phylogeny reaffirmed the basic dichotomy within the Carnivora, 1991), and a taxonomic grouping based on this character is not
for which the authors utilized the terms "Feliformia" and "Cani- supported by other lines of evidence.
formia." A further dichotomy within the Caniformia, between the Extinct genera placed in the Simocyoninae by Simpson (1945)
Cynoidea (including Canidae only) and the Arctoidea (including include Enhydrocyon, Philotrox, Euoplocyon, Protocyon, Temno-
Ursidae, Procyonidae, Mustelidae, and pinnipeds), was also reaf- cyon, Simocyon, and Brachyrhynchocyon. The first two genera are
firmed. primitive canids closely related to Mesocyon, as discussed later. Eu-
Addition of extinct carnivore groups to this analysis complicated oplocyon was assigned to the subfamily Borophaginae by Tedford
this seemingly simple picture. The possible relationships of such and Frailey (1976); Protocyon is a member of the subfamily Can-
"problem" groups as miacoids, nimravids, amphicyonids, and am- inae (Berta, 1987). Except for the species of Temnocyon now as-
phicynodontines to other carnivores were analyzed by Flynn, Neff, signed to the new canid genus Parenhydrocyon, and the species of
and Tedford (1988) and are discussed elsewhere in this volume (see Brachyrhynchocyon now assigned to the new canid genus Osborn-
summary chapter on Carnivorous Mammals, Chapter 4). None of odon or to Cynodesmus (Wang, 1994), other extinct genera listed
the alternative resultant phylogenies, however, alter the basic posi- as simocyonines by Simpson are not canids. True simocyonines
tion of the Canidae within the Carnivora or the relatively isolated (e.g., Simocyon) have been included within the family Procyonidae
position of canids among the Caniforrnia, which are depicted in (Baskin, this volume, Chapter 8); Brachyrhyncocyon (in part) and
Figure 4.1. related taxa have been removed to the family Amphicyonidae (Hunt,
this volume, Chapter 11).
The Caninae of Simpson (1945) included extant canid genera
INFRAFAMILY
(Canis, Urocyon, Vulpes, Alope.x, Fennecus, Nyctereutes, and sev-
The most recent, comprehensive classification of the Canidae is eral South American genera) and their close fossil relatives (Lep-
that of Simpson (1945), who recognized six subfamilies: Amphi- tocyon, Theriodictis). Several of the genera in Simpson's Caninae
cynodontinae, Amphicyoninae, Simocyoninae, Caninae, Otocyoni- have been lost through synonymy. Aenocyon is a synonym of Ca-
nae, and Borophaginae. Some of these subfamilies have since been nis. North American Cynodictis, in part, and Pseudocynodictis are
removed from the Canidae; others have been modified in content. synonymous with Hesperocyon. (Hesperocyon and several other
Of the genera originally included in the Amphicynodontinae by genera have been removed to the subfamily Hesperocyoninae by
Simpson (1945), Cephalogale, Hemicyon, and Dinocyon are now Wang, 1994; see later). Tephrocyon is a synonym of Tomarctus,
considered ursids by most authors (see Hunt, this volume, Chapter in part, and the latter is best considered a borophaginae, rather
10). The remaining amphicynodontine genera constitute a group of than a member of the Caninae (Tedford and Frailey, 1976). Other
 Kathleen Munthe 128

HESPERCYONINAE CAN INAE BOROPHAGINAE

Figure 7.5. Possible interrelationships among North American Tertiary canids. As the Borophaginae and Caninae are under review (Wang and Tedford,
in process), this cladogram should be viewed as a working hypothesis and thus without defining characters at all nodes. Solid lines indicate a higher
degree of confidence than dashed lines in proposed generic relationships. Several minor taxa of hesperocyonines have been excluded; see Wang (1994)
for a detailed phylogeny of that subfamily. The Cynarctoides group has been treated as a unit, pending reevaluation at the subfamily and generic levels.
Data from Berta (1981, 1987), Flynn, Neff, and Tedford (1988), Wang and Tedford (1992, 1994), R.H. Tedford and B. E. Taylor (pers. comm.), Wang
(1994), and personal observation. Key to numbered characters: (I) Fully ossified entotympanic bulla, stapedial artery lost, promontory branch of internal
carotid extrabullar and medial in position, petrosal flattened and in solid contact with basioccipital and basisphenoid bones, posterior accessory cusp on
P3 and p3, Ml with reduced parastyle, M3 lost. (2) Low bullar septum present, protocone of P4 reduced and appressed to anteromedial base of para-
cone, carnassials elongate, internal cingulum of Ml restricted to posteromedial corner of tooth. (3) Paroccipital process downturned and fused with bulla.
(4) Paroccipital process elongate and robust, rounded fossa on supraoccipital for rectus capitis dorsalis insertion. (5) Elongate rostrum, slender premolars,
quadrate molars, basined talonid. (6) UNNAMED TAXON: CANINAE PLUS BOROPHAGINAE: Broad bicuspid talonid with conical entoconid, associ-
ated development of metacone on Ml, talon of upper molars enlarged. (7) CANINAE: Mandible slender, well-spaced premolars small and simple, humerus
without entepicondylar foramen, metatarsal I reduced to proximal rudiment, feet digitigrade. (8) Metaconid of m2 enlarged and taller than the protoconid,
anterolabial cingulum of m2 enlarged and posterior cingulum present. (9) Upper incisors simple, cusplets on I 1-2 weak to absent, postorbital process broad.
( 10) Strongly arched zygoma with inverted jugals, angular process large with expanded pterygoid fossa. (11) Increased body size, upper incisors with acces-
sory cusps and 13 enlarged, frontal sinus enlarged and penetrating postorbital process. (12) BOROPHAGINAE: Premolars large and closely spaced, complex.
(13) Subquadrate Ml, cheek teeth with well-developed cingulae and accessory cusps, ml relatively short. (14) Parastyle of Ml weak to almost absent,
m2 with large metaconid, deep mandibular ramus with expanded angular process, feet digitigrade. (15) Increase in body size, premaxilla with strong
posterior, accessory cusps on incisors and premolars. (16) Molars large and subquadrate with large bunodont cusps, blade of P4 oblique, m3 enlarged.
(17) Carnassials large relative to molars, premolars with strong anterior and posterior accessory cusps. (18) I's and premolars large relative to molars, 13
with multiple accessory cusps, protocone of P4 separated from parastyle by a valley and bearing two cuspules, protocone of P4 connected to paracone by
a ridge. (19) Tendency for premolars to increase in size and molars to decrease; cheek teeth show consistently heavy wear. (20) Short face, broad palate,
frontals rise abruptly, p4 enlarged, carnassials small relative to molars, 13 with accessory cusps reduced. (21) Large to huge body size, protocone of P4
large and connected to parastyle by a ridge. (22) 13 enlarged but simple, anterior premolars reduced and simplified, entepicondylar foramen of humerus
reduced to lost. (23) Protocone of P4 reduced to absent, P4 conical but with posterior accessory cusps. (24) P4 simple and conical without accessory cusps,
entepicondylar foramen of humerus lost.

members of Simpson's Caninae are no longer considered canids. odictis, at least in part. Finally, the indeterminate arctoid carni-
Alopecocyon has been removed to the family Procyonidae (Baskin, vore Nothocyon has been removed from the Canidae by Wang and
this volume, Chapter 8), and Daphoenus and Sivacyon are now con- Tedford (1992) and is discussed later in this chapter.
sidered amphicyonids (Hunt, this volume, Chapter 11 ). European Simpson (1945) and, later, Petter (1964) placed the bat-eared
Cynodictis is a "miacoid" carnivore (Petter, 1966), as is Procyn- fox Otocyon in its own subfamily, the Otocyoninae, based on
 Canidae 129

derived characters of its chewing mechanism. These characters in- cyonid. They regarded Phlaocyon as a phylogenetic link between
clude small carnassial teeth, presence of supernumerary molars, canids and procyonids. This view was generally accepted by those
quadrate molars with low, rounded cusps, and a subangular process who subsequently described Cynarctus (now a borophaginae; see
on the dentary. A majority of morphological characters, however, earlier), Aletocyon, and Cynarctoides, all of which are dentally
indicate Otocyon is best considered a member of the Caninae (Van similar to Phlaocyon. McGrew (1938) formalized the apparent re-
Valen, 1964). lationship among these genera by classifying them within a new
Hesperocyon, Mesocyon, and Cynodesmus were listed within the procyonid subfamily, Cynarctinae. Simpson (1945) retained cynar-
Caninae by Simpson (1945), but are primitive canids possessing ctines within the Procyonidae, but moved Aletocyon and Phlao-
none of the derived characters shared by members of the Caninae cyon to the subfamily Procyoninae. Subsequent studies by Hough
or the Borophaginae. They and related taxa are best placed within (1948) showed that the auditory regions of these genera were canid-
a separate subfamily, the Hesperocyoninae, as originally proposed like, and they were formally classified as canids by Romer (1966).
by Tedford (1978) and formalized by Wang (1994). This subfamily Other additions to the Cynarctinae included Oxetocyon (Green,
encompasses several lineages within the initial radiation of canids 1954) and Bassariscops (Frailey, 1979). "Nothocyon" annectens
and is necessarily paraphyletic, but is a convenient grouping of (Peterson, 1906) has never formally been considered a cynarctine,
primitive canids. In addition to Hesperocyon, Mesocyon, and Cyn- but McGrew (1938) and Hough (1948) noted its similarity to the
odesmus, it includes Sunkahetanka, Philotrox, Enhydrocyon, and just mentioned genera and its probable erroneous assignment to
four new genera: Ectopocynus, Parenhydrocyon, Caedocyon, and Nothocyon. Cynarctines share certain dental characters with mem-
Osbornodon. bers of the Borophaginae (e.g., large, closely spaced, and complex
The subfamily Borophaginae was first proposed by Simpson premolars) that indicate relationships with, or within, that subfam-
(1945) to encompass the so-called "hyaenoid dogs". Several of the ily. Phlaocyon was discussed as a possible borophaginae by Tedford
genera placed in this subfamily by Simpson (Borocyon, Pliocyon, (1978; see earlier). The remaining cynarctines (Oxetocyon, Cynarc-
Hadrocyon, and Gobicyon) were later removed, primarily to the Am- toides, Aletocyon, Bassariscops), henceforth referred to as the "Cy-
phicyonidae (see Hunt, this volume, Chapter 11). What remained narctoides group," here are tentatively placed within the Borophag-
of Simpson's original subfamily, then, were Aelurodon, Boroph- inae, pending reevaluation.
agus, and Osteoborus. In evaluating Aelurodon, McGrew (1944) In this chapter, then, three subfamilies of canids are recognized:
recognized two distinct species groups, an A. taxoides group and an (1) Hesperocyoninae, which includes Hesperocyon, Mesocyon, and
A. saevus group. Tedford and Frailey (1976) later applied the name other primitive canids and, as currently constituted, is paraphyletic;
Prohyaena (Schlosser) to theA. taxoides group. Baskin (1980), how- (2) Caninae, which includes all extant genera of canids and their
ever, showed that the type of A. ferox, a single P4, is like that of the close fossil relatives; and (3) Borophaginae, which includes not
A. taxoides group and not like that of the A. saevus group. He con- only "hyaenoid dogs," but canids with probable predaceous, omniv-
cluded that the' name Aelurodon should be restricted to members of orous, and frugivorous habits as well. The Cynarctoides group, a
the A. taxoides group and that the name Epicyon (Leidy) should be collection of small, hypocarnivorous canids, is tentatively included
applied to the A. saevus group. Strobodon (Webb, l 969a) seems to within the Borophaginae. These three subfamilies and their con-
be allied with A. taxoides. As noted earlier, Euoplocyon and Tomarc- stituent Tertiary genera, as well as the primitive canid Prohes-
tus are now considered members of the Borophaginae (Tedford and perocyon, are catalogued here. Figure 7.5 depicts their possible
Frailey, 1976). Other additions to the subfamily include Carpocyon, relationships. Temporal ranges of Tertiary canids are illustrated in
Cynarctus, and Cormocyon. Carpocyon has not been formally rec- Figure 7.6.
ognized as a member of this subfamily, but Webb ( l 969b) discussed
its probable close relationship to Tomarctus based on dental char-
acters, and it is here considered a borophagine (see also Munthe, INCLUDED NORTH AMERICAN GENERA IN THE
1989). The genus Cynarctus, once considered a member of the FAMILY CANIDAE
subfamily Cynarctinae (see later), shares a number of characters
(e.g., complex premolars, premaxillary-frontal contact) with The locality numbers listed for each genus refer to the list of unified
members of the Borophaginae and is also assignable to that sub- localities in Appendix I. The acronyms for museum collections are
family (Tedford, 1978; Munthe, 1989). The same is true of Phlao- listed in Appendix III.
cyon (Tedford, 1978; see later). Finally, Cormocyon, a primitive The locality numbers may be listed in a couple of alternative
borophagine that apparently played a central role within the radia- ways. Parentheses around the locality (e.g., [CPI 01]) mean the taxon
tion of the subfamily, has been added to the Borophaginae, based in question at that locality is cited as an "aff." or "cf." the taxon
on material mistakenly referred to Nothocyon (Wang and Tedford, in question. Parentheses are usually used for individual species,
1992). thus implying the genus is firmly known from the locality, but the
The complicated taxonomic history of the carnivore subfam- actual species identification may be questionable. Question marks in
ily Cynarctinae was summarized by Galbreath (1956). In describ- front of the locality (e.g., ?CPlOl) mean the taxon is questionably
ing Phlaocyon, Wortman and Matthew (1899) noted its skeletal known from that locality, thus implying some doubt the taxon is
resemblances to Hesperocyon, the oldest recognized canid at the actually present at that locality, either at the genus or the species
time, and its dental similarities to Bassariscus, a primitive pro- level.
 Kathleen Munthe 130

CANIDAE INCERTAE SEDIS Hesperocyon Scott, 1890 (synonyms: Canis, in part;

Cynodictis, in part; Procynodictis, in part)
Prohesperocyon Wang, 1994 (synonyms: Hesperocyon in Type species: Hesperocyon gregarius (Cope, 1873)
part; Miacis in part) (=Canis).
Type species: Prohesperocyon wilsoni (Gustafson, 1986) Type specimen: AMNH 5297a (dentary with teeth).
(= Hesperocyon). Characteristics: Small, elongate body with long tail (see Fig-
Type specimen: TMM 40504-126 (partial skull, mandible). ure 7.4 ); forefeet subdigitigrade and claws probably retrac-
Characteristics: Fully ossified entotympanic bulla and other tile (Wang, 1993). Derived in relation to Prohesperocyon
basicranial characters diagnostic of the Canidae present, in bullar and dental characters. Low intrabullar septum
but partial intrabullar septum lacking. Teeth primitive com- present. Protocone of P4 reduced and medially positioned.
pared to Hesperocyon. Accessory cusps on P3 and p3 Internal cingulum of Ml enlarged and posteriorly posi-
small; protocone of P4 large and anteriorly placed; Ml with tioned. Shearing blade of ml elongate and anteroposteri-
well-developed parastyle; m 1 with short trigonid blade and orly oriented; trigonid and talonid of ml approximately
with talonid narrow relative to a high trigonid. Further equal in height. Metacone and paracone of M2 subequal.
differentiated from Hesperocyon by several autapomor- Protoconid and metaconid of m2 subequal.
phies (Wang, 1994). Rostrum elongate; bulla anteriorly Average length of ml: 9.23 mm.
expanded; premolars slender; metacone of M2 extremely Included species: H. gregarius (= H. paterculus) (known
reduced; metaconid of m2 reduced. from localities CP39B, CP39C, CP39F, CP40A, CP40B,
Average length of ml: 9.1 mm. CP41A, CP41B, CP68B, CP68C, CP83C, CP84A, CP84B,
Included species: P. wilsoni only (= Hesperocyon wilsoni, CP98B, CP98C, CP99A, CP99B, NP9A, NPlOB, NP24C,
"Miacis" gracilis, in part), known from locality SB44D NP24D, NP29C, NP29D, [NP32B], NP50B, NP50C,
only. PN6A, PN6B, PN6C); "H." coloradensis (= H. lippincot-
tianus, in part) (locality CP68C); "H." pavidus (localities
Comments: Described as a species of Hesperocyon by CC9C, CP68C, CP84B, CP99A, CP99B).
Gustafson (1986), H. wilsoni was removed to the new Hesperocyon sp. is also known from localities GC22,
genus Prohesperocyon by Wang (1994) in recognition of CP39IIA, CP42B, CP43, CP45A, CP84C, CP85C, NP25C,
its very primitive nature. Prohesperocyon has a canid-type NP27C.
bulla but is lacking the partial intrabullar septum found
in all other canids. Its dentition is primitive compared to Comments: Wang (1994) removed H. temnodon to "Meso-
other canids and is similar to that of "Miacis" gracilis. Pro- cyon" and H. wilsoni to Prohesperocyon, restricting con-
hesperocyon is thus a transitional animal and at the base tent of the genus to the three species listed here, separable
of the canid phylogeny (Wang, 1994; Wang and Tedford, on the basis of size and subtle morphological differences.
1994). Wang was unable to find convincing synapomorphies link-
ing these species and considered Hesperocyon to be para-
phyletic. He noted that "H." coloradensis showed initial
HESPEROCYONINAE tendencies toward the Mesocyon/Osbornodon clade and
that "H." pavidus had dental features that indicated a
Characteristics: This subfamily is a grouping of primitive canids
borophagine relationship.
distinguished primarily by a lack of synapomorphie3, as noted by
Wang (1994). A canid-type bulla is present, but some taxa lack
Ectopocynus Wang, 1994
a bony, tubular meatus. The talonid of m 1 is trenchant in most,
Type species: Ectopocynus simplicidens Wang, 1994.
which is probably a primitive character for the subfamily, whereas
Type specimen: F:AM 25431 (incomplete dentary with teeth).
a bicuspid talonid is primitive for other canid subfamilies and for
Characteristics: Lower premolars short, blunt, and robust
caniform carnivores in general (Flynn, Neff, and Tedford, 1988).
with accessory and cingular cusps extremely reduced to
Compared to Prohesperocyon, the protocone of P4 is located more
absent; pl lost. Metaconid reduced on ml and m2.
medially, the carnassial is more elongate, and the anterior cingulum
Average length of ml: 18.54 mm.
of M 1 is reduced or lost. Included species: E. simplicidens (known from localities
Comments: Most of the characters that distinguish hesperocy-
CP105, CP106); E. antiquus (localities CP48, CP84B);
onines from Prohesperocyon are also present in borophagines and
E. intermedius (locality CP48).
canines, neither of which were included in Wang's (1994) analysis.
Monophyly of the Hesperocyoninae was thus not supported. Wang Comments: The three species of Ectopocynus named by Wang
( 1994) was able to recognize, at least tentatively, several clades (1994) are represented by a total of six specimens, none of
within the Hesperocyoninae. A Mesocyon-Enhydrocyon clade and which preserves a basicranium. The status of Ectopocynus
a probable sister Osbornodon clade were best supported by the as a canid and its relationships within the family, there-
data; Parenhydrocyon and Ectopocynus represent other possible fore, cannot be evaluated adequately at this time. Wang
clad es. (1994) found no convincing dental synapomorphies
 Canidae 131

linking Ectopocynus with any particular canid lineage but Average length of ml: 15.05 mm.
noted some dental similarities between E. simplicidens and Included species: M. coryphaeus (known from localities
Hesperocyon gregarius. CC9C, CC12, PN6B, PN6C, PN6D, ?PN6E, ?PN6F); M.
brachyops ( = M. baileyi) (localities CC9C, PN6C, PN6D,
Parenhydrocyon Wang, 1994 (synonyms: Enhydrocyon, in ?PN6E); "M." temnodon (= Hesperocyon temnodon) (lo-
part; Mesocyon, in part; Temnocyon, in part) calities CP39J, CP68D, CP84A, CP84B, CP85C, CP99B,
Type species: Parenhydrocyon wallovianus (Cope, 1881) CP101, CP102).
(= Temnocyon). Mesocyon sp. is also known from locality NP51A.
Type specimen: AMNH 6858 (partial skull with teeth).
Characteristics: Paroccipital process laterally expanded and Comments: Nine species have been referred to Mesocyon
posteriorly directed. Suprameatal fossa lost. Mandible long historically (see Wang, 1994, for a review). Wang
and slender, masseteric fossa deep. Premolars narrow. De- (1994) removed several species to Parenhydrocyon (M.
rived species with dental modifications for hypercarnivory. josephi, M. robustus, M. venator) and to Cynodesmus
Average length of ml: 16.73 mm. (M. drummondanus), restricting content of Mesocyon to
Included species: P. wallovianus ( = Temnocyon wallovianus, the three species listed here. He found no convincing syna-
T. venator, Enhydrocyon wallovianus, E. venator, Meso- pomorphies linking these species but grouped them on the
cyon venator) (known from localities SB46, ?CP49B, basis of shared primitive characters. The genus Mesocyon
CP51A, CP52, CP103 or CP104A, PN6?F, G or H); is thus paraphyletic.
P. josephi (= Mesocyonjosephi, M. hortulirosae, in part)
(localities CP42D, CP46II, CP48, CP49B, CP50, CP52, Cynodesmus Scott, 1893 (synonyms: Brachyrhynchocyon,
CPlOl, PN6C, PN6D, ?PN6E); P. robustus (= Mesocyon in part; Mesocyon, in part; Sunkahetanka, in part)
robustus, M. hortulirosae, in part) (localities ?CP51A, Type species: Cynodesmus thooides Scott, 1893.
CP86C, CP86D). Type specimen: YPM-PU 10412 (partial skull, dentaries with
teeth).
Caedocyon Wang, 1994 Characteristics: Paroccipital process strong, with posterior
Type species: Caedocyon tedfordi Wang, 1994. keel. Broad M2. Bulla less inflated than in Mesocyon; pre-
Type specimen: UW 10251 (partial skull with teeth). molars more robust and more closely spaced; protocone
Characteristics: Rostrum short, broad, and deep. Frontal re- of P4 larger. Compared to Sunkahetanka, premolars less
gion slightly domed; postorbital process of frontal bone re- robust, ml shorter with unreduced metaconid.
duced; frontal sinus extensive. Enlarged I3 is caniniform. Average length of ml: 15.97 mm in C. thooides: 20.1 mm in
Tooth row short with short premolars, reduced molars. Pre- C. martini.
molars simple. Included species: C. thooides ( = Mesocyon drummondanus,
Average length of ml: none available; average length of Ml Sunkahetanka sheffieri, S. geringensis, in part, Brachyrhyn-
= 10.4mm. chocyon douglassi) (known from localities CP84B,
Included species: C. tedfordi only, known from locality CP86A, CP101, NP34B, NP36A, NP36B); C. martini (lo-
CP42D only. cality CPlOl).

Comments: Caedocyon is known from a single specimen Comments: Of the 14 species originally assigned to Cyn-
that does not preserve a basicranium, so its status as a odesmus, few are actually hesperocyonines (see Wang,
canid cannot be confirmed at this time. Wang (1994) tenta- 1994, for a review). C. vulpinus has been synonymized with
tively placed Caedocyon within the Hesperocyoninae and Leptocyon, a canine; other species have been removed to
suggested a possible affinity to Parenhydrocyon, based on Tomarctus or to other borophagine genera. Wang (1994)
characters such as simple premolars and an enlarged frontal further removed C. brachypus and C. iamonensis to the
sinus. new hesperocyonine genus Osbornodon, and considered
C. cooki to be a primitive borophagine close to Cormo-
Mesocyon Scott, 1890 (synonym: Temnocyon, in part; cyon. He restricted content of Cynodesmus to C. thooides,
Hesperocyon, in part; Hypotemnodon) transitional between Mesocyon and Sunkahetanka, and the
Type species: Mesocyon coryphaeus (Cope, 1878) (= Tem- larger and more robust C. martini, represented by only one
nocyon). specimen.
Type specimen: AMNH 6859 (skull, partial skeleton).
Characteristics: Distinguished from H esperocyon and Paren- Sunkahetanka Macdonald, 1963 (synonyms: Enhydrocyon,
hydrocyon by a long robust, ventrally directed paroccipital in part; Mesocyon, in part)
process. Primitive in having a wide M2 and lacking ro- Type species: Sunkahetanka geringensis (Barbour and
bust premolars. Derived species with round fossae on the Schultz, 1935) (= Mesocyon).
supraoccipital bone. Type specimen: UNSM 1092 (skull, mandible, skeleton).
 Kathleen Munthe 132

Characteristics: Compared to Cynodesmus, premolars more Type specimen: F:AM 27363 (skull, partial mandible, incom-
massive, metaconid on lower molars reduced. Compared plete skeleton).
to Mesocyon and Cynodesmus, skull heavier and denti- Characteristics: Paroccipital process partly ventrally directed,
tion more robust: larger preorbital process, posteriorly ex- expanding posteriorly into a sagittally oriented plate in de-
tended nuchal crest, deeper zygomatic arch, lower incisors rived species. Rostrum rather long; frontal region domed,
crowded, ml wider. Compared to Philotrox, bulla smaller with extensive sinus in later species. Upper molars quad-
and ramus shallower. rate; M2 and m2 enlarged. Lower molars with basined
Average length of ml: 17.63 mm. talonid and medially placed crestlike entoconid.
Included species: S. geringensis only (= Mesocyon gerin- Average length of ml: 18.78 mm.
gensis, Enhydrocyon geringensis) (known from localities Included species: 0. fricki (known from localities NB6C,
CP85C, CP99C, CP101). SB32B, CP108A, CP108B, CP109A, CPl 10); 0. brachy-
pus (= Cynodesmus brachypus) (locality CP51A); 0. ia-
Comments: Wang (1994) restricted content of Sunkahetanka
monensis (= Mesocyon iamonensis, Cynodesmus iamo-
to the type species by removing S. sheffieri to Cynodesmus
nensis, C. nobilis, Tomarctus nobilis) (localities GC3B,
and S. pahinsintewakpa to Enhydrocyon. He considered
?GC8A, [GC8C], GC8D, GC9A, CP104B, CP106,
Sunkahetanka transitional between the more primitive
CP107); 0. renjiei (localities CP84B, CP99B, NP50B);
Mesocyon and Cynodesmus and the more derived Philotrox
0. sesoni (= Brachyrhynchocyon sesoni) (locality CP84B ).
and Enhydrocyon.
Comments: Osbornodon is the longest lived hesperocyonine
Philotrox Merriam, 1906 (synonyms: Sunkahetanka, and parallels later canines in certain features of the skull
in part) and dentition (Wang, 1994). Notably, the rostrum is rather
Type species: Philotrox condoni Merriam, 1906. long and slender, the premolars are slender and the molars
Type specimen: UCMP 89 (skull fragment, dentary with quadrate, and the talonids of the lower molars are basined.
teeth). In Osbornodon, however, the medially placed and crestlike
Characteristics: More derived than Sunkahetanka. Postorbital entoconid replaces the hypoconid-entoconid ridge seen in
region of skull more elongate, masseteric fossa longer. canines (and in borophagines).
Sagittal crest higher, nuchal crest stronger and overhung
posteriorly. Mastoid more inflated. Premolars wider.
CANINAE
Average length of ml: 8.07 mm.
Included species: P. condoni only (= Sunkahetanka geringen- Characteristics: Derived characters that distinguish the Caninae from
sis, in part) (known from localities CP48, CP85B, CP85C, othercanids include small, simple, well-spaced premolars, a humerus
PN6D). without an entepicondylar foramen, and a metatarsal I which is re-
duced to a proximal rudiment. Certain characters of m2 were also
Enhydrocyon Cope, 1879 (synonyms: Hyaenocyon; identified by Berta (1987) as derived for the subfamily. These in-
Sunkahetanka, in part) clude an enlarged anterolabial cingulum, presence of a posterior
Type species: Enhydrocyon stenocephalus Cope, 1879. cingulum, and a metaconid taller than the protoconid.
Type specimen: AMNH 6902 (skull, dentary fragments). Comments: Primitively, canines and borophagines share several
Characteristics: Rostrum short, anterior frontal laterally dental characters that demonstrate their sister-group relationship
pinched. Paroccipital process laterally expanded; mastoid (Wang and Tedford, 1992). These include presence of a broad, bi-
process hypertrophied. Internal cusps present on upper in- cuspid talonid with conical entoconid on the lower molars, and an
cisors; I3 enlarged; Pl and pl lost. Carnassials elongate. enlarged talon, especially the hypocone, on the upper molars.
Paracone of Ml elevated. Talonid of lower molars tren-
chant, metaconid lost. Leptocyon Matthew, 1918 (synonyms: Canis, in part;
Average length of ml: 22.09 mm. Neocynodesmus, Nothocyon, in part)
Included species: E. stenocephalus (= E. oregonensis) Type species: Leptocyon vafer (Leidy, 1858) (=Canis).
(known from localities PN6D, ?PN6E); E. basilatus ( = ?E. Type specimen: USNM 126 (dentaries with nearly complete
sectorius) (localities CP52, ?NP37, PN6F); E. crassidens dentition).
(localities CP50, CP86B, CP86C); E. pahinsintewakpa (= Characteristics: Small, foxlike canid with slender jaw. An-
Sunkahetanka pahinsintewakpa) (localities CP48, CP50, terior premolars narrow, uncrowded, with small posterior
CP85C, CP86B, CP87A, CPlOl, CP102). accessory cusps. Upper cheek teeth with protocone consis-
Enhydrocyon sp. is also known from localities CP85A, tently larger and metaconule smaller than in Vulpes. Open
CP99C. trigonid and much reduced entoconid on m 1. Paraconid of
m2 low, vestigial, shelflike. Very small m3.
Osbornodon Wang, 1994 (synonyms: Brachyrhynchocyon, Average length of ml: 11.62 mm.
in part; Cynodesmus, in part; Mesocyon, in part) Included species: L. vafer (known from localities GC6A,
Type species: Osbornodonfricki Wang, 1994. GC6B, CC26B, NB7C, NB7D, NB7E, [NBS], NB28,
 Canidae 133

SB32D, SB34A, [SB48], CP76, CP107, CP114A, CP114B, Cerdocyon Hamilton Smith, 1839 (synonyms: Canis, in
CPl 140, CP115C, CP116A, NPl l);L. gregorii (=Notho- part; Dusicyon, in part)
cyon gregorii) (localities CP86C, CP87 A, CP104A); L. Type species: Cerdocyon thous (Linnaeus, 1766) (= Canis)
vulpinus (= Nothocyon vulpinus, Neocynodesmus or Cyn- (the extant crab-eating fox).
odesmus vulpinus) (localities CP86C, CP86D, CP103, Type specimen: None designated; based on Recent specimens
CP104B). of crab-eating fox from South America.
Leptocyon sp. is also known from localities ?GCl lA, Characteristics: The crab-eating fox is primarily a South
CP86B, CP86C, CP103, CP105, CP116B. American taxon with a scant North American record. It ex-
hibits several derived characters that it shares with the rac-
coon dog, Nyctereutes (Berta, 1987). These include small
Vulpes Brisson, 1762 (synonyms: Canis, in part)
carnassial relative to cheek teeth; mandibular condyle
Type species: Vulpes vulpes (Linnaeus, 1758) ( = Canis) (the
above the level of the alveolar border of cheek teeth; sub-
extant red fox).
angular lobe on dentary.
Type specimen: None designated; based on Recent speci-
Average length of ml in Tertiary specimens: none available;
mens of red fox from Europe (genus also includes North
average length of Ml in Pleistocene and Recent specimens
American red, kit, and swift foxes and several African and
is 9.5 mm.
Asian species).
Included Tertiary species: C. avius (known from locality
Characteristics: Skull with long, sharp muzzle; postorbital
CC53); ?C. n. sp. (known from the late Hemphillian of
process thin and concave dorsally; frontal bones less con-
Texas, exact locality unavailable).
vex than in Canis. Simple incisors. Canine teeth slender,
relatively longer than in Canis. Cheek teeth slender with
sharp, well-defined cusps and crests. Shares with Urocyon Canis Linnaeus, 1758 (synonyms: Aenocyon; Leptocyon,
simple I's, weak to absent cusps in Il-2, and a broad pos- in part)
torbital process (Berta, 1987). Type species: Canis familiaris Linnaeus, 1758 (the extant
Average length of ml in Tertiary specimens: 15.6 mm. domestic dog).
Included Tertiary species: V. shermanensis (known from lo- Type specimen: None designated; based on Recent specimens
calities CC37, SP3B, CP123D); V. stenognathus (localities of domestic dog from Europe (genus also includes wolves,
SB9, SB34C, SPlD, [CP116C], [CP116D]). coyotes, and jackals).
Vulpes sp. is also known from localities ?GC12II Characteristics: Large body size. Shorter, broader face than
CC17G, CC17H, CC17I, CC32A, CC32B, CC38, NB7E, Vulpes. Wide zygomatic arches; width of skull across wid-
NB2Q, NB31, (SB7), SB60, SPlA, SPlF, CP116F, CP127, est part of arches equal to at least half the length of the
PN4, PNlO, ?PN12, PN14. skull. Frontal sinuses enlarged and penetrating the pos-
torbital processes. Incisors with accessory cusps and I3
Urocyon Baird, 1857 (synonyms: Canis, in part) enlarged. Canine teeth relatively shorter than in Vulpes but
Type species: Urocyon cinereoargenteus (Schreber, 1775) thicker. Cheek teeth relatively larger than in Vulpes with
(=Canis) (the extant grey fox). blunter cusps; P4 about the same length as Ml, no para-
Type specimen: None designated; based on Recent specimens style.
of grey fox from North America. Average length of ml in Tertiary specimens: 16.8 mm in
Characteristics: Separate temporal ridges join posteriorly on Canis lepophagus; 11.69 mm in Canis davisi.
the skull to form a lyre-shaped configuration. A subangu- Included Tertiary species: C. condoni (known from local-
lar process is present on the dentary. Premolars are high ity PN14); C. davisi (= Leptocyon shermanensis) (local-
and narrow. Large protocone, small protoconule, variously ities NB31, SB9, SB34C, SPlD, SP3B, SP4B, CP116D,
developed external cingulum, and distinct metaconule are [CP116E], CP123D, PN3C, PN12, PN13, PN15); C.
present on M 1; large hypoconal shelf occupied by hypo- lepophagus (localities GC14, [GC17], NB13C, SB17,
cone and, in U. progressus, by two small cuspules. Later- [SB50], [SPlF]), SPlG, SPlH, SP5, [CP116F], CP117A,
ally expanded trigonid, paraconid reduced and displaced CP117B, CP118, CP128B, CP128C, CP130A, CP132B,
slightly anteriorly on ml. Entoconid of ml robust. Small PN4, PN23A, PN23C).
body size. Canis sp. is also known from localities ?CC40, CC43,
Average length of ml in Tertiary specimens: none available; SB10,SB14A,SB14B,SB14C,SB14E,SB14~SB58A,
average length of Ml in Tertiary specimens is 9.5 mm. SB58B, ?SB58C, SB60, SB63, ?SB65, SPlA, SPlF,
Included Tertiary species: U. progressus only (known from CP128A.
localities [SB50], [SPlF], ?CP1l7B, CP128C).
Urocyon sp. is also known from.locality NB13C. Comments: According to Berta (1987), "C." davisi is a prim-
itive canid not assignable to Canis and bearing a sister-
Comments: Undescribed specimens tentatively extend the an- group relationship to Pleistocene and Recent canids, ex-
cestry of this genus into the Clarendonian. cluding foxes.
 Kathleen Munthe 134

Chrysocyon Hamilton Smith, 1839 (synonyms: datus of Savage and Russell, 1983) (known from localities
Canis, in part) CP84B, CP99B).
Type species: Chrysocyon brachyurus (Illiger, 1815) (=Ca-
nis) (the extant maned wolf). Cynarctoides McGrew, 1938 (synonyms:
Type specimen: None designated; based on Recent specimens Cynarctus, in part)
of maned wolf from South America. Type species: Cynarctoides acridens (Barbour and Cook,
Characteristics: The maned wolf is primarily a South Amer- 1914) (= Cynarctus).
ican species but, as with Cerdocyon, there are a few North Type specimen: AMNH 82558 (Harold Cook Collection 144;
American specimens. Chrysocyon is a sister taxon of Canis right dentary fragment with ml).
(Berta, 1981, 1987) and shares with it the following charac- Characteristics: Small, with slender dentary. Heavy internal
ters: enlarged frontal sinus extending into postorbital pro- cingulum but no cusp on P4. Upper molars with stylar
cess; angular process with large fossa for branch of ptery- cusps and broad anterointernal cingulum; on Ml this cin-
goid muscle; enlarged I3 with accessory cusps and strong gulum bears a prominent cusp. Talonid of m 1 long relative
posteromedial cingulum. It can be distinguished from Ca- to trigonid; large accessory cusp at base of protoconid;
nis in having a small carnassial relative to molar size and accessory cusp present posterior to metaconid; entoconid
very elongate limbs. large.
Average length of m 1 in Tertiary specimens: none available; Average length of ml: 8.47 mm.
average length of M 1 in Pleistocene and Recent specimens Included species: C. acridens (known from localities CP54B,
is 15.6 mm. CP71, CP104A, CP105); C. mustelinus (= C. acridens?;
Included Tertiary species: C. n. sp. (known from the early to locality CP108B).
middle Blancan of Arizona and New Mexico; exact local- Cynarctoides sp. is also known from localities GC8A,
ities unavailable). NB6C, SB29B, CP104B, CP107.

BOROPHAGINAE Aletocyon Romer and Sutton, 1927

Type species: Aletocyon multicuspis Romer and Sutton, 1927.
Characteristics: Shared derived characters of the Borophaginae in- Type specimen: UC 1482 (skull).
clude a weak to almost absent parastyle on Ml, m2 with a large Characteristics: Skull with narrow muzzle and anteriorly
metaconid, strongly cuspidate premolars that are large and closely placed orbits. A small hypocone and distinct parastyle are
spaced, and a deep mandibular ram us with an expanded angular pro- present on P4. Cusps and accessory cusps of upper molars
cess (Wang and Tedford, 1992). The incisors bear accessory cusps; arranged in two parallel rows; heavy internal cingulum.
I3 is especially large and complex, except in terminal members of Average length of ml: none available; average length of Ml
the subfamily. The premaxilla and frontal bones are in sutural con- is 9.0mm.
tact and exclude contact between maxilla and nasal bones, or nearly Included species: A. multicuspis only, known from locality
so (see Munthe, 1989). Borophagines are primitive in retaining a CP51A only.
humerus with an entepicondylar foramen and a robust fibula. The
radius is short relative to the humerus; the tibia is short relative to Bassariscops Peterson, 1928 (synonyms: Phlaocyon, in part)
the femur. Metatarsal I is about half the length of metatarsal II. Type species: Bassariscops willistoni (Peterson, 1924)
Comments: This subfamily may include the Cynarctoides group, (= Phlaocyon).
a collection of small Oligocene to early Miocene canids (Oxeto- Type specimen: CM 11332 (anterior portion of skull).
cyon, Cynarctoides, Aletocyon, Bassariscops, and "Nothocyon" an- Characteristics: Smaller and less modified than other cynar-
nectens) that converge on procyonids dentally. The upper molars ctines. Longer P4 than in Phlaocyon, with small hypocone
in this group are subquadrate and tend to develop six major cusps. on internal cingulum. Upper molars wider than in Phlao-
Additional cingulae and cuspules of all gradations are present on cyon, with cingulae and extra cusps variously developed.
the carnassial and molar teeth, modifying the typical canid pattern. Only one accessory cusp on ml; m2 with anterointernal
Large, closely spaced, and often complex premolars are present. cingulum.
Average length of ml: 8.1 mm.
Oxetocyon Green, 1954 Included species: B. achoros (known from locality GC8A);
Type species: Oxetocyon cuspidatus Green, 1954. B. willistoni (locality CP73A).
Type specimen: SDSM 2980 (single upper molar).
Characteristics: Small canid with specialized, bunodont mo- Comments: Generic distinction may be unwarranted.
lars. Lingual cingulum of Ml 's with 3 distinct cusps. P4
with a parastyle but not a hypocone (Tanner, 1974). "Nothocyon" (new genus?)
Average length of ml: none available; average length of Ml Type species: "Nothocyon" annectens (Peterson, 1906).
is4.85 mm. Type specimen: CM 1602 (maxillae and dentaries with all
Included species: 0. cuspidatus only (= Cynarctoides cuspi- teeth but incisors).
 Canidae 135

Characteristics: Upper incisors subequal; I3 with a lingual per incisors arranged in a transverse row. Canine teeth short
shelf. Canine teeth long, slender, and curved. Strong cin- and curved. Blade of P4 oblique; small anterior cingu-
gulum around base of P4; incipient parastyle; cingular cusp lar cusp and hypocone present. Upper molars with low,
present posterior to protocone. Upper molars subquadrate rounded cusps, reduced external cingulum, no cingular
with six cusps and cingulae; broad talonid present on ml. cusps; anterointernal cingulum strong but without cusps;
Average length of ml: 9.5 mm. protoconule, hypocone, and two cuspules in position of
Included species: "N." annectens only (known from localities metaconule present. Lower molars low and broad. Limb
SB46, CP103, CP104, ?CP106). bones rather slender; radius and ulna about the same length
as humerus; tibia and fibula about the same length as fe-
Cormocyon Wang and Tedford, 1992 (synonyms: mur; metacarpal I about half the length of metacarpal III.
Nothocyon, in part) Average length of ml: 8.0 mm.
Type species: Cormocyon copei Wang and Tedford, 1992. Included species: P. leucosteus only, known from locality
Type specimen: AMNH 6885, skull, mandible, partial skele- CP71 only.
ton. Phlaocyon sp. is also known from localities ?GC8A,
Characteristics: A small, primitive borophagine unique in CP103, CP104A, CP104B, CP105, CP106.
possessing an elongate ml with open trigonid, better devel-
oped entoconid, and posterior cingulum (Wang and Ted- Comments: Cynodesmus minor was referred to "Phlaocyon"
ford, 1992). minor informally by Wang (1994).
Average length of ml: 10.4 mm (C. sp.).
Included species: C. copei only (known from localities PN6C, Tomarctus Cope, 1873 (synonyms: Cynodesmus, in part;
PN6D, ?PN6F). Tephrocyon)
Cormocyon sp. is also known from localities CP85C, Type species: Tomarctus brevirostris Cope, 1873.
?CP86B, ?NP36D, NP37. Type specimen: AMNH 8302 (dentary with ml).
Characteristics: Small- to medium-sized. Muzzle rather short
Comments: Wang and Tedford ( 1992) erected Cormocyon as and broad, tapering gradually. Large sagittal and lamb-
a primitive borophagine genus based on much of the ma- doidal crests, occiput with squared peak. Anterior premo-
terial originally referred to Nothocyon geismarianus (the lars with accessory cusps. Carnassial blades anteroposte-
type of N. geismarianus is not a canid). Wang (1994) infor- riorly aligned; P4 with variably developed parastyle, often
mally considered Cynodesmus cooki a primitive borophag- large. Talonid of ml bicuspid, trigonid large with small
inae close to Cormocyon. metaconid. Metaconule of upper molars large. Depth of
dentary and robustness of teeth variable.
Cynarctus Matthew, 1902 Average length of ml: 20.12 mm.
Type species: Cynarctus saxatilis Matthew, 1902. Included species: T. brevirostris (known from localities NB 18,
Type specimen: AMNH 9453 (right dentary with p3-m2). CP75B, CPllO); T. canavus (= Cynodesmus canavus) (lo-
Characteristics: Medium-sized canid with long face and slen- calities GC4B, GC8D); T.?confertus (localities [SB32B],
der dentary. Incisors large and lobed, canine teeth rather SB32C, [CP75B], [CPI 08B], [CPI! OJ); T. euthos (= Cyn-
long and curved. Anterior premolars well spaced. Well- odesmus euthos) (localities CP90A, CP114D); T. hippoph-
developed, sharp blade on P4 is obliquely aligned; dis- agus (localities [NB3D], CPI 10, CPll 1, NPl 1); T.? kel-
tinct cusp between protocone and paracone; no hypocone. logi (localities NB6C, NB17, [NB18], [NB23B], NB23C,
Upper molars large and subquadrate with large bunodont [NB31], [SB32A], [CP54B], PN8B); T. marylandica (lo-
cusps, including large metaconule and hypocone; strong, cality NC3A); T. optatus (localities CCl9, [CC20], NBl 7,
serrate cingulum internal to protocone; external cingulum [NB18], NB20A, SB29B, [SB29D], SB32A, CP54B,
weak. Large m3. Trigonid of lower molars low with two CP105, CP107, CP108A, CP108B); T. paulus (localities
accessory cusps. NB6D, NB6E, NB23B, SB32D, [CP73D]); T. propter
Average length of ml: 16.5 mm. (locality CPI l 5C); T. rurestris (localities CCl 7E, NB6C,
Included species: C. saxatilis (known from localities CP75B, CP56, [NP42], PN7, PN8B); T.? robustus (localities NB7C,
CP114A, CP114B); C. crucidens (localities CP114C, CP116A); T. temerarius (localities [CCI 7E], NB6E,
CPI 140, CP116B); C. fortidens (locality SP2A). SB30B, CPI 14A, CP114B, NP41B); T. thompsoni (=Cyn-
Cynarctus sp. is also known from localities NB6D, odesmus thompsoni; localities [NB17], CP51B, CP86E,
NB6E, SB32B, CP76, ?CPI 10, CPI 16A. CP104B, CP105).
Tomarctus sp. is also known from localities GC4C,
Phlaocyon Matthew, 1899
GC4E, GC9A, CC21A, CC23, NB6C, SB32F, SP2A,
Type species: Phlaocyon leucosteus ]'v1atthew, 1899.
CP88, CP90B, CP103, CP109A, CP114B, CP114C.
Type specimen: AMNH 8768 (skull, dentaries, skeleton).
Characteristics: Skull short and wide; dentary short and deep Comments: The characters cited here are quite variable and
with short coronoid process; orbits anteriorly placed. Up- reflective of the horizontal nature of Tomarctus. The name
 Kathleen Munthe 136

Tomarctus should probably be restricted to a few species molars crowded, slender, with accessory cusps. Strong
(e.g., T. optatus, T. hippophagus, T. brevirostris, and pos- parastyle and long blade on P4. Upper molars short and
sibly T. euthos). T.? confertus is a small form with a very wide. Posterior emargination present on Ml; hypocone in
short muzzle, quite different from other named species. T.? extreme posterolingual position so M 1/ appears "twisted"
kellogi shows dental adaptations for omnivory and may be lingually. Enamel of cheek teeth crenulated.
related to Cynarctus (Wang, 1994); T.? robustus is quite Average length of ml: none available. Average length of Ml:
close morphologically to early species of Epicyon. The 13.9 mm.
rather small, generalized species of Tomarctus that existed Included species: S. stirtoni only (known from localities CPI 14B,
through Barstovian and Clarendonian times may be gener- CPI 14C, CP114D).
ically distinct. The genus is badly in need of reevaluation. Strobodon sp. is also known from locality SB32D.

Euoplocyon Matthew, 1924 Carpocyon Webb, 1969b (synonym: Tomarctus, in part)

Type species: Euoplocyon praedator Matthew, 1924. Type species: Carpocyon limosus Webb, l 969b.
Type specimen: AMNH 18261 (partial left dentary with Type specimen: UF 12069 (two maxilla fragments with left
c-m3). P3-Ml, zygomatic arch, right P4-M2).
Characteristics: Anterior and posterior cusps on all anterior Characteristics: Medium-sized canid. Dentition with low,
premolars. Metaconid absent from ml, talonid trenchant. broad cusps tending to heavy wear. Anterior premolars
Crested trigonid and trenchant talonid present on m2. reduced in size; P3 with posterior accessory cusp, tending
Average length of m 1: 17.4 mm. to have strong lingual cingulum. Low, blunt blade, strong
Included species: E. praedator(known from locality CPl 10); parastyle, and broad protocone on P4. Massive Ml almost
E. spissidens (locality GC8D). square, with five major cusps including a large metaconule;
Euoplocyon sp. is also known from localities GC9A, continuous labial cingulum, strong labial cingulum, no sty-
NB6C, ?CP76, PN8B. lar cusps. Smaller M2 is similar.
Average length of ml: none available. Average length of Ml:
Aelurodon Leidy, 1858 (synonym: Prohyaena) 11.4 mm.
Type species: Aelurodonferox Leidy, 1858. Included species: C. limosus (known from locality GC13B);
Type specimen: USNM 523 (a single P4). C. cuspidatus (= Tomarctus cuspidatus, Cynodesmus cus-
Characteristics: Wolf-sized or larger animal with a long muz- pidatus) (localities NB23C, CPI 14A, CPI 14B, CPI 16A).
zle, deep mandible, and gently sloping forehead. Tendency Carpocyon sp. is also known from localities SB32C,
for incisors and premolars to increase in size. Anterior pre- SB32D, SB32F, ?SP2A, CP76, CPI 14D, CPI 16B.
molars complex, with accessory cusps. Protocone of P4
separated from the parastyle by a valley; protocone with Epicyon Leidy, 1858 (synonyms: Aelurodon, in part; Canis,
two cuspules and connected to paracone by a ridge. Ento- in part; Osteoborus, in part)
conid of ml much reduced; m2 with reduced paraconid, Type species: Epicyon saevus (Leidy, 1858) (=Canis).
metaconid, and talonid. Tendency for molars to decrease Type specimen: USNM 124 (dentary with p3-ml).
in size relative to P4. Characteristics: Coyote-sized to lion-sized. Face shorter and
Average length of ml: 29.85 mm. steeper than inAelurodon due to more abrupt rise of frontal
Included species: A. ferox (=A. wheelerianus; known from bones, more pronounced in later species. Incisors and ante-
localities NB23B, SB30B, SB32D, SB32EE, SB32F, rior premolars smaller and less complex than inAelurodon,
?[CP114A], CP114B, [NP41B]); A. francisi (locality but p4 large. Protocone of P4 a large, anterointernal cusp
GC4E); A. marshi (= A. taxoides?) (locality CP114B); connected to the parastyle via a ridge. Entoconid of ml
A. taxoides (localities GC6A, GC6B, NB23C, SB32B, equal in size to hypoconid. Unreduced paraconid, meta-
SB32H, SB34A, SPIA, SP2A, CP75B, CP90A, CP90B, conid, and talonid on m2. Tendency for molars to increase
CPl 12, CP114B, CP114C, CP114D, CP115B, CP116A, in size relative to P4.
CP123B). Average length of ml: 30.47 mm.
Aelurodon sp. is also known from localities CC38, Included species: E. saevus (= E. diabloensis?) (known from
NB6D, NB6E, NB29, NB42, SB32C, SPIA, CP54C, localities GC I IA, NB 7A, NB 7B, NB23C, NB27 A, NB28,
CP75C, PNl 1. SB34A, SP2A, CP90A, CP90B, CP114B, CP116A,
[CP116B], CP123B, NPll); E. diabloensis (= Osteoborus
Strobodon Webb, 1969a diabloensis) (localities CC26B, NB7D, NB7E, NB29); E.
Type species: Strobodon stirtoni Webb, 1969a. haydeni (= E. aphobus, E. mortifer) (localities GCl lA,
Type specimen: UCMP 33474 (left maxilla fragment with GCI lB, CC26B, NB7B, NB7C, NB7D, NB23C, SB6,
P2-Ml). [SB33II], [SB34A], [SB34B], [SB34C], SPIB, SPIC,
Characteristics: Similar to Aelurodon taxoides but smaller SP2A, CPI 15B, CPI 16A, CP116B, CP123C); E. injiatus
(coyote-sized) and with a narrower muzzle. Anterior pre- (locality CP90A); E. littoralis (= Osteoborus littoralis)
 Canidae 137

(localities CP32A, CP32B); E. validus (localities GCI lB, Average length of ml: 32.37 mm.
[GC12], SPIB, SPIC, SP3A, CP78, CP115C, [CP116C], Included species: B. diversidens (known from localities
CP116D, CP123C). GC14, GC18, [CC43], [CC53], SB14A, SB50, SPIF,
Epicyon sp. is also known from localitiesNB6E, NB23B, SPIH, SP5, CPI l 7A, CP117B, CPI 18, CP128A, CP128B,
SB32D, CP127, PN2A, PNIO. CP128C, PN4, PN23A, PN23C); B. dudleyi (localities
GC13B, [GC28]); B. pachyodon (locality NB12B); B. so-
Osteoborus Stirton and VanderHoof, 1933 (synonyms: lus (locality NB12A).
Hyaenognathus, in part; Pliogulo, in part) Borophagus sp. is also known from localities CC45,
Type species: Osteoborus cyonoides (Martin, 1928) ( = CP121, PN3C.
Hyaenognathus).
Comments: Great Plains and southeastern specimens are
Type specimen: KUVP 3468 (right dentary with all teeth but
larger than southwestern and West Coast specimens, which
p2 and m3).
are usually identified as B. pachyodon. Borophagus may
Characteristics: Short face with domed frontal bones and
be a continuation of the 0. cyonoides lineage.
broad palate. Incisors small and simple; canine teeth short
and stout. Anterior premolars greatly reduced and simpli-
Indeterminate borophagines
fied; p4 very large and conical, but with accessory cusps
Fossils assigned to the subfamily Borophaginae but of uncertain
on its posterior face lacking in Borophagus. Protocone
genus have been reported from localities CC17G, CC17H, CC51,
of P4 very reduced to absent, parastyle present. Meta-
NBS, SB48.
conid present on ml, unlike in Borophagus. Tooth row
crowded, teeth show heavy wear. Entepicondylar fora-
men of humerus present in oldest specimens, variable in INDETERMINATE CANIDS
younger specimens, absent in youngest specimens.
Specimens assigned to the family Canidae but of uncertain sub-
Average length of ml: 28.88 mm.
family or genus have been reported from the following localities:
Included species: 0. cyonoides (known from localities CA9,
WM26, GC13A, CC21B, CC21C, NB2, NB21, SB16, SB47, CP46,
CAIO, [NBIO], SB9, [SB58A], [SB58B], [SPIC], SPID,
CP131, NPIOE, NP34D, PN6H, PN9B, NC2.
SP3B, SP4A, CP123C, CP123D); 0. direptor (localities
SP4B, CP115C, CPI 15D); 0. galushi (locality GCl IB);
0. hilli (= 0. progressus) (localities SP4B, CP123E); 0. PROBLEMATICAL ARCTOIDS
ore (locality GC12II); 0. pugnator (localities [SP3A], Nothocyon Matthew, 1899 (synonyms: Canis, in part;
CP78, PN2B); 0. secundus (localities SB34C, CP115C, Cynodictis, in part; Galecynus, in part)
?CP1'15D). Type species: Nothocyon geismarianus (Cope, 1878)
Osteoborus sp. is also known from localities CC25B, (=Canis).
CC36, CC37, ?CC38, CC40, NB27B, NB31, SBIO, Type specimen: AMNH 6884, fragment of left dentary with
SB33II, ?SB60, SPIB, SPIC, (CPI 16C), CPI 16D, worn ml.
CPI 16E, CPI 16F, CP127, PNI 1, PN14, (PN15), and from Characteristics: Enamel surface crenulated. Premolars robust
Honduras. with lingual cingula; P4 broad relative to length. Rect-
angular Ml with caudally shifted metaconule. Trigonid
Comments: The numerous named species of Osteoborus need
of ml low; trigonid of m2 anteriorly placed and laterally
to be reevaluated. It is probable that at least some of them
compressed; paraconid of m2 greatly reduced. Similar to
are synonymous with 0. cyonoides. 0. pugnator, a larger
some hypocamivorous canids dentally (Wang and Tedford,
middle Hemphillian form, may also be a valid species.
1992).
Average length of ml: 10.75 mm.
Borophagus Cope, 1892 (synonyms: Hyaenognathus,
Included species: N. geismarianus only (known from locality
in part; Osteoborus, in part)
PN6F or PN6G).
Type species: Borophagus diversidens Cope, 1892.
Nothocyon sp. has also been identified from localities
Type specimen: UTBEG 18578 (dentary with p3-4).
?CC12, CP42D, CP51A, CP86C, CP88.
Characteristics: Head and neck large in proportion to body.
Limbs short, especially distally; no entepicondylar fora- Comments: Wang and Tedford (1992) reported that the holo-
men on humerus. First two incisors small and crowded; type of Nothocyon geismarianus is not a canid, and the
I3 and i3 larger and stouter. Canine teeth moderately long concept of this genus as a canid stemmed from materials
and stout. First two premolars reduced or lost in some indi- mistakenly referred to the type. They erected a new genus
viduals. Enlarged p4 simple and conical. Carnassial teeth and species of borophagine, Cormocyon, based on much
enlarged; P4 with no parastyle, protocone reduced to ab- of the referred material and redescribed N. geismarianus
sent. Metaconid of m I reduced. Tooth row crowded, shows as an indeterminate arctoid carnivore seemingly close to
heavy wear. Subparictis. Two species previously referred to Nothocyon,
 Kathleen Munthe 138

N. gregorii and N. vulpinus, are now considered species of duced "H." pavidus and "H." coloradensis, as well as Mesocyon
Leptocyon. N. lemur and N. latidens (see Wortman and and Osbomodon. By the Whitneyan, all major clades of hespero-
Matthew, 1899), and other taxa identified as Nothocyon, cyonines, including the first medium-sized canids, had appeared.
might be assignable to the Canidae but need to be re- Maximum diversity among hesperocyonines was reached during
assessed. the late Oligocene (early Arikareean); by the Miocene (early late
Arikareean), however, diversity was severely restricted. Taxa of the
latest Oligocene and early Miocene included seemingly formidable
BIOLOGY AND EVOLUTIONARY PATTERNS predators such as Enhydrocyon, Parenhydrocyon, and later species
of Osbornodon. They were fairly large animals, with large heads,
The evolutionary history of the Canidae is primarily a North Amer- robust skeletons, and short legs, indicative of strength rather than
ican phenomenon. Canids did not reach the Old World until the speed. Stevens ( 1991) envisioned Parenhydrocyon wallovianus
late Miocene or early Pliocene (Hemphillian) but were so success- ( = Mesocyon venator) as a scansorial predator, possibly ambushing
ful in North America that they apparently blocked any significant oreodonts from trees. Several lineages of hesperocyonines indepen-
invasion by similarly adapted members of other carnivore families dently acquired trenchant talonids, and Van Valkenburgh (1991)
during much of the Tertiary. hypothesized that hesperocyonines of the Arikareean filled the hy-
In general, canid evolution was shaped by the spread of savanna percarnivore niches left vacant by hyaenodontids and nirnravids.
and, later, grassland environments resulting from a deterioration in Osbornodon is notable among hesperocyonines in having a basined
world climate from the more tropical conditions that existed earlier talonid and may have been more mesocarnivorous in its habits.
in the Tertiary. Mesic forests, which previously had covered much A second Tertiary radiation of canids was that of the Borophagi-
of North America, became increasingly restricted and replaced over nae. Borophagines had their beginnings in primitive Oligocene taxa
large areas by open, grassy woodlands into middle Miocene times. such as Cormocyon (Wang and Tedford, 1992) and may be traced
These, in turn, were gradually displaced by grasslands, with a final back to "Hesperocyon" pavidus (Wang, 1994), but borophagine
shift to steppelike conditions occurring in the later Hemphillian, at evolution is primarily a Miocene phenomenon. An initial, small-
least on the Great Plains (see Gregory, 1971; Webb, 1977). scale radiation within this subfamily produced (assuming they are
Canids originated in North America from among "miacoid" car- borophagines) the hypocarnivores of the Cynarctoides group, which
nivores of the middle Eocene. Wang and Tedford's (1994) study of may have been an early attempt by canids to occupy a procyonidlike
basicranial and dental characters in selected caniforrn miacoids and oromnivore niche. Cynarctoids were moderately successful until the
in early canids narrowed the phylogenetic gap between these two Hemingfordian when, coincidentally, the scansorial-arboreal procy-
groups. They pinpointed the late Uintan "Miacis" gracilis as the onines made their first appearance in North America (see Baskin,
sister group of the Canidae and included both "M." gracilis and the this volume, Chapter 8). Whether cynarctoids were also scansorial-
late Bridgerian "M." sylvestris within the Cynoidea. Wang and Ted- arboreal is an open question; skeletal material has not been de-
ford estimated initiation of the cynoid clade (i.e., the cynoid-arctoid scribed.
divergence) as early Bridgerian in age (50 mya) and initiation of the Borophagines proper dominated canid history through much of
family Canidae (i.e., its divergence from "M." gracilis) as Uintan in the Miocene. Analysis of the dentition (Tedford, 1978) and postcra-
age (45 mya). nial skeleton (Munthe, 1989) has shown that borophagines were
There were three major radiations of canids in the Tertiary, each much more diverse in habit than previously suspected and can no
of which encompassed a range of adaptations. The first of these oc- longer be characterized simply as "hyaenoid dogs."
curred among the Hesperocyoninae, primarily during the Oligocene Nominal species of Tomarctus first appeared in the earliest
and early Miocene (Orellan to late Arikareean). A second, pri- Miocene, and at least some may have been similar to the modern-
marily Miocene, radiation of Tertiary canids occurred among the day jackal or coyote in dietary habits. They were of about the same
Borophaginae. During the late Miocene and Pliocene a third radia- body size and had rather generalized dentitions, and one specimen
tion of canids occurred among the Caninae. was preserved with a mass of small bones and teeth identified as the
Hesperocyon gregarius is the oldest known hesperocyonine. Its lagomorph Hypolagus, which presumably had been its meal (Green,
first undoubted appearance is in the Chadronian (late Eocene) 1948). If Tomarctus did concentrate on small prey animals, it was
(Wang, 1994), although dental records may extend its range back to probably primarily a solitary hunter. Skeletal proportions and mor-
the Duchesnean (Bryant, 1992). H. gregarius was a common carni- phology indicate Tomarctus was not an efficient cursor and may
vore of the Chadronian-Whitneyan and is represented by abundant have occupied riparian forests or more heavily vegetated areas of
materials. Study of the skeleton (Figure 7.4) of this small, primi- the savanna (Munthe, 1989).
tive canid by Wang (1993) revealed a mix of cursorial and arboreal Hypocarnivorous members of the Borophaginae include Cynar-
postcranial characters. Wang (1993) concluded that Hesperocyon ctus, Carpocyon, and possibly the earlier Phlaocyon. Hall and
was scansorial in habit, as were a number of other predators of that Dalquest's (1962) study of Cynarctus led them to conclude it was
time (Van Valkenburgh, 1985). Hesperocyon may have hunted small primarily a frugivore. Cynarctus may also have included small verte-
mammals in riparian forests and adjacent developing grasslands. brates in its diet, as do some of the more omnivorous/frugivorous ex-
An overview of subsequent hesperocyonine evolution was pro- tant canids. Webb ( l 969b) concluded that the almost ursidlike teeth
vided by Wang (1994). Initial diversification in the Orellan pro- of Carpocyon indicated a varied diet including many fruits. Little
 Canidae 139

EOCENE OLIGOCENE MIOCENE PLIOCENE

!I: r m r m !I: r m r
~ ~ ~ I!' I"
~ ~ " fl
r !? ~ i::
~ ~
,...m
~ ~
m
%
~ i •i:: ,•
m
r
r
~
m
p

"=
e !:
... ii:
ln ~ eEe !:l
'a ~ E ~ c l: ~ c t E b : c ;:: =cc: t: ~
Prohesperocyon
I I IHesperocyon
I I
I I I I I I
I I IEctopocynus I I I I I I
::c IParenhydrocyon I I I I I I
m
,,m
en ICaedocyon I I I I I I
:ti
IMesocyon I I I I I
0
0
- •!l_I
<
0
ICynodesmus I I I I I I
z 1sun~anka
z)> I I I I I I
m IPhilotrox I ..., . I I I I I
Enhydrocy7'n- -
I I I I
Osbornodon
I I I I
i----+--+-+-+--+--+-+--1--~L_e~p~tocyon
I I I
Vutpes
I I
~
z I I
z
)> I I
m
I I
I I I I I
Oxetocyon
t-"1 I I I I I
Cynarctoides
I Aletocyon
n 1 1 I I I I
Bassariscops
I I I I I I I I
"Nothocyon" annectans
m ._, i I I I I I I I
0
:ti
L'...,_ Cormocyon _:oJ_lj I I I I I I I
,,::c0 I I Cynarctus
I I
)>
G> Phlaocyon
....__....__ 1. I I I
z
)> Tomarctus
m
1Euop/ocyon I !.:..., I I I I
I I Aelurodon I
I I 1Strobodon

I I Carpocyon I
I I 1Epicyon I
I I oftfo?f 1us 1
1
I I isorophagus
L. ...J. •..., Nothocyon I
-I •..., I I I I
Figure 7.6. Temporal ranges of North American Tertiary canid genera.
 Kathleen Munthe 140

is known about the postcranial skeleton of either taxon. Cynarctus had a greater range of prey available to it than did smaller species; it
became extinct at the end of the Clarendonian (early late Miocene) could have made full use of prey animals or carcasses by consuming
with the final shift from savanna to grassland conditions; Carpocyon them bones and all, and was capable of appropriating the prey of
disappeared from the Great Plains in the later Clarendonian, but was other predators through its sheer size. The Epicyon lineage died out
extant in Florida in the late Hemphillian (latest Miocene), where it with the demise of E. validus at the end of the early Hemphillian
may have retreated with the vegetation on which it fed. (late late Miocene).
Aelurodon first appeared on the Great Plains in the Barstovian The Hemphillian borophagine Osteoborus can be derived from
(middle Miocene) and is difficult to separate from the larger Tomarc- something like Epicyon saevus and shows extreme development
tus (s.l.) species from which it was apparently derived. Its large size, of the skull and dental characters diagnostic of Epicyon. It has long
long muzzle, complex incisors and anterior premolars, and large been assumed that Osteoborus and related borophagines were bone-
carnassial teeth indicate predatory habits. The cursorial characters cracking scavengers. Their dental morphology and heavy degree of
seen in members of the Epicyon group are absent in Aelurodon, tooth wear certainly support this idea, and Dalquest (1969) found
which retained a versatile forearm (Munthe, 1989). Further, Radin- bone chips and small vertebrates in the fecal material that he as-
sky (1973) found that the brain ofborophagines lacked an enlarged signed to Osteoborus. Werdelin (1989) suggested that phylogenetic
prorean gyros, an apparent prerequisite for the development of pack constraints dictated this behavioral mode. Unlike hyaenas, which
social structure. One conclusion, then, is that Aelurodon relied on use P3 against p3/anterior p4 as bone crackers and thus maintain the
stealth and strength and was a solitary hunter of large prey an- carnassials as shearing teeth, Osteoborus used P4 against p4/anterior
imals, stalking them through vegetated areas and subduing them ml to crack bones and the carnassials thus became severely worn,
with grasping anterior teeth and versatile forearms (Munthe, 1989). limiting the versatility of the dentition. Werdelin (1989) hypothe-
No living canid hunts in this manner. Heavy apical wear on the sized that the evolutionary retention of posterior molars in canids
cheek teeth of Aelurodon indicates it may have made full use of its predetermined use of P4 and p4 as the bone-cracking teeth because
prey by consuming bones, thus offsetting the usually low hunting P3 and p3 are more anteriorly placed than they would be if posterior
success rate found in solitary hunters. Aelurodon's disappearance in molars were lost and are thus outside of the region of maximum bite
the early Hemphillian is probably related to the spread of grasslands force as determined by mechanical studies of the skull. The domed
in the late Tertiary. forehead in these borophagines dissipated the forces generated by
Epicyon existed in the southwestern United States by the mid- bone cracking. Although no living mammal lives entirely by scav-
dle to late Barstovian (middle Miocene), but did not appear on enging due to time constraints (Van Valkenburgh, 1985), advanced
the Great Plains until about the middle Clarendonian (early late borophagines exhibit a unique combination of body size and dental
Miocene). The slenderness, proportions, and morphology of their characters (Van Valkenburgh, 1988) that suggests a unique mode
limb bones indicate Epicyon saevus and the larger E. haydeni were of life. They may have been similar in habit to brown and striped
more cursorial than Aelurodon and may have inhabited more open hyaenas, which subsist primarily on carrion supplemented by fruits,
habitats (Munthe, 1989). Janis and Wilhelm (1993), who disputed insects, and small vertebrates, but likely emphasized carrion even
the existence of high-speed pursuit predators prior to the devel- more and included a larger percentage of bone in their diets. Van
opment of steppelike environments in the latest Tertiary, analyzed Valkenburgh (1989) hypothesized high carcass predictability in the
limb proportions in a wide variety of Tertiary and Recent carnivores. later Hemphillian, when open habitats prevailed and scavenging
They found that Tomarctus, Epicyon haydeni, and Osteoborus did birds would have improved carcass finding, as in the Serengeti to-
not fall into the morphospace clusters defined by present-day pur- day. Still, carrion feeders must range over large areas in search
suit, ambush, or generalist carnivores. Although they were unable of food, but the stocky and robust nature of their skeletons im-
to determine the predatory behavior of these animals, limb propor- plies Osteoborus and Borophagus had the endurance to do just that
tions were considerably more robust than those of extant canids. (Munthe, 1989). Population levels among scavengers are necessar-
The attack morphology of predators is appropriate to the escape ily low because a large area is needed to support each individual.
strategies of their prey and therefore the vegetative structure of their The impression gained from the fossil record is that Osteoborus did
habitats (Van Valkenburgh, 1985). In the case of Epicyon, we do not not occur in very great numbers and that Borophagus was espe-
know what those were, and its attack morphology may have been cially rare and scattered. Borophagines died out with the extinction
as unique as that just proposed for Aelurodon. Although Epicyon's of Borophagus in the late Blancan or early Irvingtonian (latest
morphology may seem suboptimal compared to that of extant cur- Pliocene).
sors, and it probably was not capable of maintaining speed over dis- A third radiation of canids began in the late Clarendonian/early
tance, its skeletal morphology provides evidence that running was a Hemphillian (late Miocene) among the Caninae. This radiation can
part of its behavioral repertoire, if only in a final chase following a be derived from Leptocyon (Radinsky, 1973), a small, foxlike ani-
stalk. mal that had appeared by the late early Arikareean (latest Oligocene)
Epicyon had a short muzzle, a domed forehead, and cheek teeth and was the only member of the Caninae extant during the late
with very heavy apical wear, suggesting a greater proclivity for bone Oligocene-middle Miocene. Leptocyon coexisted with boro-
cracking than Aelurodon. Extreme development of these characters phagines through much of their history, but its small size proba-
is seen in the Hemphillian E. validus. This huge, lion-sized canid bly minimized competition with most of them. Limb morphology
 Canidae 141

suggests Leptocyon was probably a better cursor than contempo- (1879). Observations on the faunae of the Miocene Tertiaries of Oregon.
raneous borophagines and may have occupied more open habitats Bulletin of the United States Geological and Geographical Survey
(Voorhies, 1969), although its apparent long-leggedness may merely of the Territories, 5, 55-69.
reflect allometric scaling (see Janis and Wilhelm, 1993). (1881). On the Nimravidae and Canidae of the Miocene period. Bulletin
of the United States Geological and Geographical Survey of the
The initial radiation of the Caninae correlates with the develop- Territories, 6, 165-81.
ment of steppelike environments in the latest Tertiary and Pleis- (1892). A hyaena and other Carnivora from Texas. Proceedings of the
tocene, but taxonomic diversity was rather low until members of Academy of Natural Sciences of Philadelphia, 44, 326-7.
the subfamily dispersed from North America. Urocyon (grey fox) Dalquest, W. W. (1969). Pliocene carnivores of the Coffee Ranch (type
Hemphill) local fauna. Bulletin of the Texas Memorial Museum, 15,
now ranges through much of North and South America. Vulpes (in-
1-44.
cluding red fox) appeared in the Old World in the early Pliocene as Evans, H. E., & deLahunta, A (1971). Miller's Guide to the Dissection of
a probable migrant from North America and is currently the most the Dog. Philadelphia: W. B. Saunders.
variable and widespread canid genus. Primitive species of Canis Flower, W. H. ( 1869). On the value of the characters of the base of the cranium
apparently invaded the Old World in the early Pliocene along with in the classification of the Order Carnivora, and on the systematic
position of Bassaris and other disputed forms. Proceedings of the
Vulpes. Radiations of canids in the Old World and in South America
'Zoological Society of London, 1869, 4-37.
(see Berta, 1987, 1988), particularly oflarge, woltlike species, were Flynn, J. J., Neff, N. A., & Tedford, R.H. (1988). Phylogeny of the Car-
primarily Pleistocene phenomena. nivora. In The Phylogeny and Classification of the Tetrapods, Vol.
2: Mammals, ed. M. J. Benton, pp. 73-116. Systematics Association
Special Volume, no. 35B. Oxford: Clarendon.
ACKNOWLEDGMENTS Frailey, D. ( 1979). The large mammals of the Buda local fauna (Arikareean:
Alachua County, Florida). Bulletin of the Florida State Museum,
Biological Sciences, 24, 123-73.
I am especially grateful to R.H. Tedford, B. Taylor, and others on Galbreath, E. C. (1956). Remarks on Cynarctoides acridens from the
the staff of the American Museum of Natural History for extensive Miocene of northeastern Colorado. Transactions of the Kansas
discussions that led to my current understanding of canids. R. H. Academy of Sciences, 59, 373-8.
Gray, J. E. ( 1821 ). On the natural arrangement of vertebrose animals. London
Tedford and J. Baskin reviewed this manuscript and provided in-
Medical Repository, 15, 296-310.
numerable constructive suggestions. Finally, special thanks are due Green, M. (1948). A new species of dog from the lower Pliocene of Cali-
C. M. Janis, for bringing this volume to fruition. fornia. University of California Publications in Geological Sciences,
28, 81-90.
(1954). A cynarctine from the upper Oligocene of South Dakota. Trans-
REFERENCES actions of the Kansas Academy of Sciences, 57, 218-20.
Gregory, J. T. (1971). Speculations on the significance of fossil vertebrates
for the antiquity of the Great Plains of North America. Hessische
Baird, S. F. ( 1857'). Mammals: general report upon the zoology of several Pa- Landes-Amt. Bodenforsch, Abhandlungen, 60, 64-72.
cific railroad routes. Reports, Explorations and Surveys for Railroad Gustafson, E. P. (1986). Carnivorous mammals of the late Eocene and early
Route from Mississippi River to Pacific Ocean, 8, 1-757. Oligocene of Trans-Pecos Texas. Texas Memorial Museum Bulletin,
Barbour, E. H., & Cook, H. J. (1914). Two new fossil dogs of the genus 33, 1---06.
Cynarctus from Nebraska. Nebraska Geological Survey, 4, 225-7. Hall, E. R., & Dalquest, W.W. (1962). A new dog-like carnivore, genus
Barbour, E. H., & Schultz, C. B. (1935). A new Miocene dog, Mesocyon Cynarctus, from the Clarendonian, Pliocene, of Texas. University of
geringensis, sp. nov. Bulletin of the Nebraska State Museum, 43, Kansas Publications, Museum of Natural History, 14, 135-8.
407-18. Hamilton Smith, C. (1839). Dogs. Canidae or genus Canis of authors. In:
Baskin, J. A. (1980). The generic status of Aelurodon and Epicyon (Car- Jardine's Naturalist's Library, 9, pp. 1-267. Edinburgh.
nivora, Canidae ). Journal of Paleontology, 54, 1349-51. Hough, J. R. (1948). The auditory region in some members of the Procy-
Berta, A (1981). Evolution of large canids in South America. University of onidae, Canidae, and Ursidae: its significance in the phylogeny of
Florida Contributions to Vertebrate Paleontology, 196, 835-45. the Camivora. Bulletin of the American Museum of Natural History,
(1987). Origin, diversification, and zoogeography of the South American 92, 67-118.
Canidae. Fieldiana: 'Zoology n.s., 39, 455-71. Hunt, R. M. (1974). The auditory bulla in Carnivora: an anatomical basis
( 1988). Quaternary evolution and biogeography of the large South Amer- for reappraisal of carnivore evolution. Journal of Morphology, 143,
ican Canidae (Mammalia: Carnivora). University ofCalifornia Press, 21-76.
Publications in Geological Sciences, 132, 1-149. (1987). Evolution of the aeluroid Carnivora: significance of auditory
Brisson, M. J. (1762). Regnum Animale in Classes IX Distribution, Sive Syn- structure in the nimravid cat Dinictis. American Museum Novitates,
opsis Methodica. Editio Alterea Auctior. Leiden: Theodorum Haak. 2886, 1-74.
Bryant, H. N. (1992). The Carnivora of the Lac Pelletier Lower Fauna Illiger, K. (1815). Uberlich der Saugethiere nach ihrer Verteilung uber die
(Eocene: Duchesneau), Cypress Hills Formation. Paleontology, 66, Weittheile. Abhandlungen der Akademie der Wissenschaften, Berlin,
847-55. 1804-11, 39-159.
Clark, J., & Guensburg, T. E. (1972). Arctoid genetic characters as related Janis, C. M., & Wilhelm, P. B. (1993). Were there mammalian pursuit preda-
to the genus Parictis. Fieldiana: Geology, 26, 1-71. tors in the Tertiary? Dances with wolf avatars. Journal ofMammalian
Cope, E. D. ( 1873). Third notice of extinct Vertebrata from the Tertiary of Evolution, I, 103-26.
the plains. Palaeontological Bulletin, 16, 1-8. Leidy, J. (1858). Notice of remains of extinct vertebrata from the valley
(1878). On some characters of the Miocene fau~a of Oregon. Proceedings of the Niobrara River, collected during the exploring expedition of
of the American Philosophical Society, 28, 63-181. 1857, in Nebraska under the command of Lieut. G. K. Warren, U.S.
 Kathleen Munthe 142

Top Engineer by F. Hayden. Proceedings of the Academy of Natural Scott, W. B. (1890). The dogs of the American Miocene. Princeton College
Sciences of Philadelphia, IO, 20-9. Bulletin, 2, 37-9.
Linnaeus, C. (1758). Systema naturae per regna tria naturae secundum (1893). The mammals of the Deep River beds. The American Naturalist,
classes, ordines, genera, species cum characteribus, dijferentiis, 27, 659-62.
synonymis, locis. Edito decima, reformata. Stockholm: Laurentii Scott, W. B., & Jepsen, G. L. (1936). The mammalian fauna of the White
Sal vii. River Oligocene, Part I: Insectivora and Carnivora. Transactions of
(1766). Systema natura per regna tria naturae sedundum classes, or- the American Philosophical Society, 28, 1-153.
dines, genera, species cum characteribus, differentiis, synonymis, Segall, W. ( 1943). The auditory region of arctoid carnivores. Publications of
locis. Edito duodecima. Stockholm: Laurentii Salvii. the Field Museum of Natural History, Zoological Series, 29, 33-59.
Macdonald, J. R. (1963). The Miocene faunas from the Wounded Knee Simpson, G. G. (1945). The principles of classification and a classification
area of western South Dakota. Bulletin of the American Museum of of mammals. Bulletin of the American Museum of Natural History,
Natural History, 125, 139-238. 85, 1-350.
Martin, H. T. (1928). l\vo new carnivores from the Pliocene of Kansas. Stevens, M. S. (1991). Osteology, systematics, and relationships of earliest
Journal of Mammalogy, 9, 233-6. Miocene Mesocyon venator (Carnivora: Canidae). Journal ofVene-
Matthew, W. D. ( 1899). A provisional classification of the freshwater Tertiary brate Paleontology, 11, 45-66.
of the west. Bulletin of the American Museum of Natural History, Stirton,R. A., & VanderHoof, V. L. (1933). Osteoborus, anew genus of dogs,
12, 19-75. and its relations to Borophagus. University ofCalifornia Publications
(1902). New Canidae from the Miocene of Colorado. Bulletin of the in Geological Sciences, 23, 175-82.
American Museum of Natural History, 16, 281-90. Tanner, L. G. ( 1974 ). Notes regarding skull characteristics of Oxetocyon cus-
(1918). Contributions to the Snake Creek fauna with notes upon the pidatus Green (Mammalia, Canidae). Transactions of the Nebraska
Pleistocene of western Nebraska. Bulletin of the American Museum Academy of Sciences, 2, 66-9.
of Natural History, 38, 183-229. Tedford, R.H. (1978). History of dogs and cats: a view from the fossil record.
( 1924 ). Third contribution to the Snake Creek fauna. Bulletin of the In Nutrition and Management of Dogs and Cats, pp. 1-10. St. Louis:
American Museum of Natural History, 50, 59-210. Ralston Purina.
(1930). The phylogeny of dogs. Journal of Mammalogy, 11, 117-38. Tedford, R.H., Barnes, L. G., & Ray, E. E. (1994). The early Miocene
McGrew, P. 0. (1938). Dental morphology of the Procyonidae with a descrip- littoral ursoid carnivoran Kolponomos: systematics and mode of life.
tion of Cynarctoides, gen. nov. Field Museum of Natural History, In Contributions in Marine Mammal Paleontology Honoring Frank
Geological Series, 6, 323-39. C. Whitmore, Jr., eds. A. Berta & T. Demere. Proceedings of the San
( 1944). The Aelurodon saevus group. Field Museum of Natural History Diego Society of Natural History, 29, 11-32.
Publications in Geology, 8, 79-84. Tedford, R. H., & Frailey, D. (1976). Review of some Carnivora (Mam-
Merriam, C. W. (1906). Carnivora from the Tertiary formations of the John malia) from the Thomas Farm local fauna (Hemingfordian: Gilchrist
Day region. University of California Publications in Geological Sci- County, Florida). American Museum Novitates, 2610, 1-9.
ences, 5, 1-64. Turner, N. H. (1848). On the evidence of affinity afforded by the skull of car-
(1913). Notes on the canid genus Tephrocyon. University of California nivorous mammals. Proceedings ofthe Zoological Society ofLondon,
Publications, Bulletin of the Depanment of Geology, 7, 359-72. 1848, 63-88.
(1930). Allocyon, a new canid genus from the John Day Beds of Oregon. VanderHoof, V. L., & Gregory, J. T. ( 1940). A review of the genus Aelurodon.
University of California Publications, Bulletin of the Department of University of California Publications, Bulletin of the Department of
Geological Sciences, 19, 229-44. Geological Sciences, 25, 143-64.
Mivart, S. G. (1890). Dogs, Jackals, Wolves and Foxes. A Monograph of the Van Valen, L. (1964). Nature of the supernumerary molars of Otocyon. Jour-
Canidae. London: R.H. Porter. nal of Mammalogy, 45, 284-6.
Munthe, K. (1989). The skeleton of the Borophaginae (Carnivora, Canidae): Van Valkenburgh, B. (1985). Locomotor diversity within past and present
morphology and function. University of California Publications in guilds of large predatory mammals. Paleobiology, 11, 406-28.
Geological Sciences, 133, 1-115. ( 1988). Trophic diversity in past and present guilds of large predatory
Peterson, 0. A. (1906). The Miocene beds of western Nebraska and eastern mammals. Paleobiology, 14, 155-73.
Wyoming and their vertebrate fauna. Annals ofthe Carnegie Museum, (1989). Carnivore dental adaptations and diet: a study of trophic diversity
4, 21-72. within guilds. In Carnivore Behavior, Ecology and Evolution, ed.
(1924). Discovery of fossil mammals in the Brown's Park formation J. L. Gittleman, pp. 410-36. Ithaca: Cornell University Press.
of Moffat County, Colorado. Annals of the Carnegie Museum, 15, (1991). Iterative evolution ofhypercarnivory in canids (Mammalia: Car-
299-304. nivora): evolutionary interactions among sympatric predators. Pale-
(1928). The Brown's Park formation. Memoirs of the Carnegie Museum, obiology, 17, 340-62.
11, 87-121. Voorhies, M. R. (1969). Taphonomy and population dynamics of an early
Petter, G. (1964). Origine du genre Otocyon (Canidae africain de la sous- Pliocene vertebrate fauna, Knox County, Nebraska. University of
famille des Otocyoninae). Mammalia, 28, 330-44. Wyoming Contributions to Geology, Special Paper, 1, 1-69.
(1966). Cynodictis, canide Oligocene d'Europe. Region tympanique et Wang, X. (1993). Transformation from plantigrady to digitigrady: functional
affinities. Annals Palaeontologie, Vertebrata, 52, 3-19. morphology of locomotion in Hesperocyon (Canidae: Carnivora).
Radinsky, L. (1973). Evolution of the canid brain. In Brain, Behavior, and American Museum Novitates, 3069, 1-23.
Evolution, ed. W. Riss, pp. 169-202. Basal: S. Karge. (1994). Phylogenetic systematics of the Hesperocyoninae (Carnivora:
Romer, A. S. (1966). Vertebrate Paleontology. Chicago: University of Canidae). Bulletin of the American Museum ofNatural History, 221,
Chicago Press. 1-207.
Romer, A. S., & Sutton, A. H. (1927). A new arctoid carnivore from the Wang, X., & Tedford, R.H. (1992). The status of genus Nothocyon Matthew,
lower Miocene. American Journal of Science, 214, 459-64. 1899 (Camivora): an arctoid not a canid. Journal of Vertebrate Pa-
Savage, D. E., & Russell, D. E. ( 1983 ). Mammalian paleofaunas of the world. leontology, 12, 223-229.
Reading, Mass.: Addison-Wesley. (1994). Basicranial anatomy and phylogeny of primitive canids and
Schreber, J.C. D. (1775). Die Saugethiere in Abdildungen nach der Nature closely related miacids (Camivora: Mammalia). American Museum
mit Beschreibungen. W Walther, Leipzig, 3, 281-590. Novitates, 3092, 1-34.
 Canidae 143

Wayne, R. K., & O'Brien, S. J. (1987). Allozyme divergence within the (1977). A history of savanna vertebrates in the New World, Part I: North
Canidae. Systematic Zoology, 36, 339-55. America. Annual Review of Ecology and Systematics, 8, 355-80.
Webb, S. D. (l 969a). The Burge and Minnechaduza Clarendonian mammal Werdelin, L. (1989). Constraint and adaptation in the bone-cracking canid
faunas of northcentral Nebraska. University of California Publica- Osteoborus (Mammalia: Canidae). Paleobiology, 15, 387-401.
tions in Geological Sciences, 78, 1-191. Wortman, J. L., & Matthew, W. D. (1899). The ancestry of certain members
(1969b). The Pliocene Canidae of Florida. Bulletin of the Florida State of the Canidae, the Viverridae and the Procyonidae. Bulletin of the
Museum, Biological Sciences, 14, 273-308. American Museum of Natural History, 12, 109-39.
 8 Procyonidae
JON A. BASKIN

INTRODUCTION

The family Procyonidae includes the subfamilies Procyoninae, Ail-

urinae, and Simocyoninae. Procyonines are the raccoons, coatis,
ringtail cats, and their Recent and fossil relatives. Ailurines are the
Old World lesser or red panda and fossil relatives; simocyonines
(Beaumont, 1982) are a restricted group of extinct procyonids re-
lated to the ailurines. These three subfamilies all originated in the
Old World, where the Ailurinae and Simocyoninae have almost
their entire record. In contrast, the major radiation of the Procyoni-
nae took place in the New World. All three subfamilies have a very
limited fossil record, even in North America (Baskin, 1982). Twelve
genera occur in the North American Tertiary. Many of these taxa
are poorly represented and are known only from isolated lower or
upper dentitions.

DEFINING FEATURES OF THE Figure 8.1. Restoration of Simocyon, by Brian Regal.

FAMILY PROCYONIDAE

Primitive procyonids have a basicranium with a Type A auditory hypoconulid. The ml primitively has a relatively prominentmeta-
bulla (Hunt, 1974) in which the ectotympanic is large and flat- conid and a basined talonid. Ml retains a postprotocrista and meta-
tened and the entotympanic is slightly inflated. There is a shallow conule, the primitive arctoid state.
suprameatal fossa in the squamosal, and an alisphenoid canal is Very little postcranial material is ascribed to fossil procyonids.
present. This morphology is primitive for arctoid and occurs in the Living procyonids are semiplantigrade to plantigrade and have five
Ailurinae and probably the Simocyoninae. In the procyonids, the toes on each foot. Bassariscus astutus, the living ring-tailed cat, has
posterior carotid foramen is more separated from the posterior lac- semiretractile claws.
erate foramen than in ursids. Procyoninae are derived relative to
other procyonids in having a Type B auditory bulla (Hunt, 1974)
with an inflated entotympanic, greater separation of the posterior SYSTEMATICS
lacerate and posterior carotid foramina, a deep suprameatal fossa,
and loss of the alisphenoid canal. A deep suprameatal fossa evolved
SUPRAFAMILY
independently in or is primitive for the Mustelidae.
The primitive procyonid dental formula is 13/3, Cl/l, P4/4, M2/2, Infraorder Arctoidea
as in primitive mustelids. The procyonid dentition (Figure 8.2) is The Arctoidea contains two superfamilies: Ursoidea (Family Ursi-
distinguished from that of mustelids in having the second molars dae and the Pinnipedimorpha) and Musteloidea (Families Mustel-
enlarged. The m2 has a short trigonid and an elongate talonid with idae and Procyonidae). Shared-derived characters of the Arctoidea

144
 Procyonidae 145

Figure 8.2. Dentition of Procyonidae. A. Bassariscus casei, left maxillary fragment, lateral and occlusal views, left mandible, lateral and occlusal views
(from Hibbard, 1952). B. Arctonasuafioridana, left mandible, lateral and occlusal views (from Baskin, 1982). C. Simocyon marshi, left mandible, lateral
and occlusal views (from Thorpe, 1921). Scale= 2 cm for Band C; 1 cm for A.

include complete ossification of the auditory bulla, loss of M3, that gave rise to the Mustelidae and Procyonidae. However, most
and reduction or loss of the trigonid on m2 by loss of the para- of these primitive taxa can be allocated either to the Procyonidae
conid, enlargement of the metaconid, reduction of the protoconid, or Mustelidae. A more restricted assemblage of late Eocene to me-
and development of an anterior cingulum. Primitive arctoid have an dial Oligocene taxa (assigned to the Mustelavinae in the Mustelidae
amphictid-type suprameatal fossa (shallow depression in squamosal) chapter, this volume, Chapter 9) probably represents the musteloid
in the auditory canal (Schmidt-Kittler, 1981), a large flattened ecto- stem group. The alisphenoid canal (present in mustelavines and
tympanic extended to form a tubular meatus, a prominent mastoid ailurines) was lost independently in procyonids and mustelids, as in
process separated from paroccipital process, and a scapula with a the ursid Ailuropoda (the giant panda) and phocids (seals).
postscapular fossa (Tedford, 1976), some or all of which may be Alternatively, the superfamily Musteloidea may not be a mono-
primitive for the Caniformia (Cynoidea plus Arctoidea). phyletic group. Some or all of the diagnostic characters could have
evolved in two or more independent radiations from primitive ur-
Superfamily Musteloidea soids such as Amphicynodon. The fossil record indicates that the
The Procyonidae and Mustelidae form the superfamily Musteloidea, suprameatal fossa has been independently lost in several groups of
a group united by shared-derived characters of the basicranium arctoid. Additionally m3 may also have been independently lost
and dentition (Tedford, 1976). Musteloids share a common ances- in different lineages radiating from the amphicynodontines. How-
try with the ursoids. In North America, ursoids (Subparictis and ever, described species of Amphicynodon are too specialized to be
Campylocynodon) and musteloids (Mustelavus) first appear in the ancestral to Amphictis (Cirot and Bonis, 1992) or to the mustelids.
Chadronian (late Eocene). In Europe, ursoids (Amphicynodon) and
musteloids (Mustelictis) first appear in the early Oligocene imme-
INFRAFAMILY
diately following the Grande Coupure (Remy et al., 1987).
Musteloids are derived relative to primitive ursoids in having en- Figure 8.3 is a cladogram of hypothesized relationships of the gen-
largement and inflation of the caudal entotympanic, increased sep- era discussed in this chapter. Baskin ( 1982, 1989) discusses the
aration of the posterior carotid foramen from the posterior lacerate relationships of North American procyonines with the European
foramen, and loss of m3. Schmidt-Kittler (1981) places most early early Miocene Broiliana, the most primitive procyonine. In this
Miocene and all older musteloids in an undifferentiated stem group chapter, these Ailurinae and Procyoninae are tentatively included
 Jon A. Baskin 146

Figure 8.3. Interrelationships of

~ c: North American genera of Tertiary
(.)
c:
~
Procyonidae. Key to characters:
c: c:
.f:!2
lo.,.; ~ ~ § <tS PROCYONIDAE. (I) m2 with short
.f:!2 Cl) (.) (.)
~

.S:2 2 :::3 ~ ~
<tS
Cl)
Cl)
.f:!2
lo.,.;
..c::
(.)
0
Cl)
<tS
c:
e :::3
Cl) c:
trigonid and elongate talonid. (2)
ml with large metaconid. (3) ml
~
0.. <tS
..c:: (.) (.)
.E c:
::::::: <tS <tS <tS with low trigonid. (4) P4 with
.o 0 ..c Cl) 0.. .E :::3
~ 8: ~
~ a..e
~
(.)

~
.+::;

~
(.)
Cl)
<tS
ca ~
LU
~
~
e
a.. 8: ~
Cl)
e
a..
internal cingulum. (5) P4 with small
parastyle. UNNAMED TAXON.
(6) M2 enlarged. UNNAMED
- 33 34 TAXON OR AILURIDAE. (7) m2
o.11Y with hypoconulid and very elongate
,,~'13,14 31,32
talonid. (8) Tall posteriorly curving
- 7,8,9 coronoid process. (9) Long tubular
auditory meatus (?plesiomorphic for
arctoids). AILURINAE. (10) Molar-
iform premolars. (11) ml with para-
- 24,25,26
conid and metaconid close together.
1,22,23 (12) Cheek teeth with accessory
cuspules. SIMOCYONINAE. (13) P4
with posterior small protocone and
15, 16, 17, 18, 19 narrow shelf. (14) ml with bladelike,
open trigonid. PROCYONINAE. (15)
Suprameatal fossa deep. (16) Alisphe-
noid canal absent (?synapomorphy
with Mustelidae). (17) Bulla inflated
6 (?synapomorphy with Mustelidae).
( J 8) m2 hypoconulid internal. (19) m2
- 1,2,3,4,5
anterior cingulum reduced. (20) ml
with hypoconulid. (21) P4 with en-
larged protocone. (22) P4 with
shortened metaconid blade. (23) ml with paraconid and metaconid close together. (24) P4 with enlarged parastyle. (25) M2 with broadened protocone.
(26) m2 with hypoconulid posterior forming heel. (27) mI trigonid widened. (28) m2 hypoconid larger than protoconid. (29) Upper molars with reduced
internal cingulum. (30) Angular process of dentary above level of tooth row. (31) Upper molars with external cingulum very reduced to absent. (32) Ml
with metacone to metaconule crest. (33) ml paraconid and metaconid very close together. (34) P2-3/p2-3 tall, anteriorly posteriorly compressed.

in the Procyonidae because of their dental characteristics. Several in front of the locality (e.g., ?CPIO 1) mean the taxon is questionably
authors have placed the Ailurinae in a family Ailuridae as the sister known from that locality; thus implying some doubt that the taxon
taxon of the Procyonidae. Others include the Ailurinae in the Ursi- is actually present at that locality, either at the genus or the species
dae. The absence of good basicranial material makes the taxonomic level. An asterisk (*) indicates the type locality.
placement of the Simocyoninae even more difficult to resolve (e.g.,
Schmidt-Kittler, 1981). The possibility that ailurines and simocy-
onines should be placed in a distinct family Ailuridae is discussed PROCYONIDAE
in the Included Genera section.
INCERTAE SEDIS

Amphictis Pomel, 1853

INCLUDED NORTH AMERICAN GENERA IN THE Type species: Amphictis ambiguus Gervais, 1876.
FAMILY PROCYONIDAE Type specimen: MNHN QU9243.
Characteristics: An alisphenoid canal is present in a skull of
Figures 8.4 and 4.5 present the stratigraphic range and geographic A. ambiguus from the late Oligocene of Quercy (Bonis,
distribution of these genera. 1976). The suprameatal fossa is shallow. M2 in described
The locality numbers listed for each genus refer to the list of uni- late Oligocene specimens is small, but three rooted, with
fied localities in Appendix I. The acronyms for museum collections the paracone and metacone small and close together.
are listed in Appendix Ill. Average length of m 1: 9 .1 mm.
The locality numbers may be listed in a couple of alternative Included North American species: ?Amphictis sp. only,
ways. Parentheses around the locality (e.g., [CPIO 1]) mean the taxon known from locality CP105 only.
in question at that locality is cited as an "aff." or "cf." the taxon
in question. Parentheses are usually used for individual species, Comments: Amphictis, from the late Oligocene and early
thus implying the genus is firmly known from the locality, but the Miocene of Europe, is the earliest known procyonid. It is
actual species identification may be questionable. Question marks noted as an early Hemingfordian (late early Miocene)
 Procyonidae 147

immigrant to North America (Tedford et al., 1987), but Bassariscus Coues, 1887
the material from the Runningwater Formation (locality Type species: Bassariscus astutus Lichenstein, 1830, the ex-
CP105) is not described. Bonis (1976) considersAmphictis tant ring-tailed cat or cacomistle.
a primitive musteloid closely related to amphicynodontines Characteristics: P4 with a prominent internal cingulum and
and near the stem of the procyonid and mustelid radiations, hypocone in all but B. parvus, the most primitive species;
as does Schmidt-Kittler (1981). However, the elongate Ml and M2 with more posteriorly directed hypocones
talonid on m2 is a procyonid synapomorphy. Beaumont than in Probassariscus; m 1 with the entoconulid merged
(1976, 1982) states that the reduced M2 of Amphictiscould with the entoconid and with a prominent hypoconulid;
have given rise to the larger M2 present in simocyonines. and m2 with an elongate but narrow talonid and with the
Amphictis displays a large amount of dental variabil- hypoconulid situated on the posteromedial margin of the
ity (Beaumont, 1976) and may be a paraphyletic genus. talonid. Bassariscus is probably derived from P. antiquus,
Certain A. antiquus (Beaumont, 1976, Figure lb, 1982), which has the entoconid absent on m2.
with m2 very elongate and with a hypoconulid, may be Average length of ml: 6.1-8.6 mm.
more closely related to the Simocyoninae and perhaps the Included Tertiary species: B. parvus (known from localities
Ailurinae. Other A. antiquus (Beaumont, 1976, Figure la) CC26B, *NB23B, CPI 14A, CPI 14B, CP114D, NPl 1); B.
may be more closely related to the Procyoninae. ogallalae (localities *CP116C, CP123D); B. casei (local-
ities NB13C, SPIF, *CP128C, [PN4]); ?B. lycopotamicus
PROCYONINAE (locality *PN7 or *PN12).
Bassariscus sp. is also known from localities CC17G,
Characteristics: Alisphenoid canal absent; entotympanic inflated;
CC32A, NB6E, NB7D, NB19C, SB9, SBIO, SB12,
suprameatal fossa deep; P4 with at least a narrow internal cingulum
SB34C, SPIA, CP121, PN9B, PN14.
and a small parastyle. Living taxa include raccoons, coatimundis,
olingos, ring-tailed cats or cacomistles, and kinkajous.
Edaphocyon Wilson, 1960
Probassariscus Merriam, 1911 Type species: Edaphocyon pointblankensis Wilson, 1960.
Type species: Probassariscus matthewi Merriam, 1911. Type specimen: TMM 31190-76.
Type specimen: UCMP 12539. Characteristics: A medium-sized procyonid, known only
Characteristics: Differs from European Broiliana, in possess- from upper dentitions. It is derived with respect to bas-
ing a better developed parastyle on P4 and a more poste- sariscines in its larger size, more inflated dental cusps,
riorly placed hypocone and stronger metaconule on Ml, enlarged and more posteriorly placed protocone on P4,
derivt;d characters shared with all other New World pro- a reduced metacone blade on P4, better developed exter-
cyonids. P4 has a narrow internal cingulum with hypocone nal and internal cingula on Ml and M2, and an elongate
absent or incipient. Probassariscus is primitive with respect premaxilla. These synapomorphies are shared with Arc-
to Bassariscus in having ml with a minute hypoconulid tonasua, Cyonasua, Nasua, and Procyon (Baskin, 1982).
and with entoconid and somewhat smaller entoconulid Edaphocyon may be the stem taxon for these procyonines.
close together on lingual side of talonid. The m2 has been Average length of ml: unknown. Length of P4: 8.7 mm.
described as having a paraconid, but this cusp is a swollen Included species: E. pointblankensis (known from locality
anterointernal cingulum (Baskin, 1989). This character is *GC4C); E. lautus (locality *CP105).
also present in Broiliana and may be a synapomorphy for
Arctonasua Baskin, 1982
these two genera, or may represent the primitive character
state for the procyonines. Type species: Arctonasuafloridana Baskin, 1982.
Broiliana, Probassariscus, and Bassariscus have a rela- Type specimen: UF 32022.
tively primitive, arctoid cusp pattern on P4-M2 that is Characteristics: A medium to large procyonid having teeth
broadly similar to the primitive Oligocene canid Hespe- with bulbous cusps. It differs from all other procyonids in
rocyon. The upper carnassial is bladelike, the upper mo- having P4 with the metacone reduced to absent and m 1 with
lars are triangular with well-developed pre- and postpro- a very closed trigonid with the paraconid and metaconid
tocristae and a small metaconule. The ml has a relatively closely appressed, with a tendency toward reduction or
open carnassial blade and a prominent metaconid. loss of the paraconid. It is derived with respect to Edapho-
Average length of ml: 6.7-7.7 mm. cyon in having P4 with a better developed parastyle, a
Included species: P. matthewi (=Rhapsodus tedfordi) (known greatly reduced paracone blade, and a larger more posteri-
from localities *NB 18, CPI 11 ); P. antiquus (localities orly situated protocone, and more quadrate upper molars.
?CC32B, ?CC38, *CPI IO, [CPll I], CPI 14A). Arctonasua and the South American Tertiary genus Cyona-
sua share eight possible synapomorphies indicating they
Comments: P. matthewi can be distinguished from are sister groups (Baskin, 1982). These include Ml with
P. antiquus by the presence of an entoconid on m2 and a reduced hypocone; P4 with a more posteriorly placed
by broader premolars with better deveioped cingula. protocone and a reduced paracone-metacone blade; and
 Jon A. Baskin 148

ml with a reduced paraconid and a well-developed, sepa- m 1 with the protoconid and metaconid subequal; and the
rate entoconulid. Arctonasua differs from Cyonasua in the angular process of the mandible above level of tooth row.
retention of a small hypocone on the upper molars, having Average length of ml: 10.5 mm.
m2 with the protoconid smaller than the metaconid, and Included species: P. biradica (known from locality *GCl IA).
the absence of an enlarged anterointernal cingulum on m 1, A questionable species of Paranasua is also known from
all primitive characters. Cyonasua may be derived from a locality CP!23C.
form similar to the primitive A. minima, which is known
from the early Barstovian of Nebraska. Procyon Storr, 1780
Average length of ml: 11.2-15.2 mm. Type species: Procyon lotor (Linnaeus, 1758) (=Ursus lo-
Included species: A. floridana (known from locality tor), the extant raccoon.
*GCl lA); A. gracilis (localities GC4C, *CP114A); A. Characteristics: A relatively short-faced procyonine with
minima (locality *CPl 11 ); A. eurybates (localities GC l 3B, anteriorly-posteriorly compressed premolars; Pl/pl single
*CPI 15D); A. fricki (locality *SPlD). rooted; P4 with a large hypocone and a very small meta-
Arctonasua sp. is also known from localities GC12, cone; Ml with the external cingulum reduced to absent;
CP90A, CP114C. ml with the paraconid and metaconid close together; and
m2 elongate.
Protoprocyon Linares, 1982 Av(llrage length of ml: 11.1-13.6 mm (P. rexroadensis).
Type species: Protoprocyon savagei Linares, 1982. Included Pliocene species. P. rexroadensis only (known from
Type specimen: F:AM 25210. localities SPIH, *CP128C).
Characteristics: A small procyonid about the size of Bas- Procyon sp. is also known from localities GC13B, GC14,
sariscus that is known only from a lower jaw; p2 and p3 GC16, GC17, CC25B, CC40, NBl3C, and questionably
are subequal; p4 is slender and broadens only slightly pos- known from localities CP128B (=Bassariscus rexroaden-
teriorly; m 1 has a low trigonid with the paraconid separated sis), PN4.
from the subequal metaconid by a narrow valley, a wide
talonid, and the entoconid and entoconulid close together; Nasua Storr, 1780
and m2 with the protoconid connected to the slightly larger Type species: Nasua nasua (Linnaeus, 1766) (= Viverra na-
metaconid by an anterior cingulum, the hypoconid larger sua), the extant coatimundi.
than the protoconid, and a very elongate talonid with a Characteristics: A long-faced procyonine with small, narrow
posteromedian hypoconulid. anterior premolars; Pl/pl occasionally double rooted, P4
Average length of ml: 6.9 mm. with a large hypocone and a small metacone and with the
Included species: P. savagei (=Lichnocyon phoenicus external cingulum absent; M 1 with the external cingulum
Baskin, 1982) only, known from locality *CP116B only. absent; ml with the paraconid and metaconid very close
together; and m2 elongate.
Comments: Contrary to previous interpretation (Baskin, Average length of ml: 10.7 mm (Hemphillian Nasua).
1982), Protoprocyon is not the sister taxon of the extant Included Pliocene species: N. pronarica (known from locality
Neotropical genus Bassaricyon (the olingo). Protoprocyon *SPlF). A questionable species of Nasua is also known
is closely related to Paranasua, Nasua, and Procyon on from locality GC13B.
the basis of the shared derived m2 morphology. The ml
morphology, with the wide talonid, is similar to Procyon
and Nasua. Bassaricyon has m2 with the primitive procy- AILURIDAE OR UNNAMED GROUP
onid morphology, that is, the talonid is not greatly elon- Characteristics: Arctoid with the alisphenoid canal present (a prim-
gate, the hypoconid is smaller than the protoconid, and itive arctoid characteristic); auditory bulla only slightly inflated
the hypoconulid is posterointernal. Decker and Wozencraft and relatively short (primitive musteloid characteristic); m3 absent
(1991) conclude that Potos (the kinkajou) and Bassaricyon (primitive arctoid characteristic); M2/m2 enlarged to greatly en-
(the olingo) are the sister group of Bassariscus, Nasua larged; and m2 with a hypoconulid.
(coatimundis), and Procyon (raccoons). Comments: The taxonomic placement of Ailurus (the Pleistocene
to Recent, southeast Asian, red or lesser panda) is controversial.
Paranasua Baskin, 1982 Consensus molecular and karyotypic phylogeny suggests it is most
Type species: Paranasua biradica Baskin, 1982. closely related to procyonines and that it separated from the New
Type specimen: UF 24829. World procyonines slightly after the time the latter split from the ur-
Characteristics: A medium-sized procyonine with large sids (O'Brien et al., 1985). Typically, lesser pandas have been classi-
double-rooted first premolars; P4 with the metacone fied in the Procyonidae because of dental similarities, although Gins-
smaller than the parastyle, a prominent posteriorly placed burg ( 1982) concludes that the teeth of ail urines are most similar to
protocone, and a hypocone; M 1 with an external cingulum ursids. Ail urines and simocyonines are tentatively united on the basis
and a reduced internal cingulum; m2 slightly smaller than of their enlarged second molars. The basicranium of simocyonines is
 Procyonidae 149

very poorly known. If the loss of the alisphenoid canal is a musteloid Type specimen: LACM/CIT 2747.
synapomorphy, then it may be necessary to recognize the Ailuridae Characteristics: Four premolars; P4 only slightly longer than
as a separate family. However, it is likely that the loss occurred inde- Ml, with a very small parastyle, a narrow internal shelf
pendently. Ailurines and simocyonines may have been derived from that extends posteriad as a cingulum, and a very small
the primitive procyonid Amphictis, which retained the alisphenoid protocone; Ml subtriangular with a low protocone con-
canal. Another possibility is that this group is a specialized ursoid nected to a small metaconule, and a prominent posteroin-
offshoot of the amphicynodontines that independently lost m3. ternal hypocone. Cl has a lateral groove, a characteristic
of Alopecocyon and Simocyon.
AILURINAE Average length of ml: unknown. Average length of P4: 11.6
mm.
Characteristics: Cheek teeth with numerous accessory cuspules; mo-
Included species: A. leardi only, known from locality
lariform premolars, especially P4/p4; Ml-2 with the metaconule
*CCI 7H only.
subequal in size to the paracone and metacone and with external
and internal cingular cusps; ml with a low trigonid and with the Comments: This taxon is known from a palate fragment with
paraconid and metaconid close together; m2 approximately equal teeth that strongly resemble those of Alopecocyon.
in length to ml; skull very convex; muzzle short; coronoid process
of mandible high and curving anterior! y. Simocyon Wagner, 1858
Type species: Simocyon robustus Wagner, 1858; junior syn-
Parailurus Schlosser, 1899 onym of S. primigenius (Roth and Wagner, 1854) (=Gulo
Type species: Parailurus anglicus (Dawkins, 1888) (=Ailurus primigenius).
anglicus). Type specimen: In the collections of the Palaeontological
Type specimen: In the collections of the Museum of the York- Museum at Munich.
shire Philosophical Society, Great Britain. Characteristics: Pl small; P2-3/pl--3 reduced to absent; P4
Characteristics: Parailurus is distinguished from Ailurus in with a very large parastyle and a low shelflike protocone
having less complete molarization of the premolars; P4 and opposite the very tall paracone; M2 relatively large; p4 with
Ml with the length equal to or greater than the width; ml a prominent posterior accessory cusp; ml large, with an
with the trigonid higher than the talonid; and poorer devel- open carnassial blade, a small metaconid on the posterior
opment of accessory cusps of the cheek teeth. Parailurus is flank of the large protoconid, a low trenchant hypoconid,
a late Pliocene immigrant from Eurasia and is represented a very slightly basined talonid, and the metaconid con-
in North America by a single upper molar. nected to internal cingulum of talonid; m2 very long; and
Averag~ length of ml: unknown. Length of Ml: 14.0 mm mandible stout. The presence or absence of an alisphenoid
(North American Parailurus). canal can not be determined (Beaumont, 1982).
Included North American species: Parailurus sp. only, known Average length of ml: 22.5 mm.
from locality PN4 only. Included North American species: S. marshi only, known
from locality *PN12 only.
SIMOCYONINA E
Comments: This large procyonoid is well represented in Eu-
Characteristics: Procyonids with a tendency to develop an open rope, but is known in North America only by a lower jaw
trigonid with an anteriorly placed, bladelike paraconid on ml. The (Thorpe, 1921).
subfamily Simocynoninae is used in the restricted sense of Beaumont
(1982) to include Simocyon, Alopecocyon, and at least some species
of Amphictis. Actiocyon is also referred to this subfamily. In the BIOLOGY AND EVOLUTIONARY PATTERNS
past, this group has been used as a taxonomic scrap basket for many
canids and amphicyonids with a trenchant talonid on the ml. True Procyonidae first appear in the late Oligocene of Europe. Their ear-
simocyonines are poorly known and are tentatively included in the liest record in North America is from the upper Harrison Formation,
Procyonidae. The lineage from the Orleanian Amphictis aff. an- probably from the late Arikareean (earliest Miocene). The specimen
tiquus to Alopecocyon to Simocyon established by Beaumont, in (AMNH 81029) is too damaged for generic identification. Procy-
which the ursoid characters of Simocyon evolve independently with onines are predominantly a North American group, first appearing
size, is reasonable. However, the structure of the basicranium in on that continent in the Hemingfordian (late early Miocene), but
these three genera needs to be accurately determined to interpret having their origin in the early Miocene of Europe. They also gave
the relationships of this group. The simocyonines may share a com- rise to a late Miocene to Pliocene South American group (Baskin,
mon ancestry with the ailurines through those species of Amphictis 1982, 1989). Procyonines today have their greatest diversity in the
having m2 with a very long talonid. Neotropics, and the fossil record suggests this was true in the past as
well. In North America, simocyonines and ail urines are known from
Actiocyon Stock, 1947 rare, isolated specimens and apparently were not able to establish
Type species: Actiocyon leardi Stock, 1947. themselves successfully on that continent. Even the more abundant
 Jon A. Baskin 150

w
z L L.81.
1.8 c:
w 2.4 -0
-8 ~I
E.BI
0
0 E Lot.Hp.
4.5
1-----------1-------f-
c:g_ _____ ...
:i 5.2
§ -Q. :::::---_--=c;:=-_-------1
_g__ =le
L.Hp.
a. 6.0 u
e
L
L.E.Hp.
E.E.Hp.

L,CI.
7.0

8.8
9.5
i,-_el=
Q
~·?- ~ _gc::: -8-
-~ -e-
w
E.CI.
11.0
a: ~ o:i:: Cii
z L.L.Ba.
w 12.5
0 M
E.L.Ba.
0 14.0

:2 E.Ba.

5.8 ~
L.Hm. .~
E.Hm.
17.5

18.8
-a.1-
e -
E LL.Ar. q:
19.2 L-....J
E.L.Ar. Procyoninae Ailurinae Simocyoninae
23.0

PROCYONIDAE AILURIDAE
w LE.Ar.
z
w L
0 27.7
0
(!' E.E.Ar.
:i 29.4
0
Wt.
E 31.9
Or.
33.4
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Ou

39.5

L.Un.

w 41.3

z
w M
E.Un.

0 45.9
0
w L.Br.
46.7

M.Br.
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
55.0
E.E.Wa.
55.5
Ck.3 55.6
w Ck.2
Ck.1
55.8
z Tl.6
56.0
56.2
w L Tl.5
Tl.4
57.0
0 Tl.3
57.3
0
w Tl.2
Tl.1
..J
<( To.3
a. To.2
To.1
E Pu.3
Pu.2
Pu.1
Pu.O

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronlan, Du.= Duchesnean, Un. = Uintan, Br.= Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti.= Tiffanian, To.= Torrejonian, Pu. = Puercan.
Figure 8.4. Temporal ranges of North American procyonid genera.
 Procyonidae 151

procyonines have the poorest fossil record of any carnivore fam- Hibbard, C. W. (1952). A new Bassariscus from the upper Pliocene of
ily, perhaps because they mainly inhabited environments not well Kansas. Journal of Mammalogy, 33, 379-81.
represented in the fossil record. Hunt, R. M., Jr. (1974). The auditory bulla in Carnivora: an anatomical basis
Living Procyoninae and Ailurinae are scansorial to arboreal. Po- for reappraisal of carnivore evolution. Journal of Morphology, 143,
21-76.
tos, the kinkajou, is a frugivore and Ailurus, the lesser panda, em-
Lichenstein, H. (1830). ErHiuterungen der Nachrichten des Fran. Her-
phasizes bamboo shoots in its diet. Their herbivory is reflected in nandez von den vierfiissigen Thieren Neuspaniens. Abhandlugen
their specialized dentitions. The remaining procyonines are mainly der Koniglichen preussishen Akademie der Wissenschaften, Berlin,
omnivorous in their diet, with Bassariscus having a preference for 89-128.
animal matter. Fossil procyonines apparently were mainly omniv- Linares, 0. J. (1982). Tres nuevos carnfvoros procionidos f6siles de! Mio-
ceno de Norte y Sudamerica. Ameghiniana, 10, 113-21.
orous as well. Advanced species of Arctonasua had powerful jaws
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundem
and bunodont, fiat-wearing teeth, which indicate they were adapted classes, ordines, genera, species cum characteribus, differentiis, syn-
for bearlike omnivory. Large Arctonasua probably weighed 30 to 40 onymis, locis. Editiodecima, reformata. Stockholm: Laurentii Salvii.
kg. Simocyon (Figure 8.1), the most atypical procyonid, had a short (1766). Systema naturae per regna tria naturae, secundem classes, or-
face, powerful jaws, and well-developed carnassials, all of which dines, genera, species cum characteribus, differentiis, synonymis,
locis. Editio duodecima, reformata. Stockholm: Laurentii Salvii.
indicate a predacious, hyaenalike mode of life. The arrival of Simo-
Merriam, J.C. (1911). Tertiary mammal beds of Virgin Valley and Thousand
cyon in North America coincided with the radiation of Osteoborus, Creek in northwestern Nevada, Part 2: Vertebrate faunas. University
a short-faced hyaenalike canid that is one of the most common car- of California Publications, Department of Geological Sciences, Bul-
nivores in Hemphillian (late Miocene) deposits, which may account letin, 6, 199-304.
for the rarity of Simocyon in North America. O'Brien, S. J., Nash, W. G., Wildt, D. E., Bush, M. E., & Benveniste, R.
E. (1985). A molecular solution to the riddle of the giant panda's
phylogeny. Nature, 317, 140--4.
Pomel, A. (1853). Catalogue Methodique et Descriptzfdes Vertebres Fossiles
ACKNOWLEDGMENTS Decouverts dans le Bassin Hydrographique Superieur de la Loire et
Surtout de son Principal Affluent, I 'Allier. Paris: J. B. Bailliere.
Remy, J. A., Crochet, 1.-Y., Sige, B., Sudre, J., Bonis, L. de, Vianey-
Acknowledgments are listed in the Mustelidae chapter
Liaud, M., Godinot, M., Hartenberger, J.-L., Lange-Badre, B., &
(Chapter 9).
Comte, B. (1987). Biochronologie des phosphorites du Quercy: Mise
a jour des listes fauniques et nouveaux gisements de mammireres
fossiles. Abhandlugen der Miinchner geowissschaften (A), IO,
REFERENCES 169-88.
Roth, J., & Wagner, A. (1854). Die fossilen knocheniiberriste von Pikermi in
Griechenland. Abhandlungen der Koniglich Bayerischen Akademie
Baskin, J. A. (1982). Tertiary Procyoninae (Mammalia: Carnivora) of North der Wissenschaften der Miinchen, 7, 371-464.
America: Journal of Vertebrate Paleontology, 2, 71-93. Schlosser, M. (1899). Parailurus anglicus und Ursus bockhi aus den
(1989). Comments on New World Tertiary Procyonidae (Mammalia, Ligniten von Bar6th-Kopecz. Mittheilungen aus dem Jahrbuch
Carnivora). Journal of Vertebrate Paleontology, 9, 110-17. der Koniglischen Ungarischen geologischen Anstalt, Budapest, 13,
Beaumont, G. de (1976). Remarques pre!iminaires sur le genre Amphictis 67-95.
Pomel (Carnivore). Bulletin de la Societe Vaudoise des Sciences Na- Schmidt-Kittler, N. (1981). Zur Stammesgeschichte der marderverwandten
turelles, 73, 171-80. Raubtiergruppen (Musteloidea, Carnivora). Eclogae Geologicae
(1982). Qu'est-ce que le Plesictis leobensis Redlich (Mammirere, Car- Helvetiae, 74, 753-801.
nivore)? Archives des Sciences (Geneve), 35, 143-52. Stock, C. (1947). A peculiar new carnivore from the Cuyama Miocene,
Bonis, L. de (1976). Decouverte d'un crane d'Amphictis (Mammalia, Car- California. Bulletin of the Southern California Academy of Sciences,
nivora) dans !'Oligocene superieur des Phosphorites du Quercy 46, 84-9.
(Lot). Academie des Sciences de Paris, Comptes-rendus (D), 283, Storr, G. C. C. (1780). Prodromus methodi Mammalium. Tiibingen: Reissian.
327-30. Tedford, R. H. (1976). Relationship of pinnipeds to other carnivores (Mam-
Cirot, E., & Bonis, L. de ( 1992). Revision du genreAmphicynodon, carnivore malia). Systematic Zoology, 25, 363-74.
de !'Oligocene. Palaeontographica, Abteilung A, 220, 103-30. Tedford, R. H., Skinner, M. F., Fields, R. W., Rensberger, J. M., Whistler,
Coues, E. (1887). Bassariscus, a new generic name in mammalogy. Science, D. P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S.
9, 516. D. (1987). Fauna! succession and biochronology of the Arika-
Dawkins, W. Boyd (1888). On Ailurus anglicus, a new Carnivore from the reean through Hemphillian interval (late Oligocene through earliest
Red Crag. Quarterly Journal of the Geological Society of London, Pliocene Epochs) in North America. In Cenozaic Mammals of North
44, 228-31. America, ed. M. 0. Woodbume, pp. 153-210. Berkeley: University
Decker, D. M., & Wozencraft, W. C. (1991). Phylogenetic analysis of Recent of California Press.
procyonid genera. Journal of Mammalogy, 72, 42-55. Thorpe, M. R. (1921). Two new fossil Carnivora. American Journal of
Gervais, P. (1876). Zoologie et paleontologie genera/es, nouvelles recherches Science, (5) I, 477-83.
sur !es animaux vertebres vivants et fossiles. 2e serie. Paris: Arthus Wagner, A. (1858). Geschichte der Urwelt, mit besonderer Beriicksichtigung
Bertrand. der Menschenrassen und des mosaischen Schopfungsberichtes, pt. 2,
Ginsburg, L. (1982). Sur la position systematique du petit panda, Ailu- Leipzig.
rus fulgens (Carnivora, Mammalia). Geobios, memoire special, 6, Wilson, J. A. (1960). Miocene carnivores, Texas Gulf Coastal Plain. Journal
247-58. of Paleontology, 34, 983-1000.
9 Mustelidae
JON A. BASKIN

INTRODUCTION

Recent mustelids include weasels, skunks, badgers, and otters. This

revision places all the extant and a majority of extinct North Ameri-
can, genera in two subfamilies, Mustelinae and Melinae. The fossil
record also includes the extinct Mustelavinae, Oligobuninae, and
Leptarctinae. Beginning in the Oligocene and continuing into the
Pleistocene, North America was repeatedly invaded by Old World
mustelids (Tedford et al., 1987). This chapter discusses 38 mustelid
genera from the North American Tertiary. At least 26 of these 38
Figure 9.1. Restoration of Aelurocyon, by Brian Regal.
genera are immigrants, which makes it very difficult to interpret
mustelid systematics from the North American fossil record alone.
INFRAFAMILY

DEFINING FEATURES OF THE The classification of the Mustelidae requires intensive reexamina-
FAMILY MUSTELIDAE tion to determine the phylogenetic relationships of suprageneric
groupings. Pocock's (1921) classification of Recent Mustelidae rec-
Mustelids (Figures 9.1, 9.2) are very small to medium-sized arctoid ognizes fifteen subfamilies, most of them monogeneric. More typ-
carnivores, generally with low braincases, wide occiputs, short jaws, ically, Recent mustelids are placed in five subfamilies: Mustelinae
small orbits, and forwardly placed carnassials (Radinsky, 1982). The (weasels, etc.), Mellivorinae (honey badger or rate!), Melinac (bad-
paroccipital and mastoid processes are widely separated. The M2 is gers), Mephitinae (skunks), and Lutrinae (otters). Most paleontol-
very reduced to absent. The m2 is reduced to absent, with a short ogists (e.g., Simpson, 1945; Piveteau, 1961) place fossil mustelids
talonid. The following features characterize primitive mustelids: in extant subfamilies, with the exception of the distinctive Miocene
The auditory bulla is moderately inflated; there is a suprameatal Leptarctinae. This has resulted in the production of polyphyletic
fossa in the squamosal; an alisphenoid canal is present; the den- taxonomic groups. Figure 9.3 is a cladogram of hypothesized rela-
tal formula is I3/3, Cl/1, P4/4, M2/2 (as in primitive procyonids); tionships of suprageneric taxa discussed in this chapter. The Sub-
Ml lacks a postprotocrista and metaconule and has a very large families Mustelavinae (represented by Mustelavus) and Oligobun-
parastyle; and ml has a reduced metaconid. The mustelid skeleton inae are used for the more primitive mustelids. The more derived
tends to be long bodied and short legged. Foot structure ranges from mustelids are placed in three subfamilies: Leptarctinae, Melinae,
semiplantigrade to digitigrade, and five toes are present on each foot. and Mustelinae.

SYSTEMATICS INCLUDED NORTH AMERICAN GENERA

IN THE FAMILY MUSTELIDAE
SUPRAFAMILY
Figures 9.4 and 4.5 present the temporal range and geographic
The monophyly of the Musteloidea (Mustelidae plus Procyonidae) distribution of these genera. Figures 9.5 and 9.6 show the North
is discussed in the Procyonidae chapter (this volume, Chapter 8). American biochrons for the genera of Mustelidae, with inferred
152
 Mustelidae 153

Figure 9.2. Comparison of A. Paleomustelid and B. Neomustelid skulls. In Promartes (A) the facial region is not reduced and the auditory bullae are
inflated. Satherium (B) has a shortened facial region. C. the skeleton of Aelurocyon. Skull length is approximately 20 cm. A and C, from Riggs, 1945; B
from Bjork, 1970.

phylogenies indicated by dotted lines linking taxa. In many in- volume, Chapter 8) because the dentition is not as derived as in more
stances, however, the actual sister taxa are Old World genera not advanced procyonids and mustelids.
included in the present study.
Included genera: Mustelavus, Mustelictis, ?Amphicticeps.
The locality numbers listed for each genus refer to the list of uni- Mustelavus is the only North American representative of this sub-
fied localities in Appendix I. The acronyms for museum collections family.
are listed in Appendix III.
Comments: The taxonomic placement of the very primitive,
The locality numbers may be listed in a couple of alternative mainly Oligocene, musteloids requires reassessment of dental and
ways. Parentheses around the locality (e.g., [CPIOl]) mean the taxon basicranial characters of Mustelavus, Mustelictis, Plesictis, Amph-
in question at that locality is cited as an "aff." or "cf." the taxon ictis, and Amphicticeps. The European Mustelictis is distinguished
in question. Parentheses are usually used for individual species, dentally from the procyonidAmphictis by its Ml with a very reduced
thus implying the genus is firmly known from the locality, but the metaconule and smaller second molars. Schmidt-Kittler (1981) in-
actual species identification may be questionable. Question marks cludes Plesictis pygmaeus, P. palmidens, P. robustus, and Pseu-
in front of the locality (e.g., ?CPlO 1) mean the taxon is questionably dobassaris riggsi in Mustelictis and states that an alisphenoid canal
known from that locality; thus implying some doubt that the taxon is present in the genotypic M. piveteaui, contrary to the observation
is actually present at that locality, either at the genus or the species of Lange ( 1970). An alisphenoid canal is present in M. riggsi. Amph-
level. An asterisk (*) indicates the type locality. icticeps (Matthew and Granger, 1924) has M 1 with a large parastyle,
reduced postprotocrista, and no metaconule; M2 very reduced; m3
ARCHAIC MUSTELOIDS absent; the suprameatal fossa very shallow; and the alisphenoid
canal probably present. These characters suggest it is a mustelavine.
MUSTELAVINAE
Mustelavus Clark, in Scott and Jepsen, 1936
Characteristics: An alisphenoid canal is present and M2 is three Type species: Mustelavus priscus Clark, in Scott and Jepsen,
rooted, both characters found in primitive arcto.ids. This paraphyletic 1936.
subfamily is tentatively included in the Mustelidae because of the Type specimen: PU 13775.
reduction of the metaconule and postprotocrista on Ml (mustelid Characteristics: Alisphenoid canal possibly present, but
synapomorphies). Alternatively, Mustelavinae may represent the "damage precludes certain determination" (Clark and
stem taxon for the Superfamily Musteloidea (see discussion in this
Guensburg, 1972, p. 45); shallow suprameatal fossa;
 Jon A. Baskin 154

Q)
co
Cl) c
~ c c c
.~ ::J t5,._ c c t5 c
.f2 ...... c
~

~
.~
~
Cl)

0...
..0
0
.Ql
0
co
a.
Q)
_J
"O
·~
I-
E
..c

~
a.
Q)
-
c
'i:::

_J
::J
~
c
Q)
'i:::
>.
..c
()
.!!2
-
~
Q)
en
::J
~
{!J
co
,
, (46-50) -(50-51)
-(13-15) 28)
/(25-27)/ (43-44)
, -(45)

(31-37)
Leptarctinae - Mustellnae (38-42)
(20-24)

Melinae (30) --

(29)

neomustelids (16-19)

- (7-12)

Mustelidae (1-6)
Figure 9.3. Interrelationships of the tribes of North American Mustelidae.
Key to characters: MUSTELIDAE. (I) M2 reduced. (2) m2 reduced,
with short talonid. (3) Ml with reduced postprotocrista and reduced metaconul
e. (4) Ml with enlarged parastyle. (5) ml with reduced metaconid.
(6) Bullae inflated. UNNAMED TAXON. (7) Alisphenoid canal absent. (8)
Ml postprotocrista and metaconule absent. (9) Ml parastyle very large.
(10) M2 very small, two rooted. (11) ml trigonid opened. (12) palate prolonged
behind molars. OLIGOBUNINAE. (13) ml relatively short with
narrow, indented talonid. (14) 13 large. (15) mandible short and deep. UNNAME
D TAXON: "NEOMUSTELIDS." (16) P4 carnassial notch ab-
sent. (17) M2 absent. (18) Suprameatal fossa very reduced to absent. (19) Postscapul
ar fossa absent. LEPTARCTINAE. (20) Pl/pl absent. (21) Ml
with prominent metacone separate from paracone. (22) ml with prominent
metaconid. (23) ml with notch between metaconid and talonid rim.
(24) ml with elongate talonid. LEPTARCTINI PLUS TAXIDIINI. (25)
Large septate bulla. (26) ml with prominent separate entoconid.
(27) vertically oriented mastoid. TAXIDIINI PLUS MEPHITINI. (28) Invasion
of mastoid by middle ear cavity. Unnamed taxon. (29) Increased sep-
aration of posterior carotid and posterior lacerate foramina. Foramen ovale partially
covered ventrally. MELINAE. (30) ml with elongate cuspidate talonid.
LUTRINI PLUS MELIN!. (31) P4 with broad internal shelf (independently evolved
in other taxa). (32) M 1 with reduced external cingulum indenteJ between
paracone and metacone. (33) ml with closed, low trigonid. (34) ml with expanded
basined talonid. (35) ml metaconi~ connected to talonid (in~e.pendently
evolved in Galictini and in some Mustelini). (36) Infraorbital foramen enlarged.
(37) Ectotympanic flattened and with tubular meatus (meph1tmes also).
MUSTELINAE. (38) MI protocone close to paracone. (39) M 1 protocone expanded
anteroposteriorly. (40) ml metaconid reduced. (41) ml talonid trenchant
(?primitive for neomustelids). (42) m2 single rooted (independently evolved in
melines). ISCHYRICHTINI. (43) ml with reduced metaconid. (44) ml
entoconid absent, talonid semitrenchant. MUSTELINI PLUS GALICTINI. (45)
ml metaconid connected to basined talonid. MUSTELINI. (46) Ml with
median constriction. (47) Ml with reduced stylar shelf. (48) bulla expanded and
inflated posteriorly. (49) foramen ovale faces anteriorly. GALICTINI.
(50) P4 with internal cingulum. (51) ml talonid expanded slightly lingually.

moderately inflated auditory bulla in contact with the base basined and elongate, and the trigonid compressed, with a
of a posteriorly projecting paroccipital process, which has narrow anterior cingulum, a very reduced paraconid, and
a broad ventral connection with mastoid process; secant the protoconid opposite the smaller metaconid.
dentition; four premolars; P4 with an anteriorly placed, Average length of ml: 5.6 mm.
rounded, conical protocone and a deep camassial notch; Included species: M. priscus only, known from locality
Ml with a subtriangular occlusal outline, a moderate-sized *CP83C only.
parastyle, a small paraconule, a very reduced postpro- Mustelavus sp. is also known from locality CP84A.
tocrista that does not connect to the very small metaconule,
and narrow anterointemal and wider posterointemal cin- Comments: Mustelavus has been placed in the Mustelidae,
gula; M2 small and three rooted; p4 with weak external Procyonidae, and most recently (Schmidt-Kittler, 1981)
and internal cingula; ml with a tall trigonid, a low basined with the primitive amphicyonid Cynodictis. Arctoid
talonid, a prominent metaconid separate from the taller synapomorphies of Mustelavus include M3 absent and
protoconid, an offset carnassial blade, a small accessory m2 with a low trigonid, reduced paraconid, and enlarged
cusp at the internal base of paraconid, a small hypoconid meta-conid. The m2 metaconid is more anterior than in
and smaller entoconid, and a notch separating the proto- amphicyonids and canids. Procyonids have a postproto-
conid and hypoconid; m2 small, double rooted, the talonid crista and a metaconule on Ml, but these features are
 Mustelidae 155

primitive for musteloids. The reduction of these features

are present in Plesictis and the Oligobuninae: P4 with car-
in Mustelavus, as well as the loss of m3, reduction of
nassial notch, suprameatal fossa, and presence of M2 (but
M2/m2, and the presence of a strong parastyle on Ml
reduced to two roots).
suggest this genus is a primitive mustelid, although these
characters do not necessarily exclude it from a sister-group
relationship with the procyonids as well. Simpson (1946) OLIGOB UNINAE
tentatively synonymizes Mustelavus with Plesictis,
Characteristics: Oligobunines are small to very large mustelids char-
although he notes that the more primitive species of Plesic-
acterized by P4 with a tall paracone, a camassial notch, and a small
tis (including M. prise us) probably deserve generic distinc-
anteriorly placed protocone; Ml with the postprotocrista and meta-
tion. Mustelavus differs from Plesictis in characters of the
conule absent, a very prominent parastyle, and primitively narrow
dentition and the auditory region (Schmidt-Kittler, 1981).
internally; M2 very small and double or single rooted; ml relatively
Mustelavus is closely related to the Old World Mustelictis.
short, with a high trigonid and a low talonid that is restricted labi-
The dentition of Mustelavus is more primitive than that of
ally; m2 double rooted to absent; I3 very enlarged; auditory bulla
Mustelictis.
with a large posterior accessory chamber; suprameatal fossa vari-
ably present to absent; posterior carotid foramen close to the pos-
UNNAMED GROUP: "PALEOMUSTELIDS" terior lacerate foramen; scapula with a teres process; palate greatly
AND "NEOMUSTELIDS" prolonged behind the molars; and a posteriorly projecting paroc-
cipital process. Except for Aelurocyon, the upper tooth row length
Late Oligocene to Recent mustelids have lost the alisphenoid canal and lower jaw length of oligobunines are not as reduced as in neo-
and have a palate that is greatly prolonged behind the molars; Ml mustelids (Radinsky, 1982).
with a large parastyle and with the postprotocrista and metaconule Included genera: Promartes, Zodiolestes, Oligobunis, Paroli-
absent; M2 two rooted to absent; and ml with a more anteriorly di- gobunis, Aelurocyon, Brachypsalis, and Potamotherium.
rected paraconid than in the Mustelavinae. Some of these characters Comments: A Eurasian taxon similar to ?Mustelictis robustus
are modified in later taxa. Paleomustelids are a paraphyletic group or perhaps Amphicticeps is closely related to the ancestry of the
and are characterized by the primitive retention of a camassial notch Oligobuninae. Oligobunines are not ancestral to neomustelids.
on P4. Mustelids of modem aspect are informally grouped herein
as neomustelids.
Promartes Riggs, 1942
Type species: Promartes olcotti Riggs, 1942.
SUBFAM ILY INCERT AE SEDIS Type specimen: FMNH 1578.
Characteristics: Small oligobunine with Pl reduced to absent;
Plesictis Pomel, 1846 P4 parastyle very small to absent; M 1 narrow internally, tri-
Type species: Plesictis genettoides Pomel, 1846. angular occlusal outline, well-developed parastylar shelf,
Type specimen: In the collections of the Musee de Lyon. paracone slightly larger than metacone; M2 small, dou-
Characteristics: A small mustelid with parasaggital crests on ble rooted; ml trigonid open, talonid semitrenchant with a
the skull; auditory bullae inflated; suprameatal fossa well low internal rim, protoconid separated from hypoconid by
developed; P4 protocone anteriorly placed (primitive); M2 a valley, metaconid small, much lower than the paraconid,
very reduced to absent; and m2 two rooted. and opposite the protoconid; m2 two rooted; auditory mea-
Average length of ml in North American species: unknown. tus not prolonged; and suprameatal fossa absent.
Smaller than Mustelavus priscus. Average length of ml: 8.3-10.4 mm.
Included North American species: ?Plesictis sp. only, known Included species: P. olcotti (probabl y= P. lepidusfi de Mac-
from locality CP86B only. donald, 1970) (known from localities CP52, CP86B,
*CP104A); P. lepidus (localities [SB29A], *CP86D,
Comments: Macdonald (1970) refers an anterior mandibu-
[CP88]); P. gemmarosae (locality *CP86B or *CP86C);
lar fragment to Plesictis sp. because of similarities to "P."
P. vantassalensis (probably = P. gemmarosae fide Mac-
priscus. Whether this specimen is congeneric with either
donald) (locality *CP52); P. darbyi (locality *CP103).
Plesictis or Mustelavus has not been determined. There are
Promartes sp. is also known from localities NB3B,
twenty-one nominate species of Plesictis from Europe. As
SB28, SB29C.
currently recognized Plesictis is almost certainly a para-
phyletic genus. Apomorphic characters shared with more Comments: This genus was established for the relatively
advanced mustelids appear in late Oligocene taxa such smaller, late Arikareean to Hemingfordian mustelids pre-
as Plesictis aff. P. pygmeus (Beaumont and Weidmann, viously referred to Oligobunis.
1981), which lacks the posterior opening of the alisphe-
noid canal, has a narrow Ml with a large parastyle and Zodiolestes Riggs, 1942
without a postprotocrista or metaconule, and has a very Type species: Zodiolestes daemonelixensis Riggs, 1942.
small M2. The following primitive musteloid characters Type specimen: FMNH 12032.
 Jon A. Baskin 156

Characteristics: The skull and dentition are similar to Pro- *CP104A); P. petersoni (locality *CP51B); P. frazieri (lo-
martes. It differs in having a longer face; premolars with cality *GC8A).
weak cingula; Pl/pl not reduced; M2/m2 more reduced; Paroligobunis sp. is also known from locality SB46.
Ml with a less prominent parastyle and a better developed
anterointernal cingulum on the protocone; ml with a some- Aelurocyon Peterson, 1907 (synonym: Megalictis
what taller metaconid; mastoid process flattened, expanded Matthew, 1907)
winglike, and almost horizontal in position; auditory bulla Type species: Aelurocyon brevifacies Peterson, 1907.
that contacts the base of the paroccipital process; and palate Type specimen: CM 1590.
only slightly expanded behind the molars. Characteristics: Very large mustelid (see Figure 9.1); cheek
Average length of ml: 10.3-12.1 mm. teeth relatively small for such a deep jaw; 13 greatly en-
Included species: Z. daemonelixensis only, known from lo- larged; Pl/pl small, other premolars broad; Ml cingulum
cality *CP103 only. slightly expanded internally, triangular parastylar shelf,
paracone much larger than metacone; M2 very small, sin-
Comments: Zodiolestes has been considered a procyonid in
gle rooted; m 1 metaconid absent, talonid narrow and se-
spite of its mustelid dentition because of the suprameatal
cant; m2 small, double rooted; flask-shaped auditory bulla
fossa. However, a suprameatal fossa is present in primi-
with a long tubular meatus, directed slightly anteriorly;
tive mustelids such as Mustelictis and Plesictis. Two un-
flattened, posteriorly projecting paroccipital process not
described skulls in the F:AM collections from a locality
in contact with the bulla.
in Wyoming have been referred to Zodiolestes; in one the
Average length of ml: 19.8-22.6 mm.
suprameatal fossa is well developed as in the genoholo-
Included species: A. brevifacies (known from localities CP52,
type; in the other, although the basicranial region is dam-
*CP104B); A. ferox (locality *CP86D).
aged, it is represented by a very shallow depression.
Comments: Megalictis has been distinguished from Aeluro-
Oligobunis Cope, 1881 cyon on minor dental characters and on characters of the
Type species: Oligobuniscrassivultus (Cope, 1879) (= Icticy- skull and dentition interpreted from reconstructions that
on crassivultus). are difficult to confirm in the fragmentary type skull of M.
Type specimen: AMNH 6903. ferox. Aelurocyon is regarded by some authors as a senior
Characteristics: Larger than Promartes and with ml rela- synonym of Megalictis. This is followed here. However, af-
tively more shortened and with a strong labial convexity ter this volume had gone to press, Hunt and Skolnick ( 1996)
caused by the indentation of a low, short talonid; Pl a determined that Megalictis has priority over Aelurocyon,
simple peg; diastemata present between anterior premo- and that there is only a single species, Megalictisferox.
lars; premolars broad posteriorly; M 1 triangular, very nar-
row internally, with a subequal paracone and metacone, a Brachypsalis Cope, 1890
prominent parastyle, and a medial-internal hypocone; M2 Type species: Brachypsalis pachycephalus Cope, 1890.
more reduced than in Promartes, double or single rooted; Type specimen: AMNH 8544.
m 1 metaconid reduced; and m2 single or fused double root. Characteristics: A relatively long-faced, medium to large
A referred specimen in the AMNH collections has a very mustelid; premolar and molar cusps tend to be inflated;
shallow suprameatal fossa. premolars crowded, upper premolars usually offset later-
Average length of ml: 12.7-14.8 mm. ally, widest medially; Pl/pl reduced to absent; P4 with a
Included species: 0. crassivultus (known from locality parastyle in later species; Ml quadrate with a prominent
*PN6F); ?0.floridanus (locality *GC8D). hypocone, protocone and paracone separate, and a broad
Oligobunis sp. is also known from locality CP105. labial cingulum that is well developed anteriorly in later
species; M2/m2 relatively large; p4 small, with a poste-
Paroligobunis Peterson, 1910 rior accessory cusp; ml with the protoconid and tall para-
Type species: Paroligobunis simplicidens (Peterson, 1907). conid close together, metaconid large, talonid large and
Type specimen: CM 1553. basined, entoconid present, protoconid separated from the
Characteristics: Similar to Oligobunis in having a shortened hypoconid by a deep notch, and metaconid separated from
ml with a strong labial convexity, but differing in having the rim of the talonid by a narrow notch. A primitive un-
ml with no entoconid, a larger pl and no diastemata be- described species in the F:AM collections from the early
tween premolars (Frailey, 1978). This distinction is prob- Hemingfordian, Aletomeryx Quarry (locality CP106) has
ably not enough to warrant generic distinction, especially a shallow suprameatal fossa that is partially underlain by a
because the entoconid in 0. crassivultus appears to be an dorsolateral extension of the ectotympanic. Brachypsalis
artifact of breakage. is larger, has a more basined talonid on ml, and larger sec-
Average length of ml: 13.5-16.4 mm. ond molars than Promartes, from which it was probably
Included species: P. simplicidens (known from locality derived.
 Mustelidae 157

Average length ofml: 12.8-16.5 mm.

oligobunines. It could be derived from a form similar to
Included species: B. pachycephalus (known from localities
Zodiolestes, which has cingula on the premolars, reduced
NB6E, NB23B, *CP114B); B. hyaenoides (?= B. pachy- M2/m2, and unreduced Pl/pl.
cephalus) (locality *CP114B);B. modicus(localities CP76,
*CPl 10, CP114A); B. matutinus (locality *CP108B); B.
obliquidens (probably = B. modicus) (locality *CPl 10). UNNAMED GROUP: "NEOMUSTELIDS"
Brachypsalis sp. is also known from localities (CC2 l C),
Characteristics: P4 carnassial notch absent (paracone connected
SP2A, CP105, CP114D, CP116A, PN8B, PN9B.
continuously with metacone); M2 absent; posterior carotid foramen
Comments: The holotype is a damaged jaw with a broken m 1. well anterior of the posterior lacerate foramen; suprameatal fossa
Webb (1969) considers "Lutra" pristina a typical member greatly reduced to absent; and postscapular fossa absent.
of Brachypsalis and establishes a new genus Brachypsa- Comments: Paragale and Plesiogale from the latest Oligocene or
loides, based on B. modicus, which he recognizes has a earliest Miocene of Europe, the earliest mustelids lacking a camas-
different morphology from other species of Brachypsalis. sial notch on P4, possess a small suprameatal fossa (Schmidt-Kittler,
Skinner, Skinner, and Gooris (1977, p. 353) point out that 1981 ). As seen in these two genera, neomustelids primitively have a
B. pristinus differs in morphology from other species of bladelike P4; a narrow Ml, with the protocone and metacone close
Brachypsalis. In the present chapter it is reassigned to together; and ml with a small metaconid and a short talonid with a
Mionictis because of characteristics of the lower camas- trenchant hypoconid and no entoconid. Mustelines primitively retain
sial, as it has been curated in the AMNH. However, the these states, which are adaptations for hypercarnivory. A triangular
upper teeth (AMNH 18268) figured by Matthew (1924) P4 with expanded internal shelf and hypocone evolved indepen-
are assignable to Brachypsalis. The cranium illustrated dently in the Leptarctini, Mephitini, Taxidiini, and Melini along
by Webb (1969) as "B." pristinus is also assignable to with other adaptations for omnivory. An inflated auditory bulla is
Brachypsalis. The type specimens of B. matutinus the primitive state for the neomustelids. A flattened ectotympanic
(Matthew, 1924) include a lower jaw with m2 (herein des- evolved independently in several groups.
ignated the lectotype) and upper dentition that are assign-
able to Brachypsalis, and a lower jaw with p4-m 1 (figured) LEPTARCTINAE
that is a Mionictis. B. angustidens Hall is also tentatively
assigned to Mionictis. Characteristics: Medium-sized, short-faced mustelids; skull short
and broad with very prominent parasagittal and lambdoidal crests
Potamotherium Geoffroy, 1833 and with zygomatic arches deep and widely expanded; auditory bul-
Type species: Potamotherium valetoni Geoffroy, 1833. lae large with bony projections extending anteromedially; the bullae
Type sp~cimen: In the Collection Julien, France. extend ventrally, but are not anteroposteriorly inflated; paroccipital
Characteristics: A medium-sized oligobunine with elongated process reduced, with limited contact with the bulla, and broad con-
frontals; premolars with cingula; Pl/pl present; P4 greatly tact with the mastoid process only on posterior surface (the latter a
broadened internally; Ml narrow, with parallel anterior derived state, not primitive fide Muizon, 1982); the external audi-
and posterior margins, a prominent parastylar shelf and a tory meatus opens on the lateral surface of the expanded posterior
well-developed internal cingulum extending anteriorly and portion of jugal process of the squamosal; posterior lacerate and
posteriorly around the protocone; M2/m2 single rooted; p4 posterior carotid foramina close together (primitive); opening of
with prominent anterior and posterior accessory cusps; ml foramen ovale faces ventrally (primitive); palate very long; Pl/pl
metaconid small, talonid narrow and semitrenchant; audi- absent; P4 with at least a small parastyle and an internal cingulum;
tory bulla small and flattened; large posterior carotid fora- M 1 with a preprotocrista extending to anterior border of paracone,
men adjacent to a very large posterior lacerate foramen; a posterointernal hypocone, and a prominent metacone separated
postglenoid foramen reduced to absent; skeleton otterlike from the protocone; ml with a prominent metaconid, long talonid,
in proportions. and a notch between metaconid and cingulum of talonid; and m2
Average length of m 1: 10.6-11.0 mm. double rooted (primitive).
Included North American species: Potamotherium sp., known
from the Hemingfordian (locality CPI 05) to the late Barsto- LEPTARCTINI
vian (locality CP 114) (Tedford et al., 1987).
Characteristics: Same as for subfamily.
Comments: Because of its otter-shaped body and dental con-
Comments: Qiu and Schmidt-Kittler (1982) describe a moder-
vergences, this genus has been included in the Lutrinae. ately deep suprameatal fossa in Leptarctus from the Tung Gur Fauna
Tedford ( 1976) regarded Potamotherium as a primitive arc- of China. In skulls of Leptarctus in the AMNH and MCZ collec-
toid transitional between mustelids and phocids and placed tions, a very shallow suprameatal fossa can be seen only in one
it in the phocid subfamily Semantorinae. Potamotherium is skull (MCZ 6223) of L. primus, the most primitive species of this
placed in the O!igobuninae because it maintains the grade genus. In all others the suprameatal fossa is absent. Leptarctus has
of dental and skeletal morphology that characterizes the a deep pocket formed by the extension of the epitympanic recess
 Jon A. Baskin 158

into the squamosal. It appears that the suprameatal fossa is lost by The early Barstovian "L. primus" from the Lower Snake
diminution in size by shallowing, rather than by being roofed under Creek Fauna (locality CPllO) (Matthew, 1924) is herein
as suggested by Schmidt-Kittler (1981). Segall (1943, Figure 10) il- referred to L. oregonensis, which has a smaller P4 with a
lustrates sinuses enclosed in the squamosal of both procyonids and less prominent hypocone and a relatively smaller Ml than
mustelids that are not homologous with a roofed-under suprameatal the holotype of L. primus.
fossa. Schmidt-Kittler (1981) notes a carnassial notch on P4 in the
European Trocharion, which he refers to the leptarctines. However,
Craterogale and L. primus both clearly lack a carnassial notch, have
UNNAMED GROUP: MELINAE
a more primitive, sectorial upper camassial, and occur earlier than
AND MUSTELINAE
Trocharion. There is no evidence of a carnassial notch in any P4 of Melinae and Mustelinae are derived with respect to Leptarctinae
Leptarctus. in having greater separation of the posterior lacerate and posterior
carotid foramina and in having the opening of the foramen ovale
Craterogale Gazin, 1936 facing at least partially anteriorly as well as ventrally.
Type species: Craterogale simus Gazin, 1936.
Type specimen: USNM 13801.
Characteristics: Craterogale is distinguished from Leptarctus SUBFAMILY INCERTAE SEDIS
in having P4 with an incipient parastyle and a knoblike Miomustela Hall, 1930
protocone and with hypocone absent and in having Ml Type species: Miomustela madisonae (Douglas, 1904)
less quadrate and with a better developed parastyle and a ( = Mustela? minor).
more internally situated hypocone. Type specimen: CM 848.
Average length of ml: unknown. Length of P4: 8.7 mm. Characteristics: Lower canine with a prominent basal cingu-
Included species: C. simus only, known from locality *CP105 lum on the posterior and medial surfaces; pl absent; p3-4
only. laterally compressed and tall with anterior and posterior
Comments: Craterogale is known from a relatively complete cingular cusps; p4 with a prominent posterior accessory
skull, which is slightly smaller than that of L. primus, the cusp; ml trigonid open, with a low paraconid subequal in
smallest and most primitive species of Leptarctus. height to the metaconid but larger than this cusp, a meta-
conid positioned slightly posterior to the protoconid, and
Leptarctus Leidy, 1857 a low, narrow, slightly basined talonid, indented labially
Type species: Leptarctus primus Leidy, 1857. and with a relatively wide notch separating the metaconid
Type specimen: ANSP 11293. from the internal cingulum; m2 double rooted.
Characteristics: P4 large with a prominent parastyle, an inter- Average length of m 1: 5.1 mm.
nal cingulum, and large hypocone; Ml quadrate in shape, Included species: M. madisonae only (known from localities
with paracone and metacone widely separated and with [CP114A], [CP114B], NPll, *NP41B).
a posterointernally situated hypocone; cl grooved antero- Miomustela sp. is also known from localities GC8D,
medially; p4 with a well-developed posterior accessory CP54B, CP105, CP107.
cusp; ml with a long talonid, oblique protoconid-paraconid
Comments: Hall (1930) relates Miomustela to the mephitines
blade, the metaconid equal in height to the paraconid, the
Pliogale and Martinogale, a relationship suggested by the
metaconid separated from the entoconid and the proto-
conid from the hypoconid by deep valleys; m2 double basal cingulum on the canine, the loss of pl, and the mor-
rooted, primitively with a broad anterior cingulum and a phology of the ml trigonid. The mastoid process is in-
vaded by the middle ear cavity in Miomustela as in the
narrow talonid. This m2 morphology is primitive for neo-
mustelids and is not a leptarctine synapomorphy. mephitines. The ml talonid, however, is shorter and nar-
Average length of ml: 10.3-11.8 mm. rower than in mephitines.
Included North American species: L. primus (= Hypsoparia Undescribed material from the Runningwater Forma-
bozemanensis Dorr, 1954) (known from localities CP76,
tion (locality CP105) (Tedford and Frailey, 1976) indi-
*CP89, CP114A, CP114B, CP116A, NP41B); L. ancipi- cates that Miomustela is a neomustelid with the carnassial
dens (localities *GC8D, CC22A); L. oregonensis (locali-
notch absent on P4 and the posterior carotid foramen posi-
ties CPl 10, *PN7); L. progressus (locality *GClOB); L. tioned midway between the posterior and median lacerate
wortmani (localities CC26B, *CP109B).
foramina. Parasagittal crests are present. The ?Miomustela
Leptarctus sp. is also known from localities GCl lA,
described from Thomas Farm (locality GC8D) and ad-
SP2A, (SP3A), (CP54B), CP105, CP114C, CP115C, ditional material from the Runningwater Formation be-
CP116B, (NP42), (PNlO). long to a small feliform carnivore, discussed later as cf.
Stenogale. Undescribed material of Miomustela is also
Comments: The species of Leptarctus are in need of revision. present at Thomas Farm.
 Mustelidae 159

MELINAE
tela or Martinogale; pl absent in genotype (P. manka is
Included tribes: Melini and Lutrini; possibly also Mephitini and described as having a small pl); p2 two rooted; ml talonid
Taxidiini. basined with a cuspidate rim and a distinct entoconid, sep-
Characteristics and comments: Talonid of ml primitively elon- arated from the metaconid by a shallow notch and from
gate and basined, with a cuspidate rim connected to a prominent the protoconid by a narrow notch; and m2 fused-double
metaconid; and ectotympanic flattened and with a tubular meatus rooted.
(except Taxidiini). Melini usually have inflated entotympanics, al- Average length ofml: 5.7-8.4 mm.
though Arctonyx has the flattened state of lutrines and mephitines. Included species: P.furlongi (known from localities *NB31,
Lutrines and melines share a P4 with a broad internal shelf; Ml SB9); P. manka (locality *CP90B).
with a reduced external cingulum indented between the paracone Pliogale sp. is also known from locality CP! 14B.
and metacone, and an accessory cusp posterointernal to the meta-
Comments: Hall (1930) recognizes Pliogale as an interme-
cone; ml with a closed trigonid (in more derived species) and the
diate between Miomustela and Martinogale.
talonid low (very low in melines), expanded and basined; and an
enlarged infraorbital foramen.
The Leptarctini, Taxidiini, and Mephitini share several dental Martinogale Hall, 1930
characteristics (see Figure 9.3); some are plesiomorphic, others may Type species: M. alveodens Hall, 1930.
be convergences. Additionally, leptarctines and taxidiines have large Type specimen: KUVP 3473.
septate bullae, vertically oriented mastoid regions, and m 1 with a Characteristics: A small mephitine with the characteristic
prominent separate entoconid. Whether the dental similarities of slender mandible with an anterior expansion; cl with a
Leptarctinae and Melinae are synapomorphies or parallel adapta- well-developed posterior and internal cingulum; premo-
tions for an omnivorous diet has not been determined. lars crowded, offset laterally in the jaw; p2 small, reduced,
supposedly single rooted in type, double rooted in referred
specimens; p4 anteriorly narrow, two rooted; ml with a
MEPHITINI
long, basined talonid with a very low anteromedial rim, a
Characteristics: The skunks are short-faced mustelines with a flat- prominent metaconid opposite the protoconid, accessory
tened, unexpanded ectotympanic and an extension of the middle ear rootlets beneath the protoconid and metaconid; and m2
cavity into the mastoid, forming an accessory chamber (a charac- double or fused-double rooted. An associated mandible
ter shared with taxidiines); the palate not greatly expanded behind and cranium from the Ricardo Formation (locality NB7C)
the molars; a bony septum separating the posterior nares; Pl/pl referred to M. alveodens has P4 with a very small, ante-
absent; the remaining premolars anteriorly-posteriorly compressed, riorly placed protocone, the internal shelf absent, a small
with a tendency toward reduction of the anterior premolars; Ml parastyle, and the paracone larger than the metacone; Ml
with a protocone-metacone crest (also present in the leptarctines), with a broad external shelf, a prominent parastyle, a small
an external shelf, and a posterointernally expanded shelf; the lower metastyle, the paracone separated from the smaller meta-
canine with a basal cingulum; and ml with an open trigonid with cone, and the hypocone with only a moderate amount of
an elongate paraconid, prominent metaconid, elongate and usually posterointernal expansion; and a moderately inflated ento-
basined talonid, accessory rootlets, the hypoconid separated from tympanic.
the protoconid by a narrow notch, and the median crest of the talonid Average length of ml: 6.6 mm.
separated from the metaconid by a deep valley. Included species: M. alveodens (known from localities NB7C,
Comments: Although today known only from the New World, CP116F, *CP123D); M. nambiana (locality *SB32D).
mephitines have an extensive Old World fossil record. Mephitines
Comments: Harrison (1983) states thatM. nambiana is equal-
and stink badgers (Mydaus) may share a common ancestor, as sug-
ly likely referable to Pliogale.
gested by Radinsky (1973). Petter (1967) summarizes the system-
atics of Old World mephitines and briefly notes their relationships
Buisnictis Hibbard, 1950
to New World forms. The earliest undoubted mephitine is the Old
Type species: B. meadensis Hibbard, 1950.
World medial Miocene Proputorius, which also has ml with an
Type specimen: UMMP 25769.
accessory root under the metaconid.
Junior synonym of B. breviramus (Hibbard, 1941 [=Bra-
chyprotoma breviramus]. Type specimen: KUVP 4609.)
Pliogale Hall, 1930
Characteristics: Small short-faced skunk with a very crowded
Type species: Pliogale furlongi (Merriam, 1911) (= Mustela premolar series; P4 with protocone anteriorly placed and
furlongi).
internal cingulum absent; ml with entoconid separated
Type specimen: UCMP 12540.
from a very reduced metaconid by a deep broad notch;
Characteristics: The ml with a third external rootlet, and a m2 double rooted; and lower canine with a basal postero-
smaller more posteriorly placed metaconid than Miomus- medial cingular cusp as in Spilogale and Martinogale.
 Jon A. Baskin 160

Average length of ml: 5.2 mm. described as being intermediate between the typical mephi-
Included species: B. breviramus (known from localities SPIF, tine and weasel morphology and suggesting possible as-
SPlH, CP128A, CP128B, *CP128C, [PN4], PN23A); B. signment to a new genus. Specimens assigned to the extant
schoffi (localities *SP4B, CPI I6F, CP123E); B. burrowsi spotted skunk, S. putorius, are first known from the latest
(locality *CPl 18). Blancan or earliest Irvingtonian.
Buisnictis sp. is also known from localities ?SB48,
CP121, CP130A. Mephitis Geoffroy Saint-Hillaire and Cuvier, 1795
Comments: Buisnictis is closely related to Martinogale; it Type species: Mephitis mephitis (Schreber, 1776) ( = Viverra
differs in having the ml with a well-developed entoconid mephitis), the extant striped skunk.
and a broader Ml with a better developed posterointer- Included pre-Pleistocene records: M? rexroadensis (*known
nal cingulum. The lower dentition is similar to the Pleis- from locality CP128A); M. mephitis (locality CP117B).
tocene skunk Brachyprotoma, which is also short faced,
Comments: The early Blancan M.? rexroadensis is described
with crowded premolars, and has ml with a relatively
from a P4 and two Ml 's. Ml has a broad posterointemal
small metaconid. However, the upper dentition indicates
cingulum with a broad posterior valley, as in M. mephi-
Brachyprotoma is more closely related to Spilogale.
tis. P4 has a more anterior protocone and a narrower lin-
Brachyopsigale Hibbard, 1954 gual shelf than recentMephitis. See other comments under
Brachyopsigale.
Type species: B. dubius Hibbard 1954.
Type specimen: UMMP 30584.
Characteristics: This genus is known from fragmentary jaw Mephitini gen. et sp. indet
fragments. It is distinguished from Buisnictis in having Indeterminate mephitines have been noted from localities NB7D,
p2 single rooted, p4 broadened posteriorly and with three SPlA, CP116E.
roots, and extreme shortening of the jaws (Dalquest, 1978).
Dalquest notes that Mephitis? rexroadensis, known only
from upper molars, is large enough to belong to B. du- TAXIDIINI
bius. Dalquest's diagnosis suggests that Brachyopsigale is Characteristics: The North American badgers are medium-sized,
closely related to, or perhaps congeneric with, Brachypro- fossorial mustelids; auditory bulla with a moderately to greatly in-
toma. Both Brachyprotoma and Promephitis have P2 ab- flated entotympanic region that is expanded posteriorly, restricting
sent. the ventral contact between the paroccipital and mastoid proctoss (a
Average length of ml: unknown. About the size of Buisnictis derived state); auditory bulla is divided by septa, and the epitym-
breviramus. panic sinus extends into the mastoid region of the periotic; Pl/pl
Included species: B. dubius only (known from localities SPlF, absent; P4 with a well-developed protocone, hypocone and internal
[CP116F], *CP128A). shelf in described genera; M 1 with a reduced parastylar shelf, broad
anterointemal and greatly expanded posterointemal cingula, acces-
Spilogale Gray, 1865 sory cuspules, and transversely aligned cusps; p4 with a prominent
Type species: Spilogale interrupta (Rafinesque, 1820) posterolateral cusp; ml with a large cuspidate talonid, an accessory
(= M ephitis interrupta),junior synonym of Spilogale puto- cusp developed external to the hypoconid, and with the paraconid
rius (Linnaeus, 1758)(= Viverraputorius), the extant spot- and subequal metaconid close together; trigonid separated from the
ted skunk. talonid by a notch; and m2 circular, with two separate or fused roots,
Characteristics: Small, long-jawed skunks with a more sec- which probably reevolved with increasing size and more massive
torial dentition than in the other living mephitine genera; dentition. Undescribed F:AM material indicates that the internal
P4 with protocone broadly attached to and opposite the shelf of the P4 evolved independently in the taxidiines.
paracone, lingual shelf extending only slightly onto the
metacone; p4 with a broader posterior cingulum than Buis-
Pliotaxidea Hall, 1944
nictis; and ml metaconid well developed (equal in height
to, but smaller than paraconid). Type species: Pliotaxidea nevadensis (Butterworth, 1916)
(= Taxidea nevadensis).
Average length of ml: 5.6-7.0 mm.
Included pre-Pleistocene records: S.? rexroadi (known from Type specimen: UCMP 22290.
localities SPlF, SP5, *CP128C); S.? microdens (locality Characteristics: Smaller than Taxidea; Ml with a small para-
*SPlF); S. putorius (localities SB14F, CP132B). style, an elongate posterointemal cingulum, and an angular
Spilogale sp. is also known from localities WM26, PN4.
posterior termination (Wagner, 1976); ml with a relatively
larger talonid; and m2 single or double rooted.
Comments: P4's of early Blancan Spilogale? have not been Average length of ml: 8.2-10.9 mm.
described. Isolated M 1's from Beck Ranch (locality SPlF) Included species: P. nevadensis (known from localities
tentatively referred to S.? rexroadi by Dalquest (1978) are *NB31, SB9, SB34C, SPlD, SP3B, CP115D, CP116E,
 Mustelidae 161

PNll, PN14); P. garberi (localities *CC25B, SB34C, double or fused-double rooted; and infraorbital foramen
CP116F). large.
Pliotaxidea sp. is also known from localities SB 10, Average length of ml: 10.1-16.4 mm.
CP115C, ?PN2B, PN14. Included North American species: M. incertus (known from
localities *CPI 10); M. elegans (locality *CPI 10); M.
Taxidea Waterhouse, 1838 letifer(locality *CPI 08B); M. pristinus (localities *CP90A,
Type species: Taxidea taxus (Schreber, 1778) ( = Ursus taxus, CP114A, CPl 140, CP115B, CPl 16A); ?M. angustidens
the extant North American badger). (locality *CC38).
Characteristics: See Pliotaxidea; in addition, Taxidea has Mionictis sp. is also known from localities SP2A;
a median hypoconid as the most prominent cusp of the CP114B, (CP116F), ?CP123A.
talonid on ml, a wider parastyle on Ml, double or fused-
double roots on m2, and a flattened ectotympanic. Taxidea Comments: Refer to the section on Brachypsalis for further
taxus is the extant North American badger. comments.
Average length of ml: 11.6-14.3 mm.
Included Pliocene records: T. mexicana (known from local- Limnonyx Crusafont Pairo, 1950
ity *SB60); T. taxus (localities SB17, CP118, CP128B, Type species: Limnonyx sinerizi Crusafont Pair6, 1950.
[SPlF]). Type specimen: In the collections of the Institut de Paleon-
Taxidea sp. is also known from localities (NB13C),
tologia "M. Crusafont" (Sabadell).
SB50, SPlH, SP4B, CP117B, CP128C, CP130A, PN4, Average length of ml: 12.1 mm.
PN23A. Included North American species: cf. Limnonyx sp. only,
known from locality NB29 only.

LUTRINI Comments: Macdonald (1956) refers several fragmentary

lower jaws to Lutra. These represent the first occurrence
Characteristics: Otters are small to very large mustelids with aquatic in North America of an otter closely related to the extant
specializations of the postcranial skeleton; a flattened entotympanic; genera. The best specimen, a lower jaw with ml, seems
P4 with a hypocone or hypoconal crest and a progressively shortened better assignable to the Old World late Miocene genus
metacone; Ml with a greatly reduced external cingulum indented Limnonyx. The ml protoconid has a posterior accessory
between the paracone and metacone, a poorly developed internal cusp, the talonid is wider than the trigonid, and the depth
cingulum, the protocone close to the internal margin of the tooth, of the mandible is slightly greater than the length of the ml.
the hypocone slightly expanded posteriorly, and the anterior margin
fiat with a small cingulum at the anterointemal comer; p 1 primitively Satherium Gazin, 1934
present; and ml with an accessory cuspule on the posterior margin Type species: Satheriumpiscinarium (Leidy, 1873) (=Lutra
of the protoconid, a shortened trigonid, and a strong metaconid piscinaria).
connecting to a broad, basined talonid. Type specimen: USNM 662.
Characteristics: Four upper and three lower premolars are
Mionictis Matthew, 1924 present; Pl small and situated posterolateral to the canine;
Type species: Mionictis incertus Matthew 1924. P4 wide, triangular in outline, and with a well-developed
Type specimen: AMNH 18263. parastyle and a broadly expanded hypoconal shelf; Ml
Characteristics: Small to medium mustelid, with three promi- smaller than the P4; p4 tall and with a small posterior acces-
nent premolars that lack accessory cusps; P4 with a tri- sory cusp; and ml with an obliquely directed protoconid-
angular occlusal outline, broad lingual shelf, and a small paraconid blade and a metastylid adjacent and posterior to
protocone; Ml with a postprotocrista, a small accessory the metaconid. This large otter resembles the Recent South
cusp on the posterior margin of tooth posterointemal to American Pteroneura.
the metacone, the paracone separate from the smaller meta- Average length of ml: 16.5-18.3 mm.
cone, a narrow external cingulum that is slightly indented, Included species: S. piscinarium (known from localities
a very broad internal cingulum expanded anteriorly and es- SP4B, CP117B, CP118, CP128C, PN4, *PN23A); S. in-
pecially posteriorly, and a fiat anterior margin; ml with a gens (localities CP128C, *PN23C).
trigonid that is primitively open and a semibasined talonid Satherium sp. is also known from locality GC15.
that is low, long, and broadly expanded internally; ml
metaconid well developed, situated posterior to the pro- Enhydritherium Berta and Morgan, 1985
toconid, and with a gently sloping posterior surface that is Type species: Enhydritherium terraenovae Berta and Mor-
continuous with the internal cingulum of talonid, which ex- gan, 1985.
tends continuously to the hypoconid; m 1 hypoconid promi- Type specimen: UF 18928.
nent, with a fiat, gently sloping internal surface, and sep- Characteristics: A large otter with a short, deep jaw; P4 trian-
arated from the protoconid by a notch; m2 circular and gular, with the parastyle absent, the paracone and metacone
 Jon A. Baskin 162

distinct, separate inflated cusps that form a de novo car- Included North American pre-Pleistocene records: M. glareae
nassial notch, and a prominent hypoconal crest; and ml (known from localities *CPl 10, CPI l 6A); M. areganensis
broad, with low inflated cusps, the metaconid larger than (locality *PN15); M. stirtani (locality *CP123A).
the protoconid, a metastylid is present, and a protostylid Martes sp. is also known from localities CC 17E, CC 17G,
absent. CCI 7H, SB9, SBlO, SB48, CPl 160, ?CP116F, NP41B.
Average length of ml: 15.8-17.1 mm.
Comments: Some fossil species referred to Martes are herein
Included North American species: E. terraenavae only
questionably reassigned to Plianictis or Sthenictis.
(known from localities WM21B, GC12II, *GC13B,
CC39B).
Enhydritherium sp. is also known from locality CC42. ISCHYRICTINI

Characteristics: Advanced forms are medium to very large

MUSTELINAE mustelids with I3 much larger than the median incisors; P4 with
a low, rounded, anteriorly placed protocone connected by a broad
Characteristics: Ml with metacone close to paracone and with an-
isthmus to the paracone and a reduced to absent parastyle; M 1 prim-
teroposteriorly expanded internal cingulum; ml with metaconid re-
itively transversely elongate, with a prominent parastyle and a well-
duced and talonid trenchant; and m2 single rooted.
developed internal cingulum that extends anteriorly and posteriorly
Comments: Mustelinae as commonly used by neomammalogists
of the large median-internal protocone; m 1 narrow and elongate with
include mustelines (weasels, martens, wolverines, etc.), galictines
a semitrenchant to trenchant talonid; m2 small and circular; and the
(grisons), and mellivorines (honey badgers), the mustelids with the
auditory bullae moderately inflated, but short, and with thick walls
most sectorial dentitions. There is no widespread consensus on the
and a tubular meatus.
relationships of the musteline genera. Forms assigned to the genus
Comments: Many investigators have supported an ancestral-
Martes, the extant marten, are among the earliest of the neomus-
descendant or sister-group relationship between Eamellivara and
telids. The relationship of Martes to other mustelines, especially
Mellivara and between Plesiagula and Gula. However, the rela-
Gulo (the wolverine), requires further study. TheNeotropical genus
tionship of the ischyrictines to Mellivara and Gula requires further
Eira (the tayra) appears to be closely related to Martes.
study.

TRIBE INCERTAE SEDIS Plionictis Matthew, 1924

Type species: Plianictis agygia (Matthew, 1901) (= Mus•ela
Dinogale Cook and Macdonald, 1962
agygia).
Type species: Dinagale siauxensis Cook and Macdonald,
Type specimen: AMNH 9042.
1962.
Characteristics: Small mustelid, with narrow premolars;
Type specimen: AMNH 81009.
Pl/pl reduced to absent; P4 parastyle very small, the pro-
Characteristics: This Martes-like taxon is known from a sin-
tocone small, knoblike, narrowly connected to paracone,
gle lower jaw with four widely spaced premolars, a very
and set more posteriorly than in Mustela, and the metacone
small p 1, and a p4 with a posterior accessory cusp. Them 1
blade relatively long; Ml parastylar shelf well developed,
differs from that of Plianictis in having a stronger meta-
transversely elongate and only slightly anteriorly-
conid and a slightly basined talonid in which the internal
posteriorly expanded internally; p4 with a posterior ac-
cingulum merges with the posterior margin of the meta-
cessory cusp (worn in type); ml trigonid much longer
conid.
than talonid, the trigonid is open, the metaconid small,
Average length of ml: 10.7 mm.
the talonid narrow, and the hypoconid trenchant and sep-
Included species: D. siauxensis only, known from locality
arated from protoconid by a notch; palate extending far
*CP108B only.
behind molars; the auditory bullae elongate, inflated me-
dially, and attached to base of paroccipital process; and a
Martes Pinel, 1792
poorly developed tubular meatus.
Type species: Martes faina (Erxleben, 1777) ( = Mustela
Average length of ml: 7.6-9.5 mm.
faina), the extant pine marten.
Included species: P. agygia (known from localities *CP76,
Characteristics: Lower jaw characteristics are those of Dina-
CP110,CP114A,CP114B,CP116A, [NPl l]);P.parvilaba
gale. P4 with a small anterior knoblike protocone that is
(localities ?CP76, *CPllO, CP114A); P. kinseyi (locality
narrowly connected to the paracone; auditory bullae ex-
*NP41B); ?P. gazini (locality *PN8B); ?P. buwaldi (local-
panded posteriorly to the paroccipital and mastoid pro-
ity *NB7C).
cesses; and Ml with the internal cingulum expanded ante-
Plianictis sp. is also known from localities CP108B,
riorly and posteriorly to give the characteristic musteline
CP114C.
figure 8 shape, the parastylar shelf small, and the paracone
and metacone close together. Comments: It is sometimes difficult to distinguish specimens
Average length of ml: 7.0-13.0 mm. of Martes from those of Plianictis. In Plianictis the ml
 Mustelidae 163

talonid is less elongate and less basined than in Martes.

Eamellivara by its shallower jaw, more trenchant premo-
The present study is only a preliminary investigation. The
lars, absence of premolar cingula, and reduction of the
species of Plianictis, especially, need a thorough revision.
metacone and hypocone of Ml. Haplictis is distinguished
from the European Ischrictis by the reduced Ml and the
Sthenictis Peterson, 1910 absence of a metaconid on ml.
Type species: Sthenictis rabustus (Cope, 1890) (= Stenagale
rabusta). Eomellivora Zdanski, 1924
Type specimen: AMNH 8541. Type species: E. wimani Zdanski, 1924.
Characteristics: Medium-sized mustelid with four upper and Type specimen: In the collections of the Paleontological In-
lower premolars; P4 with a small anterior protocone nar- stitute, University of Uppsala.
rowly connected to the paracone; Ml transversely elon- Characteristics: p 1 very reduced; p4 with well-developed an-
gate, slightly expanded anteriorly and posteriorly on the terior and posterior accessory cusps; m 1 metaconid absent,
internal margin, the small paracone and the metacone close talonid very trenchant; Ml with the paracone much larger
together, and with a wide median to anterior stylar shelf; than the metacone and a prominent protocone situated me-
m 1 compressed and elongate, with a small metaconid and a dianly and bordered anteriorly, posteriorly, and internally
narrow, semi trenchant talonid with a prominent hypoconid by a broad cingulum; and P3 three rooted.
and no entoconid; auditory bulla contacting the base of the Average length of ml: 21.9-26.3 mm.
paroccipital process, which has a broad ventral connection Included North American species: E. cf. E. wimani only
to the flattened mastoid process; there is a tubular meatus; (known from localities ?CC32B, CC38).
and the palate is greatly extended behind the molars. Eamellivara sp. is also known from localities ?(NB7D),
Average length of ml: 12.0-18.3 mm. NB27B, SP3A.
Included species: S. rabustus (known from locality *CP114B);
S. bellus (locality *CP108B); S. dalichaps (localities Plesiogulo Zdansky, 1924
*CPll 0, NPl 1); S. junturensis (locality *PNlO); S. camp- Type species: Plesiagulo brachygnathus (Schlosser, 1903)
estris (locality *CP90A); S.? lacata (localities GCllA, (= Lutra brachygnatha).
*CP90A). Type specimen: Mandible figured by Schlosser (1903).
Questionable specimens of Sthenictis sp. are also known Characteristics: A very large mustelid with simple premolars
from localities CC17G, CC17H, ?SP2A, ?CP109B, that are broader and more inflated than Sthenictis; Pl/pl
CP116D. very reduced; Ml with a greatly expanded posterointernal
lobe, paracone and metacone close together, and a broad
Comments: The genoholotype is a lower jaw with broken parastylar shelf; and ml very elongate relative to premo-
teeth. The fragmentary ml shows the small metaconid lars, with a small metaconid usually present and a broad-
and semitrenchant, elongate talonid characteristic of this ened semitrenchant talonid that is only slightly concave
genus. Sthenictis can be derived from a form like Plian- and inclined lingually at a low angle (Harrison, 1981). P.
ictis by an increase in size, the elongation of the talonid,
lindsayi has a three-rooted P3. The auditory meatus is long
and decrease in relative size of the metaconid. Sthenictis and tubular.
strongly resembles the European Laphictis. Average length of ml: 23.5-29.0 mm.
North American species: P. marshalli (known from localities
Hoplictis Ginsburg, 1961 (synonym: Beckia Bryant, 1968) GC13B, CC25B, SBl 1, SB34C, SPlD, SP3B, *CP123D,
Type species: Haplictisflarancei (Mayet, 1908). (= Trachic- [NP5B], PN14); P. lindsayi (localities CC37, *SB9, SBlO,
tis zibethaides mut.flarancei Ma yet, 1908, /schyrictis [Ha- [CP116F]).
plictis] flarancei Ginsburg, 1961 =Haplictis flarancei
Ginsburg and Morales, 1992). Canimartes Cope, 1892
Type specimen: In the collections of the MNHM. Type species: Canimartes cumminsi Cope, 1892.
Characteristics: Large mustelid with ml with a high-angle Type specimen: TMM 40287-16.
trenchant talonid and no metaconid; p 1 very small; p4 tren- Characteristics: This enigmatic genus is known only from
chant, accessory cusps not prominent. fragmentary material. Ray, Anderson, and Webb (1981)
Average length of ml: 19.4 mm. consider Canimartes a possible galictine. However, the
Included North American species: H. grangerensis only protocone of P4 is set off from the paracone by a constric-
(known from localities CC30A, *PN2A). tion and ml is similar to that of Sthenictis. Cope (1893)
Haplictis sp. is also known from localities GCl lA, illustrated Ml-2 for this taxon, but the specimens have
CP116B, PNlO. been lost. No additional material of this taxon has been
recovered to confirm the presence of two upper molars or
Comments: Assignment of Beckia to Haplictis follows Ted- even the provenance of the specimens.
ford (Tedford et al., 1987). Haplictis is distinguished from Average length of ml: 14.l mm.
 Jon A. Baskin 164

Included species: C. cumminsi only, known from locality questioned. They may instead be Plionictis, as is assumed
*SP5 only. for ?P. buwaldi.

Ferinestrix Bjork, 1970

Type species: Ferinestrix vorax Bjork, 1970. GALICTINI
Type specimen: UMMP 53343. Characteristics: Small mustelines with a deep jaw, crowded pre-
Characteristics: A robust, short-jawed mustelid with a crowd- molars, and pl absent; P4 primitively with a small anterior proto-
ed, reduced premolar series and wrinkled enamel; cl large; cone and narrow lingual cingular shelf that is more expanded in
pl probably absent; p2 probably single rooted; p4 small, derived forms; Ml with an expanded hypocone and a more sepa-
simple, and wide; ml with a large trigonid and a basined rate protocone-metacone; and ml with an open bladelike trigonid,
talonid with a cuspidate rim; and m2 small and circular. a lingually expanded and basined talonid with a posterolingual cin-
Average length of ml: 27.0 mm. gulum extending from the base of the reduced metaconid to the
Included species: F. vorax only, known from locality *PN23A hypoconid, and hypoconid separated from the protoconid by a small
only. notch.
Comments: This genus is known solely from a lower jaw. Comments: Galictines have been reviewed most recently by Ray,
Them 1 is broken lingually anterior to the metaconid. Bjork Anderson, and Webb (1981). They consider Trigonictis the most
( 1970) compares Ferinestrix with Plesiogulo and Gula, the primitive genus included in the galictines, more closely relat~d ~o
two largest North American Plio-Pleistocene mustelids. It Eira than Galictis, and derive it from the Old World Troch1ct1s.
can be derived from a form similar to Plesiogulo by further However, the upper dentition of Eira indicates it is more closely
decreasing the relative length of p4, increasing the length related to the martens.
of ml, and adding internal cingular cusps to the talonid.
Alternatively, the cuspidate rim is characteristic of melines Cernictis Hall, 1935
and the metaconid is better developed than in Plesiogulo. Type species: Cernictis hesperus Hall, 1935.
The talonid, however, is not elongate as is characteristic Type specimen: UCMP 22968.
of large melines, and wrinkled enamel is also present in Characteristics: Lower premolars crowded; p4 with a poste-
Plesiogulo. rior accessory cusp; and m I short relative to p4, hypoconid
trenchant, and talonid relatively shorter and less well basin-
ed than Trigonictis.
MUSTELINI Average length of ml: 11.2 mm.
Included species: C. hesperus only, known from locality
Characteristics: Weasels are very small to small mustelids with sec-
*CC37 only.
torial camassials; P4 protocone small and anteriorly placed; Ml
Cernictis sp. is also known from locality SB 11.
with a medial constriction and an expanded internal lobe forming a
figure-eight occlusal outline and a reduced parastyle; the auditory
bulla is expanded and inflated posteriorly, limiting contact between Trigonictis Hibbard, 1941
the paroccipital and mastoid processes; and the foramen ovale is T)'pe species: T. kansasensis Hibbard, 1941.
directed anterior! y. T)'pe specimen: KUVP 4604.
Junior synonym of T. macrodon (Cope, 1868) (= Galera
Mustela Linnaeus, 1758 macrodon).
T)'pe species: Mustela erminea Linnaeus, 1758, the extant Type specimen ANSP 11626.
Characteristics: No accessory cusps on lower premolars; Pl
stoat or ermine.
Characteristics: A small musteline with Pl/pl absent; ml present; P4 with a triangular occlusal outline, a low and
Jacks a metaconid and has a trenchant talonid; and m2 is conical protocone, an incipient hypocone, and a shallow,
basined shelf; M 1 with metaconule present, roots short and
very small.
broad, anterior margin straight, and hypoconal cingulum
Average length of ml: 5.0-6.0 mm.
Included North American pre-Pleistocene records: M. rex- expanded posteriorly; p2 double rooted; and ml with an
roadensis (=M. gazini Hibbard, 1958) (known from lo- elongate trigonid and a long basined talonid.
calities SPlF, [CP116F], CP128A, *CP128C, [CPl31], Average length of ml: 10.7-14.9 mm.
[PN4], PN23A); M.frenata (locality CP132B), M. meltoni Included species: T. macrodon ( = T. idahoensis, T. kansasen-
sis) (localities GC14, GC30, SB13, SPlG, CP117B,
(locality *CP128C).
Mustela sp. is also known from localities NB13B,
CP118, CP128C, CP130A, PN3C, PN23A, PN23C,
SB18B, ?CP116C, CP121, ?PN9B. *NC6); T. cooki (localities CPl 18, PN4, *PN23A, PN23C;
also reported from the early Irvingtonian of Florida).
Comments: Mustela is probably a Pliocene immigrant. Ear- Trigonictis sp. is also known from localities SPIH,
lier records (from localities CP116C and PN9B) should be CP116F, CP121.
 Mustelidae 165

Sminthosinis Bjork, 1970 vestigial to absent; ml trigonid open and bladelike, meta-
T)'pe species: Sminthosinis bowleri Bjork, 1970. conid absent, and talonid trenchant; m2 very reduced but
Type specimen: UMMP V52868. ovate and two rooted, cusps linear, and talonid absent; and
Characteristics: A small, short-faced grisonine with Pl small m3 absent.
and crowded posterolingual to canine; P4 with a large Average length of ml: 4.0-6.2 mm.
protocone that is more posteriorly situated and more broad- Included North American species: P. sectoria (= P. lago-
ly connected to the paracone than in Martes and a slight phaga, P. infelix) (known from localities CP68C, [CP84A],
internal cingulum; Ml similar to that of Trigonictis, but [CP88], NP24, [NP32B]); P. dorothiae (localities *CP85C,
without a metaconule; and m I with metaconid more sep- CP101); P. sanguinarius (locality *CP42D); P. cf. P. min-
arate and situated further anteriorly than in Trigonictis. uta (locality CP106).
Average length of ml: 8.0-9.5 mm. Questionable specimens of Palaeogale sp. are also known
Included species: S. bowleri only, known from locality from localities GC5, GC7.
*PN23A only.
Sminthosinis sp. is also known from localities ?CC25B, Comments: The species of Palaeogale have been reviewed
CP116F. by Bonis (1981), whose synonymies are followed here.
Palaeogale has long been considered a primitive mustelid.
Lutravus Furlong, 1932 However, according to more recent interpretation (Hunt,
Type species: Lutravus halli Furlong, 1932. 1974, 1989; Flynn and Galiano, 1982), it is closely re-
Type specimen: LACM/CIT 478. lated to the early Tertiary family Viverravidae (i.e., "Mi-
Characteristics: Upper canine with an anterolateral groove, a acoidea"). Flynn, Neff, and Tedford (1988) conclude that
characteristic of the grisonines; four upper premolars; P4 Palaeogale is best considered incertae sedis within the Car-
with a short carnassial blade, protocone opposite the para- nivora, not having any unambiguous synapomorphies with
cone and connected to it by a lingual shelf; and Ml with a either the Feliformia or Caniformia.
prominent internal cingulum lingual to the protocone, cin- Palaeogale sanguinarius was described as a species of
gulum not expanded anteriorly and only slightly expanded Plionictis by Loomis (1932). It is slightly larger than and
posteriorly, protocone slightly larger than the paracone, probably equivalent in age to P. dorothiae.
and parastylar shelf narrow. Referred lower dentition has
Stenogale Schlosser, 1887
p4 without a posterior accessory cusp; ml with an open
Type species: Stenogale gracilis (Gervais, 1876).
trigonid, metaconid small and connected directly to the in-
Type specimen: In the collections of the MNHN.
ternal cingulum of talonid, and hypoconid separated from
Comment: Material of a very small carnivoran from Thomas
the pro'toconid by a narrow notch.
Farm (locality GC8D) was initially referred to ?Miomus-
Average length of ml: 14.0 mm.
tela. Tedford and Frailey (1976) question this reference;
Included species: L. halli only (known from localities NB3 l,
they compare the Florida specimens to European Plesictis
PN12).
julieni and P. sicaulensis and note the presence of the same
taxon in the Runningwater Formation (locality CPI 05). In
INDETERMI NATE MUSTELIDA E fact, the Thomas Farm and some of the Nebraska spec-
imens represent a primitive aeluroid (sensu Hunt, 1989).
Indeterminate mustelids have been reported from localities GC5,
This is indicted by the tall paraconid blade, reduced meta-
GClOA, ?NB17, NB19C, SB7, SB14A, NPlOE, ?NC2.
conid, and the short talonid of m I, and the morphology
of the lower premolars, P4, and M 1. The tax on compares
ORDER CARNIVORA INCERTAE SEDIS best with the Old World Stenogale, but may represent a
new genus.
Two genera of small carnivorans, Palaeogale and cf. Steno gale, are
Average length of m I: 6.0 mm.
discussed at this point because they have been classified as mustelids
Included North American species: cf. Stenogale sp. only
and their correct position within the Carnivora is debatable. Both
(known from localities GC8D, CP105).
are probably feliform carnivorans.

Palaeogale von Meyer, 1846 (= Bunaelurus Cope, 1873) BIOLOGY AND EVOLUTIONARY PATTERNS
Type species: Palaeogale minuta (Gervais, 1848-1852) (=
Mustela minuta). The family Mustelidae has the highest generic and specific diversity
Type specimen: In the collections of the MNHN. of all Recent carnivoran families (with the recognition of the Her-
Characteristics: Very small carnivoran- with auditory bullae pestidae as distinct from the Viverridae). Many extant mustelids
complete, inflated, and undivided; alisphenoid canal ab- are sexually dimorphic, with males significantly larger than fe-
sent; P4 with parastyle and camassial notch; Ml internal males, which has been documented in the Pliocene Trigonictis (Za-
cingulum absent; M2 vestigial to absent; M3 absent; p 1 krzewski, 1967) and also occurs in the Miocene Leptarctus. There
 Jon A. Baskin 166

EOCENE OLIGOCENE MIOCENE PLIOCENE

m m r- m r-
' '
....
~
:;:
?
....
c
?
l?
I'
m
h ~?".~ ~ ii i::
;.
....
~i "' I !I'
r
~
!" m

~
F
!I' ,~
m
h
....
h
~
m .-

,,:cm,,:cm ..,.!;: ."'~ -c: ,~

..
r

"
e . ~ !l
... ~ E t: I: tl
In In
). .. :
!:
cc E c ~ "'
In b
. !:
In ~ b"' ~
mm
1'

~ .:
Muste/avus I
I I 1?Plesictis I I
I
IPromartes I I
Zodiolestes
Oligobunis
Paroligobunis
Aelurocyon I
1Brachypsalis I
1Potamotherium
r-
m I I ' I I
'lJ
Craterogale
-I
J>
J:J
0
-I
[ t"feptarctus'
...
~, ,~-1~~,~----
1
......... ~,1
-.)I

zJ> 1 Miom~ste/a 1
1
m P/1ogale
s:
CD
"C
:::r
...3·
2:
s:
m -I
cz
J>
m ~[

z
m
0
s:
i-·. [ I
c: 1oinogale I I
en Martes
-I 1
m
r- I I Plionictis I I
e
en I Sthenictis 1 J~
I Hoplictis ~ I 1
I I I I Eomellivora
IP/esiogulo
1Canimartes
-·
1Ferinestrix
1-
1 Musteta
I 1 I I
Cernictis
I I I MI I
Trigonictis
I I ' I
Sminthosinis
I ·~ ~ I ,..._,._..
I Lutravus
Pataeogate 1 I I --, ' I
I I I I

Figure 9.5. Temporal ranges of North American mustelid genera.

 Mustelidae 167

EOCENE -a
OLIGOCENE
I MIOCENE -a
'm EPOCH
CHAD- 0 ~
HEMING- CLAREN- HEMP-
50 ~
~ m
:D
m ARIKAREEAN BARSTOVIAN DONIAN HILLIAN 0
RONIAN FORDIAN z 0
z~
m NALMA
;!:: I' i m
<D
Im l~I ;= m I• ml ~ Ir= ml ' m I•
m z
m
(,.) !\) !\) ...... ......
~' ';> 0 01 MA
' ' ' '9' ' ' ' ' I I I I I
' qi ' ' ' ' ' ' ' ' ' ' ' ' ' '
PALAEOGALE
-m
tD
o,
)>

D cf STENOGALE
Oc
J: :JJ
:JJ0
--- o-
zO
D CJ)~

D "'U
MUSTELAVUS
)>
D r
?PLESICT/S
-
m
0
c
s:
D
--- -·-·-·-·-- - ---·-
POTAMOTHERIUM (f)
_,
D m
ZODIOLESTES
-- 0 c:
r
G5 0
PRO MARTES 0
CD OJ
c
D BRACHYPSALJS
z 0
z ()
)>
m I
AELUROCYON :0
0
OLJGOBUNIS z
--- (f)

PAROLJGOBUNIS

Figure 9.6. Primitive aeluroids and paleomustelids in North America. An arrow indicates a presumed immigration event. Solid lines indicate known
geologic range in North America. Dotted lines indicate inferred phylogenetic relationships. Dashed and dotted lines indicate known Old World ranges that
are earlier than the North American records. Dashed lines indicate possible earlier or later North American records.

are almost certainly other instances where this and other variation Valkenburgh, 1989). For example, hypercamivorous forms have
has complicated the taxonomy of fossil mustelids (e.g., the species well-developed, shearing camassials, whereas those with a more
of Promartes). Mustelids are also geographically widespread. Al- omnivorous diet have camassials modified for crushing and grind-
though predominantly Holarctic, they also occur in Africa and South ing. A cautionary note is provided by the North American badger
America. There are 11 genera and 20 species extant in North Taxidea, which as an adult subsists mainly on small mammals but has
America. a dentition adapted for both shearing and, more obviously, crushing
Recent mustelids display a large amount of behavioral and loco- (Long, 1980). Unfortunately, there is limited cranial and very little
motor diversity. They are terrestrial (including burrowers), aquatic described postcranial material of fossil mustelids to give additional
(both fresh and marine water), and arboreal. Many are highly preda- information about mustelid paleobiology.
ceous, feeding almost exclusively on warm-blooded vertebrates; Mustelids span the greatest adult size range in the Camivora
others are omnivorous, taking advantage of a wide variety of food (Nowak, 1991), from as little as 25 g for Mustela nivalis, the least
items. Morphology of the dentition, skull, and skeleton can be used weasel, to as much as 45 kg for Enhydra, the sea otter, a differ-
to characterize carnivore dietary and locomotor adaptations and ence greater than three orders of magnitude. Body mass of fossil
to recognize these behaviors in fossil mustelids (Radinsky, 1982; mustelids can be estimated by several methods. Lower camassial
Van Valkenburgh, 1985, 1987, 1988, 1989). Dental morphology, in size is a good predictor of body mass in carnivores in general and an
particular, is widely used to infer diet in fossil camivorans (Van excellent predictor for mustelids in particular (Legendre and Roth,
 Jon A. Baskin 168

-a
MIOCENE EPOCH
I PLIOCENE fii
~
HEMING- CLAREN- HEMP- 0
FOR DIAN BARSTOVIAN DONIAN HILLIAN BLANCAN 0
m NALMA
m I r m I r ml r m Ir zm
Ph Im I r
(1)
Ir
I\) ..... .....
0 01 0 01 MA
r-
m CRATEROGALE
~ ·-
:0
~
D LEPTARCTUS

~
m
MIOMUSTELA
I--
iJ PUOGALE
MARTINO GALE

;!::
iJ-- . BUISNICT/S
m
-a
I
3 if-
BRACHYOPS/GALE
z
~ --
MEPH/TIS m
if ___
SPILOGALE 0
s:
c
~
M/ONICT/S

iJ ?UMNONYX
-----
D
D
SATHER/UM
_,
(f)

~
- m
iJ ENHYDRITHERIUM
c:
0
~
g ·········
PLIOTAXJDEA

LT TAXJDEA OJ
~
0
.[l MARTES ()
D/NOGALE
--- I
PUONICT/S .[l :0
iJ D,,- EOMELUVORA 0
-------·-·----- HOPL/CTIS
---
(f)
z
;!:: STHENICTIS CAN/MARTES
c
~ iJ '- ---- D -
m
c
--- ------- - -PLESIOGULO
D
FERINESTRIX
~
m

u. _.
LUTRAVUSD
CERNICTIS
SMINTHOS/NIS

iJ TRIGONICTIS

D
MUSTELA
--
Figure 9.7. Neomustelids in North America. See Figure 9.6 caption for information on symbols.

1988). Small and medium-sized mustelids tend to have a lower body those with more compact bodies, such as badgers, skunks, and the
mass for a given carnassial size than members of other carnivore wolverine, have a lower predicted body mass than actual. Given the
families (Legendre and Roth, 1988, Figure 2). Van Valkenburgh problems of determining body shape in extinct mustelids, estimated
(1985) has developed regression equations to estimate body mass masses for fossil species must be treated with caution.
from either skull length or head-body length, based on data from The ischyrichtines Eamellivara, its close relative Perunium, and
seventy species of extant carnivorans. These equations can be used Plesiagula are perhaps the largest neomustelids. The skull of the late
to estimate body mass of extinct camivorans, although not without Miocene Old World Perunium has a condylobasal length of 192 mm,
difficulties for mustelids, given the variety of body forms within and Plesiagula includes specimens that are nearly as large. The re-
this family. Those mustelids with very elongate bodies, such as ot- gression equation for body mass versus ml area in mustelids (Leg-
ters and weasels, have predicted masses higher than actual, whereas endre and Roth, 1988) yields a body mass of 45 kg for Plesiagula
 Mustelidae 169

lindsayi, given an average ml area of 27.1x10.6 mm (Harrison, in the late early Arikareean, Monroe Creek Formation (Tedford
1981). Based on its skull length, the early Miocene oligobunine et al., 1987). Oligabunis (and Paraligabunis) can be derived from
Aeluracyan is the largest known mustelid. However, the relatively Pramartes, and Aeluracyan from Oligabunis. The skull shapes of
short m 1 of all oligobunines underestimates their size relative to that Zadialestes and probably Promartes are intermediate to those of
of the neomustelids. For example, although Plesiagula has a much mustelids and viverrids (Radinsky, 1982, Figure 3) because of their
larger ml, Aeluracyan has a larger skull. The condylobasal length of relatively long tooth row and jaw lengths, which are similar in both
an A. brevifacies skull (F:AM 25430) from the Marsland Formation genera and fall outside the range reported by Radinsky for modem
of Nebraska (locality CP104B) is 199 mm with an associated ml mustelids. The type specimens of Promartes and Zodialestes were
length of 19.8 mm, versus dimensions of 184 mm and 25.1 mm, found in the burrows of the extinct beaver Palaeacastar. Zadialestes
respectively, for a Chinese specimen of Plesiagula. Therefore the has been categorized as a semifossorial form. However, the relative
regression equation for body mass calculated from carnassial size size of the olecranon process of the ulna is smaller in Zodialestes
in living mustelids (Legendre and Roth, 1988) will give too low a (estimated at 0.20 from Riggs, 1945, Figure 43) than in semifosso-
value for Aeluracyan (and other paleomustelids). rial forms (range 0.23-0.35, Van Valkenburgh, 1987). The ratio of
The extant wolverine Gula has a body mass of7 to 32 kg (Nowak, femur to fourth metatarsal length is 2.5 in Zadialestes, which is at
1991 ). Male Gula from Norway have a condy lobasal length of about the lower limit of the range of semifossorial carnivores and closer
150 mm, an average ml length of 21.8 mm, and an average mass to the average for terrestrial forms. Zadialestes does not exhibit any
of 15 kg (Wiig, 1989). The A. brevifacies skull is about 33 percent notable specializations for semifossorial locomotor behavior, nor
longer than that of Gula. Aeluracyan ferax is significantly larger does it have the very long body and short limbs of weasels that
than A. brevifacies. The breadth across the mastoid processes is 138 actively enter burrows in search of prey. It seems more likely that
mm in A. ferox, 110 mm in the early Miocene A. brevifacies, and an Zodialestes utilized burrows as dens, as do foxes. Pramartes has
average of90 mm for Gula. Body mass in mammals usually scales to an estimated olecranon index of 0.26, which falls within the range
the cube oflinear measurements. This suggests Aeluracyan weighed of semifossorial, terrestrial (0.16-0.36), and scansorial (0.15-0.29)
at least two to three times as much as Gula and perhaps as much as carnivores (Van Valkenburgh, 1987). The humerus has a wide me-
a leopard (see Figure 9.1). dian epicondyle and a prominent lateral epicondylar ridge at its
Mustelid diversity is well reflected in the Neogene fossil record, distal end. The epicondyle is particularly well developed in burrow-
in spite of the fact that mustelids are among the rarest of fossil ing mammals such as Taxidea (Hildebrand, 1974). Promartes may
carnivores. Their usually small size, low population densities, and therefore have had a badgerlike digging ability.
preference for forested habitats may account for the paucity of fos- Aeluracyan (Figure 9.1) is perhaps the most remarkable mustelid
sil specimens. The history of the Mustelidae in North America is of all time. Its large size has already been remarked upon. The
one of numerous dispersal events from Eurasia, which are used variables used to characterize its skull morphology are within the
in the biostratigtaphic definition of Neogene North American land range of modem mustelids, with the exception of an extremely ro-
mammal "ages" in the Miocene and Pliocene epochs (Tedford et al., bust cross-sectional area of the mandible (Radinsky, 1982). Riggs
1987). Important immigration events that include mustelids mark the (1945) compares Aeluracyan to the felids (and nimravids). Nim-
start of the late Arikareean, Hemingfordian, late Hemphillian, and ravids became extinct in North America in the early Arikareean
Blancan (Tedford et al., 1987). This exchange of mustelids is very (earliest Miocene), until the reappearance of Barbaurofelis in the
one sided, with no unequivocal example of an Old World mustelid Clarendonian (late Miocene). Riggs (1945) attributes their extinc-
genus that had its origins in the New World. Vermeij (1991) and tion to displacement by Aeluracyan. Given the rarity of Aelura-
Webb (1991) discuss factors causing asymmetric biotic exchanges. cyan, especially as compared to the abundance of early Arikareean
Marshall (1985) distinguishes five types of dispersal and two of ap- nimravids, this seems unlikely; rather it seems more likely that a
parent dispersal (pseudodispersal) events, and examples of all seven mustelid was becoming more catlike in the absence of competition
kinds can probably be found for North American mustelids. Unfor- from felids or nimravids, but may have been limited by competition
tunately, the poor quality of their fossil record frequently makes from hypercamivorous canids (Van Valkenburgh, 1991). Felids ap-
it difficult to distinguish dispersants from pseudodispersants. The pear in North America in the Hemingfordian (late early Miocene),
majority of genera of North American neomustelids are dispersants after the extinction of Aelurocyan. Morphometric indexes to deter-
from the Old World. Many had a Holarctic distribution. In contrast, mine diet preference (Van Valkenburgh, 1988) in Aelurocyan are
with one exception (Eamellivara cf. E. wimani), all North American similar to the values for Plesiagula, which Van Valkenburgh placed
Tertiary species are considered endemics, although more detailed in the meat/bone diet group. Aeluracyan has a humeroradial index
comparisons with Old World species are needed. of 79 and a femorotibial index of 84 (measurements from Riggs,
In the late Eocene and Oligocene, the Canidae and Arctoidea oc- 1945), similar to those for the felid Smiladan or the nimravid Ha-
cur allopatrically for the most part: the canids exclusively in North plaphaneus (Gonyea, 1975). These values are typical of cats that
America and arctoids predominantly in Eurasia (Tedford, 1976). stalk and ambush their prey from dense cover.
Mustelavus is the only pre-Arikareean musteloid in North Amer- During the 6 million-year interval spanning the late Arikareean
ica. In the late Arikareean (earliest Miocene), five or six genera and Hemingfordian (early Miocene), there is a complete turnover
of oligobunines appear in North America. Promartes and perhaps in mustelid genera as the paleomustelids were, for the most part,
Zadialestes are immigrants. Pramartes may occur slightly earlier, replaced by neomustelids and the number of genera doubled.
 Jon A. Baskin 170

Neomustelids first appear in the earliest Miocene of Europe and the early Clarendonian (late early Miocene) to the early Blancan,
emigrate to North America in the early Hemingfordian. During this Hoplictis, Eomellivora, Plesiogulo, and Ferinestrix succeed each
time period, there are four genera of oligobunines and three of neo- other and serve as biostratigraphic markers for their respective time
mustelids in North America. In the late Hemingfordian, four addi- periods (Tedford et al., 1987). In part this may be a result of replace-
tional genera arrive, at which time there are two genera of oligob- ment, because Hoplictis may be ancestral to Eomellivora. However,
unines and six of neomustelids. Brachypsalis and Potamotherium Eomellivora and Plesiogulo may co-occur in the Old World for at
are the only two oligobunines to survive past the early Heming- least part of their ranges. The extinction of Plesiogulo apparently
fordian. Brachypsalis, which appears at that time (and can be de- occurs at about the same time (earliest Pliocene) in the Old and New
rived from Promartes), is the most successful oligobunine. It is Worlds. The origin of taxidiines is problematical. The forelimb of
represented by the largest number of specimens and ranges into the Pliotaxidea indicates that Pliotaxidea was as fossorial as Taxidea
Clarendonian. The skull and dentition of Brachypsalis are much (Wagner, 1976). In the late Clarendonian and early Hemphillian,
more like those of a canid or procyonid than a mustelid and it is ?Limnonyx and Enhydritherium, two lutrines with affinities to liv-
a comparatively light-limbed form (Galbreath, 1955). This atypi- ing otters, appear in North America.
cal mustelid morphology may account for its success in the face of The late Hemphillian to early Blancan (latest Miocene to early
the neomustelid invasion. The otterlike oligobunine Potamotherium Pliocene) is the time of greatest faunal diversity for North Amer-
first known from the latest Oligocene of Europe also appears in ican Mustelidae. Thirteen genera are present in the late to latest
the early Hemingfordian. The lutrine Mionictis, a late Hemingfor- Hemphillian. Diversity increases mainly as a result of the appear-
dian immigrant, is much better represented in North America. Its ance of new immigrants. Galictines, first represented in the New
skeleton also possesses adaptations for an aquatic way of life (Gins- World by the early Hemphillian Lutravus, are the most impor-
burg, 1961 ). The leptarctine Craterogale, an earliest Hemingfordian tant additions to the mustelid fauna. The history of the galictines
immigrant, may have given rise to Leptarctus in North America, al- is very complex, in which immigrant genera from Eurasia give
though this genus is also known from Asia. Leptarctines have highly rise to New World endemic genera (Ray, Anderson, and Webb,
distinctive skulls and dentitions. Postcranial material, other than the 1981). Today galictines are a Neotropical group. In the late Blan-
atlas and axis, has not been described. Leptarctus has frequently can, mustelid diversity declines to nine genera, the present Nearctic
been compared to Taxidea, the North American badger. Qiu and diversity. The major difference with the present is the absence of a
Schmidt-Kittler (1982) suggest that leptarctines were omnivorous mustelid in the meat/bone-eating guild following the extinction of
with a high proportion of arthropods and plants in their diets, as Ferinestrix. Gulo, the wolverine, first appears some 2 million years
in Otocyon, the bat-eared fox, and Nyctereutes, the raccoon dog. later in the Irvingtonian (middle Pleistocene). The first undoubted
Leptarctus is one of the most widespread and best represented of Mustela (weasels, etc.) appears in North America in the Blancan
the Miocene mustelids. Its extinction occurs at the time of diver- (Tedford et al., 1987).
sification of the North American badgers. Miomustela, Dinogale, A better characterization of the trophic and locomotor roles of
Martes, Plionictis, and Sthenictis all appear to have their origins in North American mustelids would help provide insight into their ex-
the Old World. tinction and evolution. It would be interesting to determine whether
From the early Barstovian to early Hemphillian (middle to late displacement as a result of competition by similarly adapted taxa
Miocene), eight to ten genera ofmustelids are represented in North was an important factor causing extinction. The major increases in
America, a continuation of the level of diversity seen in the Hem- mustelid diversity in the late Hemingfordian and late Hemphillian
ingfordian. Important events in this time period include the first ap- result from the addition of new immigrant taxa, with no apparent
pearance of mephitines, ischyrictines, and taxidiines in North Amer- major impact on native mustelids, perhaps because the newcom-
ica, but they had no major effect on generic diversity. Mephitines ers were occupying new niches. Other appearances may have been
are represented by at least two invasions (late middle Miocene and made possible as a result of replacement of a niche recently vacated
early Pliocene) from Eurasia and subsequent radiations. The ear- by extinction.
lier radiation, derived perhaps from the European Mesomephitis,
includes very small, carnivorous forms with sectorial upper camas-
sials, which are very similar to those of weasels. P liogale is the most ACKNOWLEDGMENTS
primitive representative of this group, which also includes Martino-
gale and Buisnictis, which became extinct in the Blancan (early Respectfully dedicated to the memory of Leonard Radinsky. Any-
Pliocene). The living genera Mephitis, Spilogale, Conepatus plus thing of value in these two chapters was made possible through the
the extinct Osmotherium, Brachyprotoma, and perhaps Brachyop- study of the F:AM and AMNH collections. R. Tedford freely al-
sigale may represent a second invasion of skunks and radiated in lowed me to study these specimens and shared some of his thoughts
the latest Hemphillian and Blancan from a form closely related to on arctoid relationships with me. H. Galiano enlivened my visits to
the Eurasian Promephitis (which, however, is too specialized to be New York with discussions on musteloid systematics. L. Ginsburg
ancestral to the Recent genera). of the MN is gratefully acknowledged for putting his collections at
The six genera of Miocene and Pliocene North American is- my disposal. Special thanks are due C. Janis for editorial assistance
chyrictines appear to have their origins in the Old World. From and B. Van Valkenburgh for comments on functional morphology.
 Mustelidae 171

Figures 8.1 and 9. I in the procyonid and mustelid chapters were Flynn, J. J., & Galiano, H. (1982). Phylogeny of early Tertiary Carnivora,
drawn by B. Regal. Support for these chapters was provided by with a description of a new species of Protictis from the middle
Organized Faculty Research of Texas A&I University and by Mr. Eocene of northwestern Wyoming. American Museum Novitates,
2725, 1-64.
and Mrs. Jerome Madans. C. Tipton and G. Longoria of Texas A&I
Flynn, J. J., Neff, N. A., & Tedford, R. H. (1988). Phylogeny of the Car-
University assisted with the preparation of the figures. nivora. In The Phylogeny and Classification of the Tetrapods, Vol.
2: Mammals, ed. M. J. Benton, pp. 73-116. Oxford: Systematics
Association Special Volume, no. 35B. Clarendon Press.
Frailey, D. (1978). An early Miocene (Arikareean) fauna from northcentral
REFERENCES
Florida (The SB-IA Local Fauna). Occasional Papers ofthe Museum
of Natural History, University of Kansas, 75, 1-20.
Beaumont, G., de & Weidmann, M. (1981). Un crane de Plesictis (Mam- Furlong, E. L. (1932). A new genus ofotter from the Pliocene of the northern
mifere, Carnivore) dans la Molasse subalpine oligocene fribour- Great Basin Province. Carnegie Institution of Washington Publica-
geoise, Suisse. Bulletin de la Societe Vaudoise des Sciences Na- tion, 418, 93-103.
turelles, 75, 249-56. Galbreath, E. C. (1955). Skull and skeleton of a mustelid, Brachypsalis
Berta, A., & Morgan, G. S. (1985). A new sea otter (Carnivora: Mustelidae) from the Miocene of northeastern Colorado. University of Kansas
from the Late Miocene and Early Pliocene (Hemphillian) of North Paleontological Contributions, Vertebrata, 5, 1-15.
America. Journal of Paleontology, 59, 809-19. Gazin, C. L. (1934). Upper Pliocene mustelids from the Snake River Basin
Bjork, P. R. (1970). The Carnivora of the Hagerman Local Fauna (late of Idaho. Journal of Mammalogy, 15, 137-49.
Pliocene) of Southwestern Idaho. Transactions of the American (1936). A new mustelid carnivore from the Neocene beds of northwest-
Philosophical Society, new series, 60, 1-54. ern Nebraska. Journal of the Washington Academy of Sciences, 26,
Bonis, L. de (1981). Contribution a l'etude du genre Palaeogale Meyer 199-207.
(Mammalia, Carnivora). Anna/es de Paleontologie (Vertebres), 67, Geoffroy Saint-Hillaire, E. (1833). Palaeontographie. Considerations sur
37-56. des ossemens fossiles, la plupart inconnus, trouves et observes
Bryant, L. J. (1968). A new genus of mustelid from the Ellensburg Forma- dans Jes bassins de I' Auvergne. Revue Encyclopedique, Paris, 59,
tion of Washington. Los Angeles County Museum Contributions in 76-95.
Science, 139, 1-6. Geoffroy Saint-Hillaire, E., & Cuvier, G. L. C. F. D. (1795). Memoire surune
Butterworth, E. M. (1916). A new mustelid from the Thousand Creek nouvelle division des mammiferes, et sur Jes principes qui doivent
Pliocene of Nevada. University of California Publications, Depart- servir de base dans cette sorte de travail. Magazin Encyclopedique.
ment of Geological Sciences, Bulletin, 10, 21-4. Millin, Noel et Warens, Paris, 2, 164-90.
Clark, J., & Guensburg, T. E. (1972). Arctoid genetic characters as related Gervais, P. (1848-1852). Zoologie et Paleontologie Franfaises. Paris: Arthus
to the genus Parictis. Fieldiana Geology, 26, 1-76. Bertrand.
Cook, H.J., & Macdonald, J. R. (1962). New Carnivora from the Miocene (1876). Zoologie et Paleontologie Genera/es, Nouvelles Recherches sur
and Early Pliocene of western Nebraska. Journal of Paleontology, les Animaux Vertebres Vivants et Fossiles. Paris, serie. 2, 1-72.
36, 560-7. Ginsburg, L. (1961). La faune des Carnivores miocenes de Sansan (Gers).
Cope, E. D. (1868). An addition to the vertebrate fauna of the Miocene Memoire du Museum national d'Histoire naturelle. Nouvelle serie,
period, with a synopsis of the extinct Cetacea of the United States. serie C, 9, 1-190.
Proceedings of the Academy of Natural Sciences of Philadelphia, Ginsburg, L., & Morales, J. (1992). Contribution a la connaissance des
1867, 137-8. Mustelides (Carnivora, Mammalia) du Miocene d'Europe Trochic-
(1873). Synopsis of new Vertebrata from the Tertiary of Colorado, ob- tis et Ischyrictis, genres affines et genres nouveaux. Academie des
tained during the summer of 1873. Washington, D.C.: U.S. Govern- Sciences de Paris, Comptes-rendus, 315, Serie II, 111-16.
ment Printing Office, 19 pp. Gonyea, W. J. (1975). Behavioral implications of saber-toothed felid mor-
(1879). On the genera of Felidae and Canidae. Proceedings of the phology. Paleobiology, 2, 332-42.
Academy of Natural Sciences of Philadelphia, 1879, 168-94. Gray, J.E. (1865). Revision of the genera and species of Mustelidae con-
(1881). Miocene dogs. American Naturalist, 15, 497. tained in the British Museum. Proceedings of the Zoological Society
(1890). On two new species of Mustelidae from the Loup Fork Miocene of London, 1865, 100-54.
of Nebraska. American Naturalist, 24, 950-2. Hall, E. R. (1930). Three new genera of Mustelidae from the later Tertiary
(1892). A hyena and other Carnivora from Texas. American Naturalist, of North America. Journal of Mammalogy, 11, 146-54.
26, 1028-9. (1935). A new mustelid genus from California. Journal of Mammalogy,
(1893). A preliminary report on the vertebrate paleontology of the 16, 137-8.
Llano Estacado. Geological Survey of Texas, Fourth Annual Report, (1944). A new genus of American Pliocene badger, with remarks on
1-89. the relationship of badgers of the Northern Hemisphere. Carnegie
Crusafont Pair6, M. (1950). Limnonyx un nuevo Lutrido de! Mioceno Institution of Washington Publication, 551, 9-23.
espafiol. Instituto Geologia y Minas de Espana, Notas et Comuns, Harrison, J. A. (1981). A review of the extinct wolverine, Plesiogulo (Car-
20, 131-42. nivora: Mustelidae), from North America. Smithsonian Contribu-
Dalquest, W.W. (1978). Early Blancan mammals of the Beck Ranch Local tions to Paleobiology, 46, 1-27.
Fauna of Texas. Journal of Mammalogy, 59, 269-98. (1983). The Carnivora of the Edson Local Fauna (late Hemphillian),
Dorr, J. A., Jr. (1954). Hypsoparia bozemanensis; and new genus and species Kansas. Smithsonian Contributions to Paleobiology, 54, 1-42.
of leptarctine mustelid from the late Miocene Madison Valley For- Hibbard, C. W. (1941). New mammals from the Rexroad Fauna, Upper
mation of Montana. Annals of the Carnegie Museum, 33, 179-84. Pliocene of Kansas. American Midland Naturalist, 26, 337-68.
Douglas, E. (1904). New vertebrates from the Montana Tertiary. Annals of ( 1950). Mammals of the Rexroad formation from Fox Canyon, Kansas.
the Carnegie Museum, 2, 145-99. Contributions from the Museum ofPaleontology, University ofMichi-
Erxleben, J. C. P. ( 1777). Systema regni animal is, per classes, ordines, gen- gan, 8, 113-92.
era, species, varietates, cum synonymis et historia animalium. Clas- (1954). Second contribution to the Rexroad Fauna. Transactions of the
sis I. Mammalia. Leipzig: Weygand. Kansas Academy of Science, 57, 221-37.
 Jon A. Baskin 172

(1958). A new weasel from the Lower Pleistocene of Idaho. Journal of Peterson, 0. A. (1907). The Miocene beds of western Nebraska and east-
Mammalogy, 39, 245-6. ern Wyoming and their vertebrate faunae. Annals of the Carnegie
Hildebrand, M. (1974). Analysis of Venebrate Structure. New York: Wiley. Museum, 4, 21-72.
Hunt, R. M., Jr. (1974). Daphoenictis, a cat-like carnivore (Mammalia, Am- (1910). Description of new carnivores from the Miocene of western
phicyonidae) from the Oligocene of North America. Journal of Pa- Nebraska. Memoirs of the Carnegie Museum, 4, 205-78.
leontology, 48, 1030-47. Petter, G. (1967). Mustelides nouveaux du Vallesien de Catalogne. Anna/es
(1989). Evolution of the aeluroid Carnivora: significance of the ventral de Paleontologie (Vertebres), 53, 93-114.
promontorial process of the petrosal, and the origin of the basicra- Pine!, (1792). Actes de la Societe d'histoire naturelle de Paris, 1, 55.
nium in the living families. AmericanMuseumNovitates, 2930, 1-30. Piveteau, J. (1961). Carnivora. In Traite de Paleontologie, VI, pp. 641-802.
Hunt, R. M., Jr., & Skolnick, R. (1996). The giant mustelid Megalictis from Paris: Masson.
the early Miocene carnivore dens at Agate Fossil Beds National Mon- Pocock, R. I. (1921 ). On the external characters and classification of the
ument, Nebraska: earliest evidence of dimorphism in New World Mustelidae. Proceedings of the Zoological Society of London, 1921,
Mustelidae (Carnivora, Mammalia). Contributions to Geology, Uni- 803-37.
versity of Wyoming, 31, 35-48. a
Pomel, A. (1846). Memoire pour servir la geologie paleontologique des
Lange, B. (1970). Mustelictis piveteaui, mustetide nouveau des phosphorites terrains tertiaires du departement de 1' Allier. Bulletin de la Societe
du Quercy. Anna/es de Paleontologie (Vertebres), 5, 75-88. Geologique de France, 2 ser., 3, 353-73.
Legendre, S., & Roth, C. (1988). Correlation of carnassial tooth size and Qiu Z., & Schmidt-Kittler, N. (1982). On the phylogeny and zoogeogra-
body weight in Recent carnivores (Mammalia). Historical Biology, phy of the leptarctines (Carnivora, Mammalia). Paliiontologische
1, 85-98. Zeitschrift, 56, 131-45.
Leidy, J. (1857). Notices of extinct Vertebrata discovered by Dr. F. V. Hayden, Radinsky, L. B. (1973). Are stink badgers skunks? Implications of neu-
during the expedition to the Sioux country under the command of roanatomy for mustelid phylogeny. Journal of Mammalogy, 54,
Lieut. G. K. Warren. Proceedings ofthe Academy ofNatural Sciences 585-93.
of Philadelphia, 8, 311-12. (1982). Evolution of skull shape in carnivores. 3. The origin and early
(1873). Contributions to the extinct vertebrate fauna of the western ter- radiation of the modern carnivore families. Paleobiology, 8, 177-95.
ritories. Report of the USGS of the Territories, F. V. Hayden, US Rafinesque, C. S. (1820). Annual synopsis of new genera and species of
Geologist in Charge, 14-358. animals, plants, & c discovered in North America. Annals of Nature,
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundem 1, 1-25.
classes, ordines, genera, species cum characteribus, differentiis, syn- Ray, C. E., Anderson, E., & Webb, S. W. (1981). The Blancan carnivore
onymis, locis. Editio decima, reformata. Stockholm: Laurentii Sal vii. Trigonictis (Mammalia: Mustelidae) in the eastern United States.
Long, C. A. (1980). Provisional classification and evolution of the badgers. Brimleyana, 5, 1-36.
In Worldwide Furbearer Conference Proceedings, Vol. 1, eds. J. A. Riggs, E. S. (1942). Preliminary description of two Lower Miocene car-
Chapman & D. Pursley, pp. 55-85. nivores. Field Museum of Natural History, Geological Series, 8,
Loomis, F. B. (1932). The small carnivores of the Miocene. American Jour- 59-62.
nal of Science, 24, 316-29. (1945). Some early Miocene carnivores. Field Museum of Natural His-
Macdonald, J. R. (1956). A new Clarendonian mammalian fauna from the tory, Geological Series, 9, 69-114.
Truckee Formation of western Nevada. Journal of Paleontology, 30, Schlosser, M. (1887). Die Affen, Lemuren, Chiropteren, Insectivoren, Mar-
186-202. supialier, Creodonten, und Carnivoren des europaischen Tertiars.
( 1970). Review of the Miocene Wounded Knee Faunas of southwestern Beitriige zur Paliiontologie und Geologie Osterreich-Ungarns und
South Dakota. Bulletin ofthe Los Angeles County Museum ofNatural des Orients, Vienna, part I, 6, 1-224.
History, Science, 8, 1-82. (1903). Die fossilen Saugthiere Chinas nebst einer Odontographie
Marshall, L. G. (1985). Geochronology and land-mammal biochronology of der recenten Antilopen. Abhandlungen der Koniglich Bayerischen
the Transamerican fauna! interchange. In The Great American Biotic Akademie der Wissenschaften, 22, 1-22.
Interchange, eds. F. G. Stehli & S. D. Webb, pp. 49-88. New York: Sclunidt-Kittler, N. (1981). Zur Stammesgeschichte der marderverwandten
Plenum. Raubtiergruppen (Musteloidea, Carnivora). Ecologae geologicae
Matthew, W. D. (1901). Fossil mammals of the Tertiary of northeastern Helvetiae, 74, 753-801.
Colorado. Memoirs of the American Museum of Natural History, 1, Schreber, J. C. D. (1776). Die Saugthiere in Abbildungen nach der Natur,
353-447. mit Beschreibungen. Wolfgang Walther, Erlangen, 3 (Heft 17), 290-
(1907). A Lower Miocene fauna from South Dakota. Bulletin of the 312.
American Museum of Natural History, 23, 169-219. (1778). Die Saugthiere in Abbildungen nach der Natur, mit Beschrei-
(1924). Third contribution to the Snake Creek Fauna. Bulletin of the bungen. Leipzig, vol. 3, 520-89.
American Museum of Natural History, 50, 59-210. Scott, W. B., & Jepsen, G. L. (1936). The mammalian fauna of the White
Matthew, W. D., & Granger, W. (1924). New Carnivora from the Tertiary of River Oligocene, Part I: Insectivora and Carnivora. Transactions of
Mongolia. American Museum Novitates, 104, 1-9. the American Philosophical Society, new series, 28, 1-153.
Mayet, L. ( 1908). Etude des mammireres miocenes des sables de l'Orleanais Segall, W. (1943). The auditory region of the arctoid carnivores. Field Mu-
et des faluns de la Touraine. Anna/es de l'Universite de Lyon, 24, seum of Natural History, Zoological Series, 29, 33-59.
1-316. Simpson, G. G. (1945). The principles of classification and a classification
Merriam, J. C. (1911 ). Tertiary mammal beds of Virgin Valley and Thousand of mammals. Bulletin of the American Museum of Natural History,
Creek in northwestern Nevada, Part 2: Vertebrate faunas. University 85, 1-350.
of California Publications, Department of Geological Sciences, Bul- (1946). Palaeogale and allied early mustelids. American Museum of
letin, 6, 199-304. Natural History Novitates, 1320, 1-14.
Meyer, H. von (1846). [no title: letter on various fossils.] Neues Jahrbuch Skinner, M. F., Skinner, S. M., & Gooris, R. J. (1977). Stratigraphy and
for Mineralogie, Geologie und Palaeontologie, (1846), 462-76. biostratigraphy of late Cenozoic deposits in Central Sioux County,
Muizon, C. de (1982). Les relations phylogenetiques des Lutrinae (Mustel- western Nebraska. Bulletin of the American Museum of Natural His-
idae, Mammalia). Geobios, memoire special, 6, 259-77. tory, 158, 263-370.
Nowak, R. M. (1991). Walker's Mammals of the World. Baltimore: John Tedford, R. H. (1976). Relationship of pinnipeds to other carnivores (Mam-
Hopkins. malia). Systematic Zoology, 25, 363-74.
 Mustelidae 173

Tedford, R.H., & Frailey, D. (1976). Review of Some Carnivora (Mammalia) nivora): evolutionary interactions among sympatric predators. Pale-
from the Thomas Farm Local Fauna (Hemingfordian: Gilchrist obiology, 17, 340-62.
County, Florida). American Museum Novitates, 2610, 1-9. Vermeij, G. (1991). When biotas meet: understanding biotic interchange.
Tedford, R. H., Skinner, M. F., Fields, R. W., Rensberger, J. M., Whistler, Science, 253, 1099-1104.
D. P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S. Wagner, H. (1976). A new species of Pliotaxidea (Mustelidae, Carnivora)
D. (1987). Fauna! succession and biochronology of the Arika- from California. Journal of Paleontology, 50, 107-27.
reean through Hemphillian interval (late Oligocene through earliest Waterhouse, G. R. (1838). [Untitled: On the skull of the American badger.]
Pliocene epochs) in North America. In Cenozaic Mammals of North Proceedings of the Zoological Society of London, 1838, 153-4.
America, ed. M. 0. Woodburne, pp. 153-210. Berkeley: University Webb, S. D. (1969). The Burge and Minnechaduza Clarendonian mam-
of California Press. malian faunas of North-Central Nebraska. University of California
Van Valkenburgh, B. (1985). Locomotor diversity within past and present Publications in Geological Sciences, 78, 1-191.
guilds of large predatory mammals. Paleobiology, 11, 406-28. (1991). Ecogeography and the Great American Interchange. Paleobiol-
( 1987). Skeletal indicators of locomotor behavior in living and extinct ogy, 17, 266-80.
carnivores. Journal of Vertebrate Paleontology, 7, 162-82. Wiig, 0. (1989). Craniometric variation in Norwegian wolverines Gula gulo
(1988). Trophic diversity in past and present guilds of large predatory L. Zoological Journal of the Linnean Society, 95, 177-204.
mammals. Paleobiology, 14, 155-73. Zakrzewski, R. J. (1967). The systematic position of Canimanes?
(1989). Carnivore dental adaptations and diet: a study of trophic diversity from the upper Pliocene of Idaho. Journal of Mammalogy, 48,
within guilds. In Carnivore Behavior, Ecology, and Evolution, ed. J. 293-7.
L. Gittleman, pp. 410-36. Ithaca: Cornell University Press. Zdansky, 0. (1924). Jungtertiare carnivoren Chinas. Palaeontologica Sinica,
(1991). Iterative evolution ofhypercarnivory in canids (Mammalia: Car- series C, 2, 1-155.
10 Ursidae
ROBERT M. HUNT, JR.

INTRODUCTION

The Ursidae occur in North America from late Eocene (Chad-

ronian) to present, a span of about 37 Ma. The living ursine bears
represent only a fraction of the diversity that has been discovered
in the North American fossil record. Three subfamilies (Amphi-
cynodontinae, Hemicyoninae, Ursinae) succeed each other in time
not only on this continent, but in Eurasia, and record a gradually
changing picture of ursid evolution from small generalized arctoid
camivorans, to digitigrade mid- to large-sized active predators, ter-
minating in plantigrade omnivorous ursine bears that include the
largest camivorans (arctodonts) that ever lived. Figure 10.1. The digitigrade hemicyonine ursid Phoberocyon (male and fe-
These ursid subfamilies represent separate episodes of diversi- male) from the early Miocene of North America (restoration by Ted Browne).
fication: radiations of Oligocene amphicynodonts, Miocene hemi- Phoberocyon is representative of the mid-Cenozoic radiation of cursorial
cyonines (Figure 10.1), and Miocene to Pleistocene ursines (Fig- ursids that flourished in Eurasia and North America, paralleling the later
Cenozoic canids in many dental and skeletal traits.
ure 10.2), centered in Eurasia. In fact, a most interesting aspect
of ursid evolution is the lack of phyletic continuity between North
American ursid genera. The North American Cenozoic record is form the inner and posterior margins of the bulla. In most ursids
punctuated by the repeated entrance of immigrant ursid stocks that the bulla does not inflate to any degree during ontogeny (in some
persist for a time, then become extinct. Consequently, ursid first extinct species of Cephalogale and amphicynodonts the bulla is
appearances serve as important and useful biostratigraphic first ap- somewhat inflated); hence it retains a primitive configuration. An
pearance datums in the establishment of a Cenozoic biochronology extremely similar pattern of bulla ontogeny in otariid pinnipeds in-
for North America. dicates that their auditory bulla is Type A and, in company with
dental features, demonstrates their descent from early ursids (Hunt
and Barnes, 1994).
DEFINING FEATURES OF THE FAMILY URSIDAE Early amphicynodonts are long-skulled somewhat short-snouted
animals, and later ursids reflect this ancestry in their long postorbital
distances and shortened rostra; however, both hemicyonine and ur-
CRANIAL sine radiations evolve short-skulled (brachycephalic: Cephalogale,
The structure of the basicranium is diagnostic for ursids in com- Tremarctos, Ursus malayanus) and long-skulled (dolichocephalic:
pany with key dental features. Living ursine ursids (e.g., Ursus Hemicyon, Ursus arctos, Figure 10.2A, B) species.
americanus, Hunt, 1974, p. 35, Pl. 4) possess a primitive arctoid
auditory bulla termed Type A, comprised of four ontogenetic ele-
DENTAL
ments (ectotympanic, rostral, and two caudal entotympanics). The
ectotympanic is the major element contributing to the bulla, flooring The dental formula of living ursine bears (I3/3, Cl/l, P4/4, M2/3)
most of the middle ear cavity. Medial and posterior to the ectotym- is primitive for the family and maintained throughout their evolu-
panic are three small entotympanic ossifications that coalesce to tionary history, modified only by premolar loss in derived species,

174
 Ursidae 175

A
Figure 10.2. Comparison of long-
snouted ursine (A) and hemicyonine
(B) skulls. In the ursine (C) dentition,
the upper carnassial is reduced and
the upper molars greatly lengthened as
crushing teeth; in contrast, the hemi-
cyonine (D) upper dentition maintains
a large upper carnassial with posteri-
orly placed protocone, and quadrate
upper molars produced by migration
c D of the metaconule to the posterointer-
nal margin of the tooth and develop-
ment of the adjacent cingulum. The
molars are not lengthened as in ur-
sine bears, and the marked develop-
E F ment of the talon of M2 never occurs.
In the lower dentition, ursine bears
(E) show modifications corresponding
to the upper teeth: The lower camas-
FT. sial is reduced and m2-3 are length-
H ened and enlarged. Hemicyonines (F)
maintain a large carnassial and do not
greatly enlarge or lengthen their m2-
3. Sexual dimorphism (G) is common
in ursids, where it is known to be pro-
nounced in the larger ursines such as
Tremarctini (Kurten, 1966). The enor-
mous size of some extinct Tremarctini
(Arctodus) relative to the largest living
Ursini (Alaskan brown bear) empha-
sizes their role as the largest Carnivora ever to have walked on land (H). A, from Kurten, 1969; B, from Colbert, 1939; C, D, E, F, from Frick,
1926; G, from Kurten, 1966; H, from Voorhies and Comer, 1982.

chiefly the larger forms. Whereas large canines and incisors are re- the rear border of M2 enlarges and elongates to form a distinctive
tained in living ursines, the premolars are often strongly reduced, heel found only in these camivorans, but because hemicyonine and
the molars are modified in form to produce large quadrate crushing amphicynodontine ursids never develop these talons on M2, molar
platforms, and the heel of the last upper molar (M2) is posteriorly morphology remains a key diagnostic feature. In addition, the up-
elongated. The shearing function of the camassial pair is lost as these per camassial of most ursids (except some amphicynodonts) can be
teeth are reduced in size and modified to the shape of crushing teeth. identified by the more posterior position of the protocone, which is
All living bears exhibit a dentition of this type (Figure 10.2C, E). drawn back (retracted) along the internal face of the tooth; the mo-
However, extinct ursids in many cases have more generalized lar morphology in combination with the retracted P4 protocone and
dentitions and cannot be so easily categorized, in the main due to simple premolars (lacking prominent accessory cusp development
their retention of strong shearing function of the camassials, main- and surrounded by developed cingula) are key dental traits indicat-
tenance of normal premolar batteries, and no posterior prolongation ing an ursid. The lower camassial is less diagnostic: The amphicyn-
of the rear molars (Figure 10.2D, F). How, then, are we to identify odontine m 1 has a low trigonid with equal development of its three
an ursid in the fossil record? cusps and without developed protoconid-paraconid shear; the hemi-
An ursid can be diagnosed using basicranial structure in com- cyonine m 1 develops a tall trigonid and strong protoconid-paraconid
bination with dental morphology. Most ursids are characterized by shear; the ursine m 1 maintains a low trigonid derived from its amph-
a diagnostic upper molar pattern (Figure 10.2C, D) in which the icynodont ancestry (except in Agriotherium), and minimizes shear
crown is quadrate (not triangular), with four principal cusps (para- in most lineages while enlarging m2-3. In many ursids, the second
cone, metacone, protocone, metaconule), and the protocone basin lower molar has a trigonid in which the protoconid and metaconid
opens to the rear of the tooth and is not closed by a postprotocrista. are connected by a symmetrical arcuate ridge (into which the para-
The migration of the metaconule to the posterointemal comer of the conid is often subsumed) forming the anterior border of the tooth
tooth (Beaumont, 1982), and the posteriorly open protocone basin (Figure 10.2E, F).
are useful diagnostic features of ursid molars (except for some prim- Ursids never possess greatly enlarged crushing premolars, nor
itive amphicynodonts in which metaconule.migration has not oc- do they develop hypersectorial camassial blades as in felids and
curred). The quadrate molar form is the result of absence of a strong hyaenids. Ursids always retain the rear molars (M2/m2-3); M3 is
parastyle (best developed in aeluroid camivorans) and the squaring never present. Tooth surfaces are often blunted and worn flat through
of the tooth by the migration of the metaconule. In ursine ursids, the crushing action of the teeth.
 Robert M. Hunt, Jr. 176

POST CRANIAL and there is no firm evidence that diversity approached that seen
at Quercy. The early arctoid record in North America could be ex-
Ursids retain a generalized postcranial skeleton (Figure 10.2G) in
plained as the result of occasional immigration of a few Eurasian
which modifications of the limbs for more cursorial gaits are not
arctoid species into the New World. These immigrants possibly
developed to the extent seen in specialized runners such as canids.
spring from a species-rich radiation of small arctoids that devel-
In living canids, the ability to rotate the radius around the ulna
oped in Eurasia. According to this hypothesis, Eurasia acts through
(supination-pronation) is greatly limited by bone-joint configura-
much of the Cenozoic as the principal staging ground of arctoid
tion; consequently the lower limb is more restricted in its motion
evolution, generating a diverse array of lineages, some of which
and favors a fore-aft path. In ursids, the rotatory capability of the
migrate periodically into North America. However, most of the car-
lower forelimb is retained, abduction-adduction of the upper fore-
nivoran record of the late Eocene and early Oligocene of North
limb is present; even the largest ursids do not significantly modify
America comes from the White River Group of the midcontinent.
this pattern. In the feet, the ursines are commonly plantigrade, but
Small arctoids of this age are extremely rare elsewhere in North
the hemicyonine ursids are exceptional in their digitigrade stance
America, and it should be kept in mind that arctoid diversity outside
(Figure 10.1 ). The more generalized shearing dentition of hemicy-
the North American midcontinent and in Asia remains largely un-
onines logically accompanies digitigrady, and the crushing teeth of
sampled, hence may be much richer than is currently believed. The
most ursines occurs with plantigrady. The postcranial skeleton of
known geographic distribution of late Eocene-Oligocene arctoids
the amphicynodonts was poorly known, but recent studies by Cirot
on the Holarctic landmasses may not be representative of the true
and Bonis ( 1992) have clarified its nature in Amphicynodon in which
diversity and distribution of species, but actually may more accu-
there is a semiplantigrade manus and plantigrade pes (Cirot, 1992).
rately reflect a strong sampling bias provided by the carnivore-rich
Five toes are present in fore and hind feet; lower limb bones are
fissures at Quercy and the regionally extensive, thick, fine-grained
never fused, always separate elements. Claws are nonretractile.
volcaniclastic sediments of the White River Group.
Pronounced sexual dimorphism in body size has been docu-
mented in a number of extinct and living ursids (Kurten, 1966, 1967;
Hendey, 1980) and should be suspected when assessing variation INFRAFAMILY
present in large samples ofursid fossils (Figure 10.2G). The degree
Recognition of the magnitude and diversity of the ursid radiation
of dimorphism increases with body size (Kurten, 1967), so it is not
(Figure 10.3) on the northern continents is a relatively recent event.
surprising that the largest living carnivorans were males belonging
Simpson (1945), who listed fourteen ursid genera in his classifica-
to the huge extinct Pleistocene ursid Arctodus (Figure 10.2H).
tion, included primarily the living ursines and their extinct relatives
in the family but omitted the amphicynodont and hemicyonine ur-
SYSTEMATICS sids. Moreover, most amphicynodonts and hemicyonines were clas-
sified as canids, not ursids, due to the practice at that time of using
the dentition as the principal index of relationships. With the recog-
SUPRAFAMILY
nition (Hunt, 197 4, 1977, 1987) that canids can be identified by their
Ursids are members of the arctoid Carnivora. This placement is sup- unique basicranial anatomy, it became evident that from their earli-
ported by basicranial (especially auditory bulla) structure, dental est appearance in the late Eocene until the late Miocene, the Canidae
traits, elaboration of the maxilloturbinal bones within the nasal cav- were restricted to North America. The supposed "canid" amphicyn-
ity, serum proteins such as hemoglobin, as well as features of the soft odonts and hemicyonines displayed not only basicranial and bulla
anatomy (absence of caecum and bulbourethral glands of Cowper). structure indicative of ursids, but also dental traits. It became clear
Arctoid carnivorans are first clearly identified in the late Eocene of for the first time that during the Oligocene and Miocene there had
North America (amphicynodontine ursids, amphicyonids, mustelid) been a major radiation of predatory ursids that paralleled canids
and Europe (amphicyonids) and are present in the early Oligocene in the attainment of digitigrady and shearing/crushing dentitions.
of both Eurasia and North America; they are diverse and well repre- These predaceous ursids occupied the canid niche in Eurasia, and
sented in the Quercy fissure fillings of France (Oligocene) - several many invaded North America (Figures 10.4, 10.5). Their failure to
post-Grand Coupure localities yield the oldest abundantly repre- radiate broadly in North America appears to be due to the early col-
sented arctoids (Remy et al., 1987). In the Quercy district we find onization and continued occupation of these niches by true canids.
a rich sampling of ancestral ursids, amphicyonids, mustelids and Three subfamilies of ursids (Amphicynodontinae, Hemicyoni-
procyonids, but not canids. The oldest canids occur only in the late nae, Ursinae) are recognized here (Figure 10.3). The Ursinae are fur-
Eocene (Chadronian) and early Oligocene (Orellan) of North Amer- ther divided into three tribes: Ursavini, Ursini, Tremarctini.
ica, and this geographic separation between the oldest arctoids and Tremarctini and Ursini include all living ursids (except the Giant
canids lends credence to the recognition of a basic arctoid-cynoid Panda Ailuropoda) and are the product of relatively recent Plio-
dichotomy among the early caniform carnivorans. It seems probable Pleistocene ursid radiations. A separate tribe, Ursavini, is suggested
that the initial arctoid-cynoid split is the result of spatial isolation for the Holarctic Miocene ursines: the large Indarctos and Agrio-
of the ancestral canid stock in the New World. In North Amer- therium and the small ancestral ursine Ursavus. The Hemicyoninae
ica, late Eocene and early Oligocene arctoids are present (Parictis, represents the major mid-Cenozoic (Oligo-Miocene) radiation of
Mustelavus, and a new amphicynodont genus) but are uncommon, predatory ursids that parallel canids in many aspects of their biology.
 Ursidae 177

Figure 10.3. Hypothesis of cladistic

relationships within North American
Tertiary ursid genera. Key to charac-
ters (scale bar in all illustrations is 5
cm): (I) URSIDAE: Type A auditory
PLITHOCYON TREMARCTOS bulla (Hunt, 1974); presence of ursid
8 18
upper molar pattern involving the

~
development and posterointernal
PLIONARCTOS migration of metaconule to position
17

~
posterior to protocone; no paraconule;
7 protocone-metaconule aligned to
form low prominent anteroposterior
crest that parallels similarly aligned
paracone-metacone crest; proto-
cone basin opens to the rear; small
transverse enamel ridge may join
metacone and metaconule (and join
AGRIOTHERIUM
paracone with protocone); premolars
12 small, low crowned, yet wide and
robust, surrounded by cingulum;
accessory cusps never or only very
rarely present on Pl/pl-P3/p3,
creating a very simple conical crown;
posterior accessory cusp occasionally
present in some species on p4;
URSAVUS upper carnassial with protocone
posteriorly displaced relative to other
families of Carnivora; P4 metastylar
blade never elongate; lower car-
nassial with talonid's labial margin
formed by hypoconid crest, and
lingual margin formed by crest with
entoconid-entoconulid cusps; m2
trigonid pattern diagnostic in many
PARICTIS ursid species in which metaconid-
protoconid of similar height are
anteriorly joined by a symmetrical
CLADOGRAM OF arcuate crest (which may or may not
NORTH AMERICAN TERTIARY be interrupted by a paraconid); m2
URSID GENERA metaconid-protoconid also connected
by straight transverse ridge. (2) HEMICYONINAE: Long footed (metacarpals and metatarsals elongate); digitigrade; principal
weight-bearing axis of foot passes through digits 3-4 (paraxonic). (3) Premasseteric fossa oflower jaw present. (4) Parictis only:
Small body size <2 kg; teeth low crowned; low simple conical premolars; Ml subtriangular, not quadrate, with rudimentary to
moderately developed metaconule; P4 metastylar blade short, protocone robust, lingually directed; ml trigonid low, with para-
conid little mesially advanced; if Campylocynodon is referable to Parictis, then basicranium without wide mastoid-exoccipital
shelf; auditory bulla Type A, slightly inflated; ml lengths ~6- I l mm. (5) Small to medium body size 2 to 50 kg; carnassial pair of
sectorial type; simple conical premolars; MI subtriangular to subquadrate, superficially canidlike; M2 quadrate to subquadrate;
ml paraconid advanced to form shearing carnassial blade; snout short; auditory bulla somewhat inflated; ml lengths, Old World,
~ 12-23 mm; New World, ~23-31 mm. (6) Medium body size 25 to 50 kg; carnassial pair highly sectorial; simple but tall
compressed premolars; upper molars unknown but presumably subquadrate; ml paraconid strongly advanced to form shearing
carnassial blade; snout not shortened; auditory bulla unknown; ml lengths ~27-35 mm. (7) M2 short relative to Ml length (in
Old World Hemicyon, M2 is longer); ratio of m2 width over ml width is < or = 1 (in Old World Hemicyon it is >I); M2 is
smaller in absolute size relative to M2 in Hemicyon; the lower carnassial is wide to very wide relative to its length (in Old World
Hemicyon it is narrow relative to length). (8) Medium body size 25 to 60 kg; carnassial pair sectorial but with initial development
of crushing features (widening of ml, aligned sloping ml crest formed by protoconid-metaconid-entoconid-entoconulid
minimizing shear of ml trigonid's posterior slope); premolars small but not strongly reduced; upper molars quadrate; skull
without strongly reduced snout; Type A auditory bulla not inflated; ml lengths ~25-35 mm. (9) New hemicyonine genus:
Medium to large body size 30 to 75 kg; carnassial pair sectorial (proto-meta-entoconid-entoconulid crest of ml absent); widest
ml relative to length of any known hemicyonine; ml talonid without entoconid or entoconulid (present in Plithocyon); palate
wide with relatively small teeth; premolars small; upper molars quadrate; skull without strongly reduced snout; Type A auditory
bulla not inflated; ml lengths ~29-38 mm. (10) URSINAE: M2 develops posterior talon lengthening the tooth. (11) Large body
size (60 to > 100 kg) in conjunction with slightly posteriorly extended M2 talon (distinguishing lndarctos from Agriotherium);
no premasseteric fossa; prominent ml trigonid coupled with large ml talonid and large m2; teeth low crowned (in con-
trast to higher crowned agriothere teeth); premolars reduced, especially in terminal species, but usually all present; long skulled in
 Robert M. Hunt, Jr. 178

Figure 10.3 (Cont.). comparison to the short robust skull of Agriotherium; probably short footed relative to agriotheres; ml lengths ~30--44 mm. (12)
Large body size(> 100 kg) in conjunction with quadrate Ml-2, and no or only rudimentary development of M2 talon; short wide robust rostrum; prominent
P4 parastylar cusp; anterior premolars single rooted and strongly reduced; p2 usually lost; Type A auditory bulla present; prominent distal placement of
volar carpometacarpal and tarsometatarsal ligament scars on shafts of MC4 and MT4; ml lengths ~34-51 mm. (13) Small to medium body size (2 to 60
kg); premolars reduced in size relative to molars; upper carnassial (with short metastylar blade) reduced in size relative to Ml-2; elongation of m2-3;
progressive elongation within the lineage of M2 talon (Crusafont-Pair6 and Kurten, 1976); no premasseteric fossa; basicranium and auditory bulla structure
unknown; most species probably short footed and plantigrade; ml lengths~ 12-25 mm. (14) No premasseteric fossa in ancestral stock; short footed and
primitively plantigrade (principal weight-bearing axis not between digits 3--4); tendency to elongate upper toothrow by strong posterior extension of M2
talon; reduction of ml trigonid relative to talonid; maintenance and enlargement ofm2-3 to form lengthened crushing surface for occlusion with lengthened
Ml-2; reduction of premolars. (15) Medium to large body size (50 to >100 kg); no premasseteric fossa; snout not greatly shortened as in Tremarctini;
skull narrower and more elongate than in Tremarctini; early species possess high ml trigonid and single metaconid, whereas later derived species have
low ml trigonid and twinned metaconid; m2-3 longer relative to ml than in Tremarctini (Ursus reduces ml relative to m2-3 length); premolars often lost;
trend toward emphasis of outer metapodials of the feet and development of toe-in plantigrade gait; ml lengths ~20-30 mm. ( 16) Snout strongly shortened;
premolars crowded together and reduced; characteristic pattern of enamel tuberculation appears in m2 talonid basin; toe-in gait characteristic of Ursini not
present. (17) Medium body size (about 60-80 kg); no premasseteric fossa; ml lengths ~21-24 mm. (18) Medium to large body size (about 60 to >100 kg);
premasseteric fossa develops; ml lengths~ 17-24 mm.

Amphicynodonts are the ances.tral ursid stock from which ultimately Bonis, 1992). North American amphicynodonts (Parictis, and a see-
the other subfamilies have evolved; hence Amphicynodontinae is a ond endemic genus; see below) are not identified in Eurasia; they
paraphyletic category retained for convenience until the members occur only in the White River Group of the central Great Plains
of the group are better known. and northern Rocky Mountains, and in the John Day beds of Ore-
gon. In Asia, the only known amphicynodonts are Amphicynodon
and the small endemic Amphicticeps from the Oligocene Hsanda
INCLUDED NORTH AMERICAN GENERA IN THE Go! Formation; the latter genus seems to be closely related to the
FAMILY URSIDAE rare North American early Miocene amphicynodonts Allocyon and
Kolponomos from the Pacific Northwest. Thus, although Oligocene
The locality numbers listed for each genus refer to the localities in amphicynodonts are Holarctic in distribution, the genera are en-
Appendix I (for a more detailed listing of localities, see the Key demic to Eurasia and North America: no coeval migration between
to Fossil Localities at the end of this chapter). The acronyms for Old and New World is evident.
museum collections are listed in Appendix III. Because the species-
level taxonomy of ursids is in revision at the present time, localities Parictis Scott, 1893
usually are given for genera rather than species. Type species: Parictis primaevus Scott, 1893.
Type specimen: PU 10583.
Characteristics: Small amphicynodont carnivorans with skull
AMPHICYNODONTINAE
lengths of about 6-8 cm. Dental formula I3/3, Cl/l, P4/4,
Characteristics: Small arctoid carnivorans with Type A auditory M2/3. Premolars without anterior accessory cusps, and
bulla formed chiefly by the ectotympanic; skull dolichocephalic with no or very tiny posterior accessory cusps on Pl-3/
(postorbital distance long), snout relatively short. Dental formula pl-3; posterior accessory cusp occurs on p4; cingulum
I3/3, Cl/l, P4/4, M2/3. Ml triangular to subquadrate with three encircles base of all premolars. In post-Chadronian fos-
principal cusps (protocone, paracone, metacone) and variable de- sils, premolars tend to be robust, transversely widened.
velopment of the metaconule resulting in a more quadrate tooth; The ml trigonid cusps low, and in occlusal view nearly
tendency for Ml metaconule to be incorporated into postprotocrista; at the apices of equilateral triangle (there is little or no
postprotocristajoins protocone and metaconule to form a posteroex- advancement of the paraconid to form a shearing blade
ternally directed ridge that intersects the posterior margin of Ml at in contrast to the canid Hesperocyon where the paraconid
the base of the metacone (the primitive condition, hence a more tri- has shifted anteriorly); talonid basin opens to rear, bor-
angular Ml), or forms an anteroposterior ridge which intersects the dered labially by hypoconid ridge and lingually by ento-
posterior margin of Ml some distance internal to the metacone (a conid ridge, the latter often slightly creased to form two
subquadrate Ml, hence a derived condition); protocone basin in small close-spaced cusps. The m2 length formed by long
more derived species opens to rear and is not closed by postpro- shallowly basined talonid, with low trigonid compressed
tocrista; no strongly developed Ml parastyle. M2 triangular to sub- at anterior end of tooth. Ml triangular with slightly devel-
quadrate with metaconule developed as in Ml. P4 a shearing tooth oped metaconule; postprotocrista intersects posterior mar-
with retracted or unretracted protocone and short metastylar blade. gin of tooth at base of metacone. P4 short, robust, with
P4 protocone robust (more so than in canids), protruding strongly unretracted but prominent protocone and short metastylar
linguad. The m 1 trigonid low. Premolars (except p4) lack developed blade, and surrounded by cingulum. M2 a reduced ver-
posterior accessory cusps, but are surrounded by distinct cingula. sion of M 1. Postcranials not yet described. Arctoid single-
The amphicynodonts are the stem group from which all other chambered auditory bulla.
ursids evolve. The core genera in the Old World are Amphicynodon Lower camassial length range: P. parvus, 6.4 mm; P. pri-
and Pachycynodon, both known from the Quercy fissures (Cirot and maevus, 6.8 mm (estimated from alveolus); P. gilpini,
Ursidae 179

8.0-8.5 mm; P. montanus, 7.6-7.9 mm; P. dakotensis, A rare small ?amphicynodont is known from the White
8.5-9.4 mm; P. major, 10.8 mm. Total range for Paric- River Group of the central Great Plains, based upon a
tis, 6.4-10.8 mm (N = 19). partial maxilla with P4-Ml (FMNH P27157, base of the
Included species (see Clark and Guensburg, 1972, Baskin Orella Member, Brule Formation, near Roundtop, 7 miles
and Tedford, 1996): Parictis parvus, P, gilpini, P. dakoten- N of Crawford, Dawes County, Nebraska). The specimen
sis, P. montanus, P, primaevus, P. major, P. ?personi. has been variously referred to Parictis (Clark and Guens-
Parictis is known from the following localities: Chad- burg, 1972) or to Drassonax (Baskin and Tedford, 1996).
ronian (CP39B, CP39C, CP39F, CP39IIC, CP41A, CP83C, Its Ml crown pattern is more derived than that of Parictis
CP98BB, NP24C, NP24D, NP29C); Orellan and foreshadows Ml of later ursids. The ursid character
(CP41B, CP68C, CP84A, CP99A, PN6); ?Whitneyan of Ml is evident, based upon the alignment of the pre-
(?NP50C). and postprotocristae, the developed metaconule posterior
to the protocone, the development of a hypocone from the
Comments: The teeth of the common North American am- posterolingual cingulum, and two small transverse ridges
phicynodont Parictis can be clearly distinguished from the in the protocone basin, an anterior one running from the
contemporaneous canid Hesperocyon by the form of its base of the paracone to the protocone, a posterior one run-
premolars and features of the molars. As interpreted here, ning from the base of the metacone to the base of the notch
Parictis includes a group of species in which rather slender in the postprotocrista between protocone and metaconule.
amphicynodont premolars progressively evolve into wide The P4 protocone is retracted and the tooth is surrounded
robust teeth, an important synapomorphy of the lineage. by a developed cingulum. I am in accord with Baskin and
The holotype species of Parictis, P. primaevus from the Tedford, who separate this taxon from Parictis, recogniz-
John Day beds of Oregon, has such robust broad amph- ing it as a distinct genus of early arctoid; however, its affin-
icynodont premolars, and consequently is regarded as an ity with Drassonax remains to be demonstrated.
evolved species of the genus. Baskin and Tedford (1996)
believe that the preserved features of the on! y known spec-
Allocyon Merriam, 1930
imen of P. primaevus (a partial mandible, PU 10583, age
Type species: Allocyon loganensis Merriam, 1930.
uncertain, but probably Chadronian to early Arikareean)
Type specimen: UCMP 24106.
are insufficient to provide diagnostic generic characters,
Characteristics: Mid-sized amphicynodont camivoran with
and suggest Parictis be considered a nomen dubium. They
basilar skull length of about 18 cm. Dental formula in the
refer the large Great Plains-northern Rocky Mountain hy-
only known individual is I3/?3, Cl/l, P4/4, M2/3. Ante-
podigm placed in Parictis by Clark and Guensburg (1972)
rior premolars broken, but p4 and P3 indicate they were of
to Subparictis, and create a family Subparictidae for this
simple, unspecialized amphicynodont type. P4 short with
material. The robust subequal premolars of the holotype of
retracted protocone. Ml-2 worn, thus cusp placement ob-
Parictis, however, are adequate to identify the lineage. Al-
scure, but triangular Ml form and smaller elliptical M2
though the locality and age of the holotype are uncertain,
with slightly expanded posterior margin are suggestive of
a latest Eocene to late Oligocene age is in keeping with
EuropeanAmphicynodon. The p4 simple, without posterior
the advanced stage of development of its premolars. Thus
accessory cusp, and with developed cingulum as in other
the genus Parictis is retained and comprises six species, of
amphicynodonts. Them 1 with low trigonid; paraconid not
which P. primaevus is considered at present to be restricted
anteriorly advanced as a cutting blade; talonid well de-
to the John Day beds, represented only by the holotype
veloped but wear obscures cusp pattern. The m2-3 present
mandible. I place the North American endemic amphicyn-
and developed. Snout short, postorbital distance long, as in
odonts as genera within the Holarctic subfamily Amphi-
amphicynodonts in general. Choanal canal floored by bone
cynodontinae, rather than employing a new family for the
for nearly 2 cm posterior to M2. Basicranium with wide ba-
New World forms.
sioccipital, and greatly expanded mastoid-exoccipital re-
The problematic skull described by Chaffee (1954) as
gion with enlarged mastoid and paroccipital processes. Au-
Campylocynodon, and considered by Clark and Guensburg
ditory bulla a simple arctoid ectotympanic chamber very
(1972) as referable to Parictis, may be an early amphicyn-
similar in form to the bulla of Amphicynodon. Postcranials
odont as suggested by Baskin and Tedford (1996), but the
unknown.
poor preservation of the teeth and damaged basicranium
Lower carnassial length range: A. loganensis, 14.5 mm.
remain extremely difficult to interpret. Although the form
Included species: A. loganensis only, known from locality
of the petrosal and remnants of the auditory bullae sug-
PN6C only.
gest arctoid affinities, the skull lacks the broad mastoid-
exoccipital shelf seen in Eurasian amphicynodont crania. Comments: The only known specimen of Allocyon is an asso-
We await discovery of a well-preserved cranium of Paric- ciated skull and right lower jaw collected in 1920 from the
tis, associated with lower teeth typical of the species from Logan Butte region of Oregon and described by Merriam a
the White River Group, in order to determine its diagnostic decade later. The skull form, structure of P4-M2 and ml-
basicranial structure. 3, the simple premolars, and the basicranial anatomy are
 Robert M. Hunt, Jr. 180

clearly amphicynodont in pattern. The highly specialized HEMICYONINAE

mastoid-ex occipital region, however, is not seen in any Eu-
Characteristics: Small to large arctoid camivorans with Type A bulla
ropean amphicynodont, and in Asia is observed only in the
formed chiefly by ectotympanic; skulls brachycephalic and dolicho-
much smaller skull of the rare Amphicticeps from Hsanda
cephalic, snout short. Dental formula 13/3, Cl/l, P4/4, M2/3. Ml
Go!, Mongolia (Matthew and Granger, 1924).' Allocyon
quadrate with principal cusps protocone, metacone, paracone, and
is plausibly descended from an amphicynodont species
metaconule; postprotocrista unites protocone and metaconule as an
near Amphicticeps, which in tum would be derived from
anteroposterior ridge that intersects the posterior border of Ml, re-
Eurasian Amphicynodon (which coexists with Amphicti-
sulting in a posteriorly open protocone basin; in Cephalogale, vari-
ceps at Hsanda Gol). Current age estimates would place
able development of metaconule results in subquadrate to quadrate
Amphicynodon and Amphicticeps in the early Oligocene,
Ml. P4 a well-developed shearing tooth with retracted protocone
and Allocyon in the late Oligocene (~28.5-29.5 Ma).
and short metastylar blade. The ml trigonid low to high. The prin-
Slight size reduction of the posterior molars in Amphic-
cipal derivative feature of the group is a digitigrade stance in which
ticeps, however, may indicate that the Hsanda Go! species,
both metatarsals and metacarpals are elongate; the major weight-
A. shackelfordi, is not itself in the direct line of descent.
bearing axis passes between digits 3 and 4; hence the foot is function-
ally paraxonic. Digitigrady apparently evolved once in the ancestral
Kolponomos Stirton, 1960 hemicyonine and was an important adaptive feature contributing to
Type species: Kolponomos clallamensis Stirton, 1960. this Miocene Holarctic radiation of cursorial predaceous ursids.
Type specimen: UCMP 50056. The Miocene hemicyonine radiation (Hemicyon, Plithocyon,
Characteristics (in part from Tedford, Barnes, and Ray, 1994): Phoberocyon, Dinocyon, Cephalogale, and a new undescribed
Large aquatically adapted amphicynodont derivatives in genus) includes the first ursids to attain moderate to large body
which the cheek teeth show strongly inflated principal size. Hemicyonines are found only in Eurasia and North America,
cusps, P4 molariform with large protocone, Ml with en- with a single tooth reported from the early Miocene of east Africa
larged conules; P4-M2andm1-2 expanded to large crush- (Schmidt-Kittler, 1987). They share a common basicranial structure,
ing platforms; M3/m3 absent; basilar length of skull 25-26 including the ursid Type A bulla, development of the ursid molar
cm; rostrum with marked downward flexure and forwardly and camassial pattern, and digitigrady. Hemicyonines periodically
extended premaxilla-maxilla resulting in a deeply concave invade North America in the Miocene but are extinct worldwide
anterior palatal region; enlarged infraorbital foramen; ba- by the end of the epoch (Figure 10.4 ). They are distinguished from
sioccipital very wide; a rudimentary arctoid auditory bulla the ursine bears by absence of posterior elongation of M2, main-
with both ectotympanic and entotympanic elements; an tenance of large functional camassials, and paraxonic digitigrade
enormously hypertrophied mastoid process. A few postcra- foot emphasizing digits 3--4. All hemicyonines except Cephalogale
nial bones of K. newportensis (including vertebrae, steme- possess a premasseteric fossa of the lower jaw. That the premasse-
bra, hyoid elements, partial rib) accompany a partial meta- teric fossa has evolved independently in ursids at least three times
podial and a phalanx that suggest terrestrial locomotion is demonstrated by (1) its presence in the ursines Tremarctos and
was possible in this species, but modification of the foot Arctodus, although absent in the ancestral ursine Ursavus; (2) its
had probably not progressed to the degree seen in occurrence in all derivative hemicyonines, despite its absence in the
pinnipeds. ancestral hemicyonine Cephalogale; and (3) its presence in the ur-
Lower camassial length range: K. newportensis, 27.3-27.5 sine Agriotherium, which is derived from Ursavus separately from
mm. the Tremarctini.
Included species: K. clallamensis, K. newportensis. Kolpon-
omos is known from localities WM311 and WM6.
Cephalogale Jourdan, 1862
Comments: Two species of these large arctoids have been Type species: Cephalogale geoffroyi Filhol, 1879.
found in the Pacific Northwest. Kolponomos clallamensis Type specimen: Upper and lower jaws, Museum of Lyon,
was initially based upon a rostrum (UCMP 50056) from Viret, 1929, Figure 18.
Slip Point, Clallam Bay, Washington, found in 1957. Later, Characteristics (modified from Beaumont, 1965; Bonis,
in 1988, a nearly complete cranium of this species was 1973): Small to mid-sized digitigrade hemicyonine car-
discovered at the same locality. These specimens are from nivorans with skull lengths of about 10-20 cm. Larger
the early Miocene Clallam Formation. The holotype and body size than amphicynodonts Parictis and Amphicyn-
only known specimen of K. newportensis (USNM 215070) odon, but the smallest hemicyonine genus. Ursid basicra-
comprises a cranium, mandible, and a few postcranial el- nium with Type A bulla, well ossified, completely enclos-
ements, collected from 1969 to 1976, in the vicinity of ing middle ear, with inflation of medial part ofbulla in some
Newport, Lincoln County, Oregon. It comes from the lower species. Skull dolichocephalic, like that of Amphicynodon
part of the Nye Mudstone of early Miocene age. Both the but larger in size; rostrum short, postorbital portion of skull
Oregon and Washington occurrences are believed to corre- long. Some European Aquitanian and North American
late with the late Arikareean Land Mammal age (Tedford, Hemingfordian species with broad, short-snouted, brachy-
Barnes, and Ray, 1994). cephalic skulls. Dental formula 13/3, Cl/l, P4/4, M2/3.
 Ursidae 181

Premolars laterally compressed and bladelike (C. minor) wolf-sized predators dentally near Hemicyon. It retains
to somewhat robust and transversely widened (C. ursi- primitive features of the lower teeth, coupled with a large
nus); posterior accessory cusp not present on pl, variable shearing lower camassial. Them I has an anteriorly placed
on p2-3, present on p4. No posterior accessory cusps on paraconid, a reduced retracted metaconid, with protoconid-
Pl-3. The ml with trigonid cusps more widely separated paraconid forming an effective shearing blade. The ml
in occlusal view than in primitive amphicynodonts, para- trigonid is tall, the talonid low with a strong hypoconid
conid anteriorly advanced, and together with protoconid and a low entoconid ridge, much like the pattern found in
forming an effective shearing blade; metaconid reduced Cephalogale. The p4 is preserved only in the North Amer-
and posteriorly retracted. The ml talonid with prominent ican species: It is a well-developed tooth, whereas pl-3
hypoconid ridge or crest but only low entoconid swelling are small and low. Thus there is an abrupt decrease in size
(not cuspate), the latter continuing around the posterior between p3 and p4. The m2 reverses the condition seen
margin of the talonid,joining the former to close the talonid in Cephalogale in having the protoconid more elevated
basin. The m2 with metaconid prominent, higher than pro- than metaconid. Elongate metapodials and other foot bones
toconid; paraconid strong to weakly developed, never cus- from Thomas Farm attributable to North American Phobe-
pate but can form a thickening of the ridge anterior to rocyon demonstrate that it is long-footed and digitigrade.
the protoconid. P4 with relatively short metastylar blade; Known specimens indicate an early hemicyonine branch in
P4 protocone more anteriorly placed in primitive species the Burdigalian and Hemingfordian of Europe and North
but retracted in advanced forms. Upper molars quadrate to America that manifests a shearing dentition, digitigrady,
subquadrate, with metaconule posterior to protocone, no with slightly reduced anterior premolars; it is the largest
paraconule, and posterointemal cingulum with hypoconal hemicyonine to evolve in the early Miocene, rivaled only
swelling. Metaconule (often only weakly cuspate) united by the largest species of North American Cephalogale.
with protocone by ridge; protocone basin opens to rear of Lower camassial length range: P. aurelianensis, 35 mm;
tooth, not closed by postprotocrista. Ml always wider than P. huerzeleri, 27-28 mm; P. johnhenryi, 33.2-33.6 mm.
long. M2 smaller than Ml. Teeth generally surrounded by Included species: Europe: P. aurelianensis, huerzeleri; North
developed cingula. America: P. johnhenryi, known from locality GC8D
Lower camassial length range: Europe: about 12-23 mm only.
(Bonis, 1973, p. 86); North America: 23.4-30.7 mm
(N = 18, Bridgeport Quarries and Runningwater Forma- Plithocyon Ginsburg, 1955
tion). Type species: Plithocyon armagnacensis Ginsburg, 1955.
Included species: There are about eight Old World species Type specimen: Paris Museum 244, 245, 246 (Ginsburg,
(Beaumont, 1965; Bonis, 1973). New World species are 1961, p. 60; also Ginsburg, 1980).
not presently described. Three North American species Characteristics: Small to large Barstovian hemicyonines,
differing in body size are known from New Mexico, Wyo- strong sexual dimorphism, skull dolichocephalic, basilar
ming, and Nebraska. Cephalogale is known from the fol- lengths ~15-32 cm. Ursid basicranium and Type A au-
lowing localities: late Arikareean (SB29A, CP51A, ditory bulla. Dental formula 13/3, Cl/l, P4/4, M2/3. In-
CP104A); early Hemingfordian (CP105, CP106). ternal margin of ml talonid divided into entoconid and
entoconulid; m2 protoconid and metaconid same height;
Phoberocyon Ginsburg, 1955 Ml-2 quadrate without developed talons; ml metaconid
Type species: Phoberocyon aurelianensis (Frick, 1926). not as developed as in Hemicyon and not separated from the
Type specimen: Paris Museum 228 (Ginsburg, 1955, p. 90, protoconid slope; protoconid-metaconid-entoconid-entoc-
Figure 2), ml-2. onulid aligned along a smooth regular curve; talonid basi-
Characteristics: Phoberocyon was originally described from ned, with strong groove enclosed between the hypoconid
ml-2 and some additional fragments from the Burdigalian and the two well-developed internal talonid cusps.
Sables d'Orleanais, France (Ginsburg, 1955). Ginsburg Digitigrade.
also referred a lower jaw to this genus but to a new species Lower camassial length range: P. barstowensis, 27.3-34.2
(P. huerzeleri) from the Burdigalian of Burgdorf, Switzer- mm (N = 18); P. sp. I, 25.7-25.9 mm (N = 2, Steepside
land. The genus was diagnosed by Ginsburg (1955): "Car- Quarry, California); P. sp. II, 28.2 mm (Skull Ridge, New
nivore near Hemicyon. The p3--4 with posterior accessory Mexico); P. ursinus, ~33-35 mm (N = 2). Total range for
cusps. The ml long, slender as in Hemicyon, butthe meta- North American Plithocyon, 25.7-35 mm (N = 23).
conid somewhat more reduced; posterior slope of meta- Included species: Europe: P. armagnacensis, P. statzlingii,
conid and protoconid aligned in the same plane; talonid P. bruneti; North America: P. barstowensis, P. sp. (unde-
very low, anteriorly inclined. The m2 of same width as m I, scribed, New Mexico and California), P. ursinus. North
massive, low; trigonid cusps strong, paraconid well devel- American Plithocyon is known from the following local-
oped; protoconid higher than metaconid; talonid as in ml." ities: early Barstovian (NB6C, SB32B); medial Barsto-
Few fossils of Phoberocyon have been found; however, vian (NB6E); late Barstovian (SB32D, CP113A, CPI 14C);
the taxon seems to represent Burdigalian-Hemingfordian Barstovian (?CP55).
 Robert M. Hunt, Jr. 182

Comments: Survey of Eurasian and New World hemicyo- relative to M2 length; M2 length to width. However, in two
nine dental proportions indicates that Plithocyon is fur- traits, the new Clarendonian genus is distinct from Plitho-
ther characterized by (I) a relatively shorter M2 relative to cyon: (1) ml length versus width, and (2) morphology of
Ml length than true Hemicyon (Ml-2 unknown in Phobe- the lower teeth, particularly the lower camassial. The ml
rocyon; however, the new hemicyonine genus from the of the new genus lacks the derived morphology seen in
central Great Plains is similar to Plithocyon in this trait). Plithocyon where, as Ginsburg has noted, Plithocyon is
(2) A ratio of m2 width/m 1 width of 1.0 or < 1. In characterized by a smoothly curved metaconid-entoconid-
Hemicyon, the ratio is > 1. The new genus plots along entoconulid crest. This crest is absent in the new Clarendo-
the Plithocyon regression axis but includes animals mostly nian genus in which ml shows a steep vertical surface be-
larger in absolute size. Phoberocyon specimens, which tween metaconid and the inner talonid margin. In this, the
are very rare, are intermediate between Plithocyon and Great Plains Clarendonian hemicyonine strongly contrasts
Hemicyon in this trait. (3) An M2 absolutely smaller in with the Plithocyon sample from the Barstow syncline
length and width than Hemicyon; M2 in Hemicyon is large. whose teeth show the defining features attributed by Gins-
The new genus has an M2 most like large Plithocyon burg to Plithocyon: the metaconid-entoconid-entoconulid
in absolute size, yet despite the considerable size of sev- crest of ml, the short M2 length relative to Ml length, and
eral skulls of this new taxon, M2 is relatively small when an m2 width no greater than m 1 width. This new genus is
compared to Hemicyon of similar skull size. (4) Lower unknown in the Old World.
camassials are intermediate in length/width ratios between
Hemicyon and the new genus. Hemicyon has the narrowest
ml relative to length; Plithocyon is intermediate; and the URSINAE
new genus has the widest ml relative to length. Phobe- Characteristics: Miocene and younger ursids of small to very large
rocyon overlaps both Hemicyon and Plithocyon in this body size, distinguished from hemicyonines by the development of
feature. a crushing hypocarnivorous dentition (in contrast to shearing hy-
percarnivorous hemicyonines) in which the shearing ability of the
New genus camassial pair is at first diminished and then lost (least evident in
Jndarctos andAgriotherium). This trend in camassial reduction ac-
Type species: Hemicyon barbouri Colbert, 1941.
companies the lengthening of the tuberculate molars, especially the
Type specimen: UNSM 2602.
development of a prominent talon on M2 (rudimentary in Agrio-
Characteristics: Clarendonian hemicyonines of large size and
therium, Hendey, 1980). Premolars remain small and can be re-
probable strong sexual dimorphism; skull dolichocephalic
duced and lost. The subfamily is divided into three tribes, each
(to 40 cm length) with wide palate and relatively small
independently attaining large body size: (I) Ursavini (the ursine
teeth. Ursid basicranium and Type A auditory bulla. Den-
ancestral stock Ursavus, and two independent derivative lineages,
tal formula I3/3, Cl/I, P4/4, M2/3. The ml lacks diag-
Jndarctos and Agriotherium); (2) Ursini (radiation of Ursus); and
nostic descending cusp series (metaconid, entoconid, ento-
(3) Tremarctini (Plionarctos-Tremarctos-arctodont group). Because
conulid) seen in Plithocyon. Them 1 metaconid a small but
the ancestral ursine Ursavus is placed in Ursavini, the tribe is a pa-
distinct cusp, its posterior surface descending nearly ver-
raphyletic stem group, functioning in this respect just as Amphicyn-
tically to talonid surface. Inner edge of m 1 talonid without
entoconid or entoconulid. Hypoconid only prominent ml odontinae within Ursidae.
Ursines are also distinguished from hemicyonines by foot struc-
talonid cusp, situated along labial edge oftalonid. Width of
ture. Both short-footed (most Ursavus and Ursus, large Indarctos,
m 1 relative to length greater in this tax on than in Plithocyon
Tremarctos) and long-footed (Agriotherium, Arctodus, some Ursus)
and Hemicyon. Premolars unreduced and without posterior
ursines retain a foot structure indicative of plantigrady; none appear
accessory cusps, distinguishing this taxon from Phobero-
cyon. The p2 prominent, double rooted, and part of a graded to have attained a digitigrade stance.
series from pl-4; p2 does not show reduction as in North
American Plithocyon. Digitigrade. URSAVINI
Lower camassial length range: 29.5-37.4 mm (N = 7).
Characteristics: Ursavus, the ancestral ursine, together with Indarc-
Included species: "Hemicyon" barbouri (Colbert, 1941 ),
tos and Agriotherium, comprise a Miocene Holarctic radiation of
known from the following Clarendonian localities: CP90B,
nearly twenty species, identified by the first appearance of dental
CP113B, ?CP114C, CP114D, CP116A, CP116AA.
traits diagnostic of the later radiation of ursine bears. These traits in-
Comments: The Clarendonian Great Plains new hemicyonine clude small simple anterior premolars, reduced in some species; size
genus differs from true Old World Hemicyon in having an and shearing function of P4 not as reduced as in Ursini-Tremarctini;
anteroposteriorly shorter M2 relative to Ml length, a some- M2 quadrate but not lengthened as in Ursini-Tremarctini; first ap-
what smaller M2, and a relatively greater m 1 width in com- pearance ofM2 talon; M2 talon the same length as, or shorter than,
parison to width of m2, and especially, a wider ml relative the distance from the anteroextemal comer of M2 to the metacone
tom I length. The Clarendonian Great Plains genus is indis- (i.e., heel of M2 shorter than in Ursini-Tremarctini); no premasse-
tinguishable in certain features from Plithocyon: MI length teric fossa (except in Agriotherium). Plantigrade.
 Ursidae 183

Ursavus Schlosser, 1899 Ma) mid-sized to large ursids, sexually dimorphic

Type species: Ursavus brevirhinus (Hofmann, 1887). (Crusafont-Pair6 and Kurten, 1976). Ursid basicranium
Type specimen: Partial lower jaw with p3-m3, and canine, and Type A bulla. Skull dolichocephalic, length 30-42
from Steiermark (Frick, 1926, p. 99). cm; snout short. Dental formula I3/3, Cl/I, P4/4, M2/3.
Characteristics: Early to late Miocene small to mid-sized ur- Pl/pl-P3/p3 well developed in primitive species(/. vireti
sids (Eurasia: early Burdigalian to Turolian, MN 3-12, has double-rooted P2/p2; other species single-rooted
~7-20 Ma; North America: early Arikareean to late Barst- P2/p2), and reduced in size but nearly always present as
ovian, ?29 Ma-~ 12 Ma). Basicranium not described. Short small single-rooted teeth in advanced species, in contrast
snouted, with skull lengths of ~ 10-22 cm; small sam- to Agriotherium where anterior premolars often lost. P4
ples prevent determination of sexual dimorphism. Dental robust, with parastylar cusp present but usually not as
formula I3/3, Cl/l, P4/4, M2/3. Premolars small, simple. developed as in Agriotherium. Molars low crowned. M2
Pl/pl to P3/p3 without posterior accessory cusps; p4 pos- with elongate talon (more developed than inAgriotherium)
terior accessory cusp present in some species, absent in paralleling Ursini-Tremarctini. In contrast to Ursini-Trem-
others. P4 tending to reduction relative to Ml in most arctini, Indarctos maintains large carnassial teeth as well as
species (e.g., P4 length to Ml length ratio equal to or <1 robust crushing molars. Protoconid-metaconid-entoconid-
in U. elmensis, U. primaevus, U. brevirhinus, U. orien- entoconulid of ml aligned in smooth descending curve.
talis), but no reduction in some (ratio > 1 in U. depereti Premasseteric fossa absent. Anterior lower jaw tapers to
and some U. primaevus). Molars low crowned. Upper mo- front, and not as squared off and blunt as in Agriotherium;
lars quadrate, with wrinkled enamel. The ml paraconid "chin" present only in old individuals. Plantigrade.
advanced, with metaconid-entoconid-entoconulid present, Lower camassial length range:/. vireti, 30-34.5 mm (N = 6);
forming a line of cusps on posterointemal margin of tooth. I. arctoides, 30.8-35.7mm(N = 3);/. atticus,41.6-44m m
The m2-3 slightly elongate relative to lower camassial (N = 3); I. anthracitis, 33.2 mm (N = l); Siwalik species,
length; progressive elongation of M2 talon (Crusafont- 39.5 mm (N = l); I. oregonensis, 40.3-40.5 mm (N = 2);
Pair6 and Kurten, 1976, Table 11). No premasseteric fossa. I. lagrelii, 37.5 mm (N = l); I. sinensis, 42.8 mm (N = 1).
Associated postcranial skeleton known only in dentally Total range of Jndarctos, 30-44 mm (N = 18).
aberrant, long-footed U. orientalis (eastern China, Qiu Included species: Eurasia: I. vireti, I. arctoides, I. atticus, I.
et al., 1985), which is probably plantigrade and arboreal; anthracitis; I. lagrelii and/. sinensis from north China and
all other Ursavus species probably plantigrade because re- the Siwalik species are all probably referable to I. atti-
ferred metapodials from Steinheim and Goriach are like cus. North America: /. oregonensis. North American ln-
those of the living Sun Bear Helarctos malayanus (Heiz- darctos is known from the following localities: medial
mann; 1973, p. 46). Hemphillian (GC12II, CC36, NB27B, SP3A, CP116D,
Lower camassial length range: Eurasia: U. elmensis, 12.6- PN12, PN13).
14.3 mm (N = 13); U. brevirhinus, 15-18 mm (N = 9); U.
primaevus, 18.4-21.2 mm (N = 6); U. depereti, 23.5 mm Comments: Indarctos is known from Eurasia and North
(N = !); U. orientalis, 12.2 mm (N = l); U. intermedius, America. Its evolution can be traced from small European
14.2-15 mm (N = 4); North America: very few known: species such as /. vireti to large Holarctic I. atticus and
U. pawniensis, 22-24.6 mm (N = 2); U. cf. brevirhinus, I. oregonensis. In the Old World, the Turolian Hipparion
14.3 mm (N = 1); U. cf. primaevus, 18.6 mm (N = 1). faunas seem to contain a single terminal species, I. atticus,
Total range for Ursavus, 12.2-24.6 mm (N = 38). found from Pikermi and Samos east to the Siwalik Hills
Included species: Eurasia: U. elmensis, U. brevirhinus, U. and north China. The more primitive Vallesian species, I.
primaevus, U. depereti, U. ehrenbergi, U. orientalis, U. in- vireti and I. arctoides, are known only from Europe and
termedius; North America: U. pawniensis, new species I, are the oldest known lndarctos. There is probably only a
new species II, U. cf. brevirhinus, U. cf. primaevus. North single New World species, I. oregonensis.
American Ursavus is known from the following localities:
early Arikareean (PN6C); early Hemingfordian (CP88); Agriotherium Wagner, 1837
medial or late Hemingfordian (NB 17); early or medial Type species: Agriotherium sivalensis (Cautley and Falconer,
Barstovian (PN8B, PN9A); medial Barstovian (NB24, 1836).
NPI l); late Barstovian (NB19B, NB20B, NB25A, CP76, Type specimen: British Museum M39721, a skull.
CP113A). Characteristics: Late Hemphillian ursines of large size, sex-
ually dimorphic (Hendey, 1977). Ursid basicranium and
Indarctos Pilgrim, 1913 Type A bulla. Short-snouted robust skull, somewhat brach-
Type species: Indarctos salmontanus Pilgrim, 1913. ycephalic, basilar, length about 40-45 cm. Dental formula
Type specimen: Geological Survey oflndia, Calcutta D158, I3/3, Cl/l, P3-4/2-4, M2/3. Palate wide, anterior premo-
a left maxilla. lars reduced in size, single rooted, low crowned, conse-
Characteristics: Late Miocene (Vallesian-Turolian, Old quently length of anterior tooth row between canine and
World, ~6-12 Ma; medial Hemphillian, New World, ~6-8 camassial pair much shortened. The p2 lost, often pl and
Robert M. Hunt, Jr. 184

p3 as well. P4 robust, with strong parastylar cusp and TREMARCTINI

protocone shelf. M2 with only rudimentary talon (Hendey,
Characteristics: Short-faced ursids of moderate to large size in which
1980). The ml-2 cusp pattern varies in African, Eurasian,
the distance between P4 and C, and between m 1 and c in the lower
and North American species (see below). Premasseteric
jaw is strongly reduced; hence these species are prohibited from
fossa present; symphyseal region of lower jaw ventrally
the ancestry of longer jawed Ursus. Premasseteric fossa absent in
produced as a "chin." Long-footed, plantigrade.
Plionarctos, present in Tremarctos andArctodus. In all three genera,
Lower carnassial length range: Africa: A. africanum, 37 .8-44
diagnostic tuberculation of enamel surface in talonid and trigonid
mm (N = 7, Hendey, 1980); Eurasia: 33.7-41.8 mm (N =
basins of m2, forming a Z-pattern in the talonid basin and a pair of
7); North America: 38.7-50.5 mm (N = 44).
tubercles in the trigonid basin. Neomorph cusp developed on lower
Included species: Africa: A. africanum; Eurasia: A. inter-
carnassial between hypoconid and protoconid at junction of trigonid
medium, A. insignis, A. sivalensis; North America:
and talonid.
A. schneideri. North American Agriotherium is known
Tremarctini (Plionarctos, Tremarctos, Arctodus) first appear
from the following late Hemphillian localities: CA4,
about 7 Ma in North America as the genus Plionarctos in the late
GC13, CC40, NB12A, NB35, SB9, SBlO, SB60, SPlD,
Miocene Rattlesnake fauna of Oregon (late Hemphillian). Tremarc-
SP3B, SP4B, CP115D, CPI 13C, CP123D.
tos first appears in North America at about 2.5 Ma (Opdyke et al.,
1977) in the Vallecito and Grandview faunas of late Pliocene age,
Comments: Agriotherium's outstanding traits are its anteri-
and may have evolved directly from Plionarctos. No postcranials of
orly shortened lower jaw, premasseteric fossa, rudimen-
the rare Plionarctos are known, but Pleistocene discoveries of short-
tary M2 talon, tall-crowned ml with hypoconid crest
footed Tremarctos and long-footed Arctodus indicate that they are
intersecting the trigonid's posterior slope (not confluent
plantigrade bears (Kurten, 1966, 1967). Kurten believed that the
with the labial side of the trigonid as in Indarctos). In
Tremarctini never evolved the turning-in of the forefeet seen in
North America, Agriotherium is an important Eurasian im-
the gait of living Ursini. Bears of the genus Arctodus, the largest
migrant (Lindsay, Opdyke, and Johnson, 1984), useful in
land carnivorans that ever lived, do not appear in North America
definition of an "Agriotherium Biochron" ( ~4.5-6 Ma),
until Pleistocene (Irvingtonian) time, and were derived from early
which appears to postdate the biochron of Indarctos
Tremarctini. Arctodus and Tremarctos are the only ursids to reach
(~6-8 Ma).
South America.
Lower carnassial morphology is important in indicating
the existence of regional groups of Agriotherium, which
Plionarctos Frick, 1926
can be recognized by the talonid cusp pattern of ml. Cusp
Type species: Plionarctos edensis Frick, 1926.
pattern is regionally constant: In the African species, only
Type specimen: AMNH 18115, partial maxilla with partial
two cusps (metaconid, entoconid) occur posterointernal to
M 1 and complete M2.
the protoconid on the inner margin of the talonid. Between
Characteristics: Medial Hemphillian to early Blancan ( ~7-?3
metaconid and protoconid, a steep posterior slope of the
Ma) mid-sized New World ursids, similar to small indi-
trigonid occurs. In the North American sample, in con-
viduals of the living Andean Bear (Tremarctos ornatus,
trast, protoconid, metaconid, entoconid, and entoconulid
60-150 kg). Basicranium unknown. Complete skull un-
are aligned, forming a smoothly rising curve from talonid
known; basilar skull length estimated to be about 20 cm.
to trigonid. Because the North American species possesses
No premasseteric fossa. Size range of upper carnassial
both metaconid and entoconid, it is unlikely that there
suggests sexual dimorphism. Earliest known development
has been any replacement of the metaconid in agriotheres
in North America of dental traits of living Ursini and
(Hendey, 1980). In Asia, Agriotherium from the Siwalik
Tremarctini: P4 reduced in both length and overall size
Hills and from China is much like the African form in
relative to M 1-2; Ml lengthened; M2 lengthened by growth
the cusps of the inner talonid margin of ml. However, in
of elaborate talon; rugose wrinkled enamel developed in
m2 form, the Asian and North American species are alike,
interior basins of Ml-2 and m2-3; m2-3 elongate; m2
whereas the African species is unique: The former group
rectangular with greatly lengthened talonid. Postcranial
has metaconid and protoconid about equal in height (or oc-
skeleton unknown.
casionally with metaconid slightly higher), and with both
Lower carnassial length range: P. edensis, 21.7-23.7 mm
an entoconid and entoconulid behind the metaconid. The
(N = 2, Blancan forms only).
African species, in contrast, is distinguished by a proto-
Included species: P. edensis, known from the following local-
conid higher than metaconid, by the presence of an ento-
ities: medial to late Hemphillian (GC13B, CC40, CPI 16E,
conid, but with entoconulid very weak. The distinguishing
PN12); Blancan (CC46, SBl8C).
feature of A. africanum is the dominant m2 protoconid
that exceeds in relative development any known form in Comments: Plionarctos is the earliest member of the Tremar-
Europe, Asia, or North America. The consistency of these ctini to appear in the New World. A maxilla from the Up-
cusp patterns of the lower molars suggests that Agrio- per Bone Valley Formation, Florida, shows that these bears
therium was divided into a few major regional races or were already short faced as are their direct descendants,
sibling species. Tremarctos andArctodus. In addition to the short rostrum,
 Ursidae 185

the m2 has a remarkably invariant pattern of enamel tu- The diversity of living species of Ursus did not evolve until the
berculation (Z-pattern) that is developed in individuals of Plio-Pleistocene. During the Miocene, the ancestry of Ursus is to
all three genera of Tremarctini. Plionarctos has already be found among the seven to ten species of Ursavus, a number
developed a reduced camassial pair, and strongly elongate of which are probably members of a single phyletic lineage (Heiz-
M2, diagnostic of the later Ursinae. When discovered, the mann, 1973). The earliest Old World Ursini are European Ruscinian
postcranial skeleton will probably be nearly identical to species attributed to Ursus minimus and U. boeckhi (MN 14-15,
the short-footed plantigrade Tremarctos. The rarity of the ~3.5-5 Ma). Arrival in North America of the first Ursus occurs
genus in the Great Plains, but its presence in Florida and the about 4.3 Ma in the Pliocene (early Blancan), represented by rare
West Coast, suggests it occupied moister forested habitat fossils of Ursus abstrusus from the White Bluffs fauna, Washington
and did not often penetrate more open environments. (Gustafson, 1978). The first North American occurrence of Ursini is
predated by the slightly earlier appearance of Tremarctini (Plionar-
Tremarctos Gervais, 1855
ctos) in the late Hemphillian. By the Pleistocene (Irvingtonian),
Type species: Tremarctos ornatus (Cuvier, 1825).
both tribes of ursine bears are well represented in North America
Type specimen: Museum d'Histoire Naturelle, Paris.
by Ursus, Tremarctos, and Arctodus.
Characteristics: Late Blancan to Recent (~2.5 Ma-present)
short faced mid-sized to large ursids, strongly sexually di- Ursus Linnaeus, 1758
morphic (T. fioridanus). Ursid basicranium and Type A Type species: Ursus arctos Linnaeus, 1758.
bulla. Brachycephalic, with skull lengths of 19-32 cm. Type specimen: Locale uncertain.
Distinguished from Plionarctos by development of the pre- Characteristics: North American Ursus includes early Blan-
masseteric fossa, and by a slightly longer M2 talon. These can to Recent ( ~4.3 Ma-present) mid-sized to large ursids,
are the small-toothed bears of Kurten, in which "the main sexually dimorphic (Blancan North American sample too
cusps of ml form a characteristic alternating, elongate small to decide the question of dimorphism). Ursid basi-
W-pattern" (Kurten and Anderson, 1980, p. 178). Tooth cranium and Type A bulla. Skull dolichocephalic (Blancan
size similar to Plionarctos (combined samples of both T. skulls not known, but estimated at ~20 cm); known skull
fioridanus and living T. ornatus yield measurements of lengths range from Ursus minimus (~25 cm) to U. arc-
P4/p4, Ml/ml, M2/m2 with ranges that encompass most tos (to 45 cm) and U. spelaeus (to 50 cm) (Kurten, 1976).
Plionarctos measurements for these teeth). Short-footed, Dental formula I3/3, C 1/1, P2-4/l-4, M2/3. No premasse-
plantigrade. teric fossa. The m2 shorter than ml in Blancan forms, but
Lower carnassial length range: T. ornatus, 17-20.9 mm m2 later increasing to exceed ml length; great elongation
(N = 28); T. fioridanus, 20.2-23.8 mm (N = 17) (Kurten, of m2-3 and Ml-2; Pleistocene forms exhibit twinning
1966j. ofmetaconid of ml and m2 and considerable widening of
Included species: South America: T. ornatus; North America: m2-3 relative to ml. Associated postcranials unknown in
T. fioridanus. North American Tremarctos is known from Blancan Ursus.
late B!ancan localities NB 13C and PN23C. Lower camassial length range: Blancan U. abstrusus,
20.3 mm (N = l).
Included species: North America (Blancan): U. abstrusus.
URSINI
North American Ursus is known from early to medial Blan-
Characteristics: Pliocene to Recent Eurasian and North can localities PN3C and PN23A, and late Blancan locality
American mid-sized to large ursids, sexually dimorphic. Snout not SPIH.
as short as in Tremarctini; skull longer and narrower with longer mo-
Comments: Oldest North American Ursus are occurrences
lars than in Tremarctini. Ursus, the only genus, first appears from 4
at White Bluffs, Washington, and Hagerman, Idaho. The
to 5 Ma in both Old and New World as bears of about the same gen-
White Bluffs material consists of tooth fragments of ml
eral size and dental traits. Ursini lack the small accessory cusp at the
and Ml (Gustafson, 1978); them 1 trigonid compares with
lateral margin of the lower carnassial at the junction of trigonid and
the Hagerman ml belonging to the holotype of U. ab-
talonid that characterizes Tremarctini, according to Kurten. Earliest
strusus. The Hagerman holotype lower jaw (Bjork, 1970)
Ursus in Old and New World (U. minimus, U. abstrusus) retain a
has a primitive m 1 without twinned metaconid; the m2 was
primitive lower camassial form like Ursavus with a somewhat sec-
not much if any wider than m 1. A third Blancan occurrence
torial higher trigonid, but a low trigonid soon evolves in U. etruscus
of Ursus at Cita Canyon, Texas, is recorded in a fauna! list
of the European late Pliocene. In later Ursus, several key traits of
by Johnston and Savage (1955) but is not described.
living Ursini appear: (1) twinning of the ml metaconid; (2) expan-
sion of m 1 talonid and development of elongation between trigonid
and talonid; (3) Ml-2, m2-3 become enormously lengthened and BIOLOGY AND EVOLUTIONARY PATTERNS
broadened relative to the carnassials. In living U. arctos, even the
metaconids of m2-3 are twinned. All Ursini are plantigrade and tend Our present concept of ursid evolution visualizes the Cenozoic
to toe-in with the forefeet when walking. Emphasis of metatarsals development of the family as a series of three successive radia-
4-5 begins with U. etruscus and continues in later species of Ursus. tions: amphicynodonts, hemicyonines, and ursines. Ecological roles
 Robert M. Hunt, Jr. 186

w
Z L

: =·~[ =~§~~e- -=-~-

w E.BI
0 4.6
0 E
:J
I.at.Hp.
L.Hp.
5.2 !--------------------------- n -.s
0 - - QI
_.,__
-c:Ill 111--!I)
a. 6.0
L.E.Hp.
E.E.Hp.
7.0
~ 1
i§ -·2- -~
- s, __a. e-
i-;;
_"::l_

-
L 8.8

~1-
L.CI. ':::::.. _"'(_
9.5

E.CI.
w 11.0 en -
z L.L.Ba. GI
c
w E.L.Ba.
12.5
·2
0
0 M 14.0 c::
Q
g.
31 E.Ba. u
L.Hm.
5.8
~
e ·e
GI
GI :c
~I
17.5

z~
E.Hm.
18.8
E L.L.Ar. Q.
19.2

E.L.Ar.
23.0

w
z LE.Ar.

w L
0 27.7
0
CJ E.E.Ar.

:J 29.4
_l_
0
Wt.
E 31.9
Or.
33.4
L.Ch.
34,5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5
.._____, 1-.J
L.Un.

w 41.3 Ursavini Tremarctini Ursini

z E.Un.
w
0

t
45.9
0 HEMICYONINAE
W M
L.Br. URSINAE
46.7

M.Br. AMPHICYNODONTINAE
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
55.0 -
E.E.Wa. ss.s
Ck.3 55.6
1----------------------------- ------------=--j
w Ck.2 55,8
z ~1 ~:g
w L Tl.5 570
0 Tl.4 57:3
0 ~:: 58.5
w TI.1 59.2
..J
<( To.3
a. To.2

E ~-~
Pu.2 64.5
Pu.1 65.0
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu.= Puercan.
Figure 10.4. Temporal ranges of North American Tertiary ursid genera.
 Ursidae 187

I I I
CHADRONIAN ORELLAN WHITNEY AN ARKAREEAN HEMl'JGFORDIAN BARSTOVIAN CLARENDONIA~ HEMPHLLtANI BLANCAN
0
@) @) ~ @ @: >
r

"'' -o-:.,, cs
' ' '
"'"' "'"' .... 0 ....0
:0
"'0"' z
"'"' "' "' );

®® ®CID z
m
' <
~ "' ....' >
'!'
"'....
"'"' .... 0
>
..,,
® ® ®~ 'e=<
r@, &:; 0
:0
' ...' ' ' :i:: m
- z
"'"'... "'"' 0 Gl
"' ... o~ 0
"' "co
z
z

®@:., Ei
>
0 0
:r
0
"' "'
~'W

s::
® ®~
>
'0<
0
z
' -I
:c
z>
~ '
... "'
0 !l1 >
~

'"
® ® © @ :!!
-<
0
'
... s::
E "'"' "'"' "' zGl
"'
~~
z
@ @ ®~ 0
® ® 0(/)
>o
;>;c
"'cl> "'
~ "'"' ~-< ...
'
C> :::.
0 -I
~ :r
-.2::.

® ®
'
© ©
'
@ID
'
® ~ z
m
OJ
"' - "'
' '
! '.!: "'.... "'"' ... "' "'
!? "' 9 "'0 0
:0
>
:: ....
"'°' "'"' (/)
"' 0 ;>;
>

@® 8r
l:I
© @ s::
' ~z 0
'
...."' "'"'"',,,... o~ :0
"'"' "' 0 >
8
CID ~ >
:0
N
....' 0
0 z
;>; "'cl> >
m
-< "' "' 0
0
-I
®® ~ p>m-I
'
2~
Gl '
m
z
~
"'.... cl:"'
.... - >
>
CID ;>;
>
' z
(/)
"'
~ >
(/)

"' ..,,
~ ' 0'
r
0
:0
"' ..."' "'
<O:.... N
0
Ei
>

Figure 10.5. Biogeographic ranges of ursids. Numbers in this chart refer to the locality numbers listed in the appendix at the end of this chapter.
 Robert M. Hunt, Jr. 188

played by member species within these radiations differed to a great Beaumont, G. de (1965). Contribution aI' etude du genre Cephalogale Jour-
extent: The small primitive arctoid amphicynodonts were plausible dan (Carnivora). Memoires suisses de Paleontologie, 82, 1-34.
(1982). Breves remarques sur la dentition de certains ursictes (Marn-
generalists, giving rise to larger hemicyonines that explored the
mireres). Archives des Sciences, Geneve, 35, 153-6.
niche for cursorial predators of moderate and large body size. In Bjork, P. R. (1970). The Carnivora of the Hagerman local fauna (Late
North America, canids underwent a major Neogene diversification Pliocene) of southwestern Idaho. Transactions of the American
and radiation as the ecological vicars of hemicyonines, but whereas Philosophical Society, 60, 1-54.
almost no canids reached the Old World in the Miocene, periodic Bonis, L. de (1973). Contribution al'etude des mammireres de I' Aquitanien
de I' Agenais. Rongeurs - Carnivores - Perissodactyles. Memoires
Miocene migrations of Old World hemicyonines into North America
du Museum National d'Histoire Naturelle (Serie C), Science de la
must have engendered competition between these digitigrade ursids Terre, 28, 1-192.
and canids. Hemicyonines, together with the beardogs or amphicy- Cautley, P. T., & Falconer, H. (1836). Notes on the Ursus sivalensis, a
onids, were the large terrestrial carnivorans of the early and middle new fossil species from the Sivalik Hills. Asiatic Researches, 19,
Miocene, but by the end of that epoch the hemicyonine-amphicyonid 193-200.
Chaffee, R. G. (1954). Campylocynodon personi, a new carnivore from the
hegemony ended, and felids, some canids, and ursine bears evolved
Beaver Divide, Wyoming. Journal of Paleontology, 28, 43-7.
to become the large land carnivorans of the Plio-Pleistocene. Ciro!, E. (1992). Etude Phylogenetique de Quelques Genres d'Arctoidea de
The geographic locus of ursid evolution has been the northern !'Oligocene Eurasiatique. Poitiers: These pour l'obtention du Grade
continents. South America received only the ursine Tremarctini in de Docteur, l 'Universite de Poitiers.
the Pleistocene; Africa was invaded by hemicyonines in the early Cirot, E., & Bonis, L. de (1992). Revision du genreAmphicynodon, carnivore
de ]'Oligocene. Palaeontographica, 220, 103-30.
Miocene, by agriotheres in earliest Pliocene, and by Ursus in the
Clark, J., & Guensburg, T. E. (1972). Arctoid genetic characters as related
Quaternary. Eurasia, by virtue of its diverse ursid record, was cer- to the genus Parictis. Fieldiana: Geology, 26, 1-71.
tainly involved in the progressive unfolding of the primary ursid Colbert, E. H. (1939). Carnivora of the Tung Gur Formation of Mongolia.
radiations. At the moment, North America appears to have been Bulletin of the American Museum of Natural History, 76, 47-81.
the recipient of episodic pulses of migration from Eurasia. But one (1941). The ancestral ursid Hemicyon in Nebraska. Bulletin of the Uni-
versity of Nebraska State Museum, 2, 49-57.
must be cautious, for much of the North American Neogene mam-
Crusafont-Pair6, M., & Kurten, B. (1976). Bears and bear-dogs from
mal record is derived from sediments deposited in open savanna or the Vallesian of the Valles-Penedes Basin, Spain. Acta Zoologica
savanna-parkland settings of the continental interior, and the more Fennica, 144, 1-29.
mesic forested environments in coastal regions occupied by many Cuvier, F. (1825). Histoire Naturelle des Mammiferes, Vols. I-VII (1818-
ursid species are poorly represented. For example, recent discoveries 1842).
Filhol, H. (1879). Etude des mammireres fossiles de Saint-Gerand-le-Puy
ofursids in the northwestern United States indicate a greater diver-
(Allier). Bibilotheque de I'Ecole des Hautes Etudes, Section des Sci-
sity than earlier believed. A significant number of North American ences Naturelles, Paris, 19, 107.
fossil ursids are known only from the Pacific Northwest as single Frick, C. (1926). The Hemicyoninae and an American Tertiary bear. Bulletin
occurrences (the Pliocene panda Parailurus from Washington, and of the American Museum of Natural History, 56, 1-119.
Oligocene Allocyon, Parictis, and earliest Ursavus from Oregon) Gervais, F. L. P. (1855). Mammiferes. AnimauxNouveaux, ou Rares, Recueil-
lis Pendant !'Expedition dons les Parties Centrales de l'Amerique du
or by only a very few specimens (early Miocene Kolponomos
Sud. Paris: P. Bertrand.
from Oregon and Washington, Pliocene Ursus from Idaho and Ginsburg, L. (1955). De la subdivision du genre Hemicyon Lartet. Bulletin
Washington), which suggests how little we know of carnivoran di- de la Societe geologique de France (6th Serie), 5, 85-99.
versity in these Cenozoic faunas of the Pacific Northwest. (1961). La faune de carnivores Miocenes de Sansan. Memoires du
Ursids today are seen in a much different and more inclusive Museum National d'Histoire Naturelle (Serie C), Sciences de la
perspective relative to G. G. Simpson's view of 1945. The family Terre, 9, 1-190.
(1980). Plithocyon bruneti nov. sp., Hemicyoninae (Ursidae, Carnivora,
concept has been extended far beyond the morphologically uniform Mammalia) du Miocene de France. Compte Rendu sommaire Societe
and dentally monotonous living ursine bears to encompass a series geologique de France, 1980 (6), 232-5.
of Neogene radiations that more accurately reveal the Ursidae as Gustafson, E. P. (1978). The vertebrate faunas of the Pliocene Ringold
a group experimenting with both cranial and postcranial functional Formation, south-central Washington. Bulletin ofthe Museum ofNat-
ural History of the University of Oregon, 23, 1-62.
strategies throughout their approximately 40-million-year history.
Heizmann, E. P. (1973). Die Carnivoren des Steinheimer Beckens. B. Ursi-
In this respect, recent research on fossil ursids is representative of dae, Felidae, Viverridae sowie Erganzungen und Nachtrage zu den
new paleontological discoveries derived from the other families of Mustelidae. Palaeontographica, Suppl. 8 (SB), 1-95.
Carnivora, where diversity and evolutionary pattern are proving to Hendey, Q. B. (1977). Fossil bear from South Africa. South African Journal
be richer, more complex, and varied than conveyed by earlier work. of Science, 73, 112-16.
(1980). Agriotherium (Mammalia, Ursidae) from Langebaanweg, South
Africa, and relationships of the genus. Annals of the South African
Museum, 81, 1-109.
REFERENCES Hofmann, A. (1887). Uber einige Siiugetierreste aus der Braunkohle von
Voitberg und Steieregg bei Weiss, Steiermark. Jahrbuch k.k. Geo!.
Baskin, J., & Tedford, R.H. (1996). Small arctoid and feliform carnivorans. Reichsanst., Wein, 37, 201-18.
In The Terrestrial Eocene-Oligocene Transition in North America, Hunt, R. M., Jr. ( 1974 ). The auditory bulla in Carnivora: an anatomical basis
eds. R. J. Emry & D.R. Prothero, pp. 486-97. Cambridge: Cambridge for reappraisal of carnivore evolution. Journal of Morphology, 143,
University Press. 21-76.
 Ursidae 189

(1977). Basicranial anatomy of Cynelos Jourdan (Mammalia, Car- bearing section of Anza-Borrego State Park, California. Quaternary
nivora), an Aquitanian amphicyonid from the Allier Basin, France. Research, 7, 316-29.
Journal of Paleontology, 51, 826-43. Pilgrim, G. (1913). The correlation of the Siwaliks with mammal hori-
( 1987). Evolution of the aeluroid Carnivora: significance of auditory zons of Europe. Records of the Geological Society of India, 43,
structure in the nirnravid cat Dinictis. American Museum Novitates, 264-326.
2886, 1-74. Qiu, Z., Yan, D., Jia, H., & Wang, B. (1985). Dentition of the Ursavus skele-
Hunt, R. M., Jr., & Barnes, L. G. (1994). Basicranial evidence for ursid ton from Shanwang, Shan dong Province. Vertebrata Palasiatica, 23,
affinity of the oldest pinnipeds. In Contributions in Marine Mammal 264-75.
Paleontology Honoring F. C. Whitmore, Jr., eds. A. Berta & T. A. Remy, J. A., Crochet, J.-Y., Sige, B., Sudre, J., Bonis, L. de, Vianey-
Demere. Proceedings of the San Diego Society of Natural History, Liaud, M., Godino!, M., Hartenberger, J.-L., Lange-Badre, B., &
29, 57-67. Comte, B. (1987). Biochronologie des phosphorites du Quercy: Mise
Johnston, C. S., & Savage, D. E. (1955). A survey of various Late Cenozoic a jour des listes fauniques et nouveaux gisements de mammireres
vertebrate faunas of the Panhandle of Texas, Part I: Introduction, fossiles. Miinchner Geowissenschaften Abhandlungen (A), 10,
description oflocalities, preliminary fauna] lists. University of Cali- 169-88.
fornia Publications in Geological Sciences, 31, 27-50. Schmidt-Kittler, N. (1987). The Carnivora (Fissipedia) from the Lower
Jourdan, C. ( 1862). Description de restes fossiles de grand Mammireres. Miocene of East Africa. Palaeontographica, Part A, 197, 85-126.
Rev. Soc. Sav., 1, 126-30. Schlosser, M. (1899). Uber die Biiren und bareniihnlichen Formen des eu-
Kurten, B. (1966). Pleistocene bears of North America, Part 1: Tremarctos, ropiiischen Tertiiirs. Palaeontographica, 197, 95-146.
spectacled bears. Acta Zoologica Fennica, 115, 1-120. Scott, W. B. (1893). On anew musteline from the John Day Miocene. Amer-
(1967). Pleistocene bears of North America, Part 2: Arctodus, short-faced ican Naturalist, 27, 658-9.
bears. Acta Zoologica Fennica, 117, 1-60. Simpson, G. G. (1945). The principles of classification and a classification
(1969). Cave bears. Studies in Speleology 2, 13-24. of mammals. Bulletin of the American Museum of Natural History,
(1976). The Cave Bear Story. New York: Columbia University Press. 85, 1-350.
Kurten, B., & Anderson, E. ( 1980). Pleistocene Mammals ofNorth America. Stirton, R. A. ( 1960). A marine carnivore from the Miocene Clallam Forma-
New York: Columbia University Press. tion, Washington: its correlation with nonmarine faunas. University
Lindsay, E. H., Opdyke, N. D., & Johnson, N. M. (1984). Blancan- of California Publications in Geological Sciences, 36, 345-68.
Hemphillian land mammal ages and Late Cenozoic mammal dis- Tedford, R. H., Barnes, L. G., and Ray, C. E. (1994). The early Miocene
persal events. Annual Review of Earth and Planetary Sciences, 12, littoral ursoid camivoran Kolponomos: systematics and mode of life.
445-88. In Contributions in Marine Mammal Paleontology Honoring F. C.
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundem Whitmore, Jr., eds. A. Berta & T. A. Demere. Proceedings of the San
classes, ordines, genera, species cum characteribus, differentiis, syn- Diego Society of Natural History, 29, 11-32.
onymis, locis. Editio decima, reformata. Stockholm: Laurentii Sal vii. Viret, J. (1929). Les faunes de mammireres de l' Oligocene superieur de la
Matthew W. D., & Granger, W. (1924). New Carnivora from the Tertiary of Limagne bourbonnaise. Annales de l'Universite de Lyon (n.s.), 47,
Mongolia. American Museum Novitates, 104, 1-9. 1-328.
Merriam, C. W. (1930). Allocyon, a new canid genus from the John Day Voorhies, M. R., & Corner, R. G. (1982). Ice Age Superpredators. University
beds of Oregon. University of California Publications, Bulletin of of Nebraska State Museum Notes, 70, 1-4.
the Department of Geological Sciences, 19, 229-44. Wagner, A. (1837). Paliiontologische Abhandlungen. Gelehrte Anzeigen her-
Opdyke, N. D., Lindsay, E. H., Johnson, N. M., & Downs, T. (1977). The ausgegeben von Mitgliedern der k. hayer, Akadamie Wissenschaften
paleomagnetism and magnetic polarity stratigraphy of the mammal- Miinchen,5,334-5.
 Appendix: Key to fossil localities

LISTING OF LOCALITIES BY NUMBER 15. White River Group, Beaver Divide, Fremont County,
Wyoming (CP39IIC)

Chadronian (Parictis)
Orellan (undescribed ?amphicynodont genus)
1. Chadron Formation, Seamen Hills area, Niobrara County,
Wyoming (CP41A) 16. Orella Member, Brule Formation, near Roundtop, Dawes
2. Chadron Formation, Flagstaff Rim, Natrona County, Wyom- County, Nebraska (CP99A)
ing (CP39B, C, F)
3. Chadron Formation, Stebbins Ranch, Custer County, South
Dakota (CP83C) Orellan (Parictis)
4. Chadron Formation, Bohling Ranch, Shannon County, South
Dakota (CP83C) 17. White River Group, Chimney Canyon, Logan County, Col-
5. Chadron Formation, sect. 12, T42N, R45W, Shannon County, orado (CP68C)
South Dakota (CP83C) 18. Orella Member, Brule Formation, sects. 12 and 14, T33N,
6. Chadron Formation, Quinn Draw, Shannon County, South R54W, Sioux County, Nebraska (CP99A)
Dakota (CP83C) 19. ?Orella Member, Brule Formation, Everson Ranch, Dawes
7. Chadron Formation, Big Corral Draw, Shannon County, County, Nebraska (?CP99A)
South Dakota (CP83C) 20. Scenic Member, Brule Formation, Cottonwood Pass area,
8. Chadron Formation, Little Corral Draw, Pennington County, Washabaugh County, South Dakota (CP84A)
South Dakota (CP83C) 21. Scenic Member, Brule Formation, Big Corral Draw, Shannon
9. Chadron Formation, Warbonnet Creek, Sioux County, County, South Dakota (CP84A)
Nebraska (CP98B or C) 22. White River Group, NE of Walker Ranch, Niobrara County,
10. Chadron Formation, Arner Ranch, Sioux County, Nebraska Wyoming (CP41B)
(CP98B or C) 23. John Day Formation, Camp Creek, Silver Wells, Oregon
11. Chadron Formation, sect. 2, T32N, R52W, Dawes County, (PN6B, ?Orellan)
Nebraska (CP98B or C) 24. John Day Formation, Oregon (PN6, age uncertain)
12. Renova Formation, Pipestone Springs Main Locality,
Jefferson County, Montana (NP24C)
13. Renova Formation, Little Pipestone Creek, Jefferson County, ?Whitneyan (Parictis)
Montana (NP24D)
14. Unnamed unit, Douglas Creek, Powell County, Montana 25. Fitterer Ranch, SW of Dickinson, Stark County, North Dakota
(NP29C) (?NP50C)

190
 Appendix: Key to fossil localities 191

Late Arikareean (Cephalogale) Barstovian (?Plithocyon)

26. Zia Sand Formation, Standing Rock Quarry, Sandoval Coun- 42. Unnamed unit, 10 miles NE of Saratoga, Carbon County,
ty, New Mexico (SB29A) Wyoming (CP55)
27. Upper Harrison beds, University Quarry, Agate National Mo··
nument, Sioux County, Nebraska (CP104A)
28. Upper Harrison beds, 16 Mile District, Goshen County, Late Arikareean (Kolponomos)
Wyoming (CP51A)
43. Clallam Formation, Slip Point, Clallam Bay, Clallam County,
Washington (WM3II)
Early Hemingfordian (Cephalogale) 44. Nye Mudstone, near Newport, Lincoln County, Oregon
(WM6)
29. Unnamed unit, Bridgeport Quarries, Morrill County,
Nebraska (CP106)
30. Sh-101B, Runningwater Formation, Sheridan County, Early Arikareean (Ursavus, Allocyon)
Nebraska (CP105)
31. Runningwater Formation, Cephalogale Ash Locality, sect. 45. John Day Formation, Logan Butte area, Crook County,
29, T29N, RSI W, Dawes County, Nebraska (CP105) Oregon (PN6C)
32. Runningwater Formation, Gregory Ranch, sect. 24, T30N,
R48W, Dawes County, Nebraska (CP105)
33. Runningwater Formation, Hemingford Quarry 12D, sect. 12, Early Hemingfordian (Ursavus)
T28N, R49W, Box Butte County, Nebraska (CP105)
46. Batesland Formation, Flint Hill (UCMP Locality V3417),
Bennett County, South Dakota (CP88)
Early Hemingfordian (Phoberocyon)

34. Sinkhole filling, Thomas Farm, Gilchrist County, Florida Medial or late Hemingfordian (Ursavus)
(GC8D)
47. Unnamed unit, UCR Locality RV7043, Massacre Lake I,
Washoe County, Nevada (NB 17)
Early Barstovian (Plithocyon)

35. Barstow Formation, Steepside Quarry (upper level), San Early or medial Barstovian (Ursavus)
Bernardino County, California (NB6C)
36. Barstow Formation, Turbin Quarry, San Bernardino County, 48. Sucker Creek Formation, CIT Locality 44, Malheur County,
California (NB6C) Oregon (PN9A)
37. Tesuque Formation, Skull Ridge Member, Espanola Basin, 49. Butte Creek Volcanic Sandstone, CIT Locality 57, Red Basin,
New Mexico (SB32B) Malheur County, Oregon (PN8B)

Medial barstovian (Plithocyon) Medial Barstovian (Ursavus)

38. Barstow Formation, "Hemicyon" Quarry, San Bernardino 50. Wood Mountain Formation, Yost Farm, Saskatchewan,
County, California (NB6E) Canada (NPI I)
51. Stewart Valley Formation, UCMP Locality V6020, Mineral
County, Nevada (NB24)
Late Barstovian (Plithocyon)

39. Tesuque Formation, Pojoaque Member, Pojoaque Bluffs lo- Late Barstovian (Ursavus)
calities of Frick Laboratory, Espanola Basin, New Mexico
(SB32D) 52. Valentine Formation, UNSM Locality Wt-15A, Webster
40. Ogallala Formation, UNSM Locality Fr-20, Franklin County, County, Nebraska (CP113A)
Nebraska (CP113A) 53. Ogallala Formation, Tapir Hill, Weld County, Colorado
41. Valentine Formation, Verdigre Quarry, Knox County, (CP76)
Nebraska (CPI 14C) 54. Ogallala Formation, Uhl Pit, Weld County, Colorado (CP76)
 Appendix: Key to fossil localities 192

55. Unnamed unit, Jersey Valley, Pershing County, Nevada 84. Unnamed unit, Tehuichila, Hidalgo, Mexico (CA4)
(NB25A) 85. Ogallala Formation, Coffee Ranch, Hemphill County, Texas
56. Humboldt Formation, UCMP Locality V4709, Elko County, (SP3B)
Nevada (NB 19B) 86. Ogallala Formation, Guymon-Optima quarries, Texas County,
57. Unnamed unit, High Rock Canyon 2, Washoe County, Nevada Oklahoma (SPID)
(NB20) 87. Ogallala Formation, Axtel Quarry, UCMP Locality V5319,
Randall County, Texas (SP4B)
88. Ogallala Formation, Edson Quarry, Sherman County, Kansas
Clarendonian (New genus) (CP123D)
89. Ogallala Formation, Lost Quarry, Wallace County, Kansas
60. Ogallala Formation, Lessig Canyon, Brown County, (CP123D)
Nebraska (CP116AA) 90. Ogallala Formation, near UNSM Locality Sm-101, Sherman
61. Ogallala Formation, Dutch Canyon, Brown County, Nebraska County, Nebraska (CP113C)
(CPI 16AA) 91. Snake Creek Formation, ?Johnson Member, Sioux County,
62. Ogallala Formation, June Quarry, Brown County, Nebraska Nebraska (CP1 I5D)
(CP114D) 92. Bone Valley Formation, Polk and Hillsborough Cos., Florida
63. Ogallala Formation, Lucht Quarry, Brown County, Nebraska (GC13)
(CP114D)
64. Ogallala Formation, White Point Quarry (Deep Creek),
Brown County, Nebraska (CPI 14D) Medial Hemphillian (lndarctos)
65. Ogallala Formation, Burge Quarry, Cherry County, Nebraska
(CP114D) 93. Rattlesnake Formation, East Fork of John Day River, Grant
66. Ogallala Formation, Little Beaver B Quarry, Cherry County, County, Oregon (PN12)
Nebraska (CPI 16A) 94. Unnamed unit, Juniper Creek Canyon, Malheur County,
67. Ogallala Formation, Devil's Jump Off area, Cherry County, Oregon (PN13)
Nebraska (CPI 14D or CPI 16) 95. Coal Valley Formation, Smiths Valley, Lyon County, Nevada
68. Ogallala Formation, west of Boiling Spring Bridge, Cherry (NB27B)
County, Nebraska (CPI 14C or D) 96. Mulholland Formation, UCMP Locality V73148, Contra
69. Ogallala Formation, Turtle Canyon, Sheridan County, Costa County, California (CC36)
Nebraska (CPI 16AA) 97. Ogallala Formation, Box T Quarry, Lipscomb County, Texas
70. Ogallala Formation, UNSM Locality Bn-103, Banner (SP3A)
County, Nebraska (CP113B) 98. Ogallala Formation, UNSM Locality Ft-40, Frontier County,
71. Ogallala Formation, Wolf Creek area, Shannon County, South Nebraska (CP1l6D)
Dakota (CP90B) 99. Sinkhole in Ocala Limestone, Withlacoochee 4A, Marion
County, Florida (GC12II)

Late Hemphillian (Agriotherium)

75. Coso Formation, USGS Locality Ml430, Inyo County, Medial to Late Hemphillian (Plionarctos)
California (NB 12A)
76. Eden Formation, UCMP Locality 3269, Riverside County, 100. Eden Formation, Riverside County, California (CC40)
California (CC40) I 01. Rattlesnake Formation, UCMP Locality 3045, Grant County,
77. ?Hay Ranch Formation, Carlin High Quarry, Elko County, Oregon (PN12)
Nevada (NB35) I 02. Ogallala Formation, Bear Tooth Slide, Brown County,
78. Big Sandy Formation, Claybank Quarry, Mohave County, Nebraska (CP1 I6E)
Arizona (SB9) 103. Upper Bone Valley Formation, Palmetto Mine, Polk County,
79. Big Sandy Formation, Birdbone Quarry, Mohave County, Florida (GC13B)
Arizona (SB9) 104. Upper Bone Valley Formation, Fort Green Mine, Polk
80. Quiburis Formation, Old Cabin Quarry, Pima County, County, Florida (GC13B)
Arizona (SBlO)
81. Quiburis Formation, Redington Quarry (= Bingham Ranch),
Pima County, Arizona (SBlO) Blancan (Plionarctos)
82. Quiburis Formation, Camel Canyon, Pinal County, Arizona
(SBlO) 105. Unnamed unit, South Red Knoll Quarry, Graham County,
83. Unnamed unit, Yepomera, Chihuahua, Mexico (SB60) Arizona (SB I 8C)
 Appendix: Key to fossil localities 193

I06. Unnamed unit, Aguanga horizon, Riverside County, 8. Chadron Formation, Little Corral Draw, Pennington County,
California (CC46) South Dakota
9. Chadron Formation, Warbonnet Creek, Sioux County,
Nebraska
Late Blancan (Tremarctos) 19. ?Orella Member, Brule Formation, Everson Ranch, Dawes
County, Nebraska
107. Palm Springs Formation, Anza-Borrego Desert, California 22. White River Group, NE of Walker Ranch Niobrara County,
(NBI3C) Wyoming
108. Glenns Ferry Formation, Grandview fauna, Owyhee
County, Idaho (PN23C)
Parictis montanus

12. Renova Formation, Pipestone Springs Main Locality,

Early to medial Blancan (Ursus) Jefferson County, Montana
13. Renova Formation, Little Pipestone Creek, Jefferson County,
109. Glenns Ferry Formation, Hagerman fauna, Twin Falls Montana
County, Idaho (PN23A) 14. Unnamed unit, Douglas Creek, Powell County, Montana
110. Ringold Formation, White Bluffs fauna, Franklin County,
Washington (PN3C)
Parictis personi

I5. White River Group, Beaver Divide, Fremont County,

Late Blancan (Ursus)
Wyoming

111. Ogallala Formation, Cita Canyon fauna, Randall County,

Texas (SPIH) Parictis primaevus

23. John Day Formation, Camp Creek, Silver Wells, Oregon

(genoholotype)
LISTING OF LOCALITIES BY SPECIES

Parictis dakotensis
Parictis sp.
1. Chadron Formation, Seamen Hills area, Niobrara County,
2. Chadron Formation, Flagstaff Rim, Natrona County,
Wyoming
Wyoming
3. Chadron Formation, Stebbins Ranch, Custer County, South
10. Chadron Formation, Arner Ranch, Sioux County, Nebraska
Dakota
11. Chadron Formation, sect. 2, T32N, R52W, Dawes County,
4. Chadron Formation, Bohling Ranch, Shannon County, South
Nebraska
Dakota
17. White River Group, Cedar Creek beds, Chimney Canyon,
7. Chadron Formation, Big Corral Draw, Shannon County,
Logan County, Colorado
South Dakota
18. Orella Member, Brule Formation, sects. 12 and 14, T33N,
8. Chadron Formation, Little Corral Draw, Pennington County,
R54W, Sioux County, Nebraska
South Dakota
20. Scenic Member, Brule Formation, Cottonwood Pass area,
Washabaugh County, South Dakota
Parictis major 21. Scenic Member, Brule Formation, Big Corral Draw, Shannon
County, South Dakota
5. Chadron Formation, sect. 12, T42N, R45W, Shannon County, 24. John Day Formation, Oregon
South Dakota
25. Fitterer Ranch, SW of Dickinson, Stark County, North Dakota

Parictis parvus Undescribed ?amphicynodont genus

6. Chadron Formation, Quinn Draw, Shannon County, South 16. Orella Member, Brule Formation, near Roundtop, Dawes
Dakota County, Nebraska

Parictis gilpini Allocyon loganensis

1. Chadron Formation, Seamen Hills area, Niobrara County, 45. John Day Formation, Logan Butte area, Crook County,
Wyoming Oregon
 Appendix: Key to fossil localities 194

Kolponomos clallamensis Plithocyon sp.

43. Clallam Formation, Slip Point, Clallam Bay, Clallam County, 40. Ogallala Formation, UNSM Locality Fr-20, Franklin County,
Washington Nebraska
41. Valentine Formation, Verdigre Quarry, Knox County,
Kolponomos newportensis Nebraska

44. Nye Mudstone, near Newport, Lincoln County, Oregon

?Plithocyon

Cephalogale new species I 42. Unnamed unit, 10 miles NE of Saratoga, Carbon County,
Wyoming
26. Zia Sand Formation, Standing Rock Quarry, Sandoval Coun-
ty, New Mexico
Ursavus new species I
Cephalogale new species II 45. John Day Formation, UCMP Locality V6632, Logan Butte
area, Crook County, Oregon
27. Upper Harrison beds, University Quarry, Agate National
Monument, Sioux County, Nebraska
28. Upper Harrison beds, 16 Mile District, Goshen County, Ursavus new species II
Wyoming
46. Batesland Formation, Flint Hill (UCMP Locality V3417),
Bennett County, South Dakota
Cephalogale new species III 48. Sucker Creek Formation, CIT Locality 44, Malheur County,
29. Unnamed unit, Bridgeport Quarries, Morrill County, Oregon
Nebraska
30. Sh-lOlB Runningwater Formation, Sheridan County, Ursavus cf. brevirhinus
Nebraska
31. Runningwater Formation, Cephalogale Ash Locality, sect. 47. Unnamed unit, UCR Locality RV7043, Massacre Lake I,
29, T29N, R51W, Dawes County, Nebraska Washoe County, Nevada
32. Runningwater Formation, Gregory Ranch, sect. 24, T30N, 52. Valentine Formation, UNSM Locality Wt-15A, Webster
R48W, Dawes County, Nebraska County, Nebraska
33. Runningwater Formation, Hemingford Quarry 12D, sect. 12,
T28N, R49W, Box Butte County, Nebraska Ursavus cf. primaevus
49. Butte Creek Volcanic Sandstone, CIT Locality 57, Red Basin,
Phoberocyon johnhenryi
Malheur County, Oregon
34. Sinkhole filling, Thomas Farm, Gilchrist County, Florida 50. Wood Mountain Formation, Yost Farm, Saskatchewan,
Canada
51. Stewart Valley Formation, UCMP Locality V6020, Mineral
Plithocyon new species
County, Nevada
35. Barstow Formation, Steepside Quarry (upper level), San
Bernardino County, California
Ursavus pawniensis (cf. depereh)
36. Barstow Formation, Turbin Quarry, San Bernardino County,
California 53. Ogallala Formation, Tapir Hill, Weld County, Colorado
37. Tesuque Formation, Skull Ridge Member, Espanola Basin, 54. Ogallala Formation, Uhl Pit, Weld County, Colorado
New Mexico 55. Unnamed unit, Jersey Valley, Pershing County, Nevada
56. Humboldt Formation, UCMP Locality V4709, Elko County,
Nevada
Plithocyon barstowensis
57. Unnamed unit, High Rock Canyon 2, Washoe County,
38. Barstow Formation, "Hemicyon" Quarry, San Bernardino Nevada New genus
County, California 60-71. All listed specimens from Nebraska and South Dakota
may belong to a single species.

Plithocyon ursinus
Agriotherium schneideri
39. Tesuque Formation, Pojoaque Member, Pojoaque Bluffs lo-
calities of the Frick Laboratory, Espanola Basin, New Mexico 75-92. All listed specimens probably belong to this species.
 Appendix: Key to fossil localities 195

Indarctos oregonensis Tremarctos sp.

93-99. All listed specimens probably belong to this species. 107-108. Material undescribed; species uncertain.

Plionarctos edensis Ursus abstrusus

100-106. All listed specimens probably belong to this species. 109-111. All listed specimens may belong to this species.
 11 Amphicyonidae

ROBERT M. HUNT, JR.

INTRODUCTION North America were without substantiated phyletic connection be-

tween ancestral and descendant genera. In this chapter I advance a
The Amphicyonidae occur in North America from late new hypothesis to account for the evolution and geographic distri-
Eocene (~40 Ma) to early late Miocene (~9 Ma), a span of 31 bution of amphicyonids. This hypothesis uses an improved knowl-
million years (Fig. I I.I). In the North American land mammal edge of biogeography of amphicyonid fossils on the northern con-
biochronology, this diverse family extends from the Duchesnean tinents as a significant new source of information in deciphering
to the end of the Clarendonian and is especially well represented in their phylogenetic relationships, while at the same time recognizing
the Miocene (~5-24 Ma). In the Old World, amphicyonids survive the lack of reliable synapomorphic traits to ally subgroups with-
no longer than in North America: The last European records are in the family. Simply stated, the distribution of amphicyonid fossils
Vallesian (Dinotheriensande, Mainz Basin, Germany, Kuss, 1965, in North America and Europe suggests the existence of two separate
Tobien, 1980; Valles-Penedes Basin, Spain, Crusafont-Pair6 and radiations of beardogs, both initiated in the late Eocene/Oligocene,
Kurten, 1976), and the last representative may occur at about 9 Ma at the European termed "amphicyonine," the North American iden-
Kohfidisch, Austria (latest Vallesian, Beaumont, 1984; Mein, 1989). tified as "daphoenine." Thus I employ the subfamilies as mono-
The group is extinct by 13 Ma in eastern Asia and by about 7 Ma phyletic groups arising and radiating in different geographic areas,
on the Indian subcontinent (Barry and Flynn, 1989). In Africa, am- yet sharing a common ancestral population (the stem amphicy-
phicyonids are no longer evident after about 11 Ma, but future dis- onid stock) sometime in the Eocene. The distribution of amph-
coveries may improve this sparse record as Miocene faunas become icyonid genera in time in North America and Europe is central
better known. to the proposed concept of amphicyonid evolution advocated in
Two North American subfamilies (Daphoeninae, Amphicyoni- this chapter: North American daphoenines are endemic to the New
nae) have been recognized, but as commonly employed they are World, first appearing about 40 Ma and continuing to about 17 Ma.
in fact arbitrary groupings without well-founded support. Because European amphicyonines originate about the same time (40 Ma) in
the Amphicyoninae simply succeeded the Daphoeninae in time, the the Old World and continue to 9 Ma in Europe, but amphicyonine
subfamilies appeared to be evolutionary grades, but the grades in lineages periodically invade North America as immigrants during

Figure 11.1. The giant amphicyonid Amphicyon ingens and its prey, a dromomerycid, mid-Miocene of the North American midcontinent. Restoration by
Paula Denham.
196
 Amphicyonidae 197

the late Oligocene and Miocene. All native daphoenines are extinct distinctive genera. The transitions between groups occur within rel-
by 17 Ma and are replaced by Old World amphicyonines. From this atively brief periods of time. The late Eocene-early Oligocene fea-
point in time until their North American extinction (~9 Ma), the tures the plesiomorphic Daphoenus and several archaic beardogs
beardogs of the New World are all derived from migrant Old World of small size ( <25 kg) that comprise the Daphoeninae: This as-
amphicyonines whose roots occur in European lineages. semblage shows only a remote relationship to the initial amphicy-
I also recognize two additional subfamilies, the North Ameri- onine radiation occurring in Europe at this same time. In the late
can Temnocyoninae and European Haplocyoninae. The dental spe- Oligocene and early Miocene the North American record is char-
cializations of these two groups provide distinctive synapomorphic acterized by temnocyonine amphicyonids (7-50 kg), recognizable
features that allow them to be identified and set apart. New World by their distinctive dentitions. Temnocyonines are the first beardogs
temnocyonines are plausibly derived from a haplocyonine immi- to attain moderate body size and became extinct in North America
grant in the late Oligocene, although each subfamily evolves its by 18 Ma. They are paralleled in Europe by their sister group, the
own unique dental morphologies in Europe and North America. Haplocyoninae, that can be recognized by similar dental specializa-
Abundant amphicyonid fossils have been found in both Europe tions.
and North America where the family displays a progression from However, in the late Oligocene and earliest Miocene (~32-22
generalized species of small body size to very large forms of a more Ma), only temnocyonine and daphoenine beardogs are found in
specialized nature. Amphicyonids are also known from Asia and North America. Whether this hiatus reflects an actual absence of
Africa where their remains are rare but of great interest in demon- amphicyonines is uncertain. The 22-32 Ma interval in the Great
strating affinity to European taxa. No amphicyonids are known from Plains and in Oregon is represented chiefly by fine-grained volcani-
South America, and their extinction in North America preceded the clastic loessic sediments in which camivorans and mammals in gen-
interchange of faunas that occurred in the Americas during the Plio- eral are poorly represented; hence this scarcity of fossils would not
Pleistocene. be surprising in such relatively barren sediments. At present, how-
In North America, the Duchesnean (late Eocene) fossils indicate ever, I favor the concept that the carnivoran fauna of North America
that the oldest known daphoenine amphicyonids are small carnivo- includes only temnocyonines and endemic daphoenines until the
rans less than 5 kg in body weight, about the size of a swift fox immigration into North America of multiple amphicyonine lineages
(Vulpes velox). Chadronian through Whitneyan (late Eocene-late from 18-22 Ma (early Miocene). This opinion is based on the re-
Oligocene) amphicyonids lack diversity, attain only moderate body peated discovery of temnocyonines and daphoenines in the Great
size no larger than the coyote Canis latrans (~ 25 kg), and are known Plains and in the John Day basin and absence of other beardogs
primarily from rocks of the White River Group in the midcontinent despite over a century of collecting.
and Rocky Mountain basins. However, in the Arikareean and Hem- In the early Miocene of North America, a new group of amphicy-
ingfordian (late Oligocene-early Miocene), daphoenine and amph- onid genera (amphicyonines Cynelos and Ysengrinia, daphoenine
icyonine amphicyonids oflarge size (100-200 kg) appear in North Daphoenodon) appears for the first time. They are notably bigger
America, diversifying to include wolflike and bearlike species that (~20-100 kg) than earlier species and include the first large North
parallel the living canid Canis lupus and the ursid Ursus arctos. American beardogs. The amphicyonines are immigrants from the
The largest North American species, which are bearlike in their Old World, recognizable in the European record, but not evident as
postcranial skeletons, are attributable to Amphicyon and occur in yet in Asia because of a sparser record and the scattered fragmen-
the mid-Miocene (late Hemingfordian and early Barstovian). Such tary nature of the Asian fossils. Daphoenodon is a North American
huge amphicyonine beardogs also are recorded from Eurasia and endemic. These genera increased in size in North America during
Africa where, together with large hemicyonine ursids, they were the early Miocene, but most became extinct by the end of this in-
the largest terrestrial carnivorans of the Miocene. terval. In the late early Miocene, these genera are joined by a new
There are no certain records of amphicyonids or hemicyonine ur- and important amphicyonine immigrant, a true Amphicyon from the
sids after the Clarendonian (early late Miocene) Land Mammal age Old World at about 18 Ma.
in North America. The succeeding Hemphillian (late late Miocene) By the mid-Miocene a characteristic suite of amphicyonids be-
mammal fauna is dominated by large ursids (Agriotherium, Indarc- came established in the New World, comprising Amphicyon, Cyne-
tos) that appear to fill the vacuum left by the extinction of amphi- los, and Pliocyon. Together with hemicyonine ursids, these are the
cyonids and hemicyonines, and also by large felids (Machairodus, common large carnivorans of the late Hemingfordian and early
Nimravides) and hyaenalike (Osteoborus) and wolfl.ike (Aelurodon) Barstovian where some species became enormous (>200 kg). They
canids. Within the late Miocene the North American fossil record were abruptly replaced in the late Barstovian and Clarendonian
documents a significant shift in the dominant large land camivorans by the terminal North American beardogs, the Pseudocyon-Ischy-
from the amphicyonid-hemicyonine fauna of the Clarendonian to rocyon group, both genera approaching or attaining body weights
an Agriotherium-Indarctos-Machairodus fauna in the Hemphillian. of over 100 kg.
The disappearance of hemicyonine ursids and amphicyonids is es- As in the case of ursids, an interesting aspect of amphicyonid
pecially striking because they were an integral part of the North evolution is the lack of phyletic continuity between North American
American camivoran fauna since the Hemingfordian age, persisting genera. The North American record is punctuated by the repeated
during an interval of about 10 million years. entrance of immigrant stocks from the Old World that persist for
The amphicyonid fauna of North America· changes through a a time, then become extinct. Consequently, first appearances and
succession of temporal intervals each characterized by a group of extinctions of amphicyonids can contribute to the recognition of
 Robert M. Hunt, Jr. 198

important biochronologic datum events in the Cenozoic of North the bulla of Daphoenus medial to the ectotympanic: In the genoholo-
America. type skull of Daphoenus vetus Leidy, the ectotympanic is separated
from the petrosal promontorium by a space that would accommo-
date a cartilaginous caudal entotympanic plate, and in addition there
DEFINING FEATURES OF THE is a conspicuous flange on the medial rim of ectotympanic that must
FAMILY AMPHICYONIDAE have contacted the dorsal rim of such a plate. Recent new evidence
demonstrates the presence of an ossified caudal entotympanic in a
temnocyonine beardog: A young adult amphicyonid cranium was
CRANIAL discovered in the Upper John Day beds of Oregon (Fingerut et al.,
Figure 11.2 illustrates some aspects of amphicyonid osteology. The 1993) with intact bullae displaying caudal entotympanics fused to
structure of the basicranium is diagnostic for amphicyonids in com- the ectotympanic as in Type A arctoid bullae (Hunt, 1974a, Pl. 4),
pany with key dental features. Amphicyonids for which the basi- demonstrating that the ontogenetic elements of at least some amph-
cranium is known possess a primitive arctoid auditory bulla termed icyonids were assembled as in living ursids.
Type A (Hunt, 1974a). In contrast to ursids, the number of ontoge- In all amphicyonids, the ectotympanic is the major element con-
netic elements comprising the bulla is uncertain because we lack ju- tributing to the bulla, flooring the middle ear cavity. In amphicy-
venile skulls demonstrating the nature and number ofbulla elements. onids the bulla does not inflate to any degree during ontogeny, and
There is no doubt that an ectotympanic bone is present because it is in this respect is quite similar to bulla development in ursids. We can
preserved as a separate element loosely attached to the squamosal in characterize the bulla of amphicyonids and ursids as the plesiomor-
several skeletally plesiomorphic amphicyonids (Daphoenus, Cyn- phic arctoid type, in which the ectotympanic is able to nearly fully
odictis). However, discrete rostral and caudal entotympanics have encompass and cover the petrosal promontorium during develop-
not been identified in juveniles, and well-preserved basicrania of ment. Thus complete enclosure of the middle ear space requires
juvenile amphicyonids are uncommon. There is reason to believe, only the addition of bone at the periphery of the ectotympanic as
however, that a caudal entotympanic element was incorporated in the animal grows to adulthood. This peripheral addition is currently

Figure 11.2. Representative skele-

tal elements of North American
Miocene amphicyonids. A. Cranium,
Amphicyon, early Hemingfordian,
F Nebraska. B, C. Sexual dimorphism
in European mid-Miocene Amphicyon
upper canines and femora (redrawn
from Ginsburg, 1961). D, E. Hindfeet
(D) and forefeet (E) of amphicyonids
are paraxonic with the weight-bearing
axis passing between digits 3 and
4 (Daphoenodon, early Miocene,
Nebraska). F. Amphicyonid humeri
have a strongly developed deltopec-
toral crest and epicondylar ridges
permitting abduction-adduction and
weight-bearing of the upper forelimb
and strong extension/Hexion of the
lower limb and manus (Ysengrinia,
early Miocene, Nebraska). G. In
amphicyonids the radius retains the
ability to rotate around the ulna allow-
ing pronation-supination of the manus
(Daphoenodon, early Miocene,
Nebraska). H. Calcanea of large am-
phicyonids (Ysengrinia, Amphicyon,
Miocene, Nebraska) are similar to
c those of large living ursids, massive,
without elongation, suggesting an am-
G bulatory gait and short bursts of speed
J (temnocyonine and some Daphoen-
odon species evolve elongate slender
calcanea and limbs indicative of a cur-
sorial gait). J. Amphicyonid astragali
generally lack cursorial modifications
such as placement of the distal
condyle directly below a strongly
ridged trochlea (Daphoenodon, early Miocene, Nebraska). D through H modified from Peterson (1910). Scale bar in all drawings= 1 cm.
 Amphicyonidae 199

believed to occur by means of discrete entotympanic elements, as bosphenic character of the molars is maintained (parallel evolu-
demonstrated in living ursids (Hunt, 1974a, Pl. 4), and also probably tion of molars and camassial form in amphicyonids and canids is
includes a small amount of peripheral extension of the ectotympanic largely responsible for the erroneous attribution of amphicyonids
itself. The discovery of the John Day cranium is important in sug- to the Canidae in earlier studies). In many advanced amphicyo-
gesting that the assembly of the amphicyonid bulla closely follows nine beardogs the posterior molars are enlarged to produce broad
the ursid pattern. crushing platforms (Ml < M2 > M3). However, in contrast to living
In the amphicyonid basicranium, only a simple ectotympanic ursids, these platforms are not created by adding a talon to the upper
crescent is known in Eocene and Oligocene species. A more hemi- molars or by anteroposterior elongation and squaring of the molars.
spherical bony bulla that is the product of an ectotympanic with On the contrary, in amphicyonids with enlarged posterior molars
caudal entotympanic(s) appended to its medial and posterior edges the entire tooth is proportionately increased in size so that it appears
appears for the first time in the early Miocene. In Europe such a bull a as an enlarged replica of the original (M2 may become as large or
is first encountered in the amphicyonine Cynelos from St.-Gerand, larger than Ml, yet its typical plesiomorphic form is retained).
France (Hunt, 1977); in North America in the late Arikareean The shearing function of the camassial pair is maintained in am-
daphoenine Daphoenodon from the den site at Agate National Mon- phicyonids. This is true of even the large bearlike forms, hence
ument, Nebraska (Hunt, Xue, and Kaufman, 1983); and in the John contrasts with living ursids in which the shearing function of the
Day temnocyonine beardog from Haystack Valley, Oregon, pre- camassials has been lost. In large bearlike amphicyonids the car-
sumed to be of Arikareean age (Fingerut et al., 1993). Middle and nassials become heavy and massive, losing the slender, gracile form
late Miocene amphicyonid basicrania show that this low hemispher- of the carnivorous lineages, yet shear surfaces are still developed on
ical bulla is retained by the family and is modified in several lineages such teeth. Blunted cusps in older individuals, however, testify to
by the addition of an elongate bony external auditory meatus to pro- the processing of hard materials in their diet. The retracted proto-
duce a low, flask-shaped auditory bulla as in living and many extinct cone diagnostic of ursid camassials does not occur in amphicyonids
ursids. However, contrary to ursids, several amphicyonid lineages that maintain a typical, more anteriorly placed protocone (except in
develop an extension of the hypotympanic sinus in the floor of the some temnocyonines).
ectotympanic that invades the bony meatus in order to increase mid- Consequently, an amphicyonid can be identified using a com-
dle ear volume. Expansion of the auditory bulla by relative growth bination of basicranial structure and molar morphology. Beardogs
of the caudal entotympanic as seen in canids and most aeluroid are characterized by a diagnostic upper molar pattern in which the
camivorans never occurs in amphicyonids. Ml crown is basically triangular (not quadrate), with three principal
Skull form in amphicyonids departs very little from a basic plan cusps (paracone, metacone, protocone). The protocone basin does
in which a short snout is associated with a long postorbital seg- not open to the rear of the tooth and is closed by a postprotocrista;
ment (Fig. 11.2), a configuration that reflects the plesiomorphic arc- in ursids the molar crowns become quadrate, a metaconule is de-
toid skull form. Short-skulled (brachycephalic) amphicyonids are veloped to square the tooth crown, the protocone basin opens to
known but are the exception; most genera are long skulled (dolicho- the rear, and the postprotocrista is not developed. The migration of
cephalic ). When large body mass evolved in some lineages, the the metaconule to the posterointemal comer of the tooth seen in ur-
heads of these animals often became disproportionately enlarged. sids (Beaumont, 1982) and the posteriorly open protocone basin are
Consequently, the temporal muscle masses can be enormous, result- useful diagnostic features of ursid molars not seen in amphicyonids.
ing in exaggerated development of the sagittal crest in some species The talons or heels developed on the posterior edge ofM2 in ursine
(e.g., Amphicyon ingens). bears never occur in amphicyonids.
Temnocyonines are the only amphicyonids to possess enlarged
crushing premolars paralleling living hyaenids. No amphicyonids
DENTAL
develop hypersectorial camassial blades as in felids and hyaenids
The plesiomorphic dental formula of amphicyonids (13/3, in which metaconid and talonid are lost, but some temnocyonines
Cl/I, P4/4, M3/3) is the same as the basic eutherian formula and is and Ischyrocyon evolve shearing lower carnassials in which the
maintained in many amphicyonid lineages throughout their evolu- ml metaconid is lost and the talonid contains only a single tren-
tionary history, modified only by premolar loss in derived species chant cusp (hypoconid). However, in the temnocyonines such lower
and by loss of the M3 independently in several genera. M3 loss camassials accompany upper camassials in which the protocone
is characteristic of an entire subfamily, the Temnocyoninae, and is well developed, even enlarged (in contrast to felids), indicating
also characterizes the Ischyrocyon lineage as well as some species that these amphicyonids have developed a hybrid camassial func-
of Daphoenodon. Whereas large canines and incisors are retained tion combining both shearing and crushing in a manner unique to
in all amphicyonids, the premolars can be strongly reduced, par- Camivora.
ticularly Pl-3/pl-3. Such premolar reduction is typical of many
amphicyonines but not daphoenines or temnocyonines.
POSTCRANIAL
Molar modification in amphicyonids follows one of two direc-
tions. In daphoenine, temnocyonine, haplocyonine, and plesiomor- Although amphicyonids retain a generalized postcranial skeleton
phic amphicyonine lineages, the molars decrease in size from first in which modification of the limbs and feet for more cursorial
to third (M3 can be lost). This dental type parallels modem canids gaits is not developed to the extent seen in specialized runners
in which M2 is usually slightly smaller than Ml and the basic tri- such as canids, the family nevertheless includes several independent
 Robert M. Hunt, Jr. 200

experiments directed toward improved cursoriality. In North Amer- nature of the auditory bulla and petrosal, as well as the dentition.
ica, the two most prominent were the temnocyonines, which evolved Because amphicyonids are extinct, important biochemical traits and
an unusual cursorial adaptation of the forelimbs, and some species of diagnostic features of the soft anatomy (such as absence of caecum
Daphoenodon, in which the limbs became elongate and cursorially and bulbourethral glands of Cowper) cannot be determined. In keep-
adapted in a manner similar to large living canids such as wolves. ing with other arctoid camivorans, however, they possess a promi-
Cursorial amphicyonids are long-limbed, long-footed, digitigrade nent baculum in males, evidence of strong sexual dimorphism, and
animals but lack the extreme appression of the metapodials found the teres fossa on the ventral border of the scapula.
in living canids. In North America, arctoid camivorans appear in the late Eocene
The largest amphicyonids are in most cases scaled-up Miocene Duchesnean Land Mammal age where the first representatives are
descendants of plesiomorphic Oligocene species and display mas- amphicyonids belonging to the genus Daphoenus (Dawson, 1980).
sive bearlike skeletons characterized by heavier, robust limb ele- The oldest sampling of a diverse early arctoid assemblage in the
ments lacking cursorial adaptations (Fig. 11.2). These large fossil record occurs in the Quercy fissures of France (late Eocene-
beardogs are short footed and presumably plantigrade or at most Oligocene) where dental and cranial material record the early an-
subdigitigrade just as living ursines. In canids the ability to rotate cestry of amphicyonids, ursids, procyonids, and mustelids, but not
the radius around the ulna (supination-pronation) is greatly lim- canids. The oldest canids occur only in the late Eocene (Chadro-
ited by bone-joint configuration, and consequently the lower limb nian) and early Oligocene (Orellan) of North America, and this
is restricted in its motion to a fore-aft path. In amphicyonids the ro- geographic isolation of ancestral canids in the New World supports
tational capability of the radius and the ability to abduct/adduct the the recognition of a pre-Oligocene split within caniform carnivorans
massive upper forelimb are maintained; metapodials and digits are into canoid versus arctoid branches.
not strongly appressed. There is herein a parallel with ursids: The In Asia, the oldest arctoids are ( 1) amphicynodont ursids found in
ursine bears are commonly plantigrade, but the hemicyonines are ex- the Oligocene sediments of Inner Mongolia at Hsanda Go!, and (2)
ceptional among ursids in their digitigrade stance; similarly, a num- rare records of poorly preserved Oligocene amphicyonids. A pos-
ber of large amphicyonine amphicyonids are plantigrade, whereas sibly latest Eocene amphicyonid is known from the Bai-se Basin
temnocyonines and daphoenines are commonly digitigrade. The car- of south China, indicating that the family may be of equivalent
nivorous shearing dentition of hemicyonines logically accompanies antiquity in both Asia and Europe. Although there is as yet no evi-
digitigrady, and the crushing teeth of most ursines occurs with planti- dence that diversity of Asian arctoids approached that of Quercy in
grady. Among the amphicyonids there is a similar correlation: Shear- Europe, this may be the result of the lack of adequate sampling in
ing carnivorous dentitions are usually found in digitigrade forms, Asian sediments.
and enlarged crushing molars often accompany short-footed planti- The Oligocene arctoid record in North America can be explained
grade species. as the result of occasional immigration of Eurasian arctoid species
Five toes are present in fore- and hind feet; a paraxonic stance into the New World. These immigrants spring from a species-rich
(digits 3 and 4 of equal length) is the rule in digitigrade species. radiation of small arctoids that developed in Eurasia during the
Lower limb bones are never fused, always separate mobile ele- Oligocene and Miocene. It seems likely that Eurasia acts through
ments. Claws are nonretractile, but the intermediate phalanges of much of the Cenozoic as the principal staging ground of arctoid
amphicyonids are often moderately asymmetric. evolution, generating a diverse array of lineages, some of which
Pronounced sexual dimorphism in body size has been docu- migrate periodically into North America. Amphicyonids, however,
mented in amphicyonids. In North America, evidence oflarge males appear quite early in the fossil record of both Europe and North
and smaller females is known from samples of late Eocene America: Amphicyonid skulls with diagnostic basicrania and teeth
Brachyrhynchocyon dodgei, Oligocene Daphoenus vetus and occur both at Quercy in France and in the oldest levels of the White
hartshornianus, early Miocene Daphoenodon superbus, and large River Group of the Great Plains. Where the family originated is
species of middle Miocene Amphicyon, all from western Nebraska. uncertain, except to say that the Holarctic region must have been
Degree of dimorphism is known to increase with body size (Kurten, the spawning ground for amphicyonids as well as arctoids because
1969), so it is not surprising that the largest amphicyonids display these groups are unknown in the early Cenozoic of the southern
strong dimorphism in teeth and postcranial skeleton. Ginsburg and continents.
Telles-Antunes (1968) describe marked dimorphism in large Euro-
pean Amphicyon.
INFRAFAMILY
Amphicyonids radiated broadly on the northern continents from
SYSTEMATICS their first appearance in the late Eocene until the Miocene when
they attained their greatest diversity. This diversity is manifest par-
ticularly at the generic level (Figure 11.3). The grouping of genera
SUPRAFAMILY
into higher infrafamilial categories is at best a tenuous procedure.
Amphicyonids are members of the arctoid Carnivora. This place- This situation is due to the maintenance of a conservative skele-
ment is primarily supported by basicranial structure, particularly the tal anatomy in most amphicyonids, amounting to the lack of strong
 Amphicyonidae 201

w
TIME w <( AMPHICYONINAE TIME
8 Ma- <( z w r-~~~~~~~~_,/"-~~~~~~~~~~-
z z <(
-8 Ma

z z

II'
10- 0 -10
w >-
12- 0 (.) z MIGRATION
I 0 0 EVENTS z -12
a... z >- 0
14- <( >-
Cl ::? (.) .-? 5-~ (.)
0
-14

16- A w 0 <C-4- 0
:::>
I i-
_J
a...
w
(/) ?
-16

18-
l":l <(
a..
-18

20-

22-

24-
z
li
...
Io._
I <
I 0
z
0
>-
(.)
<(
::E
::E
z
0
>-
(.)
0
I

::!:
z
a:
(!l
z
w
z
0
>-
~
J:
Q._
::E
,_

(/)
0
...J
w
z
? -- - - - - _,
-20

-22

-24
0 <( z
(/)
<( >-
26- >- ::E
>- (.)
(.) ::E -26
0 w
t-

-1-
J: (/)
28- (.)
~ :::>
z
>- t- z
w
(/)
(/)
-·· -28
(.)
J: 0 Q._
30- a: z
>- w (/)
:::>
J:
Q._ ----· 0
z -30
J: 0 <(
32- (.) J: z 0
0
>-
<( Q._
w <( (.) -32
a: <( 0 a:
aJ J: <( 0
0 Q._ Q._ 0
34-

II
<( :::> -34
0 w
(/)
Q._
36- NORTH AMERICA -36
'
---·
38-
- .... - -...
~ -38
EUROPE
40Ma-
-40Ma

' ---... ---- - - ---,. ---- - --- ... --. -'

Figure 11.3. Hypothes.ized relationships of North . American and European Cenozoic Amphicyonidae with time ranges and migration events:
(1) 29.5 Ma, haplocyomnes. (2) 19.2 Ma, amph1cyonmes Cynelos, Ysengrinia. (3) 18-18.5 Ma, Amphicyon. (4) 15.5 Ma, Pseudocyon. (5) ?14-15 Ma
I schyrocyon. '

skeletal specializations that would distinguish and set apart such European haplocyonines, which have similar dental specializations
subgroups, and the seemingly endless parallelisms that evolve in (Fig. 11.3). The most plesiomorphic temnocyonine comes from the
dentition and other skeletal traits. This problem has been well John Day beds, Logan Butte, Oregon, and can be derived from Euro-
articulated by Beaumont (1984) in evaluating the European fos- pean haplocyonines. The European haplocyonine and North Ameri-
sil amphicyonids: "[Their] systematics is badly obscured due to the can temnocyonine radiations proceed independently, determined by
great monotony in the morphology of the fossil remains, the frag- regionally distinct dental patterns; however, the first temnocyonine
mentary nature of the material (numerous but often isolated teeth) to appear in North America is probably a late Oligocene Eurasian
which hinders the definition of the diverse species and their possible immigrant.
variability, and finally because of the great dispersion of the fossils Previous infrafamilial classifications of amphicyonids have
in collections throughout the world."
grouped the early North American genera Daphoenus, Daphoenic-
However, among amphicyonids, the North American temnocy- tis, and Brachyrhynchocyon (=Daphoenocyon) in a subfamily
onines (late Oligocene-early Miocene) can be identified from de- Daphoeninae and all later genera in the Amphicyoninae. These sub-
rived dental features and set apart as a subfamily. Recognized by tall, families are essentially temporal grades, not clades: Almost no well-
robust, well-spaced premolars, by loss of M3, and by lingually hy- founded anatomical evidence has been marshaled in support of them.
pertrophied upper molars with swollen lingual ~ingula that surround Daphoenines, however, can be redefined to create a monophyletic
a centrally placed protocone, temnocyonines are closely related to group with geographic integrity and some degree of morphological
 Robert M. Hunt, Jr. 202

identity. The similarity among the crania of North American fissures, but a difference in geologic age may have been obscured
daphoenines suggests they represent a local radiation. None of these by combining samples from fissures of different ages in museum
genera appear in the Old World. Morphological attributes that ally collections under the single designation "Quercy."
the group are few because plesiomorphic features are retained and Consequently, it is possible that a Eurasian amphicyonine am-
simply scaled upward in size. Nevertheless, some features unite the phicyonid radiation stems from early species of Cynodictis and
daphoenines: (1) Premolars do not undergo significant reduction as that North American daphoenine amphicyonids are derived from
they do in many amphicyonines; (2) in daphoenines p4 tends to early forms of Daphoenus. No immigrant taxa entering Eurasia from
elongation, with a squared posterior border, but in amphicyonines is North America can be presently identified. However, in North Amer-
shorter on average with a more rounded posterior border; (3) digiti- ica several Old World amphicyonine genera (Cynelos, Ysengrinia,
grady is common in daphoenines, less so in amphicyonines; (4) Ml Amphicyon, Pseudocyon) appear abruptly in the Miocene record.
tends to be longer along its labial margin in early daphoenines rela- There are no plausible ancestral forms on this continent for these
tive to early amphicyonines (compare Daphoenus with Cynodictis). beardogs. These must be immigrants from Eurasia and are unlikely
Beyond these morphological features, the most important evi- to be derived from plesiomorphic North American genera such as
dence supporting the hypothesis of amphicyonid evolution adopted Daphoenus because plausible ancestral species already occur at an
here is the geographic distribution of the fossils. The pattern of earlier time in Europe.
genera on the northern continents is arresting. The restriction of
daphoenines to North America and confinement of early amphicy-
onines to the Old World, followed by the sudden appearance of Old INCLUDED GENERA IN THE
World amphicyonine genera in the Miocene of North America, is FAMILY AMPHICYONIDAE
interpreted as the principal support for the hypothesis. The early
basal or stem amphicyonid stock probably underwent geographic The locality numbers listed for each genus refer to the localities in
separation in the late Eocene and early Oligocene, and subsequent Appendix I (for a more detailed listing of localities, see the Key
migrations to the New World only became possible in the late to Fossil Localities at the end of this chapter). The acronyms for
Oligocene and at various times in the Miocene when continental museum collections are listed in Appendix III. Because the species-
connections and climatic factors became favorable. Why the ex- level taxonomy of amphicyonids is in revision at the present time,
change of species was unidirectional is unexplained, but the fossil localities usually are given for genera rather than species.
evidence of these immigration events is convincing and unequiv-
ocal. Independent tests of this hypothesis will occur as the Asian
DAPHOENINAE
fossil record becomes better established. Currently, Asian amphicy-
onids have affinities with those of Europe and can be provisionally Characteristics: Small ( <25 kg) to large (100 kg) arctoid carnivo-
classified as amphicyonines. rans with plesiomorphic auditory bulla formed by an ossified ecto-
Among the daphoenines, Brachyrhynchocyon and Daphoenictis tympanic element with probable cartilaginous entotympanic plate
are both specialized taxa without descendants, the former a short- attached by its ventral margin to the medial rim of ectotympanic
faced, broad-skulled carnivoran that probably was in part duro- and by its dorsal margin to the petrosal promontorium (this is the
phagous, the latter a hypercarnivorous, possibly catlike animal most plesiomorphic bulla yet known among amphicyonids and is
known from a single species. Daphoenus, in contrast, is a plesiomor- also found in European Cynodictis); a more advanced bulla occurs
phic genus with multiple species represented by large samples. It in Daphoenodon in which the middle ear is enclosed by a rudimen-
is plausibly ancestral to Daphoenodon of the North American early tary ectotympanic bulla that may include ossified entotympanics;
Miocene. The small Paradaphoenus is regarded as an early offshoot the bulla is unknown in Daphoenictis and Brachyrhynchocyon but
of Daphoenus. These five genera constitute the members of a revised was probably similar to the Daphoenus bulla, based on impres-
subfamily Daphoeninae. sions made by the bulla on the basicranium. Skull dolichocephalic
The closest morphological analogues to Daphoenus in dentition in Daphoenus and probably in Daphoenictis but brachycephalic in
and cranial features are species of Cynodictis from the Quercy fis- Brachyrhynchocyon and Daphoenodon; snout relatively short in all
sures of France. Skulls of Cynodictis from Quercy represent the genera (considered a plesiomorphic arctoid trait). Dental formula
oldest European amphicyonid crania, and these show the embayed 13/3, Cl/l, P4/4, M3/3 for Daphoenus and Brachyrhynchocyon but
basioccipital bone and characteristic petrosal, ectotympanic, and molarnumber uncertain for Daphoenictis. M2-3/3 in Daphoenodon
molar structure seen in North American Daphoenus. The highly in which the youngest species lose M3.
specialized Brachyrhynchocyon and Daphoenictis are unknown in The Daphoeninae is considered here as a monophyletic North
Eurasia and are considered North American endemics. Similarly, American endemic subfamily and has been diagnosed in earlier
in Europe specialized endemic amphicyonine amphicyonids occur studies (Hunt, 1974b, p. 1032): M2-3 relative to Ml not enlarged in
with Cynodictis in the Quercy fissures. The age of the larger, more contrast to amphicyonines in which M2-3 are enlarged crush-
advanced beardogs remains in question because some Quercy fis- ing teeth with amplified surface area; no reduction of premolars;
sures appear to contain mammals as young as late Stampian or Aqui- p4 unreduced, often elongate, with squared posterior border; au-
tanian. Some of the more derived larger Quercy amphicyonids may ditory bulla preserved only as an ossified ectotympanic crescent,
in fact be descendants of more plesiomorphic taxa also found in the loosely attached to the skull, without addition of any ossified
 Amphicyonidae 203

entotympanic elements and without lateral prolongation into a bony Daphoenines occur in abundance only in the White River Group
external auditory meatus (the Daphoenodon bulla is evolved be- of the central Great Plains and northern Rocky Mountains, but a
yond the plesiomorphic daphoenine bulla in enclosing the middle small sample derives from the John Day beds of Oregon, and rare
ear by means of a hemispherical bony capsule that may include ento- individuals are known from the Tecuya beds of California, the Vieja
tympanic elements); lack of expansion of the bulla posterior to the Group of Texas, the Cypress Hills of Saskatchewan, and the I-75
mastoid process; inferior petrosal venous sinus deeply excavated locality in northern Florida. They are unknown in Europe despite
into edge of basioccipital; medial edge of petrosal in only slight a good amphicyonid record. In Asia, the only known amphicyonid
contact with margin ofbasioccipital, not sutured to the basioccipital fossils of Oligocene age are a few rare fragmentary discoveries
as seen in canids. from mid-Oligocene sites attributed to ?Cynodictis and Amphicyon
Hough (1948) also diagnosed the subfamily (placed by her at (Russell and Zhai, 1987). This material is too poorly known to war-
family rank) using plesiomorphic features: generalized "canid" (ac- rant comment. Thus, although late Eocene-Oligocene amphicyonids
tually plesiomorphic caniform) dentition; "feline" (=unspecialized) are Holarctic in distribution, little is known about Asian links be-
postcranial skeleton; short facial region of skull yet cranium elon- tween North America and Europe; the nature of early Asian amph-
gate; presence ofM3; lack of accessory cusps on anterior premolars. icyonids and potential migration between the Old and New World
Or she employed traits that depend on body size or sex differences, prior to the Miocene and the center(s) of origin of the family are not
such as development of the sagittal crest and relative size of the clarified by Asian evidence. However, the marked endemism of Eu-
braincase. None of Hough's characters are exclusive todaphoenines, ropean and North American late Eocene-Oligocene amphicyonids
either occurring in amphicyonines or lacking clear definition. argues for absence of migration between the two geographic regions.
Morphology of the teeth indicates that Daphoenus and Brac-
hyrhynchocyon are closely related; teeth of Daphoenictis appear less Daphoenictis Hunt, 1974b
like those of the former two genera, but basicranial anatomy, espe- Type species: Daphoenictis tedfordi Hunt, 1974b.
cially the embayed basioccipital, suggests relationship. In Type specimen: NMC 9205.
Daphoenus and Brachyrhynchocyon the Ml is triangular to sub- Characteristics: Small amphicyonid carnivorans with skull
quadrate with an expanded lingual cingulum posterointernal to the lengths of about 10-12 cm. Dental formula I?/?, Cl/l,
protocone. Three principal cusps (protocone, paracone, metacone) P?4/4, M?2/3. Recognized by their distinctive dentition
form the apices of the triangle, and both paraconule and metaconule (postcranial skeleton unknown), which differs from the
are present as small wedge-shaped cuspules at the bases of the para- species of Daphoenus and Brachyrhynchocyon in the fol-
cone and metacone, respectively. The paraconule and metaconule lowing features: p2 with prominent posterior accessory
are situated at the labial terminations of the protocristae and are cusp; p3 taller yet anteroposteriorly shorter than p4; pos-
incorporated in them. Because the protocone is placed close to the terior accessory cusp on p3 well developed; prominent
anterior edge of the Ml, the postprotocrista is longer than the pre- anterior and posterior cingulum cusps on p2-4; premo-
protocrista. A weak constriction in the postprotocrista lingual to lars bladelike, narrow, with accessory and cingulum cusps
the metaconule may segregate a cusplike segment of the postpro- producing a leafl.ike appearance in lateral view; ml meta-
tocrista just posterobuccal to the protocone. The protocone basin conid greatly reduced, paraconid advanced, trigonid much
of the Daphoenictis Ml, a more derived genus, opens to the rear longer than talonid, talonid reduced, with single centrally
and is not closed by a postprotocrista; this is a specialization of the placed knoblike cusp; m2 with prominent protoconid, re-
sectorial dentition in this carnivore. duced paraconid, metaconid absent, talonid with single
M2 in these daphoenines is smaller than Ml, subquadrate, with centrally placed cusp, talonid reduced relative to trigonid;
less surface relief. The expression of the para- and metaconules as molars more adapted for shearing than in Daphoenus and
well as the protocone is subdued relative to Ml: The conules in both Brachyrhynchocyon.
genera are weakly developed or absent, although in Brachyrhyn- The genus is known from (1) the teeth, either isolated
chocyon there are very few specimens with an unworn M2. M2 is or preserved in jaw fragments, and (2) a posterior cranium
unknown as yet in Daphoenictis but was probably reduced relative including the auditory region.
to Ml. P4 is a shearing tooth with a prominent protocone and short Lower carnassial length range: D. tedfordi, 14.2-16.6 mm
metastylar blade. In the lower carnassial the paraconid blade is not (N = 6).
advanced as it is in canids and amphicyonines; hence the trigonid Included species: D. tedfordi (known from localities SB44B,
cusps are more closely grouped. Premolars (except p4) lack de- CP39B, CP39C, CP43, CP98B or C, NPIOB, NP24C).
veloped posterior accessory cusps but are surrounded by distinct
cingula. Brachyrhynchocyon Loomis, 1936 (in Scott and
Old World amphicyonid fossils from Quercy attributed to Cyn- Jepsen, 1936)
odictis are similar in skull morphology and dentition to the North Type species: Brachyrhynchocyon dodgei (Scott, 1898) (=
American Daphoenus, suggesting a common origin for amphicy- Daphoenocyon ).
onines and daphoenines in the Eocene. Daphoenictis and Brachy- Type specimen: PU 11422.
rhynchocyon are too specialized to give rise to any of the later amph- Characteristics: Small amphicyonid carnivorans with skull
icyonids, and both lineages terminate at the end of the Chadronian. lengths of about 14-17 cm. Skull form exhibits strong
 Robert M. Hunt, Jr. 204

dimorphism, interpreted as sexual: males broad skulled, with skull lengths of about 14-16 cm, 17-20 cm, and 24
massive; female skulls narrower, more gracile. Dental for- cm, respectively. Skull form and canine teeth dimorphic,
mula 13/3, Cl/l, P4/4, M3/3. Recognized by brachycepha- interpreted as sexual: males larger, with robust rostrum
lic skull and distinctive dentition that differs from the and canines; females smaller, with more gracile rostrum
species of Daphoenus and Daphoenictis in having an- and canines. Dental formula 13/3, Cl/l, P4/4, M3/3. Rec-
teroposteriorly short, wide, robust premolars relative to ognized by dolichocephalic skull and distinctive dentition
elongate narrow premolars in the latter two genera. In with anteroposteriorly elongate, narrow premolars (espe-
bothDaphoenus and Brachyrhynchocyon the premolars in- cially P2-3/p2-3) that differ from the brachycephalic skull
crease in size from pl to p4, and posterior accessory cusps and short, wide, robust premolars of Brachyrhynchocyon.
are developed on p2-4; the posterior accessory cusp of In both Daphoenus and Brachyrhynchocyon the premo-
p4 is always prominent, progressively diminishing in size lars increase in size from pl to p4, and posterior accessory
in the anterior direction on p2-3 (in both genera, posterior cusps are developed on p2-4; the posterior accessory cusp
accessory cusps on p2-3 may be weak to absent in some in- of p4 is always prominent, progressively diminishing in
dividuals). Also, in Brachyrhynchocyon the principal cusp size in the anterior direction on p2-3 (in both genera, pos-
of p2-4 is more anteriorly placed than in Daphoenus; that terior accessory cusps on p2-3 may be weak to absent
is, the short Brachyrhynchocyon premolars are evolved by in some individuals). Also, in Daphoenus the principal
apparent reduction of the anterior part of the premolar such cusp of p2-4 is centrally placed, not anteriorly placed as
that the principal cusp lies near the front of the tooth. in Brachyrhynchocyon. In Daphoenictis the p3 is taller
Daphoenictis differs from Daphoenus and Brachyrhyn- than p4, and p2-4 have much more elaborately developed
chocyon in its p2-4 with prominent posterior accessory posterior accessory and cingulum cusps.
cusps and p3 taller yet anteroposteriorly shorter than p4; The lower molars emphasize the external cusps relative to the
in the latter two genera, p3 is never as tall as p4. internal: The ml-2 protoconid and hypoconid are domi-
The ml metaconid remains prominent and unreduced in nant. The ml metaconid is prominent, not reduced or re-
Brachyrhynchocyon, and the paraconid blade is not ad- tracted relative to the protoconid; the paraconid blade is not
vanced. The ml talonid has a prominent hypoconid ridge advanced (contemporaneous canids have an advanced ml
that dominates the heel; from this ridge the talonid surface paraconid). The ml talonid is basined (no basin in Daphoe-
slopes downward to the lingual border without develop- nictis and only a weak to moderately developed basin in
ment of a strongly basined talonid as seen in Daphoenus. Brachyrhynchocyon): A prominent hypoconid ridge dom-
The m2 protoconid and metaconid are equally developed inates the heel; the internal face of this ridge slopes down-
(in Daphoenus the protoconid is larger than metaconid), ward to a basin bordered internally by a rim that includes
the paraconid low; the hypoconid is expressed as a low the entoconid and adjacent cuspules. The m2 protoconid
ridge sloping internally to a fl.at talonid surface; there is is usually more developed than the metaconid (in some
no entoconid. The talonid is lower than (and the same size individuals these cusps are nearly equally developed) and
or somewhat smaller than) the trigonid. Thus m2 differs the paraconid low; the anterobuccal comer of m2 is en-
only in detail from that of Daphoenus but is much different larged and the posterior border squared, giving m2 a more
from the derived m2 of Daphoenictis. quadrate appearance than seen in Brachyrhynchocyon.
The genus is known from fossil remains of (1) 5 to 7 As in the case of ml, the m2 hypoconid is expressed
skulls, some with associated lower jaws; (2) numerous as a low ridge sloping internally to a basined talonid;
isolated mandibles and maxillae belonging to 40-50 in- there is an internal enamel rim formed by a weak ento-
dividuals; and (3) a few postcranial skeletons, indicating a conid; the talonid is lower than the trigonid. Thus m2 of
robust, muscular, wolverinelike camivoran. Daphoenus differs only in detail from that of Brachyrhyn-
Lower camassial length range: B. dodgei, ~13-17.0 mm chocyon but is much different from the derived m2 of
(N = 32); B. montanus, 11.9 mm (N = 1). Daphoenictis.
Included species: B. dodgei (known from localities SB44C, The genus is known from fossil remains of ( 1) over sixty
CP39B,CP39C,CP39~CP39G,CP40A,CP41A,CP42A, skulls, many with associated lower jaws; (2) numerous
CP68B, CP83A, CP83B, CP83C, CP98B or C, CP98C, rostra, isolated mandibles, and maxillae; and (3) several
NPlOB, NP29C, NP23C, NP24C, NP24E); B. montanus postcranial skeletons, indicating in the North American
(locality NP24C). midcontinent (White River Group) a large and small species
of digitigrade, gracile, subcursorial arctoid camivoran.
Daphoenus Leidy, 1853 This is the most common genus of North American
Type species: Daphoenus vetus Leidy, 1853. amphicyonid.
Type specimen: USNM 658-659. Lower camassial length range: Daphoenus (Whitneyan-early
Characteristics: Very small (D. hartshornianus) to small (D. Arikareean, Great Plains), 15.4-18.1 mm (N=3);
vetus) to mid-sized (D. n. sp.) amphicyonid carnivorans (Whitneyan-early Arikareean, John Day), 14.4-18.5 mm
Amphicyonidae 205

(N = 5); (Orellan), 13-17.1 mm (N = 81); (Chadronian-

Plains sample, the two John Day skull morphs are not male
Duchesnean), 10.9-15.4 mm (N = 14).
and female but more probably two species.
Included species: D. vetus, D. hartshornianus, D. lambei,
Unfortunately, the specimens of John Day Daphoenus
D. transversus, D. socialis, D. n. sp. Daphoenus is known
were primarily collected in the nineteenth century with
from the following localities: Duchesnean (SB44B,
only minimal data as to their placement in the John Day
CP29D, CP39IIA, NP9); Chadronian (CP39D, CP41A,
succession. As a result we know almost nothing of the rel-
CP83B, CP83C, CP98C, CP98B or C, NPlOB); Orellan
ative stratigraphic position of the fossils of the two species.
(CP40B, CP41B, CP68C, CP84A, CP99A, NP29E,
However, recent collections by Ted Fremd and his asso-
NP50B); Whitneyan localities (CP84B, CP99B); Whit-
ciates at John Day National Monument have produced
neyan or Arikareean (CC13, CPlOOII, PN6, PN6C,
teeth of Daphoenus that are much younger than the up-
PN6D).
per range limit of the Great Plains Daphoenus vetus, and
suggest that the John Day Daphoenus persists as a relict
Comments: The Great Plains sample of Daphoenus is almost lineage in the Pacific Northwest.
entirely derived from the White River Group of Nebraska, The National Park Service John Day collections also
South Dakota, Wyoming, and Colorado. Sample size is demonstrate the existence of a third very large species
sufficient to recognize a small D. hartshornianus Cope of Daphoenus in the John Day beds low in the strati-
and a larger D. vetus Leidy. Definition of these species graphic section. In Fremd 's Unit C, over 100 m below the
is greatly improved by the larger samples available after Picture Gorge ignimbrite, a maxilla of Daphoenus with
decades of collecting by various institutions in the White unusually large teeth was found in Blue Basin at John
River beds. In particular, the nature of sexual dimorphism Day National Monument. This fossil (JODA 1411) estab-
in amphicyonids has been clarified on the basis of these lishes this large species of Daphoenus low in the John
fossils.
Day sequence; it is equivalent in size to the largest known
In the John Day beds of Oregon there are two identifi- species of Daphoenus from the Great Plains dated at about
able species of Daphoenus represented by adequate ma- 28.6 Ma. Thus three species of Daphoenus occur in the
terial: D. transversus, a small species, and D. socialis, a
John Day beds, and future discoveries may establish their
somewhat larger animal. The first-named is the size of stratigraphic range more accurately.
the Great Plains Daphoenus hartshornianus Cope; in fact, The continuation of the Daphoenus vetus lineage into
Cope assigned the holotype rostrum toD. hartshornianus, the Whitneyan of the Great Plains is evidenced by several
but in 1899 Wortman and Matthew made the rostrum the mandibles in the Protoceras channels of South Dakota.
type of "Paradaphoenus" transversus. This animal is not These indicate a form larger than Orellan vetus but smaller
part of the Paradaphoenus lineage that comprises much than the terminal Great Plains species ofDaphoenus known
smaller carnivorans (see Paradaphoenus, this chapter). It from a single enormous cranium in the uppermost White
is a hartshornianus-sized Daphoenus but differs from the River Group of Wildcat Ridge, western Nebraska, dated at
Great Plains species in a more sectorial dentition that mim- about 28.6 Ma.
ics Oligocene canids. This dentition can be recognized by There is also evidence of the small D. hartshornianus
a more advanced paraconid blade and slightly retracted lineage in the Whitneyan of the Great Plains. Two spec-
metaconid of ml, and by the more abbreviated protocone imens, a maxilla with Ml-2, P4, and alveoli for C-P3
of p4. There also appears to be more of a difference in (LACM 16933) and a maxillary fragment with P4-Ml
m2 trigonid versus talonid height. A second larger species (AMNH 1195b), represent the hartshornianus lineage in
of Daphoenus includes the holotype mandibles of "Peri- the Whitneyan of South Dakota. A fragmentary mandible
cyon" socialis Thorpe from the Haystack Valley area of the with ml, p2, p4 (AMNH 85880) from the Upper Brule of
John Day beds. Several additional fragmentary specimens Nebraska may represent this same species.
and isolated teeth can be assigned to this D. vetus-sized The vetus lineage ranges from about 35-36 Ma to
beardog. To this hypodigm I also refer a partial cranium 28.6 Ma as currently recognized in the Great Plains White
(YPM 10064) with an inflated frontal region and broad River Group. The small hartshornianus ranges less cer-
snout similar to D. vetus but differing sufficiently in dental tainly from 35-36 Ma to perhaps 30-32 Ma in the Great
traits to warrant recognition as a distinct morphospecies, Plains. Both species are probably descended from the small
D. socialis. Thus two skull morphs of Daphoenus in the Duchesnean species and therefore may extend to about
John Day region correspond to the two morphs found in 40 Ma. Thus, by combining the dating of Great Plains and
the Great Plains: D. socialis (YPM 10064) has an inflated John Day fossils, the range of Daphoenus in North Amer-
frontal region and broad snout, whereas D. transversus ica extends from about 40 Ma to about 27 Ma.
(AMNH 6851, the holotype of Wortman and Matthew)
has a more gracile skull with less frontal inflation and a Paradaphoenus Wortman and Matthew, 1899
narrower snout. Based on comparison with the larger Great Type species: Paradaphoenus cuspigerus (Cope, 1878).
Robert M. Hunt, Jr. 206

Type specimen: AMNH 6852. the John Day beds of Oregon (AMNH 6852, 6853), one
Characteristics: Very small amphicyonid camivorans with with associated mandibles, originally described by Cope in
skull lengths ofabout 9-11 cm. Dental formula 13/?3, C 1/1, 1878, both with well-preserved basicrania. AMNH 6852
P4/4, M3/3. Very little change in body size and skull pro- is Cape's type of "Canis" cuspigerus, which was made
portions from first appearance of the genus in the Orellan to the genoholotypic species of Paradaphoenus by Wortman
the most advanced species from the John Day beds of Ore- and Matthew (1899). I use the species minimus for the
gon (Whitneyan or early Arikareean) and Arikaree Group Great Plains hypodigm, distinguishing it for the present
of the Great Plains (early Arikareean). The general form of from the John Day species cuspigerus by the more en-
the premolars is much like that of Daphoenus but smaller in larged M2/m2 in the John Day form. No postcranials are
size. The lower premolars are laterally compressed, blade- associated with this cranial and dental material.
like, and are known in only two mandibles (UNSM 26130, Lower camassial length range: P. minimus, 8.3-9.2 mm (N
AMNH 6852); they become slightly more robust in the = 6), P. cuspigerus, 9.3 mm (N = 1).
John Day species. A prominent posterior accessory cusp Included species: P. cuspigerus (known from localities in
is developed only on p4, may be very weakly developed on PN6), P. minimus (localities CP84A, CP85, CC99A,
p3, and is absent on pl-2 in the two available mandibles. In CP99A or B, CP99B or C).
P. minimus, the oldest and most plesiomorphic species, the
ml is distinguished from early contemporaneous canids by Comments: The presence or absence of an M3 has been a
the closed trigonid in which the paraconid is only slightly persistent problem with this genus and led Cope initially
advanced. The ml talonid is basined, the hypoconid and to place the two John Day skulls in different genera. In
entoconid forming low ridges bordering the basin. The m2 fact, we can determine that both John Day skulls retain a
is plesiomorphic in Orellan P. minimus in which the trigo- small M3. The tooth itself is preserved in AMNH 6853,
nid and talonid are of equal size. The protoconid, meta- but only a small vestige of the M3 alveolus remains in
conid, and paraconid form a small equilateral triangle; the the genoholotype skull (AMNH 6852). In Great Plains P.
talonid surface is slightly lower than the trigonid and is minimus there is no M3; however, on one side a vestige of
basined, bordered by low ridges formed by the hypoconid an alveolus may be present.
and entoconid. In the John Day P. cuspigerus the enlarged Basicranial structure is distinct from contemporaneous
M2 causes the m2 trigonid to be crowded to the front of the canids. Canids develop a narrow basioccipital bone and a
tooth and the basined talonid is exaggerated, forming most large hemispherical auditory bulla well fused to surround-
of the occlusal surface. This type of m2 is derived from the ing skull bones. InParadaphoenus the basioccipital is wide
plesiomorphic m2 of P. minimus in which trigonid and and unconstricted because there is no development of an
talonid are the same size. A key feature of m2 is its thick- enlarged inflated bulla. In the John Day skulls an audi-
ened anterobuccal cingulum that forms a protuberant cor- tory bulla is preserved and demonstrates that it is a small,
ner on the tooth, similar to the same feature in Daphoenus. slightly inflated ectotympanic element (it is doubtful if
Unworn upper molars are preserved in the holotype any caudal entotympanic element contributed to the bulla).
skull of P. minimus: Ml is very similar in form to Ml The petrosal is very similar in form and placement to that
of Daphoenus. Para- and metacone are placed far laterad; of Daphoenus. My comparison of Paradaphoenus crania
hence there is no stylar shelf, although a small parastylar with Quercy Cynodictis reveals that both show the amphi-
region is present. Both para- and metaconules are present, cyonid pattern, but dental features of Paradaphoenus differ
as is the small cuspule anterointemal to the metaconule that from all Cynodictis known to me in details of P4, Ml, and
develops on the postprotocrista (also in Daphoenus). The lower molar form, suggesting that Paradaphoenus proba-
V-shaped protocone is surrounded by a prominent postero- bly evolved in North America.
medially thickened lingual cingulum. M2 is a miniature of
Ml even to the presence of tiny para- and metaconules, Daphoenodon Peterson, 1909
and the para- and metacone are similarly placed far toward Type species: D. superbus (Peterson, 1907).
the buccal margin. The upper camassial shows a reduced Type specimen: CM 1589.
protocone that is a hallmark of the genus; P4 is short and Characteristics: Mid-sized to large amphicyonids with skull
robust in both species, not elongate and bladelike, and is lengths of about 23-27 cm in the oldest species ranging to
always surrounded by a sharply defined cingulum. The about 32-34 cm in the youngest. Dental formula 13/3, C l/l,
upper premolars lack posterior accessory cusps. P4/4, M3/3 in oldest species, reduced to M2/3 in youngest.
The genus is known from fossil remains from two geo- In adequately sampled species, dimorphism is evident. In
graphic areas: (1) six mandibles and a maxilla from the the sample of D. superbus from the Agate National Mon-
Orellan, Whitneyan, and early Arikareean of the Great ument carnivore den site, the males are significantly larger
Plains, including a skull (AMNH 39099) from the Orel- in dental and postcranial dimensions (the large Daphoen-
lan of South Dakota that Hough (1948) made the holo- odon periculosis Cook is simply a male of D. superbus).
type of "Daphoenus" minimus and (2) two skulls from The premolars of Daphoenodon never become reduced
 Amphicyonidae 207

as they do in many amphicyonine lineages. The hallmark D. robustum (lower Runningwater Formation), 28.6-29.0
of the genus is its wolflike dentition, progressive size in- mm (N = 2); D. n.sp. III (upper Runningwater Formation),
crease, and development of cursorial postcranial traits. It 29.4-33.4 mm (N = 11). Daphoenodon is known from lo-
is the least "bearlike" of the Miocene amphicyonids, and calities GC8A, GC8B, GC8C, CP51A, CP52, CP104A,
its terminal species must have been a formidable preda- CP!04B, CP105.
tor best described as an enormous parallel to the modem
timber wolf. Comments: Daphoenodon differs in its broader skull form
Although premolars are retained inDaphoenodon, these from the youngest species of Daphoenus, but whether
camivorans do not develop enlarged or broadened premo- Daphoenodon is a direct descendant, hence a North
lars for durophagy as in hyaenids. Premolars progressively American endemic, is still in doubt. We currently lack
increase in size from anterior to posterior but retain "nor- fossils linking the last known occurrence of Great Plains
mal" proportions; p2 and p3 are of lesser height than p4, Daphoenus (at ~28.5 Ma) and the first record of Daphoen-
not taller as in many temnocyonines. The premolars show odon (at ~22 Ma), an interval of 6.5 million years. The
a prominent posterior accessory cusp only on p4. teeth of Daphoenodon remain basically plesiomorphic as
The ml metaconid remains prominent and is never lost, the genus increases in size and dental dimensions through-
but may be somewhat reduced in size and is slightly re- out its early Miocene range (it serves as an index taxon for
tracted relative to the protoconid. The paraconid blade the North American early Miocene, ~ 17-24 Ma).
of ml is advanced; hence the P4 protocone is forwardly
placed to make room for the blade. The ml talonid has
a prominent hypoconid ridge that dominates the heel; the TEMNOCYO NINAE
hypoconid ridge is laterally placed on the talonid, and in- Characteristics: Small to moderately large arctoid camivorans with
ternal to the hypoconid is a narrow lingual shelf lacking very small, rudimentary auditory bulla formed entirely by the ec-
an entoconid. However, in a large sample from a restricted totympanic, or with plesiomorphic Type A bulla formed by ecto-
stratigraphic level (the Agate beardog den site and Amer- tympanic and bony entotympanics as in living ursids; skull dolicho-
ican Museum-Cook Quarry), we find some exceptional cephalic, frontal region often inflated. Dental formula 13/3, Cl/l,
individuals with a basined ml talonid produced by eleva- P4/4, M2/3. Ml is particularly diagnostic of the subfamily: In its
tion of a low entoconid ridge, demonstrating that the ml fully evolved form it displays a conical protocone isolated at the
talonid configuration varies and should not be too rigidly center of the lingual half of the tooth, in essence a knoblike cusp
interpreted. centered in a flat expanse of enamel. Surrounding this isolated pro-
The m2 protoconid and metaconid are both developed tocone is a strongly thickened lingual cingulum, so enlarged that
(the protoconid is larger than metaconid). The paraconid the tooth in occlusal view takes on a figure-eight shape, constricted
is a low cusp at the anterointemal comer of m2; it is con- at its midpoint, its buccal part the para-metacone region, its lingual
nected to the protoconid by a C-shaped enamel rim on the part the protocone region with prominently inflated cingulum.
anterior margin of the m2 trigonid. The hypoconid is ex- Temnocyonines (and their sister subfamily in Europe, Haplocy-
pressed as a low ridge sloping internally to a flat talonid oninae) possess tall, robust premolars; this is particularly true of
surface; there is no entoconid. The talonid is lower than p2-4 and P2-3, which in older individuals often show strong, api-
(and the same size or somewhat smaller than) the trigonid. cal wear similar to that seen in living durophagous hyaenids in which
Thus m2 form is very similar to m2 form in Daphoenus. the principal premolar cusps are worn to flat platforms by contact
The prominence of the hypoconids of the lower molars and with hard materials.
the absence of entoconids illustrates the importance of a The camassial pair remains functionally prominent throughout
mortar-pestle crushing mechanism confined to the molars the history of the subfamily and is always flanked by a large Ml and
whereby the prominent hypoconids occlude within deep smaller M2 in the maxilla (M3 is always lost) and by a prominent
protocone basins of the upper molars. This relation is typ- m2 and smaller m3 in the mandible. The trigonid of m 1 is strongly
ical of the Daphoenodon lineage. developed, becoming robust in derived species, and the ml talonid
The genus is known from (1) 8 skulls, all but one with is occupied by an enlarged hypoconid (there is no entoconid or
associated lower jaws; (2) many mandibles, maxillae, iso- basined talonid). In some species the m 1 metaconid is lost, and this
lated teeth, and postcranial bones belonging to over 45 in- feature is important in the recognition of certain lineages. The buccal
dividuals; and (3) a few entire skeletons, the most complete cusps of ml-2 are always emphasized: The strong ml hypoconid
the genoholotype female and a juvenile male, both found is followed in series by a developed m2 protoconid and hypoconid,
in the Agate camivoran den site by 0. A. Peterson in 1905. forming a line of three prominent crushing cusps on ml-2. Some
Lower camassial length range: 22.7-33.4 mm (N = 35). species possess an m2 metaconid but never an entoconid.
Included species: D. n.sp. I (Harrison Formation), lower car- An important derivative feature of the group is a digitigrade
nassial length 22.7 mm (N = l); D. superbus (basal Up- stance in which the radius and ulna become modified for restricted
per Harrison beds), 24.2-27.4 mm (N = 9); D. n.sp. II fore-aft motion, and both metatarsals and metacarpals are elongate.
(higher Upper Harrison beds), 25.8-30.6 mm (N = 6); The major weight-bearing axis passes between digits 3 and 4; hence
 Robert M. Hunt, Jr. 208

fore- and hindfeet are functionally paraxonic. The radius-ulna pair is Haplocyon from France show remarkable similarity to the oldest
distinguished by an unusual bony stop mechanism in the few skele- North American Temnocyon from Logan Butte, Oregon, yet are
tons associated with dentitions. This stop limits the movement of the somewhat more plesiomorphic in molar pattern, suggesting that
radius around the ulna, preventing any marked degree of pronation- Haplocyon migrated to North America and established the tem-
supination. Furthermore, the form of the distal humerus indicates nocyonine lineage in the late Oligocene. Despite the evident dental
that the forelimb was not everted at the elbow as in the large bear- similarity between these subfamilies, my inspection of European
like amphicyonids but was held in a more vertical alignment as in fossils indicates that haplocyonines evolved in isolation in Europe,
living cursorial canids. These features produce a more restricted temnocyonines in North America, after sharing an initial common
fore-aft movement of the forelimb, suggesting that temnocyonines ancestry in the Old World.
were cursorially adapted. Digitigrady probably evolved once in the Temnocyonines became extinct by about 19 Ma in North America,
ancestral temnocyonine and may have been an important adaptive the youngest known individuals occurring in the uppermost Arika-
feature contributing to their late Oligocene-early Miocene North ree Group in the later early Miocene (they are known in the late
American radiation. Arikareean but are absent from Hemingfordian faunas). In Europe
The early Miocene temnocyonine radiation includes the first am- the haplocyonines are reported to persist a much longer time until
phicyonids to attain moderate to large body size. Basilar lengths of the Pontian of Melchingen (Kuss, 1960), but this occurrence, based
the skulls of the largest individuals of Mammacyon and Temnocyon on a single tooth, is doubtful. The last reliable occurrence of haplo-
attain 30 cm and about 34 cm, respectively (the largest species of cyonines is in the Les Beilleaux fauna of France (Ginsburg, Huin,
Daphoenus has a basilar skull length of~ 24 cm in the late Oligocene and Locher, 1981), placed by Mein (1989) very early in European
at 28.5 Ma). Neogene mammal zone MN3. This early interval of MN3 is dated
Temnocyonines are found only in North America, haplocyonines at about 20-22 Ma (Steininger, Bernor, and Fahlbusch, 1989) and
only certainly identified in Europe. Asian temnocyonines have not provides an upper limit to the range of haplocyonines consistent
been reported, but the dental morphology of Gobicyon indicates that with their occurrence and stage of evolution in North America.
it is a member of the temnocyonine-haplocyonine clade and may be
derived from a European haplocyonine. Temnocyon Cope, 1878
Temnocyonines share a common basicranial structure typical of Type species: Temnocyon altigenis Cope, 1878.
all amphicyonids in which the basioccipital bone is embayed for an Type specimen: AMNH 6855.
enlarged inferiorpetrosal venous sinus; the basioccipital embayment Characteristics: Small to moderately large digitigrade temno-
is seen in the oldest and most plesiomorphic temnocyonines from the cyonines with basilar skull lengths of 18-~ 34 cm. Species
John Day beds of Oregon. The auditory bulla is preserved in two increase in body size through time (?Whitneyan through
individuals of different lineages (Temnocyon, Mammacyon), and late Arikareean). Largest species are late Arikareean car-
in both it is an extremely rudimentary ectotympanic bone without nivorans the size of large wolves. Known basicrania pre-
any evident enlargement. This rudimentary bulla coupled with the serving the auditory bulla demonstrate (1) in a plesiomor-
derived dentition is a hallmark of most members of the group. phic species a rudimentary ossified ectotympanic bulla,
The most plesiomorphic auditory bulla found in living arctoid car- and (2) in an advanced species a Type A bulla with fused
nivorans is the Type A bulla that occurs in its most rudimentary state ectotympanic and entotympanic elements. Distinguished
in ursids, ailurids, and some mustelids (Hunt, 1974a). The assembly from Mammacyon by maintenance of plesiomorphic den-
of the Type A bulla is best documented in living Ursus americanus, tal traits in all species of the genus, including the large ter-
where it is formed primarily by the ectotympanic with medial and minal Arikareean forms: These include retention of meta-
posterior walls contributed by a rostral entotympanic and two cau- conid on ml, normally configured shearing P4 without
dal entotympanic ossifications. Once these four bulla elements fuse enlarged protocone or massive quality of Mammacyon,
together in the adult, the bulla takes the form of a subhemispheri- laterally compressed elongate premolars, absence of elon-
cal bony capsule that becomes firmly ankylosed to the surrounding gate m2, and retention of plesiomorphic proportions of the
basicranial bones. In one as yet undescribed temnocyonine lineage teeth, more so than in Mammacyon. Postcranial skeleton
from the John Day beds, a Type A bulla has evolved and represents known only in two individuals: Limb elements, including
the only temnocyonine in which such an advanced bulla has been metapodials, indicate these are digitigrade cursorial car-
found (Fingerut et al., 1993). Its significance is in the demonstration nivorans.
that a temnocyonine independently evolved the Type A bulla. In all Lower carnassial length range: T. n.sp. I, 17.5 mm (N = 1),
other temnocyonine lineages the bulla remains a single small ecto- T. altigenis, 19.0-19.7 mm (N = 4), T. ferox, 22.2 mm
tympanic bone, wedged into the middle ear region, appearing too (N = 1), T. percussor, 23.4-24.2 mm (N = 2), T. n.sp. II,
small for the space it occupies, and unfused to surrounding bones 29.5 mm (N = 1).
in adults. Included species: There are about six New World species,
There is no evidence of intercontinental migration of temnocy- confined nearly entirely to the John Day beds of Oregon
onines to the Old World. Temnocyonine dental patterns are restricted and the Arikaree Group of the central Great Plains of North
to North America. However, the origin of temnocyonines may be America. Several New World species are not presently de-
rooted in the European haplocyonines. Mandibles of Aquitanian scribed but are included in a forthcoming study. The North
 Amphicyonidae 209

American species appear to be grouped in two lineages, one two known skulls and all postcranials; hence the genus is
evolving in the John Day region of the northwestern United defined by the Great Plains Arikaree hypodigm. A max-
States, the second evolving in the central Great Plains. Tem- illa from the John Day beds (Haystack Valley) constitutes
nocyon is known from localities CC9C, CP51A, CP103, the only firm record of the genus in Oregon, but its ex-
CP104A, PN6C or D, PN6F or G, PN6G. act stratigraphic placement in the John Day sequence is
uncertain. The genoholotype, M. obtusidens, probably is
Mammacyon Loomis, 1936 of mid-Arikareean age, and the larger, more evolved form
Type species: Mammacyon obtusidens Loomis, 1936. from the late Arikareean can be regarded as a new species
Type specimen: AM 34-41. descended from M. obtusidens.
Characteristics: Mammacyon was originally described by
Loomis from a nearly complete skull and partial postcra-
AMPHICYONINAE
nial skeleton from Arikaree rocks of the Wounded Knee
area, South Dakota. Although no mandible was found with The subfamily is employed here to include a presumed monophyletic
the holotype, other lower jaws in the Arikaree Group from aggregate of genera stemming from the European Cynodictis. All
South Dakota and Nebraska can be confidently referred to Old World beardog genera are included in this category with the
the genus. These fossils establish Mammacyon as a den- exception of haplocyonines. Amphicyonids that migrated in the
tally more derived taxon than the dentally conservative Miocene to Africa and to southeast Asia are considered amphi-
Temnocyon. Species of Mammacyon possess an elongate cyonines (Amphicyon, Cynelos). The rare beardogs recorded from
m2 relative to Temnocyon and also tend to lose the meta- central and eastern Asia (including China) and the Indian subconti-
conid of the lower carnassial (which is retained in Temno- nent also belong to the subfamily. The most commonly encountered
cyon ). In the terminal species the premolars become more amphicyonines exist as long ranging lineages in Europe and include
robust, wider, more massive. Several skulls indicate that the genera Amphicyon, Cynelos, Ysengrinia, Pseudocyonopsis, and
the frontal region is strongly inflated, but because of small Pseudocyon. All of these appear abruptly at various points in the
sample size it is not certain if this is simply a male trait. Miocene of North America yet are unknown in the Oligocene of the
In addition, the upper carnassial and Ml of Mammacyon New World. These unheralded arrivals in North American Miocene
are specialized for crushing as well as shear: The lingual faunas are interpreted as migration events of importance to Neogene
half of Ml is much enlarged and has a centrally placed biochronology.
isolated protocone and hypertrophied lingual cingulum. Bio geographic data provide the strongest argument for the valida-
This is the first specialized crushing dentition evolved by tion of the subfamily. Derived anatomical features allying the genera
amphicyonids. Although the upper carnassial still retains are few due to the marked plesiomorphic dentition and skeleton of
an effective shearing blade, the tooth becomes particularly amphicyonids. Amphicyonines tend to reduce or lose the premolars,
massive and robust, with an enlarged protocone and inter- evolve a short p4 with rounded posterior border, enlarge the pos-
nal cingulum, and is intended for crushing hard materials terior molars, maintain a short-footed plantigrade to subdigitigrade
in company with the anterior premolars. stance without elongation of the lower forelimb bones, and often
Elongate metapodials and limb bones associated with are markedly sexually dimorphic.
the holotype skull of Mammacyon demonstrate that it is
long footed and digitigrade. Postcranial bones of two other Ysengrinia Ginsburg, 1965
individuals referred to Mammacyon from Arikaree rocks Type species: Ysengrinia gerandiana (Viret, 1929)
of southeastern Wyoming confirm that Mammacyon is a Type specimen: None designated, but based on an m2 from
digitigrade lineage with a massive crushing dentition. The Weissenau described by Schlosser (1899) and two man-
late Arikareean species of Mammacyon and Temnocyon dibles (one with m2) from St.-Gerand described by Viret
are the largest members of their respective lineages and (1929, p. 120). The m2 was considered by Viret to be
also the largest temnocyonines to have evolved. particularly diagnostic, and a rostrum with a partial upper
Lower carnassial length range: M. n.sp. I, 21.2 mm (N = 1), dentition was also referred.
M. obtusidens, 24.3 mm (N = 1), M. n.sp. II, 27.8 mm Characteristics: Mid-sized to large amphicyonids with a basi-
(N = 1). lar skull length of 31 cm in the only known North American
Included species: Mammacyon is known from localities skull, an advanced species. Although no complete Euro-
GC8B, GC8A, CP52, CP85C, CP86B, CP102, PN6C or D. pean skull is known, an estimate of the basilar length of a
single individual represented by a rostrum (Viret, 1929, Pl.
Comments: There is probably a single evolvingMammacyon VII, Figure 2) from St.-Gerand is 24-25 cm. Skull exhibits
lineage in North America. A small undescribed mandible strong inflation of the frontal region in both European and
from the Sharps Formation of South Dakota first shows the North American examples. The North American skull is
specialized dental traits that identify the genus in North broad snouted, massive, with expanded frontal region and
America. In the central Great Plains, Mammacyon occurs enlarged canines and is interpreted to be a male; female
throughout the Arikareean, and this sample includes the skulls are presumed to be narrower, more gracile. Sexual
Robert M. Hunt, Jr. 210

dimorphism is present in both dentition and in postcranial Measurements of North American specimens indicate
skeleton, based on the large North American sample from that the p4 is becoming shorter through time while the
the Bridgeport Quarries, Morrill County, Nebraska. Dental ml becomes larger and more robust. Reduction in anterior
formula 13/3, Cl/l, P4/4, M3/3. premolars also accompanies this trend. In the Bridgeport
The genus is recognized by its dolichocephalic skull, Quarries sample, dimorphism is evidenced by the presence
distinctive Ml-2, and reduced anterior premolars. The of both slender, shallow as well as massive, deep mandibles
mandible is anteriorly deep and massive in some individ- together with short, robust metapodials and limb bones
uals, probably males. In North America, Ysengrinia has of two size classes. Thus the North American Ysengrinia
been found only in the central Great Plains. This is the appears to be characterized by large, heavy-bodied males
only large North American amphicyonid with reduced P2- and more gracile, light females, similar in this respect to
3/p2-3 that occurs in the Arikaree Group before the first certain living ursids.
appearance of Amphicyon in the superjacent Hemingford Outside of North America, Ysengrinia has been iden-
Group. tified only in Europe. The genus was distinguished from
The lower carnassial is a massive, robust tooth. The ml Pseudocyon by Ginsburg using upper teeth (Ml), supple-
metaconid remains prominent, unreduced to somewhat re- mented by careful distinctions in the lower teeth, primarily
duced, and is slightly retracted; the paraconid blade is ad- m2 (as discussed earlier). Ginsburg based his new genus
vanced. The short ml talonid has a prominent hypoconid on "Pseudocyon" gerandianus Viret, and added to it two
ridge that dominates the heel; internal to the hypoconid is other species: Y. tolosana from the Aquitanian of Paul-
a lingual shelf with occasional low tubercles but no strong hiac and Y. depereti from the Burdigalian of Chilleurs-
development of an entoconid. aux-Bois. In Europe the skull is known only from a re-
The m2 protoconid and metaconid are developed, but ferred rostrum from St.-Gerand (Viret, 1929, Pl. VII). Ml
the protoconid is larger. The paraconid is weak to absent. in the skull is unusual, having a pear-shaped form (Gins-
The short, abbreviated talonid is dominated by a strong burg, 1965). Premolars anterior to the upper carnassial are
hypoconid ridge somewhat labially placed (strongly labi- quite reduced, just as is the pl-3 in Viret's mandible from
ally placed in the European Y. depereti); there is no ento- St.-Gerand.
conid. The talonid is lower than the trigonid. Although the European material comprises few spec-
The genus is known from the remains of ( 1) about imens, comparison with the much larger North Ameri-
7-10 individuals from the Agate Monument bone bed and can sample does support a relationship between the two
stratigraphic equivalents (basal Upper Harrison), western groups of fossils. The North American sample shows the
Nebraska; (2) about 3-4 individuals from the Upper Harri- robust m 1 with short talonid dominated by a labially placed
son beds of southeastern Wyoming; and (3) 8 mandibles, 1 prominent hypoconid; a short m2 with trigonid compris-
maxilla, numerous isolated teeth and postcranial elements ing the anterior half of the tooth; reduction of upper and
of at least 14 individuals from the Bridgeport Quarries, lower premolars (except P4/p4); a similarity in Pl-3 form
Nebraska. between the St.-Gerand rostrum and the only North Amer-
Lower camassial length range: all species, 30.0-35.4 mm ican skull from the upper Arikaree Group of southeast
(N = 24). Wyoming; a low, nearly fiat M2 with tonguelike lingual
Included species: Probably three North American species, cingulum; a triangular Ml, robust, with a thick tonguelike
as yet undescribed. Ysengrinia is known from localities lingual cingulum appended to the posterointemal comer
CP51A, CP104A, CP106. of the tooth.
In the North American sample, many individuals (prob-
Comments: Ysengrinia was first described by Ginsburg able males) possess massive mandibles that are nearly as
( 1965) as a European genus and was carefully distinguished deep at their anterior end as in the region below the mo-
by him from Pseudocyon. Pseudocyon has a differently lars: No other North American early Miocene amphicyonid
configured m2 relative to Ysengrinia, the former based shows this trait. There is also a characteristic placement of
on a mandible from Sansan and a referred mandible from two mental foramina close together, high on the side of the
Ponsan-Soubiran. The m2 of Pseudocyon has a trigonid mandible, one below the anterior root of p3, the other at a
crowded to the front of the tooth, followed by a long talonid point between the single root of p 1 and anterior root of p2.
with a relatively fiat surface; the hypoconid is low and All North American specimens are from western Ne-
placed far toward the labial border of m2. In Ysengrinia braska and southeastern Wyoming. Upper Arikaree rocks
the m2 trigonid occupies the entire anterior half of the of Goshen County, Wyoming, have produced the only
tooth; the talonid is short but with a low hypoconid also known complete skull and associated mandible of the
labially placed. The long heel ofm2 and low p4 distinguish genus. It is probably a male with a massive rostrum and
Pseudocyon from Ysengrinia. The North American sample mandible, large canines, reduced anterior premolars, an
of Ysengrinia has the m2 structure of European members expanded frontal region but only a small braincase. The
of the genus, for example, Y. tolosana (Basel Museum Pa oldest record of Ysengrinia in North America is from the
951, Paulhiac). basal Upper Harrison beds at Agate National Monument
 Amphicyonidae 211

and other waterhole bone beds at the same stratigraphic twice the length of P4, and the enlarged rectangular M2
level in northwest Nebraska. It may have been an impor- approaches the size of the more triangular Ml. In both
tant scavenger-predator at these waterholes. From higher early and mid-Miocene Cynelos from North America, the
stratigraphic levels within the Upper Harrison beds, there Ml-3/P4 length ratio is remarkably constant at 1.9-2.0,
are only three known occurrences: the associated skull and whereas in North American Amphicyon this ratio is 2.3-
mandible from Goshen County; isolated teeth and postcra- 2.4, indicating the relatively greater enlargement of the
nials from a locality east of Spoon Butte; and a partial second and third molars in Amphicyon.
postcranial skeleton found west of Spoon Butte. Other dental features of Cynelos include a tendency in
The largest North American sample of Ysengrinia is also some species to reduce the size of p3 so that it is the same
the largest in the world and comes from the Bridgeport size or smaller than p2, which does not occur in Amphi-
Quarries, Morrill County, Nebraska. This appears to be cyon. The slender, elongate mandibles have pl-3 without
the last North American occurrence of the genus, includes accessory cusps, p4 with or without a posterior accessory
the largest individuals, and is of probable early Heming- cusp. The premolars are slender, laterally compressed teeth
fordian age (the previously mentioned North American without any enlargement for durophagy; in mid-Miocene
occurrences are all late Arikareean). Known only from the species the first premolar may be lost.
Upper Harrison beds and from the Bridgeport Quarries, The ml has a medially inflated protoconid with a small
the estimated range of the genus in North America is from metaconid situated on its posterointemal slope. The ml
about 18-19.2 Ma. talonid is long, developed, and dominated by a prominent
hypoconid laterally placed in the early Miocene species
Cynelos Jourdan, 1862 and more central in the North American mid-Miocene
Type species: Cynelos lemanensis (Pomel, 1846). forms. The m 1 entoconid is smaller than the hypoconid
Type specimen: Skull with associated mandible from Langy, and may be little developed; an anteroposterior basin is
near St.-Gerand, France (mid-Aquitanian); see Ginsburg, present between hypoconid and entoconid that opens to
1977, p. 59. the rear of the talonid and is not closed by a talonid ridge.
Characteristics: Mid-sized to large amphicyonids with di- The m2 is rectangular, with trigonid cusps crowded to-
agnostic upper and lower molars and a bearlike postcra- ward the anterior end of the tooth, the protoconid slightly
nial skeleton in which the metapodials are short in earlier larger than metaconid, the two cusps separated by a con-
species, somewhat longer in later North American forms. striction and often joined anteriorly by a smoothly curving
Limbs and feet not elongate as in Daphoenodon but rather C-shaped crest. The paraconid is absent or vestigial. The
short, less cursorial. Basilar skull length about 24-27 cm m3 is rectangular with two closely apposed roots.
in Aquitanian species of Europe, 30 cm in North American The upper molars have prominent paracone, somewhat
early Hemingfordian species, 33-34 cm in the late Hem- smaller metacone, and a distinct but much lower proto-
ingfordian species C. idoneus, and 39-44 cm in the termi- cone, all surrounding a protocone basin. A small para-
nal North American species C. sinapius of early Barstovian conule and often pronounced metaconule are present. A
age. tonguelike lingual cingulum characterizes Ml-2. These
Skull form and canine teeth dimorphic: males larger with ro- traits are found in both Cynelos and Amphicyon so that it
bust canines; females smalier, with more gracile canines. is primarily the much larger M2 (and M3) that separates the
Dental formula 13/3, Cl/l, P4/4, M3/3. Recognized by latter larger beardog from Cynelos. The terminal species
dolichocephalic skull, narrow rostrum, and characteris- of Cynelos in North America (C. sinapius) from the early
tic molars in which the M2-3/m2-3 are enlarged relative Barstovian of western Nebraska has developed the largest
to Ml/ml yet remain smaller in size when compared to M2 of the lineage relative to Ml, but this tooth still does
contemporaneous species of North American Amphicyon. not attain the size found in North American large species
Premolars are reduced in some forms. The auditory bulla of Amphicyon (A.frendens, A. ingens).
preserved in one of the earliest species (C. lemanensis, The metapodials of Cynelos are short as in living ursids,
Aquitanian, St.-Gerand, France, Hunt, 1977) completely especially in the earlier species, becoming more elongate
encloses the middle ear space and is configured as in the in the mid-Miocene C. idoneus. This contrasts with the
Type A bulla ofliving ursids (Hunt, 197 4a). This is the first uniformly short, massive metapodials of North American
European amphicyonid lineage to develop a completely and Old World Amphicyon.
ossified Type A bulla, appearing between 20-23 Ma. Lowercarnassial length range: C. sp. (Agate Monument bone
Cynelos in North America and Europe is easily recog- bed), 21.3 mm (N = l); C. n.sp. I (Bridgeport Quarries),
nized by its teeth: It is the first amphicyonid to enlarge 21.0-25.3 mm (N = 12); C. n.sp. II (Runningwater
the second and third molars relative. to the camassials and Formation), 26.4-27.1 mm (N = 2); C. idoneus (Sheep
Ml. This already occurs in relatively small early Miocene Creek Formation), 25.6-30.2 mm (N = 5); C. sinapius
species in Europe, North America, and Africa (basilar (early Barstovian of Nebraska, Colorado, New Mexico,
skull length, 22-23 cm, C. lemanensis, Hunt, 1977). In Texas), ~31-39.7 mm (N = 17); C. caroniavorus (Thomas
these early Miocene species the length ofMl-M3 is nearly Farm), ~15 mm (N = 1).
 Robert M. Hunt, Jr. 212

Included species: Probably six North American species, three ened rostrum, and P3 rotated into transverse alignment in
as yet undescribed. Cynelos is known from localities two of the three known crania, p3 likewise turned outward
GC4E, GC8D, NB6B, SB32B, CP75B, CP104A, CP105, in some mandibles. Frontal region of cranium slightly ex-
CP106, CP108A, CP108B, CPl 10. panded. Vestigial M3 occurs in some individuals. Strong
dimorphism is not evident, males averaging about the same
Comments: In North America, Cynelos is found in the early in skull size relative to presumed females. In the upper
and medial Miocene, in Europe in the Oligocene to late dentition, I3 is large, Il-2 much smaller. Premolars (Pl-3,
Miocene (Ginsburg, 1977), and in the early Miocene of pl-4) tend to be low and lack anterior or posterior ac-
Africa (Schmidt-Kittler, 1987). The African fossils were cessory cusps. P4 is small relative to Ml as in Cynelos.
first named Hecubides by Savage (1965), but the fossils There is a prominent P4 protocone but no parastyle. Ml is
from east Africa clearly are immigrant forms of Cynelos subquadrate with a low protocone connected to very weak
from Eurasia: Hecubides is a junior synonym of Cynelos. meta- and paraconules by low protocristae. M2 is smaller
Cynelos is an immigrant to North America and Africa in the than Ml, and its paracone is considerably larger than meta-
early Miocene, appearing almost simultaneously on both cone. The lingual half of M2 is flat with a prominent thick
continents at about 19-20 Ma. Because of its presence in lingual cingulum. The molars are reminiscent of Cynelos.
North America it also must have occupied parts of Asia, The short rostrum observed in the three known skulls is
but no fossil record yet exists. correlated with the length of the upper toothrow, ranging
Cynelos first appears in North America as a single iso- from 87-92.8 mm. The palate is broader than in Cynelos
lated carnassial (CM 2200, Hunt, 1972) from the water- despite a relatively shorter skull in Pliocyon.
hole bone bed at Agate National Monument, Nebraska, late The lower carnassial is robust, wide, with an inflated
Arikareean (19.2 Ma). It is represented by a large sample protoconid and reduced metaconid. Its talonid is as wide
from the Bridgeport Quarries, Nebraska, early Hemingfor- as the m2 trigonid and has a dominant trenchant hypoconid
dian (~18-18.8 Ma) and also occurs in the lower part of the placed slightly laterad and an internal talonid shelf without
early Hemingfordian Runningwater Formation, Nebraska. a developed entoconid. The m2 trigonid has a prominent
It next appears in the Sheep Creek Formation (C. idoneus, protoconid, slightly lower metaconid, and no paraconid:
late Hemingfordian) and ends in the Olcott Formation (C. Only a smooth arcuate C-shaped crest forms the front of
sinapius, early Barstovian). Nearly the entire North Amer- the trigonid as in Cynelos. The m2 talonid can be squared
ican sample of C. idoneus and C. sinapius comes from the off or tapered; the hypoconid is the sole cusp and is laterally
Snake Creek-Sheep Creek district in Sioux County, Ne- placed.
braska, and the Pawnee Buttes area, Colorado. Postcranials from the Olcott Formation include slightly
A small species of Cynelos (C. caroniavorus) occurs elongated metapodials, somewhat longer than those of
at the early Hemingfordian Thomas Farm site in north Amphicyon. Its skeleton was robust, slightly elongated in
Florida (Tedford and Frailey, 1976) but is found nowhere the limbs, but without the heavy and massive elements
else in North America. It is probably allied to the small C. and short proportions of Sheep Creek and Olcott
schlosseri of Europe. Amphicyon.
In Europe, Cynelos first occurs in the Quercy fissures (C. Lower camassial length range: Pliocyon medius (early
piveteaui, C. crassidens) and in numerous Aquitanian sites Barstovian, Nebraska), 26.3-31.7 mm (N = 12); (early
of France (C. lemanensis, C. rugosidens). The Burdigalian Barstovian, California), 28.9-31 mm (N = 2); P. robustus
species C. schlosseri and C. helbingi are best known from (Bone Valley, Florida), 33.2 mm (N = l); P. sp. (Sheep
the Bavarian fissures at Wintershof-West where large sam- Creek Formation, Nebraska), 30.3-~32 mm (N = 2).
ples were described by Dehm (1950), but they are also Included species: P. medius (late Hemingfordian to early
present in the Orleanian of the Loire Basin (MN3-4, C. Barstovian, Nebraska and California, known from locali-
schlosseri-helbingi; MN5, C. bohemicus, Ginsburg, 1989) ties NB6C, CPI 08A, CPI 08B, CPl 10, CPl 11 ); P. robustus
where the genus extends to upper MN5 (~16 Ma). Some (?Barstovian, Florida, locality GClOB).
European authors suggest that the lineage may continue as
"Amphicyon" steinheimensis in MN7, persisting into the Comments: Pliocyon was recognized and is currently defined
late Miocene (Ginsburg, 1977, p. 94). almost entirely on the basis of the large sample from the
Sheep Creek-Snake Creek District of Sioux County, Ne-
Pliocyon Matthew, 1918 braska. Rather poor material including several fragmentary
Type species: P. medius Matthew, 1918. mandibles is known from the late Hemingfordian Sheep
Type specimen: AMNH 17207. Creek Formation; however, the genus is firmly established
Characteristics: Mid-sized amphicyonids with skull lengths on the well-preserved crania, mandibles, and postcranials
of 29-30 cm (only three crania known, all from Sioux collected from various quarries in the Olcott Formation
County, Nebraska). Dental formula I3/3, Cl/1, P4/4, M2/3 of early Barstovian age, belonging to the "Lower Snake
with en echelon crowding of premolars common in a short- Creek" fauna of Matthew.
 Amphicyonidae 213

Pliocyon also occurs in the Barstow Formation, Mo- dens (late Hemingfordian, Nebraska), ~33.5-39.8 mm (N
jave Desert, California, where it is known from several = 11); A. ingens (early to medial Barstovian, Nebraska,
mandibles. The genus was also reported from the Bone Colorado, New Mexico, California, and Oregon), 36-44.9
Valley Formation of Florida (Berta and Galiano, 1984): mm (N = 20).
The mandible on which a new species was based measures Included species: Three North American species, one of these
larger in various dental dimensions than Great Plains or as yet undescribed (early Hemingfordian to medial Barsto-
California Pliocyon and may be a larger and/or younger vian of Nebraska, Colorado, New Mexico, and California,
form, but does seem best referred to the genus. ~14.5-18 Ma in North America). Amphicyon is known
The teeth of Pliocyon are similar to those of Cynelos. It from localities GC4C, NB6C, NB6D, SB32B, CP75C,
is likely that Pliocyon is a derivative of the Cynelos lineage, CP105, CP108A, CP108B, CPl 10, PN8B.
possibly diverging in the early Hemingfordian.
Pliocyon is unknown in Eurasia or Africa and is regarded Comments: The oldest fossils legitimately attributed to Am-
as a short-lived endemic North American group confined phicyon are of Stampian age (mid- to late Oligocene) in
to the mid-Miocene. western Europe (Ginsburg and Telles-Antunes, 1968). In
Asia the oldest record of Amphicyon is from the early
Amphicyon Lartet, 1836 Miocene of Vietnam (Ginsburg et al., 1992). The ma-
Type species: A. major Blainville, 1841. jority of Asian specimens attributed to the genus need
Type specimen: MP Sa 1. restudy to determine their true relationships. In Africa the
Characteristics: Mid-sized to very large amphicyonids with only record of Amphicyon is in the mid-Miocene Arris-
skull lengths of 23-25 cm (early Hemingfordian), 37-39 drift fauna of Namibia (Hendey, 1978) where a magnifi-
cm (late Hemingfordian), and about 42-52 cm (early to cent mandible with teeth demonstrates that large species of
mid-Barstovian). Dental formula I3/3, Cl/l, P4/4, M3/3 Amphicyon migrated to the limit of the continent, the south-
with reduction of premolar size and height but without ernmost point colonized by amphicyonids anywhere in the
loss of premolars. There is no premolar crowding as seen in world.
Pliocyon. Marked hypertrophy of posterior molars (m2-3, In North America there is no plausible New World an-
m2-3), producing the largest such teeth evolved within the cestor for Amphicyon at its first appearance in the early
family. Ml-3 length commonly greater than two times the Hemingfordian of Nebraska at about 18 Ma. This earliest
length of P4 as a result of molar expansion. Enlarged M2 North American species is undescribed, is a probable im-
somewhat wider transverse! y than M 1; M2 subtriangular migrant from the Old World, and can be compared with
in early species, rapidly becoming subrectangular in later early Burdigalian Amphicyon from Europe discussed by
larger forms, and projecting lingually farther than the in- Ginsburg and Telles-Antunes (1968). They placed the Eu-
ner edge of Ml. M2 protocone connected to paraconule ropean fossils in A. giganteus, a long-ranging European
but metaconule isolated as in European Amphicyon. P4 species known from early Burdigalian to early Helvetian.
short, wide, robust, with weak to moderate development of Late HemingfordianAmphicyon is represented by a sample
protocone and small parastylar cusp. The lower carnassial of the large A.frendens from the Sheep Creek Formation of
has a swollen inflated appearance, always retains a meta- Nebraska. Its descendant in the early Barstovian faunas of
conid, and has a wide talonid with prominent, somewhat Nebraska and Colorado is the giant A. ingens, the largest
laterally placed, ridgelike hypoconid. The m2 is rectangu- of the North American amphicyonids. The youngest fos-
lar, with trigonid elevated above talonid, protoconid larger sils of this species in North America occur in faunas from
than metaconid, and a small vestigial paraconid (the arcu- the early part of the late Barstovian (or medial Barstovian)
ate crest at the anterior end of the m2 trigonid of Cynelos is such as the Horse and Mastodon Quarry, Pawnee Creek
absent). A somewhat laterally placed low hypoconid dom- Formation, Colorado (~14-15 Ma).
inates the m2 talonid. M3/m3 are flat massive teeth with a These large Amphicyon must have been formidable pre-
subdued occlusal topography. dators, weighing over 200 kg, with massive skulls, robust
Sexual dimorphism is present, particularly developed in shearing camassials, great crushing molars, and power-
the youngest and largest species. The postcranial skeleton ful limbs capable of short bursts of speed in pursuit of
is important to identification of the genus: The limb skele- prey. Following the appearance of Ysengrinia in the early
ton is massive and robust, very similar to the limb elements Miocene, these are the first large bearlike camivorans to
of large living ursids. Metapodials and digits are short, evolve in North America, attaining maximum size in the
never elongate, and specializations for cursoriality such as mid-Miocene. Amphicyon was replaced in the late Barsto-
lengthened lower limb segments and elongate metapodi- vian and Clarendonian by other large amphicyonids of
als are absent. The gait was ambulatory, the limbs and feet the Pseudocyon-Jschyrocyon group, and they in tum were
with powerful heavy musculature. supplanted by the huge ursids Agriotherium and lndarc-
Lower camassial length range: Amphicyon n.sp. (early Hem- tos of the Hemphillian. During the mid-Miocene, Amphi-
ingfordian, Nebraska), 31.2-32.2 mm (N = 3); A. fren- cyon was accompanied in the central Great Plains by large
 Robert M. Hunt, Jr. 214

species of Cynelos, the latter averaging somewhat smaller sites ofnorth-central Nebraska that display the characteris-
in size. Both genera are Old World immigrant stocks that tic "thaumastocyonine" wear pattern, reduced premolars,
reached North America in the early Miocene and became and the carnassial and molar morphology of the European
extinct in the mid-Miocene in the New World. specimens. The wear pattern is produced by extremely tall
European Amphicyon is best known from the genoholo- paracones on Ml-2 and a tall Ml metacone. In the lower
typic species A. major from Sansan, France (Ginsburg, teeth the m 1 trigonid and talonid are also high, as is the
1961). Referred postcranials accompanying the type den- m2 trigonid. During the bite, the tall ml trigonid fits into a
tition indicate a carnivore with ambulatory gait, lacking deep embrasure in front of Ml, shearing against the broad
cursorial specializations: short stout metapodials and ra- internal surface of the upper carnassial. P4 protocone is re-
dius, astragalus and calcaneum, all similar to those ele- duced, and the m 1 paraconid cuts a wear groove in its pos-
ments in North American Amphicyon. terior face. The typical "thaumastocyonine" pattern, how-
ever, is created by the tall Ml paracone that cuts a wear
Pseudocyon Lartet, 1851 groove in the anterolateral face of the ml talonid. Behind
Type species: P. sansaniensis Lartet, 1851. this facet the Ml metacone, not quite as tall as the paracone,
Type specimen: MP Sa 207. cuts a prominent wear surface on the anterolateral face
Characteristics: Mid-sized to large amphicyonids with the of the m2 trigonid (a facet is also cut by the metacone down
only known skull having a basilar length of 37 cm (early the posteroexternal corner of the m 1 talonid just below the
Clarendonian, Nebraska). Dental formula I3/3, C 1/1, P4/4, ml hypoconid). Finally, the tall M2 paracone creates a
M3/3 (M3 is retained contrary to terminal forms of lschy- wear groove on the outer face of m2 between its trigonid
rocyon). Sexual dimorphism highly probable. Identified and talonid. The elevated massive m 1 hypoconid fits firmly
by dental features, especially a diagnostic pattern of wear into the Ml protocone basin: a dental mortar-pestle device.
facets observed on the carnassials and molars produced The high m2 trigonid is focused to an edge formed by its
by vertical shear. In the Old World this has been termed protoconid-metaconid: this edge fits between the posterior
the "thaumastocyonine" pattern and is found in European border of Ml and anterior border of M2, and the heel of the
specimens of Pseudocyon (Kuss, 1965). The anterior pre- m2 is applied to the shallow protocone basin of M2. The
molars (Pl-3, pl-3) are strongly reduced, to a greater de- three primary wear grooves found on the ml-2 are cut by
gree than those oflschyrocyon, and this premolar reduction the tall, nearly vertical labial cusps of Ml-2 and constitute
is already present in the geochronologically oldest Pseu- the essence of the "thaumastocyonine" pattern. This wear
docyon from North America. The p4 has both anterior and pattern is so distinctive that Viret (1929) created the genus
posterior accessory cusps, whereas lschyrocyon lacks the Amphicyonopsis for an m 1 from La Grive that displayed it:
anterior cusp. The m2 trigonid is shifted anteriorly and is the tooth was later assigned to Pseudocyon by Kuss (1965,
elevated above the talonid (this forward shift and elevated p. 134) and Amphicyonopsis placed in synonymy.
trigonid are absent inlschyrocyon). The subrectangular M2 There is no doubt that some mandibles with teeth show-
is always smaller than the triangular Ml, differing in this ing the "thaumastocyonine" wear pattern from early Clar-
respect from Amphicyon in which M2 > Ml. endonian sediments of Nebraska (e.g., Lucht Quarry, F:AM
No postcranials are certainly associated with dentitions 96627) are nearly identical in mandibular and dental mor-
in the North American sample. phology to specimens of Pseudocyon sansaniensis from
The genus is known from (1) a skull without associated Europe (MN7, Ponsan-Soubiran, Kuss, 1965, Figure 78).
lower jaws, and a damaged but associated partial mandible Upper teeth from early Clarendonian sites in Nebraska du-
and maxilla, and (2) at least eight mandibles, a few isolated plicate the upper molar of the Ponsan-Soubiran individual,
teeth, but no certainly associated postcranial bones. including the tall M 1 paracone that creates the principal m 1
Lower carnassial length range: 30.9-~40 mm (N = 10). wear groove. Pseudocyon ranges in North America from
Included species: In North America there appears to be a sin- the early Barstovian Olcott Formation fauna and medial
gle lineage that at present has not been formally named. Barstovian Hottell Ranch quarries (UNSM Loe. Bn-12) of
In Europe a similar situation seems to occur: members of western Nebraska to the medial and late Barstovian sites
the European lineage are placed in P. sansaniensis (Kuss, in the Valentine Formation of north-central Nebraska, end-
1965). Pseudocyon is known from localities SB32D, ing in early and late Clarendonian sites of the Ash Hollow
CPl 10, CP114A, CP114B, CP114D, CP115B, CP116A, Formation in northern Nebraska. This range is estimated
CP116B. to span an interval from about 15.5-9 Ma.
A problem in the recognition of Pseudocyon in North
Comments: Pseudocyon is a European amphicyonid, origi- America arises because of the dental and cranial simi-
nally described from Sansan (France), first appearing in larity to North American lschyrocyon. lschyrocyon was
zone MN4 and continuing into MN7, hence from about originally defined using the youngest forms from the late
13-17 .2 Ma. It was recognized in North America from am- Clarendonian, which are enormous animals with massive
phicyonid dentitions in late Barstovian and Clarendonian shearing dentitions (Matthew, 1902; Matthew and Gidley,
Amphicyonidae 215

1904). Whereas the large late Clarendonian species is eas- The last North American occurrence appears to be an iso-
ily recognized, Barstovian and early Clarendonian fossils lated ml (F:AM 54232) from the late Clarendonian Xmas-
are sometimes difficult to distinguish from contemporane- Kat Quarries local fauna of Nebraska ( ~9 Ma).
ous Pseudocyon. However, various lines of evidence de- In North America the Pseudocyon-lschyrocyon group
rived from examination of the large American Museum abruptly replaces the large Amphicyon in the 14-15 Ma
sample of these amphicyonids, as well as key specimens interval. Amphicyon appears to be extinct by 14.5 Ma in
in the University of Nebraska collections, allow definition North America. This is a striking change in the large car-
of two separate lineages, Pseudocyon and lschyrocyon. nivoran fauna and occurs during the transition from early
The reasons for segregating the two genera are as fol- to medial Barstovian faunas of the Great Plains region.
lows:

(1) Mandibles with well-preserved dentitions from the lschyrocyon Matthew and Gidley, 1904
Burge Quarry (early Clarendonian, Nebraska) repre- Type species:/. gidleyi (Matthew, 1902).
sent both genera: (a) a mandible (F:AM 54225) sim- Type specimen: AMNH 10671.
ilar to the genoholotypic mandible of Pseudocyon from Characteristics: Mid-sized to large amphicyonids with skulls
Sans an, with extreme reduction of p 1-3, small p4, deep having basilar lengths of 28-36 cm (Barstovian, N = 8)
"thaumastocyonine" wear groove on ml (made by Ml and 34-47 cm (Clarendonian, N = 6). Dental formula I3/3,
paracone), and other typical facets present on m2. The Cl/l, P4/4, M2-3/3 (M3 is lost in many Clarendonian an-
m2 trigonid cusps are anteriorly shifted and the talonid imals). Sexual dimorphism present based on two sizes of
is a low, long shelf as in European Pseudocyon; (b) a teeth at several sites. Identified by dental and cranial fea-
second mandible (F:AM 54212) with pl-4 all present tures, especially low, robust, unreduced premolars (may be
(p2-3 low but well developed), worn ml without pro- crowded), shearing carnassials in which P4 rotates labiad
nounced "thaumastocyonine" wear groove, and a large, with increasing age so that the protocone root protrudes out
broad m2 with longer trigonid than in Pseudocyon. This of the palate as an exposed strut, Ml >M2>M3, M2 sub-
jaw appears to be lschyrocyon. Both animals are old, rectangular and more of a flat crushing tooth. Them I meta-
based on tooth wear, yet in lschyrocyon wear has flat- conid is gradually lost in time (absent in all late Clarendo-
tened the ml talonid and m2; in Pseudocyon the ml nian lschyrocyon), and them 1 talonid is eventually domi-
talonid remains trenchant, and m2 retains the high an- nated by a large centrally placed hypoconid. The m2 pos-
terior trigonid and low talonid. In addition, F:AM 54212 sesses a lowertrigonid than inPseudocyon, and the trigonid
duplicates the morphology of the late Clarendonian gi- is not crowded to the front of m2 as in the latter genus.
ant, lschyrocyon gidleyi but is simply smaller. The wear facets observed on the carnassials and molars
produced by vertical shear can simulate the "thaumasto-
(2) A large sample of amphicyonids of a single species de-
cyonine" pattern, particularly on ml, but the pattern is not
rived from the Hemicyon Stratum and Quarry, Barstow
developed as strongly as in Pseudocyon. The final wear
Formation, California, by Frick Laboratory field par-
stages in old individuals produce flat wear surfaces on the
ties, shows a uniform dental morphology of the Jschy-
ml talonid and the m2 trigonid and talonid, thereby dif-
rocyon pattern, without the features typical of Pseudo-
fering from the "thaumastocyonine" pattern of aged Pseu-
cyon. These fossils occur below a tuff dated at 13.4 +/-
docyon. Although the Ml para- and metacones are tall as
0.7 Ma that occurs above Hemicyon Quarry (Tedford et
in Pseudocyon, the m2 para- and metacones are not as
al., 1987). This sample is an excellent ancestral popula-
tall as in that genus. The terminal Clarendonian form of
tion preceding in time the Clarendonian lschyrocyon of
lschyrocyon has a large robust skull with shortened bas-
Nebraska: it correlates well with Jschyrocyon from the
icranial region (thus distinct from Amphicyon and Cyne-
Valentine Formation of Nebraska (Cornell Dam, Crook-
los), expanded frontal region, and a broadened snout with
ston Bridge, and Devil's Gulch members).
incisors arranged in an arc. The I3 is enormous, nearly
The European time range of Pseudocyon ( ~ 13-l 7.2Ma) as large as the canine. In Pseudocyon the upper canine
slightly precedes the time range of the genus in North has a thin, sharp posterior ridge, and the tooth itself is
America (~9-15.5 Ma). The genus conceivably immi- laterally compressed; the lschyrocyon canine is without
grated to North America about 15.5 Ma. At its first ap- the sharp posterior ridge and is more rounded in cross-
pearance in North America, it occurs at a single early section. The plesiomorphic amphicyonid dentition is sim-
Barstovian site in Nebraska. In the medial Barstovian it ply enlarged inlschyrocyon without the development of the
is represented by very large carnivores from at least three widened crushing molars of large Amphicyon. Large hypo-
localities: a huge mandible from the Hottell Ranch quar- tympanic sinuses occur in the floor of the auditory bullae of
ries (UNSM Loe. Bn-12), Nebraska; a comparably large lschyrocyon as inAmphicyon and Cynelos, which is a par-
pair of mandibles from the Espanola Basin, New Mexico allel development in these large amphicyonids.
(Pojoaque Bluffs); and a partial mandible and enormous lschyrocyon ranges in time from small forms from the
proximal femur from Norden Bridge Quarry, Nebraska. Hemicyon Stratum of the Barstow Formation, California,
Robert M. Hunt, Jr. 216

and the Norden Bridge fauna of the Valentine Formation, gidleyi. In addition, Hadrocyon Stock and Furlong (1926)
Nebraska, to the huge animals of the Gidley Horse Quarry from the Ricardo fauna, California, is a junior synonym of
in the Clarendon beds, Texas. Its temporal span is approx- lschyrocyon.
imately 14-9 Ma. These are the youngest well-sampled lschyrocyon and Pseudocyon share a similar skull form,
amphicyonids in North America. In the late Barstovian and including a common molar morphology, broad rostrum and
Clarendonian, Jschyrocyon and Pseudocyon are the prin- frontal region, and abbreviated basicranium. Both genera
cipal amphicyonid lineages in North America. A smaller also have developed a massive, united mastoid/paroccipital
animal as yet undescribed comes from this same time in- process. The ancestry of Pseudocyon probably lies in Cyne-
terval but is known from only two specimens. No cer- los: both genera share a derived m2 morph in which the
tain records of beardogs have been documented in the trigonid is crowded to the anterior end of the tooth. The
Hemphillian. earliest North American lschyrocyon also has molars and
The genus lschyrocyon is known from (1) numerous camassials like those of Cynelos, but it lacks the "crowded"
skulls: of particular note is the B arstovian sample from the m2 trigonid of Pseudocyon. The origin of lschyrocyon is
Hemicyon Stratum, Barstow syncline, California, and a uncertain but may lie in an early form of North American
Clarendonian sample from (a) the Ash Hollow Formation or Asian Cynelos. Late Barstovian lschyrocyon from Ne-
of northern Nebraska and southern South Dakota, and (b) braska shows a pronounced correspondence in its dentition
the Clarendon beds, Texas. Some skulls are accompanied to the contemporaneous European Cynelos steinheimensis
by associated mandibles; (2) over thirty mandibles, many from Germany (MN7, ~13.4 Ma).
isolated teeth, and postcranial elements. It is striking that the North American amphicyonids of
Lower camassial length range: Hemicyon Stratum sample, the late Barstovian and Clarendonian are carnivores with
27.3-31.5 mm (N = 15); Crookston Bridge and Devil's enlarged plesiomorphic shearing dentitions, in essence hy-
Gulch members of Valentine Formation, 29.5-35 mm percamivores, and are not the large hypocamivorous bear-
(N = 6); Burge member of Valentine Formation, 33.6- dogs with crushing molars typical of the late Hemingfor-
40.2 mm (N = 7); late Clarendonian sites, Texas, Nebraska, dian and early Barstovian interval (Amphicyon, Cynelos).
South Dakota, 38.6-46.1 mm (N = 10); late Clarendonian The replacement of the large crushing-toothed beardogs by
Ricardo fauna, California, 44.0 mm (N = 1). the seemingly more dentally plesiomorphic Pseudocyon-
Included species: In North America there is a single lin- lschyrocyon group remains unexplained. However, these
eage, I. gidleyi, ranging from the medial Barstovian to the large Clarendonian amphicyonids are succeeded in the
late Clarendonian, based on the holotype skull and par- Hemphillian by large ursids (lndarctos, Agriotherium)
tial skeleton described by Matthew. lschyrocyon gidleyi which must have adopted much the same niche among the
could be considered as a single lineage extending back in large predatory Camivora as utilized by the great beardogs.
time to the first appearance of the genus in North Amer- In the Old World, amphicyonids survive no longer than
ica. Whether it is necessary to name the earlier part of in North America: the last European record is about 9 Ma
the lschyrocyon lineage as a separate species is an open at Kohfidisch, Austria (Beaumont, 1984). The family is
question. The genus is unknown outside of North Amer- apparently extinct by 13 Ma in eastern Asia and by about
ica. lschyrocyon is known from localities NB6D, NB6E, 7 Ma on the Indian subcontinent (Siwaliks). In Africa,
NB7B,NB27II, SP2, CP90A, CP114B, CPI 14C, CP114D, amphicyonids are not found after about 11 Ma, but future
CP115B, CP116A. discoveries may establish the survival of relict species as
in the Siwaliks.
Comments: Anlschyrocyon was first described from a partial In Europe the youngest amphicyonids are Vallesian spe-
mandible from northwest Nebraska named "Aelurodon" cies documented primarily by isolated teeth (Kuss, 1965;
meandrinus by Hatcher (1894). I follow Webb (1969) in Beaumont, 1984): (l) huge camassials and enlarged crush-
considering Hatcher's species a nomen dubium due to ing posterior molars referred to Amphicyon; (2) smaller
the fragmentary state of the teeth. Although it is evident molars reminiscent of Cynelos; and (3) occasional M2s
that the fossil is probably lschyrocyon, the jaw could not that call to mind M2 in P seudocyon and I schyrocyon, which
be used to define a species within the genus. Therefore, is transversely wide and anteroposteriorly narrow, yet
"Dinocyon" gidleyi (AMNH 10671) becomes the geno- smaller than Ml. These teeth suggest that large hypercar-
typic species, based on a skull and partial skeleton de- nivorous beardogs much like the Pseudocyon-lschyrocyon
scribed by Matthew (1902). The genus itself is based on group also existed in Eurasia in the Vallesian interval but
a juvenile mandible with nearly complete dentition from are poorly known.
the Little White River drainage of South Dakota, AMNH
10802, the holotype of I. hyaenodus Matthew and Gidley
(1904). The species gidleyi has priority over hyaenodus for BIOLOGY AND EVOLUTIONARY PATTERNS
this large late Clarendonian beardog. "Pliocyon" walkerae
Johnston and Christian (1941), based on a skull from the The 30-million-year history of amphicyonids demonstrates their lon-
Clarendon beds, Texas, is also a synonym of lschyrocyon gevity as a viable group of predatory camivorans (Figures 11.3, 11.4).
 Amphicyonidae 217

w
z
w L L.BI. 1.8
2.4
E.BI
0
4.5
0 E LaL Hp.
:i L.Hp.
5.2
6.0
ll. LE.Hp.
7.0

L
E.E.Hp.
... c:
-o-
c:

~1=
L.CI.
-~
9.5

w
E.CI.
11.0
_.g e::io. -
z
w
LL.Ba. i
-;f_ 'fi
0 M
E.L.Ba.
12.5
....
II) -

0 14.0

:iii e.aa.

L.Hm.
...
17.5
E.Hm.
18.8
E LL.Ar.
19.2

E.L.Ar.
23.0

w
z
w
L.E.Ar.

L
0 27.7
0
CJ E.E.Ar. c:
:i 0
0 29A
-~-
Wt.
.!!8
E
Or.
31.9
~-s.- -~

i-tr~
33.4
L.Ch.
-"5 -.c:: -
ij
34.5
L M.Ch.
35.S Q. -111 _Q._
E.Ch.
37.1 ~ -~ ~ -
Du

39.5

L.Un.

w 41.3
z
w E.Un.
0 45.9 DAPHOENINAE TEMNOCYONINAE AMPHICYONINAE
0
w M
L.Br.
45.7

M.ar.
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2

w
z
w L
0
0
w 11.1
..J
<( To.3
ll. To.2
To.1
E Pu.3
Pu.2
Pu.1
Pu.o

Bl.=Blancan, Hp.= Hemphillian, Cl.= Clarendonlan, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Ulntan, Br.= Bridgerlan, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonlan, Pu. = Puercan.
Figure 11.4. Temporal ranges of North American amphicyonid genera.
 Robert M. Hunt, Jr. 218

They first appeared on the northern continents, spreading south- adaptive features of limbs and feet. Two broad skeletal categories
ward into Africa and southeastern Asia in the early Miocene, but appear to have evolved from the generalized amphicyonid postcra-
failed to enter South America due to their extinction before the nial skeleton found inDaphoenus and other early species: (1) Some
establishment of a land connection between the Americas in the lineages (daphoenines, temnocyonines) emphasized cursorial adap-
Pliocene. The earliest species are animals of small body size, less tations: elongation of lower limb segments, limitation of ability to
than 20 kg. Progressive size increases are evident in most lin- pronate/supinate the forelimb, and elongation of the feet including
eages, culminating in medial and late Miocene carnivorans exceed- appression of the metapodials; (2) others maintained the proportions
ing 200 kg. of the plesiomorphic postcranial skeleton yet scaled its dimensions
The replacement of amphicyonids in the late Miocene of Europe for large body size, resulting in a massive, heavily muscled, bearlike
and North America by large ursids, felids, and canids may be nearly skeletal frame (amphicyonines).
synchronous at about 9 Ma. Relict species survive into the latest In contrast to living ursids, amphicyonids never entirely lose the
Miocene in south Asia (7 Ma) and perhaps Africa. There is no shearing function of the carnassials, most maintaining a plesiomor-
reliable Pliocene record of the family anywhere in the world. phic arctoid dentition in which shear is focused at the carnassial lo-
Feeding preferences of beardogs must have ranged widely from cus and crushing/shearing are accomplished by postcamassial molar
carnivorous to omnivorous and would have been adjusted to the teeth. Dental reduction in amphicyonids is commonly restricted to
size of prey based on the average size of the beardog. The strong loss or size decrease of the premolars. Scavenging of carcasses was
sexual dimorphism, particularly for the larger species, may indi- probably within the domain of many amphicyonid species; whether
cate that the males were polygynous and wide ranging. Postcranial some of these animals favored such a life mode is not known from
skeletons demonstrate that locomotor patterns differed, based on the fossil evidence.

iii" iii" ~ ~ ~
(j"
(j" ~ !!!. .,
!!!. 2..
., ., ~ z (j"
~
~ !!:..
~ -ii8.,
.. !g,
"Iii
e·=
c c
c- c~ .5 .. "i-o..
c
i! ]i E·i
.,_ u'ti "Iii8
~ "ii!-~ 0 ~m j~ GI-
.c 0..
c,..
.c 0.. =~ (J
(J
:; 1a t: ... u .c
MA ~ e., c~ !§
(J (J t: ...
0 I!! ~ 1i ~! ~~ ~I!!
.. t:
0.. 0
u.i
J:l;~ ~~ Cl ~oll ZCl
UI ~
Cl Cl Cl Cl z z
1.8 ,

~
c: ' I
Q)
"'c:
0
I
/
/
I
jffi I 1,

x
w 2.4
zw

IX ~
c: I
(.) >-
"i:
"'ffi
0
I
0 ~as

xx x
::::i 4.5
Q.
(;)., Ea. .!!!c:
5.2
j :c
,.__ a>=
J:
I
I
'

IX /
/

~ x
/ /
c. c: I /
/
s~:m I
j J: '.E 1,
/
,/
6.0

~ IX
.2' '
m ffi
t><
a.
w~= Figure 11.5. Biogeographic ranges
_j J: :c
7.0
~ 6. c:
. , / . of North American amphicyonid gen-

IX
/ / / /
w E.!!! era. A "box" (for a particular time
/ /
w~:E 1, 1,
/
,, / I/ period in a particular biogeographic
8.8 /
I:. c:
/
region) that has a cross through it
I / Is, / I
JB ~ -~ Is means no fossil localities are known
zw 9.5 jQ .g
w 1...
I
1,
/ Ps
I/
/
II
/
1X for that time period from that area;
/
(.) c: c: Is, I I a single dashed line through the box
/
0
:5
~2:?-~
~"'
~u
c:
.g t>< Is Is
Ps / I I means only scant fossil information

x
II II II is available (usually only a single,
11.0 , , / /

x
Q) b /
I small, locality). Key: Daphoeninae are
/ / / /
jf? c: Is, Ps
I I
~ -~ / /
,, / / in roman type: Br = Brachyrhyn-

x
_j 1, It
1, II II
12.5 / chocyon, Dd = Daphoenodon, Di =
.-
2l b ls,P~ /
i
Daphoenictis, Ds = Daphoenus, Pd =

14.0
...J
w~·~
!!! c: Is Ps
IX I;/
I
Paradaphoenus. Temnocyoninae are
b I in italics: Ma = Mammacyon, Te =

15.8
~f? c:
"'"'"'
w al ·s; I
/ Am,
Cy,
?Pc
,
Am,
Pc
Am,
Cy
x Am, Cy,
Pc,Ps
Am I Temnocyon. Amphicyoninae are in
boldface: Am = Amphicyon, Cy =

x
Cynelos, Is= lschyrocyon, Pc= Plio-

IX
Clc:
/ I
c:"' /
.sd~ 'E0 /
Am,Cy, Pc / I cyon, Ps = Pseudocyon, Ys = Ysen-
"'Q)
...J J: - / [/ II grinia.
17.5
 Amphicyonidae 219

Br, Di,
Br, Di, Ds
Ds

Ds Ds

Figure 11.5. (Cont.)

The European Oligocene radiation of amphicyonids produced is followed in the later Miocene by additional amphicyonine migra-
distinct species not seen in North America; similarly the less species- tion events (Figure 11.3).
rich North American Oligocene radiation is marked by its own en- The influx of amphicyonines into North America is not matched
demic taxa (Figure 11.5). Despite this, the strong similarity in cranial by invasion of the Old World by North American species. Thus the
anatomy betweenDaphoenus and Cynodictis is evidence of the unity abrupt appearance of amphicyonid and ursid carnivores in North
of the family and their descent from a common ancestor at some America during the Miocene epoch seems to mark a series of im-
time in the Eocene. Most Miocene amphicyonids in Europe can be portant migration pulses from Eurasia, a unidirectional movement of
traced directly into Oligocene European species, demonstrating the arctoid carnivorans in which earlier species are periodically replaced
continuity of these lineages on that continent (the amphicyonine ra- by new arrivals during the Neogene. North America seems incapable
diation). As the Asian record improves, one would predict a diverse of evolving its own large arctoid carnivorans in the Miocene, relying
amphicyonine fauna that may or may not be diluted to some degree at times of fauna) turnover on resupply from the arctoid homeland
by New World daphoenines. But in North America the early am- of Eurasia.
phicyonid record is dominated by daphoenines and temnocyonines,
and amphicyonines are absent. Then, in the early Miocene, when
only a single daphoenine lineage (Daphoenodon) remains in North REFERENCES
America, typical Old World amphicyonine genera appear almost
simultaneously in the North American midcontinent. This apparent Barry, J., & Flynn, L. (1989). Key biostratigraphic events in the Siwa-
movement of amphicyonids to the New World in the early Miocene lik sequence. In European Neogene Mammal Chronology, eds.
 Robert M. Hunt, Jr. 220

E. H. Lindsay, V. Fahlbusch, & P. Mein, pp. 557-71. New York: (1977). Basicranial anatomy of Cynelos Jourdan (Mammalia, Car-
Plenum. nivora), an Aquitanian amphicyonid from the Allier Basin, France.
Beaumont, G. de (1982). Breves remarques sur la dentition de certains Ur- Journal of Paleontology, 5, 826-43.
sides (Mammiteres). Archives des Sciences, Geneve, 35, 153-6. Hunt, R. M., Jr., Xue X.-X., & Kaufman, J. (1983). Miocene burrows of
( 1984). Des dents d 'Amphicyon (Mammitere, Carnivore, Urside) du Tur- extinct beardogs: indication of early denning behavior of large mam-
olien basal de Kohfidisch, Burgenland, Autriche. Archives des Sci- malian carnivores. Science, 221, 364-6.
ences, Geneve, 37, 75-83. Johnston, C. S., & Christian, W. G. (1941). Pliocyon walkerae, a new
Berta, A., & Galiano, H. (1984). A Miocene amphicyonid (Mammalia: Car- Pliocene canid from Texas. Journal of Paleontology, 15, 56-60.
nivora) from the Bone Valley Formation of Florida. Journal of Ver- Jourdan, C. (1862). Description de restes fossiles de grands Mammiteres.
tebrate Paleontology, 4, 122-5. Rev. Soc. Sav., 1, 126-30.
Blainville, H. M. D. (1841). Osteographie et Description Iconographique Kurten, B. (1969). Sexual dimorphism in fossil mammals. In Sexual Di-
des Mammiferes Recents et Fossiles (Carnivores), 2 vols. Paris. morphism in Fossil Metazoa and Taxonomic Implications, ed. G.
Cope, E. D. (1878). On some characters of the Miocene fauna of Oregon. Westermann, pp. 226-33. Stuttgart: International Union of Geolog-
Paleontological Bulletin, 30, 1-16. ical Sciences, Series A, No. I.
Crusafont-Pair6, M., & Kurten, B. (1976). Bears and beardogs from the Kuss, S. (1960). Haplocyonoides ponticus n. sp., ein neuer aberranter Canide
Vallesian of the Valles-Penedes Basin, Spain. Acta Zoologica Fen- aus dem Unterpliocan von Melchingen. Ber. Natulforschung Gesell.
nica, 144, 1-29. Freiburg i. Br., 50, 247-50.
Dawson, M. R. (1980). Paleontology and geology of the Badwater Creek (1965). Revision der Europfilschen Amphicyoninae (Canidae, Car-
area, central Wyoming, Part 20. Annals of the Carnegie Museum, 49, nivora, Mammalia) ausschliesslich der voroberstampischen Formen.
79-91. Sitzungsberichte der Heidelberger Akademie der Wissenschaften,
Dehm, R. (1950). Die Raubtiere aus dem Mittel-Miocan (Burdigal- Jahrgang 1965, I. Abhandlung, pp. 1-168.
ium) von Wintershof-West bei Eichstatt in Bayern. Abhandlun- Lartet, E. (1836). Nomenclature des mammiteres et des coquilles qu'il a
gen der Bayerischen Akademie der Wissenschaften (Mathematisch- trouves dans un terrain d'eau douce pres de Simorre et de Sansan
natwwissenschaftliche Klasse), n.f., Heft 58, 1-141. (Gers). Bulletin de la Socihe geologique de France, 7, 217-20.
Fingerut, M., Taylor, D., Hunt, R. M., & Stepleton, E. (1993). A new beardog (1851). Notice sur la colline de Sansans. Annuaire du departement du
from the Upper John Day beds, Oregon: evidence for the anatomical Gers, pp. 1-42.
structure of the auditory bulla in amphicyonid carnivorans. Proceed- Leidy, J. (1853). Untitled communication. Proceedings of the Academy of
ings of the Nebraska Academy of Sciences, /03rd Annual Meeting, Natural Sciences of Philadelphia, 6, 392-4.
Abstracts, p. 58. Loomis, F. B. (1931). A new Oligocene dog. American Journal of Science,
Ginsburg, L. (1961). La faune de carnivores Miocenes de Sansan. Memoires 22, 100--2.
du Museum National d' Histoire Nature/le (Serie C), Sciences de la (1936). Three new Miocene dogs and their phylogeny. Journal of Pale-
Terre, 9, 1-190. ontology, 10, 44-52.
(1965). L' "Amphicyon" ambiguus des Phosphorites du Quercy. Bulletin Matthew, W. D. (1902). A skull of Dinocyon from the Miocene of Texas.
du Museum National d'Histoire Naturelle (Serie 2), 37, 724-30. Bulletin of the American Museum of Natural History, 16, 129-36.
(1977). Cynelos lemanensis (Pomel), carnivore urside de l'Aquitanien (1918). Contributions to the Snake Creek fauna. Bulletin of the American
d'Europe. Anna/es de Paleontologie (Vertebres), 63, 57-104. Museum of Natural History, 38, 183-229.
(1989). The faunas and stratigraphical subdivisions of the Orleanian in Matthew, W. D., & Gidley, J. (1904). New or little known mammals from
the Loire basin (France). InEuropeanNeogene Mammal Chronology, the Miocene of South Dakota, Parts I-III. Bulletin of the American
eds. E. H. Lindsay, V. Fahlbusch, & P. Mein, pp. 157-76. New York: Museum of Natural History, 20, 241-68.
Plenum. Mein, P. (1989). Updating of MN zones. In European Neogene Mammal
Ginsburg, L., Huin, J., & Locher, J.-P. (1981). Les carnivores du Miocene Chronology, eds. E. H. Lindsay, V. Fahlbusch, & P. Mein, pp. 73-90.
inferieur des Beilleaux a Savigne-sur-Lathan (Indre-et-Loire). Bul- New York: Plenum.
letin du Museum National d' Histoire Naturelle (Serie 4), 3, 183-94. Peterson, 0. A. (1907). The Miocene beds of western Nebraska and eastern
Ginsburg, L., Le Van Minh, Kieu Qui Nam, & Din Van Thuan. (1992). Wyoming and their vertebrate fauna. Annals of the Carnegie Muse um,
Premieres decouvertes de vertebres continentaux dans le Neogene 4, 21-72.
du Nord du Vietnam. Comptes Rendus de l'Academie des Sciences, (1909). A new genus of carnivores from the Miocene of western Ne-
Paris, 315, 111-16. braska. Science, 29, 620--1.
Ginsburg, L., & Telles-Antunes, M. (1968). Amphicyon giganteus, carnassier (1910). Description of new carnivores from the Miocene of western
geant du Miocene. Anna/es de Paleontologie, 54, 1-32. Nebraska. Memoir of the Carnegie Museum, 4, 205-78.
Hatcher, J. B. (1894 ). On a small collection of vertebrate fossils from the Pomel, A. (1846). Memoire pour servir a la geologie paleontologique des
Loup Fork Beds of northwestern Nebraska, with a note on the geology terrains tertiaires du departement de I' Allier. Bulletin de la Societe
of the region. American Naturalist, March 1894, 236-48. geologique de France, 3, 353-73.
Hendey, Q. B. (1978). Preliminary report on the Miocene vertebrates from Russell, D. E., & Zhai, R.-J. (1987). The Paleogene of Asia: mammals and
Arrisdrift, South West Africa. Annals of the South African Museum, stratigraphy. Memoires du Museum National d'Histoire Naturelle,
76, 1-41. Serie C, 52, 1-488.
Hough, J. R. (1948). A systematic revision of Daphoenus and some allied Savage, R. J. (1965). Fossil mammals of Africa, 19: The Miocene Carnivora
genera. Journal of Paleontology, 22, 573-600. of east Africa. Bulletin of the British Museum (Natural History),
Hunt, R. M., Jr. (1972). Miocene amphicyonids (Mammalia, Carnivora) Geology, 10, 239-316.
from the Agate Spring Quarries, Sioux County, Nebraska. American Schlosser, M. (1899). Ueber die Baren und barenahnlichen Formen des eu-
Museum Novitates, 2506, 1-39. ropaischen Tertiars. Stuttgart: Palaeontographica, 13-14.
(1974a). The auditory bulla in Carnivora: an anatomical basis for reap- Schmidt-Kittler, N. (1987). The Carnivora (Fissipedia) from the Lower
praisal of carnivore evolution. Journal of Morphology, 143, 21-76. Miocene of east Africa. Palaeontographica (Abt. A), 197, 85-126.
(1974b). Daphoenictis, a cat-like carnivore (Mammalia, Amphicy- Scott, W. B. (1898). Notes on the Canidae of the White River Oligocene.
onidae) from the Oligocene of North America. Journal of Paleontol- Transactions of the American Philosophical Society, 19, 327-415.
ogy, 48, 1030-47. Scott, W. B., & Jepsen, G. L. (1936). The mammalian fauna of the White
 Amphicyonidae 221

River Oligocene (Insectivora and Carnivora). Transactions of the Hemphillian (late Oligocene through earliest Pliocene epochs) in
American Philosophical Society (n.s), 28, 1-149. North America. In Cenozoic Mammals of North America, ed. M. 0.
Steininger, F., Bernor, R., & Fahlbusch, V. (1989). European Neo- Woodburne, pp. 153-210. Berkeley: University of California Press.
gene marine/continental chronologic correlations. In European Tobien, H. (1980). Taxonomic status of some Cenozoic mammalian local
Neogene Mammal Chronology, eds. E. H. Lindsay, V. Fahlbusch, faunas from the Mainz Basin. Mainzer geowissenschaften Mitt., 9,
& P. Mein, pp. 15-46. New York: Plenum. 203-35.
Stock, C., & Furlong, E. L. (1926). New canid and rhinocerotid remains from Vire!, J. (1929). Les faunes de mammiteres de !'Oligocene superieur de la
the Ricardo Pliocene of the Mohave Desert, California. University Limagne Bourbonnaise. Anna/es de l' Universite de Lyon (n.s.), 47,
of California Publications in Geological Sciences, 16, 43-60. 1-313.
Tedford, R. H., & Frailey, D. (1976). Review of some Carnivora (Mam- Webb, S. D. (1969). The Burge and Minnechaduza Clarendonian mam-
malia) from the Thomas Farm local fauna (Hemingfordian: Gilchrist malian faunas of Nebraska. University of California Publications in
County, Florida). American Museum Novitates, 2610, 1-9. Geological Sciences, 78, 1-191.
Tedford, R.H., Skinner, M. S., Fields, R. S., Rensberger, J.M., Whistler, D. Wortman, J., & Matthew, W. D. (1899). The ancestry of certain members
P., Galusha, T., Taylor, B. E., Mcdonald, J. R., & Webb, S. D. (1987). of the Canidae, Viverridae & Procyonidae. Bulletin of the American
Fauna! succession and biochronology of the Arikareean through Museum of Natural History, 12, 109-39.
 Appendix: Key to fossil localities

Numbers in parentheses indicate the unified localities for the entire 12. Chadron Formation, Crazy Johnson Member, Indian Creek,
volume listed in Appendix I. Pennington Co., South Dakota (CP83B)
13. Chadron Formation, Peanut Peak Member, Shannon Co.,
South Dakota (CP83C)
Duchesneau (Daphoenus) 14. Chadron Formation, Seamen Hills area, Niobrara Co.,
Wyoming (CP41A)
I. Wagon Bed Formation, Badwater Loe. 20, Fremont Co., 15. White River Formation, Shirley Basin, Natrona Co.,
Wyoming (CP29D) Wyoming (CP39D)
2. Wagon Bed Formation, Badwater Wood Locality, Fremont 16. Cypress Hills Formation, Calf Creek local fauna, Cypress
Co., Wyoming (CP29D) Hills, Saskatchewan (NPlOB)
3. West Canyon Creek, Beaver Divide, Fremont Co., Wyoming
(CP39IIA)
4. Cypress Hills Formation, Lac Pelletier lower fauna, Cypress Chadronian (Daphoenictis)
Hills, Saskatchewan (NP9A)
5. Chambers Formation, Porvenir local fauna, Presidio Co., 20. Chadron Formation, Norman Ranch local fauna, Sioux Co.,
Texas (SB44B) Nebraska (CP98B or C)
21. Chadron Formation, Brecht ranch local fauna, Dawes Co.,
Nebraska (CP98B or C)
22. White River Formation, Mink Creek local fauna, Teton Co.,
Duchesneau (Daphoenictis)
Wyoming (CP43)
23. White River Formation, Flagstaff Rim, 10' above Ash E,
6. Chambers Formation, Porvenir local fauna, Presidio Co.,
Natrona Co., Wyoming (CP39C)
Texas (SB44B)
24. White River Formation, Flagstaff Rim, Dry Hole Quarry, Na-
trona Co., Wyoming (CP39B)
25. Renova Formation, Pipestone Springs Main Locality,
Chadronian (Daphoenus)
Jefferson Co., Montana (NP24C)
26. Cypress Hills Formation, Cypress Hills, Saskatchewan
7. Chadron Formation, Chadronia Pocket, Dawes Co., Nebraska
(NPlOB)
(CP98C)
8. Chadron Formation, Arner Ranch area, Sioux Co., Nebraska
(CP98B or C)
9. Chadron Formation, sect. 15, T.33N, R.53W (Sx-35), Sioux Chadronian (Brachyrhynchocyon)
Co., Nebraska (CP98B or C)
10. Chadron Formation, Pasture 34 (Dw-107), Dawes Co., Ne- 30. Chadron Formation, Arner Ranch area, Sioux Co., Nebraska
braska (CP98B or C) (CP98B or C)
11. Chadron Formation, Everson Ranch, Sioux Co., Nebraska 31. Chadron Formation, Whitehead Creek, Sioux Co., Nebraska
(CP98B or C) (CP98B or C)

222
 Key to fossil localities 223

32. Chadron Formation, near Sugarloaf, Sioux Co., Nebraska 62. Roundtop area, Dawes Co., Nebraska
(CP98B or C) 63. Badland Creek, Hat Creek Basin, Sioux Co., Nebraska
33. Chadron Formation, Brecht Ranch local fauna, Dawes Co., 64. Dry Creek, Hat Creek Basin, Sioux Co., Nebraska
Nebraska (CP98B or C) 65. Prairie Dog Creek, Hat Creek Basin, Sioux Co., Nebraska
34. Chadron Formation, Chadronia Pocket, Dawes Co., Nebraska 66. Cedar Creek, Hat Creek basin, Sioux Co., Nebraska
(CP98B or C) 67. NW of Crawford, Sioux Co., Nebraska
35. Chadron Formation, Rook Ranch (Dw-104), Dawes Co., Ne- 68. Plunkett-Parsons Ranch area, N of Harrison, Sioux Co., Ne-
braska (CP98B or C) braska
36. Chadron Formation, Pasture 34 (Dw-107), Dawes Co., Ne- 69. 10 miles N of Harrison, Sioux Co., Nebraska
braska (CP98B or C) 70. Sect. 11, T.33N, R.54W, Sioux Co., Nebraska
37. Chadron Formation, Bartlett West #1, Dawes Co., Nebraska 71. 1.5 miles SW of Dout Ranch, Sioux Co., Nebraska
(CP98B or C) 72. Brecht Ranch area, Dawes Co., Nebraska
38. Chadron Formation, Ahearn and Crazy Johnson Members, 73. Whitehead Creek, Sioux Co., Nebraska
Indian Creek, Pennington Co., South Dakota (CP83A, 74. Warbonnet Creek, Sioux Co., Nebraska
CP83B) 75. Geike Ranch, N of Harrison, Sioux Co., Nebraska
39. Chadron Formation, Hutenmacher Table, Peanut Peak Mem-
ber, Pennington Co., South Dakota (CP83C)
40. Chadron Formation, Breaks, Crazy Johnson Member, Custer
Scenic Member, Brule Formation (localities nos. 80-97 are
Co., South Dakota (CP83B)
CP84A):
41. Chadron Formation, Seamen Hills area, Niobrara Co.,
Wyoming (CP41A) 80. Big Corral Draw, Shannon Co., South Dakota (Lower
42. Chadron Formation, Yoder local fauna, Goshen Co., Nodular Zone)
Wyoming (CP42A) 81. Spring Creek Basin, Pennington Co., South Dakota
43. Chadron Formation, Irvine Bridge area, Converse Co., 82. East side of Hamey Springs Range, SE of Sheep Mountain,
Wyoming (CP40A) South Dakota
44. White River Formation, Flagstaff Rim, below Ash B, Natrona 83. Chamberlain Pass, SE of Scenic, Pennington Co., South
Co., Wyoming (CP39B) Dakota
45. White River Formation, Flagstaff Rim, lower part (Ash B to 84. 6 miles W of Interior, Jackson Co., South Dakota
F), Natrona Co., Wyoming (CP39C) 85. Little Corral Draw, Shannon Co., South Dakota (Lower
46. White River Formation, Flagstaff Rim, upper Part (Ash F to Nodular Zone)
J), Natrona Co., Wyoming (CP39F) 86. Red Shirt Table, Shannon Co., South Dakota
47. White River Formation, Bates Hole/Ledge Creek, Natrona 87. 2 miles and 3 miles E of Scenic, Pennington Co.,
Co., Wyoming (CP39G) South Dakota
48. White River Formation, Cedar Creek area, Logan/Weld Cos., 88. 11 miles S of Scenic, Shannon Co., South Dakota
Colorado (CP68B) 89. Cain Creek, Pennington Co., South Dakota
49. Renova Formation, Pipestone Springs Main Locality, Jeffer- 90. Bear Creek Basin, Pennington Co., South Dakota
son Co., Montana (NP24C) 91. S Arrow Wound Table, Shannon Co., South Dakota
50. Renova Formation, Sage Creek, Beaverhead Basin, Montana 92. Sheep Mountain area, Shannon Co., South Dakota
(NP24E) 93. Cottonwood Pass area, Jackson Co., South Dakota
51. Climbing Arrow Formation, Thompson Creek, Three Forks 94. 1 mile E of Babby Butte, Shannon Co., South Dakota
Basin, Montana (NP23C) 95. 6 miles S of Scenic, Shannon Co., South Dakota
52. Unnamed unit, Hadcock Ranch, Sewell Lake area, Montana 96. 3.5 miles S of Scenic, Pennington Co., South Dakota
(NP29C) 97. 7-8 miles N of Rockyford, South Dakota
53. Cypress Hills Formation, Calf Creek local fauna, Cypress 98. White River Formation, NW of Whitman, S-Bar Creek, Nio-
Hills, Saskatchewan (NPlOB) brara Co., Wyoming (CP41B)
54. Chambers Formation, Little Egypt local fauna, Reeves Bone 99. White River Formation, about 6 miles SE of Douglas, Con-
Bed, Presidio Co., Texas (SB44C) verse Co., Wyoming (CP40B)
100. White River Formation, Bald Butte, Niobrara Co., Wyoming
(CP41B)
Orellan (Daphoenus) l 01. White River Formation, Wulff Ranch, Douglas area, Con-
verse Co., Wyoming (CP40B)
102. White River Formation, Walker Ranch area, Niobrara Co.,
Orella Member, Brule Formation (all locality CP99A):
Wyoming (CP41B)
60. 2.5 miles N of Chadron, Dawes Co., Nebraska 103. White River Formation, 8 miles SE of Douglas, Converse
61. Harrison Ranch, Sioux Co., Nebraska Co., Wyoming (CP40B)
 Key to fossil localities 224

104. White River Formation, 8 miles E of Douglas, Converse Co., Whitneyan or early Arikareean
Wyoming (CP40B) (Paradaphoenus)
105. White River Formation, Greeley/Pawnee Buttes area, Weld
Co., Colorado (CP68C) 130. John Day Formation, John Day River valley, Oregon (PN6)
106. White River Formation, N of Stone Ranch, Yuma Co., Col- 131. Brule Formation, Scottsbluff Co., Nebraska (possibly Orel-
orado (CP68C) lan) (CP99A or B)
107. White River Formation, Cedar Creek, Logan Co., Colorado 132. Brule Formation, sect. 36, T.20N, R.56W, Banner Co.,
(CP68C) Nebraska (CP99B or C)
108. White River Formation, Casement Ranch, Logan Co., Col- 133. Sharps Formation, Wounded Knee area, LACM Loe. 1981,
orado (CP68C) South Dakota (CP85C)
109. White River Formation, Canyon Ferry area, Montana
(NP29E)
110. White River Formation, Little Badlands, Dickinson area, Whitneyan or early Arikareean
Stark Co., North Dakota (NP50B) (Temnocyon)

135. John Day Formation, Logan Butte, Crook Co., Oregon (PN6C
Whitneyan (Daphoenus) orD)
136. John Day Formation, John Day valley, Turtle Cove, Grant
114. Brule Formation, Roberts Draw, NE of Harrison, Sioux Co., Co., Oregon (PN6C or D)
Nebraska (CP99B) 137. John Day Formation, Morgan's Place, UCMP Loe. 874, Grant
115. Brule Formation, Wolff Lake, South Dakota (CP84B) Co., Oregon (PN6C)
116. Brule Formation, NE of Indian Stronghold, South Dakota
(CP84B)
117. Brule Formation, Big Corral Draw, Shannon Co., South
Dakota (CP84B)
Arikareean (Temnocyon)

138. John Day Formation, McGinnis Ranch, Grant Co., Oregon

Whitneyan or Arikareean (Daphoenus) (PN6ForG)
139. John Day Formation, Haystack Valley, Wheeler Co., Oregon
118. John Day Formation, John Day River valley, Oregon (PN6) (PN6G)
119. John Day Formation, Camp Watson, John Day River valley, 140. Harrison Formation, 0.5 miles W of Agate, Sioux Co., Ne-
Oregon (PN6) braska (CP103)
120. John Day Formation, Turtle Cove, John Day River valley, 141. Upper Harrison beds, American Museum - Cook Quarry,
Oregon (PN6) Sioux Co., Nebraska (CP104A)
121. John Day Formation, Haystack Valley, John Day River valley, 142. Sespe Formation, Kew Quarry, Las Posas Hills, Ventura Co.,
Oregon (PN6) California (CC9C)
122. John Day Formation, LACM Loe. 1843, Malheur Co., Ore- 143. Upper Harrison beds, near Guernsey, Platte Co., Wyoming
gon (PN6) (CP51A)
123. John Day Formation, Blue Basin, Unit C, Grant Co., Oregon
(PN6C)
124. John Day Formation, Sheep Rock area, ?Unit K, Grant Co., Arikareean (Mammacyon)
Oregon (PN6D)
125. Tecuya Formation, Tecuya Canyon, Kem Co., California 150. Sharps Formation, LACM Loe. 1872, south of Wolf ranch,
(CC13) Shannon Co., South Dakota (CP85C)
126. Arikaree Formation, Birdcage Gap (Mo-106), Morrill Co., 151. Arikaree Formation, Porcupine Creek, Shannon Co., South
Nebraska (CPIOOII) Dakota (CP86B)
152. Arikaree Formation, LACM Loe. 1964, Wounded Knee area,
Shannon Co., South Dakota (CP86B)
Orellan (Paradaphoenus) 153. Monroe Creek Formation, Warbonnet Creek, Sioux Co., Ne-
braska (CP102)
127. Brule Formation, Orella Member, UNSM Loe. Sx-14, Sioux 154. John Day Formation, Haystack Valley, Wheeler Co., Oregon
Co., Nebraska (CP99A) (PN6C orD)
128. Brule Formation, Orella Member, UNSM Locs. Sx-25, Sx- 155. Arikaree Formation, N of Keeline, Niobrara Co., Wyoming
26, Sioux Co., Nebraska (CP99A) (CP52)
129. Brule Formation, Scenic Member, SW of Sheep Mt., near 156. Unit 9 Limestone, Martin-Anthony Roadcut, Marion Co.,
Scenic, South Dakota (CP84A) Florida (GC8B)
 Key to fossil localities 225

157. SB-IA local fauna, 1 mile N of Live Oak, Suwanee Co., 182. ?Upper Harrison beds, Morava Ranch Quarry, Box Butte Co.,
Florida (GC8A) Nebraska (CP104A)
183. Upper Harrison beds, Harper Quarry, Sioux Co., Nebraska
(CP104A)
Medial and late Arikareean 184. Upper Harrison beds, Spoon Butte, Lay Ranch, Goshen Co.,
(Daphoenodon) Wyoming (CP51A)
185. Upper Harrison beds, 25-Mile District, Goshen Co.,
158. Brooksville, Hernando Co., Florida (GC8B) Wyoming (CP51A)
159. Franklin Phosphate Co. Pit no. 2, Alachua Co., Florida
(GC8C)
Early Hemingfordian (Ysengrinia)
160. Buda local fauna, Alachua Co., Florida (GC8A)
161. Harrison Formation, Van Tassell Creek, Niobrara Co.,
186. Unnamed lithic unit, Bridgeport Quarries, Morrill Co., Ne-
Wyoming (CP52)
braska (CP106)
162. Upper Harrison beds, Agate Monument den site (Quarry 3),
Sioux Co., Nebraska (CP104A)
163. Upper Harrison beds, Harper Quarry, Sioux Co., Nebraska Late Arikareean (Cynelos)
(CP104A)
164. Upper Harrison beds, American Museum-Cook Quarry, 187. Upper Harrison beds, Quarry 1, Agate Monument, Sioux Co.,
Sioux Co., Nebraska (CP104A) Nebraska (CP104A)
165. Upper Harrison beds, University Quarry, Sioux Co., Ne-
braska (CP104A)
166. Upper Harrison beds, Carnegie Quarry 1, Sioux Co., Ne-
braska (CP104A) Early Hemingfordian (Cynelos)
167. Upper Harrison beds, 18-Mile District, Goshen Co.,
Wyoming (CP51A) 188. Unnamed lithic unit, Bridgeport Quarries, Morrill Co., Ne-
168. Upper Harrison beds, Royal Valley District, Niobrara Co., braska (CP106)
Wyoming (CP51A) 189. Runningwater Formation, Runningwater Quarry, Box Butte
169. Upper Harrison beds, 5 miles SE of Chugwater, Platte Co., Co., Nebraska (CP105)
Wyoming (CP5 IA) 190. Karst sinkhole, Thomas Farm Quarry, Gilchrist Co., Florida
170. Upper Harrison beds, ?Niobrara Canyon, Sioux Co., Ne- (GC8D)
braska (CP104B)
Late Hemingfordian (Cynelos)

192. Sheep Creek Formation, Thomson, Hilltop and Long Quar-

Early Hemingfordian (Daphoenodon)
ries, Sioux Co., Nebraska (CP108A and B)
171. Runningwater Formation, Brown Quarry, Sioux Co.,
Nebraska (CP105)
Early Barstovian (Cynelos)
172. Runningwater Formation, Whistle Creek-Niobrara River di-
vide, Sioux Co., Nebraska (CP105)
194. Unnamed lithic unit, Pawnee Buttes area, Weld Co., Colorado
173. Runningwater Formation, Aletomeryx Quarry, Cherry Co.,
(CP75B)
Nebraska (CP105)
195. Olcott Formation, East and West Sinclair Draws, Sioux Co.,
174. Runningwater Formation, Marsland Quarry, Box Butte Co.,
Nebraska (CPI 10)
Nebraska (CP105)
196. Olcott Formation, Echo Quarry, Sioux Co., Nebraska(CPl 10)
175. Runningwater Formation, Dunlap Camel Quarry, Dawes Co.,
197. Tesuque Formation, Skull Ridge Member, Espanola Basin,
Nebraska (CP105)
New Mexico (SB32B)
176. Runningwater Formation, Hemingford Quarries 7A, 7B, 12B,
198. Fleming Formation, Cold Spring fauna, Washington Co.,
12D, 23, Box Butte Co., Nebraska (CP105)
Texas (GC4E)
199. Barstow Formation, "Rak Division," San Bernardino Co.,
California (NB6B)
Late Arikareean (Ysengrinia)

180. Upper Harrison beds, University Quarry, Agate Monument, Late Hemingfordian (Pliocyon)
Sioux Co., Nebraska (CP104A)
181. Upper Harrison beds, American Museum-Cook Quarry, 200. Sheep Creek Formation, Thomson and Long Quarries, Sioux
Sioux Co., Nebraska (CP104A) Co., Nebraska (CP108A and B)
 Key to fossil localities 226

Early Barstovian (Pliocyon) 221. Barstow Formation, Green Hills fauna, San Bernardino Co.,
California (NB6C)
20I. Olcott Formation, East and West Sinclair Draws, Sioux Co., 222. Barstow Formation, Turbin, Steepside, and Valleyview Quar-
Nebraska (CPI 10) ries, San Bernardino Co., California (NB6C and D)
202. Olcott Formation, Echo, Humbug, and Boulder Quarries, 223. Barstow Formation, Skyline and Oreodont Quarries, San
Sioux Co., Nebraska (CPI 10) Bernardino Co., California (NB6C and D)
203. Unnamed lithic unit, Observation Quarry, Dawes Co., Ne- 224. Butte Creek Volcanic Sandstone, Red Basin local fauna, Mal-
braska (CPI I I) heur Co., Oregon (PN8B)
204. Barstow Formation, "Green Hills," San Bernardino Co., Cal-
ifornia (NB6C)
Early Barstovian (Pseudocyon)

Barstovian or Clarendonian (Pliocyon) 230. Olcott Formation, Trojan Quarry, Sioux Co., Nebraska
(CPllO)
205. Bone Valley Formation, exact age uncertain, Polk Co., Florida
(GClOB)
Medial Barstovian (Pseudocyon)

Early Hemingfordian (Amphicyon) 231. Ogallala Formation, Hottell Ranch Quarries, Banner Co.,
Nebraska (CPI I4A)
206. Runningwater Formation, Dunlap Camel Quarry, Dawes Co., 232. Valentine Formation, Norden Bridge Quarry, Brown Co.,
Nebraska (CP105) Nebraska (CP114A)
207. Runningwater Formation, Marsland and Hovarka Quarries, 233. Tesuque Formation, Pojoaque Member, Pojoaque Bluffs,
Box Butte Co., Nebraska (CP105) Espanola Basin, New Mexico (SB32D)
208. RunningwaterFormation, Hemingford Quarries 7B and I2B,
Box Butte Co., Nebraska (CP105)
209. Runningwater Formation, Cottonwood Quarry, Dawes Co., Late Barstovian (Pseudocyon)
Nebraska (CP105)
234. Valentine Formation, Ripple Quarry, Cherry Co., Nebraska
(CPII4B)
Late Hemingfordian (Amphicyon) 235. Ogallala Formation, Hazard Homestead Loe. C, Hitchcock
Co., Nebraska (CPI I4B)
2I2. Sheep Creek Formation, Thomson and Long Quarries, Sioux
Co., Nebraska (CP108A and B)
2I3. Sheep Creek Formation, Thistle and Greenside Quarries, Early Clarendonian (Pseudocyon)
Sioux Co., Nebraska (CP108A and B)
236. Valentine Formation, Burge Member, Burge Quarry, Cherry
Co., Nebraska (CPI I4D)
Early or medial Barstovian (Amphicyon) 237. Valentine Formation, Burge Member, Lucht Quarry, Brown
Co., Nebraska (CP114D)
2I5. Olcott Formation, East and West Sinclair Draws, Sioux Co., 238. Valentine Formation, Burge Member, June Quarry, Brown
Nebraska (CPI 10) Co., Nebraska (CPI I4D)
2 I 6. Olcott Formation, Echo, Grass Roots (= Quarry 6), Boulder, 239. Valentine Formation, Burge Member, Midway Quarry zone,
Prosynthetoceras, Humbug, and Mill Quarries, Sioux Co., Cherry Co., Nebraska (CPI I4D)
Nebraska (CPI 10) 240. Ash Hollow Formation, Cap Rock Member, Horsethief
2I 7. Pawnee Creek Formation, Horse and Mastodon Quarry, Weld Canyon no. 2, Brown Co., Nebraska (CPI I6A)
Co., Colorado (CP75C)
2I8. Pawnee Creek FomJation, West Quarry, Weld Co., Colorado
(?CP75C) Late Clarendonian (Pseudocyon)
2I9. Trinity River Pit no. I, Texas (GC4C)
220. Tesuque Formation, Skull Ridge Member, Espanola Basin, 241. Snake Creek Formation, Laucomer Member, Olcott Quarry,
New Mexico (SB32B) Sioux Co., Nebraska (CP115B)
 Key to fossil localities 227

242. Ash Hollow Formation, Merritt Dam Member, Kat Quarry Clarendonian (Ischyrocyon)
horizon, Cherry Co., Nebraska (CPI I6B)
243. Ash Hollow Formation, Merritt Dam Member, Clayton 255. Valentine Formation, Burge Member, Burge Quarry, Cherry
Quarry, Brown Co., Nebraska (CP116B) Co., Nebraska (CPI 14D)
256. Valentine Formation, ?Burge Member, Paleo Quarry, Sheri-
dan Co., Nebraska (CP114D)
257. Ash Hollow Formation, Cap Rock Member, Little Beaver B
Medial or late Barstovian (Ischyrocyon) Quarry, Cherry Co., Nebraska (CP116A)
258. Ash Hollow Formation, Cap Rock Member, 2-2.5 miles W
245. Barstow Formation, First Division, Hemicyon Stratum, San of Burton, Keyapaha Co., Nebraska (CPI I6A)
Bernardino Co., California (NB6E) 259. Ash Hollow Formation, Hollow Hom Bear Quarry, Todd Co.,
246. Barstow Formation, Leader, Mayday, Hailstone, and Hidden South Dakota (CP90A)
Hollow Quarries, San Bernardino Co., California (NB6D) 260. Ash Hollow Formation, Big Spring Canyon, Bennett Co.,
South Dakota (CP90A)
261. Ash Hollow Formation, Mission local fauna, Fox Ranch,
Mellette Co., South Dakota (CP90A)
Late Barstovian (Ischyrocyon) 262. Ash Hollow Formation, 3 miles E of Rosebud, Todd Co.,
South Dakota (CP90)
250. Valentine Formation, Ripple Quarry, Cherry Co., Nebraska 263. Snake Creek Formation, Laucomer Member, Olcott Hill,
(CP114B) Sioux Co., Nebraska (CPI 15B)
251. Valentine Formation, Devil's Gulch Member, Devil's Gulch 264. Ogallala Formation, Skillet Creek, L.I7, Quarry I, Donley
Horse Quarry, Brown Co., Nebraska (CPI I4C) Co., Texas (SP2A)
252. Valentine Formation, Devil's Gulch Member, Mizner Slide, 265. Ogallala Formation, Adam Rizley Ranch, Donley Co., Texas
Cherry Co., Nebraska (CP114C) (SP2A)
253. Valentine Formation, Devil's Gulch Member, Verdigree 266. Unnamed lithic unit, 7 miles W of Blair Junction, Esmeralda
Quarry, Knox Co., Nebraska (CPI I4C) Co., Nevada (NB2711)
254. ?Valentine Formation, upper channel S of Nenzel Quarry, 267. Ricardo Formation (= Dove Spring Formation), Last Chance
Cherry Co., Nebraska (CPl 14) Gulch, Kem Co., California (NB7B)
12 Nimravidae
LARRY D. MARTIN

INTRODUCTION

Nimravids appeared simultaneously in the late Eocene of North

America and Eurasia. This suggests that they had an early Holarc-
tic distribution (Martin, 1989). They were catlike in their general
form (short faces, sectorial carnassials, reduction of posterior mo-
lars, retractile and hooded claws) and, although probably as capable
of climbing trees as modem-day leopards (Van Valkenburgh, 1985),
prey capture and most of their daily activity probably took place on
the ground. Like modem felids, they were probably adapted to the
open spaces between trees on the edge of forests or in parklands,
and the diversification of catlike carnivores correlates well with the Figure 12.1. Restoration of Barbourofelisfricki.
early development of parklands and savannas in the late Eocene of
North America.
Although nimravids must ultimately be derived from carnivores the Barbourofelinae, first appeared in the early late Miocene (early
with ordinary canines, the earliest known forms (Dinictis and Clarendonian) and were extinct worldwide by the end of the epoch.
Hoplophoneus) were saber-toothed. As in felids, three types of ca- The diversity of the Barbourofelinae may have been limited by the
nines were present (Martin, 1980): short, round canines (conical- then numerous felids (true cats), which may have been occupying
toothed); highly elongated, flattened canines with fine serrations niches that barbourofelines might otherwise have utilized, and they
(dirk-toothed); and somewhat elongated, more coarsely serrated never attained the diversity seen in the Nimravinae.
flattened canines (scimitar-toothed). The dirk-toothed forms (Eu-
smilus, Hoplophoneus, andBarbourofelis) tended to be short legged
and more plantigrade, whereas the scimitar-toothed forms (Dinic- DEFINING FEATURES OF THE
tis, Pogonodon, and Nimravus) were more cursorially adapted. The FAMILY NIMRAVIDAE
only known conical-toothed nimravid, Dinaelurus, may have been
a cheetahlike cursor.
CRANIAL
The family Nimravidae may be thought of as containing an early
radiation, the Nimravinae, and a later one, the Barbourofelinae. The skull is catlike with a short face, but never with the elongated
Nimravines became extinct in North America during the latest postorbital cranium seen in primitive felids. The width across the
Oligocene or earliest Miocene (the youngest definite records are zygoma is broad in comparison to the skull length. All forms have
in the late early Arikareean, but questionable remains are assigned a posterior carotid foramen (Hough, 1953) and/or an inferior pet-
from the early late Arikareean). There is a long interval (around 5 rosal sinus (Neff, 1983; Bryant, 1991). There is no bilaminarseptum
million years) from the start of the early late Arikareean (early early bulla in the auditory chamber. There is a small vertical septum in the
Miocene) until near the end of the late Hemingfordian (late early anteromedial comer of the middle ear cavity (Bryant, 1991) and a
Miocene) when there were no catlike carnivores in North Amer- hypertrophied and complex styliform process along with an anterior
ica, although true felids first appeared in the later Hemingfordian bony lamina of the caudal entotympanic (Hunt, 1987). The palate is
(see Martin, this volume, Chapter 13). The last group of nimravids, reduced and the vomer exposed (Bryant, 1991). All nimravids retain

228
 Nimravidae 229

divergent opinions suggest a need for caution in formulating an

opinion of relationship. Retractile and hooded claws, a large dis-
tinct parastyle on the deciduous camassial (DP3), and loss of M3
in even the most primitive forms are evidence that the nimravids
are at least feloids, and Bryant's (1991) taxonomic scheme is not
unreasonable. According to Neff (1983), the relationship between
the Barbourofelinae and the Nimravinae may also be distant. Bio-
geographically this may make some sense: The Nimravinae have
their earliest occurrence and greatest diversity in North America,
whereas the Barbourofelinae quite clearly originated in Eurasia.
Figure 12.2. Restoration of the skeleton of Barbourofelis fricki.
INFRAFAMILY

primitive cranial features that are modified in modem felids, includ- Cope (1881) proposed the Family Nimravidae. Besides the nominate
ing a large postglenoid foramen, the condyloid foramen separated subfamily, Kretzoi (1929) proposed Pogonodontinae, Dinictinae,
from the posterior lacerate foramen, and an alisphenoid canal. and Hoplophoneinae. Most of these subfamilies contain only one or
two genera. Schultz, Schultz, and Martin (1970) tried to resolve the
problem by using a tribal system, although one of their tribes, the
DENTAL
Barbourofelinae, has since been elevated to subfamily rank (Bryant,
The incisors lack cingular cusps; with the exception of Dinaelu- 1991). Only two subfamilies are recognized in this chapter, Nim-
rus, the upper canine is enlarged, flattened, and serrated; the lower ravinae and Barbourofelinae, and there is no division into tribes.
canine is reduced. M2-3 are absent, and Ml is reduced with an ec- Hypotheses of relationships within the Nimravidae are very much
toloph on the posterior margin. The parastyle on P4 is absent in most influenced by traits connected with the evolution of the saberlike ca-
forms, but present in Eusmilus and members of the Barbourofelinae. nines. It is clear that the bladed upper canines must be derived at
The deciduous carnassial (DP4) has a large parastyle. All nimravids some level and such teeth have evolved independently in a number
whose deciduous dentition is known show delayed eruption of en- of different mammals. Therefore, it seems reasonable to assume that
larged deciduous canines (Bryant, 1991). The m3 is absent. the basal (as yet unknown) nimravid was a conical-toothed form,
even though the earliest known nimravids are all saber toothed. This
leaves open the question of whether Dinaelurus, the only known
POST CRANIAL
conical-toothed nimravid, separated off the main stem before the
The primitive members of the group are all plantigrade with some evolution of saber teeth or whether it represents a later character
capability for wrist and ankle rotation (see Figure 12.2). Van Valken- reversal for this trait. If the conical teeth do not represent a charac-
burgh (1985) is probably right in her assertion that most nimravids ter reversal, and if Dinaelurus is indeed a nimravine, this implies
retained capabilities for tree climbing. The feet of the longer-limbed that saber teeth evolved twice within the Nimravidae (within both
forms such as Nimravus, Pogonodon, andDinaelurus have not been subfamilies), andBarbourofelis may be even more distantly related
described, but it is likely that these taxa were digitigrade. The ungual to the main line of nimravid evolution than Dinaelurus (Neff, 1983).
phalanges are hooded and retractile in all known cases. Figure 12.3 illustrates a hypothesis of relationships between nim-
ravid genera. At the specific level only Hoplophoneus has had an
extensive review (Simpson, 1941; Hough, 1949). Scott and Jepsen
SYSTEMATICS ( 1936) also provide a discussion of the Oligocene forms.

SUPRAFAMILY
INCLUDED NORTH AMERICAN GENERA IN THE
Nimravids appear to be very similar to modem cats (felids) and occur FAMILY NIMRAVIDAE
earlier in time. Many authors have postulated a close phylogenetic
relationship, deriving felids from some generalized nimravid (Cope, The locality numbers listed for each genus refer to the list of unified
1880; Matthew, 1910; Schultz, Schultz, and Martin, 1970). Other localities in Appendix I. The locality numbers may be listed in a
authors consider the two families to be sister taxa (Piveteau, 1931; couple of alternative ways. The acronyms for museum collections
Martin, 1980). More recently, the two families have been consid- are listed in Appendix III.
ered as more distantly related. Bryant (1991) considered the Nim- Parentheses around the locality (e.g., [CPlOl]) mean the taxon
ravidae to be the sister group of the other feloids (felids, viverrids, in question at that locality is cited as an "aff." or "cf." the taxon in
and hyaenids), and Flynn and Galiano (1982) suggested a sister- question. Parentheses are usually used for individual species, thus
group relationship to the caniforms (canids, ursids, mustelids, and implying the genus is firmly known from the locality, but the actual
procyonids). Neff (1983) went so far as to make the Nimravidae species identification may be questionable. Question marks in front
the sister taxon to all other extant carnivore families. Such widely of the locality (e.g., ?CPlOl) mean the taxon is questionably known
 Larry D. Martin 230

Nanosmilus/
Hoplophoneus Eusmilus
Dinictis Nimravus
7
10 12
Dinaelurus Pogonodon
Barbourofelis
6 13
4 11
9
5
3

Figure 12.3. Proposed nimravid relationships. Key to numbers on cladogram: (1) Delayed eruption of deciduous upper canine and consequent delay
in eruption of the permanent canine (Bryant, 1991). (2) BARBOUROFELINAE: Grooved upper canines and completely ossified auditory bullae. (3)
NIMRAVINAE: Incomplete auditory bullae with a thick endotympanic containing a central region of cancellous bone (Neff, 1983). (4) Domed skull with
enlarged external and internal nares. (5) Bladelike upper canines. (6) Enlarged paramastoid processes. (7) Shortened distal limb segments. (8) Parastyle on
the upper carnassial. (9) Elongated distal limb segments. (10) Canine relatively elongated. {11) Canine shortened and coarsely serrated. (12) Masseteric
torus (Toohey, 1959) on the mandible. (13) Enlarged dependent flange on the mandible.

from that locality, thus implying some doubt that the taxon is actually formula 13/3, Cl/l, P3/3, Ml/2 (D.felina); canine shorter
present at that locality, either at the genus or the species level. and broader than in Hoplophoneus, but longer and nar-
rower than in Nimravus; protocone on P4 even with the
paracone; parastyle absent on P3-P4; mandible with small
NIMRAVINAE
dependent flange. Occiput inclined; paramastoid process
Characteristics: Incomplete auditory bulla; large, thick entotym- only slightly larger than the paraoccipital process. Distal
panic with an inner layer of cancellous bone; endotympanic fails segments of limbs relatively more elongate than in Hoplo-
to join the ectotympanic; parastyle (when present) not formed from phoneus.
the cingulum; distinct teres major process on the scapula (Hunt, Average length ofml: 16.8 mm.
1987; Bryant, 1991). Included species: D .felina Leidy, 1854 (= Daptophilus squa-
lidens Cope, 1873; Dinictisfortis Adams, 1895; D. bomb-
Nimravus Cope, 1879 (synonyms: Archaelurus; ifrons Adams, 1895) (known from localities CP68C,
Machairodus, in part; Dinictis, in part) CP83C, CP84B, CP98A, CP98C, CP99A, NPlOB); D. cy-
Type species: Nimravus brachyops (Cope, 1878) (=Macha- clops Cope, 1879.
irodus). Dinictis sp. (including unsynonymized speciesD. mega-
Type specimen: Lectotype AMNH 6935, partial right ramus. loides) is also known from localities CP39C, CP39F,
Characteristics: Scimitar-toothed nimravid about the size of CP40A, CP40B, CP41A, CP68B, CP84A, CP98B, NP24E,
a leopard. Dental formula 13/3, Cl/l, P3-4/2-4, Ml/1-3; NP50A, NP50B.
upper canine coarsely serrated; protocone reduced on P4;
ml with a distinct talonid, m2 and m3 sometimes present. Pogonodon Cope, 1880 (synonyms: Hoplophoneus, in part;
Palatine anterior to small Ml; alveolar torus (Toohey, Dinictis, in part)
1959) under p4-m3; masseteric fossa elongated (extending Type species: Pogonodon platycopsis (Cope, 1879) (=Ho-
under p4). plophoneus).
Average length of ml: 27.5 mm. Type specimen: AMNH 6938, skull and mandible.
Included species: N. brachyops only (= Machairodus brachy- Characteristics: Scimitar-toothed nimravid with a high sagit-
ops Cope, 1878; Archaelurus debilis Cope, 1879; Nim- tal crest, mountain lion- to jaguar-sized. Dental formula
ravus gomphodus Cope, 1880; N. confertus Cope, 1880; 13/3, Cl/l, P3/3, Ml/1; upper canine short, broad, and
N. meridianus Stock, 1933; N. bumpensis Scott and Jepsen, coarsely serrated; protocone reduced on P4; m2 lost; dis-
1936; N. altidens Macdonald, 1950) (known from localities tinct flange on mandible. Paramastoid process larger than
CC9C, CC12, CP84B, CP85C, CP86A, CP86B, CP99B, in Dinictis but not as enlarged as in Hoplophoneus.
CP101, NP51A, PN6C, PN6D, PN6E or F). Average length of m 1: 24 mm (P. platycopsis ).
Included species: P. platycopsis Cope, 1880 (= Hoplopho-
Dinictis Leidy, 1854 (synonym: Daptophilus) neus platycopsis Cope, 1879; Dinictis platycopsis Adams,
Type species: Dinictisfelina Leidy, 1854. 1896; Pogonodon davisi Merriam, 1906; P. cismontanus
Type specimen: AMNH 455, skull. Thorpe, 1920; P serrulidens Eaton, 1922) (known from lo-
Characteristics: Scimitar-toothed nimravid ranging in size calities CP41B, ?CP41C, CP84A, CP84B, CP85C, NP37,
from that of a large lynx to that of a mountain lion. Dental PN6C, PN6D, PN6E); P. eileenae (=Dinictis eileenae
 Nimravidae 231

Macdonald, 1970) (locality CP85C); P. paucidens cusps; p4 with large paraconid; ml with small metaconid;
( = Dinictis paucidens Riggs, 1895) (locality CP99A). dependent flange on mandible weakly developed. Skull
narrow; glenoid fossa level with the gum line; occiput in-
Comments: Pogonodon is sometimes included in Dinictis, clined posteriorly.
but it seems to have had a long separate history beginning Length of ml: 14.6 mm.
in the Orellan (early Oligocene).
Included species. N. kurteni only, known from locality
CP99A only.
Dinaelurus Eaton, 1922
Type species: Dinaelurus crassus Eaton, 1922. Eusmilus Gervais, 1876 (synonyms: Egmoiteptecela;
Type specimen: YPM 10518, skull. Hoplophoneus, in part; Machairodus, in part)
Characteristics: Conical-toothed nimravid with unserrated Type species: Eusmilus bidentatus (Filhol, 1873) (=Macha-
canines and domed skull, approximately jaguar-sized. irodus).
Dental formula 13/?, Cl/?, P2/?, Ml/?; incisors small, pro- Type specimen: Canine and mandible in the Museum d'His-
tocone reduced on P4. Short face, broad zygoma, external toire Naturelle, Paris.
and internal nares enlarged. Characteristics: Dirk-toothed nimravid, domestic cat- to-
Average length of ml: unknown. Average length of P4: 21.8 bobcat-sized. Dental formula 13/3, Cl/l, P2/l-2, Ml/l;
mm. large parastyle on P4; il reduced; p2 absent; ml lacking
Included species: D. crass us only, known from locality PN6C talonid. Greatly enlarged flange on mandible; vertical oc-
only. ciput.
Average length of ml: 13.0 mm (E. olsontau).
Hoplophoneus Cope, 1874 (synonyms: Machairodus, Included species: E. cerebralis (Cope, 1880) (=Hoplophone-
in part; Eusmilus, in part; Drepanodon; Dinotomius) us cerebralis) (known from Logan Bluff, John Day Forma-
Type species: Hoplophoneus primaevus (Leidy, 1851) tion [within locality PN6]); E. belli (=Hoplophoneus belli
( = Machairodus). Stock, 1933) (locality CC9C); E. olsontau (=Egmoitepte-
Type specimen: USNM 99, skull and mandible. cela olsontau Macdonald, 1963) (locality CP85C).
Characteristics: Dirk-toothed nimravid with finely serrated Eusmilus sp. is also known from localities ?CP39C,
canines, mountain lion- to leopard-sized. Dental formula CP84B, CP99B, PN6D.
13/3, Cl/l, P2-3/2, Ml/1; parastyle absent or indistinct
on P4; large dependent flange on mandible. High sagittal
crest; vertical occiput; enlarged mastoid process. BARBOUROFELINAE
Average length of ml: 18 mm (H. primaevus). Characteristics: grooved upper canines; all teeth serrated; P4 with a
Included species: H. primaevus (Leidy, 1851) (= Machairo- parastyle; lower cheek tooth row extended lateral to the central axis
dus primaevus Leidy, 1851; Drepanodon primaevus Leidy, of the mandibular ramus. Origin for superficial masseter enlarged;
1857; Machairodus oreodontis Cope, 1983; Hoplophoneus ossified, nonseptate auditory bulla. Distal segment of limbs short;
oreodontis Cope, 1874; H. robustus Adams, 1896; H. in- all feet plantigrade.
solens Adams, 1896; H. latidens Thorpe, 1920; H. marshi
Thorpe, 1920; H. molossus Thorpe, 1920) (known from Barbourofelis Schultz, Schultz, and Martin, 1970 (synonyms:
localities CP68B, CP68C, CPP84A, CP84B); H. occiden- Albanosmilus, in part; Ischyrosmilus, in part; Eusmilus,
talis Leidy, 1869 (= Dinotomus atrox Williston, 1895) (lo- in part)
calities CP84A, CP84B); H. sicarius (=Eusmilus sicar- Type species: Barbourofelis fricki Schultz, Schultz, and
ius Sinclair and Jepsen, 1927); H. dakotensis (= Eusmilus Martin, 1970.
dakotensis Hatcher, 1895); H. mentalis Sinclair, 1921 Type specimen: USNM 7600, skull and right ramus.
(= H. oharrai Jepsen, 1926) (localities CP28A, CP39F, Characteristics: Dirk-toothednimravid with large mandibular
CP83B, CP83C, CP98A, CP98B, CP98C, CP99A, flange, ranging from size of a small mountain lion to that of
NP24C). an African lion. Dental formula 13/3, Cl/l, P2/2, Ml/l; P3
H oplophoneus sp. is also known from localities CP39C, reduced, protocone absent on P4; P4 with a preparastyle;
CP41A. metaconid absent from m 1. Cranium short with the occiput
almost vertical; infraorbital foramen greatly enlarged for
Nanosmilus Martin, 1992 the passage of the deep masseter muscle; postorbital bar
Type species: Nanosmilus kurteni Martin, 1992. present; brain small with large olfactory lobes.
Type specimen: USNM 25502, skull and mandible. Average length of ml: 38.5 mm (B.fricki).
Characteristics: Dirk-toothed nimravid with serrated postca- Included species: B. fricki (known from localities NB27B,
nine teeth and incisors, approximately bobcat-sized. Den- [SPIC], ?CP116D, CP123C); B. whitfordi (=Eusmilus
tal formula 13/3, C 1/1, P3/2, Ml/1; reduced protocone and whitfordi Barbour and Cook, 1915; lschyrosmilus osborni
small parastyle on P4; P3 double rooted with small anterior Merriam) (localities NB7C, NB7D, NB7E, SP2A, [CP78],
 Larry D. Martin 232

CP116A, CP116B); B. morrisi Schultz, Schultz, and Mar-

tin, 1970 (localities GCl lB, NB23C, ?SP2B, ?CP77,
CP90A, CP1l6B, and from the Clarendonian of New Mex-
ico); B. lovei Baskin, 1981 (localities GCl lA, [CC26B],
[SPlB]).
Barbourofelis sp. is also known from locality GCl lB.

INDETERMINATE NIMRAVIDS
Remains assigned to the family Nimravidae have been reported from
the following localities (see also Martin on Felidae, this volume,
Chapter 12, for indeterminate "felids" that might be assignable to
Barbourofelis): GC8A, SB43E, SB44C, ?CP103, NP24D.

BIOLOGY AND EVOLUTIONARY PATTERNS

The nimravids appeared in the late Eocene of North America at the

same time, or a little later, than their first appearance in Eurasia,
indicating an early Holarctic distribution. The early forms showed
some arboreal adaptations and probably occupied the typical "cat
niche" in the spaces around or between trees. The first North Ameri- Figure 12.4. A-D from left to right: Conical-toothed felid, Panthera (A-B,
D. P. lea: C. P. atrox); scimitar-toothed felid, Machairodus coloradensis;
can forms already showed a division into a more cursorially adapted,
dirk-toothed nimravid, Barbourofelis fricki. A. Labial and cross-sectional
scimitar-toothed form, Dinictis, and a shorter-limbed, dirk-toothed views of upper canines. B. Posterior views of mandibles showing the low-
form, Hoplophoneus. Because true cats (felids) are not known from ering of the coronoid process in saber-toothed carnivores. C. Labial views
this time, there were a great variety of size classes of nimravids, of p2-ml showing the progressive overlap of p4 on ml. D. Crown view of
ranging from the size of an ordinary house cat up to about the size of canines and incisors showing their enlargement and rearrangement in dirk-
toothed carnivores. E. Skull and first two cervical vertebrae in Barbourofelis
a jaguar. The largest prey species available (excluding brontotheres
fricki showing the gape and the route of the rectus capitus muscle. F. Cross-
in the late Eocene) would approach the size of a cow; consequently, section through the carnassials of Barbourofelis morrisi showing the thick
larger sized nimravids did not develop at this time. The smallest outer enamel on P4 and inner enamel on ml. The ml rotates toward the
form, Eusmilus, was among the most specialized. The late Eocene- upper carnassial in order to maintain a self-sharpening mechanism. (A-D
Oligocene fauna in North America was dominated by oreodonts and modified from Martin, 1980.)
they, along with small rhinocerotoids, were probably the principal
prey of the nimravids.
The upper canines of saber-toothed carnivores do not really re- toothed or scimitar-toothed carnivores. Scimitar-toothed nimravids
semble sabers. Real sabers have the cutting edge on the outside tended to be longer limbed and more like regular conical-toothed
of the curve, whereas the cutting edge on the canine teeth is on cats in their body proportions. Like modem felids, they must have
the inside. Saber-tooth canines more closely resemble the Arabian stalked their prey until they were close enough to overtake it with
curved daggers called jambiyas. Jambiyas are specialized for throat a short run. Unlike doglike carnivores, cats (and, presumably, other
cutting, which was probably the function of saber teeth (Martin, catlike carnivores) rely on catching prey with a short pursuit that
1989). The development of canines that cut while they bite is the relies on very rapid acceleration. This mode of attack requires some
fundamental saber-tooth adaptation. The cutting was probably done cover for the initial stalk, and cats have to change their fundamen-
when the tooth was being driven into the prey, thereby restricting tal hunting strategy to succeed in very open habitats. The periodic
torsional forces and permitting a longer and more slender canine. At development of treeless environments in the late Cenozoic could
the same time that this presumed ability to bite through large necks have been a major factor in the extinction of catlike carnivores.
was developed, the ability to bite normally was degraded, as was Modem cats kill by biting the neck or throat. Saber teeth probably
the ability to capture small prey. As a result, saber-toothed carni- make the throat bite more effective (Martin, 1980, 1989). Aker-
vores were sensitive to environmental changes that reduced average sten ( 1985) singles out the abdomen as a likely target of attack. This
prey size, and the ecomorphological type has been redeveloped and may have been true for scimitar-toothed carnivores, but it is unlikely
subsequently suffered extinction many times (Martin, 1989). Figure that dirk-toothed forms could even successfully bite the abdomen
12.4 illustrates the differences in the dentitions of conical-, scimitar- of a large animal, let alone make a mortally deep wound (Martin,
and dirk-toothed carnivores. 1980).
Dirk-toothed nimravids were short limbed with badger or bearlike All catlike carnivores probably use some modification of a killing
proportions (see Figure 12.1 ). They must have ambushed rather than bite to the neck (Eisenberg and Leyhausen, 1972) that includes a
pursued their prey and were probably solitary hunters. Their brains learned throat bite. The size of the victim's neck limits the effec-
tended to be smaller than those of either comparably sized conical- tiveness of this mode of attack, and a well-defended prey animal
 Nimravidae 233

w
zw L L.81. u
2.4
() E.BI
45
0 E Let.Hp.
:i
a.
L.Hp.
L.E.Hp.
...
62

7.0

L
E.E.Hp.
L.CI.
...
0.5

E.CI.
11.0
w
zw LL.Ba.
125
() E.L.Ba.
M
0 14.0

:iE E.Ba.

L.Hm.
...
17.S
E.Hm.
18.B
E LL.Ar.
192

E.L.Ar.
23.0

w
zw L.E.Ar.

L
()
'ZT.7
0
CJ E.E.Ar.
:i 29A
0
WL
E 31.8
Or.
33A
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5
NIMRAVINAE BARBOUROFELINAE
L.Un.

w 41.3-

zw E.Un.

g L.Br.
45.9

W M 46.7

M.Br.
49.0
E.Br.
....
L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
........
w
z
c~•
c~
C~1
Tl.8
........
55.6

562
~ L
Tl.6
Tl.4
Tl.3
0 Tl.2
w Tl.1
..J
o( To.3
a. To.2

E ~:1
Pu.2
Pu.1
Pu.o

Bl.=Blancan, Hp.= Hemphilllan, Cl.= Clarendonlan, Ba.= Barstovian, Hm.= Hemlngfordlan, Ar.= Arikareean,
=
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronlan, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerlan, Wa. wasatchian
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan. '

Figure 12.5. Temporal ranges of North American nimravid genera.

with a large neck would involve a high risk factor for a predator. The long but superficial wound that would have been made by saber-
The anatomical targets of the throat bite (carotid artery, jugular vein, toothed carnivores brings the posterior edge of the canine medial
larynx) are relatively superficial but ventrally located. It is very awk- to the ventral comer of the throat, just behind the carotid artery.
ward to bite directly toward the ventral side of.the neck, and most The curvature of the canine would cause it to cut out the comer
killing bites used by modem cats are made to the side of the throat. of the throat, severing the carotid and causing nearly instantaneous
 Larry D. Martin 234

death. An attack on the abdomen would require predatory behaviors The paramastoid processes became enlarged in saber-toothed car-
unknown among modem carnivores. nivores. This occurred at the same time as the insertion for the
The distance that a carnivore can open its jaws determines which digastric was enlarged on the lower jaw. It seems a reasonable spec-
anatomical structures of the prey species are available for attack by ulation that an enlarged digastric would have been necessary to
whatever teeth are used for the killing bite. The front of the mouth pull the lower jaw past the normal gravity-assisted vertical drop.
always has wider gape than the back and can engulf larger struc- Matthew (1910) suggested that the enlarged paramastoids were re-
tures. The depth to which the bite can be effective depends on the lated to their function as insertion points for head depressors, the
length of the canines and would limit the size of prey that could stemomastoid and cleidomastoid. Both functions may have been
easily be attacked. involved in the enlargement of these processes.
It is important to distinguish between the potential rotational arc During the late Eocene through the early Oligocene of North
of the lower jaw and the effective gape that determines the size America, dirk-toothed nimravids ranged in size from that of a house
of objects that can enter the mouth. The effective gape is the arc cat (Nanosmilus kurteni), through that of a mountain lion (Hoplo-
measured tip to tip between the upper and lower canines, and it phoneus primaevus), to that of a jaguar (Hoplophoneus occiden-
is always much less than the rotational arc. In fact, Barbourofelis talis). Eusmilus may have survived into the early Miocene and be-
fricki, with a potential rotational arc of over 110°, had an effective came extinct without much increase in size. Hoplophoneus became
gape of 60°, about the maximum gape found in living large conical- extinct at the end of the early Oligocene. Contemporaneous withHo-
toothed cats (see Figure 12.4E). Maintaining even this amount of plophoneus was a mountain lion-sized scimitar-toothed nimravid,
effective gape must have required heroic measures in dirk-toothed Dinictis, which may have given rise to the jaguar-sized Pogonodon
carnivores (Martin, 1980, 1984). that continued into the latest Oligocene and perhaps into the earli-
Dirk-toothed carnivores must possess a large mouth, and the mas- est Miocene. Nimravus was similar to Dinictis but was somewhat
seteric muscle must be moved posteriorly, both to accommodate the larger, with scimitar teeth and a skeletal structure a little more like
muscle itself and to reduce stretching of the fibers. The coronoid that of a modem felid. This genus also possibly survived into the
process must also be shortened (see Figure 12.4B) to reduce the de- earliest Miocene. Dinaelurus, known only from the early Oligocene,
gree of stretch experienced by the temporalis muscle. Dirk-toothed was the only conical-toothed nimravid and, although Dinaelurus is
carnivores also lower the glenoid process, and the occipital region only known from a skull, its enlarged external and internal nares are
is rotated forward to a nearly 90° angle in the most derived forms. suggestive of a long-limbed cheetahlike pursuit predator.
The latter change is related to an increase in the size of the rectus All members of the Nimravinae had become extinct during the
capitus muscles and an elongation of their areas of insertion on the earliest Miocene at the latest, and the Barbourofelinae lack an-
spine of the axis vertebra (Figure 12.4E). These muscles elevate the tecedents in known North American Oligocene faunas. Thus Bar-
head and thereby increase the distance through which the lower jaw bourofelis must represent a separated later immigration of nim-
can be rotated (depressed) before it meets the chest. In dirk-toothed ravids. Catlike carnivores were absent in North America during most
carnivores with large mandibular flanges, the restriction imposed by of the early Miocene, although other carnivores may have taken on
the flanges contacting the chest when the lower jaw was depressed a more catlike predatory behavior during this time interval (Van
to its maximum limit could have posed a serious problem. In con- Valkenburgh, 1991). True felids first appear as new immigrants in
trast, the dirk-toothed true felid, Smilodon, reduced the mandibular the later Hemingfordian around 17 million years ago. Barbourofelis
flange through time and progressively developed a more inclined did not appear until the Clarendonian (early late Miocene) nearly
occiput. This condition in Smilodon would seem to invalidate part 6 million years later. The earliest Barbourofelinae were Old World
of the argument by Emerson and Radinsky (1980) that seeks to ex- species about the size of a large lynx, but there was a rapid in-
plain the vertical occiputs of saber-toothed carnivores on the basis crease in size in North America up to about that of an African lion
of the mechanical efficiency of the temporalis muscle. (see Figure 12.1). Barbourofelis coexisted with a scimitar-toothed
Feeding must have been a special problem for saber-toothed car- felid, Nimravides, and a variety of conical-toothed felids (Baskin,
nivores because they would not have been able to bite by introducing 1981). The extinction of Barbourofelis in the late Miocene of North
food into the front of the mouth. Specialized incisors (see Figure America marked the end of nimravids worldwide.
12.4D) probably acted to tear the meat away from the bone until
it could be inserted into the side of the mouth to be sliced off by
highly derived camassial teeth. The most extreme example is pro-
REFERENCES
vided by Barbourofelis where the upper carnassial lacks a protocone
and is elongated by means of an enlarged parastyle and a neomor-
Akersten, W. A. (1985). Canine function in Smilodon (Mammalia; Felidae;
phic preparastyle or ectostyle. This enormous tooth occludes against Machairodontinae). Contribution of the Science and History Mu-
both the lower fourth premolar and the lower camassial. These two seum, Los Angeles County, 356, 1-22.
lower teeth contact and overlap each other forming a single func- Baskin, J. A. (1981). Barbourofelis (Nimravidae) and Nimravides (Felidae)
tional shearing surface (see Figure 12.4C). The carnassials were kept with a description of two new species from the late Miocene of
Florida. Journal of Mammalogy, 62, 122-39.
continually sharp by wearing against each other, forming a theoge-
Bryant, H. N. (1991). Phylogenetic relationships and systematics of the
sis (see Figure 12.4F). This sharpening relationship was apparently Nimravidae (Carnivora). Journal of Mammalogy, 72, 56-78.
maintained by the spreading of the lower jaws and the rotation of Cope, E. D. (1878). On some characters of the Miocena fauna of Oregon.
the lowercarnassials (Martin, 1989; Bryant, 1991). Proceedings of the American Philosophical Society, 18, 63-78.
 Nimravidae 235

(1879). On the genera of Felidae and Canidae. Proceedings of Leidy, J. (1854). Remarks on a new species of mammal from Nebraska,
the Academy of Natural Sciences of Philadelphia, 1879, Dinictis felina. Proceedings of the Academy of Natural Sciences
168-94. of Philadelphia, 7, p. 127.
(1880). On the extinct cats of America. The American Naturalist, 14, Martin, L. D. (1980). Functional morphology and the evolution of cats.
833-58. Transactions of the Nebraska Academy of Sciences, 8, 141-54.
(1881). On the Nimravidae and the Canidae of the Miocene Period. (1984). Phyletic trends and evolutionary rates. Carnegie Museum ofNat-
Bulletin of the United States Geological and Geographical Survey ural History Special Publications, 8, 526-38.
of the Territories, 6, 165-81. (1989). Fossil history of the terrestrial Carnivora. In Carnivore Behav-
Eaton, G. I. (1922). John Day Felidae in the Marsh collection. American ior, Ecology and Evolution, ed. J. L. Gittleman, pp. 536-68. Ithaca:
Journal of Science, 5, 425-52. Cornell University Press.
Eisenberg, J., & Leyhausen, P. (1972). The phylogenesis of killing behavior (1992). A new miniature saber-toothed nimravid from the Oligocene of
in mammals. Zeitschrift fur Tierpsychologie, 30, 59-93. Nebraska. Annals Zoologici Fennici, 28, 341-8.
Emerson, S. B., & Radinsky, L.B. (1980). Functional analysis of sabertooth Matthew, W. D. (1910). The phylogeny of the Felidae. Bulletin of the Amer-
cranial morphology. Paleobiology, 6, 295-312. ican Museum of Natural History, 28, 289-316.
Filhol, H. (1873). Machairodus bidentatus. Bulletin de la Societe des Sci- Neff, N. A. (1983). The Basicranial Anatomy of the Nimravidae (Mammalia:
ences Physiques et Naturelles de Toulouse, 205. Carnivora): Character Analyses and Phylogenetic Inferences. Un-
Flynn, J. M., & Galiano, H. (1982). Phylogeny of early Tertiary Carnivora, published Ph.D. Thesis. New York: City University.
with a description of a new species of Protictis from the middle Piveteau, J. (1931). Les chats des Phosphorites du Quercy. Anna/es de
Eocene of Northwestern Wyoming. American Museum Novitates, Paleontologie, 20, 105-63.
2725, 1-64. Schultz, C. B., Schultz, M. R., & Martin, L. D. (1970). A new tribe of saber-
Gervais,P. (1876). Zoologie etpaleontologie genera/es, nouvelles recherches toothed cats (Barbourofelini) from the Pliocene of North America.
sur les animaux vertebres vivants et fossiles, 2e series. Arthus Bulletin of the University of Nebraska State Museum, 9, 1-31.
Bertrand. Scott, W. B., & Jepsen, G. L. (1936). The mammalian fauna of the White
Hough, J. ( 1949). The subspecies of Hoplophoneus: a statistical study. Jour- River Oligocene, Part 1: Insectivora and Carnivora. Transactions of
nal of Paleontology, 23, 536-55. the American Philosophical Society, 28, 1-153.
(1953). Auditory region in North American fossil Felidae: its significance Simpson, G. G. (1941). The species of Hoplophoneus. American Museum
in phylogeny. United States Geological Survey Professional Paper, Novitates, 1123, 1-21.
243-G, 95-1!5. Toohey, L. (1959). The species of Nimravus (Carnivora, Felidae). Bulletin
Hunt, R. M., Jr. (1987). Evolution of the aeluroid Camivora: significance of of the American Museum of Natural History, 118, 71-112.
auditory structure in the nimravid cat Dinictis. American Museum Van Valkenburgh, B. (1985). Locomotor diversity in past and present guilds
Novitates, 2886, 1-74. of large predatory mammals. Paleobiology, 11, 406-28.
Kretzoi, N. (1929). Materialen zur phylogenetischen Klassification der (1991). Iterative evolution ofhypercarnivory in canids (Mammalia: Car-
Aeluroideen. lnternational Congress of Zoology, Session 10 (1927), nivora): evolutionary interactions among sympatric predators. Pale-
Budapest, Part 2, 1293-355. obiology, 17, 340-62.
13 Felidae
LARRY D. MARTIN

INTRODUCTION

Many researchers have regarded Proailurus from the early Miocene

of France as an early representative of the Felidae. It has an elon-
gated cranium and a viverridlike body form. The elongated cranium
seems to be retained in the earliest known North American felid and
undescribed species of Pseudailurus from the late early Miocene
(late Hemingfordian) of Nebraska (Radinsky, 1975). Early felids
rapidly differentiated into small house cat-sized forms, leopard-
sized forms, and saber-toothed (scimitar-toothed) forms the size of
modem lions. During their early radiation they occurred together
with the last surviving nimravid, Barbourofelis. By the end of the
Figure 13.1. Restoration of M achairodus coloradensis.
Miocene the Felidae remained as the only group of catlike carni-
vores, and by the Pleistocene they developed a giant dirk-toothed
saber-toothed ecomorph, Smilodon.
This tendency for two rather distantly related families of car-
nivores, Felidae and Nimravidae, to evolve similar ecomorphs in
parallel is one of the most remarkable stories in evolution. Both fam-
ilies contain carnivores with retractile claws that can be divided into
conical-toothed forms (with "regular" canines) and saber-toothed
forms (although conical canines are the more normal condition in
felids, and there was only one conical-toothed nimravid). The par-
allels among the saber-toothed forms were accomplished in very
similar detail. The canines were similar in their hypertrophy in both
groups and followed the same evolutionary design. This was also Figure 13.2. Skeletal restoration of Machairodus coloradensis.
true of the pattern of premolar and molar reductions as well as the de-
velopment of serration on the teeth. In both families saber-toothed
primitive forms. The carotid artery is reduced. The auditory bulla is
forms can be divided into two different ecomorphological types:
ossified with a bilaminar septum bulla. A cruciate sulcus is present
dirk-toothed (with highly elongated canines), short-legged forms,
on the brain. In later forms the postglenoid foramen is reduced and
and scimitar-toothed (with less elongated canines), longer-legged
the condyloid foramen shares a common opening with the posterior
forms. (See also Martin, this volume, Chapter 12, on Nimravidae.)
lacerate foramen. The alisphenoid bone is lost in later forms.

DEFINING FEATURES OF THE FAMILY FELIDAE DENTAL

Strongly developed "felid grooves" are present on the canines. The
CRANIAL incisors have cingula cusps. The parastyle is always strongly devel-
The skull is short faced and the cranium elongated in the more oped on the upper carnassial, and the lower carnassial has a deep
236
 Felidae 237

notch. Dental formulal3/3, Cl/1, P2-4/2-3, Ml/1. (See Figure 12.4 remaining in the Felinae. In a few cases additional subfamilies have
in Martin, this volume, Chapter 12, on Nimravidae.) been proposed for the cheetah (Acinonyxinae) and the primitive
cats (Pseudailurinae). Within the cladogram shown in Figure 13.3,
a two-part division of the Felidae is possible. One branch, includ-
POST CRANIAL
ingAdelphailurus, Dinofelis, Meganteron, Smilodon, Machairodus,
The skeleton is broadly primitive with few specialized cursorial Homotherium, and Dinobastis, is close to the conventional Mac-
adaptations, and the entepicondylar foramen is retained. More de- hairodontinae (without the inclusion of the nimravids). The other
rived features include digitigrade feet (Smilodon is the exception) branch is the modem Felinae. Acinonyx, Pelis, and Panthera are not
with hooded and retractile claws (except for the cheetah) (see Fig- included on the cladogram because they either did not reach North
ure 13.2). America (Acinonyx) or did not do so until the Pleistocene (Pelis and
Panthera). Neff (1982) suggests that Smilodon might actually be
more closely related to the roaring cats (genus Panthera) than to the
SYSTEMATICS other Machairodontinae. It is clear that saber-toothed morphologies
have evolved several times, and the saber-toothed character must
be used with caution to define groups like the Machairodontinae.
SUPRAFAMILY
Tedford (1978) and Neff (1982) have produced slightly different
See discussion by Martin, this volume, Chapter 12, on Nimravidae. cladograms, but do not provide a list of characters.
Herrington (1985) considered the relationship of the living
Felidae and examined the relationships of some extinct forms in-
INFRAFAMILY
cluding the so-called American "cheetah," Miracinonyx. Contrary to
Pocock ( 1917) provides the basis for modem felid classification. Adams (1979), both Herrington and Van Valkenburgh, Grady, and
No worker has examined all the Miocene-Pliocene forms, and con- Kurten (1990) came to similar conclusions with those expressed
siderable revision may be expected, although there is an extensive in Martin, Gilbert, and Adams (1977), and in Glass and Martin
literature on the various taxa (see, e.g., Kurten, 1952; Ginsburg, (1978), that Miracinonyx represents a parallel of cheetahlike traits
1961; Beaumont, 1964; Schultz and Martin, 1970, 1972; Martin in an animal closely related to the American puma, Puma concolor.
and Schultz, 1975). Werdelin (1985) used ratio diagrams to come to the conclusion that
The traditional division of the Felidae into two subfamilies Pelis rexroadensis and Pelis lacustris may lack close modem rela-
(Felinae and Machairodontinae), with sometimes a third (Pantheri- tives, although many other workers associate them with the genera
nae) is found widely in the literature on cat systematics. Tradition- Lynx and Puma, respectively, as I have done here. There are two
ally, all of the saber-toothed Carnivora (Nimravidae and Felidae) comprehensive reviews of the Felidae, Herrington (1985) and Salles
are lumped int0 the Machairodontinae (often Machairodontidae), ( 1992), and many opposing views still can be found concerning both
the roaring cats placed in the Pantherinae, with everything left over the living and the fossil taxa.

@iJ
Homotherlum/Dinobastis

Pratifelis
Puma
19 Miraclnonyx
Nimravides
18 20
5 16
2
6

Figure 13.3. Proposed felid relationships. (1) "FELINAE" PLUS MACHAIRODONTINAE: Bilaminar septum bulla, cruciate sulcus on the brain. (2)
Elongated cranium. (3) Enlarged lower camassial metaconid. (4) Enlargement of the angular process. (5) Elongation of the limbs, bladelike upper canine
with little reduction of the lower canine. (6) Reduction of the metaconid on the lower camassial. (7) MACHAIRODONTINAE: Upper canines enlarged. (8)
Distinct groove on the interior surface of the upper canine. (9) Lower canine reduced. (10) Upper canine long and bladelike. (I I) Upper canines serrated. (12)
Distal limbs shortened, tail shortened. (13) Upper canines coarsely serrated. (14) Face elongated. (15) Limbs elongated, tail shortened. (16) UNNAMED
TAXON (DERIVED FELINAE): Rounded canines with deep external grooves. (17) Short tail and three lateral grooves on the upper canine. (18) Small
domed head. (19) Small camassials. (20) Enlarged ectotympanic contribution to the auditory bulla, elongated limbs. Tedford (1978) and Neff (1982) show
slightly different relationships but do not list characters. Smilodon is a descendant of Megantereon and Dinobastis a descendant of Homotherium.
 Larry D. Martin 238

INCLUDED NORTH AMERICAN GENERA IN THE Nimravides Kitts, 1958 (synonyms: Machairodus, in part;
FAMILY FELIDAE Pseudailurus, in part)
Type species: N imravides catacops is (Cope, 18 87) (=Macha-
The locality numbers listed for each genus refer to the list of unified irodus).
localities in Appendix I. The acronyms for museum collections are Type specimen: AMNH 8318, mandibular symphysis.
listed in Appendix III. Characteristics: Scimitar-toothed cats ranging in size from
The locality numbers may be listed in a couple of alternative ways. that of a jaguar to that of a mountain lion. Dental formula
Parentheses around the locality (e.g., [CPlOl]) mean the taxon in 13/3, Cl/I, P2/2, Ml/l. Upper canines serrated posteriorly
question at that locality is cited as an "aff." or "cf." the taxon in and weakly or unserrated anteriorly; lower canines less
question. Parentheses are usually used for individual species, thus reduced than in other saber-toothed carnivores; dependent
implying the genus is firmly known from the locality, but the actual flange on the mandible; ml with strong talonid. Limbs
species identification may be questionable. Question marks in front elongated.
of the locality (e.g., ?CPl 01) mean the tax on is questionably known Average length of ml: 25.4 mm (N. catacopsis).
from that locality, thus implying some doubt that the taxon is actually Included species: N. catacopsis (=Machairodus catacopsis
present at that locality, either at the genus or the species level. Cope, 1887, Pseudailurus thinobates Macdonald, 1942;
N. nebraskensis Beaumont, 1990) (known from locali-
ties GCl lB, CC25B, CC26B, ?CC32B, [SB33II], SB60,
FELINAE
CP116D, CPl23C, CP127): N. galiani Baskin, 1981 (lo-
Pseudailurus Gervais, 1848-1852 (synonyms: Metailurus, cality CP116D); N. pediomus Macdonald, 1948 (locality
in part; Felis, in part)
CPl 16B).
Type species: P seudailurus quadridentatus (Blainville, 1841)
Nimravides sp. (including unsynonymized species N.
(=Felis). lovei) is also known from localities GCl lA, SPlC, SP2A,
Type specimen: Partial ramus in the Museum d'Histoire Na-
SP3A, SP3B, CP78, CP90A, CP114B, CP114D, CP116C.
turelle, Paris, No. 601.
Characteristics: Small, primitive conical-toothed cats with Lynx Kerr, 1792 (synonym: Felis, in part)
elongated cranium. Dental formula 13/3, C l/l, P4/3, Ml/1. Type species: Lynx lynx (Linnaeus, 1758) (=Felis), (the ex-
Auditory bulla small; mastoid process not connected to tant lynx or bobcat).
the paraoccipital process by a crest; posterior lacerate and Characteristics: Small, conical-toothed cats with a short face.
condylar foramina separate. Dental formula 13/3, Cl/l, P2/2, Ml/1. Three distinct
Average length of ml: 16.5 mm (P. intrepidus). grooves on upper canines.
Included North American species: P. intrepidus Leidy, 1858 Average length of ml: 15.2 mm (L. rexroadensis).
(known from localities NBS, NB23B, CP56, CP75B, CP76, Included species: L. rexroadensis Stephens, 1959 (known
[CP90A], CPl 10, CP114B, CP115B); P. marshi Thorpe, from localities GC13B, SB50, SP4B, CP116F, CP128C);
1922 (localities [SB32A], CPl 10, CP114B); P. aeluroides L. kurteni Schultz and Martin, 1972 (locality CPl20); L.
Macdonald, 1954 (localities CPllO, [NPll]); P. hibbardi proterolyncis Savage, 1941 (localities SB9, SPlD, SP3B).
Dalquest, 1968 (localities SB9, SP3B); P. sinclairi Lynx sp. is also known from localities ?SBlO, SB16,
Matthew, 1918 (locality CPllO); "P." stouti Schultz and SB37, SPlF, SPlH, PN4.
Martin, 1972 (locality CP76).
Pseudailurus sp. is also known from localities GC4C, Comments: Werdelin (1985) accepted the small Miocene
GCllB, CC17E, CC17G, CC17H, CC171, ?CC26B, felid L. proterolyncis as a member of Lynx but did not
CC32B, CC37, NB6C, NB6E, NB7C, NB7D, NB20, necessarily accept L. rexroadensis, which he considered
?NB23A, NB23C, NB3 l, NB33, SB30A, SB32D, SB34C, a synonym of Puma lacustris. The presence of three felid
?SB60, SPlB, SPlD, SP2A, ?SP2B, CP78, CP108A, grooves on the upper canine of L. rexroadensis confirms its
CP108B, CP114A, CP123C, PN2A, PN7, PNSB, PNlO, inclusion within the genus Lynx (Martin, in preparation).
PNll, PN13.
Puma Jardine, 1834 (synonym: Felis, in part)
Pratifelis Hibbard, 1934 Type species: Puma concolor (Linnaeus, 1771) (=Felis), the
Type species: Pratifelis martini Hibbard, 1934. extant cougar or mountain lion.
Type specimen: KUVP 3156, partial ramus. Characteristics: Conical-toothed cats with relatively small
Characteristics: Conical-toothed cat the size of a leopard. broad skulls, heavily built lower jaws with small
Dental formula l?/3, C?/l, P?/2, M?/l; distinct paraconid camassials, and relatively short limbs. Dental formulal3/3,
and metaconid onp3; large ml with enlarged talonid; mas- Cl/l, P3/2, Ml/l.
seteric fossa extending under m 1. Average length of ml: 15.7 mm (P. lacustris).
Average length of m 1: 29 .5 mm. Included species: P. lacustris Gazin, 1933 (=P. idahoensis)
Included species: P. martini only, known from locality only (known from localities CC53, SB14F, SB15, SB31C,
CP123B only. SPlF, SPlH, SP5, CP128C, PN4, PN23A, PN23C).
 Felidae 239

Comments: When Miracinonyx studeri was considered to both anteriorly and posteriorly. Dental formula 13/3, Cl/l,
be a member of the genus Puma, it was more difficult to P2/2, Ml/l. Lower canines more reduced than in Nim-
regard the contemporary "Felis" lacustris as a mountain ravides: usually some slight development of mandibular
lion. I now think this is the case, and that we had late Pleis- flange; ml with small talonid.
tocene representation of all three of the large nonpantherine Average length of ml: 31.9 mm (M. coloradensis).
conical-toothed cats in North America. Included North American species: M. coloradensis Cook,
1922 (known from localities [NB 10], SB9, ?SB60, SP3B,
Miracinonyx Adams, 1979 (syqonyms: Puma, in part; CP78, CP115C, CP123D); M. tanneri (=M. coloradensis
Felis, in part) tanneri) Martin and Schultz, 1975 (locality CP116E).
Type species: Miracinonyx trumani (Orr, 1969) (=Felis). Machairodus sp. is also known from localities: GC12II,
Type specimen: Western Speleological Institute no. p3a/450, CC25B, CC37, NB27B, SB58, SP3A, SP4B, CP116F,
skull and mandible. CP116E, PN14, PN15.
Characteristics: Long-legged, conical-toothed cat with small
canines. Dental formula 13/3, Cl/l, P3/2, Ml/l. P4 with Comments: There has been some confusion between Macha-
reduced protocone. Enlarged internal and external nares; irodus and Nimravides (Martin and Schultz, 1975). Nim-
domed skull; enlarged ectotympanic contribution to audi- ravides evolved in North America, whereas Machairodus
tory bullae. is a later immigrant.
Average length of ml: unavailable. Average length of P4:
23.8 mm. Meganteron Croizet and Jobert, 1928 (synonyms:
Included species: M. trumani Savage, 1960 only, known from Machairodus, in part; lschyrosmilus, in part)
locality SPlH only. Type species: Meganteron cultridens (Cuvier, 1824)
(=Machairodus).
MACHAIRODONTINAE Type specimen: Lecotype specimen IGF 816, right upper ca-
nine.
Adelphailurus Hibbard, 1934 Characteristics: Puma- to leopard-sized dirk-toothed cat, with
Type species: Adelphailurus kansensis Hibbard, 1934. unserrated or weakly serrated upper canines. Dental for-
Type specimen: KUVP 3462, maxillary. mula 13/3, C 1/1, P2/2, M 1/1. Ectoparasty le absent or poorly
Characteristics: Mountain lion- to lion-sized cat, with upper developed on P4; slight dependent flange on mandibular
canines enlarged and flattened internally, and with a dis- ramus; prominent postorbital process; postglenoid process
tinct anterior-internal groove. Dental formula 13/3, Cl/l, enlarged.
P3/2,,Ml/l. Metacone blade of P4 elongated; strong pos- Average length of ml: 19.8 mm (M. hesperus).
terior cingula on premolars. Postcranial skeleton similar to Included species: M. hesperus (=Machairodus? hesperus
Puma. Gazin, 1933) only (in North America) (known from local-
Average length of ml: 16.6 mm. ities GC13B, GC14, GC15, GC18, NB13C, SB9, SBlO,
Included species: A. kansensis only (known from localities SBll, SB60, CP117B, ?CP128C, CP131, PN23A).
SB9, SBlO, SB60, CP123D).

Dinofelis Zdansky, 1924 (synonym: Panthera, in part) Homotherium Fabrini, 1890 (synonyms: Ischyrosmilus;
Type species: Dinofelis abeil Zdansky, 1924. Machairodus, in part)
Type specimen: Partial cranium in the University of Uppsala Type species: Homotherium crenatidens Fabrini, 1890.
Paleontological Institute. Type specimen: Upper canine, from the Villafranchian of
Characteristics: Jaguar-sized scimitar-toothed cats. Dental Italy.
formula 13/3, Cl/l, P2/2, Ml/l. Moderately enlarged and Characteristics: Jaguar-sized scimitar-toothed cats with up-
flattened, but unserrated, upper canines. P2 absent, P3 re- per canine coarsely serrated anteriorly and posteriorly.
duced, and P4 with large accessory cusps. Dental formula 13/3, Cl/1, P2/2, MO/l. Protocone highly
Average length of ml: 23.7 mm (D. palaeoonca). reduced on P4; distinct dependent flange on mandible.
Included species: D. palaeoonca (=Panthera palaeoonca Shorter face than Machairodus; highly elongated limbs;
Mead, 1945) only (in North America), known from lo- short tail.
cality SP5 only. Average length of ml: 25.8 mm (H. crusafonti).
Included North American species: H. ischyrus (=Machairo-
Machairodus Kaup, 1833 dus? ischyrus Merriam, 1905) (known from localities
Type species: Machairodus aphanistus Kaup, 1833. CC45, CP118); H. idahoensis (=lschyrosmilus idahoen-
Type specimen: Mandibular ramus jn the Hesses Museum, sis Merriam, 1918) (locality PN23C); H. crusifonti (=!.
Darmstadt. crusifonti Schultz and Martin, 1970) (localities CPll 7 A,
Characteristics: Lion-sized (at least North American species) CP117B ); H. johnstoni (=I. johnstoni Maw by, 1965) (lo-
scimitar-toothed cats with upper canihes coarsely serrated cality SPlH).
 Larry D. Martin 240

w
z L L.BI.
1.8 II) II)
w
0
0 E
E.BI

Lat.Hp.
2.4

-
-~
: .;.
el1I __ o ___
-c:--
::J -
s::::~o....
l!
- -"g~ .I!!
c:--1;.
(11 ...
-II)-
::i
-
a
::i
D..
L.Hp.
L.E.Hp.
-ir -·~- - :§. - i5 - -'iii -~
7.0 -II) - - -111- - -'fi
==~-=
E.E.Hp. QI - - - -l::i)-
L L.CI.
a.a
8.5
-~
-·s -~.: ~ -~
E.CI.
11.0
e
-e
w
zw -~
s....~
LL.Bo.
12.6
0
0
:5
M
t:.L.Ba.

E.Ba.
14.0 -
~
::J
5.8 QI
L.Hm.
17.5
If.
E.Hm.
18.8
E L.L.Ar.
19.2 FELINAE MACHAIRODONTINAE
E.L.Ar.
23.0

w
z
w
L.E.Ar.

L
0 27.7
0
CJ E.E.Ar.
::i 29.4
0
Wt.
E 31.9
Or.
33A
LCh.
34.5
L M.Ch.
35.5
E.Ch.
37.1
Du

39.5

L.Un.

w 41.3
zw E.Un.

0
0
....
w M
L.Br,
48.7

M.Br.
49.0
E.Br.
BOA

L.WL
53.5
E M.WL
54.2

w
z
w
0
L -
0
w
..J
To.3
ct To.2
D..
To,1
E Pu.3
Pu.2
Pu.1
Pu.O

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonlan, Ba.= Barstovian, Hm.= Hemlngfordlan, Ar.= Arlkareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un. = Ulntan, Br.= Brldgerlan, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti.= Tiffanian, To.= Torrejonlan, Pu.= Puercan.
Figure 13.4. Temporal ranges of North American felid genera.
 Felidae 241

Homotherium sp. is also known from localities GC14, (Meganteron and the Pleistocene Smilodon). Conical-toothed cats
SB50, SPIH, CP128C, PN3, ?PN23A. and especially the scimitar-toothed saber-toothed cats evolved large,
probably short-pursuit forms (Panthera, Machairodus, Dinofelis,
Homotherium). The abilities of catlike camivorans to use their fore-
INDETERMINATE FELIDS
limbs to grapple with prey may have given them some evolutionary
Remains assigned to the family Felidae have been reported from advantages with large prey over doglike carnivorans. This would be
the following localities (those tagged * may be of the nimravid especially true for solitary hunting, where a single individual must
Barbourofelis, as they are found within the appropriate time range): immobilize the prey and deliver a killing bite.
GC28, CC21A, CC21B, CC38*, NB6E, NBll*, NB21, SB12, There were two major radiations of felids in North America. A late
SB13, SB14A, SB18B, PNll *, PN12*, PN13*, NC2. Miocene diversification originated with Pseudailurus, giving rise to
lion-sized scimitar-toothed forms (Nimravides and Machairodus),
on the one hand, and leopard- to house cat-sized conical-toothed
BIOLOGY AND EVOLUTIONARY PATTERNS forms (Pseudailurus, "P." stouti, and Adelphailurus), on the other.
"Pseudailurus" stouti is a small cat at the Pseudailurus evolutionary
With the extinction of the nimravine nimravids by the earliest level with possible affinities to some of the small South American
Miocene, there was a long period in North America between the felids (Glass and Martin, 1978). There were no dirk-toothed felids
early late Arikareean and late Hemingfordian when catlike carni- for most of the Miocene: Apparently that role was taken by the dirk-
vores were absent. The early Miocene was also a period of great toothed nimravid, Barbourofelis. However, the dirk-toothed felid
diversity in arctoid (doglike) carnivores and was the time of the first Meganteron first appeared in the latest Miocene (late Hemphillian)
major radiation of fossorial mammals into early grasslands. Catlike shortly following the extinction of Barbourofelis in the late early
carnivores rely on vegetational cover to a large extent to assist in prey Hemphillian.
capture, with an ambush mode of predation where it is important to It seems fairly clear that Meganteron gave rise to Smilodon (the
avoid early detection by the prey. The early Miocene reduction of Pleistocene dirk-toothed felid), and Homotherium toDinobastis (the
vegetational cover may have favored musteloid and arctoid carni- Pleistocene scimitar-toothed felid). M eganteron and Smilodon had
vores better suited to digging out or to pursuing prey, although some short legs and probably relied on ambush to capture their prey. Ho-
of these carnivores may have adopted more catlike modes of pre- motherium and Dinobastis were both long-legged predators prob-
dation in the absence of true catlike carnivores (Van Valkenburgh, ably capable of pursuit. According to Radinsky (1975), they had
1991). enlarged optical centers in the brain, similar to those found in mod-
The earliest true cats had relatively short legs and long bodies and em cheetahs. The Irvingtonian (early Pleistocene) saw the arrival
were about the size of a large lynx or a small mountain lion. This of Panthera to North America. According to Herrington (1985),
morphology iS' suggestive of tree-climbing capacities, a behavior the cave lion (P. spelaeus), the tiger (P. tigris), the American lion
that has been retained in present-day small cats, many of which are (P. atrox), and the giant jaguar (P. onca augusta) all occurred in
specialized predators on birds. Later, larger cats probably hunted on North America during the late Pleistocene. There was also a small
the ground, with a predation mode of a highly developed stalking Pleistocene cat, Felis ammicola, in the southeast United States .
behavior using vegetational cover to get close to their prey. Prey
capture probably involved a single bound or a short pursuit, as in
present-day ambush predation cats. All cats have the capacity for REFERENCES
rapid acceleration for short distances, so it is not surprising that
sprinters like the present-day cheetah have been evolved on several Adams, D. B. (1979). The cheetah: Native American. Science, 205, 1155-8.
occasions (this may have been the hunting mode of the cheetahlike Beaumont, de, G. (1964). Remarques sur la classification des Felidae. Ecolo-
Miracinonyx). However, most cats are not capable of distance pur- gae Geologicae Helvetiae, 57, 837-45.
Blainville, de, H. M. (1841). Osteographie et description iconographique
suit and must get close to their prey before the final dash (the cheetah
des Mammiferes recents et fossil es (Carnivores), 2 vols. Paris.
has more stamina than other present-day cats but does not have the Cope, E. D. (1887). A saber-toothed tiger from the Loup Fork beds. American
endurance capacity of pursuit-hunting canids). Cats therefore re- Naturalist, 21, 1019-20.
quire brush or trees to conceal their approach and do not do well Croizet, J.B., & Jobert, A. C. G. (1828). Recherches sur /es ossements fossil es
in open habitats where long-distance pursuit by dogs is a more ef- du departement du Puy-de-Dome, 1, Delahayes, Paris, 224 pp.
Cuvier, G. (1824). Recherches sur /es ossementsfossiles, oii /'on retablit !es
fective hunting strategy. Dogs also display social hunting behaviors
caracteres de plusieurs animau.x dont !es revolutions du globe ant
that are found only in the lion among modem cats. In North America detruit les especes (2nd ed.). Paris, 5, 547 pp.
the overall Cenozoic trend toward a reduction of forests has tended Fabrini, E. (1890). Machairodus (Meganthereon) de Valdamo superiore.
to be detrimental to catlike camivorans and to favor doglike forms Comitzto Geologico; Bollettino, XXI, 121-44, 161-77.
(see Martin, 1989). Gervais, P. (1848-1852). Zoologie et Paleontologie Fran9aise (10 ed.).
The large cats can be divided into ambush predators and short- Paris.
Ginsburg, L. (1961). La Faune des carnivores Miocenes de Sansan (Gers).
pursuit forms. Dirk-toothed saber-toothed cats, with their short legs Memoires du Museum d'Histoire Naturelle, 9, 1-190.
and semidigitigrade feet, must have been the ultimate ambush preda- Glass, G. E., & Martin, L. D. (1978). A multivariate comparison of some
tors. They represent a single radiation among North American felids extant and fossil Felidae, Camivora. Carnivore, 1, 80--8.
 Larry D. Martin 242

Herrington, S. J. (1985). Phylogenetic Relationships of the Wild Cats of Martin, L. D., & Schultz, C. B. (1975). Scimitar-toothed cats, Machairodus
the World. Unpublished Ph.D. Thesis. Lawrence: University of andNimravides, from the Pliocene of Kansas and Nebraska. Bulletin
Kansas. of the University of Nebraska State Museum, 10, 55-63.
Hibbard, C. W. ( 1934). Two new genera of Felidae from the middle Pliocene Neff, N. A. (1982). The Big Cats. New York: Abrams.
of Kansas. Transactions of the Kansas Academy of Science, 37, Orr, P. C. (1969). Fe/is trumani, a new radiocarbon dated cat skull from Crypt
239-55. Cave, Nevada. Bulletin of the Santa Barbara Museum of Natural
Jardine, W. (1834). The Naturalist's Library: Mammalia, Vol.II: The Natural History, 2, 1-8.
History of the Felinae. Edinburgh: W. H. Lizars. Pocock, R. I. (1917). The classification of existing Felidae. Annals of the
Kaup, J. J. (1833). Description d'ossements fossiles de mammireres in- Magazine of Natural History, 20, 329-50.
connus jusqu 'a present qui se trouvant du Museum grand-ducal de Radinsky, L. (1975). Evolution of the felid brain. Brain, Behavior, Evolution,
Darmstadt. Second Cahier 40, Darmstadt, 31 pp. 11, 214-54.
Kerr, R. (1792). The Animal Kingdom; or Zoological Systems of the Cele- Salles, L. 0. (1992). Felid phylogenetics: extant taxa and skull morphology
brated Sir Charles Linnaeus. Class I. Mammalia and Class II. Birds. (Felidae, Aeluroidea). American Museum Novitates, 3047, 1-67.
London: J. Murray. Schultz, C. B., & Martin, L. D. (1970). Machairodont cats from the Early
Kitts, D. B. (1958). Nimravides, a new genus of Felidae from the Pliocene Pleistocene Broadwater and Lisco local faunas. Bulletin of the Uni-
of California, Texas, and Oklahoma. Journal of Mammalogy, 39, versity of Nebraska State Museum, 9, 1-36.
368-75. (1972). Two lynxlike cats from the Pliocene and Pleistocene. Bulletin of
Kurten, B. (1952). The Chinese Hipparion fauna. Societas Scientiarum the University of Nebraska State Museum, 9, 197-203.
Fennica, Commentationes Biologicae, 13, 1-81. Tedford, R.H. (1978). History of dog and cats: a view from the fossil record.
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundum In Nutrition and Management ofDogs and Cats, pp. 1-10. St. Louis:
classes, ordines, genera, species cum characteribus, differentis, Ralston Purina.
synonymis, locis. Editio Decima, Reformata. Stockholm: Laurentii Van Valkenburgh, V. ( 1991 ). Iterative evolution of hypercarnivory in canids
Salvi!. (Mammalia: Carnivora): evolutionary interactions among sympatric
(1771). Car. 2. Linne mantissa plantarum altera generum editionis VI. predators. Paleobiology, 17, 340-62.
2nd. specierum editionis II (mantissa prioris additamenta). Regni Van Valkenburgh, B., Grady, F., & Kurten, B. (1990). The Plio-Pleistocene
Animalis Appendix. Vol. 2. cheetahlike cat Miracinonyx inexpectatus of North America. Journal
Martin, L. D. (1989). Fossil history of the terrestrial Carnivora. In Carnivore of Vertebrate Paleontology, 10, 434-54.
Behavior, Ecology, and Evolution, ed. J. L. Gittleman, pp. 536--68. Werdelin, L. (1985). Small Pleistocene felines of North America. Journal of
Ithaca: Cornell University Press. Vertebrate Paleontology, 5, 194-210.
Martin, L. D., Gilbert, B. M., & Adams, D. B. (1977). A cheetahlike cat in Zdansky, 0. (1924). Jungtertiare Carnivoran Chinas. Paleontologica Sinica,
the North American Pleistocene. Science, 165, 981-2. (C) 11, 1-55.
14 Hyaenidae

ANNALISA BERTA

INTRODUCTION

Only one member of the family Hyaenidae, the genus Chas-

maporthetes, is known to have entered North America in the early
Pliocene across the Bering Land Bridge (Figure 14.1), and is known
from the southern United States and Mexico. The earliest diverging
species of the genus Chasmaporthetes exilus recorded from the Tur-
olian (late Miocene) of China and the greater diversity of hyaenids
in the Old World suggests that Chasmaporthetes originated in Eura-
sia (Kurten and Werdelin, 1988). Chasmaporthetes disappeared in
North America during the early Pleistocene shortly after its arrival.
Its ecologic replacement in North America may have been the chee-
tah and/or the dire wolf.
The dentition of Chasmaporthetes, including sharp premolars,
large camassials, and a deep mandible (Figure 14.2), resembles that
of the modem spotted hyaena, Crocuta. The dentition suggests that
this fossil hyaenid apparently filled a similar ecologic role as an
active predator rather than as a strict bone-crushing scavenger more
Figure 14.1. Restoration of the North American hyaenid, Chasmaporthetes
typical of the genus Hyaena. This suggestion is supported by the
?ossifragus from Florida, by Henry Galiano (from Berta, 1981).
elongate, slender limbs of Chasmaporthetes that indicate cursorial
capabilities. DENTAL
The early members of the family are characterized by elongate,
DEFINING FEATURES OF THE slender, trenchant premolars (Figure 14.2). Later hyaenids (except
FAMILY HYAENIDAE the aardwolf, Proteles) tend to modify the premolars into heavy,
conical, bone-crushing structures. Extant hyaenids possess Ml/1
with a loss of postcamassial crushing function and great reduction
CRANIAL of and loss of the Ml metaconid (Flynn, Neff, and Tedford, 1988).
There are no synapomorphies that diagnose all fossil and extant
members of this family (Werdelin and Solounias, 1991). Defining
POSTCRANIAL
features shared among extant hyaenids include an auditory bulla
with a greatly enlarged entotympanic chamber and a septum bullae Early hyaenids have elongate, slender limbs (Figure 14.2C). Later
either fusing the two layers very early or losing the entotympanic hyaenids acquire robust limbs and have shorter hindlimbs than fore-
contribution entirely (character state uncertain for all hyaenids) limbs. The claws, unlike those of other aeluroids (except Herpesti-
(Flynn, Neff, and Tedford, 1988). dae), are blunt and nonretractile.

243
 Annalisa Berta 244

0 2 3 4 5 cm

c ~
,,"'\1 l 1''~
-rl

~
u
'
i \1\
\\I

II I
!

~!'1 \
I ii
I,::
Im-~

0 2 3 A 5cm 0 l 2 3 A 5 cm

Figure 14.2. Osteology of Chasmaporthetes ? ossifragus. A. Upper dentition. B. Lower dentition. C. Postcrania: left metatarsal II, right metatarsal IV, right
tibia. Drawn by Wendy Zomlefer (from Berta, 1981). Scale= 5 cm.

SYSTEMATICS opinion was followed by many workers including Galiano and

Frailey (1977), Hendey (1978), Chen and Schmidt-Kittler (1983),
and Werdelin and Solounias (1991).
SUPERFAMILY
Although recognizing the possibility that hyaenids may be poly-
Classic anatomical studies of the nineteenth century recognized the phyletic, Hunt (1987) proposed within a cladistic framework a mono-
Hyaenidae as a family of aeluroid carnivores (Flower, 1869; Mivart, phyletic herpestid-hyaenid clade. Hunt in this same paper further
1882). Hyaenids were considered by most workers to be derived distinguished hyaenids from all other aeluroids on the basis of a
from viverrids (civets, etc.) (Savage, 1978). Mivart (1882) believed uniquely derived bullar pattern (see previous section). In Hunt's
they were most similar to herpestids (mongooses), a proposal later scheme, the Felidae and Viverridae form a monophyletic clade that is
formalized by Hunt (1987). allied with the Hyaenidae-Herpestidae clade. A different hypothesis
Thenius (1965) was the first to question the monophyly of hyae- of relationships for hyaenids was proposed by Wozencraft (1989),
nids in his separation of percrocutids from all other hyaenids. This who suggested a sister-group relationship between hyaenids and
 Hyaenidae 245

Figure 14.3. Interrelationships of the Hyaenidae. Key to characters at nodes (modified from Werdelin and Solounias, 1991 ): (1) ml with reduced protoconid;
ml paraconid and protoconid equal/subequal in height; reduction in length of ml talonid and posterior molars. (2) Narrower premolars. (3) M2/2 absent. (4)
Anterior face of p3-4 concave. (5) Elongated P4 metastyle. (6) Short P4 metastyle; reduced anterior accessory cusps on P4; large contribution of maxillary
to anterointemal end of the zygomatic arch.

felids. Most recently, Flynn, Neff, and Tedford (1988) concluded Frailey (1977) support Lycyaena as the sister taxon of Chasmapor-
that the conflicting distribution of derived characters among aeluroids thetes. Qiu (1987) holds a similar view but considers Lycyaena a
(Herpestidae, Hyaenidae, Felidae, and Viverridae) led to presenta- member of the Chasmaporthetes stem group. Kurten and Werdelin
tion of their relationships as unresolved. (1988) offer Thalassictis (synonym Hyaenictitherium, Lycyaena)
as the closest relative. The latter proposal is contrary to Galiano
and Frailey (1977) because it includes Hyaenictitherium, a taxon
INFRAFAMILY considered by these authors to be linked to more derived hyaenids
Despite the poor preservation of the type edentulous jaw, Chasma- (Hyaena and Crocuta). In the most comprehensive cladistic anal-
porthetes was recognized by Hay (1921) as a hyaenid. Stirton and ysis of hyaenids yet undertaken, Werdelin and Solounias (1991)
Christian (1940) described a well-preserved ramus with a complete (see Figure 14.3) ally Chasmaporthetes with Hyaenictis as sister
dentition as the type of Ailuraena johnstoni, which they recognized taxa on the basis of their similarly derived premolar morphology
as a hyaenid with catlike features. The following year they syn- in which the anterior face of p3-4 is convex. Chasmaporthetes is
onymized Ailuraena with Chasmaporthetes (Stirton and Christian, distinguished from Hyaenictis by its elongate P4 metastyle.
1941). Although Simpson (1945) placed Chasmaporthetes in the Fe-
lidae, incertae sedis, all later workers have followed Hay's classifi-
cation of the genus in the Hyaenidae. INCLUDED NORTH AMERICAN GENERA IN THE
There is disagreement regarding the relationship of Chasma- FAMILY HYAENIDAE
porthetes to other hyaenids. The European hyaenid Euryboas was
recognized as a close relative of Chasmaporthetes by numerous The locality numbers listed for each genus refer to the list of unified
workers including Galiano and Frailey (1977) and Hendey (1975, localities in Appendix I. The acronyms for museum collections are
1978). Most recently, Qiu (1987), Kurten and Werdelin (1988), and listed in Appendix III.
Werdelin and Solounias (1991) have suggested these two genera are
congeneric, with Chasmaporthetes as the senior synonym. Chasmaporthetes Hay, 1921 (synonym: Ailuraena Stirton
Aside from the problem of the status of Eury boas, there is a differ- and Christian, 1940)
ence of opinion regarding hyaenid interrelationships. Galiano and Type species: Chasmaporthetes ossifragus Hay, 1921.
 Annalisa Berta 246

Type specimen: USNM 10023, edentulous left jaw from Cal REFERENCES
Verde Mine, near Anita, Coconino Country, Arizona.
Characteristics: Hyaenid of moderate size; tendency for broad Berta, A. (1981). The Plio-Pleistocene hyaena Chasmaporthetes ossifragus
rostrum and stepped forehead; cheek teeth elongate, slen- from Florida. Journal of Vertebrate Paleontology, 1, 341-56.
der, and trenchant; Pl large; P2 and 3 with large anterior Chen, G., & Schmidt-Kittler, N. (1983). The deciduous dentition of Percro-
accessory cusps; long, deep mandible; lower tooth row cuta Kretzoi and the diphyletic origin of the hyaenas (Carnivora,
Mammalia). Palaontologische Zeitschrift, 57, 159---09.
straight or slightly curved with imbricate premolars; p4
Ewer, R. F. (1977). Some adaptive features in the dentition of hyaenas. Annals
elongate with large accessory cusps; ml metaconid absent, and Magazine of Natural Hist01y, series 123, 7, 188-94.
talonid unicuspid and trenchant; elongate, slender limbs Flower, W. H. (1869). On the value of the characters of the base of the cra-
(modified from Galiano and Frailey, 1977; Berta, 1981; and nium in the classification of the Order Carnivora and the systematic
Kurten and Werdelin, 1988). position of Bassaris and other disputed forms. Proceedings of the
Zoological Society of London, (1869), 4-37.
Average length of ml: 24.0-29.0 mm.
Flynn, J. J., Neff, N. A., & Tedford, R. H. (1988). Phylogeny of the Car-
Included species: C. ossifragus Hay, 1921 (=Ailuraena john- nivora. In The Phylogeny and Classification of the Tetrapods, Vol.
stoni Stirton and Christian, 1940, C.johnstoni Stirton and 2: Mammals, ed. M. J. Benton, pp. 73-116. Systematics Association
Christian, 1941; see Kurten and Werdelin, 1988, and Special Volume no. 35B. Oxford: Clarendon Press.
Werdelin and Solounias, 1991) (known from localities Galiano, H., & Frailey, D. (1977). Chasmaporthetes kani, new species from
China, with remarks on phylogenetic relationships of genera within
SB15,SB16,SB17,SB6 3,SB64,SB1H,SB14A ,SP1H,
the Hyaenidae (Mammalia: Carnivora). American Museum Novi-
?CP128A, CP 128C); Chasmaporthetes ?ossifragus (=Cha- tates, 2632, 1-16.
smaporthetes undescribed new sp. Werdelin and Solou- Hay, 0. P. (1921). Descriptions of species of Pleistocene Vertebrata, types
nias, 1991) (locality GC14). or specimens most of which are preserved in the United States Na-
tional Museum. United States National Museum Proceedings, 59,
599-642.
BIOLOGY AND EVOLUTIONARY PATTERNS Hendey, Q. B. (1975). Relationships of North American hyaenas. South
African Journal of Science, 73, 187.
(1978). Late Tertiary Hyaenidae from Langebaanweg, South Africa, and
The narrow, sharp teeth and elongate, slender limbs of Chasmapor-
their relevance to the phylogeny of the family. South African Museum
thetes suggest that it filled a different ecologic role than the bone- Annals, 76, 265-97.
crushing, scavenging habitus more typical of the striped and brown Hunt, R. M., Jr. (1987). Evolution of the Aeluroid Carnivora: significance of
hyaenas, Hyaena. Chasmaporthetes more nearly resembles Crocu- auditory structure in the Nimravid cat Dinictis. American Museum
ta, which has elongate camassials more highly modified as shearing Novitates, 2886, 1-74.
Kurten, B., & Anderson, E. (1980). Pleistocene Mammals ofNorth America.
blades in accordance with its more predaceous habit (Ewer, 1977).
New York: Columbia University Press.
Galiano and Frailey (1977) suggested that Chasmaporthetes and Kurten, B., & Werdelin, L. (1988). A review of the genus Chasmaporthetes
the extinct borophagine dog, Aelurodon taxoides, may have occu- Hay, 1921 (Carnivora, Hyaenidae). Journal of Vertebrate Paleontol-
pied similar ecologic roles, a role now filled by wolves and the ogy, 8, 46---06.
hunting dog, Lycaon. Morphologic features shared among these two Mivart, St. George. (1882). On the classification and distribution of the
Aeluroides. Proceedings of the Zoological Society, London, (1882),
groups include a long face, large Pl, trenchant premolars, and a deep
340--404.
horizontal ramus. A predator-prey relationship between Chasma- Qiu, Z. (1987). Die Hyaeniden aus dem Ruscinium und Villafranchium Chi-
porthetes and the giant ground squirrel, Paenemarmota, has been nas. Munchner Geowissenschaftliche Abhandlungen, Reihe A, Ge-
suggested by a number of workers (see Kurten and Anderson, 1980, ologie und Palaontologie, 9, 1-110.
and references cited therein) based on their co-occurrence at several Savage, R. J. G. (1978). Carnivora. In Evolution of African Mammals, eds.
V. J. Maglio & H.B. S. Cooke, pp. 249---07. Cambridge: Harvard
localities.
University Press.
The fossil record suggests immigration of Chasmaporthetes from Simpson, G. G. (1945). The principles of classification and a classification
Eurasia to North America across the Bering Land Bridge prior to of mammals. Bulletin of the American Museum of Natural History,
3.5 Ma (Berta, 1981). The center of radiation and dispersal of this 85, 1-350.
fossil hyaenid was the southern United States and Mexico. Chasma- Stirton, R. A., & Christian, W. G. (1940). A member of the Hyaenidae from
the upper Pliocene of Texas. Journal of Mammalogy, 21, 445-8.
porthetes became extinct in Eurasia during the early Pleistocene
(1941). Ailuraena Stirton and Christian referred to Chasmaporthetes Hay.
(late Villafranchian) and survived in North America slightly longer, Journal of Mammalogy, 28, 198.
becoming extinct in the early Irvingtonian. Hyaenids were ecologi- Thenius, E. (1965). Zur Stammesgeschichte der Hyanen. Zeitschrift fur
cally replaced on the North American continent by active predaceous Saugetierkunde, 31, 293-300.
carnivores, which included the American cheetah, Pelis trumani, Werdelin, L., & Solounias, N. (1991). The Hyaenidae: taxonomy, systemat-
ics and evolution. Fossils and Strata, 30, 1-104.
and the dire wolf, Canis dirus; the appearance of both in the middle
Wozencraft, W. C. (1989). The phylogeny of the Recent Carnivora. In Carni-
Pleistocene is preceded by the last record of Chasmaporthetes in vore Behavior, Ecology and Evolution, ed. J. L. Gittleman, pp. 495-
the early Pleistocene. 535. Ithaca: Cornell University Press.
 Part III: Archaic ungulates and ungulatelike mammals

15 Archaic ungulates and ungulatelike mammals

CHRISTINE M. JANIS, J. DAVID ARCHIBALD, RICHARD L. CIFELLI, SPENCER G. LUCAS,
CHARLES R. SCHAFF, ROBERT M. SCHOCH, and THOMAS E. WILLIAMSON

INTRODUCTION SYSTEMATICS

The term "ungulate" is virtually synonymous with "large placental

herbivore," although "ungulate" also can have taxonomic
UNGULATES VERAE
associations, referring to a group believed to be monophyletic. Al- Over the past few decades there has been a revolution in our notion
though modem ungulates are primarily specialized folivorous her- about what constitutes an "ungulate," especially with the inclusion
bivores (i.e., browsers and grazers), and only a few have a more of cetaceans as ungulates in the taxonomic sense. The most recent
generalized omnivorous diet (e.g., pigs), this was not the ecology comprehensive paper to review the phylogeny of the entire group is
of the original radiation of ungulates. During the early Tertiary, that of Prothero, Manning, and Fischer (1988) (see also Prothero,
in the Paleocene and the early Eocene, there were few specialized 1993). A major conclusion, supporting previous suggestions, was a
foliovores. Most archaic ungulates ("condylarths") have bunodont documented phylogeny indicating the basal ungulates, the Condy-
cheek teeth suggestive of a broadly omnivorous diet. Those larger larthra, to be paraphyletic. Another major conclusion, previously
mammals often portrayed as rather clumsy analogues to present-day suggested but not documented (e.g., Van Valen, 1978) was the rel-
rhinos and hippos, such as the uintatheres and pantodonts, are now atively primitive phylogenetic position of the Artiodactyla, more
regarded as not being ungulates in the taxonomic sense. In contrast, with a point of divergence prior to any condylarth family, rather
some animals that can be included within the grandorder Ungulata than being derived from a condylarth group such as arctocyonids
were playing ecological roles that are not usually associated with or hyopsodontids (see discussion in Van Valen, 1971). They ad-
the concept of ungulates, such as the carnivorous mesonychids. ditionally demonstrated the derived phylogenetic position of pro-
Modem orders of terrestrial ungulates, such as Artiodactyla, Peri- boscideans and their allies, long referred to rather pejoratively as
ssodactyla, and Proboscidea, are not known until the Eocene. De- paenungulates (= "near ungulates") (Simpson, 1945) or "subungu-
spite their first appearance in the early Eocene (or perhaps latest lates" (e.g., Romer, 1966). Prothero, Manning, and Fischer (1988)
Paleocene in the Old World), it was not until the late Eocene that also reviewed the current evidence concerning the supposed ungu-
members of modem orders attained large size and began assuming late affinities of the aardvark (Tubulidentata) (e.g., Patterson, 1975)
the role of large, specialized herbivores. The early Tertiary mam- and concluded that it cannot be included within the ungulate clade
mals grouped together in this section represent an assemblage of (see also discussion later).
two different types of animals: those apparently playing an un- Prothero, Manning, and Fischer (1988) only briefly discussed the
gulatelike role herbivorous prior to the emergence of the modem affinities of the numerous different types of endemic South Ameri-
ungulate orders, such as pantodonts and uintatheres, and animals can mammals (see Cifelli, 1993, for a recent review of this group).
that can be classified as ungulates in the taxonomic sense, the infa- As this book concerns only North American ungulates, however,
mous "Condylarthra," that may have been playing roles adopted by the problems of the affinities of these South American groups need
no subsequent true ungulate, such as the carnivorous mesonychids. not be of concern here, especially as the supposedly notoungulate
The predominance of these taxa in the earlier Tertiary and the very arctostylopids of North America have been assigned to a separate
different nature of the range of apparent behaviors and ecologies of order, the Arctostylopidae, so do not have affinities with this South
later Tertiary ungulates may be reflective of early Tertiary habitats American group (Cifelli, Schaff, and McKenna, 1989; Cifelli and
that were very different, in terms of structure and food availability, Schaff, this volume, Chapter 21).
from those found after the thermal maximum of the early middle Figure 15.1 illustrates a current "consensus phylogeny" of how
Eocene. the various "condylarth" taxa are related to other orders of

247
 Christine M. Janis et al. 248

PAENUNGULATA

CETE TETHYTHERIA

Figure 15.1. Consensus of the interrelationships of major groups of ungulates: based on Archibald (this volume, Chapter 20), Prothero, Manning, and
Fischer (1988), and Domning, Ray, and McKenna (1986). Key: "Condylarth" taxa are underlined; * indicates taxon not found in the North American
Tertiary; #indicates aquatic or semi-aquatic taxon not considered in this volume.

ungulates. Archibald (this volume, Chapter 20) provides an ex- Artiodactyla, Cetacea, Hyracoidea, Perissodactyla, Proboscidea,
tensive discussion about which taxa are now considered to belong Sirenia, and Tubulidentata. It is beyond the scope of this chapter to
within the Ungulata and which were previously included but have elaborate on the various opinions pertaining to this issue. However,
now been removed. Note that our concept of the diversity of basal Figure 15.2 provides a synopsis of the opinions of several authors,
ungulates has been considerably changed since prior to the 1960s, who have reviewed either the higher relationships among all eu-
following the removal of the omnivorous/carnivorous arctocyonid therians (Novacek, 1982, 1986; Novacek and Wyss, 1986; Wyss,
and mesonychid "condylarths" from an association with the superfi- Novacek, and McKenna, 1987; McKenna, 1987) or for those of the
cially similar Creodonta and Carnivora (see discussion in summary ungulates alone (Prothero, Manning, and Fischer, 1988), regarding
chapter on carnivorous mammals, this volume, Chapter 4). the possible relationships of these seven clades.
A complicating issue is whether the current expanded taxon Un- Morphological studies (Novacek, 1982, 1986, 1992; Novacek and
gulata is monophyletic. The apparent morphological synapomor- Wyss, 1986; Prothero, Manning, and Fischer, 1988) support the
phies are rather limited and may be subject to homoplasy (see monophyly of the Ungulata, with the exception of Tubulidentata,
discussion in Archibald, this volume, Chapter 20). Note that the which sometimes clusters with other eutherians (Figure 15.2D, E,
possession of hoofed terminal phalanges is not an ungulate synapo- I, and J). Although possibly an ungulate or the sister taxon of ungu-
morphy (Prothero, Manning, and Fischer, 1988). Novacek, Wyss, lates, aardvarks are not further considered here. Few of the molec-
and McKenna (1988) point out that ungulates have long been rec- ular studies depicted in Figure 15.2, however, indicate monophyly
ognized on a general gestalt of craniodental design for herbivores, for the Ungulata (even excluding Tubulidentata), but they do sug-
which has led to the past inclusion of many nonungulate groups in gest the presence of at least two major groups of "ungulates." One
the Ungulata (see discussion later). Some of these issues may be re- group includes Artiodactyla-Cetacea-Perissodactyla (A-C-P) and
solved as species of the probable sister tax on of Ungulata are known the other includes Hyracoidea-Proboscidea-Sirenia (H-P-S). Note
from the Late Cretaceous of western Asia (Nessov, Archibald, and that this clustering need not be at variance with the tree based on
Kielan-Jaworoska, 1997). morphology depicted in Figure 15.1 (if hyracoids were regrouped
A way of testing the issue of ungulate monophyly would be with tethytheres): Certain molecular procedures, such as distance
to compare various morphological and molecular evidence from measures, could group together a symplesiomorphic A-C-P versus a
the seven apparently monophyletic clades of extant eutherians that synapomorphic H-P-S. However, the current apparent inconsistency
have been considered members of the Ungulata. These taxa are between morphological and molecular data makes us wonder if it is
 Archaic ungulates and ungulatelike mammals 249

A F

B G

c H

D
'?I
-<,.~'Q x ~~ e,,<t- ~.t.,,,<1:- ~</;- ~</;-

E J

Figure 15.2. Recent interpretations for the relationships of the seven lineages (orders) of extant eutherians that have been variously allied in the taxon
Ungulata. A triangle represents a monophyletic clade of nonungulate eutherians whose internal relationships are not discussed here, but can be found in the
cited reference. The numbers (e.g., TUB+ 3, 2 others, etc.) indicate the number of additional ordinal-level clades not shown in the figure. See discussion
in Archibald (this volume, Chapter 20) for additional comments. Abbreviations: Ungulates: ART (Artiodactyla), CET (Cetacea), HYR (Hyracoidea), PER
(Perissodactyla), PRO (Proboscidea), SIR (Sirenia), and TUB (Tubulidentata); nonungulate eutherians: car (Carnivora), chi (Chiroptera), ede (Edentata),
and pri (Primates). A. Based on alpha hemoglobin (Wyss, Novacek, and Mc Kenna, 1987). B. Based on myoglobin (Wyss, Novacek, and Mc Kenna, 1987). C.
Based on myoglobin (McKenna, 1987). D. Based on morphology (Novacek, 1982). E. Based on morphology (Prothero, Manning, and Fischer, 1988). F. Based
on beta hemoglobin (Wyss, Novacek, and McKenna, 1987). G. Based on alpha crystallin A chain (Wyss, Novacek, and McKenna, 1987). H. Based on alpha
crystallin A chain (McKenna, 1987). I. Based on morphology (Novacek and Wyss, 1986). J. Based on unique cranial and jaw morphology (Novacek, 1986).

wise, at present, to formalize an association of "higher ungulates" McKenna (1987), drawing from both dental and molecular ev-
as the revised "Paenungulata" (linking peris_sodactyls with the old idence, has most recently discussed whether these two clades
concept of paenungulates; see Figure 15. l)(see also Prothero, 1993, (A-C-P and H-P-S) form an expanded and monophyletic Ungulata.
for the notion of a grouping of "Altungulata," and discussion in In 1975, he argued, mostly on the basis of some Cretaceous and
Archibald, this volume, Chapter 20). early Tertiary mammals, that the primitive number of postcanine
 Christine M. Janis et al. 250

teeth in eutherians was five premolars and three molars, rather than larger size and more obviously ungulatelike herbivorous diet than
only four premolars and three molars as traditionally assumed. The taeniodonts or tillodonts (which more closely resemble variants on
presence of five antemolar teeth in Eocene sirenians (Damning, large rodents). They were both originally considered to be members
Morgan, and Ray, 1982; see Luckett, 1993, for a contrasting view of the Proboscidea until Cope (1875) united them in their own order,
of homologies), plus support from studies on myoglobin and the Amblypoda. Features of both the teeth and the tarsal joint, however,
alpha crystallin A chain, suggests, according to McKenna (1987), suggest that they are not of ungulate affinities (see Cifelli, 1983;
that Hyracoidea-Proboscidea-Sirenia (and probably Tubulidentata) Lucas, 1993). The periptychid "condylarths" were long allied with
constitute a monophyletic clade that split very early in eutherian evo- the Amblypoda as well until they were removed to the Condylarthra
lution rather than having any special affinity with other ungulates. by Simpson (1937) on the basis of detailed studies of the dentition
This clade (with some modification in contents since its concep- and the tarsus.
tion) was named the Tethytheria by McKenna (1975). (Note that in Pantodonts have been considered as firmly removed from the
Figure 15.1 we limit the notion of the Tethytheria to the grouping ungulates for a couple of decades, with only a few current dis-
of Proboscidea, Sirenia and Desmostylia, following Damning, Ray, senters, following McKenna's (1975) review of mammalian phy-
and McKenna, (1986).) This very early split for the tethytheres is a logeny, and are currently considered to be related to didelphodon-
very intriguing possibility, but we remain swayed by the other ev- tines (grandorder Ferae) (see Lucas, 1993; this volume, Chapter
idence (some of which is noted in Archibald, this volume, Chapter 18). Dinoceratans are more problematical: Lucas (1993; Lucas and
20) discussed by Novacek (1982, 1986), Novacek and Wyss (1986), Schoch, this volume, Chapter 19) advocates their removal from the
Prothero, Manning, and Fischer (1988), and Prothero (1993) that ungulates, uniting dinoceratans with pyrotheres (a group of enig-
argues for the monophyly of the Ungulata. matic South American "ungulates") in the Uintatheriamorpha, re-
One major disagreement is in the placement of the Hyracoidea. lated to Pseudictops-like anagalids. In contrast, Prothero, Manning,
Both gross morphology and the molecular data tend to ally hyra- and Fischer (1988; see also Prothero, 1993) follow McKenna and
coids with the tethytheres in some permutation (Figure 15.2; see Manning ( 1977) in uniting uintatheres into an expanded "Paenungu-
also Shoshani, 1993). Prothero, Manning, and Fischer ( 1988), how- lata" (also including perissodactyls). They would place Dinocerata
ever, argue that hyracoids are perissodactyls and have their closest plus pyrotheres between Phenacodontidae and Embrithopoda on the
alliance among the palaeotheres, citing dental, cranial, and postcra- cladogram in Figure 15.L
nial characters. Prothero (1993) reviews subsequent evidence on
this issue, concluding that hyracoids are at least the sister taxon to
perissodactyls, a view also supported by Fischer and Tassy (1993) EVOLUTIONARY AND
(although note that they now term the group hyracoids plus peris- BIOGEOGRAPHICA L PATTERNS
sodactyls as the Perissodactyla, renaming the traditional notion of
the Perissodactyla the Mesaxonia). Note, however, that although Because archaic ungulates comprise more than half of the known
different molecular procedures have yielded a variety of different species of mammals in the early Paleocene (Archibald, 1983), their
eutherian phylogenies, none has suggested a link between perisso- patterns of early radiation must be strongly reflective of the radi-
dactyls and hyracoids. Figures 15.1 reflects our current uncertainty ations of mammals as a whole following the terminal Cretaceous
about the interrelationships of higher ungulates, in placing perisso- extinctions. Maas and Krause ( 1994) have recently reviewed the cur-
dactyls in a polychotomy with the various other ungulates previously rent literature on changes in mammalian community structure and
grouped as "subungulates." fauna! turnover in the North American Paleocene and early Eocene.
Their analyses support the notion that the characteristic Paleocene
community structure was basically in place by the end of the Puer-
UNGULATELIKE MAMMALS
can (earliest Paleocene), but do not support a previously suggested
The "ungulatelike mammals" covered in this section include the Tae- turnover between the early and the late Paleocene (between Tor-
niodonta, Tillodontia, Pantodonta, and Dinocerata. In rejonian and Tiffanian Land Mammal ages). In the late Paleocene
contrast to the other three groups, taeniodonts have not been consid- (late Tiffanian and Clarkforkian) change is again seen, with an in-
ered as ungulates by past workers. Some attempts were made to ally crease in overall body size and an increase in herbivores/folivores
them with edentates, mainly because of the loss of tooth enamel and (Maas and Krause, 1994, and references therein).
fossorial postcranial specializations of derived taeniodonts, but these
features were clearly derived convergently in the two groups. They
are currently considered to be of leptictimorph (including palaeo-
EARLY PALEOCENE FAUNAS
ryctids, leptictids, and pantolestids) affinities (Schoch, 1986; Lucas, In the earliest Paleocene, Puercan 0 and Puercan 1, none of the
Schoch, and Williamson, this volume, Chapter 16). Tillodonts have ungulatelike mammals groups are known (see Figure 15.3), but
long been considered as having "condylarth" affinities, but may now among the "condylarths" there were representatives of triisodon-
be best considered as the sister tax on to pantodonts (see Lucas, 1993; tines (basal Cete, originally considered to be arctocyonids), hyop-
Lucas and Schoch, this volume, Chapter 17). sodontids, periptychids, and the basal genus Protungulatum. Out
Pantodonts and uintatheres (Dinocerata) have a more substan- of ten genera with definite records in this time interval, at least
tial history of being placed among the ungulates because of their three (the triisodontine Ragnarok, the periptychids Maiorana and
 Archaic ungulates and ungulatelike mammals 251

6 iii' iii' II. II. II.

<( ~ 6 ~ ~
rn ~ ~ z 6
~
~
... _u c
e-::- c... '(ii"'
c
c·-
Ul

E-·i ... -
Ul
c
E·~
Ul !!:.
'lii
~i
Ul
~ (ij 'le
i::i e -I! ·.::fl u"'
0
... Ul
c"' i! IXI Gi=
.clXI
......
GI -
.c II. ... II. GI -
.c II.
u"' 0

MA r:: e 8i ... GI
= ~8 t: 1ii :; 1U :I"' ; 1ii t: 1ii u .c
ui
"'~ <:( u5j <(E z0 ...GI rn0 ... ...GI z0 ...GI
GI 0 GI U "' t:
~all rn ...
...J :I II. 0
CJ CJ CJ CJ CJ CJ z :Z
34.5
~,:,@
"O cU ·-
!J!
35.5 ~Be
37.1 I
39.5 c
Q) .m St,
-cU ·-
c
_J :::i
Figure 15.3. Biogeographic ranges
of ungulatelike mammals and Euthe-
/ ria incertae sedis. A "box" (for a
/ particular time period in a particu-
lar biogeographic region) that has a
fil, / cross through it means no fossil lo-
/ calities are known for that time pe-
riod from that area; a single dashed
Tr / line through the box means only
Ti, Tr, I scant fossil information is available
By, Un /
I (usually only a single, small, local-
ity). Key: Taeniodonta are in uni-
Es, Tr, ?Tr,/
talicized roman type. "Conoryctidae"
Cr, /Cr
are in regular roman type: Cn =
By Conoryctes, Ct= Conoryctella, Hf=
Hueifanodon, On = Onychodectes,
Sc = Schochia. Stylinodontidae are
in plain type underlined: Ee =
Ectoganus, ~ = Psittacotherium,
.fil = Stylinodon, Wo = Wortma-
nia. Tillodontia are in plain-type ital-
ics: An = "Anchippodus", Es = Es-
thonyx, Mg = Megalesthonyx, Tl =
Tillodon, Tr = Trogosus. Pantodonta
are in boldface italics: Bl = Bary-
lambda, Cl = Caenolambda, Cr =
Coryphodon, Hl = Haplolambda,
fl = lgnatiolambda, Ll = Lep-
tolambda, Pl = Pantolambda, Ti
= Titanoides, P? = pantodont in-
det. Dinocerata are in plain bold-
face: By = Bathyopsis, Eo =
Eobasileus, Pd= Prodinoceras, Ty=
Tethyopsis, Un = Uintatherium, U?
= Uintathere indet. Eutheria incer-
tae sedis are in boldface italics under-
lined: Cy = Cyriacotherium, De =
Deltatherium, M = ldiogenomys,
Mi = Mingotherium.

Mimatuta, and possibly also the hyopsodontid Oxyprimus) are not between the earliest Puercan and Puercan 2, together with the num-
known from the later Puercan, and only two genera (the periptychid berof taxa unique to the Puercan 0-1 interval, reinforces Van Valen 's
Anisonchus and the mioclaenid Promioclaenus) survived past the (1978) notion of an earliest Paleocene "Mantuan" fauna! zone (for
end of the Puercan (see Figures 15.4, 15.5). an alternative view of early Puercan fauna! zontations, see Archibald
A rather distinct change was seen by the Puercan 2 interval. Tae- and Lofgren, 1990, and Lofgren, 1995). An increase in body size
niodonts made their first appearance with the primitive members of mammals is also seen between earliest and later Puercan faunas
of both the conoryctids (Onychodectes and Schochia) and the more (Van Valen, 1978).
derived stylinodontids (Wortmania) (see Figure 15.3). Arctocyonid Although there was some fauna! turnover, Puercan 2 and 3 ap-
and mioclaenid ungulates also made their first appearance; perip- pear to be fairly similar time intervals regarding the diversity of
tychids and triisodontids remained diverse. This fauna! difference ungulates and taeniodonts. As discussed in the summary chapter on
 Christine M. Janis et al. 252

6 iii' iii' IL IL IL z 6
E. < ~ ec: e 2- ~ !!:.. ~
<
2- iii _
._ 2- c:
c:
c: ·- c: .5
t/I t/I
c: c: .5
t/I
U
"lii
GI "lii
"lii '(ij
"' = ... "' "! :i ... "' :eu .c3:
-
i; ...
-g ee Cll u"'
0 'EGI t/IIll ~m m
.cm GI -
.c II.
:; 'Iii
- II.
GI -
.c II.
'Iii
"'
0
u
:t: u8 t: - :;om GI -Ill
c: t: "' t:
II. 0
ui
MA ~~~ :s
0"'
z I!! rn ... GI
rn0 ... u I!! z0 ...GI z
CJ ~oll CJ CJ CJ CJ CJ z
56.0
~ <O
:::: c::
i= -~
56.2
~an
:::: c::
i=.~
57.0
~.q-
:::: c::
i= -~
57.3
~ (Y")
:::: c::
i= -~
58.5
~ (\J
:::: c::
i= .!!l
59.2
'
~~
:::: c::
i= -~
60.9
.::
0
-~("")
w ~ c::
zw 61.3 ~-~
(.) .::
0
0 ·~C\I
w ~ c::
..J
<t 62.5 ~ -~
Cl. .::
0
·~T""
~ c::
63.8
~ -~
c::
rl
Q;
::i "'
Cl..
64.0 c::
cU
<..l
(i; N
::i
Cl..
64.5
c::
cU
!::?
Q) ~

::i
Cl..
65.0 c::
cU
<..l
Q; 0
::i
Cl.. Figure 15.3. (Cont.)
65.1

carnivorous mammals (this volume, Chapter 4), early members of Mountains), whereas others (Anisonchus) were known only from
the Carnivora are known from only a couple of specimens in the the Great Plains. Similarly, the hyopsodontids, the triisodontine
Puercan, and even in the later Paleocene were small, generalized Goniacodon, and the mioclaenids Bubogonia and Litaletes appeared
predators. Puercan triisodontines such as Eoconodon may have been to be limited to the Great Plains; the triisodontine Platymastus was
predators on the other small mammals of the time. found only in the southern Great Basin (southern Rocky Mountains)
Some biogeographical differences are apparent in the later Puer- (see Figures 15.4, 15.5).
can: Mioclaenids were much more abundant (at least in terms of
generic diversity) in the southern Great Basin(= southern Rocky
MID-PALEOCENE FAUNAS
Mountains/San Juan Basin; see discussion in the Introduction) than
in the Great Plains; certain periptychids (Hemithlaeus, Gillisonchus) The distinct change seen by Maas and Krause ( 1994) between Puer-
were known only from the southern Great Basin ( = southern Rocky can and Torrejonian faunas is highly apparent in the ungulates and
 Archaic ungulates and ungulatelike mammals 253

6 iii' iii' IL II. IL

~ ec: e 2- ~ z 6
< E. 6 ~
2- "lii
_u
._
111~ _ c:
c:
'(ij
;c: ·-:a c: .5
t/I t/I
c:
E·~
t/I !!:..
"lii
i; ... "' ... "' ~:! u'iJ
:eu .c3: u"'0
-g ee Cll -!·c::GI
- fl u"'
GI -
~m GI -

-
- t/I
0
GI "'
c: 0
.cm .c II. 'E II. .c II.
'Iii '5 1ii 'S 1U 'Iii
uu ~m...
t: t:
MA 5j E :; z0 ...GI z0 ...GI ui
~~~ Uc:( ~I!! GI
~oll rn0 ... "' t:
II. 0

37.1
CJ CJ CJ CJ CJ CJ z z
/
/
t!g_

/
t!g, Mx
/
Figure 15.4. Biogeographic ranges
/ of more carnivorous archaic ungu-
fu'
I lates. A "box" (for a particular time
I I period in a particular biogeographic
HI, region) that has a cross through it
t!g, ~
/ I
/
means no fossil localities are known
.E§,§y for that time period from that area;
Tp, / a single dashed line through the box
HI, / means only scant fossil information
Mx, f!l is available (usually only a single,
small, locality). Key: Basal taxa are
in boldface italics: Pu = Protungu-
latum. Arctocyonidae are in regular
boldface: Ac = Anacodon, Ar= Arc-
tocyon, Ca= Carcinodon, Ch= Chri-
acus, Co = Colpoclaenus, Dm =
Desmatoclaenus, Dt = Deuterogon-
odon, La= Lambertocyon, Lo= Lox-
olophus, Mt= Mentoclaenodon, Mm
= Mimotricentes, Ot = Oxytomodon,
Pn = Princetonia, Pt = Prothrypta-
codon, Tp = Thryptacodon. Cete are
in plain typeface. Basal Cete are in
regular plain type: Ba= Baioconodon,
Eo = Eoconodon, Go = Goniacodon,
Mc = Microclaenodon, Oc = Oxy-
claenus, Pl= Platymastus, Ra= Rag-
narok, St= Stelocyon, Ti= Triisodon,
Wy = ~olestes. Hapalodectidae
are in plain-type italics: Hi = Ha-
palodectes. Mesonychidae are in plain
type underlined: Ak = Ankalagon,
Di = Dissacus, Hg = Harpagolestes,
Mx = MesonyX.-Pa = Pachyaena,
Sy = Synoplotherium.

ungulatelike mammals. Rose (1981) discussed the diverse nature dominant predator in the Torrejonian and Tiffanian. Arctocyonids
of North American Torrejonian mammal faunas, attributing this in increased in diversity throughout the Torrejonian, with the appear-
part to a warming trend in the Northern Hemisphere. Both cono- ance of the wolf-sizedArctocyon in Torrejonian 3 (see Figure 15.4).
ryctid and stylinodontid taeniodonts persisted but were now rep- The phenacodont Tetraclaenodon first appeared at the start of the
resented by new genera, including the larger, more specialized, Torrejonian, and the more derived phenacodontids Ectocion and
hypsodont stylinodontid Psittacotherium. Psittacotherium first ap- Phenacodus appeared by end of this land mammal age (see
peared in Torrejonian 2, as did the earliest pantodont (Pantolambda) Figure 15.5).
and the eutherian of possible tillodont affinities, Deltatherium (see The appearance of pantodonts and phenacodontids marked the
Figure 15.3). Torrejonian 2 also marked the first appearance of the first appearance of mammals with distinctly lophed teeth that might
coyote-sized mesonychid (Cete) Dis sac us, which was probably the be expected to have had a folivorous diet. Other workers have
 Christine M. Janis et al. 254

u m m ii:' ii:' ii:'

~ z u~
~ ~ ~ !2.
< ~ <
_u c: c: Ul Ul
c:
Ul e:.
!2. 'liica ca~ c: ·u; c: ·- c: .5
._ E·i 'lii 'lii
~ ~'lii
... ca ; :a ca
"! :! GI
- U GI ca
-~
GI -
0 ~m .c:m .c: .c: Q. 0

-g ee Cll ......
ca·- u GI ca Q. - Q. :E 3: u
u8 t:
o -ca :; 1U :; 1U c:
GI -ca t: 'Iii u.c:
ca t:
E :!::
u I!! z0 ...GI ui
~ I!!
-c: GI
MA :s z I!! rn0 ...
GI
Q. 0
~~~ ~ci: CJ ~oll CJ CJ CJ CJ CJ z z
56.0
~ <O
:::: c::
F.!!!
56.2
~ lO
:::: c::
F.!!!
57.0
~ "d" Ac, Ar, Ch,
:::: c:: Tp
F.!!!
57.3
~ ('"j
Ar, Ch, Co,
:::: c:: Mm,Tp
F.!!!
58.5
~C\J
/
:::: c:: /
F.!!! /
59.2
' Ar, Ch, Ar, Ch, Co,
~~
:::: c:: Co, Mm Dm,Mm, Tp,
F.!!! Qi
60.9
C: Ar, Ch, Co,
w 0
-~('")
z Mm, Pt, Tp,
w 0 c:: Qi
(.) I- .!!!
0 61.3 c::'
w 0
..J -~C\I
<t
Cl. 0 ffi
I-·-
62.5
C:
0
-~.,.. /
0 ~
I-·-
/ Oc
63.8 c::
ca Pu,
<..l Dm,Lo,
Q; (".) Ch,
::::l Go
(L Eo, Oc
64.0 c:: Pu,
Dm,Lo,
rl Ca,Dm,
Q; N Oc Lo, Pt,
::::l
(L Ba, Eo, Oc
64.5
c::
Pu,
~
Q) ~

::::l
(L
65.0
rl
Q; 0
::::l
(L
65.1 Figure 15.4. (Cont.)

remarked on the fact that a mammalian community containing a the southern Great Basin (southern Rocky mountains) (although the
diversity of truly herbivorous (folivorous) mammals was not appar- picture may be biased by the lack of central Great Plains localities
ent until the later Eocene (e.g., Collinson and Hooker, 1987; Wing in Torrejonian 2).
and Tiffney, 1987; Stucky, 1990). Maas and Krause (1994) present a
category of "herbivore/folivore" mammals present since the earliest
LATE PALEOCENE FAUNAS
Paleocene, but it is not apparent from their paper which taxa they
are actually including in this category. In contrast to the observations of Maas and Krause (1994), con-
Biogeographical differences are not as apparent in the Torrejonian siderable change can be observed between the Torrejonian and
as in the later Puercan, although a few genera (such as Deltatherium the Tiffanian in the ungulates and ungulatelike mammals. Rose
and the cetans Triisodon and Ankalagon) appear to be restricted to (1981) also discusses a reduction in the abundance of ungulate
 Archaic ungulates and ungulatelike mammals 255

u iii' iii' ii:'

~
ii:'
!2.
ii:'
z u~
< ~ u ~ ~ ~
e:.
!2. 'lii iii_
._ !2. c:
c:
'(ij
c:
c: ·- c: .5
._ ca
Ul Ul
c:
E·i
Ul

-:a 'lii
~ ca ca ... ca ... Ul
GI ca ~~ u ca

-
GI -
"! .g 0 ca
-c: Ul ~m .c:m .c: GI
.c: - 0
e u GI 0
uu ~ 1ii :; -;
Q. - Q.
c: -
Q.
'Iii
:E 3:
u.c: u
MA
-g e Cll 5i E
- GI :;
t:
o -ca
z I!! rn0 ...
GI 0 GI
rn ... ~~
t:
z0 GI... ca t: ui
~~~ uci: CJ ~oll CJ CJ CJ CJ CJ
Q. 0
z z
35.5
Hy
37.1
?Hy

Hy Hy

/
Figure 15.5. Biogeographic ranges
Hy / of more omnivorous/herbivorous ar-
chaic ungulates. A "box" (for a par-
Hy, / ticular time period in a particular bio-
/ I geographic region) that has a cross
Pc
through it means no fossil localities
I
Hy, are known for that time period from
/
that area; a single dashed line through
the box means only scant fossil infor-
mation is available (usually only a sin-
/
gle, small, locality). Key: Hyopsodon-
tidae are in boldface: Al = Aletodon,
Do = Dorraletes, HI = Haplaletes,
Hm = Haplomylus, Hy= Hyopso-
dus, Lm = Litomylus, Op= Oxyprim-
nus, Ut = Utemylus. Bulbulodentata
are in plain typeface. Mioclaenidae
are in plain type underlined: Ap =
Apheliscus, Bu = Bubugonia, Ch =
Choeroclaenus, .§! = Ellipsodon, .!J
= Litaletes, Mc = Mioclaenus, Pd
= Phenacodaptes, Pm = Promio-
claenus, Ps = Protoselene, Tn = Tiz-
natzinia. Periptychidae are in regu-
lar plain type: An = "Anisonchus,"
Ca = Carsioptychus, Cc = Cona-
codon, Es = Escatepos, Et = Ecto-
conus, Gi = Gillisonchus, He =Hap-
loconus, He = Hemithlaeus, Ma =
Maiorana, Mm = Mimatuta, Oc =
Oxyacodon, Pe = Periptychus, Tn =
Tinuviel. Phenacodonta are in plain-
type italics: Cp = Copecion, Ee= Ec-
~ tocion, Me = Meniscotherium, Pc =
.E Phenacodus, Tr = Tetraclaenodon.
~~
..!!! c:: Arctostylopidae are in boldface italic:
56.0 L......;0
::...:·=!!!~'--~-'"-~~--'"-~--'~~~-"''--~--''---"~--'----'~---'~~~---':.ll..--lJL..__..l Ar= Arctostylops.

species in the Tiffanian in comparison to the Torrejonian, and links Tiffanian 1, but the more derived stylinodontid Ectoganus appeared
this to cooling climatic conditions. The cast of Tiffanian ungulate- possibly as early as Tiffanian 3. Dinoceratans also first appeared in
like mammals was rather different from that of the Torrejonian: Tiffanian 3, in the form of the genus Prodinoceras (see Figure 15.3).
The pantodont Titanoides originated in Torrejonian 3 and persisted Among the more carnivorous archaic ungulates, triisodontines
almost throughout the Tiffanian, and many new pantodont taxa disappeared after the earliest Tiffanian, and mesonychids were rep-
emerged in the Tiffanian. Pantodonts reached a peak in diversity resented only by Dissacus, but arctocyonids remained diverse (see
in the later Tiffanian with four (possibly five) sympatric genera Figure 15.4). Among the more herbivorous ungulates, Tetraclaen-
and with the additional appearance of the pantodontlike mammal odon did not survive past Tiffanian 1, but the phenacodontids
Cyriacotherium. Conoryctine taeniodonts were unknown after Phenacodus and Ectocion persisted throughout the Tiffanian and
 Christine M. Janis et al. 256

ii:' ii:'
<
u
~ <
iii'
~
iii'
!!!.
ii:'
!!!. 2. ~ ze:. u~
2. c: c: Ul Ul
c:
E·~
Ul
...
'Iii ......2.
"iii E·~ c: ·- c: .5 Ul

......= .r:......:s ..."'"'

"iii 'iij 'Iii
~ "iii.~ "' ... Ul GI - ... - GI -
:eu.r:3: u"'
U GI 0
......GI 0 ~m Q;
.r:m Q. ... Q. .r: Q.

-g ee .,. u GI "'
c: 0 t: ... c: ... t: 1ii
Cll 5j E uu om
z ... 0 "'
:s GI
rn ... 0 GI
rn ...
GI Ill
u !!! z0 ...GI "' t: u.i
MA ~~~ Uc:( ~oll CJ CJ CJ CJ CJ
Q. 0
z z
56.0
~ <O
:::: c
i= .~
56.2
~ lO
:::: c
i= .~
57.0
~ "d"
EQ,
Do,Lm,
:::: c
i= .~ Ee, Pc
Ee, Pc
57.3
Al, Do, HI, Lm,
~ ('"j
?U, E!!l. El;,
Al,Lm,
:::: c Ee, Pc
i=.~ Ee, Pc
58.5
Hl,Lm, Ee, Pc /
~ C\J .Em. E§., /
:::: c
i= .~ Ee, Pc, /

59.2 Hl,Lm,
'
~~ !J..
:::: c An, Pe, An,
i= .~
Ee Ee, Te
60.9
c
w 0
-~('")
z
w 0 ~
(.) I-·-
61.3
0
w 0
.::
..J -~C\I
<t ... c
Cl. ~.~
62.5
.::0
-~~
0 ~
1-.-
63.8
c
cU
<..l
An, He,
Q;
::::l "'
Cl.. Oc, Tn
64.0 c
cU Lm, ?Op
<..l
Q; N 6!L ill
::::l
Cl.. An, Oc
64.5
c Op, Op,
cU
~
Q) ~

::::l
Cl..
65.0 c
cU
<..l
Q; 0
::::l
65.1 Cl.. Figure 15.5. (Cont.)

into the Eocene. Periptychus was the only remaining periptychid, as the dinoceratan Prodinoceras, the taeniodont Ectoganus, and the
surviving through Tiffanian 4. Hyopsodontids remained relatively archaic ungulates Mimotricentes, Dorraletes, and Phenacodaptes,
diverse, with the addition of more dentally specialized taxa such as were apparently restricted to the Great Plains region.
Aletodon and Dorraletes. Mioclaenids persisted, with the addition Maas and Krause (1994) note a significant change in mammalian
of more derived taxa such as Phenacodaptes (see Figure 15.5). community structure in the late Paleocene, in the latest Tiffanian
There is again some suggestion ofbiogeographical isolation of the and Clarkforkian (but note that they lump the intervals Tiffanian 6
southern Great Basin (southern Rocky Mountains): The pantodont and Clarkforkian 1 together in their analysis). They argue that this
Ignatiolambda, the taeniodont Psittacotherium, and the archaic change was related both to immigration events (for example, the
ungulates Utemylus (Hyopsodontidae) and Periptychus were known appearance of the first rodents) and to climatic warming. A number
only from this area during the Tiffanian. In contrast, other taxa, such of larger-sized mammals made their first appearance in the latest
 Archaic ungulates and ungulatelike mammals 257

Tiffanian, including the predatory oxyaenids (see summary chapter rhinoceros-sized homed genera such as Tetheopsis and Uintatherium
on carnivorous mammals, this volume, Chapter 4). appeared in the later Bridgerian (see Figure 15.3).
Several new type of ungulates and ungulatelike mammals first Despite the great radiation of both the creodonts (see Gunnell,
appeared at this time: the first tillodont, Esthonyx, in Tiffanian 6, and this volume, Chapter 5) and the archaic carnivorans (see Flynn,
the enigmatic ungulate Arctostylops in Tiffanian 5. The pantodont this volume, Chapter 6), the diversity of the carnivorous archaic
Coryphodon, a prominent member of early Eocene faunas, first ap- ungulates also increased in the early Eocene. The mesonychid Dis-
peared in Clarkforkian 1. Although pantodonts retained a moderate sacus, persisting from the mid-Paleocene, was joined by the ha-
diversity throughout the Clarkforkian, by the earliest Eocene only palodectid Hapalodectes, another mesonychid Pachyaena (a larger,
Coryphodon remained. Ectoganus persisted through the interval as more specialized potential predator), and the smaller, question-
the only surviving taeniodont (see Figure 15.3). Among the archaic able cetan, Vllyolestes. Arctocyonids, including the genus Chria-
ungulates, Dissacus persisted together with a modest diversity of cus (which first appeared at the same time as Dissacus), remained
arctocyonids (see Figure 15.4). A moderate diversity was retained moderately diverse through the Wasatchian, but by the early Bridge-
among the hyopsodontids, with the Clarkforkian appearance of the rian only Thryptacodon remained, making its final appearance. The
genera Haplomylus and Hyopsodus, that survived into the Eocene. Bridgerian was also the time of the greatest diversity of mesony-
The mioclaenids persisted in Apheliscus and Phenacodaptes, but chids, with the addition of larger, specialized taxa such as Mesonyx
note that Archibald (this volume, Chapter 20) has some doubts as and Harpagolestes, although hapalodectids made a final appearance
to the true mioclaenid affinities of these taxa. Perhaps the most in the middle Bridgerian (see Figure 15.4). A similar pattern of con-
significant herbivore fauna! change in the Clarkforkian was the ad- tinuation of latest Paleocene trends into the Wasatchian, but turnover
dition of phenacodontid taxa with more complexly lophed teeth, in the Bridgerian, is seen in the more herbivorous archaic mammals,
indicative of a more herbivorous diet requiring shearing rather than with the loss at the turnover between early and middle Eocene of
crushing, such as Copecion and Meniscotherium (see Figure 15.5). the hyopsodontid Haplomylus, the mioclaenid Apheliscus, and the
Sampling of areas outside of the central Great Plains is too poor dur- phenacodontid Copecion (see Figure 15.5).
ing this time interval to examine patterns of biogeographical varia- This pattern of diversification among the ungulates and ungulate-
tion, but note the survival of the pantodont Bary lambda into the late like mammals paralleled that seen among the carnivorous mam-
Clarkforkian of the southern Great Basin (southern Rocky Moun- mals (this volume, Chapter 4): an early Eocene flourishing, but the
tains), known previously from the mid-Tiffanian of the central Great turnover from early to middle Eocene bringing faunal changes, loss
Plains. of smaller, more archaic forms, but also the diversification of some
larger, more specialized forms before the extinction of the lineages.
EOCENE FAUNAS Given the fact that the major immigrations to North America were
in the earliest Eocene, but that climatic cooling commenced in the
Although both derived carnivores (hyaeodontids; see Chapter 4) and early middle Eocene, this strongly suggests that it was not compe-
derived ungulates (artiodactyls and perissodactyls; see Chapters 22 tition with these immigrants that caused the demise of the larger
and 35) first appeared as immigrants to North America in the early archaic North American mammals, but rather the effects of climatic
Eocene, there appeared to be little immediate effect on the exist- change, probably the reduction of tropical-like forest habitats.
ing generic diversity of archaic larger herbivorous and carnivorous Note that, in parallel with the carnivorous mammals, some archaic
mammals; however, their relative abundance decreased (Van Valen herbivores appear to have indulged in late middle Eocene "Uintan
and Sloan, 1966; Rose, 1981). It appears likely that the climatic experimentation," presumably in response to climatic changes. Al-
warming of the early Eocene allowed for diversification of both en- though most ungulatelike mammals were extinct by the start of
demic and immigrant animals, at least in North America (archaic the Uintan (with the exception of the highly specialized taeniodont
mammals apparently suffered more of a decrease in diversity in the Stylinodon ), uintatheres were at the height of their generic diversity
early Eocene of Europe: Russell, 1975). (as, of course, their name suggests), with three large, specialized
Ungulatelike mammals showed little change in generic diversity taxa. The large, carnivorous mesonychids persisted into the Duch-
in the earliest Eocene over the condition in the latest Paleocene, esnean, although with reduced diversity after the middle Bridge-
with the exception of the loss of all pantodonts except Coryphodon rian. Finally, although the herbivorous phenacodontids dwindled in
(which was a common element of the fauna) and the enigmatic diversity in the early middle Eocene, and were extinct by the Uintan,
pantodontlike Cyriacotherium. Diversity was increased later in the the hyopsodontid Hyopsodus persisted (although as a rare fauna! el-
early Eocene by the addition of the tillodonts Megalesthonyx and ement) into the early Chadronian. There appears to be no discernible
Trogosus and the taeniodont Stylinodon. Some changes were seen pattern ofbiogeographic variation among ungulate and ungulatelike
in the early middle Eocene: The taeniodont Ectoganus had disap- mammals during the Eocene.
peared, and Coryphodon and the tillodonts Esthonyx and
Megalesthonyx were known only from the early Bridgerian, al-
SUMMARY
though Tillodon appeared in the middle Bridgerian. Perhaps the
most notable event was an increase in the diversity of dinoceratans: There appear to have been several distinct faunal groupings of
The uintathere Bathyopsis first appeared in the late Wasatchian archaic ungulates and ungulatelike mammals through the early
(which also marked the last appearance of Prodinoceras), and Tertiary. An early Paleocene (Puercan) fauna was composed of
 Christine M. Janis et al. 258

triisodontine, hyopsodontid, periptychid, arctocyonid, and mioclae- Cifelli, R. L. (1983). The origins and affinities of the South American Condy-
larthra and early Tertiary litopterna (Mammalia). American Museum
nid archaic ungulates, together with conoryctid and primitive stylin-
Novitates, 2772, 1-48.
odontid taeniodonts. A case could be made for a distinct early "Man- (1993). The phylogeny of the native South American ungulates. In Mam-
tuan" fauna within this assemblage, comprising some unique genera mal Phylogeny: Placentals, eds. F. S. Szalay, M. J. Novacek, & M.
of archaic ungulates and lacking arctocyonids, mioclaenids, and tae- C. McKenna, pp. 195-216. New York: Springer-Verlag.
niodonts. Cifelli, R. L., Schaff, C.R., & McKenna, M. C. (1989). The relationships
of the Arctostylopidae (Mammalia): new data and interpretation.
A mid-Paleocene fauna may be apparent from the early Torre-
Bulletin of the Museum of Comparative Zoology, 152, 1-44.
jonian through the late Tiffanian. This fauna was heralded by the Collinson, M. E., & Hooker, J. J. ( 1987). Vegetational and mammalian fauna!
appearance of pantodonts, derived stylinodontid taeniodonts, the changes in the Early Tertiary of southern England. In The Origin of
phenacodont Tetraclaenodon, and carnivorous hapalodectid archaic Angiosperms and Their Biological Consequences, eds. E. M. Friis
ungulates such as Dissacus. The fauna could possibly be subdi- & W. G. Chaloner, pp. 259-303. Cambridge: Cambridge University
Press.
vided into an earlier, Torrejonian, one, and a later, Tiffanian, one.
Cope, E. D. (1875). On the order Amblypoda. American Naturalist, 9, p. 427.
Dinoceratans first appeared during the Tiffanian, and hyopsodon- Domning, D. P., Morgan, G. S., & Ray, C. E. ( 1982). North American Eocene
tids diversified, but triisodontines and the conoryctine taeniodonts sea cows (Mammalia: Sirenia). Smithsonian Contributions to Pale-
were now exinct. During the Tiffanian there was also a decline in obiology, 52, 1-69.
the diversity of periptychids (with their eventual extinction) and Domning, D. P., Ray, C. E., & McKenna, M. C. (1986). Two new Oligocene
desmosty Ii ans and a discussion of tethytherian systematics. Smith-
mioclaenids.
sonian Contributions to Paleobiology, 59, 1-56.
Little evidence exists for a sharp fauna! transition at the Pale- Fischer, M. S., & Tassy, P. (1993). The interrelation between Proboscidea,
ocene/Eocene boundary; rather, there appears to be a fauna! as- Sirenia, Hyracoidea, and Mesaxonia: the morphological evidence. In
semblage that first appeared in the latest Tiffanian and carried on Mammal Phylogeny: Placentals, eds. F. S. Szalay, M. J. Novacek, &
through into the early Eocene. The latest Tiffanian heralded the M. C. McKenna, pp. 217-34. New York: Springer-Verlag.
Lofgren, D. L. (1995). The Bug Creek problem and the Cretaceous-Tertiary
first appearance of the tillodonts and arctostylopids, and throughout
transition at McGuire Creek, Montana. University of California Pub-
the latest Paleocene there was a moderate diversity of phenacodon- lications in the Geological Sciences, 140, 1-185.
tids, arctocyonids, mesonychids, hyopsodontids, and taeniodonts. Lucas, S. G. (1993). Pantodonts, tillodonts, uintatheres, and pyrotheres are
By the start of the Eocene, pantodont diversity had been reduced to not ungulates. In Mammal Phylogeny: Placentals, eds. F. S. Szalay,
a single genus, Coryphodon, arctostylopids were extinct, and larger M. J. Novacek, & M. C. McKenna, pp. 182-94. New York: Springer-
Verlag.
mesonychids had made their first appearance. A moderate increase
Luckett, W. P. (1993). An ontogenetic assessment of dental homologies
in tillodont diversity occurred toward the end of the early Eocene. in therians. In Mammal Phylogeny: Placentals, eds. F. S. Szalay,
Finally, a middle Eocene fauna! assemblage represents the decline M. J. Novacek, & M. C. McKenna, pp. 182-204. New York: Springer-
in diversity and eventual extinction of almost all of these groups by Verlag.
the late Uintan. Some groups appeared to decline more rapidly than Maas, M. C., & Krause, D. K. (1994). Mammalian turnover and community
structure in the Paleocene of North America. Historical Biology, 8,
others: The early Bridgerian records the last arctocyonid and the last
91-128.
of several long-lived genera: Coryphodon (Pantodonta), Ectoganus
McKenna, M. C. (1975). Toward a phylogenetic classification of the Mam-
(Taeniodonta), and Ectocion (Phenacodontidae). In contrast, uin- malia. In Phylogeny of the Primates: A Multidisciplinary Approach,
tatheres and mesonychids appear to have experienced "last ditch" eds. W. P. Luckett & F. S. Szalay, pp. 21-46. New York: Plenum.
diversification, with the appearance of larger, specialized taxa in the (1987). Molecular and morphological analysis of high-level mammalian
interrelationships. In Molecules and Morphology in Evolution: Con-
Uintan. Past the Uintan, the mesonychid Harpagolestes persisted
flict or Compromise?, ed. C. Patterson, pp. 55-93. Cambridge: Cam-
into the Duchesnean, and the hyopsodontid Hyopsodus persisted bridge University Press.
into the early Chadronian. McKenna, M. C., & Manning, E. (1977). Affinities and paleobiogeographic
significance of the Mongolian Paleogene genus Phenacolophus.
Geobios, Memoire special, 1, 1-85.
Nessov, L.A., Archibald, J. D., and Kielan-Jaworoska, Z. (1997). Ungulate-
ACKNOWLEDGMENTS like mammals from the Late Cretaceous of Uzbekistan and a phylo-
genetic analysis of a higher taxon of ungulatelike mammals. Bulletin
We thank Leigh Van Valen for comments on the original manuscript. of the Carnegie Museum of Natural History (in press).
Novacek, M. J. (1982). Information for molecular studies from anatomi-
cal and fossil evidence on higher eutherian phylogeny. In Macro-
molecular Sequences in Systematic and Evolutionary Biology, ed.
REFERENCES M. Goodman, pp. 3-41. New York: Plenum.
(1986). The skull of leptictid insectivorans and the higher-level classifi-
Archibald, J. D. (1983). Structure of the K-T mammal radiation in North cation of eutherian mammals. Bulletin of the American Museum of
America: speculations on turnover rates and trophic structure. Acta Natural History, 183, 1-111.
Palaeontologica Polonica, 28, 7-17. (1992). Mammalian phylogeny: Shaking the tree. Nature, 356, 121-5.
Archibald, J. D., & Lofgren, D. L. (1990). Mammalian zonation near the Novacek, M. J., & Wyss, A. R. (1986). Higher-level relationships of
Cretaceous/Tertiary boundary. In Dawn of the Age of Mammals in the recent eutherian orders: morphological evidence. Cladistics,. 2,
the Northern Part of the Rocky Mountain Interior, North America, 257-87.
eds. T. M. Bown & K. D. Rose, pp. 31-50. Boulder, Colo.: Geological Novacek, M. J., Wyss, A. R., & Mc Kenna, M. C. (1988). The major groups
Society of America, Special paper 243. of eutherian mammals. In The Phylogeny and Classification of the
 Archaic ungulates and ungulatelike mammals 259

Tetrapods, Vol. II: Mammals, ed. M. J. Benton, pp. 31-71. System- Shoshani, J. (1993). Hyracoidea-Tethytheria affinity based on myological
atics Association Special Volume no. 35B. Oxford: Clarendon. data. In Mammal Phylogeny: Placentals, eds. F. S. Szalay,
Patterson, B. (1975). The fossil aardvarks (Mammalia: Tubuliden- M. J. Novacek, & M. C. McKenna, pp. 235-56. New York: Springer-
tata). Bulletin of the Museum of Comparative Zoology, 147, Verlag.
185-237. Simpson, G. G. (1937). The Fort Union of the Crazy Mountain Field and
Prothero, D. R. (1993). Ungulate phylogeny: molecular vs. morphologi- its mammalian fauna. United States National Museum Bulletin, 169,
cal evidence. In Mammal Phylogeny: Placentals, eds. F. S. Szalay, 1-287.
M. J. Novacek, & M. C. McKenna, pp. 173-81. New York: Springer- ( 1945). The principles of classification and a classification of mammals.
Verlag. Bulletin of the American Museum of Natural History, 85, 1-350
Prothero, D.R., Manning, E. M., & Fischer, M. (1988). The phylogeny of the Stucky, R. (1990). Evolution of land mammal diversity in North America
ungulates. In The Phylogeny and Classification of the Tetrapods, Vol. during the Cenozoic. In Current Mammalogy (2nd ed.), ed. H. H.
II: Mammals, ed. M. J. Benton, pp. 201-34. Systematics Association Genoways, pp. 375-432. New York: Plenum.
Special Volume no. 35B. Oxford: Clarendon. Van Valen, L. (1971). Toward the origin of artiodactyls. Evolution, 25,
Romer, A. S. (1966). Vertebrate Paleontology (3rd ed.). Chicago: University 523-9.
of Chicago Press. (1978). The beginning of the Age of Mammals. Evolutionary Theory, 4,
Rose, K. D. (1981). Composition and species diversity in Paleocene and 45-80.
Eocene mammal assemblages: an empirical study. Journal of Verte- Van Valen, L., & Sloan, R. E. (1966). The extinction of the multituberculates.
brate Paleontology, 1, 367-88. Systematic Zoology, 15, 261-78.
Russell, D. E. (1975). Paleoecology of the Paleocene-Eocene transition in Wing, S. L., & Tiffney, B. H. (1987). The reciprocal interaction of an-
Europe. In Approaches to Primate Paleobiology, ed. F. S. Szalay, pp giosperm evolution and tetrapod herbivory. Review of Palaeobotany
28-61. Basel: Karger. and Palynology, 50, 179-210.
Schoch, R. M. (1986). Systematics, functional morphology and macroevo- Wyss, A. R., Novacek, M. J., & McKenna, M. C. (1987). Amino acid se-
lution of the extinct mammalian order Taeniodonta. Bulletin of the quence versus morphological data and the interordinal relationships
Peabody Museum of Natural History, 42, 1-307. of mammals. Molecular Biology and Evolution, 4, 99-116.
16 Taeniodonta
SPENCER G. LUCAS, ROBERT M. SCHOCH, and THOMAS E. WILLIAMSON

INTRODUCTION

The Taeniodonta is an archaic order of eutherian mammals known

solely from lower Paleocene to middle Eocene strata of North Amer-
ica. The sister group of the Taeniodonta may lie among the Late
Cretaceous "leptictimorphs" (perhaps the "palaeoryctids" or "pan-
tolestids"). The nine currently recognized genera of taeniodonts
have been allocated to two families (Schoch, 1986). The para-
phyletic conoryctids were small- to medium-sized (5-15 kg),
generalized, omnivorous mammals. The monophyletic stylinodon-
tids were medium- to relatively large-sized (10-110 kg) upland
rooters and grubbers that fed predominantly on underground food
resources. Taeniodonts are a relatively rare component of the pale-
ofaunas in which they occur. They reached their maximum species
diversity during the early Paleocene (Torrejonian), but the cono-
ryctids went extinct soon thereafter. The stylinodontids persisted at
low diversity and absolute numbers of individuals until the middle
Eocene (Uintan) when they, too, went extinct. In the middle Eocene
the stylinodontids may have been at a competitive disadvantage with
contemporaneous suoidlike artiodactyls.
Since the first taeniodont was described in 1874 (Marsh, 1874),
there have been four major studies of the order: Wortman (1897), ~l
Matthew (1937), Patterson (1949), and Schoch (1986). This chapter ~
is a summary of Schoch (1986; see also Schoch, 1981, 1982, 1983,
l 984a, 1985) with some modifications based on Williamson and '1-
Lucas (1990) and Lucas and Williamson (1993).

DEFINING FEATURES OF THE

ORDER TAENIODONTA

Taeniodonts are united by the following characteristics: relatively

narrow upper molars, with protocones, protoconules, and metaco-
nules small and placed far lingually, paracones and metacones mod-
erate sized, punctate, and placed far labially with reduced stylar
shelves; pre- and postcingula lacking on upper molars; hypocone Figure 16.1. Restoration oftaenoidonts, by Brian Regal. A. Onychodectes.
absent or developed by a splitting off from the protocone; B. Stylinodon.

260
 Taeniodonta 261

1 TAENIODONTA
Figure 16.2. Interrelationships of the Taeniodonta (from Lucas & Williamson, 1993). Character states corresponding to the node points are as follows:
( 1) Narrow upper molars with protocones, protoconules, and metaconules small and placed far lingually, paracones and metacones moderate sized, punctate,
and placed far labially with reduced stylar shelves; pre- and postcingula lacking on upper molars; hypocone absent or developed by a splitting off from
the protocone; lower molars lack cingulids; trigonids and talonids of all molars subequal in size (length and width); trigonids bear subequal protoconids
and metaconids; trigonids only slightly higher than talonids; molars subequal in size or decrease in size posteriorly; slight crown hypsodonty of the cheek
teeth (characterized by lingual enamel extension on the uppers and labial enamel extension on the lowers) with an increased grinding function. (2) Stylar
shelf and styles on P3-4 absent. (3) Ml-2 narrow. (4) P4 molarized. (5) P3-4 have a transverse loph consisting of two cusps, the paracone and protocone.
(6) STYLINODONTIDAE: Premolars set obliquely to transversely; canines very large; skull and mandible very short and deep; large, laterally compressed
and recurved claws on the manus; large and robust limb bones.

lower molars lack cingulids; trigonids and talonids of all molars morphs (including the palaeoryctids sensu Jato) or perhaps among
subequal in size (length and width); trigonids bear subequal proto- pantolestids that may themselves be closely related to leptictimorphs
conids and metaconids; molars decrease in size posteriorly; and (see Schoch, 1986, Chapter 7). The taeniodonts appear to be nonun-
hypoconulid/talonid not expanded on m3. Taeniodonts can also gulate tokotheres (see McKenna, 1975; Schoch, 1984b), but the
be distinguished by their crown hypsodont cheek teeth. In rela- more detailed affinities of the order are unclear.
tively primitive forms (Onychodectes, Conoryctella, Conoryctes, In the past a special relationship was suggested between the tae-
Huerfanodon, Schochia, and Wortmania), this takes the form of niodonts and the edentates or the taeniodonts and the tillodonts. In
"tooth-base" or "crown" hypsodonty characterized by the labial both cases, such relationships were based on relatively superficial
extension of the enamel on the lower cheek teeth and the lingual similarities between the groups (fossorial adaptations and loss of
extension of the enamel on the upper cheek teeth. In more ad- tooth enamel in both edentates and derived taeniodonts; develop-
vanced taeniodonts (Psittacotherium, Ectoganus, and Stylinodon), ment of large gliriform, but nonhomologous, tusks in tillodonts and
this form of hypsodonty is combined with "root" hypsodonty in taeniodonts) and have since been thoroughly discredited.
which the roots of the cheek teeth fuse and become evergrowing
(crown hypselodonty of Janis and Fortelius, 1988). Taeniodonts are
INFRAORDINAL
also characterized by the possession of a leptictimorph astragalo-
calcaneal morphology. A hypothesis of the phylogenetic relationships of the taeniodonts is
presented in Figure 16.2 (from Lucas and Williamson, 1993). An
evolutionary scenario for the taeniodonts is discussed later.
SYSTEMATICS

INCLUDED NORTH AMERICAN GENERA

SUPRAORDINAL
IN THE ORDER TAENIODONTA
On the basis of evidence from the astragalar-calcaneal complex, the
ear region, and dentition, the sister group of the Taeniodonta may The locality numbers listed for each genus refer to the list of uni-
lie among Procerberus-like or Cimolestes-1ike and allied lepticti- fied localities in Appendix I. Question marks in front of the locality
 Spencer G. Lucas et al. 262

.. : , ·....:
c~-
:-:@ ~ ~~)~ ~(~~
. . . ,, ~"' ~~iJJ
Figure 16.3. Osteology of Onychodectes tisonensis. A. Skeleton (Scale= 10 cm). B. Skull (Scale = 2 cm). C. Dentition (Scale= 1 cm).

(e.g., ?CPI 01) mean the taxon is questionably known from that fused, so the radius could rotate on the ulna. The tail of Onychodectes
locality. The acronyms for museum collections are listed in is long and heavy.
Appendix III. The carpals and tarsals of the conoryctids are unreduced, un-
fused, and alternating. Both forefeet and hindfeet bear five digits,
but the medial and lateral digits (1 and 5) are somewhat reduced.
"CONORYCTIDAE" The astragalar-calcaneal complex of Onychodectes is of the lep-
"Conoryctidae" here refers to a paraphyletic group of primitive tictimorph morphotype characterized by modifications for extreme
taeniodonts. The term is used here for convenience in dis- plantar flexion. The unguals of the manus and pes of conoryctids
cussion. are relatively small, unfissured claws.
Cranially and postcranially, the conoryctids are relatively primi-
tive, generalized mammals (Figure 16.3). The skull of Onychodect- Onychodectes Cope, 1888
es is long and narrow and quite insectivorelike with a long muzzle, Type species: Onychodectes tisonensis Cope, 1888.
long nasal bones, large premaxillae, a low sagittal crest, absence Type specimen: AMNH 3405, right and left maxillae with
of postorbital processes, a long shallow dentary, a moderately large P4-M3, left dentary with m2 and associated right astra-
coronoid process, and a condyle set just above the tooth row. In more galus.
advanced conoryctids (such as Conoryctes), the face is shorter and Characteristics: Small taeniodonts; teeth moderately hyp-
deeper and the mandible is more robust. sodont (relatively less hypsodont than in Conoryctella);
Conoryctid cheek teeth are characterized by crown hypsodonty P4 nonmolariform with a well-developed protocone, para-
with the enamel extending far labially on the lower cheek teeth cone, and incipient metacone, parastyle, stylocone, and
and far lingually on the upper cheek teeth. The cusps of the molars metastyle; P4 metastylocone small to moderately well de-
are relatively low and easily lost with wear, forming broad grind- veloped; lower molar trigonids bear large, subequal, and
ing surfaces. Conoryctid premolars are sectorial teeth separated in sharply punctate protoconids and metaconids with only
the jaw so the upper and lower teeth mesh tightly with each other. slightly smaller, lingually placed paraconids; lower mo-
The canines of Onychodectes are of moderate size, but the canines lar talonids bear high and punctate hypoconids, slightly
increased in relative size during conoryctid evolution. smaller and punctate entoconids, and smaller hypoconulids.
The postcranial skeleton of conoryctids is relatively primitive Average length of m2: 6.3 mm.
with few specializations. The limb bones are moderately long and Included species: 0. tisonensis (= 0. rarus) only (known
robust. The femur bears three trochanters, and the tibia and fibula from localities SB23A, SB23B, CPIB).
are unfused. The humerus approaches what Gregory (1910, p. 249)
termed the "primitive fossorial type," which is broad distally and Conoryctella Gazin, 1939
retains an entepicondylar foramen, whereas the deltoid crest is well Type species: Conoryctella dragonensis Gazin, 1939.
developed, but not flattened, and projects anteriorly. The olecranon Type specimen: USNM 15704, left maxilla with damaged
of the ulna is relatively large and robust. The ulna and radius are not P4-M2 and part of P3 alveolus.
 Taeniodonta 263

Characteristics: Small taeniodonts, larger than Onychodectes Included species: H. torrejonius (known from locality
but smaller than Conoryctes; teeth more hypsodont than SB23G); H. polecatensis (locality CP13B).
Onychodectes but less hypsodont than Conoryctes; P4 non- A questionable Huerfanodon sp. is also known from
molariform with well-developed protocone and paracone, locality NPl9C.
but only an incipient metacone; stylocone and parastyle
absent on P4 or only slightly developed (in contrast to Schochia Lucas and Williamson, 1993
Onychodectes and Conoryctes in which they usually are Type species: Schochia sullivani Lucas and Williamson, 1993.
moderately well developed); symphysis of lower jaw un- Type specimen: NMMNHP-900, right P3-M2.
fused; lower canines relatively large, heavily invested with Characteristics: Schochia is a taeniodont similar in size to
enamel, triangular in cross section, transversely compress- Wortmania but distinguished by its very transverse molars
ed and tending toward the rootless condition seen in with large parastyles; Schochia differs from Onychodectes,
advanced taeniodonts; p4 submolariform with a single pro- Conoryctella, Conoryctes, and Huerfanodon in having bul-
toconid anteriorly and a moderately developed talonid pos- bous P3-4 paracones and protocones of subequal size and
teriorly; lower molar paraconids relatively large. no stylar shelves/styles on these teeth. Schochia is smaller
Average length ofm2: 7.6 mm. than and lacks the extreme hypsodonty and bunodont,
Included species: C. dragonensis (known from locality multicuspid Ml-2 of Psittacotherium, Ectoganus, and Sty-
CPIC); C. pattersoni (locality CPlC). linodon.
Average length of m2: unknown. Length of M2: 6.1 mm.
Conoryctes Cope, 1881 Included species: S. sullivani only, known from locality
Type species: Conoryctes comma Cope, 1881. SB23A only.
Type specimen: AMNH 3395, left dentary fragment with
p4-m2, alveolus for p2, roots of p3 and m3, isolated lower
STYLINODONTIDA E
right canine.
Characteristics: Medium-sized taeniodonts, about the size of Characteristics: Taeniodonts with lower premolars set obliquely to
Huerfanodon; teeth relatively hypsodont; canines lack in- transversely in the mandible; canines very large; skull and
ternal grooves; P3 nonmolariform, bears a simple large mandible very short and deep; large, laterally compressed and re-
paracone and a minute to small metacone and lingual cin- curved claws on the manus; and large and robust limb bones.
gulum; P4 molariform with a large protocone, smaller Stylinodontids (Figure 16.4) are characterized by large heads with
paracone, still smaller metacone, and small parastyle, sty- short, deep faces and mandibles, a shortened and widened snout,
locone, metastyle, and metastylocone; mesostyle absent on large mastoid processes, a prominent sagittal crest, and a wide, high,
P4; M1-3 bear large protocones and smaller, conical, labi- triangular-shaped occiput. The mandible is deep and extremely ro-
ally placed paracones and metacones; Ml-3 mesostyles bust with a well-fused mandibular symphysis, a large, heavy coro-
vary from absent to moderately well developed; p4 sub- noid process, a rugose, slightly inflected jaw angle (primarily for
molariform, trigonid bears a single large protoconid and attachment of the M. pterygoideus), and a large internal shelf or
simple talonid; lower molars with small paraconids. pit behind the symphysis for the origin of the tongue musculature.
Average length of m2: 9.5 mm. The main emphasis of the stylinodontid dentition is on the incisors,
Included species: C. comma only (known from localities canines, and anterior premolars.
SB23G, SB23H, NP19C). The canines and upper incisors became greatly enlarged, rootless,
and subgliriform with enamel limited to the anterior part of the teeth,
Huerfanodon Schoch and Lucas, 1981 the posterior part being enamel free. Thus the canines are chisel-like
Type species: Huerfanodon torrejonius Schoch and Lucas, teeth with an anterior cutting surface and a posterior crushing and
1981. grinding surface. The anterior premolars complement the canines
Type specimen: USNM 15412, partial skull and right dentary. as cutting blades, whereas the posterior premolars become rootless
Characteristics: Medium-sized taeniodonts, approximately and evergrowing pegs for grinding, with bands of enamel limited to
the size of Conoryctes: P3 submolariform with a large their lingual and labial faces. The crowns of the posterior premolars
paracone and small, but distinct metacone, protocone, and and molars are transversely bilophodont, but with moderate wear
hypocone; Ml-2 with well-developed and protruding mes- the cusps were obliterated, and only dentine pegs with incomplete
ostyles connected to the metacones by slightly cuspidate bands of enamel around their perimeters remained. This evidently
premetacristae, distinct and cuspidate parastyles, ectocin- was the condition in which the teeth were used for the majority of
gula, and metastyles, and broad trigon basins with shallow an individual animal's life.
valleys and cuspidate lingual margins bearing moderate- The stylinodontid skeleton (Figure 16.4) is a modified version of
sized protocones, paraconules, and metaconules; lower ca- the generalized, primitive skeleton of Onychodectes. Major modifi-
nine with internal groove; lower molars with prominent cations include increased size, plus the development of an extremely
paraconids and paracristids. squat, robust skeleton and powerful forelimbs with large, laterally
Average length of m2: 9.9 mm. compressed and recurved claws on the manus. The neck is short
 Spencer G. Lucas et al. 264

)
0
Figure 16.4. Osteology of Stylinodon mirus. A. Skeleton (Scale= 10 cm). B. Skull (Scale= 2 cm). C. Dentition (Scale= 1 cm).

and stout. The humerus is large and robust with a prominent del- Wortmania Hay, 1899 (synonym: Hemiganus)
topectoral crest, supinator ridge and teres eminence, and large epi- Type species: Wortmania otariidens (Cope, 1885) ( = Hemi-
condyles. The radius and ulna are short and robust, and the radius ganus otariidens).
could rotate around the ulna. The olecranon is large and robust sug- Type specimen: AMNH 3394, partial skull, lower mandible,
gesting the presence of powerful ftexors for retraction of the fore- and partial skeleton.
limb. The femur, tibia, and fibula are all moderately stout and robust Characteristics: Medium-sized taeniodonts (smaller than Ps-
and the tibia and fibula are unfused. The tail is long and heavy. ittacotherium but larger than Conoryctes) with relatively
The stylinodontid manus bears large, laterally compressed and low-crowned teeth; all teeth relatively shallow rooted; up-
recurved claws on short, robust metacarpals and phalanges. The me- per and lower canines enlarged and oval in cross section;
dial and lateral digits (l and 5) are greatly reduced on the man us. The skull and mandible robust; lower premolars bear a single
articular surfaces of the metacarpals and phalanges allowed a high anteroexternal conid and a well-developed posterointernal
degree of extension and fiexion, but minimized medial-lateral move- cingulid ("talonid"); lower molars with both trigonids and
ments. The carpus is alternating in more primitive stylinodontids talonids compressed anteroposteriorly and expanded trans-
(such as Psittacotherium), whereas in more advanced forms (e.g., versely, with wear forming two blunt, transverse lophs;
Stylinodon) the carpus is more nearly serial. The magnum is greatly trigonids wider than talonids.
enlarged in Stylinodon. The large third metacarpal rests fully against Average length of m2: 8.2 mm.
the distal surface of the magnum proximally; the proximal surface Included species: W. otariidens only (known from localities
of the magnum in turn rests against the distal surface of the lunar. A SB23A, SB23B, ?CPlB, ?CPIC).
large component of the stress placed on the man us may have passed
directly through these bones. Psittacotherium Cope, 1882
The stylinodontid pes bears five well-developed digits arranged Type species: Psittacotherium multifragum Cope, 1882.
in an arc. The tarsus is serial, and the metatarsals are short and Type specimen: AMNH 3413, mandible with left i3, C, right
thick. Most of the stress placed on the pes passed through the serial p2, ml, m2, roots of right i3, c, and alveoli for left m2-3
tarsus, the short metatarsals and associated sesamoids, and onto and right pl, m3.
the substrate. The short digits bear large, broad, and only slightly Characteristics: Medium-sized taeniodonts with greatly en-
recurved claws. This arrangement provided a firm support for the larged, rooted, subgliriform canines with enamel limited
stylinodontid hindfoot. to the anterior face of the tooth and with both crown and
 Taeniodonta 265

root greatly elongated; upper canine with external groove; moderately worn teeth without enamel entirely around
I3 enlarged, deeply rooted, and caniniform; lower incisors their perimeter; posterior premolars and molars with only
of moderate size and rooted; cheek teeth moderately hyp- thin strips of enamel labially and lingually after moderate
sodont with relatively shallow roots (as compared to Ec- wear.
toganus); posterior cheek teeth double rooted or single Average length of m2: 11.4 mm (S. mirus, m2 unknown for
rooted with traces of the fused roots; upper premolars S. inexplicatus, but the M2 of S. inexplicatus has a length
suboval in cross section and transversely elongated bear- of 6.5 mm); note that S. mirus was a larger beast than E.
ing a large protocone and paracone connected by low to gliriformis, although the length of m2 is shorter in S. mirus.
incipient transverse anterior and posterior crests; upper Included species: S. mirus (= S. cylindrifer) (known from
molars tritubercular, suboval in cross section and trans- localities SB22B, SB22C, CP6B, CP25H, CP25J, CP27B,
versely elongated bearing small paracones and metacones CP27D, CP34D, CP38B); S. inexplicatus (locality CP38B).
placed far labially and anteriorly, and large protocones Stylinodon sp. is also known from localities SB43A,
lingually; minutely cuspidate postmetaconule wings well ?CP64C.
developed (especially on M2-3) and extending postero-
labially to a point posterior and just lingual of the meta-
cone; pl reduced and single rooted; p2 single rooted and
INDETERMINA TE TAENIODONTS
placed transversely in the lower jaw, bearing a large exter- Fragmentary remains, possibly assignable to taeniodonts, have been
nal (labial) conid and a smaller internal conid; p4 submo- identified at localities CP14B, CP25G.
lariform with a transversely widened trigonid and small,
lingually placed talonid; ml-3 decrease in size posteriorly;
ml-3 moderately bilophodont with broader trigonids than BIOLOGY AND EVOLUTIONARY PATTERNS
talonids, talonids placed lingually.
Average length of m2: 9.2 mm. The skull and body lengths of the conoryctids range from about
Included species: P. multifragum (= P. aspasiae) only (known 58-68 cm, and their weights probably ranged from about 5-15 kg.
from localities SB23D, SB23DD, SB23E, SB23F, SB23G, In terms of its general morphology, Onychodectes (Figure 16.IA),
SB23H, SB39B, SB39C, SB39D. CP13B, CP14A, NPl 9C, the best known conoryctid, is quite similar to, although larger and
NP19IIA). more robust, than the common opossum Didelphis marsupialis.
Onychodectes was a noncursorial, plantigrade mammal that may
Ectoganus Cope, 1874 (synonym: L<impadophorus) also have been reasonably adept at climbing. Based on their denti-
Type species: Ectoganus gliriformis Cope, 1874. tions, the conoryctids may have been generalized omnivores. Cono-
Type specimen: USNM 1137, upper and lower deciduous and ryctids may have eaten a fair-sized component of insect, mam-
permanent tooth fragments. mal, and bird matter, but taking the increased grinding function
Characteristics: Medium to large-sized taeniodonts; canines of their molars into consideration, they may also have eaten a large
enlarged, rootless, and compressed posteriorly with enam- amount of plant matter. In this connection, their relatively heavy
el limited to the anterolabial aspect; cheek teeth moderately forelimbs and claws may have allowed them to harvest subsurface
to extremely hypsodont; ml-3 transversely bilophodont; food items.
p3-4 submolariform to molariform with talonids lingually The skull and body lengths of the stylinodontids range from
placed and narrower than the trigonids; ml-3 transversely about 65-130 cm, and their weights probably ranged from about
bilophodont with subequal trigonids and talonids. 10-110 kg. Stylinodontids (Figure 16.IB) were characterized by
Average length ofm2: 14.0 mm (E. gliriformis), 11.7 mm (E. short, wide skulls with massive mandibles and large canines. The
copei). stylinodontid body was solid and heavy, the limbs were stout, robust,
Included species: E. gliriformis (= E. simplex, E. lobdelli) and powerful, and the forefeet bore large, laterally compressed and
(known from localities GC20, SB24, CPI3H, CPI 7 A, recurved claws. In terms of modern analogues, an advanced stylin-
CP20A, CP20B, CP20C, CP26C, CP28B, CP62B, odontid (such as Ectoganus or Stylinodon) might be thought of as
CP64B); E. copei (localities CP20A, CP26C, CP31A). an aardvark with the head of a pig.
Ectoganus sp. is also known from locality SB25A. Stylinodontids were probably active diggers, rooters, and grub-
bers, feeding on tubers and other underground food resources. There
Stylinodon Marsh, 1874 also seems to be no reason why the stylinodontids might not have uti-
Type species: Stylinodon mirus Marsh, 1874. lized their powerful masticatory apparatus to take advantage of car-
Type specimen: YPM 11095, right and left dentary fragments rion and other animal nourishment if they happened to stray across
with partial right p3-ml, left p3 and alveoli for right p2, it. Stylinodontids may have been primarily open-country, upland
m2-3 and left p2, p4-ml; labial enamel fragment of left pl. forms, perhaps one of the first upland radiations of mammals.
Characteristics: Taeniodonts with the dental formula I2/1, The earliest known (Puercan, earliest Paleocene) taeniodonts are
C 1/1, P4/4, M3/3; all of the teeth evergrowing and rootless; already clearly distinguishable as either conoryctids (Onychodectes)
 Spencer G. Lucas et al. 266

w
z L L.BI.
1.8

w E.BI
2.4
0 4.5
0 E LstHp.
:i L..Hp.
5.2
6.0
a. LE.Hp.
7.0
E.E.Hp.
L 8.8
L.CI.
9.5

E.CI.
11.0
w
z LL.Ba.
w E.L.Ba.
12.5
0 M
0 14.0
:g E.Ba.

5.8
L.Hm.

17.5
E.Hm.
18.8
E L.L.Ar.
19.2

E.L.Ar.
23.0

w L.E.Ar.
z
w L
0 27.7
0
CJ E.E.Ar.

:i 29.4
0
WI.
E 31.9
o•.
33.4
L.Ch.
34.5
L M.Ch.

E.Ch.
35.5
CONORYCTIDAE STYLINODONTIDAE
37.1

Ou

39.5

L.Un.

w 41.3

z
w E.Un.

0 45.9
0
w M
L.Br.
46.7

M.Br.
c::
49.0 -0
E.Br. "8
-
50.4 -C::
L.Wa.
~
Cl)
53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0
55.5
-
Ck.3 55.6
w Ck.2
z
w L
Ck.1
Tl.6
Tl.5
....
55.8
56.0

57.0 Ill

.gc::
Tl.4
0 Tl.3
57.3
.!! -111-
-::::: -
0 56.5 u Ill
u -- o,=
.!I
Tl.2
w Tl.1 59.2
~ -.!! - .!!!
...I 60.9 -u-
<I:
a.
To.3
To.2
61.3 0
ti - -~E
-o ...0
;>.. - ~ - ti0
~::i
62.5
To.1
.c:
~,-
~
E Pu.3 -C:: -
c:: - -0 -
Pu.2
Pu.1
Pu.o
64.5
65.0
65.1
0 - -0- :i::: ~·=
Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar.= Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. = Wasatchlan,
Ck. = Clarkforkian, Ti.= Tiffanian, To.= Torrejonian, Pu.= Puercan.

Figure 16.5. Temporal ranges of taeniodont genera.

or stylinodontids (Wortmania). Although no known form of Ony- Predominant evolutionary trends that are discernible among the
chodectes can be considered a true ancestor of later or more ad- conoryctids include increase in overall body size, increase in the rel-
vanced taeniodonts, in some ways Onychodectes may be thought ative size of the head with a shortening and deepening of the face and
of as representative of the general morphology of the ancestral tae- mandible, relative enlargement of the canines, slight increase in the
niodont. degree of crown hypsodonty seen in the cheek teeth, and a tendency
 Taeniodonta 267

toward the molarization of the posterior premolars and a reduction Gazin, C. L. ( 1939). A further contribution to the Dragon Paleocene fauna
of the anterior premolars. After flourishing during the Puercan and of central Utah. Journal of the Washington Academy of Sciences, 29,
early Torrejonian, they were extinct by the end of the Torrejonian 273-86.
Gregory, W. K. (1910). The orders of mammals. Bulletin of the American
(late early Paleocene). In an ecological sense conoryctids may have
Museum of Natural History, 28, 1-524.
tended toward becoming carnivores, herbivores, and rooters and Hay, 0. P. (1899). On the names of North American fossil vertebrates.
grubbers; not pursuing any of these strategies particularly well, we Science, 9, 593-94.
speculate that they were outcompeted on these three major fronts Janis, C. M., & Fortelius, M. (1988). On the means whereby mam-
by creodonts, condylarths, and the stylinodontids, respectively. mals achieved increased functional durability of their dentitions,
with special reference to limiting factors. Biological Reviews, 63,
Among the stylinodontids the major evolutionary trends included 197-230.
an increase in overall body size; increasing hypsodonty of the teeth; Lucas, S. G., & Williamson, T. E. (1993). A new taeniodont from the Pale-
the development of crown hypselodonty; chewing and grinding em- ocene of the San Juan Basin, New Mexico, and the phylogeny of the
phasis shifted to the anterior dentition; bilophodonty of the molars Taeniodonta. Journal of Mammalogy, 74, 175-9.
elaborated; shortening and deepening of the face and jaw; enlarge- Marsh, 0. C. (1874). Notice of new Tertiary mammals. American Journal
of Science, 7, 531-4.
ment of the unguals in general and the development and enlargement
Matthew, W. D. (1937). Paleocene faunas of the San Juan Basin, New
of large, laterally compressed and recurved claws on the manus; Mexico. Transactions of the American Philosophical Society, new
modifications of the carpal and tarsal series toward a more serial series, 30, 1-510.
condition; and increase in the relative robustness of the limb bones. McKenna, M. C. (1975). Toward a phylogenetic classification of the Mam-
The stylinodontids evolved relatively quickly (by the Torrejonian) malia. In Phylogeny of the Primates, eds. W. P. Luckett & F. S. Szalay,
pp. 21-46. New York: Plenum.
all of their major distinguishing characteristics; later forms merely
Patterson, B. ( 1949). Rates of evolution in taeniodonts. In Genetics, Paleon-
specialized along the same lines. After experiencing their heyday tology, and Evolution, eds. G. L. Jepsen, G. G. Simpson, & E. Mayr,
during the Torrejonian to Wasatchian (early Eocene), stylinodontids pp. 243-78. Princeton: Princeton University Press.
persisted at low diversity and absolute numbers of individuals until Schoch, R. M. (1981). Taxonomy and biostratigraphy of the early Tertiary
their extinction during the Uintan (middle Eocene). One can specu- Taeniodonta (Mammalia: Eutheria). Geological Society of America
Bulletin, 92, 933-41.
late that the last taeniodonts were driven to extinction by newly ap-
(1982). Phylogeny, classification and paleobiology of the Taeniodonta
pearing suoidlike artiodactyls. The middle Eocene taeniodonts may (Mammalia: Eutheria). Proceedings of the Third North American
have been at a competitive disadvantage with the larger-brained (and Paleontological Convention, 2, 465-70.
in some cases larger-bodied) contemporaneous artiodactyls, such as (1983). An endocranial cast of Ectoganus, an early Tertiary taeniodont.
achaenodonts, which were superficially similar to taeniodonts and Journal ofMammalogy, 64, 180-3.
(1984a). Systematics, functional morphology and macroevolution of the
whose modem suid analogues dig, root, and grub as stylinodontids
extinct mammalian order Taeniodonta. Dissertation Abstracts Inter-
probably did. national, 44, 2692-3.
(l 984b ). Classification of the vertebrates. In Vertebrate Paleontology,
ed. R. M. Schoch, pp. 4-11. New York: Van Nostrand Reinhold.
(1985). Preliminary description of a new Late Paleocene land-mammal
REFERENCES fauna from South Carolina, U.S.A. Postilla, Peabody Museum, Yale
University, no. 196, 1-13.
Cope, E. D. (1874). Report upon vertebrate fossils discovered in New (1986). Systematics, functional morphology and macroevolution of the
Mexico, with description of new species. Geographical Exploration extinct mammalian order Taeniodonta. Yale University Peabody Mu-
and Survey West of the JOOth Meridian, Appendix FF of the Annual seum of Natural History Bulletin, 42, 1-307.
Report of the Chief of Engineers (U.S. Army)for 1874, pp. 589-606 Schoch, R. M., & Lucas, S. G. (1981). New conoryctines (Mammalia; Tae-
(Separatum, pp. 1-18). niodonta) from the middle Paleocene (Torrejonian) of western North
(1881). Mammalia of the lowest Eocene. American Naturalist, 15, America. Journal of Mammalogy, 62, 683-91.
829-31. Williamson, T. E., & Lucas, S. G. (1990). A new taeniodont from the early
(1882). A new genus of Tillodonta. American Naturalist, 16, 156-7. Paleocene (Puercan) of the San Juan Basin, New Mexico, and the
(1885). The mammalian genus Hemiganus. American Naturalist, 19, paraphyly of the Conoryctida. Journal of Vertebrate Paleontology,
492-3 . 10 (suppl. to no. 3), p. 49A.
(1888). Synopsis of the vertebrate fauna of the Puerco epoch. Transac- Wortman, J. L. (1897). The Ganodonta and their relationship to the Edentata .
tions of the American Philosophical Society, new series, 16, 298-361. Bulletin of the American Museum of Natural History, 9, 59-110.
17 Tillodontia
SPENCER G. LUCAS and ROBERT M. SCHOCH

INTRODUCTION

The Tillodontia is an extinct order of eutherian mammals known

from the late Paleocene to middle Eocene of North America, early
Paleocene to late Eocene of China, middle Eocene of Pakistan, and
the early Eocene of Europe. The tillodonts produced no known de-
scendants after the Eocene and were most closely related to the
extinct order Pantodonta. Tillodonts were medium- to large-sized
mammals, probably feeding primarily on roots and tubers in warm
temperate to subtropical habitats. In some respects the adaptive mor-
phology of the tillodonts converged on that of the taeniodonts, and
members of these two orders may have competed for similar re-
sources (Figure 17 .1 ). Tillodonts were relatively common and show
maximum species diversity in earliest Eocene faunas of North Amer- Figure 17.1. Restoration of Trogosus, by Brian Regal.
ica, but became progressively rarer throughout the later early Eocene
and early middle Eocene until their last known occurrence in the cingulids, bear distinct metastylids, and have labially extended
middle Bridgerian. enamel (crown hypsodonty); complete postcranial skeleton (no el-
In this chapter we deal primarily with the North American forms, ements lost or fused); digits bear moderately large and recurved
although in discussing the phylogeny and evolution of the tillodonts claws (see Figure 17.2).
we also address the Asian and European genera. The most com- For a thorough description of the known osteology of the tillo-
prehensive monograph on the Tillodontia is by Gazin (1953); other donts, see Gazin (1953).
significant articles on North American tillodonts are Rose (1972),
Gingerich and Gunnell (1979), and Stucky and Krishtalka (1983).
The taxonomy of North American tillodonts is reviewed in these SYSTEMATICS
four papers.
SUPRAORDINAL
DEFINING FEATURES OF THE Tillodonts were long associated with ungulates as supposed "condy-
ORDER TILLODONTIA larth" derivatives. For example, Gazin (1953) and Van Valen (1963)
identified arctocyonid "condylarths" as the ancestors of tillodonts.
Small to moderately large eutherian mammals (skull length ranges More recently, tillodonts have been considered closely allied to
from 5 to 37 cm) characterized by elongate rostrum and deep mandi- pantodonts (Chow and Wang, 1979). We tentatively support this
bular symphysis; anteroposteriorly short basicranial region; 12/i2 hypothesis by noting that the dilambdodont dentitions of tillodonts
large in most genera; Pl/pl small or absent; P4/p4 molariform; mo- and pantodonts unite them as sister taxa with respect to a dental out-
lars dilambdodont; upper molars rectangular (subequal in length taxon such as Didelphodus. But, unlike Chow and Wang (1979),
labially and lingually) with labially placed protocones and dis- we recognize Tillodontia and Pantodonta as separate, albeit closely
tinct hypoconal crests (flares) posterolingually; lower molars lack related, orders of mammals.

268
 Tillodontia 269

B c D ties between the two nominal genera, we place K. shantunensis in

the genus Trogosus as the new combination Trogosus shantunen-
~1 I
sis (Chow, 1963) and regard Kuanchuanius as a junior subjective
synonym of Trogosus.
1 Meiostylodon and Plesiesthonyx are very poorly known, and
I
j
therefore the systematic positions of these genera are uncertain.
Meiostylodon may be primitive relative to Esthonyx in bearing rela-
tively transverse (broad and short) upper molars with deep ectoflexi.
It is not known if Meiostylodon retained Pl/pl.
Plesiesthonyx munieri is based on only three teeth. Lemoine
(1889, 1891) recognized that these teeth are similar to those of Es-
thonyx yet they differ in several features. Gazin (1953), Gingerich
and Gunnell (1979), and Stucky and Krishtalka (1983) considered
P. munieri to be merely a very distinct species of Esthonyx; however,
in doing so they greatly expanded the concept of Esthonyx. Until
P. munieri is better known, we prefer to retain the genus Plesi-
esthonyx for this unique species, distinguished by the following
characters: cheek teeth relatively inflated with bulbous cusps (au-
tapomorphy of the genus?); upper molars with well-developed an-
terior cingula (primitive relative to Esthonyx?); p4 with slightly
crenulated enamel; p4 paraconid positioned well forward of pro-
toconid; m3 paraconid closely appressed to metaconid (autapo-
morphy of the genus?); and m3 hypoconulid lobe short and broad
Figure 17.2. Selected parts of the skeleton of Trogosus: A. Left scapula. (primitive?).
B. Left humerus. C. Left radius. D. Left femur. E. Left tibia-fibula.
F. Left manus. G. Skull plus lower jaw (after Gazin). Scale bar= 2 cm.

INCLUDED NORTH AMERICAN GENERA

IN THE ORDER TILLODONTIA
INFRAORDINAL
A hypothesis of the phylogenetic relationships of the tillodont gen- The locality numbers listed for each genus refer to the list of unified
era is shown in Figure 17.3. In the cladogram we have also included localities in Appendix I. The acronyms for museum collections are
taxa from outside of North America: Lofochaius Zhou et al., 1973 listed in Appendix III.
from the early Paleocene ofGuangdong, China; Basalina Dehm and The locality numbers may be listed in a couple of alternative ways.
Oettingen-Speilberg, 1958 (see Lucas and Schoch, 1981) from the Parentheses around the locality (e.g., [CP!Ol]) mean the taxon in
middle Eocene of Pakistan; M eiostylodon Wang, 197 5 from the early question at that locality is cited as an "aff." or "cf." the taxon in
Paleocene of Hunan, China; Plesiesthonyx Lemoine, 1891 from the question. Parentheses are usually used for individual species, thus
early Eocene of France; and Adapidium Young, 1937 from the ?late implying the genus is firmly known from the locality, but the actual
Eocene of Shanxi, China. species identification may be questionable. Question marks in front
We consider Kuanchuanius shantunensis Chow, 1963 from the of the locality (e.g., ?CPlOl) mean the taxon is questionably known
middle Eocene of Shan dong, China, to be a small species of Trogosus from that locality, thus implying some doubt that the taxon is actually
(Schoch and Lucas, 1982) for the following reasons. Kuanchuanius present at that locality, either at the genus or the species level.
shantunensis, based on a right dentary bearing m2-3, has a lower
dental formula of i3/cl/p3/m3, and i2 is greatly enlarged and root-
ESTHONYCHIDAE
less, as in North American Trogosus. Kuanchuanius, like Trogosus,
has a strongly fused mandibular symphysis that extends to under the Characteristics: Medium to large tillodonts (skull length
trigonid of ml. The anterior cheek teeth of Kuanchuanius are closely ranges from 10-37 cm); i2 relatively larger than in Basalina and
spaced and p2 is single rooted, as in North American Trogosus. The rootless in derived forms; Pl absent; Ml-3 with narrow stylar
crown morphology of the preserved m2-3 of Kuanchuanius also shelves; Ml-2 parastyles and metastyles less prominent than in
is identical to that of specimens of North American Trogosus; in Lofochaius (i.e., ectoflexus shallow); Ml-3 posterior cingula large
both, the cristids obliquae join the protocristids just labial of the with strong hypoconal flares; Ml-3 trigon basins complex and
metastylids. In Kuanchuanius there is a small entoconulid just an- conules large and distinct; I2/i2 large; I3 small or absent; pl ab-
terior to the entoconid on the lingual edge of the m2 talonid; this sent; mandibular symphysis extends under p3 or ml.
cuspid also is present on unworn lower molars of North American We consider both Lofochaius and Basalina to each be represen-
Trogosus (e.g., AMNH 17500, a partial left m2 of Trogosus sp. from tative of distinct families (left here formally unnamed), known at
the Huerfano Formation, Colorado). Based on these close similari- present only from single genera, distinguished as follows.
 Spencer G. Lucas and Robert M. Schoch 270

~ (:-
~ § ~
§ ~
r.;
.;:,
...."$ ~
0
"' 8
·~
~ g>
r.; ~ ~ "9
... .Cl>
~
Ill 'ti ~
.;:, ~ ~ ~
~ ~ ~
-z ~ ct 7
~
0
"'
Ill
Q)

" 3 8

Figure 17.3. Interrelationships of tillodonts. Character states corresponding to the node points include (I) Far labially placed protocones on the upper
molars; molariform P4/p4; rectangular upper molars (subequal in length labially and lingually); reduction of P2-3/p2-3; labial extension of enamel (crown
hypsodonty) on the lower cheek teeth; distinct metastylids on ml-3; elongated rostrum; deep mandibular symphysis; and short basicranium. (2) Large i2; i3
and pl reduced. (3) i3 lost; p2 single rooted; mandibular symphysis under p4. (4) ESTHONYCHIDAE: Further enlargement of 12/i2; Pl/pl lost; reduced
stylar shelves on Ml-3; development of large posterior cingula with distinct hypoconal flares on Ml-3. (5) TROGOSINAE: Further enlargement ofl2/i2;
further reduction of ii, i3, cl, and p2; reduced labial cingula on P4-M3; talonids reduced on p3-4; cristids obliquae of ml-3 join metalophids just labial of
metastylids. (6) Mesostyles present on Ml-3. (7) Twinned metastylids on ml-3. (8) Greatly enlarged and rootless 12/i2; greatly reduced stylar shelves on
P4-M3; well-developed enamel loop (cf. Gazin, 1953, p. 66) in trigon basins of M!-3; enlarged hypoconal flare (crest) of Ml-3; reduced anterior cingula
on Ml-3; enlarged parastyles on M13; deepened (under ml) and thick, well-fused mandibular symphysis; greatly enlarged body size. (9) Loss of ii, i3;
great elongation of the rostrum; M3 reduced in width; cheek teeth relatively brachydont; fold between external styles ofMl-3 relatively compressed.

Lofochaius family: Small tillodonts (skull length approximately Type specimen: USNM 1103, left dentary fragment with p3,
5 cm); Pl small; Ml-3 relatively broad and short with relatively ml-3, and two associated incisors.
broad stylar shelves and distinct labial cingula; M 1-2 parastyles and Characteristics: Medium- to large-sized esthonychines (den-
metastyles moderately prominent (i.e., ectoflexus deep); tally larger than Meiostylodon) with upper molars narrower
Ml-3 posterior cingula small with weak hypoconal flare; Ml-3 than those of Meiostylodon and with smaller parastyles
trigon basins simple and conules small, but distinct. Basalina family: and metastyles (i.e., relatively shallow ectoflexi); anterior
Small tillodonts (inferred skull length about 10-20 cm); i2 large cingula of upper molars not as well developed as in Ple-
and rooted; i3 absent; cl relatively large; pl very small; mandibular siesthonyx; cheek tooth cusps less bulbous than those of
symphysis extends under p4 trigonid. Plesiesthonyx; p4 without crenulated enamel like those of
Plesiesthonyx; p4 paraconid closer to protoconid and m3
paraconid farther from metaconid than in Plesiesthonyx;
ESTHONYCHINAE
m3 hypoconulid lobe long and narrow relative to Plesi-
Characteristics: Small to medium-sized tillodonts (skull length esthonyx.
ranges from 10-20 cm); I3/i3 small; Cl/cl, Il/il, and P2 large Average length of m2: 7.3-11.0 mm.
relative to Trogosinae; p3-4 with relatively broad talonids; cristid Included species (see Gingerich and Gunnell, 1979; Stucky
obliqua-protocristid junction on ml-3 positioned relatively labial; and Krishtalka, 1983): E. bisulcatus (= E. burmeisterii,
M 1-3 labial cingula larger than those of Trogosinae. E. acer) (known from localities SB24, CP20A, CP20B,
CP20C,CP23A,CP25B, CP25C,CP25D,CP27B, CP27C,
Esthonyx Cope, 1874 CP27D, CP27E, CP28A, CP64A, CP64B); E. xenicus (lo-
Type species: Esthonyx bisulcatus Cope, 1874. calities ?CP14E, CPl 7B, CP18B); E. ancylion (localities
 Tillodontia 271

CPI 7B, CP18B); E. grangeri (= E. latidens) (localities Tillodon with moderate diastema; i2-p2 closely spaced;
CP18B, CP20A); E. gunnelli (=Azygonyx gunnelli) (local- upper cheek teeth increase in width posteriorly; upper
ity CP19); E. spatularis (localities CP19, CP20A, CP63); molars lack mesostyles, with parastyles well developed
E. acutidens (localities SB22A, SB22B, SB22C, CP20A, relative to metastyles and well-developed enamel flexure
CP20C, CP20D, [CP20E], CP25G, CP25H, [CP64C]). in trigon basins; metastylids of lower molars medium to
Esthonyx sp. is also known from localities CC50, CPl 4E, large; mandibular symphysis extends to under ml-2.
CP18B, CP20B, CP20E, CP25E, CP25F, CP26B, CP26C, Average length ofm2: 17.5-26.3 mm.
CP62B, CP62C. Included species (see Gazin, 1953; Robinson, 1966; Stucky
and Krishtalka, 1983): T. castoridens (= T. hyracoides)
Comments: Gingerich and Gunnell (1979) most recently re-
(known from localities CP32B, CP34C); T. latidens (local-
vised the genus Esthonyx. These authors distinguished two
ities SB22C, [CP25J], CP31E, CP34A, CP34C, NP12); T.
putative lineages within the genus: E. xenicus-E. ancylion-
grangeri (= T. hillsii) (locality SB22C); T. vetulus (nomen
E. grangeri lack symphysial fusion and bear a double-
dubium) (locality CP34C); T. minor (nomen dubium) (lo-
rooted p2; E. spatularius-E. bisulcatus-E. acutidens have
cality CP34C).
a fused mandibular symphysis and bear a single-rooted p2.
Trogosus sp. is also known from localities (CC2), CP5A,
Gingerich (1989) erected the new genusAzygonyx (type
CP20F, CP27E, CP31B, CP31C, CP31D, NP14.
species A. gunnelli) and included all Clarkforkian species
of Esthonyx in this genus. However, we believe the only
difference between Esthonyx and Azygonyx, the unfused Tillodon Gazin, 1953 (synonym: Tillotherium)
manibular symphysis in Azygonyx, is best regarded as a Type species: Tillodonfodiens (Marsh, 1875).
feature variable among Esthonyx, not as a feature distinc- Type specimen: YPM 11087, skull, lower jaws, and some
tive of a new genus. fragmentary remains of other portions of the skeleton.
Characteristics: Large trogosines; similar to Trogosus, but
with the following differences: skull relatively longer; ii, i3
TROGOSINAE absent; moderate anterior and posterior diastema separate
Characteristics: Small to large tillodonts (skull length ranges from Cl from I3 and P2; cl and p2 small and well spaced;
10-37 cm); ii and i3 very small or absent; cl and p2 small; p3- upper and lower cheek teeth more brachydont than those
4 talonids small; cristid obliqua-protocristid junction on ml-3 just of Trogosus; folds between external styles of upper cheek
labial of metastylid; Ml-3 labial cingula small; mandibular symph- teeth relatively compressed; M2 widest tooth of the upper
ysis strongly fused and extends to under p3-m2. cheek tooth series.
Average length ofm2: 20.0-21.8 mm.
Megalesthonyx Rose, 1972 Included species: T.fodiens only, known from locality CP34C
Type species: Megalesthonyx hopsoni Rose, 1972. only.
Type specimen: YPM 18767, left dentary with il-2, cl, p2-
Anchippodus Leidy, 1868, nomen dubium
m3, alveolus for i3, symphysial region with right il-2 and
Type species: Anchippodus riparius Leidy, 1868.
alveoli for right i3 and cl; associated left I2 and various
Type specimen: ANSP 10338, left m2?
bones of the forefeet.
Characteristics: ANSP 10338 is a left lower molar that may
Characteristics: Medium-sized trogosines, intermediate in
represent either Trogosus or Tillodon.
size between Adapidium and Trogosus; i2 deeply rooted
Length of m2?: 20.9 mm.
(but not rootless); il, i3, cl, p2 small and well separated in
Included species: A. riparius only, known from locality EMl
jaw; p2 double rooted; upper molars with well-developed
only.
mesostyles, beads of enamel in trigon basins and mod-
erately developed anterolingual cingula; metastylids of
lower molars small; mandibular symphysis extends to BIOLOGY AND EVOLUTIONARY PATTERNS
under p3.
Length of known m2: 16.2 mm. Major trends in tillodont evolution include increase in overall body
Included species: M. hopsoni only (known from localities size, enlargement and elongation of the snout, deepening and fusion
SB22B, CP20D). of the mandibular symphysis, increase in relative size of I2/i2 (the
tusklike incisors), reduction or loss of the remaining incisors, ca-
Trogosus Leidy, 1871 nines, and anterior premolars, and a general trend toward increasing
Type species: Trogosus castoridens Leidy, 1871. crown hypsodonty, especially of the lower molars.
Type specimen: ANSP 10337, right -and left rami of the Lofochaius of the Paleocene of China may approximate the ances-
mandible with both i2s, a m2, a portion of a ml, and roots tral morphotype of the Tillodontia. From aLofochaius-like form may
or alveoli of other teeth. have evolved a Meiostylodon-like tillodont, which subsequently
Characteristics: Large trogosines; I2/i2 rootless; i2, i3, Cl/cl, gave rise to Esthonyx. A Meiostylodon-like or Esthonyx-like tillo-
P2/p2 small; P2/p2 single rooted; rostrum short relative to dont appears to have emigrated from Asia to North America, perhaps
 Spencer G. Lucas and Robert M. Schoch 272

w
z L L.81.
1.8
w E.BI
2.4L-

0 4.5 L..-
0 E
:::J ~~~~P· 52
6,0 L..-
a. L.E.Hp.
7.0 '--
E.E.Hp.
8.8 1--
L L.CI.
9.5 1--

E.CI.
11.0 '-
w
z L.L.Ba.
w E.L.Ba.
12.5 L--

0 M ~
0 14.0

~ E.Ba.

15.8 I--

L.Hm.
17.5 1--
E.Hm.

E
18.8 -
L.LAr.
19.2 -
E.l.Ar.

- 23.0

w L.E.Ar.
z
w L
0 27.7L-
0
CJ E.E.Ar.

:::J 29.4 L -
0
.-.Wt.
E 31.9 '--
0,.
- L.Ch. 33.4

34.5 L -
L M.Ch.
35.5 L-
E.Ch,
37.1 L -

Du

39.5 L -

L.Un.
~

w 41.3

z E.Un.
w
0
0
45.9 -
M
L.Br.
_11
W 46.7 -
M.Br.

E. Br.
49.0 '-- -
50.4 I -

L.Wa.
53.5 '--
><
-::.- - -
E M.Wa.
54.2 ~
6
-51-
E.Wa.
55.0 -
E.E.Wa. 55.5
-.n ,. .
L---~.1-------
1-
-----------------:--=--=::-=-:::--==--:=--j
Ck.3 55.6 L...-

~~:~ =·~ ~
0
zW
W
0
L
Tl.6
Tl.5
Tl.4
it!
56:2
57.0
57.3
58.5
L--

.
i-

I . I

ESTHONYCHINAE TROGOSINAE
~ n1_ ::~:
~ ~=~ 61.3
62.5
....

E ~:·.~ 63.B
64.0
Pu.2 64.5
Pu.1 65,0
Pu.O 65,1

Bl.=Blencen, Hp.= Hemphilllen, Cl.= Clerendonien, Be.= Berstovien, Hm.= Hemingfordien, Ar.= Arikereeen,
Wt. =Whitneyen, Or.= Orellen, Ch.= Chedronien, Du.= Duchesneen, Un.= Uinten, Br.= Bridgerien, We.= Wesatchien,
Ck. = Clerkforklen, TI.= Tiffenien, To.= Torrejonien, Pu. = Puercen.

Figure 17.4. Temporal ranges of North American tillodont genera.

via Beringia, before the latest Paleocene (Schoch and Lucas, 1982). ican Esthonyx-like form. An Esthonyx-like form may also have
Basalina appears to be descended from an early side branch of been ancestral to the trogosines. Within the trogosines there were
the tillodonts, endemic to Asia, which retained pl (contra Stucky two Eocene clades: the Megalesthonyx-Adapidium clade and the
and Krishtalka, 1983). European Plesiesthonyx may represent an- Trogosus-Tillodon clade. The presence of both of these clades in
other side branch that is close to, and derived from, a North Amer- China and North America indicates Asian-North American
 Tillodontia 273

interchange during the latest Wasatchian (early Eocene) and/or early Naturwissenschaftliche Klasse, Abhandlungen. Neue Falge Heft, 91,
Bridgerian (middle Eocene). 1-54.
The tillodonts, especially the later derived forms (Trogosus and Gazin, C. L. (1953). The Tillodontia: an Early Tertiary order of mammals.
Smithsonian Miscellaneous Collections, 121, no. 10, 1-110.
Tillodon), were relatively stout, compact, and muscular animals
Gingerich, P. D. (1989). New earliest Wasatchian mammalian fauna from
(Figure 17.1). Trogosus, for example, may have been almost bear- the Eocene of northwestern Wyoming: composition and diversity in a
like in gross appearance and weighed about 150 kg or more. On rarely sampled high-floodplain assemblage. University of Michigan
the digits the tillodonts bore moderately large and recurved claws; Papers on Paleontology, 28, 1-97.
the second incisors formed large, evergrowing tusks; the rostrum Gingerich, P. D., & Gunnell, G. F. (1979). Systematics and evolution of
the genus Esthonyx (Mammalia, Tillodontia) in the early Eocene of
was elongated; and the mandible heavy and deep. In many ways
North America. Contributions from the Museum of Paleontology.
tillodonts developed a morphology that was superficially similar to University of Michigan, 25, 125-53.
that of derived taeniodonts (e.g., Stylinodon: See Schoch, 1986). As Leidy, J. (1868). [Notice of some remains of extinct pachyderms.] Proceed-
in taeniodonts, it has been suggested that tillodonts used their tusks ings of the Academy of Natural Sciences. Philadelphia, 20, 230-3.
and claws to unearth roots and tubers; perhaps they were somewhat (1871). [Remarks on fossil vertebrates from Wyoming.] Proceedings of
the Academy of Natural Sciences. Philadelphia, 23, 113-6.
piglike in their feeding habits. As Gingerich and Gunnell (1979)
Lemoine, V. (1889). Considerations generales sur !es vertebres fossiles des
noted, the heavy wear often observed on tillodont dentitions sug- environs de Reims et specialement sur Jes mammireres de la faune
gests that tillodonts may have ingested a considerable amount of Cernaysienne. Comptes Rendus Congress International Zoologique,
soil and grit with their food. Grooved wear patterns on the sides pp. 233-79.
of the tillodont tusks indicate they may have been used to pull up (1891). Etude d'ensemble sur Jes dents des mammireres fossiles des
environs de Reims. Bulletin de la Societe Geologique de France, 19,
vegetation or strip leaves or buds off stems.
263-90.
Tillodonts, in the form of Esthonyx, were moderately common Lucas, S. G., & Schoch, R. M. (1981). Basalina, a tillodont from the Eocene
in late Paleocene and early Eocene faunas of North America, but of Pakistan. Baverische Staatssammlung fiir Paliiontologie und His-
suffered a drastic decline during the late early Eocene and early torische Geologie, Mitteilungen, Munich, 21, 89-95.
middle Eocene. Tillodonts are last known in North America from a Marsh, 0. C. (1875). Notice ofnew Tertiary mammals, IV. American Journal
of Science, 9, 239-50.
few individuals of Trogosus and Tillodon from Bridger B beds. It has
Robinson, P. (1966). Fossil Mammalia of the Huerfano Formation, Eocene
been suggested that the last of the North American tillodonts may of Colorado. Yale University Peabody Museum of Natural History
have been outcompeted by taeniodonts (Schoch, 1986). They may Bulletin, 21, 1-95.
have also been in fairly direct competition with contemporaneous Rose, K. D. (1972). A new tillodont from the Eocene Upper Willwood
suidlike artiodactyls, such as achaenodonts. Formation of Wyoming. Postilla, Peabody Museum, Yale University,
155, 1-13.
Schoch, R. M. (1986). Systematics, functional morphology and macroevo-
lution of the extinct mammalian order Taeniodonta. Yale University,
REFERENCES Peabody Museum of Natural History Bulletin, 42, 1-307.
Schoch, R. M., & Lucas S. G. (1982). The distribution and paleobiogeogra-
Chow, M. (1963). Tillodont materials from Eocene of Shantung and Honan. phy of the Tillodontia (Mammalia, Eutheria). Geological Society of
Vertebrata Palasiatica, 7, 97-104. America, Abstracts with Programs, 14, 349.
Chow, M., & Wang, B. (1979). Relationship between the pantodonts and Stucky, R. K., & Krishtalka, L. (1983). Revision of the Wind River faunas,
tillodonts and classification of the order Pantodonta. Vertebrata Early Eocene of central Wyoming, Part 4: The Tillodontia. Annals
Palasiatica, 17, 37-48. of the Carnegie Museum, 52, 375-91.
Cope, E. D. (1874). Report upon vertebrate fossils discovered in New Mex- Van Valen, L. (1963). The origin and status of the mammalian order Tillodon-
ico, with description of new species. Geographical Exploration and tia. Journal of Mammalogy, 44, 364-73.
Survey West of the JOOth Meridian. Appendix FF of the Annual Re- Wang, B. (1975). Paleocene mammals of Chaling Basin, Hunan. Vertebrata
port of the Chief of Engineers (U.S. Army) for 1874, pp. 589-606 Palasiatica, 13, 154-62.
(Separatum, pp. 1-18). Young, C. C. (1937). An early Tertiary vertebrate fauna from Yuanchu.
Dehm, R., & Dettingen-Spielberg, T. (1958). Palaontologische und ge- Bulletin of the Geological Society of China, 17, 413-38.
ologische Untersuchungen im Tertiar von Pakistan, 2: Die mit- Zhou, M., Zhang, Y., Wang, B., & Ting, S. (1973). New mammalian gen-
teleocanen Saugetiere von Ganda Kas bei Basal in Nordwest era and species from the Paleocene of Nanhsiung, N. Kwangtung.
Pakistan. Baverische Akademie der Wissenschaften Mathematisch- Vertebrata Palasiatica, I, 31-5.
18 Pantodonta

SPENCER G. LUCAS

INTRODUCTION

The Pantodonta is an extinct order of mammals represented by

Paleocene-Eocene fossils from western Europe, North America, and
eastern Asia. Pantodonts first appeared in Asia during the early Pa-
leocene and disappeared during the middle Eocene, leaving no de-
scendants. All, except perhaps the most primitive pantodonts, were
herbivores, and pantodonts were either the largest or among the
largest mammals of their time (see Figure 18.1). The adaptive ra-
diation of pantodonts was diverse and encompassed mammals as
different as relatively small (7 kg) arboreal herbivores (Archaeo-
lambda) and large (650 kg) ground slothlike terrestrial herbivores
(Barylambda ).
This chapter focuses on the North American Pantodonta, al-
though when discussing pantodont phylogeny and evolution I also
discuss some of the Asian genera. Monographic treatments of the
Pantodonta are Simons (1960) and Lucas (1984a).
Figure 18.1. Restoration of Coryphodon, by Brian Regal.

DEFINING FEATURES OF THE SYSTEMATICS

ORDER PANTODONTA
SUPRAORDINAL
The P3-4 have V-shaped ectolophs (i.e., the pre- and postparacristae
for a V that opens labially); p2-4 have broad metalophids; Ml-3 When Cope (1873) introduced the suborder Pantodonta, he named
parastyles do not project anterior to the protocone-paracone line; and it for one genus, Coryphodon (= "Bathmodon"). Earlier, European
ml-3 metalophids and metaconids are much taller than paracristids workers considered Coryphodon to be a tapiroid perissodactyl (e.g.,
and paraconids. Owen, 1845; Hebert, 1856). Cope (1873) united his new suborder
In addition to these features, which represent a suite of shared de- with Marsh's (1873) Dinocerata and the Proboscidia (sic) vera in
rived character states of the Pantodonta with respect to their nearest the order Proboscidia (sic). Cope (1875) subsequently separated the
relative, the didelphodontines (Lucas, 1982), a number of features Dinocerata and Pantodonta from the Proboscidea and placed them
can be considered typical of pantodonts. These include a complete in a new ungulate order, Amblypoda. He justified this separation
postcranial skeleton (no lost or fused elements), an "amblypod" car- on the basis of pedal characters. Cope (1882) added the suborder
pus and tarsus, and the absence of any sort of cranial protuberances. Taligrada to the Amblypoda, basing it on Pantolambda, and es-
The osteology of the Pantodonta is best described in Cope (1877), tablished an evolutionary sequence that remained in the literature
Matthew (1937), Simons (1960), and Lucas (1984a). for decades. According to Cope, evolution within the Amblypoda
 Pantodonta 275

progressed from Pantolambda ("Puerco-Eocene") to Coryphodon hypothesis of pantodont relationships along the lines advocated by
(Wasatch-Eocene) to Uintatherium (Bridger-Eocene). Marsh (1884) McKenna (1975). It is that scheme which is followed here (see Lu-
did not contest these conclusions, although he proposed the names cas, 1984a, for a complete discussion of these hypotheses).
Amblydactyla (= Amblypoda) and Coryphodontia (= Pantodonta)
because he claimed Cope's names Amblypoda and Pantodonta were INFRAORDINAL
preoccupied.
Cope (1875) suggested that the Diplarthra (Perissodactyla and Ar- A hypothesis of the phylogenetic relationships of North
tiodactyla) were descendants of an unknown group of amblypods American pantodont genera and non-North American pantodont
with bunodont teeth that he termed Amblypoda Hyodonta. Cope families (Lucas, 1982) is shown in Figure 18.2. All North American
( 1897) later concluded that the Periptychidae (originally considered pantodonts are eupantodonts united principally by the W-shaped
to be condylarths) were the Amblypoda Hyodonta. Thus he trans- ectolophs of their upper molars (secondarily lost in Coryphodonti-
ferred the Periptychidae from the Condylarthra to the Amblypoda dae ). Titanoides stands apart from other North American pantodonts
as a family of the Taligrada coordinate with the Pantolambdidae. and may be most closely related to Asian pantolambdodontids. The
Osborn (e.g., 1898) most fully developed Cope's ideas about the re- barylambdids form a distinct clade of graviportal, ground slothlike
lationships of the Amblypoda. Particularly significant in this regard forms with reduced anterior dentitions. Pantolambdids are the sister
was Osborn's scheme of amblypod dental homologies that rendered taxon of the coryphodontids.
simple his and Cope's envisioned evolutionary progression from Two mammals, Deltatherium and Cyriacotherium, which have
Pantolambda to Coryphodon to Uintatherium. been included in the Pantodonta by some authors, are excluded
Wood (1923), however, reinterpreted the homologies of the cusps here. Deltatherium Cope, 1881a (= Lipodectes Cope, 1881b) had
of uintathere cheek teeth so as to make it improbable that Uin- a long and troubled history of being allied with various mam-
tatherium was a descendant of Coryphodon. Furthermore, new dis- malian taxa (Matthew, 1937, pp. 68-69) until Matthew (1937) thor-
coveries of Paleocene pantodonts and uintatheres in the western oughly described Deltatherium and allied it with Chriacus in the
United States (Simpson, 1929, 1937; Jepsen, 1930; Patterson, 1933, Arctocyonidae. Deltatherium then "remained" an arctocyonid un-
1934, 1935, 1939) failed to bridge the morphological gap between til McKenna (1975, p. 37), without discussion, included it in the
uintatheres and pantodonts. The result was Simpson's (1937) recog- Pantodonta. Van Valen (1978, p. 63) considered Deltatherium an
nition of the Pantodonta as an order separate from the Dinocerata, arctocyonid near the ancestry of pantodonts. However, I exclude
and his elimination of the Amblypoda as a higher taxon to contain Deltatherium from the Pantodonta because it lacks the synapomor-
these two groups. Simpson (1937) also followed Matthew (1897, phies I use to unite the Pantodonta listed here.
1937) in placing the periptychids in the Condylarthra, far removed Nevertheless, I do consider Deltatherium a sister taxon of the
from any close relationship to the pantodonts. Despite this, Simpson Pantodonta and note that its buttressed mandible as well as other,
( 1945) subsequently advocated a somewhat less close relationship of almost certainly primitive, features (e.g., large canines, tall sagittal
the Pantodonta and Dinocerata within his superorder Paenungulata. crest, deep postorbital constriction) lend it a primitive pantodont
Simpson's (1945) concept of pantodont relationships remained gestalt. Despite these resemblances, Deltatherium shares derived
essentially unmodified for three decades. During this time, Simons features with tillodonts (pl absent, relatively molariform p4, Ml-2
(1960), in a comprehensive revision of North American Paleocene anterior and posterior cingula, Ml-3 paracone and metacone ribs)
pantodonts, presented a classification of pantodonts only slightly that suggest it may be more closely related to tillodonts than panto-
different from Simpson's and arrived at no conclusion regarding donts, although I do not include it in the Tillodontia.
their relationships to other orders of mammals. Wheeler (1961) sug- Deltatherium fundaminis is known from localities SB23E,
gested that pantodonts and uintatheres may be closely related, but SB23F, SB23G, and SB23H.
also undertook no explicit changes in Simpson's (1945) conclusions. Rose and Krause ( 1982) referred their new genus Cyriacothe rium
Significant modification of Simpson's (1945) classification, and (for which they proposed the new family Cyriacotheriidae) to the
the phylogenetic relationships it reflected, only appeared thirty years superfamily Pantolambdoidea within the Pantodonta. They based
later when McKenna (1975) removed the pantodonts from the un- this assignment on the following features shared by Cyriacotherium
gulates. He considered the Pantodonta didelphodontine derivatives and "pantolambdoids":
far removed from a condylarth ancestry or a close relationship to
a W-shaped ectoloph and the absence of a hypocone in the
uintatheres. However, Szalay (1977) and Van Valen (1978, 1988)
upper molars, dilambdodont lower molars with a deep hy-
continued to maintain a condylarth ancestry for pantodonts.
poftexid (the cristid obliqua running toward or meeting the
Chow and Wang (1979) followed McKenna (1975) in remov-
metaconid), a lingual hypoconulid and rudimentary or absent
ing pantodonts from a condylarth ancestry. But, unlike McKenna,
entoconid, a tendency toward molarization of the posterior
they considered pantodonts derivatives of a Deltatheridium-like
lower premolars, relatively simple, small incisors that in-
mammal and recognized three pantodont clades: (1) taligrades, in-
crease in size from I 1 to I3, and a fused (co-ossified) mandibu-
cluding Asian (Bemalambdoidea) and North American (Pantolamb-
lar symphysis. (Rose and Krause, 1982, pp. 42-43.)
doidea) branches; (2) coryphodonts, derived from phenacolophids;
and (3) tillodonts, previously considered an order separate from I reject Rose and Krause's (1982) inclusion of Cyriacotherium
Pantodonta. Lucas (1982) most recently presented a phylogenetic in the Pantodonta and prefer to ally this taxon with the Dermoptera
 Spencer G. Lucas 276

2 5 6

PANTODONTA

Figure 18.2. Interrelationships of the Pantodonta. Character states that correspond to the numbered node points are as follows: (1) PANTODONTA:
P3-4 have v-shaped ectolophs (i.e., the pre- and postparacristae form a v that opens labially); p2-4 have broad metalophids; Ml-3 parastyles
do not project anterior to the protocone-paracone line; and m 1-3 metalophids and metaconids much taller than paracristids and paraconids.
(2) EUPANTODONTA: Complete Ml-2 lingual cingula; m3 broader than m2, which is broader than ml. (3) Broad Ml-3 protocones; relatively
shallow Ml-3 ectoflexi; w-shaped ectolophs on Ml-2 or v-shaped Ml-2 ectolophs derived from a w-shaped one (except the pantolambdodon-
tid Harpyodus). (4) Ml-3 trigon basins narrow; Ml metastyle longer than parastyle; claws on phalanges of manus and phalanges relatively long
(known only for Archaeolambda and Titanoides). (5) TITANOIDEIDAE: Ml-2 w-shaped ectoloph united at the mesostylar cusp; P3-4 lack post-
protocristae; upper canine long and saberlike in presumed males and meets a broad shearing surface on the lower canines; mandibular symphysis
probably unfused. (6) Canines and premolars small so that the skull has a long and narrow rostrum and weak sagittal and nuchal crests; p3-m3
paraconids and metaconids of subequal height; M3 metacone lingually positioned; m3 talonid cuplike; lower jaw shallow and symphysis weak. (7)
Ml-2 w-shaped ectoloph similar to that of Titanoides but with shorter styles and less overall breadth; oblique (not vertical) labial slopes on ml-3.
(8) p2 large relative to the other lower premolars; ml-2 talonids broader than trigonids; strong hypoconulid lobe on m3. (9) PANTOLAMBDIDAE: p3-4
talonids cup shaped; p4 paraconids and metaconids of subequal height and closely appressed. (I 0) p 1 sectorial; skull dolicocephalic; dorsal vertebral series
relatively short; lack of a neck on a broad astragalus. ( 11) CORYPHODONTIDAE: Skull roof broad with parasagittal crests above orbits; premaxilla large
and heavy; postglenoid processes long and close to mastoid processes; canines very large and together with spatulate incisors form an arc that flares outward
in hippolike fashion; Ml-3 subcircular to square; Ml-3 paracones labial and small (anterior v of w-shaped ectolophs lost); Ml-3 protolophs connect
protocones to parastyles; M3 broader than M2, which is broader than M 1; m3 and m2 of equal breadth and broader than ml; ml-2 subbilophodont, and m3
subbilophodont except in some primitive Coryphodon. A large number of derived postcranial features of Coryphodon distinguish it from the Pantolambdidae,
but the postcrania of the other coryphodontids are virtually unknown. These features include tail short and first few caudal vertebrae fused; scapula with
large acuminate process; well-developed supracondyloid ridge on humerus; lunar flattened proximodistally; centrale and scaphoid fused; short and stout
phalanges on manus and pes; ilia broadly expanded posteriorly; and tibial facet of astragalus extends distally. (12) BARYLAMBDIDAE: Small canines;
relatively molariforrn p4s; P3-4 metastyles that are long and with protocones that are displaced anteriorly; ml-3 cristids obliquae climb the metalophids;
skulls small relative to body size; nasals short and broad.

for the following reasons: 3. On the Ml-3 of Deltatherium, the cristid obliquatrigonid con-
1. Among the features cited by Rose and Krause (1982) that Cyr- tact is positioned medially, and in Cyriacotherium and pantodonts
iacotherium and pantodonts share, the following are primitive (as it is only slightly more lingual. Thus the positions of the cristida
Rose and Krause noted) and cannot be used to corroborate their hy- obliquae of pantodonts and Cyriacotherium may be closer to the
pothesis that Cyriacotherium is a pantodont: absence of a hypocone primitive condition than stated by Rose and Krause (1982).
on Ml-3; relatively simple, small incisors that increase in size from 4. Fusion of the mandibular symphysis has arisen independently
Il/i 1 to I3/i3. in several, separate clades of mammals and cannot be used to
2. If Didelphodus and/or Deltatherium are used as out-taxa that corroborate assignment of Cyriacotherium to the Pantodonta. Like-
represent the primitive condition for assessing character-state wise, a simple tendency to molarize posterior premolars is present
polarities among pantodonts, the lingual hypoconulid and rudimen- in many separate mammalian lineages.
tary or absent entoconid on m3 of Cyriacotherium may also be 5. The W-shaped ectolophs of the upper cheek teeth of
primitive (Rose and Krause, 1982, Table 5). Cyriacotherium only superficially resemble those ofpantodonts and
 Pantodonta 277

differ in the following significant details: ectoloph crests cuspidate; The locality numbers may be listed in a couple of alternative ways.
ectoloph V's narrow anteroposteriorly; mesostyles as labial as the Parentheses around the locality (e.g., [CPIOl]) mean the taxon in
para- and metastyles; and mesostyles enlarged and twinned. Fur- question at that locality is cited as an "aff." or "cf." the taxon in
thermore, W-shaped molar ectolophs that differ from each other in question. Parentheses are usually used for individual species, thus
detail, and also differ in detail from the W-shaped ectolophs of Cyr- implying the genus is firmly known from the locality, but the actual
iacotherium, probably arose independently three times within the species identification may be questionable. Question marks in front
Pantodonta (Lucas, l 984a). of the locality (e.g., ?CPlOl) mean the taxon is questionably known
6. Cyriacotherium also differs from pantodonts in possessing from that locality, thus implying some doubtthat the tax on is actually
more molariform premolars than any pantodont. And, in Cyria- present at that locality, either at the genus or the species level.
cotherium, the P3-4 have W-shaped ectolophs, whereas P3-4 of all
pantodonts have V-shaped ectolophs.
7. In a cladogram of the taxa Pantodonta, Cyriacotherium, and TITANOIDEIDAE
Dermoptera (Cynocephalus plus plagiomenids), Cyriacotherium Characteristics: Medium to large pantodonts (length of Ml-3 =
shares the following derived features with dermopterans: P2-4 very 59.0-82.0 mm) in which Il/il < I2/i2 <I3/i3; short diastemata are
molariform via the development of W-shaped ectolophs similar to present anterior and posterior to Cl; Pl is double rooted;
those seen on Ml-3 and by the addition of accessory stylar cusps P2-4 have subquadrate bases and "isolated (lacking prominent pre-
labial to the distinct paracones and metacones; relatively molari- and postprotocristae) protocones anterior to a transverse midline
form p3-4 which bear large talonids that have widths subequal to through the tooth; Ml-2 paracones and metacones are of subequal
the trigonid widths; Ml-3 mesostyles large, labially positioned, size and their para- and metastyles extend far labially; M3 parastyle
and twinned or cleft; M3 metastyle relatively labial; enamel of is prominent, but the metacone and metastyle are much reduced;
molariform cheek teeth crenulated; upper molariform cheek teeth ml-3 lack metastylids, have large hypoconids and small but distinct
bear deep, transverse valleys (between the anterior and posterior entoconids; mandibular symphysis not fused and not buttressed;
V's of the ectolophs) that separate the paraconules from the mandibular angle not inflated; coronoid process short; sagittal crest
metaconules. single and prominent; premaxillae small with slender descending
Rose and Krause (1982) considered the possibility that Cyria- rarni; zygomatic arches slender and bowed outward; head relatively
cotherium might have dermopteran affinities, but rejected any close large; tail short; scaphoid and central fused; high round lunar; long
relationship between the two groups. They based this rejection phalanges and claw-bearing ungual phalanges (see Figure 18.3).
primarily on the lack of the following derived plagiomenid char-
acters in Cyriacotherium: in the molariform cheek teeth "a trend Titanoides Gidley, 1917
away from a W-shaped ectoloph and toward a proliferation of stylar Type species: Titanoides primaevus Gidley, 1917.
cusps"; "a more labially oriented cristid obliqua"; "large, bilobed
Type specimen: USNM 7934, right p3-m2, left m2, and sym-
lower incisors"; "oblique or skewed conformation of the crowns
physeal fragments. PU 16490, cranial fragments including
of the cheek teeth"; "paraconid reduced and trigonid compressed";
right Il-M3, apparently represents the same individual as
"lower incisors decrease in size from il to i3," and "ectocingulids USNM 7934 (cf. Simons, 1960, pp. 33-34).
well developed on lower cheek teeth" (Rose and Krause, 1982, Characteristics: Same as for Titanoideidae.
p. 41). Furthermore, according to Rose and Krause (1982, p. 41), Average length of m2: 23.2-27.9 mm.
Cyriacotherium lacks the following derived features, among oth- Included species: T. primaevus (known from localities
ers, seen in Cynocephalus: labial orientation of the cristid obli- CP13G CP16C, CP24B, CP62A, CP62B, NP3F, NP47C,
qua, reduced paracristids, and highly modified lower incisors. Thus, NP48); T. majus (locality CPI 7 A); T. gidleyi (locality
on the basis of the lack of these dermopteran features, Rose and CPl3E); T. zeuxis (localities SB39B, CP62A, NP19IIC);
Krause ( 1982) concluded that Cyriacotherium can be excluded from T. simpsoni (locality NP19C).
the Dermoptera. However, as noted earlier, Cyriacotherium still Titanoides sp. is also known from localities CP16A,
shares more derived features with dermopterans than it does with CP17A, CP62B, NP3D, NP7D, NP19IIA, NP19IIC,
pantodonts. Its relationships are thus better sought among the Der- NP19IID, NP20E.
moptera than among the Pantodonta. Cyriacotherium argyreum is
known from locality NP7D; C. psammium is known from local- Comments: This genus is oversplit and in need of revision
ities CP17B and CP18B; and Cyriacotherium sp. is known from (see Simons, 1960; Schiebout, 1974; Holtzman, 1978; and
localities CP13F and CP13G. Lucas, 1984b).

PANTOLAMBDIDAE
INCLUDED NORTH AMERICAN GENERA IN THE
ORDER PANTODONTA Characteristics: Small- to medium-sized pantodonts (length of
Ml-3 = 30.0-63.0 mm) in which Il/il <I2/i2 <I3/i3; there is a
The locality numbers listed for each genus refer to the list of unified short diasterria anterior to or a long diastema posterior to Cl but
localities in Appendix I. The acronyms for museum collections are not both; Pl is single rooted; P2-4 have subtriangular bases and
listed in Appendix III. protocones that are on a transverse line through the paracones and
 Spencer G. Lucas 278

Figure 18.3. Osteology of Titanoides primaevus. A. Skeleton, right lateral view (scale = 10 cm). B. Right Cl and Pl-M3, occlusal view (scale =
1 cm). C. Skull and lower jaw, right lateral view (scale= 2 cm) (all after Simons).

are connected to pre- and postprotocristae; Ml-2 paracones and Caenolambda Gazin, 1956
metacones of subequal size and para- and metastyles do not extend Type species: Caenolambda pattersoni Gazin, 1956.
far labially (i.e., ectoflexus shallow); M3 parastyle is prominent and Type specimen: USNM 21036, skull.
metacone plus metastyle much reduced; ml-3 have metastylids, Characteristics: Large pantolambdids (length ofMl-3 = 63.0
large hypoconids, and indistinct to small entoconids; the mandibular mm) that differ from Pantolambda in their dolicocephalic
symphysis is fused and the anterior portions of the horizontal rami skull, sectorial pl (and corresponding post-Cl diastema),
are thickened (buttressed); the mandibular angle is not inflated and relatively short dorsal vertebral series, and broad astragalus
the coronoid process is short; a single and prominent sagittal crest; that lacks a distinct neck.
relatively large premaxillaries with heavy descending rami; zygo- Average length of m2: 20.0 mm.
matic arches are stout and bowed outward; head small relative to Included species: C. pattersoni (known from locality CPI 6A);
body; tail long but not massive; pelvis lacking any specializations for C. jepseni (locality CP13E). A third, unnamed species of
graviportality; scaphoid and centrale not fused; lunar low and prox- Caenolambda from the Black Peaks Formation, Big Bend
imal end not rounded; relatively long phalanges and fissured ungual National Park, Texas (locality SB39A) is currently under
phalanges. study.
Caenolambda sp. is also known from localities SB39B,
Pantolambda Cope, 1882 NP3.
Type species: Pantolambda bathmodon Cope, 1882.
Type specimen: AMNH 3956, mandibular fragment with Comments: Gingerich and Childress (1983, p. 151) consid-
p3-ml. ered Caenolambda a barylambdid and questioned whether
Characteristics: Small pantolambdids (length of Ml-3 = it is generically separable from Haploiambda and Bary-
30.0--46.0 mm) that lack the dolicocephalic skull, secto- lambda. However, it is clear that Caenolambda is not a
rial p 1 (and corresponding post-C 1 diastema), short dorsal barylambdid and thus cannot be considered a synonym of
vertebral series, and broad astragalus which lacks a neck any of the genera in that family. Among other features
that are diagnostic of Caenolambda. that clearly set Caenolambda apart from the barylamb-
Average length of m2: 10.3-19.5 mm. dids are its dolicocephalic skull with a prominent sagittal
Included species: P. bathmodon (known from locality crest; its buttressed mandible; its low coronoid process; the
SB23H); P. cavirictum (localities SB23GG, SB23H, structure of its p3--4 (see earlier); its short dorsal vertebral
[CP13B], [CP15A]); P. intermedium (localities SB23E, region; the lack of any graviportal specializations of the
NP19C) (see Simons, 1960, and Lucas, 1984b). limbs and pelvis; and the absence of any modification of
Pantolambda sp. is also known from localities NP20A, the caudal vertebrae for weight bearing. The sectorial pl of
NP46, ?NP47B, ?NP47C. Caenolambda thus is not a sexually dimorphic feature, as
 Pantodonta 279

Gingerich and Childress (1983) imply, but instead a dental Included species: H. quinni (known from localities CP13G,
specialization of a distinct taxon. CP62A, CP62B); H. simpsoni (locality CP2) (see Simons,
1960; Miller, 1986).
BARYLAMBDIDAE Leptolambda Patterson and Simons, 1958
Characteristics: Medium to large pantodonts (length of Type species: Leptolambda schmidti Patterson and Simons,
Ml-3 = 42.0-85.0 mm) in which Il/il >I2/i2>I3/i3; there is a 1958.
short diastema anterior to, but not posterior to, Cl; Pl is single Type specimen: FMNH P 26075, incomplete skull and partial
rooted; P2--4 have triangular bases and protocones that are on a skeleton.
transverse line with their paracones and are connected to pre- and Characteristics: Medium-sized barylambdids (length of
postprotocristae; Ml-2 and metacones of subequal size and para- Ml-3 = 64.0-73.0 mm) in which the cl and pl are of
and metastyles do not extend labially (ectoflexus very shallow); M3 similar shape and size, the incisors are broad and trian-
parastyle prominent and metastyle reduced but metacone relatively gular, Pl is relatively broad, M3 is small, the skull has a
large; ml-3 have metastylids, large hypoconids, and small to indis- very wide and long rostrum, the dorsal vertebrae have low
tinct entoconids; mandibular symphysis fused but no buttressing of neural spines, the skull is very small relative to the body,
the rami; no inflation of mandibular angle; coronoid process long; the third trochanter of the femur is vestigial, and the caudal
sagittal crest single and low; premaxillaries small with weak de- vertebrae lack haemapophyses.
scending rami; zygomatic arches robust but do not flare outward; Average length ofm2: 21.5-26. 7 mm.
head very small relative to body; tail long and heavy with modifica- Included species: L. schmidti only (known from localities
tions for weight bearing; massive pelvis with graviportal structure; CP13G, CP26B) (see Simons, 1960).
scaphoid and centrale not fused; lunar large and rounded; phalanges
Comments: Gingerich and Childress (1983, p. 143) consid-
short with fissured ungual phalanges.
ered Leptolambda to be a junior subjective synonym of
Barylambda but failed to take into account the many differ-
Barylambda Patterson, 1937 (synonym: Titanoides, in part) ences, especially in the postcrania, between the two genera
Type species: Barylambda faberi (Patterson, 1933) (= Ti- (cf. Patterson and Simons, 1958).
tanoides faberi).
Type specimen: FMNH Pl4637, nearly complete skeleton.
lgnatiolambda Simons, 1960
Characteristics: Large barylambdids (length of Ml-3 =
Type species: Ignatiolambda bamesi Simons, 1960.
76.0-85.0 mm) with relatively small canines, shallow
Type specimen: AMNH 55400, maxillaries with C, P2-M3,
P3--4 ectoflexi, and broad (square) M3 protocones.
mandible with left and right c, p2-m3, and a left manus.
Average length of m2: 25.2-29.6 mm.
Characteristics: Small barylambdids (length of Ml-3 = 53
Included species: B. faberi (known from localities CP62A,
mm) in which the P3--4 ectoflexi are very shallow, Ml-3
CP62B); B.jackwilsoni (locality SB39B); B. churchilli (lo-
labial cingula are large, M3 posterior cingulum is large and
cality CP13F) (see Simons, 1960; Schiebout, 1974;
its protocone is broad (square), and the p4 talonid is very
Gingerich and Childress, 1983).
large.
Barylambda sp. is known from localities SB39E, CPI 7 A.
Average length of m2: 20.0 mm.
Comments: Gingerich and Childress (1983, p. 145) consid- Included species: I. bamesi only (known from localities
ered Barylambdajackwilsoni to be a junior subjective syn- SB20A, SB20B).
onym of Caenolambda jepseni. However, I accept
Comments: Gingerich and Childress ( 1983, p. 151) suggested
Schiebout's (1974) assignment of B. jackwilsoni to Bary-
that Ignatiolambda may be a synonym of Haplolambda.
lambda. Certainly the structure of p3--4, the lingually open
However, the two genera are readily separated as indicated
m3 talonid, and lack of evidence of a mandibular buttress
in their diagnoses here.
of the holotype lower jaw fragment of B. jackwilsoni pre-
clude its assignment to Caenolambda.
CORYPHODONTIDAE
Haplolambda Patterson, 1939 Characteristics: Large pantodonts (length of Ml-3 = 70.0-80.0
Type species: Haplolambda quinni Patterson, 1939. mm) distinguished from other pantodonts by the possession of broad
Type specimen: FMNH P15542, anterior half of skeleton. skull roofs with parasagittal crests above the orbits, large and heavy
Characteristics: Small barylambdids (length ofMl-3 = 42.0 premaxillae, long postglenoid processes that are close to the mas-
mm) in which the cl and pl are of similar shape and size, toid processes, very large canines that, with the incisors, form an arc
the incisors are very broad and triangular, M3 is small, Pl that flares outward in hippolike fashion, large and spatulate incisors,
is relatively broad, and the M3 metacone is vestigial. subcircular to square Ml-3 that have small, labially positioned para-
Average length ofm2: 19.0 mm. cones (anterior V of W-shaped ectolophs absent), Ml-3 protolophs
 Spencer G. Lucas 280

D \~'<,_..:;;;,c~.,;y;c,- -7 - ~ --
• ,';~ .· . . . • • 5 .·.;,;"·~'>·:~~~~~
Figure 18.4. Osteology of Corypho-
rt~~~
\J?J /lV ·.·...-.Cr-~~--"· don anthracoideus. A. Skeleton,
.·. .J U'f right lateral view (scale = 10 cm;
after Osborn). B-C. Left Pl-M3 (B)
l and right p2-m3 (C), occlusal views

(,~· '.:2::/_,-~~-~i.;.~p-·· .:
(scale= I cm; after Osborn). D. Skull
and lower jaw, right lateral view
(scale= 2 cm; after Simons).

connect protocones to parastyles, M3 broader than M2, which is Type specimen: BMHM 27848, right dentary fragment with
broader than Ml, m3 and m2 of equal breadth and broader than ml, incomplete m2 and complete m3.
ml-2 subbilophodont and m3 subbilophodont except in some prim- Characteristics: Small- to medium-sized coryphodontids
itive Coryphodon; tail short and first few caudal vertebrae fused, (length ofMl-3 = 70-99 mm) with long P2-4 ectolophs
scapula with large acuminate process, well-developed supracondy- that are longer than those of Eudinoceras and Asioco-
loid ridge on humerus, lunar flattened proximodistally, centrale and ryphodon, about the same relative length as those of
scaphoid fused, short and stout phalanges on manus and pes, ilia Hypercoryphodon, and thus are shorter than those of
broad anteriorly and tibial facet of astragalus extends distally (see Heterocoryphodon. The P2-4 protocones of Coryphodon
Figure 18.4). are not isolated like those of Asiocoryphodon and Eu-
Coryphodon is the only genus of the family found in North dinoceras. Ml-3 of Coryphodon are the least bilophodont
America (it is also known from Europe and Asia). Four other of any coryphodontid, in that postprotocristae or meta-
coryphodontid genera, Eudinoceras, Hypercoryphodon, Asiocory- conules are present (the only exception to this is C. simus)
phodon, and Heterocoryphodon, are known from the Eocene of Asia. and the premetacristae are short relative to the protolophs.
M3 of Coryphodon is an oval tooth with a distinct poste-
Coryphodon Owen, 1845 rior crest, and thus is less bilophodont than M3 of other
Type species: Coryphodon eocaenus Owen, 1845. coryphodontids. The lower canine of Coryphodon is not
 Pantodonta 281

curved outward, nor are its il-3 extremely broad, unlike the Pantodonta (contra Muizon and Marshall, 1987; Marshall and
Eudinoceras. The p2-4 trigonids of Coryphodon are rela- Muizon, 1988).
tively long (unlike Eudinoceras), and its ml-3 paracristids From their Asian origin, pantodonts emigrated into North and
and cristids obliquae are usually better developed than are South America during the early Paleocene (no later than the North
those of other coryphodontids. The skull roof of Corypho- American Torrejonian). The distribution thereby established for
don is broader than that of Hypercoryphodon, Asioco- pantodonts is similar to that achieved by uintatheriamorphs at the
ryphodon, and Heterocoryphodon, and its skull is not as same time (Schoch and Lucas, 1985; Lucas, 1986). Thus panto-
dolicocephalic as that of Hypercoryphodon. lambdodontoids were present in Asia (Harpyodus, Pastoralodon,
Average length of m2: 26.0-41.0 mm. Archaeolambda, Altilambda, Pantolambdodon), North America (Ti-
Included species: C. simus, C. molestus, C. eocaenus, C. an- tanoides), and South America (Alcidedorbigniya) by the end of the
thracoideus, C. oweni, C. lobatus, C. subquadratus, and C. Paleocene, much as Dinocerata (Prodinoceras) were in Asia and
proterus. The synonymy of these species is extensive and is North America and xenungulate pyrotheres ( Carodnia) were present
given by Lucas (1984a, 1984c). Coryphodon is ubiquitous in South America at that time (Lucas, 1986). Indeed, similar dis-
at all Wasatchian localities and present at many Clark- tributions may also exist for edentates and condylarths during the
forkian sites. Lucas (1984b) provides a complete listing of Paleocene, establishing an Asian-North American-South American
Coryphodon occurrences. To be brief, one or more species generalized track (Lucas, 1986, and references cited therein).
of Coryphodon are present at localities GC21II, SB21, The Asian (and South American) evolution of pantolambdodon-
SB22A, SB22B, SB22C, SB24, SB26A, SB40, CP4, toids encompassed relatively small mammals (<10 kg) that, if the
CP13H, CP14E, CP17A, CP17B, CP18B, CP19, CP20A, skeleton of Archaeolambda tabiensis is typical (Huang, 1977), were
CP20B,CP20C,CP20D,CP20E,CP22C,CP23A,CP24B, lightly built, arboreal herbivores. North American Titanoides, in
CP24C, CP26A, CP26B, CP26C, CP25D, CP25E, CP26F, contrast, was a relatively large (estimated mass of 150 kg for Ti-
CP25G,CP26H,CP26C,CP26D,CP27A,CP27B,CP27C, tanoides primaevus), terrestrial, clawed quadruped with saberlike
CP27D,CP27E,CP28B,CP28C,CP62C,CP63,CP64A, upper canines. This animal may have done some digging to obtain
CP64B, CP64C, NP14, NP49. its food (Coombs, 1983).
The remaining North American pantodonts form a monophyletic
group that I refer to informally as the "eupantodonts." Their evo-
INDETERMINATE PANTODONTS
lutionary radiation began in North America during the Torrejonian
Fragmentary remains ascribed to pantodonts are known from local- with Pantolambda, a terrestrial, quadrupedal browser and its some-
ities CC50, CP14C, NP20B. what aberrant (note especially the sectorial pl), although function-
ally similar sister taxon, Tiffanian-Clarkforkian (late Paleocene)
Caenolambda.
BIOLOGY AND EVOLUTIONARY PATTERNS The Tiffanian-Clarkforkian barylambdid pantodonts were con-
fused early with Asian archaeolambdids because both families con-
The origin and early evolution of the Pantodonta is a subject of dis- verged on some dental features, especially reduction of the anterior
agreement (compare Lucas, 1982, l 984a with Muizon and Marshall, dentition. However, postcranially, these families are very different.
1987, and Van Valen, 1988). I believe that pantodonts originated in Barylambda is the best known barylambdid and was functionally
Asia from a didelphodontinelike ancestry during the earliest Pa- somewhat analogous to the large ground sloths of the Quaternary.
leocene. In so doing I (1) consider the Chinese Paleocene genera The graviportal pelvis, massive hind limbs, heavy tail, and caudal
Bemalambda and Hypsilolambda to be the first pantodonts (contra haemopophyses suggest that Barylambda was capable of bipedal
Van Valen, 1988); (2) consider the Chinese stratigraphic horizons browsing. Leptolambda shows some of these modifications as well,
(especially in the Nanxiong basin, Guangdong) from which these but the two smaller barylambdid, Haplolambda and Ignatiolambda,
taxa are derived to be earliest Paleocene, older than the North Amer- are too poorly known postcranially to allow firm conclusions on
ican Torrejonian (contra Sloan, 1987); and (3) do not consider the their mode of life.
fossil mammals from the El Molino Formation at Tiupampa, Bo- From the standpoint of abundance of fossils, longevity, diver-
livia, to be Late Cretaceous, but instead of Paleocene age (contra sity, and breadth of geographic distribution, Coryphodon was the
Marshall and Muizon, 1988; Marshall et al., 1985). I thus agree most successful pantodont. It was the largest land mammal of the
with Van Valen (1988) on the Paleocene age of the Tiupampa Clarkforkian-Wasatchian (latest Paleocene-early Eocene). Coryph-
mammals, but do not follow him in considering dinosaur fossils from odon was a terrestrial, graviportal, subdigitigrade quadruped with an
the El Molino Formation to be of Paleocene age. Indeed, I have seen adult body weightofl50-300 kg (see Figure 18.1). There is no com-
no convincing evidence presented that there are dinosaur fossils in pelling evidence that it was amphibious, other than the abundance
the El Molino Formation temporally equivalent to or younger than of its fossils in some ftuvial and lacustrine deposits. Postcranially,
the Tiupampa mammals. Alcidedorbignya from Tiupampa probably Coryphodon most closely resembled the living pygmy hippopota-
is a pantodont (cf. Muizon and Marshall, 1987) and is morphologi- mus Hexaprotodon. The extinction of Coryphodon at the end of the
cally most similar to Harpyodus ( = Wanalia Van Valen, 1988) from Wasatchian marked the disappearance of pantodonts in North Amer-
the early to late Paleocene of China. It thus is not the most prim- ica. But, in Asia, other coryphodontid genera persisted through the
itive pantodont, nor does it support inclusion of the Tillodontia in middle Eocene.
 Spencer G. Lucas 282

w
z L L.81.
u
w E.81
2.4
0 4.5
0 E L•t.Hp.
5.2
:.i l.Hp.
6.0
D. LE.Hp.
7.0
E.E.Hp.
L L.CI.
...
8,8

E.CI.
11.0
w
zw L.L.Ba.
12.5
E.L,Ba.
0 M
0 14.0
:g E.BL

16.8
L.Hm.
17.5
E.Hm.
18.B
E LL.Ar.
19.2

E.L.Ar.
23.0

w l.E.Ar.
zw
L
0
0
27.7 - - -
CJ E.E.Ar.
:.i 29.4
0
Wt.
E 31.9
O•.
33.4
L.Ch.
34.5
L M.Ch.
35,5
E.Ch.
37.1

Du

39.5

L.Un.

w 41.3

z E.Un.
w
0 45.9
0
w M
L.Br.
46,7

M.Br.
49,0

E.Br.
50.4
r::
L.Wa. .g
53.5 -0
E M.Wa.
54.2 Ill -{l- ~
Ill
'ti -
-a.~
E.Wa. 'ti .Q .Q .Q 0
55.0 .a - -e- - e - e -o
E.E.Wa.
55.5 - - - - - - = - - - - - - - , - - - - - - e ---111--- 111--.!!! ------=-+----!
1 --111-
.g
~~~ i1 =~ ~l~ - ll ~ ~I==
Ck.3 55.6
w
zw
Ck.2
Ck.1
Tl.6
........ ::::g:
55,8
.Q
e = !l - ~
-Er=
- - =~!
L Tl.5
g &:
-
57.0
0 Tl.4 57.3 -111 - .!!!
=~
Tl.3
0 -'Qr?- 0
-·] -
58.5
Tl.2
w T1.1 59.2 r::
_r:: - QI

=
..I 60.9
~
=~- -
To.3
<I: To.2
61,3
D. 62.5
E To.1
Pu.3 -
Pu.2 54.5
Pu.1 85.0 PANTOLAMBDIDAE TITANOIDEIDAE BARYLAMBDIDAE CORYPHODONTIDAE
Pu.O 55.1

Bl.=Blencen, Hp. = Hemphillien, Cl.= Clerendonlen, Be.= Berstovlen, Hm.= Hemlngfordien, Ar. = Arikereeen,
Wt. =Whitneyen, Or.= Orellen, Ch.= Chedronien, Dli.= Duchesneen, Un.= Uinten, Br.= Bridgerien, We.= Wesetchien,
Ck.= Clerkforkien, Tl.= Tlffanlen, To. = Torrejonlen, Pu.= Puercen.

Figure 18.5. Temporal ranges of pantodont genera.

 Pantodonta 283

REFERENCES (1937). Paleocene faunas of the San Juan Basin, New Mexico. Trans-
actions of the American Philosophical Society, new series, 30,
Chow, M., & Wang, B. (1979). Relationship between the pantodonts and 1-510.
tillodonts and classification of the order Pantodonta. Vertebrata McKenna, M. C. (1975). Toward a phylogenetic classification of the
PalAsiatica, 17, 37-48. Mammalia. In Phylogeny of the Primates, eds. W. P. Luckett &
Coombs, M. C. (1983). Large mammalian clawed herbivores: a comparative F. S. Szalay, pp. 21-46. New York: Plenum.
study. Transactions of the American Philosophical Society, 73, 1-96 . Miller, W. E. (1986). A new species of pantodont, cf. Haplolambda
Cope, E. D. (1873 ). On the short-footed U ngulata of the Eocene of Wyoming. simpsoni (Mammalia) from Utah. Journal of Paleontology, 60,
Proceedings of the American Philosophical Society, 13, 38-74. 1138-42.
(1875). On the order Amblypoda. American Naturalist, 9, 427. Muizon, C. de, & Marshall, L. G. (1987). Le plus ancien Pantodonte (Mam-
(1877). Report upon the extinct Vertebrata obtained in New Mexico by malia), du Cretace superieur de Bolivie. Comptes Rendus Academie
parties of the expedition of 1874. Geographical Surveys West of the Science Paris, 304, 205-8.
JOOth Meridian (Wheeler Survey), 4, 1-370. Osborn, H. F. (1898). Evolution of the Amblypoda, Part 1: Taligrada and
( 1881 a). Mammalia of the lower Eocene beds. American Naturalist, 15, Pantodonta. Bulletin of the American Museum of Natural History,
337-8. 10, 169-218.
(188lb). Notes on Creodonta. American Naturalist, 15, 1018-20. Owen, R. ( 1845). Odontography; or a Treatise on the Comparative Anatomy
(1882). Two new genera of the Puerco Eocene. American Naturalist, 16, of the Teeth, etc. London, 655 pp.
417-8. Patterson, B. (1933). Anew species of the amblypod Titanoidesfrom western
(1897). The position of the Periptychidae. American Naturalist, 31, Colorado. American Journal of Science, 26, 415-25.
335-6. (1934). A contribution to the osteology of Titanoides and the relation-
Gazin, C. L. (1956). Paleocene mammalian faunas of the Bison basin ships of the Amblypoda. Proceedings of the American Philosophical
in south-central Wyoming. Smithsonian Miscellaneous Collections, Society, 73, 71-101.
131, 1-57. (1935). Second contribution to the osteology and affinities of the Pa-
Gidley, J. W. (1917). Notice of a new Paleocene mammal, a possible rel- leocene amblypod Titanoides. Proceedings of the American Philo-
ative of the titanotheres. Proceedings of the United States National sophical Society, 75, 143-62.
Museum, 52, 431-5. (1937). A new genus Barylambda, for Titanoides faberi, Paleocene
Gingerich, P. D., & Childress, C. G. (1983). Barylambda churchilli, a new amblypod. Field Museum of Natural History Geological Series, 6,
species of Pantolambdidae (Mammalia, Pantodonta) from the late 229-31.
Paleocene of western North America. University of Michigan Pale- (1939). New Pantodonta and Dinocerata from the upper Paleocene of
ontological Contributions, 26, 141-55. Colorado. Field Museum of Natural History Geological Series, 6,
Hebert, E. (1856). Recherches sur le faune des premiers sediments Terti- 351-84.
aires Parisiens (mammiferes pachydermes du genre Coryphodon). Patterson, B., & Simons, E. L. (1958). A new barylambdid pantodont from
Anna/es des Sciences Naturelles, 6, 87-136. the late Paleocene. Breviora, 93, 1-8.
Holtzman, R. (1978). Late Paleocene mammals of the Tongue River Forma- Rose, K. D., & Krause, D. W. (1982). Cyriacotheriidae, a new family of
tion, western North Dakota. North Dakota Geological Survey Report early Tertiary pantodonts from western North America. Proceedings
of Investigation, 65, 1-88. of the American Philosophical Society, 126, 26-50.
Huang, X. (1977'. Archaeolambda fossils from Anhui. Vertebrata PalAsiat- Schiebout, J. A. (1974). Vertebrate paleontology and paleoecology of Pa-
ica, 15, 249-60. leocene Black Peaks Formation, Big Bend National Park, Texas.
Jepsen, G. L. (1930). New vertebrate fossils from the lower Eocene of the Bulletin of the Texas Memorial Museum, 24, 1-88.
Bighorn basin, Wyoming. Proceedings of the American Philosophi- Schoch, R. M., & Lucas, S. G. (1985). The phylogeny and classification
cal Society, 69, 117-31. of the Dinocerata (Mammalia, Eutheria). Bulletin of the Geological
Lucas, S. G. (1982). The phylogeny and composition of the order Pantodonta Institutions of the University ofUppsala, 11, 31-58.
(Mammalia, Eutheria). Proceedings of the Third North American Simons, E. L. (1960). The Paleocene Pantodonta. Transactions of the Amer-
Paleontological Convention, 2, 337-42. ican Philosophical Society. new series, 50, 1-99.
(l 984a). Systematics, Biostratigraphy and Evolution of Early Cenozoic Simpson, G. G. (1929). A new Paleocene uintathere and molar evolution in
Coryphodon (Mammalia, Pantodonta). Unpublished Ph.D. Thesis. the Amblypoda. American Museum Novitates, 387, 1-19.
New Haven: Yale University. (1937). The Fort Union of the Crazy Mountain Field and its mammalian
(1984b). Taxonomic status of Titanoides simpsoni Simons, 1960 fauna. United States National Museum Bulletin, 169, 1-287.
and some observations on Pantolambda Cope, 1882 (Mammalia, (1945). The principles of classification and a classification of mammals.
Pantodonta). New Mexico Journal of Science, 24, 46-52. Bulletin of the American Museum of Natural History, 85, 1-350.
(1984c). Synopsis of the species of Coryphodon (Mammalia, Sloan, R. E. (1987). Paleocene and latest Cretaceous mammal ages,
Pantodonta). New Mexico Journal of Science, 24, 33-42. biozones, magnetozones, rates of sedimentation, and evolu-
(1986). Pyrothere systematics and a Caribbean route for land-mammal tion. Geological Society of America Special Paper, 209, 165-
dispersal during the Paleocene. Revista Geologia de America Cen- 200.
tral, 5, 1-35. Szalay, F. S. (1977). Phylogenetic relationships and a classification of the
Marsh, 0. C. ( 1873 ). On the gigantic fossil mammals of the order Dinocerata. eutherian Mammalia. In Major Patterns in Vertebrate Evolution, eds.
American Journal of Science, 3rd series, 5, 117-22. M. K. Hecht, P. C. Goody, & B. M. Hecht, pp. 315-74. New York:
(1884). Dinocerata. A monograph of an extinct order of gigantic mam- Plenum.
mals. United States Geological Survey Monographs, 10, 1-237. Van Valen, L. (1978). The beginning of the age of mammals. Evolutionary
Marshall, L. G., & Muizon, C. de (1988). The dawn of the age of mammals Theory, 4, 45-80.
in South America. National Geographic Research, 4, 23-55. (1988). Paleocene dinosaurs or Cretaceous ungulates in South America?
Marshall, L. G., Muizon, C. de, Gayet, M., Lavenu., A., & Sige, B. (1985). The Evolutionary Monographs, 10, 1-79.
Rosetta Stone for mammalian evolution in South America. National Wheeler, W. H. (1961). Revision of the uintatheres. Peabody Museum of
Geographic Research, 1, 274-88. Natural History Yale University Bulletin, 14, 1-93.
Matthew, W. D. (1897). A revision of the Puerco fauna. Bulletin of the Wood, H. E., II (1923). The problem of the uintathere molars. Bulletin of the
American Museum of Natural History, 9, 259-323. American Museum of Natural History, 48, 599-604.
19 Dinocerata
SPENCER G. LUCAS and ROBERT M. SCHOCH

INTRODUCTION

The Dinocerata (uintatheres) were large mammals known from fos-

sils collected in Paleocene-Eocene continental deposits in Asia and
the western United States. They left no known descendants. Uin-
tatheres were relatively large (about 175-4,500 kg) quadrupedal
herbivores that first appeared in North America during the late Pa-
leocene (Tiffanian) and became extinct during the middle Eocene
(Uintan). Although uintatheres were never very speciose, they in-
clude some of the first truly large terrestrial herbivores of the Mam-
malia. They also include some of the most distinctive, ergo grotesque,
mammals - the large, homed uintatheres of the middle Eocene (see
Figure 19.1).
In this chapter we focus on the North American uintatheres, al-
though discussion of the phylogeny and evolution of the Dinocerata
necessitates some attention to the Asian genera. Monographic stud-
ies of the Dinocerata are those of Marsh (1885a, 1885b), Flerov
(1957, 1967), and Wheeler (1961); Schoch and Lucas (1985) pre-
sented a genus-level taxonomic revision, phylogeny, and classifica-
tion of the Dinocerata that is followed here.

DEFINING FEATURES OF THE

ORDER DINOCERATA

Moderately large to very large (skull length ranges from

30-95 cm), eutherian mammals, characterized by Pl/pl extremely
small or absent; P2/p2 submolariform; P3-4/p3-4 molariform; Figure 19.1. Restoration of Uintatherium, by Brian Regal.
p2-m3 with metastylids; p3-m3 metalophids oriented obliquely
and inclined posteriorly; p3-m3 with large, anteriorly positioned
and isolated hypoconids and hypoconid crests; rnl-3 with entoconid relatives, the Pyrotheria (Lucas, 1986), a number of other features
crests and hypoconulid crest, except in derived uintatheres; infra- can be considered typical of uintatheres. These include a "complete"
mandibular flange present in most (presumed) males (see Figures postcranial skeleton (no elements lost or fused), an "amblypod" tar-
19.3 and 19.4). sus and carpus, and the presence of cranial protuberances of some
In addition to these features, which represent a suite of shared de- sort in most members of the order. The osteology of the Dinocerata
rived character states of the Dinocerata with respect to their nearest is best described in Marsh (1885b) and Flerov (1957, 1967).

284
 Dinocerata 285

SYSTEMATICS People's Republic of China and the Mongolian People's Republic

(Prodinoceras is also known from North America); Gobiatherium
Osborn and Granger, 1932, from the middle Eocene of the People's
SUPRAORDINAL
Republic of China and Kirgiziya; and Uintatherium Leidy, 1872,
Tong and Lucas (1982), Schoch and Lucas (1985), and from the middle Eocene of the People's Republic of China (Uin-
Lucas ( 1986) advocated the hypothesis that Dinocerata and Pyrothe- tatherium is also known from North America).
ria (including Xenungulata) form a clade (Uintatheriamorpha of Prodinoceras Matthew, Granger, and Simpson, 1929 (= Proba-
Schoch and Lucas, 1985) that shares a close common ancestry with a thyopsis Simpson, 1929, Bathyopsoides Patterson, 1939, Mongolo-
Pseudictops-1ike anagalid. This hypothesis removes uintatheres and therium Flerov, 1952, Prouintatherium Dorr, 1958, Jiaoluothe-
pyrotheres from the Paenungulata or Ungulata, where they were long rium Tong, 1978,HouyanotheriumTong, 1978,PyrodonZhai, 1978,
placed. It is tantamount to identifying uintatheres as giant homed Phenaceras Tong, 1979, Ganatherium Tong, 1979; see Schoch and
bunnies if Lagomorpha and Pseudictops are closely related. Lucas, 1985, for justification of these synonymies) the most primi-
tive uintathere and the primitive sister taxon of all other uintatheres.
The remaining genera ofuintatheres (namely, the higher uintatheres:
INFRA ORDINAL
Gobiatherium, Bathyopsis, Uintatherium, Tetheopsis, and Eobasile-
A hypothesis of the phylogenetic relationships of the genera of us) are united in the Uintatheriidae. Uintatheriids are easily distin-
Dinocerata, as revised by Schoch and Lucas (1985), is shown in guished from Prodinoceras by their larger size, lack of upper in-
Figure 19.2. Genera known from outside of North America and cisors, generally graviportal limb structure, development of cranial
included in the cladogram are Prodinoceras Matthew, Granger, protuberances and concomitant changes of cranial structure, as well
and Simpson, 1929, from the late Paleocene-early Eocene of the as some modifications of the cheek teeth. Gobiatherium represents

Ill
E Ill
::i
E ::i
·;: "ilj ..!!!
Ill
~
::i
·;: ·; Q)
.c;
Q.
0
"ii;
Ill
Q) Q) Q. ..... Q) .Q
(.)
0
.c;
.....
0
::... .sr:: .c; 0
r::
=s ·-
Ill
.Q
...
.c;
Ill :s ~
UJ
7
...
0
Q..
0
(!)
o:i

3
6

Figure 19.2. Interrelationships of the genera of Dinocerata. Character states that correspond to the numbered node points are as follows: (I) P2/p2
submolariform; P3-4/p3-4 fully molariforrn; p2-m3 with metastylids; p3-m3 metalophids oriented obliquely and inclined posteriorly; p3-m3 with
large, anteriorly positioned and isolated hypoconids and hypoconid crests; p3-m3 entoconid crests and hypoconulid crests present on lower molars
(lost in derived dinoceratans); inframandibular flange present in most males. (2) UINTATHERIIDAE: Relatively large size; upper incisors absent; p3-
m2 paraconids and paracristids greatly reduced; lower molars relatively short with entoconid crests and hypoconulid crests greatly reduced; lower
incisors bilobed; parasagittal crests present; astragalus lacks a neck; some form of cranial protuberance or inflation of cranial bone in the nasal re-
gion. (3) UINTATHERIINAE: Upper canines absent and lower canines greatly reduced; P2 with a prominent cingular cusp on its anterolabial comer;
skull long, thin, and shallow with a flat sagittal region; nasals strongly arched with a bony septum connecting their anterior ends to the premaxillae;
zygomatic arches broad and flare outward in the region of the glenoid fossae; lower jaw long and shallow with a tall ascending ramus and no inframandibu-
lar flange; metacarpals relatively long and slender. (4) Cranial horns present; Ml-3 with distinct hypocones. (5) Cranial horns large and deep basin present
between the temporal crests and anterior to the occipital crest; lower canine incisiform; p3-m3 essentially lack paracristids; p3-m3 entoconids rarely distinct
and entoconid and hypoconid crests absent; skeleton large and fully graviportal; tail short. (6) Skull long and narrow; portion of skull anterior to maxillary
horns relatively elongate; parietal horns relatively close to occiput. (7) Portion of skull anterior to maxillary horn extremely elongate; maxillary horns above
the premolars.
 Spencer G. Lucas and Robert M. Schoch 286

an aberrant Asian side branch of the uintatheriids, characterized by INCLUDED NORTH AMERICAN GENERA IN THE
its lack of upper canines, reduction of the lower canines, and thin, ORDER DINOCERATA
shallow skull with a strongly inflated nasal region. The remain-
ing uintatheriids, the Uintatheriinae, are distinguished from Gob- The locality numbers listed for each genus refer to the list of uni-
iatherium by their cranial horns and distinct upper molar hypocones. fied localities in Appendix I. Parentheses around the locality (e.g.,
Bathyopsis is the most primitive uintatheriine, and we consider it the [CP101]) mean the taxon in question at that locality is cited as an
plesiomorphic sister taxon of the three higher uintatheriine genera. "aff." or "cf." the taxon in question. The acronyms for museum
These three genera are distinguished from Bathyopsis by their larger collections are listed in Appendix III.
size, large paired maxillary and parietal horns, deeply basined skull
roof and various modifications of the cheek teeth that, among other PRODINOCERATIDAE
things, produce nearly bilophodont lower molars. Among the higher
uintatheriines, we consider Uintatherium the plesiomorphic sister Characteristics: Same as for Prodinoceras (see below).
taxon of a clade uniting Tetheopsis and Eobasileus. This clade is
characterized by long and narrow (dolicocephalic) crania, a length- Prodinoceras Matthew, Granger, and Simpson, 1929
ening of the portion of the skull anterior to the maxillary horns, and (synonyms: Probathyopsis Simpson, 1929; Bathyopsoides
the positioning of the parietal horns close to the occiput. Eobasileus Patterson, 1939)
is readily distinguished from Tetheopsis as the most derived uin- Type species: Prodinoceras martyr Matthew, Granger, and
tatheriine by the extreme elongation of the portion of the skull an- Simpson, 1929.
terior to the maxillary horns and the location of these horns above Type specimen: AMNH 21714, crushed palate and lower
the premolars. jaws.

:',',::··· --- __ _
<:.::,~)~:~

c E

Figure 19.3. Osteology of Prodinoceras: A. Skeleton. B. Skull. C. Lower jaw. D. Upper cheek teeth. E. Lower cheek teeth. (A-C after Flerov, D after
Simpson, E after Tong.) Scale bars= 40 cm for A, 10 cm for B-C, and 1 cm for D-E.
 Dinocerata 287

Characteristics: Relatively small to medium-sized dinocer- ferences that he listed between the two species (larger size
atans (length of skull ranges from approximately 30-50 cm) of skull, wider occiput, and better developed maxillary
with a relatively nonmolariform P2; full set of upper and horns) arguably reflect sexual dimorphism (cf. Schoch and
lower incisors; entoconids moderately to very distinct on Lucas, 1985) and are not considered by us to justify sepa-
p3-m3; entoconid and hypoconid crests present on ml-3; ration at the specific level.
ml-3 relatively elongated; skull with single sagittal crest;
astragalus bears a short, but distinct, neck (see Figure 19.3). Uintatherium Leidy, 1872 (synonym: Titanotherium,
Average length ofm2: 15.0-20.0 mm. in part)
Included species: Prodinoceras (Figure 19.3) is in need of Type species: Uintatherium anceps (Marsh, 1871).
a taxonomic revision at the species level; four North Type specimen: YPM 11030, skull fragments, thoracic ver-
American species are currently considered valid (Rose, tebrae, and left tibia.
1981; Schoch and Lucas, 1985): P praecursor (Simpson, Characteristics: Large uintatheriids (skull 69-85 cm long)
1929) (= P. successor Jepsen, 1930, P hobackensis [Dorr, with a relatively broad skull with the parietal horns well
1958]) (known from localities CP13H, CPl 7A, CPl 7B, anterior of the occiput, and the maxillary horns above the
CP18B, CP20C, CP22C); P. lysitensis Kelley and Wood, canines such that the ratio of the interhorn distance to the
1954 (locality CP27B); P newbilli Patterson, 1939 (local- distance from the maxillary horns of the anterior tip of
ities CP62A, CP62B); P harrisorum (Patterson, 1939) (= the nasals ranges from 1.5-2.2, and the ratio of the inter-
Bathyopsoides harrisorum) (localities CP13H, CP62A). horn distance to the distance from the parietal horns to
Prodinoceras sp. is also known from localities CP13G, the occipital crest ranges from 1.2-2.3 (cf. Wheeler, 1961,
CP14E, CP17A, CP18A, (CP20A), CP24B. Figure 1).
Average length of m2: 28 mm.
Comments: For a different view of prodinoceratid taxonomy, Included species: U. anceps (Figure 19.4) only (known from
see Thewissen and Gingerich (1987). localities [CC3], CP6A, CP6B, CP34D, CP38A, CP38B,
CP38C, CP64II).
UINTATHERIIDAE
Comments: See Wheeler, 1961, p. 2728, for a complete syn-
Characteristics: Large uintatheres lacking upper incisors; onymy of this species, although we believe this synonymy
p3-m2 paraconids and paracristids greatly reduced; ml-3 relatively bears critical reevaluation.
short with entoconid crests and hypoconulid crests greatly reduced;
lower incisors bilobed; parasagittal crests present; some form of Tetheopsis Cope, 1885 (synonym: Loxolophodon, in part)
protuberance ~evelopment or inflation of the cranial bones in the Type species: Tetheopsis speirianus (Osborn, 1881).
nasal region; astragalus lacks a neck (see Figure 19.4). Type specimen: PU (now in YPM) 11079, skull.
Characteristics: Large uintatheriids (skull 75-95 cm long)
Bathyopsis Cope, 1881 with a long and narrow skull, in which the parietal horns
Type species: Bathyopsis fissidens Cope, 1881. are near the occiput and the maxillary horns are above
Type specimen: AMNH 4820, lower jaw. the diastemata, so that the ratio of the interhorn distance
Characteristics: Relatively small uintatheriids (skull length to the distance from the maxillary horns to the anterior
approximately 35-50 cm) with small horns and a vestigial tip of the nasals ranges from 1.2-1.8, and the ratio of the
pl (variable?). Bathyopsis is also distinguished by its rel- interhorn distance to the distance from the parietal horns to
atively high, but anteroposteriorly compressed, trigonids the occipital crest ranges from 2.1-3.5 (cf. Wheeler, 1961,
with extremely small paraconids and paracristids on ml-3; Figure 1).
ml-3 metastylids that are large, but closely appressed to Average length of m2: 31.0 mm.
the trigonids; relatively indistinct ml-3 entoconids and Included species: T. speirianus (known from localities
no hypoconulid crests; no sagittal crest, but rudimentary CP38B, CP38C); T. ingens (Marsh, 1885a) (locality
parasagittal crests and small parietal, frontal, and maxil- CP38C).
lary horns; cranial table slightly basined; skull relatively
shallow; and temporal fossa visible in dorsal view. Comments: Wheeler (1961) last revised Tetheopsis and rec-
Average length of m2: 21.0 mm. ognized two valid species.
Included species: B. fissidens Cope, 1881 (known from lo-
Eobasileus Cope, 1872b (synonym: Loxolophodon, in part)
calities SB22C, CP25H, CP27D, CP27E); B. middleswarti
Type species: E. comutus (Cope, 1872a).
Wheeler, 1961 (localities CP34A, CP34C).
Type specimen: AMNH 5040, skull, right scapula, assorted
Bathyopsis sp. is also known from localities SB22C,
vertebrae, a rib, pelvis, and right femur.
CP3 l C, CP31E.
Characteristics: Large uintatheriids (skull approximately
Comments: Wheeler ( 1961) recognized two species of Bathy- 85-95 cm long), with the maxillary horn located above
opsis, B. fissidens and B. middleswarti: However, the dif- the premolars such that the ratio of the interhorn distance
 Spencer G. Lucas and Robert M. Schoch 288

Figure 19.4. Osteology of Uinta-

therium (after Marsh): A. Skeleton. B.
E Skull. C. Lower jaw. D. Upper cheek
teeth. E. Lower cheek teeth of Uin-
tatherium. Scale bar = 20 cm for A,
10 cm for B-C, and I cm for D-E.

to the distance from the maxillary horn to the anterior tip INDETERMINATE UINTATHERES
of the nasals ranges from 0.8-0.9, and the ratio of the in-
Fragmentary remains ascribed to uintatheriids are known from lo-
terhorn distance to the distance from the parietal horn to
calities CC4 and CP36B.
the occipital crest ranges from 2.1-3.1.
Average length of m2: 33.0 mm.
Included species: E. comutus only (known from localities
CP6A, CP38C, CP38D, CP65). BIOLOGY AND EVOLUTIONARY PATTERNS

Comments: Wheeler (1961) last revised Eobasileus and rec- Prodinoceras is the most primitive uintathere, and its essentially
ognized only one valid species. simultaneous appearance in Asia and North America near the end
 Dinocerata 289

w
z L L.BI.
u
w E.BI
2.4
0
0 E Lat.Hp.
::::i L.Hp.
D.. L.E.Hp.
E.E.Hp.
L LCI.

E,CI.
11.0
w
z L,L.Ba.
w 12.6
0 E.l.Ba.
M
0 14.0

:!! E.Ba.

15,8
L.Hm.

17.5
E.Hm.
18.8
E l.L.Ar.
19.2

E.L.Ar.
23.0

w
zW L
LE.Ar.

0 27.7
0
CJ E.E,Ar.
::::i 29.4
0
WI.
E 31.9
o..
33.4
L.Ch.
34.5
L M.Ch.
35.5
E.Ch,
37.1

Ou

39.5

- .§...
L,Un.
41.3
w Cll
z E.Un. s
w
0 45.9
- .ll!
c::
0
W M
L.Br.
46.7
:s
M.Br.
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.

60,9
To.3 61.3
To.2
62.5 PRODINOCERATIDAE UINTATHERllDAE
To.1
E Pu.3
63.8
54.0
Pu.2 54.5
Pu.1 55.0
Pu.a 55.1

Bl.=Blancan, Hp.= Hemphlllian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordlan, Ar.= Arlkareean,
Wt. =Whltneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. = wasatchian,
Ck. = Clarkforklan, TI. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 19.5. Temporal ranges of North American dinoceratan genera.

 Spencer G. Lucas and Robert M. Schoch 290

of the Paleocene implies a land connection between the two con- REFERENCES
tinents at that time. A close phylogenetic relationship between the
Dinocerata and Asian Pseudictops (Schoch and Lucas, 1985), how- Cope, E. D. (1872a). Notice of proboscidians from the Eocene of southern
ever, suggests that uintathere origins may have been in Asia. A Wyoming. Proceedings of the American Philosophical Society, 12,
close relationship between the Dinocerata and the Pyrotheria (sensu 580.
(1872b). Notices of new Vertebrata from the upper waters ofBitterCreek,
Lucas, 1986) indicates a dispersal route for an uintathere-pyrothere Wyoming Territory. Proceedings of the American Philosophical So-
common ancestor between North and South America before the end ciety, 12, 487-8.
of the Paleocene. (1881 ). The Vertebrata of the Eocene of the Wind River Basin. American
By the end of the early Eocene, more advanced uintatheriid Naturalist, 5, 74-5.
(1885). The genera of the Dinocerata. American Naturalist, 19, 594.
dinoceratans had appeared in North America, arguably the descen-
Dorr, J. A., Jr. (1958). Prouintatherium: new uintathere genus, earliest
dants of prodinoceratids. Prodinoceratids were relatively small Eocene, Hoback Formation, Wyoming, and the phylogeny of the
(body mass approximately 175-300 kg), generalized, plantigrade Dinocerata. Journal of Paleontology, 32, 506--16.
quadrupeds that lacked cranial protuberances, large saberlike up- Flerov, K. K. (1952). Novye Dinocerata iz Mongolii (New Dinocerata from
per canines, and the fully herbivorous dental specializations of the Mongolia). Doklady Akademii Nauk SSSR, 86, 1029-32.
(1957). Dinotseraty Mongolii (The Dinocerata of Mongolia). Trudy Pa-
uintatheriids. The early and middle Eocene uintatheriines (namely
leontologicheskovo Instituta Akademia Nauk SSSR, 86, 1-82.
Uintatherium, Tetheopsis, and Eobasileus) were large herbivorous ( 1967). Dinocerata ofMongolia. Jerusalem: Israel Program for Scientific
(probably browsing) forms that may have filled an ecological niche Translations (English translation of Flerov, 1957).
in some ways analogous to that of some living rhinoceroses Jepsen, G. L. (1930). New vertebrate fossils from the lower Eocene of the
(Figure 19.1). Bighorn Basin, Wyoming. American Philosophical Society Proceed-
Uintatheriid evolution in North America, best displayed by these- ings, 69, 117-31.
Kelley, D.R., & Wood, A. E. (1954). The Eocene mammals from the Lysite
quence Bathyopsis-Uintatherium-Eobasileus, involved the follow- Member, Wind River Formation of Wyoming. Journal of Paleontol-
ing changes: ogy, 28, 337-66.
Leidy, J. (1872). On some new species of Mammalia from Wyoming. Pro-
1. Increase in body size to produce middle Eocene taxa the size ceedings of the Academy of Natural Sciences, Philadelphia, 1872,
of living rhinoceroses (estimated body mass of Uintatherium 167-9.
is 3,000-4,500 kg). Lucas, S. G. (1986). Pyrothere systematics and a Caribbean route for land
mammal dispersal during the Paleocene. Revista Geologica de Amer-
2. Development of a fully graviportal, digitigrade limb struc-
ica Central, 5, 1-35.
ture.
Marsh, 0. C. (1871). Notice of some new fossil mammals from the Tertiary
3. Evolution of cranial protuberances culminated by the pos- formation. American Journal of Science (3rd series), 2, 35-44.
session of paired maxillary and parietal horns in the derived (1885a). The gigantic mammals of the order Dinocerata. United States
uintatheriids. Concomitant changes in skull structure include Geological Survey Fifth Annual Report, 243-302.
the development of parasagittal crests, deep basining of the (1885b). Dinocerata, a monograph of an extinct order of gigantic mam-
mals. United States Geological Survey Monograph, 10, 1-237.
parietal region of the skull roof, narrowing and deepening
Matthew, W. D., Granger, W., & Simpson, G. G. (1929). Additions to the
of the palatal vault of the roof of the mouth, and relative fauna of the Gashato Formation of Mongolia. American Museum
telescoping of the skull (dolicocephaly). Novitates, 376, 1-12.
4. Loss of the upper incisors, relative increase in size of the Osborn, H. F. (1881). A memoir upon Loxolophodon and Uintatherium,
upper canines, and reduction of the lower canines such that two genera of the sub-order Dinocerata. Contributions from the
E. M. Museum of Geology and Archaeology of Princeton College, 1,
they become incisiform. It thus seems highly likely that the
5-44.
uintatheriids possessed a mobile upper lip analogous to that Osborn, H.F., & Granger, W. (1932). Coryphodonts and uintatheres from the
found in living rhinoceroses. Mongolian expedition of 1930. American Museum Novitates, 552,
5. Accentuation of the essentially bilophodont pattern of the 1-16.
cheek teeth, especially by the loss of oblique cristids in the Patterson, B. (1939). New Pantodonta and Dinocerata from the upper Pa-
leocene of western Colorado. Geological Series. Field Museum of
lower molars.
Natural History, 87, 1-9.
Rose, K. D. (1981). The Clarkforkian land-mammal age and mammalian
North American Eocene uintatheriids represent the end point of
fauna! composition across the Paleocene-Eocene boundary. Univer-
uintatheriid evolution. In Asia, the divergent genus Gobiatherium sity of Michigan Papers on Paleontology, 26, 1-197.
presents some parallels with the North American uintatheriids but Schoch, R. M., & Lucas, S. G. (1985). The phylogeny and classification
strongly suggests separation of North American and Asian uin- of the Dinocerata (Mammalia, Eutheria). Bulletin of the Geological
tatheriids during the early Eocene. The presence of Uintatherium Institutions of the University of Uppsala, 11, 31-58.
Thewissen, J. G. M., & Gingerich, P. D. (1987). Systematics and evolu-
in the middle Eocene of China suggests free passage of derived
tion of Probathyopsis (Mammalia, Dinocerata) from the late Pale-
uintatheriids between Asia and North America during the middle ocene and early Eocene of western North America. Contributions
Eocene. The extinction of the Dinocerata during the middle Eocene from the Museum of Paleontology. The University of Michigan, 27,
remains unexplained. 195-219.
 Dinocerata 291

Tong, Y. (1978). Late Paleocene mammals of the Turfan basin, Sinkiang. ings of the Third North American Paleontological Convention, 2,
Memoirs of the Institute of Vertebrate Paleontology and Paleoan- 551-6.
thropology Academia Sinica, 13, 82-101. Wheeler, W. H. (1961). Revision of the uintatheres. Yale University Peabody
(1979). Fossils of early Eocene Dinocerata from the Chijiang basin. Museum of Natural History Bulletin, 14, 1-93.
In Mesozoic and Cenozoic Redbeds in Southern China, pp. 395-9. Zhai, R. (1978). Two new early Eocene mammals from Sinkiang with re-
Beijing: Science Press. marks on the age of the Gashato Formation. Memoirs of the Insti-
Tong, Y., & Lucas, S. G. (1982). A review of Chinese uintatheres tute of Vertebrate Paleontology and Paleoanthropology, Academia
and the origin of the Dinocerata (Mammalia, Eutheria). Proceed- Sinica, 13, 102-16.
20 Archaic ungulates ("Condylarthra")

J. DA YID ARCHIBALD

INTRODUCTION

Archaic ungulates or "condylarths" are the most taxonomically di-

verse and numerically abundant eutherians in the early Tertiary of
North America. Through both the Puercan and Torrejonian Land
Mammal ages (early-mid Paleocene), they constituted more than B
half of the known species of mammals (Archibald, 1983). From the
end of the Torrejonian (mid-Paleocene), archaic ungulates dwin-
dled in taxonomic diversity, but not always numerical abundance,
until their disappearance in North America in the Duchesnean Land
Mammal age (late middle Eocene) (Figure 20.7). This is, of course, a
pseudo-extinction (Archibald, 1993) because archaic ungulates are
in reality an amalgam of species that gave rise to a variety of other ex-
tinct and extant mammals, which have here been termed ungulates.
Starting from their earliest appearance in the Late Cretaceous
of Asia, North America, and possible Europe, ungulatomorphs -
represented by "zhelestids" (Archibald, 1996b; Nessov, Archibald,
and Kielan-Jaworowska, 1997) - show tendencies toward lower
c
crowned, more bunodont dentitions than did their contemporary
eutherians, which usually maintained and sometimes accentuated
a sectorial slicing and piercing dentition. When true archaic un-
gulates appear in either the latest Cretaceous or earliest Paleocene
of North America, they continue to amplify the dental trends seen
in the "zhelestids." Little is known of the earliest ungulates be-
yond their dentition (Nessov and Kielan-J aworowska, 1991; Nessov,
Signogneau-Russell, and Russell, 1994; Nessov, Archibald, and
Kielan-Jaworowska, 1997). Much could probably be gleaned con-
cerning their diet from scanning electron microscopic studies of their
dental wear patterns, but for now the most that can be said is that
the earliest archaic ungulates were probably omnivorous, possibly \~ .
~~··~---
subsisting on soft plants and invertebrates, and perhaps fruits.
In 1881, E. D. Cope (188 ld) proposed the name Condylarthra as
a subunit of Perissodactyla to include archaic mammals that are now
allocated to Phenacodontidae, Periptychidae, and Hyopsodontidae.
Figure 20.1. Restorations of archaic ungulates (by Brian Regal):
In 1884 (Cope, 1884b), he removed Condylarthra from the Perisso-
A Mesonychid Mesonyx sp. (based on skeletal reconstruction of Carroll,
dactyla because of dissimilarities in pedal structure and expanded 1988, after Scott, 1888). B. Phenacodontid Meniscotherium chamense (based
the concept of Condylarthra to include additional phenacodontids on skeletal reconstruction of Gazin, 1965). C. Arctocyonid Chriacus sp.
(meniscotheres) and periptychids, and taxa that are now included in (after Rose, 1987).
292
 Archaic ungulates ("Condylarthra") 293

Arctocyonidae, Taeniodonta, and Artiodactyla. As Prothero, Man- homoplasies (convergences and reversals). This can confound eco-
ning, and Fischer (1988) have observed, with this expansion, Cope logical or adaptational assessments. Thus neither our understanding
established the traditional view of Condylarthra as the archetypal of the evolutionary relationships nor the biology of the organisms
taxon of ungulates. is adequately satisfied.
To Cope's list of condylarths have been appended a host of other Other authors (e.g., McKenna, 1975) have tended to disassemble
taxa including additional members of the taxa listed in the preced- Condylarthra because it does not constitute a monophyletic taxon
ing paragraph as well as Mesonychidae, Didolodontidae, Parox- (including the most recent common ancestor and all descendants)
yclaenidae, Tricuspidontidae, and Phenacolophidae (see Carroll, and have allocated the various subunits of the original order to the
1987, for a listing of most taxa). Further, other taxa that less com- taxa with which they share a presumably more recent common
monly have either been placed within or allied with Condylarthra ancestry. Condylarthra is then retained only for such taxa as the
most notably include Pantodonta, Taeniodonta, Tillodontia, Pen- Phenacodontidae, the family central to Cope's original concept of
tacodontidae, and Deltatherium. Prothero, Manning, and Fischer the Condylarthra.
( 198 8) have provided the most recent review of the history of Condy- Finally, there are those authors (e.g., Prothero, Manning, and
larthra as a taxonomic concept. Suffice it to note here that in the more Fischer, 1988) who have taken the final step in arguing that the
than 100 years since the proposal of this taxon, Condylarthra and concept of Condylarthra should be abandoned altogether, not only
the taxa included in it have gone through considerable modification. because it is nonmonophyletic, but also because of "all the erro-
Certainly one of the most fundamental questions is how Condy- neous connotations it has acquired." Several of the reasons that these
larthra should be utilized, if at all, in formal taxonomy. As Insec- authors note for this suggestion are as follows: (1) Use of such a
tivora has served in the past as the wastebasket for many suppos- wastebasket taxon for these more archaic ungulates can obscure
edly primitive eutherians, Condylarthra has served as the dumping knowledge of more important derived ungulate characters; (2) the
ground for real or perceived so-called primitive ungulates. The re- emphasis on primitive condylarth or eutherian attributes masks im-
sult has been the recognition of a taxon whose members are linked portant divergences among early ungulate lineages; (3) there is un-
almost exclusively by symplesiomorphies (shared primitive char- certainty about what is meant by "condylarth"; and (4) as Prothero,
acter states) for early ungulates or even for Eutheria as a whole. Manning, and Fischer (1988, p. 228) note, "Despite claims to the
Furthermore, this taxon has also included more distantly related contrary (Van Valen, 1978), 'condylarths' are not a clear example
taxa, some of which not only share these same primitive eutherian of an 'adaptively unified group.' For example, the teeth of Arcto-
character states, but are also convergent in other characters found cyon suggest an omnivore, while Meniscotherium was apparently a
in ungulates. If one can eliminate the latter class of taxa, then the specialized herbivore that lived in herds."
problem becomes one of attempting to unify condylarths with their Another even more striking adaptive contrast is between the dimi-
respective sister taxa among both extinct and extant ungulates. This nutive insectivorous or herbivorous anisonchine periptychid Oxya-
is not a new att~mpt but dates to the time of Cope and before. codon** (see Systematics section for symbol explanation), with a
Within the past twenty years there has been renewed interest skull measuring less than 10 cm, and the immense carnivorous Asian
in elucidating the higher relationships within the fossil and living mesonychid Andrewsarchus, with a skull measuring over 80 cm.
Eutheria (e.g., McKenna, 1975; Szalay, 1977; Novacek, 1986; No- Thus I fully agree with Prothero, Manning, and Fischer (1988) that
vacek and Wyss, 1986). One result has been a corollary interest in Condylarthra should be abandoned because it not only misrepre-
attempting to relate condylarths to other eutherians (see Prothero, sents our understanding of phylogenetic history, but it also obscures
Manning, and Fischer, 1988, for a fuller discussion). Although some our attempts to understand the paleoecology of these organisms.
authors are not explicit in indicating included taxa or in providing With abandonment of the term and concept "Condylarthra," the
uniting characters, a consensus does appear that condylarths are part problem arises of how to refer to this assortment of tax a. This should
of a larger taxon, Ungulata. This is, however, further complicated really only present a problem for the nonspecialist. Two possibil-
by arguments over the monophyly of Ungulata. This issue is treated ities are to retain the term Condylarthra (or condylarthrans) in an
briefly later. There is general agreement that condylarths are ungu- informal manner only (as in vertebrates and non-vertebrates or Cali-
lates, but considerable divergence of opinion exists as to whether to fornians and non-Californians), or in the form "Condylarthra," with
recognize the Condylarthra formally (usually as an order), to use it the quotation marks signifying a nonmonophyletic taxon. The sec-
in a modified form, or to abandon it altogether. ond convention (see Gauthier, Estes, and de Queiroz, 1988, for a
Formal recognition of Condylarthra as an order (e.g., Van Valen, discussion) conserves usage familiar to the nonspecialist but calls
1978) has been the more traditional view. The order has been viewed attention to the fact that this taxon is not monophyletic. Even with
as a grade of more or less equally primitive eutherians that con- quotation marks, the term retains so much historical baggage that
stituted the radicle from which most extant and extinct ungulates it seems more prudent to use it only informally, if at all. For this
arose. As Prothero, Manning, and Fischer (1988) discussed, for- reason, if a term is needed, I favor "archaic ungulate," where "ar-
mal recognition of such a supposedly "adaptively unified group" chaic" refers to the geochronologically older fossil taxa that are
or grade does not stand up to closer scrutiny. One should have a generally the more basally splitting clades within the presumably
working hypothesis of relationship or recency of common ancestry monophyletic taxon Ungulata. Archaic is not used here to indi-
before attempting to elucidate adaptational or ecological signifi- cate that the taxon possesses fewer derived character states be-
cance. If not, the result will most often be a taxon unified on the cause, as noted previously, some archaic ungulates such as Menis-
basis of symplesiomorphies (shared primitive character states) or cotherium possess a derived (selenodontlike) dentition. The idea of
 J. David Archibald 294

an "archaic ungulate" as utilized here has been separately and sim- Several authors have recently reviewed either the higher relation-
ilarly advocated by Prothero, Manning, and Fischer (1988). ships among all eutherians (Novacek, 1982, 1986, 1992; Novacek
and Wyss, 1986; McKenna, 1987; Wyss, Novacek, and McKenna,
1987) or for the ungulates (Cifelli, 1983; Prothero, Manning, and
DEFINING FEATURES OF ARCHAIC UNGULATES Fischer, 1988), and most of the following characters are from these
authors.
Because the archaic ungulates or "condylarths" do not share any
characters other than symplesiomorphies for a wider grouping of
CRANIAL
eutherians or synapomorphies for all ungulates, it is not possible
to provide any phylogenetically meaningful defining characters (or Recent and fossil ungulates display a tremendously varied and of-
diagnosis) for this group of mammals. This is further complicated ten bizarre cranial morphology. This variation includes not only
because individual monophyletic clades of archaic ungulates show a wide range of cranial protuberances evolved convergently within
either autapomorphous (unique) character states or character states and among various ungulate clades (artiodactyls, perissodactyls, and
suggesting alliance with other extinct and extant ungulates. The re- arsinotheres), but also a tremendous variation in cranial architecture,
sult is that the following comments on features of archaic ungulates especially when cetaceans are included among the ungulates, as war-
are more a review of the relatively few character states that help to ranted by mounting evidence from the fossil and molecular studies
diagnose all ungulates. Characters helpful in defining lower level (e.g., Novacek, 1982, 1992; McKenna, 1987; Prothero, Manning,
taxa are listed under Included Genera of North American Archaic and Fischer, 1988; Thewissen, 1994). In fact, cetaceans probably
Ungulates. share a more recent common ancestry with some extant terrestrial

F -----------

,&;~
~--j-

Figure 20.2. Cranial reconstructions and restorations: A. Protungulatum** donnae* (cranium conjectural, after Szalay, 1969). B. Hyopsodontid Hyopsodus
sp. (after Gregory, 1951). C. Periptychine periptychid Periptychus sp. (after Gregory, 1951). D. Phenacodontan Tetraclaenodon sp. (after Gregory, 1951). E.
Mesonychid Harpagolestes uintensis (after Szalay, 1969). F. ArctocyonidLoxolophus sp. (after Gregory, 1951 ). G. ArctocyonidArctocyonferox (after Gidley,
1919). H. Periptychine periptychid Ectoconus sp. (after Gregory, 1951).1. Phenacodontid Phenacodus primaevus (after Gregory, 1951). J. Phenacodontid
Meniscotherium sp. (after Gregory, 1951). Scales are 1 cm. unless otherwise indicated. Illustration by Brian Regal.
 Archaic ungulates ("Condylarthra") 295

ungulates than some extant terrestrial ungulates share with each synapomorphy of ungulates that these authors noted is from the
other. Thus finding cranial (or other) synapomorphies that all ungu- work of Novacek (1986), who indicated that all ungulates lack a
lates, or even all their more distant ancestors possess, is a formidable mastoid foramen except for some artiodactyls that apparently have
task. Figure 20.2 illustrates some of the variety of cranial morphol- reacquired it. Novacek (1986) indicated that perissodactyls possess
ogy seen in archaic ungulates. a mastoid foramen; Prothero, Manning, and Fischer (1988) indicate
Prothero, Manning, and Fischer (1988) indicate that according to that it reappears only in a few palaeotheres. Didelphids have con-
Wible (1986, 1987), in ungulates the superior ramus of the stapedial vergently lost this foramen (Novacek, 1986). Novacek (1986) listed
artery is no longer situated between the petrosal and squamosal, but several other characters that, although present in most ungulates,
is absent (most ungulates) or passes through the tegmen tympani do not appear to be unique for this group, including "weak occipi-
of the petrosal (Arctocyon, Arctocyonides, and Pleuraspidotherium, tal exposure of the mastoid; subarcuate fossa shallow; squamosal-
Russell, 1964). The composition of the auditory bulla may also sinus canals vestigial or absent; mandibular condyle relatively high"
be of some use in uniting some or all of the ungulates. Accord- (Novacek, 1986, p. 97).
ing to Prothero, Manning, and Fischer (1988), all living ungulates
except hyraxes and some perissodactyls possess an auditory bulla
DENTAL
composed of the ectotympanic. There is a problem, as these au-
thors acknowledge, in that many fossil ungulates appear to lack (or Prothero, Manning, and Fischer (1988), and Prothero (1993), cited
have lost during fossilization) an ossified auditory bulla. The third three dental characters that are presumably synapomorphic for

A
B

Figure 20.3. Dentitions of A. Protungulatum** donnae*, right P3-4, Ml-3 and p2-4, ml-3 (after Carroll, 1987, from Sloan and Van Valen, 1965).
B. Conacodontine periptychid Conacodon entoconus, left P2-4, Ml-3 and p2-4, ml-3 (after Matthew, 1937). C. Periptychine periptychid Periptychus
carinidens*, left C, Pl alveolus, P2-3, MJ-4 and p2-4, ml-3 (after Matthew, 1937). D. Hyopsodontid Hyopsodus powellianus, left P2-4, Ml-3 and
right p3-4, pl-3 (after Gregory, 1951). E. Hyopsodontid Haplaletes disceptatrix, right P2-4, Ml-3 and p2-4, ml-3 (Simpson, 1937). Scales are 1 cm.
Illustration by Brian Regal.
 J. David Archibald 296

A
,,...,...._
~

c~-.,.
• ,. _, ii. ,o
• D 0 ~-~

Figure 20.4. Dentitions of A. Hapalodectid Hapalodectes leptognathus, Bl, Cl, P4, M3 and i2, cl, p4, m3 (Szalay, 1969). B. Cetan Triisodonantiquus, right
P2-4, Ml-3 and pl-4, ml-3 (Matthew, 1937). C. Phenacodontid Ectocion osbomianus, right P3-4, Ml-3 and pl-4, ml-3 (Rose, 1981). D. Mioclaenid
Mioclaenus turgidus, palate with P2-4, Ml-3 and left pl-4, ml-3 (Matthew, 1937). Scales are 1 cm. Illustration by Brian Regal.

ungulates: teeth more bunodont with relatively low cusp relief (as POST CRANIAL
compared to Purgatorius, Gypsonictops, and other Late Cretaceous
eutherians); lower molar trigonids shortened anteroposteriorly; and The single postcranial synapomorphy for all ungulates noted by
m3 with hypoconulid large and posteriorly projecting. Nessov, Prothero, Manning, and Fischer (1988) was that the astragalus has
Archibald, and Kielan-Jaworowska (1997) indicated that these char- a short, more robust head, presumably in comparison to other early
acters states are not unique to Ungulata; the first two are also found Tertiary eutherians. As an additional "defining" character, Novacek
in the mostly Late Cretaceous "zhelestid" ungulatomorphs; the third ( 1986) noted that the terminal phalanges or unguals are elongated but
character state appears in at least one zhelestid. not fissured, becoming variously modified in later lineages. In ad-
Novacek (1982, 1986) and Novacek and Wyss (1986) mentioned dition, some earlier ungulates such as Arctocyon appear to have fis-
the first and third of these synapomorphies as well as that the sured unguals (Prothero, written comm., 1989). The idea ofhooflike
hypocone on the upper molars is consistently well developed and unguals accounts, of course, for the name "Ungulata." Prothero,
the lower molars have a swollen metaconid. I question the first of Manning, and Fischer (1988) discussed a variety of other postcranial
these two characters. Although early artiodactyls probably had a true features that are shared (presumably in a derived fashion) by many
hypocone, later artiodactyls developed a "hypocone" presumably but not all ungulates: clavicle lost in most ungulates except archaic
via the metaconule. The acquisition of a hypocone may well have artiodactyls, arctocyonids, and hyopsodontids (these authors did not
occurred several times within Ungulata as suggested by Prothero, separate mioclaenids and hyopsodontids as I do here); humerus with
Manning, and Fischer (1988). reduced acromion and coracoid process in mesonychids, whales,
Nessov, Archibald, and Kielan-Jaworowska (1997) recognized and all "higher" ungulates (except meniscotheres); humerus with
the following seven dental and one dentary synapomorphies for reduced deltopectoral crest in most higher ungulates; ulna with pos-
Ungulata, although these character states are manifest in combi- terior border straight or curved posteriorly and flexed away from the
nation only in the earliest ungulates: molar conules with little or limb axis in mesonychids and "higher" ungulates rather than with a
no internal wings; metacingulum formed by postmetaconular crista curved and posteriorly convex ventral border and an olecranon that
continuing on to the metastylar region; parastylar region reduced flexes toward the limb axis; femur with a well-developed, distally
with one cusp; premolars four or fewer in number, although sire- shifted third trochanter in most "advanced" ungulates (except in
nians have five upper premolars (Gingerich et al., 1994), which is artiodactyls, hyraxes, and some gravigrade taxa). Figure 20.5 illus-
regarded as a reversal; upper molars rectangular in occlusal view; trates skeletal reconstructions of certain taxa of archaic ungulates.
parastylar groove (and lobe) reduced; conules closer to the proto- Another historically important character has been the presence of
cone than to the midposition of the crown; and mandibular condy le a serial rather than an alternating pattern of carpels. Novacek and
dorsal to the occlusal surface of the tooth row. Figures 20.3 and Wyss (1986) used the serial ("in-line") arrangement of the carpal
20.4 illustrate the dental morphology of certain taxa of archaic elements and the consequent loss of contact between the lunar and
ungulates. unciform to unite hyracoids, proboscideans, and sirenians. Fischer
 Archaic ungulates ("Condylarthra") 297

Figure 20.5. Skeletal reconstructions of archaic ungulates: A. Periptychine periptychid Ectoconus sp. (after Gregory, 1951). B. Hyopsodontid Hyopsodus
paulus (after Gazin, 1968). C. Arctocyonid Chriacus sp. (after Rose, 1987). D. Phenacodontid Phenacodus primaevus (after Gregory, 1951). E. Mesonychid
Mesonyx sp. (after Carroll, 1987, from Scott, 1888). F. Phenacodontid Meniscotherium chamense (after Gazin, 1965). Scales are 5 cm. Illustration by
Brian Regal.

( 1989) was less certain whether the serial arrangement in the carpus families and genera a more traditional "evolutionary" approach is
(his taxeopody) can be used to align these three taxa. He favored the taken. The reason for this is that several recent papers have treated
idea that the serial arrangement in the carpus is a derived character the higher level relationships among various ungulates cladistically
state for the common ancestor of hyracoids, proboscideans, sireni- (see references in previous section), and one in particular, Prothero,
ans, phenacodontids, perissodactyls, plus some others. According Manning, and Fischer (1988), attempted to include many genera
to this thesis, the alternating carpal arrangement in perissodactyls is of archaic ungulates in the cladistic analysis. In contrast, relatively
derived relative to the serial condition. few recent papers have attempted a cladistic analysis within more
exclusive clades (families and genera) of North American archaic
ungulates (some exceptions: Rigby, 1980, 1981; Archibald, Rigby,
SYSTEMATICS and Robison, 1983; Archibald, Schoch, and Rigby, 1983). The result
of this discrepancy in approach is that at higher levels of analysis
HIGHER LEVEL RELATIONSHIPS readers can assume that noted characters are hypothesized synapo-
The relationships of the various groups of archaic ungulates within morphies for the included taxa, whereas at lower levels they cannot:
the Ungulata is addressed in Chapter 15 of this volume on archaic Caveat emptor. Many portions of the cladogram in Figure 20.6 fol-
ungulates and ungulatelike mammals. low Prothero, Manning, and Fischer (1988). Again by contrast, the
characteristics for the 7 family levels and 69 generic levels pre-
sented here are based on work ranging from over 100 years old
LOWER LEVEL RELATIONSHIPS
to very recent cladistic studies. Because of this situation, readers
The systematic approach in this chapter is not theoretically uniform. may encounter discrepancies or contradictions among various taxa.
For the higher level relationships among the archaic ungulates, a This dilemma is inevitable in a synthetic study like this that relies
cladistic approach has been attempted, whereas within the various on other works of varying antiquity, extensiveness, and systematic
 J. David Archibald 298

Alostera plus other "ZHELESTIDAE" Figure 20.6. Interrelationships of

ARTIODACTYLA North American archaic ungulates
Protungulatum** ("condylarths") and other selected
Carcinodon
Oxytomodon
ungulates. An asterisk indicates
Desmatoclaenus a metataxon and quotation marks
Princetonia l>
Prothryptacodon p; indicate that the taxon is nonmono-
Chriacus d phyletic. Hatched lines connected to a
Deuterogonodon ~ node indicate that the taxon is incertae
Loxo/ophus z
Mimotricentes ~ sedis at that node; a hatched line and
Lambertocyon m a question mark not connected but
Mentoclaenodon
Anacodon directed toward a node indicate that
Arctocyon the taxon is questionably included in
Co/poc/aenus
Thryptacodon
the tax on at that node. Numbers on the
Aletodon cladogram refer to the following list of

]I
Dorrafetes
Hap/a/etes
synapomorphies for each node. Some
Haplomy/us portions of the cladogram (especially
Hyopsodus for Artiodactyla, Arctocyonidae,
Litomy/us
Oxyprimus parts of Cele, Phenacodontidae, and
Utemylus Paenungulata) and synapomorphies
Bubogonia
Choeroclaenus 1-11, 24-30, 37-42 are taken
Ef/ipsodon largely from Prothero, Manning,
Lita/etes
Mioclaenus and Fischer (1988) and references
Promioclaenus therein. Synapomorphies 31-36 are
Protosetene
Tiznatzinia
from Zhou et al. (1995). There has
Aphe/iscus been no attempt to reconfirm them
Phenacodaptes
in the present study. The remaining
Escatepos
Maiorana portions of the cladogram and
Tinuvie/ the synapomorphies are from the
"Mimatuta"
Hemithlaeus present study and references therein.
Ectoconus "m::::c Synapomorphies for terminal taxa
>"Carsioptychus" ~n'ii
Periptychus (genera and included species) are not
Gilfisonchus ~ given, and the question of monophyly
>"Anisonchus" ~
for many of them has never been
Haploconus
Conacodon examined. The synapomorphies are
Oxyacodon**
Platymastus
as follows: (I) UNGULATA: superior
Wyo lest es ramus of stapedial artery shifted to
Baioconodon petrosal or lost; mastoid foramen
~~;l~~::n ~ lost; bulla (if present) composed
Oxycfaenus ~ of ectotympanic; teeth relatively
Goniacodon 5
Eoconodon ~ bunodont with low cusp relief; lower
Triisodon g molar trigonids shortened anteropos-
Microclaenodon
:J m~ S n
m ~
teriorly·, M3 with hypoconulid large,
Hapalodectes
Dissacus
Ankalagon
J:;::
!)l
~
c:
posteriorly projecting; astragalus
with shorter, more robust head.
Pachyaena ~ \;
Synoplotherium r; :;;! (2) ARTIODACTYLA: lower molar

Jn
Harpagolestes i1i trigonids very narrow, with paraconid
Mesonyx l>
GET ACEA m and metaconid closely appressed;
Tetrac/aenodon ~- alisphenoid canal lost; lacrimal with
Phenacodus
Copecion ~
m ffi~ -1s;: enlarged pars facialis; orbitosphenoid
Ectocion 8 8 2l enlarged, separating frontal sinus
Meniscotherium 8
z
o
0
~
o from alisphenoid; pes paraxonic, wit
· h
PAENUNGULATA :::l ~ )>
~ )'; double ginglymi, enlarged third and
m fourth metapodials, and symmetri-
cally reduced second and fifth metapodials. (N .B.: Mounting evidence [see Novacek, 1992] suggests Artiodactyla is the sister tax on to whales, which is the
other alternative given here.) (3) Hypocone added to upper molars; femur with enlarged third trochanter enlarged and distally shifted; tympanic aperture of
canal is facialis anterior to fenestra oval is. (4) ARCTOCYONIDAE: Upper molars with complete lingual cingulum (N .B.: This synapomorphy of Prothero,
Manning, and Fischer (1988) may not be valid for all taxa included the Arctocyonidae in the present study). (5) M3 with enlarged anterolingual cingulum;
size larger(?). (6) Parastyles and metastyles reduced (on molars?). (7) M3 with postcingular cusp. (8) Upper molars square .. (9) Enamel crenulate. (10~
Labial cingulum rounded; canines enlarged and saber like, serrated posteriorly; mandibular condyle low; p3-4 serrated postenorly. (11) Tegmen tympam
inflated. (12) HYOPSODONTIDAE: p4 molarized; molar paraconid labially shifted and reduced. (13) Premolar inflation. (14) MIOCLAENIDAE:
Premolars and molars simplified; M3/m3 reduced. (15) P4/p4 large and distinctive. (16) PERIPTYCIDDAE: Protocone (and sometimes hypocon~)
base with lingual expansion and apex shifted labially; lower molars with internal (and sometimes external) cingulids; molar cusps appressed; talomd
length shortened relative to width; molar paraconid on or near midline or absent. (17) Preceding five synapomorphies further developed. ( 18) Upper mo-
lars with pericone; P4 protostyle very reduced or absent. (19) Size increase. (20) Premolars with very well developed ribbing. (21) Posterior premolars with
 Archaic ungulates ("Condylarthra") 299

Figure 20.6 (Cont.). either large, distinct, almost columnar paraconid or paraconid lost apparently through fusion to trigonid. (22) Upper molar width
reduced (for Haploconus and some "Anisonchus"). (23) Ml-3 with hypocone large and situated very lingually; P3 protocone absent (convergent with
Haploconus); protocone precingulum encircled or nearly encircled by precingulum, which contacts hypocone. (24) CETUNGULATA: Mastoid with weak
occipital exposure; stapedial artery lost; posterior lacerate foramen large and coalesced with cochlear fissure; postglenoid foramen small or absent; acromion
and coracoid processes on scapula reduced; clavicle lost; ulna with ventral border straight or concave; humerus with deltoid crest reduced. (25) CETE:
protocones blunt, conical. (26) M3 small with small hypocone; p4 with posterolingual accessory cusp; M3 parastyle small and paracone large; p4 paraconid
deflected posteriorly. (27) M3 metacone small. (28) Ml-2 parastyle lost. (29) Paracone larger; all cusps tall and conical; overall size increase. (30) Trigonid
high; entoconid very reduced. (31) Lower cheek teeth narrow; Ml-2 parastyles long; p2-3 diastema (some taxa). (32) p4 taloid trenchant and square shaped.
(33) mi metaconid reduced or absent. (34) P3 two rooted; m2 metaconid reduced or absent. (35) M3 very small. (36) M3 absent; P4-M3 labial cingula
reduced or absent. (3 7) CETACEA: incisors all parallel with cheek teeth; medial lambdoid crest semicircular; nasals retracted; protocones small; accessory
cusps large. (38) P4 protocone area broadened. (39) P4 protocone and metacone enlarged; aquaeductus cochleae ventral, slitlike; humerus with medial
epicondyle and entepicondylar foramen reduced; astragalar/cuboid contact reduced. (40) PHENACODONTIDAE: petrosal epitympanic sinus posterior;
upper molars with mesostyle. (41) Metaloph added between hypocone and metaconule. (42) PAENUNGULATA: Astragalus with navicular facet flattened;
thoracic vertebrae 19 or more; M3/m3 larger than M2/m2.

theory. Furthermore, the analysis presented here is not as rigorous of other families and the wealth of native South American ungu-
as one might like because it has not been subjected to any computer lates or meridiungulates have been excluded. For the latter group
programs (e.g., MacClade, PAUP) that might find more parsimo- in particular, Cifelli (1983) has most recently considered possible
nious branching sequences. Finally, the large numbers of taxa ver- relationships.
sus the relatively few numbers of characters allows only the most In 1996, Archibald (1996b) named the taxon Ungulatomorpha to
preliminary assessment of relationships. encompass not only Ungulata as used in this chapter but also the
For most genera included in this study, there are no cladistic sister tax on "Zhelestidae" best known from the Late Cretaceous (Co-
diagnoses (or definitions). Hence in the cladogram (Figure 20.6) niacian) of Asia. These taxa, as well as other "zhelestids" from North
there are no synapomorphies for the species included in these gen- America and possibly Europe, were discussed in detail by Nessov,
era. In at least three instances, "Mimatuta," "Carsioptychus," and Archibald, and Kielan-J aworowska (1997). Of note here is that one
"Anisonchus" (see Figure 20.6), the taxa are paraphyletic rather of these "zhelestids," Alostera saskatchewanensis (Fox, 1989), has
than monophyletic; that is, some of the species within in each genus been recognized from not only the uppermost Cretaceous localities
share a more recent common ancestor with species not included in but also from Bug Creek Anthills (locality NP15A), which is now
their genus (see Figure 5 in Archibald, 1994). The tradition of plac- thought to be earliest Paleocene. The specimens of A. saskatchewa-
ing such taxa (paraphyletic or polyphyletic) in quotation marks is nensis from Bug Creek may be reworked, as is now argued for most
followed here. taxa also known from the Upper Cretaceous Hell Creek Formation
In at least tw? other cases - Protungulatum and Oxyacodon - (Lofgren, 1995; Archibald, 1996a), or this species may be a true
neither paraphyly nor monophyly could be determined. The term harbinger of the archaic ungulates that began their radiation at the
"metataxon" has been used for such situations (Gauthier, Estes, and same time in North America (and possibly elsewhere).
de Queiroz, 1988). For instances such as this, Archibald (1994)
more specifically defined these kinds of metataxa as mixotaxa, or
in this case, mixogenera. Most species in these mixogenera pos- INCLUDED GENERA OF NORTH AMERICAN
sess autapomorphies that argue for their species monophyly, but ARCHAIC UNGULATES
the species as a group lacks apomorphies that argue for the mono-
phyly of the genus. Such mixotaxa are designated by two aster- In the following discussion, taxa of archaic ungulates ("condy-
isks (Protungulatum** and Oxyacodon**). Further, some species, larths") are listed in their order of appearance in the classification
whether in monophyletic, paraphyletic, or metaphyletic genera, may given in the chapter appendix. Only the more inclusive taxa in this
lack autapomorphies that unite the included specimens. Thus these table that include species of archaic ungulates are discussed here.
species are metaphyletic. Such metaspecies are designated by a sin- The locality numbers listed for each genus refer to the list of uni-
gle asterisk (Archibald, 1993, 1994). In the following discussion fied localities in Appendix I. The acronyms for museum collections
of genera - Protungulatum**, "Mimatuta," Ectoconus, "Carsiopty- are listed in Appendix III. Taxa followed by two asterisks are mixo-
chus," Periptychus, Gillisonchus, "Anisonchus," Haploconus, and taxa (it cannot be determined whether or not they are monophyletic).
Oxyacodon** - each possesses at least one metaspecies. There is The locality numbers may be listed in a couple of alternative ways.
little question that many more paraphyletic and metaphyletic taxa Parentheses around the locality (e.g., [CP101]) mean the taxon in
exist among the archaic ungulates outlined in this chapter, but with- question at that locality is cited as an "aff." or "cf." the taxon in
out more detailed analyses, they remain hidden. question. Parentheses are usually used for individual species, thus
It must be emphasized that this study includes only taxa of ar- implying the genus is firmly known from the locality, but the actual
chaic ungulates that have at least one North American represen- species identification may be questionable. Question marks in front
tative. Archaic ungulates have been reported from all continents of the locality (e.g., ?CP101) mean the taxon is questionably known
except Antarctica and Australia, thus the cladogram (Figure 20.6) from that locality, thus implying some doubt that the tax on is actually
and classification (chapter appendix) are not complete. A number present at that locality, either at the genus or the species level.
 J. David Archibald 300

Protungulatum** Sloan and Van Valen, 1965 genus is placed in a tritomy with Arctocyonidae (sensu
Type species: Protungulatum donnae Sloan and Van Valen, Prothero, Manning, and Fischer [ 198 8]) and the remaining
1965. Ungulata (excepting the possibly earlier branching artio-
Type specimen: SPSM 62-2028, left dentary with p2, ml-3. dactyls) (see Figure 20.6).
Characteristics:

Small arctocyonids that are most similar to Oxyclaenus. P3-4 ARCTOCYONIDAE (GIEBEL, 1855) MURRAY, 1866
with a relatively narrow paracone; P3 with a small protocone; Characteristics and comments:
P4 with a prominent pointed protocone, and usually a trace
of a metacone. M2/m2 wider transversely than Ml/ml. Pro- Teeth primitive, tritubercular, varying towards tuberculosec-
minent labial cingula on P3-M3; prominent lingual cingula torial or bunodont. No carnassials or specialized shearing
on Ml-3. Upper molars relatively transverse; hypocone ab- teeth. Premolars mostly simple, large, acute; incisors, small.
sent or weak swelling on the posterior cingulum; apex of Skull moderately long, brain case small, sagittal and occipi-
protocone nearly half the distance to the labial border of the tal crests strong, occiput narrow and high, tympanic bulla not
tooth. [Lower fourth premolar] submolariform, with tall pro- ossified. Limbs of moderate length, ulnar and fibular shafts
toconid, large but low paraconid, and moderately large meta- relatively stout, toes 5-5, the hallux and pollex a little re-
conid; talonid simple. Lower molars with short and relatively duced, phalanges narrow and long, the unguals compressed
narrow talonids, usually with an incomplete lingual border; and clawlike, astragalus with little or no trochlear groove,
lingual cingula ~e absent and labial cingula are weak; para- the head wedged between tibia and fibula, astragalar foramen
conid moderately large and lingual. [Lower third molar] nar- large, fibulocalcanear facet distinct and fibulo-astragalar facet
rower than m2 and with projecting hypoconulid. Cusps on facing almost as much proximad as ectad. Tibiale present in
all teeth high for an arctocyonid; enamel not wrinkled except one, probably in all the genera. Centrale united to scaphoid in
sometimes on cingula. (Sloan and Van Valen, 1965, p. 226; Claenodon, unknown in others. (Matthew, 1937, pp. 15-16;
see Figures 20.2A, 20.3A.) see Figures 20.2F, G, 20.3A, 20.5C.)

Matthew's 1937 diagnosis remains a reasonable description of

Average length of ml: 3.7-5.0 (est.) mm.
arctocyonids; however, at the time of his work they were generally
Included species: P. ** donnae* (known from localities
regarded as creodonts, hence the reference to the Jack of "carnassials
[NP6], [NP7A], NP15A, NP15B, NP15C, NP16A); P.**
or specialized shearing teeth" in arctocyonids that are by contrast
gorgun (locality NP15C); P.** mckeeveri (locality
present in creodonts. Earlier workers (Ameghino, 1901; Kretzoi,
NP16A); P.** sloani (locality NP16C).
1943) had referred Arctocyonidae to Condylarthra, but it was not un-
Protungulatum ** sp. is also known from locality NPl 7
til the 1960s (Patterson, in Patterson and McGrew, 1962; Van Valen,
as well as other localities.
1966) that this family was referred to Condylarthra where it was
Comments: A noteworthy modification of Arctocyonidae by regarded as the stem family from which other archaic ungulates
Prothero, Manning, and Fischer (1988) is the exclusion of derived. Various subfamilies have been recognized for this family.
the earliest known ungulate, Protungulatum**, from Arcto- For example, Simpson (1945) included Oxyclaeninae (including
cyonidae. They placed this genus in a tritomous relation- Chriacinae), Arctocyoninae, Triisodontinae. Van Valen (1978) rec-
ship with Hyopsodontidae (they included Mioclaenidae) ognized an additional subfamily, Loxolophinae, which incorporated
and Periptychidae. They argued for the monophyly of this several genera traditionally placed within Oxyclaeninae plus several
clade based only on the supposed bulbousness of the para- new taxa named in the same paper.
cone and protocone on P3-4, a character state that in fact is One paper addressing the inter- and intrarelationships of Arcto-
not found in all hyopsodontids and is a difficult judgment cyonidae is that of Prothero, Manning, and Fischer (1988). These
call in the case of these teeth in Protungulatum**. Placing authors took the seemingly radical step of removing the triisodon-
this genus within Arctocyonidae seems equally insupport- tines as well as the genera Oxyclaenus (traditionally an oxyclae-
able. As recognized by Prothero, Manning, and Fischer nine) and Baioconodon (a loxolophine, according to Van Valen,
(1988), this family includes most of the genera of oxy- 1978) and placing them with their presumptive sister taxa and/or
claenines and arctocyonines and is united by a complete descendants within the clade Cete. Their Cete includes the triisodon-
lingual cingulum on upper molars. Although small ante- tines, mesonychids, and Cetacea. This idea of linking triisodontines
rior and posterior cingula (and hypocone) are present in to Cete is also found in Van Valen (1978, Figure 4), although he
Protungulatum**, the cingula are not complete lingually. continued to place this subfamily within the Arctocyonidae, and
Prothero, Manning, and Fischer (1988) also cite several in McKenna (1975), but this relationship was discussed only to
cranial and postcranial character states that all ungulates the familial level. The probable linkage of triisodontines, mesony-
(including Protungulatum**) are supposed to possess in chids, and cetaceans was apparently first suggested by Van Valen
comparison to artiodactyls. Because only the dentition, (1966).
isolated ear regions, and tarsals of this genus have been Arctocyonidae is a European family (Russell, 1964). Arctocy-
described, it seems premature to speculate further on the onidae is also reported from Asia (Russell and Zhai, 1987), but the
cladistic relationships of Protungulatum**. Hence this included genera are here referred to Mesonychidae.
 Archaic ungulates ("Condylarthra") 301

Desmatoclaenus Gazin, 1941b forwardly-inclined trigonids and broadly basined talonids.

Type species: Desmatoclaenus hermaeus Gazin, 1941 b. The lingual position of the paraconid, the anteroposterior
Type specimen: USNM 16202, greater portion of upper and shortness of the trigonid, and the breadth and shallowness of
lower dentition. the talonid distinguish this form from any species of Thryp-
Characteristics: tacodon. The low trigonids, rectangular shape of m2-3, and
more rounded molar cusps and crests distinguish this
Simple P3; P4 with low, independent parastyle, no intermedi-
form from any species of Chriacus. (Gingerich, 1989,
ate conules; upper molars lack mesostyle, cingulum discon- p. 43.)
tinuous internally, small hypocone and intermediate conules;
Ml square; M2 wide with prominent parastyle; M3 small Average length of ml: unknown. Length of m2: 5.3 mm.
with poorly developed metacone and no hypocone, parastyle Included species: P. yalensis only (known from localities
smaller than in anterior molars. [Lower third premolar] lin- CP13G, CP18B, CP19).
ear with low anterior cusp; p4 with distinct low paraconid,
large protoconid and closely associated metaconid, talonid Comments: Although upper molars were mentioned by Gin-
with well developed hypoconid and small entoconid; lower gerich (1989), no figures, descriptions or measurements
molars with well developed lingually situated paraconids and were provided. Without such information it is not possible
distinct trigonid basin, as in Protogonodon, and small ante- to place Princetonia more precisely within the phyloge-
netic framework of arctocyonids.
rior cingulum; m3 relatively small with cuspidate entoconid-
hypoconulid crest. (West, 1976, p. 11.)
Prothryptacodon Simpson, 1935
Average length of ml: 6.0 (est.)-8.3 (est.) mm. Type species: Prothryptacodonfurens, Simpson 1935.
Included species: D. hermaeus (= D. paracreodus) (known Type specimen: USNM 9260, right dentary with p4, ml-3,
from localities CPlA, CPlB, CPlC); D. protogonioides and alveoli.
(localities SB23A, SB23B); D. dianae (localities SB23A, Characteristics:
SB23B); D. mearae (localities CP16A, ?NP19IIC).
Canine semiprocumbent, root extending beneath premolars
Desmatoclaenus sp. has also been reported from local-
ities NPlC, NP7C. (as in Thryptacodon). [Lower first and second premolar]
spaced widely. [Lower fourth premolar] similar to Thrypta-
Comments: When described by Gazin in 194lb, he noted codon. Molar trigonids higher than in Thryptacodon, para-
that Desmatoclaenus is intermediate in appearance be- conids reduced and in nearly same position as in Thrypta-
tween that of the arctocyonid Protogonodon and of the codon, but more distinct, higher on crown, trigonids less
phenacodontid Tetraclaendon. West (1976) referred Des- basined and with fewer accessory cuspules. Only one distinct
matoclaenus to the Phenacodontidae. More recently, inner talonid cusp, the entoconid (two in Thryptacodon).
Van Valen ( 1978) placed this tax on in the Arctocyonidae. In (Simpson, 1935, pp. 233-234.)
his "provisional phylogenies" (Figures 3 and 4), he placed
Average length of ml: 4.5-5.2 mm.
Desmatoclaenus as central to the origin of not only phenac-
Included species: P. furens (known from localities CP14A,
odontids but also most extant ungulates (except artiod-
NP19C); P. albertensis (locality NPlA); P. ambiguus (lo-
actyls and cetaceans) and most extinct ungulates, including
cality SB23H); P. hilli (localities [CP13C], [CP13D],
the host of South American forms. Cifelli (1983) agreed
CP14A).
with Van Valen's placement of Desmatoclaenus in Arc-
tocyonidae, noting that although it could serve as a suit-
able morphotype for the phenacodontids, it lacks any fea- UNNAMED TAXON: CHRIACUS + DEUTEROGONODON
tures diagnostic of that family. He further indicated that + LOXOLOPHUS + MIMOTRICENTES +
it possesses "several [unspecified] loxolophine speciali- LAMBERTOCYON + MENTOCLAENODON +
zations" that unite it with a species, D. protogonioides, ARCTOCYONINAE
which Van Valen (1978) had recently referred to the
Chriacus Cope, 1883a (synonyms: Epichriacus Scott, 1892,
genus.
Metachriacus Simpson, 1935; Spanoxyodon Simpson, 1935;
Tricentes [in part], Cope, 1884a)
Princetonia Gingerich, 1989 Type species: Chriacus pelvidens (Cope, 188 lf) (= Lipodec-
Type species: Princetonia yalensis Gingerich, 1989. tes pelvidens).
Type specimen: YPM-PU 23629, right dentary with p4 root, Type specimen: AMNH 3097, left dentary with p4, ml-3.
m2, and most of m3. Characteristics:
Characteristics:
First and second upper molars with strong spurlike hypocone
Small arctocyonid distinct from all others in combining a projecting inwards and backwards, and smaller protostyle
relatively long, narrow p4 lacking a metaconid, with rela- on M2, giving them a quadrate outline with concave sides.
tively short and broad rectangular molars having low, broad, Intermediates small. Cusps higher than in the two preceding
 J. David Archibald 302

genera [Protochriacus (now Loxolophus) and Tricentes (now Loxolophus Cope, 1885 (synonyms: Protogonodon Scott,
in part Chriacus and in part Mimotricentes], trigonid more 1892; Paradoxodon Scott, 1892; Protochriacus Scott, 1892)
raised. Paraconid somewhat reduced, P3 and p4 with well Type species: Loxolophus adapinus (Cope, 1885) (= Chria-
developed deuterocone in the larger species. (Matthew, 1897, cus hyattianus fide Matthew, 1937).
p. 272; see Figures 20.IC, 20.5C.) Type specimen of Loxolophus adapinus: AMNH 3124, crus-
The generic and specific synonymies are those of Van hed left dentary with ml-3. Type specimen of Chria-
Valen (1978). See also comments under Mimotricentes. cus hyattianus: AMNH 3121, distorted left maxilla with
Ml-3.
Average length of ml: 4.7-6.5 mm. Characteristics:
Included species: C. pelvidens (known from localities SB23E,
SB23F, SB23G, SB23GG, SB23H, CP13B, CP14A, Incisors small; canines large, crowns sharp, recurved, not
CP14B, CP15B, CP16A, CP16B, CP21A, NPlC, NP3C, compressed; Pl/pl one-rooted, small, close behind canines;
NP3D, NP19IIA, NP19IIC); C. calenancus (locality P2 two-rooted, compressed; P3-4 triangular, three-rooted;
NP16C); C. orthogonius (= Metachriacus punitor) (lo- protocone minute on P3, large on P4; molars tritubercular
calities CP13C, CP14B, NPlC, NP19C, ?NP19IIA); C. with rudimentary hypocone and no protostyle; cusps low,
katrinae (locality CP13B); C. baldwini (= Tricentes cras- conules small, M2 largest, M3 little reduced; lower premo-
sicolidens, Metachriacus provocator, Chriacus truncatus, lars compressed, with small heels, paraconids well devel-
C. schlosserianus, Spanoxyodon latrunculus) (localities oped, protoconid and metaconid subequal, entoconid strong,
SB23E, SB23F, SB23G, SB23GG, SB23H, SB39B, hypoconulid very rudimentary on ml-2, well-developed on
SB39C, CPIC, [CP13C], CP14C, [NPlC], NP19C); m3; heel of m3 narrower but longer than trigonid .... Limbs
C. gallinae (localities SB24, CP20C, [CP62C], CP63, creodont type, humerus with high deltoid crest abruptly end-
[CP64C]); C. oconostotae (locality CP13E); C. meta- ing, distal end wide, with entepicondylar foramen; femur with
cometi (locality SB20A); C. badgleyi (locality CP19); well-developed third trochanter, second trochanter internal
C. n. sp. (locality NP3C). in position; tibia with moderate cnemial crest fading gradu-
Chriacus sp. has also been reported from localities ally towards the middle of the shaft; astragalus with head of
CPI 3H, CPI 7B, CPI SB, (CP20A), CP22A, CP25A, primitive type, wedged between tibia and fibula, no trochlear
CP26C, CP62B, NPlB, ?NP3A, NP7D, NP19IIA, groove, tibial facet limited backwardly, astragular foramen
(NP20B). well-developed, fibular facet facing obliquely outward, head
of astragalus oval, neck distinct. Metapodials slender, inner
digit semi-opposable but not large nor so long as the oth-
UNNAMED TAXON: DEUTEROGONODON + ers; phalanges long and slender, unguals clawlike. (Matthew,
LOXOLOPHUS 1937, pp. 42-43; see Figure 20.2F.)
Deuterogonodon Simpson, 1935 (from Gidley, ex. ms.) Average length of ml: 5.7-10.0 mm.
Type species: Deuterogonodon montanus Simpson, 1935. Included species: L. hyattianus (known from localities
Type specimen: USNM 6160, right maxilla with fragmentary SB23A, SB23B, CP61B); L. priscus (locality SB23B); L.
Ml-2 and complete M3, left dentary with m2 and m3 pentacus (localities SB23B, CPlA, CPIC, NPl 7); L. kim-
talonid. betovius (localities SB23A, SB23C, ?CP61B); L. spiekeri
Characteristics: (localities [CPIA], CPlC); L. criswelli (locality CP14A);
Dentition basically arctocyonid and resembling Protogon- L. schizophrenus (localities NP7B, [NP7C]).
odon and Claenodon. Distinct, small hypocone on M2-3 (at Loxolophus sp. has also been reported from localities
least), cingula almost completely circling these teeth. Small CPlB, NP17.
but well-defined mesostyle present. Parastyle of M3 a dis-
tinct, strongly projecting cusp. Lower molars with trigonid Mimotricentes Simpson, 1937 (synonyms: Tricentes
only slightly higher than talonid, metaconid smaller than but [in part] Cope, 1884a)
as high as protoconid. Paraconid very small, subconical, on Type species: Mimotricentes subtrigonus (Gidley in Simpson,
anterior slope of metaconid. Talonid basin open with continu- 1935) (= Tricentes latidens = Mioclaenus subtrigonus
ous crescentic lop hid differentiated into three apices. Enamel Cope, 188ld, fide Van Valen, 1978).
wrinkled, but little or no tendency to form accessory cuspules. Type specimen of Tricentes latidens: USNM 9269 left den-
(Simpson, 1935, p. 233.) tary with canine, p2-m3.
Type specimen of Mioclaenus subtrigonus: AMNH 3227,
Average length of ml: 10.5 (est.)-12.0 (est.) mm. part of skull with most of palate and cheek teeth.
Included species: D. montanus (known from locality Characteristics:
NP19B); D. noletil (localities ?SB23C, SB23E, SB23F,
?SB23G). Dentition I?/?, Cl/l, P3/3, M3/3. [According to Van Valen,
Deuterogonodon sp. is also questionably known from 1978, some species have four lower (and upper?) premolars.]
locality SB23D. Canines of moderate size, round-oval, rather short and stout,
 Archaic ungulates ("Condylarthra") 303

a short but well-marked diastema behind them. Premolars crested lower molars with more open trigonids. (Gingerich,
rather short and high-pointed, not compressed, the basal cusps 1979, p. 525.)
comparatively small; the molars low, rounded, quadrate with
low subcircular cusps and more or less encircling cingula; Average length of ml: 6.0-6.7 mm.
M3/m3 unreduced. Enamel on all teeth remarkably rugose Included species: L. eximius (known from localities SB39C,
with short vertical wrinkles dominant. P2/p2-3 simple and CP13F, ?CP13G, CP14C, CP14D, CP15B); L. ischyrus
high, P3 trigonal, P4 with large inner cusp and rudimentary (locality CP62B).
basal cusps at external angles, p4 with rudimentary inner Lambertocyon sp. has also been questionably reported
cusp and small heel. Upper molars, M 1-2, rounded quadrate, at locality CP24A.
the hypocone and conules distinct but much smaller than the
three main cusps; stylar cusps absent or very rudimentary. Mentoclaenodon Weigelt, 1960
M3 rounded trigonal, no hypocone or metaconule, the meta- Type species: Mentoclaenodon walbeckensis Weigelt, 1960.
cone reduced, sub-median. M2 considerably larger than Ml, Type specimen: Wa/314, left dentary with p3-m3, alveoli for
M3 somewhat smaller. Lower molars with paraconid well canine and p2.
developed on ml, smaller on m2, minute on m3, but sub- Characteristics:
median on all and not merged with metaconid. Entoconids
and hypoconulids distinct and rounded on all three molars, Dental formula I?3/?3, Cl/1, P4/3, M3/3. Lower premolars
hypoconids larger, also rounded. [Second lower molar] a little elongated, but little compressed laterally; anterior basal cusp
larger than ml, heel of m3 comparatively short, the three mo- weak or absent; p3-4 talonid slightly hollowed and is pro-
lars of subequal length, but m2 a little wider than the others. vided with two small cusps; metaconid absent. Lower molars
Talonids moderately basined larger than trigonids but notably with a high trigonid and a deeply hollowed talonid; ento-
wider on ml only. (Matthew, 1937, p. 60.) conid and hypoconid in the form of crests, not cuspidate;
hypoconulid vestigial or absent. Paraconid of ml distinctly
Average length of ml: 4.9 (est.)-5.7 mm. lingual, small, but low and not confluent with the metaconid.
Included species: M. subtrigonus (= M. latidens, M. angusti- Paraconid weak or absent on m2-m3 and situated lingually
dens, and ?M. elassus) (known from localities SB23C, when present.
SB23DD, SB23E, SB23F, SB23G, SB23H, CPlC, CP13C, Upper molars provided with a well-developed rim. Proto-
CP14A, CB15A, CP21A, CP26A, NP19C); M. mirielae conule weak or absent on Ml-M2; weak hypocone, low and
(locality SB23A); M. fremontensis (localities CP13E, "crestiforme"; M3 elongated transversely.
CP16A, CP16B, NP19IIA). Dentary enamel rugose. Horizontal branch of the mandible
Mi"';otricentes sp. has also been reported from localities high and laterally compressed, its antero-ventral border show-
CP13B, CP13E, (CP13G), CP15B, CPlSA, ?NPlC, ing a small expansion below the root of the canine. (Trans-
NP19IIC, (NP20B), NP20D. lated from Russell, 1964, p. 219.)

Comments: Van Valen (1978) argued that the type of Tri- Average length of ml: 12.2 mm.
centes, T. crassicollidens Cope, 1884a, is a synonym of Included species: M. acrogenius only (known from localities
Chriacus baldwini (Cope, 1882c). Van Valen (1978) did CP13E, CP16A, CP16B).
not argue for the conservation of the name Tricentes, but
rather placed the other species that had been referred to Tri- Comments: In his discussion of this species, Gazin (1956)
centes into the genus Mimotricentes Simpson, 1937. Most noted that Claenodon acrogenius is similar to C. ferox
of the generic and specific synonymies follow Van Valen but near the upper size limit for the latter taxon. All of
(1978). the characters noted by Gazin (1956) presumably pertain
to species-level distinctions and hence are not included
here.
Lombertocyon Gingerich, 1979
Mentoclaenodon is a European genus (Russell, 1964,
Type species: Lambertocyon eximius Gingerich, 1979.
1980). Van Valen (1978) indicated, however, that Claen-
Type specimen: YPM-PU 19576, crushed rostrum with upper
odon aero genius Gazin, 1956, from Bison Basin, Wyoming,
dentition and left and right dentaries.
Characteristics: belongs to Mentoclaenodon, although no justifications
were given. Although not fully endorsing Van Valen's
Differs fromArctocyon, Thryptacodon, and most other arcto- (1978) conclusions, Russell (1980) accepted the provi-
cyonids in having relatively short upper and lower jaws, with sional referral of Claenodon acrogenius to M entoclaenodon.
deep mandibular rami, short broad premolars, stout round
canines, nearly vertically implanted incisors. The upper mo-
ARCTOCYONINAE GIEBEL, 1855
lars have reduced paraconules, and nearly continuous lingual
cingula. Lambertocyon is most similar to Mimotricentes, but Anacodon Cope, 1882a
differs from it in being larger, in lacking well-developed para- Type species: Anacodon ursidens Cope, 1882a.
conules on the upper molars, and in having more sharply Type specimen: AMNH 4261, parts of dentaries.
 J. David Archibald 304

Characteristics: recognized by Gidley (1919): N. montanenesis andN. lati-

dens. Van Valen treated the latter species as a synonym
Crowns of molar teeth flattened, rugose, cusps obscured, pre- of the first. (Simpson, 1937, had previously synonymized
molars 2/2 much reduced. Lower jaw flanged anteriorly, ca- Neoclaenodon with Claenodon.) Van Valen ( 1978) referred
nine and incisors reduced and crowded, upper canine proba- the other species of Neoclaenodon to Colpoclaenus (see
bly laniary. (Matthew, 1915a, p. 13.) this taxon). This usage is followed here. Cifelli (1983, p.
Average length of ml: 12.0 (est.) mm. 32) provides a further and slightly different discussion of
Included species: A. ursidens (known from localities SB24, this taxonomic confusion (see Figure 20.2G).
CP20A, CP20B, CP20C, CP27B, [NP20E]); A? nexus (lo-
calities CP13G, CP24B, CP26C). Colpoclaenus Patterson and McGrew, 1962 (synonym:
Anacodon sp. is also reported from locality CP20A. Neoclaenodon Scott, 1892 [in part])
Type species: Colpoclaenus keeferi, Patterson and McGrew,
Arctocyon Blainville, 1841 (synonyms: Claenodon Scott, 1962.
1892; Neoclaenodon Gidley, 1919, in part) Type specimen: MCZ 8355, left m3.
Type species: Arctocyon primaevus Blainville, 1841. Characteristics:
Type specimen: CRL 956, nearly complete skull.
Enamel of molars strongly wrinkled, all crests crenulated,
Characteristics:
accessory cuspules numerous. Principal cusps of upper M
Dental formula I3/3, Cl/1, P4/3-4, M3/3. Lower premolars high, massive, tightly grouped; central basin small; protocone
elongated but little compressed laterally; usually there is a nearly central in position with long lingual slope; conules
small anterior basal cusp and the talonid is slightly hollowed. large, blunt, separated by grooves from principal cusps; hypo-
There is no metaconid on p4. The paraconid of ml is well cone small on Ml-2, rudimentary on M3; cingula strong,
lingual, small but elevated, confluent with the metaconid. not continuous around protocones; external cingulum inter-
Paraconid vestigial orabsent on m2-3. [Lower second molar] rupted by labial continuation of sharp cleft between paracone
generally larger at the level of the trigonid than at the level of and metacone; upper M wide relative to length. Trigonids of
the talonid. Trigonid of molars low. From the protoconid and lower M high, short; paraconid crest well developed; poste-
the metaconid begin, in a medial direction, well-developed rior crest connecting protoconid and metaconid present; pro-
transverse crests, often joined. Crowns of the lower and upper toconids and metaconids with short, centrally situated crests
molars elevated and somewhat swollen; cusps low and sub- running lingually and labially, respectively, forming a third,
bunodont. Roots of the lower molars widened transversely. transverse trigonid crest; deepest portion oftalonid basin near
Roots of the upper molars long, developed antero-posteriorly lingual side; hypoconulid of m3 long, broad, high, cuspidate.
and diverging to the point of being obliquely implanted in the (Patterson and McGrew, 1962, pp. 2-3.)
maxilla. Hypocone present (vestigial or absent only on M3);
the apex of the cusp placed almost at the same level as the Average length of ml: 8.0 (est.) mm.
apex of the protocone. Trigon of the upper molars shallow. Included species: C. keeferi (known from localities CP13E,
CP15A, [NPlC], NP3C); C. silberlingi (= Claenodon vec-
Mandible thickened in the region of m2-3. (Translated from
ordensis) (localities NP19A, NP19C); C. procyonoides (lo-
Russell, 1964, p. 138.)
calities SB23H, [NP20D]).
Average length of ml: 11.0-12.0 mm. Colpoclaenus sp. has also been reported from localities
Included species: A. montanensis (= (?)Neoclaenodon lati- NP2,NP3D.
dens) (known from localities CP13C, CP16A, NPlB,
[NPlC], NP19IC, NP19IIA, NP19IIC); A. ferox (locali- Comments: When named by Patterson and McGrew (1962),
ties SB23G, SB23GG, SB23H, SB39C, CP13E, CP14A, the only referred species was Colpoclaenus keeferi. In
CP15A, CP16A, CP16B, CP16C, CP21B, [NP3C], 1978, Van Valen added two additional species, C. silber-
[NP20D]); A. mumak (locality CP14C). lingi (incl. Claenodon vecordensis) and C. procyonoides,
Arctocyon sp. has also been reported from localities both of which had originally been placed in Neoclaenodon
SB23G, CP13B, CP13G, CP14C, CP14D, CP24A, NPlC, by Gidley (1919) and Matthew (1937), respectively.
NP3F, NP19IIA.
Thryptacodon Matthew, 1915a
Comments: Arctocyon is primarily a European genus (Rus- Type species: Thryptacodon antiquus Matthew, 1915a.
sell, 1964, 1980). Van Valen (1978) argued that Claen- Type specimen: AMNH 16162, maxilla and dentary and parts
odon Scott, 1892, is a synonym of Arctocyonides Lemoine, of radius and ulna.
1891, rather than of Arctocyon. Russell (1980) retained Characteristics:
Claenodon as a synonym of Arctocyon. This usage is fol-
lowed here. Van Valen (1978) also placed within Arcto- Upper molars low-crowned, quadrate-oval or rounded, cusps
cyonides two of the species of Neoclaenodon originally round conic, hypocone prominent on Ml-2, enamel rugose,
 Archaic ungulates ("Condylarthra") 305

M3/m3 somewhat reduced, round oval; P4 trihedral with Cingula absent or weakly developed and no crest from para-
small deuterocone, distinct para- and metastyles. Lower mo- conid to lingual surface as in Oxyacodon. Hypoconulid less
lars broad with very small submedian paraconid and four developed. [Lower third molar] unreduced. (Gazin, 1941b,
subequal opposite principal; cusps. Heavy external cingula p. 26.)
on lower molars; heavy encircling cingula on upper molars.
Anterior premolars slender; canines long, compressed and Average length of ml: unknown. Length of m2: 3.5 mm.
ridged posteriorly. Skull short, comparatively large brain- Included species: 0. perissum only, known from locality
case, skeleton relatively large, resembling that of Miacinae. CPlC only.
(Matthew, 1915a, p. 7.) Comments: "Van Valen (1978) synonymized Oxytomodon
Average length of ml: 6.0-6.7 (est.) mm. Gazin, 194l(b) with Litomylus Simpson, 1935, but the for-
Included species: T. antiquus (= T. loisi, T. olseni, T. pseu- mer is distinct in the better developed, lingually placed
darctos) (known from localities CPI 3H, ?CPl 7A, CPl 7B, lower molar paraconid (as Gazin observed) and in the dis-
CP18A, CP18B, CP20A, CP20B, CP20C, CP24B, CP25A, tinct height differential between the trigonid and talonid.
CP25G, CP26A, CP27B, CP27D, CP27E, CP63, NP20E); These primitive features suggest that Oxytomodon is closer
T. australis (localities SB20A, SB24, CP13E, CP14C, to the Oxyclaeninae, to which I refer it" (Cifelli, 1983,
p. 35).
CP14D, CP16B, CP16C, CP21B, CP22A, CP22B,CP24A,
[CP62B], NP7D, NP19IIC, NP20D, NP47B, [NP47C]); T.
demari (= T. belli) (localities CP14C, CP14D, CP16A, HYOPSODONTIDAE TROUESSART, 1879
CP16B, CP16C, NP19IIA); T. barae (locality CP19).
Characteristics:
Thryptacodon sp. has also been reported from localities
CP13G, NP47C. P4/p4 not inflated, generally moderate in size, cuspidate,
somewhat molariform but never exactly so, p4 generally with
ARCTOCYONIDAE? ametaconid and wide but imperfect talonid basin. Molar para-
conids median to subinternal, not fusing with metaconids,
Carcinodon Scott, 1892
entoconids distinct and high, talonid basins closed. M3/m3
Type species: Carcinodonfilholianus (Cope, 1888) (=Mio- little or not reduced. Ml-2 with definite hypocone, small in
claenus filholianus).
earlier and large in later forms, sharply distinct from tip of
Type specimen: AMNH 3205, left dentary with p3-4, ml-3; protocone. (Simpson's 1937 definition of Hyopsodontinae,
right dentary with ml-3. p. 241; Figures 20.2B, 20.3D, E, 20.5B.)
Characteristics: In reference to Carcinodon aquilonius:
"Smaller than C.filholianus (Cope, 1888); paraconid more As constituted here, Hyopsodontidae excludes Mioclaenidae. As
anterad-projecting; talonid of p4 bulging slightly buccad" discussed under the latter taxon, some or all of the mioclaenids ap-
(L. Russell, 1974, p. 28). pear to form a monophyletic taxon whose closest sister taxon may
Average length of ml: 4.0 (est.) mm. be Periptychidae. The assignment of genera to Hyopsodontidae fol-
Included species: C. filholianus (known from locality lows that of Cifelli (1983), although he also included mioclaenines
SB2 3B); C. aquilonius (localities NP7B, NPl 7). in the family. Cifelli included Hyopsodus, Litomylus, Haplaletes,
Aletodon, and with grave doubt, Haplomylus. Subsequently, Gin-
Comments: Van Valen (1978) synonymized the genotypic gerich (1983) described two new genera, Dorraletes and Utemylus,
species Carcinodon filholianus with Oxyclaenus simplex which he referred to Hyopsodontidae. To this family, I would also
and C. aquilonius with Prothryptacodon albertensis and questionably add Oxyprimus. Van Valen (1978) included this genus
referred this latter species to the genus Oxyprimus. in Arctocyonidae, but noted a probable relationship with Hyop-
Johnston and Fox (1984) discussed these proposals and sodontidae as advocated by Archibald (1982).
argued that C. aquilonius and P. albertensis are separate As noted by Cifelli (1983, p. 33), Hyopsodontidae served as
taxa and that, further, the latter species is best retained in the wastebasket taxon for "generally small, primitive ungulates not
Prothryptacodon rather than being referred to Oxyprimus. clearly specialized enough to warrant allocation to other, more ad-
I concur. If, however, C. filholinaus is a synonym of Oxy- vanced suprageneric groups" until this role was assumed by the Arc-
claenus simplex, then Carcinodon is a synonym of Oxy- tocyonidae. There still remains the question of what if any apomor-
claenus and is no longer available for C. aquilonius. phic character states these genera share. This cannot be answered
with any certainty. Two possibilities might be the trend toward mo-
Oxytomodon Gazin, 1941b
larization of the p4 and the labial shift and eventual reduction of
Type species: Oxytomodon perissum Gazin, 1941b.
the molar paraconid. For the p4, attention has focused on whether
Type specimen: USNM 16183, left m2-3.
possession of a metaconid is primitive for ungulates specifically and
Characteristics:
therians generally.
Lower teeth slender with cusps high and distinct. Paraconid Some authors, such as Cifelli (1983), have argued that possession
on m2 and m3 lingual in position and close to metaconid. of a p4 metaconid is primitive for ungulates, whereas Archibald
 J. David Archibald 306

(1982) argued a metaconid was acquired within the Ungulata and Type specimen: USNM 9555, right dentary with p3, ml-3.
hence is derived for only some ungulates. The "zhelestids" from Characteristics:
the Late Cretaceous of western Asia are the probable sister taxon
to Ungulata. They show a small, but distinct metaconid on the ulti- [Lower fourth premolar] not trenchant, with distinct para-
mate (p5?) premolar, suggesting that Cifelli is correct (e.g., Nessov, conid and metaconid and relative! y large basined heel. Molars
Archibald, and Kielan-Jaworowska, 1997). According to this as- bunodont, cusps rather rounded and low. Paraconids reduced
sessment, the reduction of the p4 metaconid in some species of and median. M3/m3 somewhat reduced, hypoconulid of m3
Aletodon (Gingerich, 1983) would be considered to be autapomor- slightly projecting, markedly less than in Litomylus. Amph-
phous (Cifelli, 1983). icones of P3-4 more rounded than in Litolestes [now included
Hyopsodontidae is also known from Europe (Russell, 1964 ), Asia in the Erinaceidae, Insectivora; Krishtalka, 1976]. P4 with
(Russell and Zhai, 1987), and questionably South America (Muizon rudimentary metacone. Upper molars rounded. Hypocones
and Marshall, 1991). of Ml-2 distinct, but smaller than in Litomylus. M3 less
transverse and more rounded. (Simpson, 1935, p. 243; see
Aletodon Gingerich, 1977 Figure 20.2E.)
Type species: Aletodon gunnelli Gingerich, 1977.
Type specimen: UM 66301, right dentary with p4, ml, m3. Average length of ml: 2.5-3.0 mm.
Characteristics: Included species: H. disceptatrix (known from localities
[SB23H], SB39C, CP13B, CP13C, CP13D, CP14A,
Aletodon gunnelli is easily distinguished from other Hyop- CP22A, NP19C, [NP20B], NP20D); H. serior (localities
sodontidae by its relatively long, simple, pointed p4, which [CP13E], CP16C); H. pelicatus (localities CP16A,
Jacks any distinct paraconid, metaconid, or entoconid acces- [CP21B]; H. andakupensis (localities CPIB, NP16C).
sory cusps. (Gingerich, 1977, p. 238.) Haplaletes sp. may be present at localities CP21A,
Gingerich (1983) transferred Platymastus mellon (Van NP20B. Reports in Archibald et al. (1987) for Haplaletes
Valen, 1978) to Aletodon. at localities CP22B and NP47B pertain to Dorraletes.
Average length of ml: 3.9-5.0 mm.
Included species: A. gunnelli (known from localities CP13H, Haplomylus Matthew, 1915b
?CP14E, CP17B, CP18B, CP26C, CP62B, [CP62C]); A. Type species: Haplomylus speirianus (Cope, 1880) (=Hy-
quadravus (localities CP13E, CP22A, CP24A, NP47B);A. opsodus speirianus).
conardae (locality CP14D); A. mellon (locality SB20A). Type specimen: AMNH 4190, dentary with ml-3.
Characteristics:
Dorraletes Gingerich, 1983
Type species: Dorraletes diminutivus (Dorr, 1952) (=Hap- [L]ower molars consist of four cusps, somewhat obliquely
laletes diminutivus). set, but less so than in Hyopsodus, lower, with a distinctly
Type specimen: UM 27231, left maxilla with P4, Ml-2. basined heel. The last molar is considerably reduced. The
Characteristics: fourth premolar is more compressed and elongate than in
Hyopsodus, with the principal cusp distinctly twinned, and a
Differs from Haplaletes in being smaller and having a more
narrow heel, and small anterior basal cusp. The upper teeth are
symmetrical Ml and M2, each with a small hypocone pos-
of ovate-trigonal outline, with hypocone well developed but
terior to the protocone and often with a smaller pericone an-
from the posterior wing of the protocone, so that it does not
terior to the protocone. [Lower fourth premolar] is relatively
project posterointernally. The external cusps are rather small,
longer and narrower. [Lower first molar] has a more open
with distinct para-, meso-, and metastyles; the conules are
trigonid, with the paraconid well anterior to the metaconid.
small but distinct. The fourth premolar is triangular with the
There is no paraconid on m2 or m3, and all lower molars lack
principal cusp central, and three minor cusps at the internal,
a labial cingulid. M3 and m3 are markedly reduced in size
poster-external, and antero-external angles. The last upper
by comparison with M2 and m2. The occluding cusps and
molar is greatly reduced and simplified. (Matthew, 1915b,
basins on all molars are more rounded and Jess angular than
p. 313.)
those of Haplaletes. This combination of characteristics dis-
tinguishes Dorraletes from other hyopsodontid condylarths
Average length of ml: 2.3-2.6 mm.
as well. (Gingerich, 1983, pp. 244-45.)
Included species: H. speirianus (known from localities
Average length of ml: 2.1 mm. GC21II, CP20A, CP23A, CP25A, CP25B, CP25C,
Included species: D. diminutivus only (known from localities CP25D, CP27 A, CP28C, CP63, CP64A, CP64B); H. simp-
CP22B, CP24A, NP20E, NP47B). soni (localities CPI3H, [CP17A], CPI 7B, CP18B, CP22C,
CP26C, [CP62B], [CP62C]).
Haplaletes Simpson, 1935 Haplomylus sp. is also reported from localities CP23A,
Type species: Haplaletes disceptatrix Simpson, 1935. CP26D.
 Archaic ungulates ("Condylarthra") 307

Hyopsodus Leidy, 1870 which is about the same height as the protoconid, and the
Type species: Hyopsodus paulus Leidy, 1870. somewhat lower entoconid also show a slightly oblique trans-
Type specimen: USNM 1176, right dentary with p4 fragment, verse arrangement ... crests from the hypoconid, however,
ml-3. give this cusp a distinctly selenodont appearance ... crist[id]
Characteristics: obliqua extends directly to the metaconid ... [Lower third
molar] differs from the anterior molars only in the char-
Hyopsodus exhibits a full complement of 44 teeth ... incisors, acter of the talonid, which is more elongate and tapering,
canine, and first premolar above and below are comparatively with a very large and posteriorly projecting hypoconulid.
simple ... single-rooted teeth ... the larger II with its elon- (From Gazin, 1968, pp. 44-50; see Figures 20.2B, 20.3D,
gate, curved root ... has a nearly erect, somewhat elongate 20.5B.)
and tapering, enamel covered crown ... the smaller I2, I3,
C, and Pl are very much alike, increasing in size posteriorly Average length of ml: 3.5 (est.)-4.8 mm.
to the canine, although not reaching the height of II ... P2 Included species: H. paulus (known from localities NB 14,
is a two-rooted tooth, more elongate anteroposteriorly than SB22C, SB43A, CP5, CP6A, CP20F, CP27D, CP27E,
Pl, often with a very weak anterior style and a somewhat [CP31E], [CP32B], CP34B and/or C, [CP36A], [CP36B],
more distinct posterior style ... the cingulum may be de- CP38A, CP38B, [CP38C]); H. miticulus (localities SB24,
veloped posterointernally ... P3 is invariably a three-rooted SB40,CP20A,CP20B,CP20C,CP20D,CP25A,CP25B,
tooth with a prominent, single outer cusp and a narrow lin- [CP25C], [CP25D], [CP25E], CP25H, [CP27A], CP27B,
gual talon exhibiting a smaller but usually prominent deute- CP63, [CP64B], [NP49]); H. loomisi (localities CP19,
rocone . . . in P4 the primary cusp and the deuterocone are CP20B, CP20C, [CP23A], CP25A, CP25B, CP25C, CP63,
about equal in height ... but there is no evidence of trito- NP49); H. powellianus (localities SB25A, CP20B, CP20C,
cone or evidence of an hypocone ... the upper molars are CP20D, CP27B, CP27C, [CP64C]); H. wortmani (locali-
low bunodont teeth and the anterior two are nearly quadri- ties SB21, SB22A, SB22B, SB22C, ?SB24, [SB40],
lateral with six well-defined cusps ... in the first two [mo- [CP23B], CP25F, CP25G, CP25H, CP27B, CP27C,
lars] the conical paracone and metacone are of equal size and CP27D, CP34B and/or C, [CP64B], [CP64C]); H. mi-
joined together and to the anteroexternal and posteroexter- nor (localities CP20A, CP20B, CP20C); H. latidens (lo-
nal styles by a thin crest which in occlusal view is a nearly calities CP20A, CP20B, CP20C, CP20D, CP34B and/or
straight line, except at the styles where there is a sharp but C, [CP64A], [CP64B]); H. mentalis (localities CP25D,
very short outward flexure, and where this crest joins the [CP25G], CP25H, [CP28B], CP34B); H. wolcottianus
prominent outer cingulum ... there is no mesostyle ... lin- (localities SB22A, SB22B, [CP25F], [CP25H], CP27D,
gually the protocone has about the same height as the outer CP34B and/or C); H. minusculus (localities CP5, CP25J,
cusps, but appears larger, exhibiting a more obtuse conical CP25K, [CP27D], [CP27E], CP30, CP31E, CP34B and/or
form ... an anterolateral crest from the protocone joins the C); H. marshi (locality CP34D); H. despiciens (locali-
small protoconule ... the metaconule is a small, isolated ties CP34D, CP65); H. lepidus (localities CP34B and/or
cone about midway between the protocone and metacone C, CP34D, CP35); H. uintensis (= H. fastigatus) (local-
in early and middle Eocene ... in later Eocene the meta- ities [SB42B], SB43A, SB43B, CP6A, CP29B, CP29C,
conule has become aligned with the hypocone ... [Lower [CP36B], CP37 A, CP37B, CP39IIB); H. sholemi (locality
second premolar], as well as being larger than pl, is elon- NP23A); H. tonksi (locality CP32B). (Hyopsodus simplex
gate and relatively broader posteriorly ... anterior portion of is an indeterminate synonym of H. loomsi, H. latidens, or
p3 is very much like p2, but posterolingually it shows de- H. minor according to Gingerich, 1976).
velopment of a third crest to the cingulum ... [Lower third Hyopsodus sp. is also reported from localities (CCI),
premolar], as well as p4 is invariably two-rooted ... [Lower (CC50), CP6A, CP18B, CP20A, CP20B, CP20F, CP26D,
fourth premolar], is larger and relatively broader anteriorly CP27E, CP28B, CP28C, CP30, CP31A, CP31B, CP31C,
than p3, with a somewhat stronger, often more lingually di- CP31D, CP34A, CP36B, CP64A.
rected anterior crest ... it is distinctive in the development
Comments: Hyopsodus is also reported from Europe (Russell,
of the metaconid which is high, relatively lingual to the pro-
1964) and Asia (Russell and Zhai, 1987).
toconid ... there appears to be a progressive development
of the talonid of p4 . . . the lower molars . . . with com- Litomylus Simpson, 1935
paratively high conical cusps tending toward a selenodont
Type species: Litomylus dissentaneus Simpson, 1935.
pattern ... the trigonid portion consists essentially of two
Type specimen: USNM 9425, left dentary with p3-4, ml-3.
cusps, the protoconid and metaconid in a distinctly oblique Characteristics:
arrangement ... crest from the anterior wall of the proto-
conid turns sharply inward and downward, tending to be- [Lower fourth premolar], trenchant, paraconid rudimentary,
come nearly parallel to the posterior wall of the trigonid ... metaconid distinct, separated from the protoconid by a small
on the talonid portion of the anterior molars the hypoconid, pit. Molar cusps rather bunodont but acute, paraconids
 J. David Archibald 308

reduced and median, M3/m3 little reduced, hypoconulid of Haplomylus, but differs in lacking any trace of a metacone on
m3 projecting. Hypocones of Ml-2 relatively large and in- P4, in having more rounded, bunodont cusps, in having rela-
ternal. M3 transverse, triangular. (Simpson, 1935, p. 243.) tively broader M 1 and M2, and in having a more rounded M3
with a distinct posterior cingulum. (Gingerich, 1983, p. 252.)
Average length of ml: 2.9-3.7 (est.) mm.
Included species: L. dissentaneus (= L. scaphicus and L. Average length of ml: unknown. Length of Ml: 4.0 mm.
scaphiscus) (known from localities SB23G, SB23H, Included species: U. latomius only, known from locality
CP13B, CP13C, ?CP13D, CPl4A, CP14B, CP16A, SB20A only.
CP16B, NPlC, NP19C, NP19IIA, [NP19IIB]); L. osce-
olae (locality SB23G or SB23H); L. aequidens (locality
SB23G and/or SB23H); L(?) alphamon (localities NP3E, BULBULODENTATA NEW
NP3F); L. ishami (localities CP14C, CP14D, [CP21B], Bulbulodentata is named to include Mioclaenidae and Periptychi-
[NP20D]); L. orthronepius (localities NP7B, NP 17); L. n. dae. The name is from Greek using the diminutive for "bulbous"
sp. (locality NPlC). and the word for "teeth." It is in reference to the swollen aspect of
Litomylus sp. is also reported from localities CP13E, the premolars found in both families. This may be a synapomorphy
?CB15A, ?CP15B, CP22A, CP26A. shared by the members of Bulbulodentata.
Comments: According to Van Valen (1978), Oxytomodon
Gazin, 194lb is a synonym of Litomylus. Cifelli (1983) MIOCLAENIDAE OSBORN AND EARLE, 1895
regards this genus as separate and referable to Arctocy-
onidae. The latter convention is followed here. Characteristics and comments:

P4/p4 more or less enlarged and sometimes inflated. [Lower

Oxyprimus Van Valen, 1978
fourth premolar] generally relatively simple, with small talo-
Type species: Oxyprimus erickseni Van Valen, 1978.
nid. Molar paraconids reduced, internal, fusing with meta-
Type specimen: UMVP 1561, right ml.
conids. Molar talonids generally open, entoconids reduced,
Characteristics and comments:
fusing with hypoconulid and becoming vestigial. M3/m3 of-
Trigonid walls nearly vertical; p4 rather narrow; molar talo- ten more or less reduced. Ml-2 with very weak or no hypo-
nids larger than in Protungulatum, upper molars rather rect- cone, posterior cingulum tending to run to tip of protocone.
angular. (Van Valen, 1978, p. 53.) (Simpson's 1937 definition ofMioclaeninae, p. 232; see Fig-
Archibald (1982, p. 195) noted that "in contrast to Protun- ure 20.4D.)
gulatum donnae, Oxyprimus erickseni has ( 1) a large p4 meta-
conid, (2) a longer p4 relative to ml, (3) a relatively longer and Although originally recognized as a family by Osborn and Earle
wider talonid on m 1-2, (4) a probably more close! y appressed in 1895, this taxon was reduced to a subfamily ofHyopsodontidae by
paraconid and metaconid on the lower molars, and (5) a rel- both Matthew (1937) and Simpson (1937). In neither the brief com-
atively shorter talonid on m3." These character states were ments of Matthew ( 1937) nor the longer commentary of Simpson
argued to be apomorphic relative to their respective states (1937) is it made clear why these authors grouped the hyopsodon-
in P.** donnae* and although found in some other primitive tids and mioclaenids in a single family. As Cifelli (1983) points
ungulates are best represented in Hyopsodontidae. The most out, as visualized by Matthew (1915b, 1937), Hyopsodontidae (in-
recent information suggests that the first and fourth, and pos- cluding Mioclaeninae) was a wastebasket group for small primitive
sibly the third and fifth, of these character states are primitive ungulates that could not be aligned with more clearly derived taxa.
for ungulates (Nessov, Archibald, and Kielan-Jaworowska, In 1978, Van Valen returned the mioclaenines to familial sta-
1997). tus. Little reason was given for this change, but in his "provisional
phylogeny" in his Figure 2, it appears that he felt hyopsodontine
Average length of ml: 3.5-? mm. hyopsodontids and mioclaenids have their ancestry from separate
Included species: 0. erikseni (known from localities [NP7 A], taxa within the Arctocyonidae as he recognizes this latter family.
NP15C, NP16A); 0. galadrielae (localities CP12A, The recognition of Mioclaenidae does appear to have considerable
[CP13A]); 0. putorius (localities CP12A, [CP13A]). merit, and it seems plausible that a majority of the North Ameri-
Oxyprimus sp. may occur at localities NPlC, NP17. Luo can taxa traditionally placed in this family do form a monophyletic
(1991) argues that 0. erikseni is present at locality NPl5A. tax on.
Several probably derived character states - mild to extreme in-
Utemylus Gingerich, 1983
flation of the premolars, simplification of premolars and molars by
Type species: Utemylus latomius Gingerich, 1983.
loss of various cusps, and reduction in size of M3/m3 - may unite at
Type specimen: YPM-PU 14583, left maxilla with Ml-3.
least some of the genera in this family. The other characters recog-
Characteristics:
nized by Simpson in his 1937 diagnosis of mioclaenines (see earlier)
Differs from all Hyopsodontidae except Haplomylus in hav- may also prove to be useful, but for now appear to represent ple-
ing a postprotocingulum in place of a hypocone. Resembles siomorphic conditions. According to Cifelli (1983), the characters
 Archaic ungulates ("Condylarthra") 309

just described appear to unite a cohesive group including Choero- posteriorly and more bunodont, mesostyle lacking, parastylar
claenus, Ellipsodon, Tiznatzinia, and Mioclaenus, although he notes lobe broader and more protruding, internal cingulum com-
that Choeroclaenus does not have a reduced M3/m3. Johnston and plete; lower molars with ectocingulid, metaconid proportion-
Fox ( 1984) subsequently described Bubogonia saskia from the Puer- ately much larger, paraconid not as reduced and not labially
can of Saskatchewan. They indicate that it is most similar to Cho- shifted, trigonid taller than in P. griphus and P. opisthacus;
eroclaenus among mioclaenines and may be viewed as a structural apices of cusps and crests blunter, talonid basin shallower, en-
antecedent of this genus. tocristid lower and m3 hypoconulid less projecting than in P.
The remaining genera referred to this family, Promioclaenus, opisthacus. Similarities with lower molars of Choeroclaenus
Litaletes, andProtoselene, cannot be aligned with the preceding taxa turgidunculus (Cope, 1888) also evident, but lower molars
with much confidence. Although the M3/m3 is reduced in Promio- of B. saskia larger, less swollen and less reduced paraconid.
claenus as in most of the preceding taxa, it shares with Litaletes a (Johnston and Fox, 1984, p. 202.)
relatively molarized p4. Cifelli (1983) suggests that this is probably
Average length of ml: unknown. Average length of m2:
best viewed as the primitive condition for ungulates, thus telling
5.9mm.
little concerning affinities.
Included species: B. saskia only, known from locality NP7B
Where Mioclaenidae should be placed within Ungulata remains
only.
uncertain, although a possible sister-taxon relationship with Perip-
tychidae is worthy of further exploration. The single character or
Choeroclaenus Simpson, 1937
character complex that suggests this relationship is the trend to in-
Type species: Choeroclaenus turgidunculus (Cope, 1888) (=
flate the premolars. Prothero, Manning, and Fischer (1988) united
Mioclaenus turgidunculus).
Protungulatum**, Periptychidae, and Hyopsodontidae as an unre-
Type specimen: AMNH 3291, left maxilla with P4, Ml-2.
solved tritomy based on the character "P3-4 protocone and para-
Characteristics:
cone bulbous." These authors included both mioclaenines and hy-
opsodontines in Hyopsodontidae. It is a judgment as to how bulbous P4/p4 bulbous, inflated. [Lower fourth premolar] without an-
(or inflated) a tooth should be before it suggests affinities with taxa terior cuspule, paraconid or metaconid, talonid very small,
that possess unmistakably bulbous teeth. I cannot agree with these with one faintly crested cuspule. Molar paraconids small but
authors that Protungulatum** or hyopsodontines show this trend distinct, nearly confluent with metaconids but not wholly in-
and hence cannot find any basis (other than symplesiomorphies) to ternal. Entoconids distinct and about as high as hypoconulids,
unite them. As a very tentative assessment, Mioclaenidae and Perip- molar talonids basined. M3/m3 little or not reduced. [Lower
tychidae are united on the admittedly weak evidence of premolar third molar] with projecting hypoconulid. Ml-3 transverse,
inflation. with sharp external, anterior, and posterior cingula. Hypocone
Two related genera, Phenacodaptes andApheliscus, have at times indistinct, posterior cingulum tending to connect with proto-
been aligned with the mioclaenine or hyopsodontine Hyopsodonti- cone tip. M3 with well-developed metacone. Conules small,
dae (e.g., McKenna, 1960; Rose, 1981 ). McKenna ( 1960) suggested distinct. Cusps low but sharp and clear-cut, crenulations and
a relationship between mioclaenids (a subfamily in Mc Kenna, 1960) proliferation of minor cuspules slight or absent. (Simpson,
such as Promioclaenus and Apheliscus; Rose (1981) drew particu- 1937, p. 232.)
lar attention to similarities of the latter genus to the hyopsodontid,
Average length of ml: 4.0 mm.
Litomylus dissentaneus. Gazin (1959) and Bown (1980) suggested
Included species: C. turgidunculus only (known from local-
Phenacodaptes and Apheliscus are pentacodontines. The posses-
ities SB23A, SB23B).
sion of relatively simple posterior premolars, slight inflation of the
P4, and a small or no hypocone on upper molars suggests mio-
Ellipsodon Scott, 1892 (synonym: Bomburia
claenid affinities. Accordingly, these two genera are very question-
Van Valen, 1978)
ably aligned with Mioclaenidae.
Type species: Ellipsodon inaequidens (Cope, l 884a) (= Tri-
Mioclaenidae is also known from South America (Muizon and
centes inaequidens).
Marshall, l 987a, l 987b, who included Mioclaeninae within Hyop-
Type specimen: AMNH 3095, parts of skull with left P2-3,
sodontidae).
Ml-3 and right P4, M2-3.
Characteristics:
Bubogonia Johnston and Fox, 1984
Type species: Bubogonia saskia Johnston and Fox, 1984. Molar paraconids internal, but usually nearly if not entirely
Type specimen: UA 15105, an incomplete right dentary with indistinguishable. Entoconids indistinguishable from hypo-
m2-3. conulids, reduced to a low, bordering rim. Reduction of
Characteristics: M3/m3 extreme. Molar trigonid cusps (ml-2) closely ap-
pressed, turgid. Lower jaws short and stout. Premolars short
Molars similar to Protoselene griphus (Gazin, 194l[b]) and and wide, crowded (obliquely set in rami), simple with basins
Protoselene opisthacus (Cope) 1882c, but differing as fol- of heels shallow and poorly defined but wide. (Wilson, 1956,
lows: on M3, B. saskia, protocone less compressed antero- p. 112.)
 J. David Archibald 310

Average length of ml: 3.8-5.8 (est.) mm. Promioclaenus Trouessart, 1904

Included species: E. inaequidens (known from localities Type species: Promioclaenus acolytus (Cope, 1882c) (=Hy-
SB23E, SB23F, SBGG, SB23H); E. priscus (localities opsodus acolytus). (Note: Wilson, 1952, listed P. lemuro-
SB23A, SB23B); E. witkoi (locality SB23B); E. yotankae ides [Matthew, 1897) as the genotypic species. Gazin,
(locality ?SB23C); E. grangeri (localities ?SB23B, SB23F, 1956, notes that P. acolytus was listed first by Trouessart,
CPlG). 1904, under his proposed genus Promioclaenus, hence this
Ellipsodon sp. may also occur at locality CPlB. The species is the genotypic species.)
occurrence of Ellipsodon sp. reported by Archibald et al. Type specimen: AMNH 3208, left maxilla with P3-4, Ml-2,
(1987) at Swain Quarry (locality CP14A, their locality 38) left dentary with p3-4, ml-3.
was an editorial error and should have referred to Litaletes. Characteristics:

Comments: Van Valen (1978) named the new genus, Bom- P4/p4 cuspidate, more or less enlarged, but not really inflated.
buria, including in it only one species based on Matthew's The p4 without distinct paraconid, metaconid absent or rudi-
(1937) Ellipsodon priscus. Cifelli (1983) argued that Bom- mentary, protoconid inflated rather than bladelike; talonid
buria was indistinguishable from Ellipsodon as the latter basined and relatively large. Molar hypocones weak (or ab-
tax on is currently recognized and accordingly returned this sent?), metaconules present. Molar trigonids with paraconids
species to Ellipsodon. internal, fusing with metaconids; moderately separated meta-
conids and protoconids. Molar talonids with relatively wide
Litaletes Simpson, 1935 (synonym: Jepsenia Gazin, 1939) basins, entoconids generally indistinct and fusing with hypo-
Type species: Litaletes disjunctus Simpson, 1935. conulids (especially on m2). M3/m3 moderately reduced. M3
Type specimen: USNM 9323, right dentary with c, pl-4, with reduced metacone. Rami relatively long and slender.
ml-3 (m3 is broken). (Wilson, 1956,p.115.)
Characteristics: Average length of ml: 3.3-5.0 (est.) mm.
[Lower fourth premolar] with bladelike main cusp, distinct Included species: P. acolytus (= P. aquilonius) (known from
paraconid and relatively large metaconid, talonid small with localities SB23E, SB23F, SB23G, SB23GG, SB23H,
narrow, rudimentary, open basin. Molar paraconids distinct, SB39A, CPlA, CP13B, CP13C, CP13D, CP14A, CP21A,
small, internal. Entoconids indistinct, fusing with hypo- CP26A, NP19C); P. lemuroides (localities SB23E, SB23F,
conulids, especially on m2. P4 with rudimentary metacone, ?SB23G, SB23H, CPlC, CP21A); P. sheperdi (locality
strong metastyle. Ml-2 with distinct hypocones. M3/m3 not CPlC); P. wilsoni (localities SB23A, SB23B, CPlA); P.
reduced, M3 with strong metacone. (Simpson, 1937, p. 238.) pipiringosi (localities CP16A, CP16B).
Promioclaenus sp. has also been reported from localities
Average length of ml: 3.9-4.8 (est.) mm. CPlA, CB15A, ?CP16A, NP2.
Included species: L. disjunctus (known from localities CPlC,
CP13C, [CP21A], NP19C); L. stembergi (locality CPlC); Protoselene Matthew, 1897 (synonym: Dracoclaenus
L. mantiensis (locality CP14A); L. ondolinde (locality Gazin, 1939)
CP13B); L. gazinensis (locality CPlA).
i
Type species: Protoselene opisthacus (Cope, 1882j) (=Mio-
Litaletes sp. is questionably recognized at localities claenus opisthacus).
CP22A, NP7B, NP47A. Type specimen: AMNH 3275, left dentary with p4, ml-3,
right dentary with ml-3, and unspecified premolars.
Mioclaenus Cope, 1881c Characteristics:
Type species: Mioclaenus turgidus Cope, 1881c.
Premolars not much inflated, lower ones trenchant. Fourth
Type specimen: AMNH 3135, left maxilla with P4, Ml-2,
upper premolar with strong internal cusp (deuterocone) and
left dentary with p4, ml-3, right dentary with p4, ml.
distinct posteroexternal cusp (tritocone ). Third premolar near-
Characteristics:
ly simple. Last lower premolar with strong heel, anterior
[C]anines small, premolars and molars very robust, low- ones simpler. Entoconid well developed on lower molars.
crowned, with simple cusp construction, premolars large, (Matthew, 1897, p. 317.)
much inflated, the cusps convexly conical, molars simple,
Average length of ml: 5.5-5.7 (est.) mm.
with hypocones, paraconids, entoconids and basal cingula
Included species: P. opisthacus (known from localities
more or less obsolete or wholly absent; last molar consider-
SB23E, SB23F, SB23G, SB23GG, SB23H, SB39B,
ably reduced. (Matthew, 1937, p. 195; see Figure 20.4D.)
CP26A); P. bombadili (locality SB23A); P. griphus (lo-
Average length of ml: 6.3 mm. calities SB23C, CPlC); P ?. novissimus (localities CP15A,
Included species: M. turgidus only (known from localities CP16A).
SB23C, SB23E, SB23F, SB23G, SB23GG, SB23H). Protoselene sp. has also been mentioned as possibly
Mioclaenus sp. might also occur at locality NB 1. being present at locality CP13E.
 Archaic ungulates ("Condylarthra") 311

Tiznatzinia Simpson, 1936a Average length of ml: 2.1 (est.)-2.5 mm.

Type species: Tiznatzinia vanderhoofi Simpson, 1936a. Included species: A. insidiosus (known from localities SB24,
Type specimen: UCMP 31264, left dentary with p4, ml-2. CP20A, CP27B); A. nitidus (localities CP13H, ?CP14E,
Characteristics: CPI 7A, CPI 7B, CPl 8A, CP 18B, CP24B, CP25B, CP26C,
CP63); A. wapitiensis (localities CP20A, CP25B).
[Lower fourth premolar) little enlarged, slightly higher than Apheliscus sp. is also reported from localities CP20A,
ml and not wider, with distinct but single-cusped heel, and CP20B, CP20C, CP27D, CP62C, CP63, CP64A.
very small incipient metaconid. Molars low-crowned, cusps
relatively low and blunt, trigonids little elevated above talo- Phenacodaptes Jepsen, 1930
nids. Paraconids more or less reduced, but always present, Type species: Phenacodaptes sabulosus Jepsen, 1930.
closely approximated and usually almost directly anterior to Type specimen: YPM-PU 13302, right dentary with pl-4,
metaconids, partly confluent with the latter. Trigonids with ml-2.
shallow basins, nearly or quite closed. Deep talonid basins Characteristics: The pl single rooted, laterally compressed,
open through a narrow notch on the inner side, hypoconids with small posterior basal cusp; p2 double rooted, recurved,
large, entoconids and hypoconulids smaller and poorly differ- has minute cusp positioned midway on anterior side and
entiated, frequently with other minute cuspules on the talonid single sharp cusp on posterior border; p3 similar in size
crescent. (Simpson, 1936a, pp. 7-8.) to p2 but wider, anterior cusp positioned lower on tooth,
and heel flatter and wider; p4 much larger than p3 with
Average length of ml: 3.2 mm. anterolingual basal cingulum, main cusp large and sharp
Included species: T. vanderhoofi only (known from localities pointed with anterior ridge and three posterior ridges, mid-
SB23A, SB23B). dle one leading to prominent posterior basal cusp; ml
with trigonid slightly wider and higher than talonid, pro-
Comments: When he named Tiznatzinia, Simpson (1936a)
toconid and metaconid subequal and almost opposite each
also referred Mioclaenus turgidunculus and Ellipsodon
other, paraconid small and lingual, hypoconid shorter and
priscus to this genus. Subsequently, Simpson ( 1937) made
slightly less stout than protoconid, entoconid smaller than
T. turgidunculus the type and only species of Choero-
hypoconid, hypoconulid distinct but joined to entoconid by
claenus. Van Valen (1978) synonymized Tiznatzinia with
short rib; external and anterior cingula variable; m2 simi-
Promioclaenus. He referred the type species, T. vander-
lar to ml except in being larger, trigonid wider relative to
hoofi, to the latter genus and referred E. priscus to a new
talonid, and lacking a paraconid; m3 longer and narrower
genus, Bomburia. Cifelli ( 1983) resurrected Tiznatzinia for
than m2, also lacks paraconid, hypoconulid projects from
the type species, T. vanderhoofi, and synonymized Bom-
the rear, metaconid larger than protoconid. (After Jepsen,
buria priscus with Ellipsodon priscus.
1930.)
Average length of ml: 2.8 mm.
Included species: P. sabulosus only (known from localities
MIOCLAENIDAE? APHELISCINAE (MATTHEW, 1918)
CP13G, CP14C, CP14D, CP15B, [CP17A], CP18A,
Apheliscus Cope, 1875a CP26B, CP26C).
Type species: Apheliscus insidiosus (Cope, 1874) (=Proto-
tomus insidiosus).
PERIPTYCHIDAE COPE, 1882I
Type specimen: "The type belongs in the National Museum
collections, but is missing" (Matthew, 1918). Characteristics: Protocone (and sometimes hypocone) base with
Characteristics: Molar cusps bunodont; ml with small para- slight to considerable lingual expansion and apex shifted
conid, m2-3 only with paracristid; lower molars with well- slightly to considerably labially; internal and sometimes external
developed anterior cingulids and remnants of ectocingulids cingulids present on lower molars; molar cusps slightly to notice-
in the hypoflexids; upper molars roughly triangular with ably appressed and talonid length shortened relative to width; molar
prominent posterior and labial cingula and a somewhat paraconid on or near midline and can be absent, posterior premolars
weaker anterior cingulum; Ml-2 with moderately well- with slight or very considerable expansion (bulbous). The last char-
developed hypocone; P4/p4 large and distinctive, p4 with acter state is shared with mioclaenids (see Figure 20.2C, H, 20.3B,
single high trigonid cusp and sharply defined talonid heel, C, 20.5A).
P4 with a large, inflated labial cusp and a lower and smaller Periptychidae is questionably known from Asia (Russell and
lingual cusp. These character states pertain to Clarkforkian Zhai, 1987) and South America (Muizon and Marshall, 1991 ).
specimens of Apheliscus nitidus as described by Rose
(1981). According to Rose (1981), younger Wasatchian- "Mimatuta" Van Valen, 1978 (synonym: Earendil
aged specimens referable to this genus are smaller and Van Valen, 1978)
have less prominent postcingulae and smaller hypocones Type species: Mimatuta morgoth Van Valen, 1978.
on Ml-2. Type specimen: UMVP 1560, right maxilla with M2.
 J. David Archibald 312

Characteristics: Average length of ml: 5.0 mm.

Included species: H. kowalevskianus (known from localities
Primitive periptychids: upper molars quite transverse, proto- SB23A, SB23B).
cone apex central, hypocone absent to moderate, its base not Hemithlaeus sp. is also reported from locality NP16B.
projecting lingually; molar paraconid large, entoconid small.
(Van Valen, 1978, p. 62.) Ectoconus Cope, 1884b
Type species: Ectoconus ditrigonus (Cope, 1882c) (= Perip-
At least "M." morgoth* shows the beginnings of trends
tychus ditrigonus).
that are found in later periptychids but not seen in Prot-
Type specimen: AMNH 3798, right dentary fragment with
ungulatum** donnae*: protocone base with slight lingual
m2.
expansion and apex shifted slightly labially, weak internal
Characteristics: Large periptychines; much less crenulation
cingulid present on lower molars, molars cusps slightly
of enamel than in Carsioptychus or Periptychus; upper mo-
appressed and talonid length short relative to width,
lar ectoflexus may be present because of slight develop-
molar paraconid on or near midline, posterior premolars
ment of para- and metastylar region, which both some-
with some expansion (Archibald, Schoch, and Rigby,
times bear cuspules; last three upper premolars (at least)
1983).
possess a protocone; P3-4 possess well-developed para-
Average length of ml: 3.9-4.1 mm.
and metaconules; p3-4 lack paraconid(?) and metaconid(?)
Included species: "M." morgoth* (known from localities
and not as bulbous as in Carsioptychus or Periptychus (see
NP15B, NP15C, NP16A); "M." minuial* (= E. undomiel)
Figures 20.2H, 20.5A). Ectoconus is questionably known
(localities CP12A, NP16A); "M." makpialutae (locality
from Asia (Russell and Zhai, 1987).
CP12B).
Average length of ml: 10.0 (?)-10.5 mm.
"Mimatuta" sp. is also reported from locality NP7 A.
Included species: E. ditrigonus* (known from localities
Luo (1991) argued that"M." morgoth* is present at locality
SB23A, SB23B, CPlA, CP61B); E. symbolus (locality
NP15A.
CPlA, CPlB).
Comments: Cifelli (1983) suggested that Earendil undomiel Ectoconus sp. also is reported from locality CPlA.
Van Valen (1978) may be a junior synonym of"Mimatuta"
minuial*. After examining casts and photographs of both "Carsioptychus" Simpson, 1936b
taxa I could find no substantive differences. Accordingly, Type species: Carsioptychus coarctatus (Cope, 1883b) (=
I formalize Cifelli's suggestion of synonymy. Periptychus coarctatus).
"Mimatuta" is questionably known from South America Type specimen: AMNH 3775, "teeth of the lower jaw, viz.:
(Muizon and Marshall, 1991). one incisor, three premolars, and two molars, two of the
latter imperfect" (Cope, 1883b, p. 168).
Characteristics: Large periptychines; as in Periptychus: ena-
UNNAMED TAXON: PERIPTYCHINAE + mel on molars and especially premolars highly crenulated,
ANISONCHINAE + CONACODONTINAE all four upper premolars (in species where known) pos-
sess protocone; in comparison to Periptychus: upper pos-
PERIPTYCHINAE OSBORN AND EARLE, 1895 terior premolars and molars transversely wider in occlusal
Characteristics: Small to large periptychids; upper molars possess view with protocone only slightly labial of pericone and
pericone (Haploconus comiculatus autapomorphously developed hypocone on molars, posterior premolar paraconid(?) and
this character state); P4 protostyle very reduced or absent. metaconid(?) slightly smaller with latter cusp not forming
Although the subfamily has not been thoroughly reviewed, a pos- as distinct of a crestlike connection to the protoconid.
sible nesting of taxa can be suggested as following: Carsioptychus Average length of ml: 9.0 (?)-10.0 mm.
and Periptychus are sister taxa, Ectoconus is the sister taxon of these Included species: C. coarctatus* (known from localities
two taxa, and Hemithlaeus is the most distant outgroup of the pre- SB23A, SB23B, SB39II); C. matthewi (localities SB23B,
ceding taxa (see Figures 20.2C, H, 20.3C, 20.5A). CP61B); C. hamaxitus (localities CPIA, CPlB).

Hemithlaeus Cope, 1882i Periptychus Cope, 1881a

Type species: Hemithlaeus kowalevskianus Cope, 1882i. Type species: Periptychus carinidens Cope, 188la.
Type specimen: AMNH 3587, right P2-M2, left pl-2, ml-2. Type specimen: AMNH 3620, dentary fragments with right
Characteristics: Small periptychine; upper molars transver- dp4 and left dp3.
sely wider than in any other periptychines; little or no Characteristics: Large periptychines; as in Carsioptychus:
crenulation of enamel on any teeth, p3-4 lack paraconid enamel on molars and especially premolars highly crenu-
and metaconid, but relatively more bulbous than in Ec- lated, all four upper premolars (in species where known)
toconus. All except possibly the last character state are possess protocone; in comparison to Carsioptychus: up-
primitive for Periptychinae or a more inclusive taxon. per molars subrectangular in occlusal view with protocone
 Archaic ungulates ("Condylarthra") 313

only slightly labial of pericone and hypocone, posterior Included species: "A." sectorius (known from localities
premolar paraconid(?) and metaconid(?) slightly larger SB23C, SB23E, SB23F, SB23G, SB23GG, SB23H,
with latter cusp forming more crestlike connection to the CP13B, CP13C, CP14A, CP14B, CP15A, CP16A,
protoconid (see Figures 20.2C, 20.3C). NP19C, [NP20B],NP47A); "A." athelas* (="A." eowynae
Average length of ml: 10.3-12.0 mm. Van Valen, 1978) (localities CPlA, CPlB, CP61B); "A."
Included species: P. carinidens* (known from localities oligistus* (localities CPlB, [CPlC], NP7B, NP16B,
SB23C, SB23D, SB23DD, SB23E, SB23F, SB23G, NP16C); "A." dracus* (locality CPlC); "A." onostus* (lo-
SB23GG, SB23H, CP14A, CB15A); P. gilmorei (locality calities CPlA, CPlC); "A." fortunatus (locality GC19);
CPlC); P. superstes (localities SB20A, ?SB39A, SB39B, "A." willeyi* (locality CP14A).
[SB39D]). "Anisonchus" sp. is also known from localities (NB 1),
NP17. The report of Archibald et al. (1987) that "Anison-
chus" occurs at localities SB23A and SB23B is incorrect.
ANISONCHINAE OSBORN AND EARLE, 1895
These localities have yielded on! y Gillisonchus gillianus*.
Characteristics: Small periptychids; paraconid on posterior premo-
lars is either large, distinct, and almost columnar (Gillisonchus Comments: The complexity within this genus was suggested
and "Anisonchus") or is absent, probably because of fusion to the by Van Valen's (1978) recognition of two subgenera,
anterior aspect of the trigonid (all species of Haploconus except "Anisonchus" (Anisonchus) and "A." (Mithrandir), and by
possibly H. elachistus); P3 protocone present (Gillisonchus and Rigby's (1981) examination of the genus and his nam-
"Anisonchus") or absent (all species of Haploconus except possibly ing of a new genus, Gillisonchus, for the species "A."
H. elachistus). gillianus*. Cifelli (1983) argued that "A." oligistus*, which
is Van Valen's (1978) type species for the subgenus Mith-
Gillisonchus Rigby, 1981 randir, is not generically distinct from "A." gillianus*, but
Type species: Gillisonchus gillianus (Cope, 1882h) (=Hap- is generically distinct from "A." sectorius. He argued that
loconus gillianus). Mithrandir should be elevated to generic rank with the re-
Type specimen: AMNH 3543, fragments of maxilla and den- sult that Gillisonchus is a junior synonym. Cifelli's (1983)
tary and a few skeletal remains. assessment may be correct, but no justifications were given.
Characteristics: Small anisonchine. Until a more thoroughly documented analysis is completed
and published for Anisonchus, I prefer to retain Gillison-
The hypocone and lingual part of the anterior cingulum close- chus. The matter is even more complicated because one or
ly approximate the apex of the protocone in the upper molars; more of the species now included inAnisonchus appears to
the upper, molar conules (crest confluences) are located near share a more recent common ancestry with species of Hap-
the bases of the paracone and metacone; the preparaconule loconus than with other species of Anisonchus. This ren-
wing and postmetaconule wing are confluent with, but do not ders Anisonchus, in any present permutation, as nonmono-
interrupt, the anterior and posterior cingula; the lower premo- phyletic (hence the name is enclosed in quotation marks).
lars are inflated at the base and are compressed apically but
not to the degree that the whole tooth is compresses in species
Haploconus Cope, 1882g
of Anisonchus; the paraconid of p3-4 is low and prominent
Type species: Haploconus lineatus Cope, 1882g (= Mio-
on the anterior median face. (Rigby, 1981, p. 97.)
claenus angustus Cope, 188lc,fide Matthew, 1937).
Average length of m 1: 4 .4 mm. Type specimen of Haploconus lineatus: AMNH 3425, partial
Included species: G. gillianus* only (known from localities skull and dentaries with right P3-M3, left P4-M3, right
SB23A, SB23B). p3-m3, and left ml-m3.
(Note: The notation in Archibald et al., 1987, Table 3.1 Type specimen of Mioclaenus angustus: AMNH 3477, right
indicating that "Anisonchus" occurs at these localities is dentary with p4-m3.
incorrect. These occurrences pertain to Gillisonchus.) Characteristics: Small anisonchines; P3 protocone absent, ex-
cept possibly in H. elachistus (this may be dp3 in the spec-
"Anisonchus" Cope, 1881c imen preserving this tooth site); paraconid on posterior
Type species: "Anisonchus" sectorius (Cope, 188 lc) (=Mio- premolars absent, possibly by fusion to the anterior aspect
claenus sectorius). of the trigonid, except possibly in H. elachistus.
Type specimen: AMNH 3527, right maxilla and dentary with Average length of ml: 3.5 (?)-5.0 mm.
P4-M2/p4-m2. Included species: H. angustus* (known from localities
Characteristics: Small anisonchines; all species of "Anison- SB23D, SB23E, SB23F, SB23G, SBGG); H. comiculatus
chus" for which appropriate specimens are known have a (locality SB23E); H. elachistus* (localities CPlA, CPlB);
protocone on P3, and have a large, distinct, almost colum- H. inopinatus* (localites SB39II, CPlC).
nar paraconid on posterior premolars. Haploconus sp. has also been identified at localities
Average length of ml: 4.0-6.0 mm (approx.). CPlA, CP14A.
 J. David Archibald 314

CONACODONTINAE ARCHIBALD, SCHOCH, AND RIGBY, Included species: O.** apiculatus (known from localities
1983 SB23A, SB23B, CPlA); O.** agapetillus (= Fimbrethil
ambaronae Van Valen, 1978) (localities SB23A, NP7B);
Characteristics: Small periptychids; hypocone on Ml-3 large and O.** priscilla* (locality SB23A); O.** ferronensis* (local-
situated very lingually; protocone on P3 absent; probably greater
ities CPlA, NP16B); questionably O.** marshater(local-
tendency for precingulum to encircle protocone and contact hypo-
ities CPlB, NP16C).
cone (Archibald, Schoch, and Rigby, 1983b; see Figure 20.3B).
Oxyacodon** n. sp. is known from locality (CPlA).

Comments: The most recent data suggest there are no synap-

Conacodon Matthew, 1897
Type species: Conacodon entoconus (Cope, 1882h) (=Hap- omorphies that unite all species of Oxyacodon**. All spec-
ies of this genus and the monophyletic clade Conacodon
loconus entoconus).
Type specimen: AMNH 3462, right maxilla with P3-M3. (all three species) form a polytomy; hence Oxyaco-
Characteristics: Small conacodontine; paraconids usually ab- don** is a mixotaxon (see Archibald, 1994).
sent on p3-4 except in C. cophater; premolars inflated and Van Valen (1978) named Oxyacodon** josephi from
P4/p4 larger than Ml/ml. Stylar shelf on Ml-3 narrow Mantua Lentil. Archibald, Rigby, and Robison (1983) in-
and often discontinuous; metacingulum and postcingu- dicated that it does not possess a lingually expanded hypo-
lum usually forming continuous shelf with postmetaconule cone as in Oxyacodon** (and Conacodon) and hence there
wing not or only slightly contacting metacingulum. As are no synapomorphies to link it with conacodontines.
compared to Oxyacodon**, paraconid smaller on posterior
molars, talonid length is reduced, both partial labial and
PERIPTYCHIDAE?
lingual cingulum can be present, and apices of trigon,
trigonid, and talonid are more closely appressed Escatepos Reynolds, 1936
(Archibald, Schoch, and Rigby, 1983; see Figure 20.3B). Type species: Escatepos campi Reynolds, 1936.
Average length of ml: 3.5-5.6 mm. Type specimen: UCMP 36659.
Included species: C. entoconus (known from localities Characteristics:
SB23A, SB23B, CP61B); C. cophater (localities SB23A,
SB23B); C. kohlbergeri (locality SB23A); C. utahensis [r]ight maxilla retains five teeth, two molars and three pre-
(localities CPlA, CPlB). molars; left maxilla, the fractured remains of three teeth ....
Conacodon sp. is also reported from localities CP61A, Mandible apparently collapsed in center of palate at time
CP61B. of fossilization and cemented there. (Reynolds, 1936,
p. 207.)
Comments: The first publication of the name Conacodon ap-
Average length of ml: unknown. Length of Ml: 4.0 mm.
peared in Matthew, 1897, but without any indication that
Included species: E. campi only, known from locality SB23A
this was a newly proposed genus. The first substantial ac-
count of Conacodon appeared in Matthew, 1937. only.
Comments: The affinities of this apparently poorly preserved
specimen are not well established. Reynolds (1936) orig-
Oxyacodon** Osborn and Earle, 1895 inally described it as an oxyclaenine creodont; Van Valen
Type species: Oxyacodon apiculatus Osborn and Earle, 1895. (1978) considered it to be a synonym of Oxyacodon**
Type specimen: AMNH 816, crushed and distorted left den- agapetillus. Archibald, Rigby, and Robison (1983) could
tary with p4, ml-2. not detect any expanded hypocone on the specimen and
Characteristics: Small conacodontines; paraconid small, sit- hence did not accept its referral to Oxyacodon** agapetil-
uated low and slightly lingual of midline on p3-4 or can be lus. Van Valen (pers. comm., 1993) suggested that the type
absent; premolars slightly inflated but P4/p4 smaller than and only known specimen is a deciduous dentition of a
or subequal to Ml/ml. Stylar shelf Ml-3 wider than in
periptychid.
"Anisonchus," Conacodon, Haploconus, and Hemithlaeus;
metacingulum and postmetaconule wing usually forming Maiorana Van Valen, 1978
continuous shelf with little or contact of metacingulum and Type species: Maiorana noctiluca Van Valen, 1978.
postcingulum. Paraconid on ml-3 small and on midline; Type specimen: YPM-PU 16667, skull fragment with left C,
talonid length not as reduced as in aforementioned genera;
P3-M3.
narrow partial labial cingulum present but lingual cingu- Characteristics:
lum usually absent. Apices of trig on, trigonid, and talonid
cusps not as closely appressed as in aforementioned genera Upper molars only weakly transverse, protocone apex tall and
(Archibald, Rigby, and Robison, 1983; Archibald, Schoch, central, hypocone small, its base not projecting lingually; mci-
and Rigby, 1983). lar paraconid and entoconid large; canines small; mandibular
Average length of ml: 2.9-4.2 mm. condyle high. (Van Valen, 1978, p. 61.)
 Archaic ungulates ("Condylarthra") 315

Average length of ml: not published. Length of M2: 4.5 mm. departure might at first seem quite radical, if continued analysis
Included species: M. noctiluca only, known from locality upholds these cladistic relationships, it seems reasonable to place
CP12A only. taxa such as mesonychids and some arctocyonids with those taxa
such as cetaceans with which they share the most recent common
Comments: Cifelli (1983) argued that although Maiorana is
ancestor rather than with taxa such as other "condylarths" that are
similar to "Mimatuta," it does not have the "shallow lingual
similar by virtue of retaining primitive characters (no matter how nu-
slope on the upper molars" characteristic of periptychids.
merous). There is, however, a considerable problem in discovering
His preference was to refer Maiorana to the Oxyclaeninae
synapomorphies that unite such a morphologically heterogeneous
because it was primitive. I prefer not to use primitive char-
taxon. Prothero, Manning, and Fischer (1988) briefly discuss these
acters for reference to a taxon. Cifelli may well be correct,
characters. (See also comments under Characteristics for [Family]
but for now it seems more prudent to retain Maiorana as
Arctocyonidae; see Figures 20.lA, 20.2E, 20.4A, B, 20.5E.)
incertae sedis within the Periptychidae.

Tinuviel Van Valen, 1978 Baioconodon Gazin, 1941a

Type species: Tinuviel eurydice Van Valen, 1978. Type species: Baioconodon denverensis Gazin, 1941 a.
Type specimen: UMVP 2226, left m3. Type specimen: USNM 16621, right dentary with m2-3.
Characteristics: Characteristics:

Upper molars weakly transverse, pericone as large as hypo- Trigonid of lower molars moderately elevated, with para-
cone and placed symmetrically lingual to protocone; molar conid lingual and well defined. Ridge extending anterolin-
paraconid nearly as large as metaconid; cingular cusp lin- guall y from hypoconidjoins posterior wall oftrigonid at dis-
gual of paraconid; cusps rather blunt, protocone wearing from tinctly lingual point. [Lower third molar] unreduced, with
apex ratherthan from labial surface. (Van Valen, 1978, p. 61.) hypoconulid crenulate and talonid basin exhibiting rugae.
External cingulae developed. (Gazin, 194la, p. 292.)
Average length of ml: unavailable. Length of m3: 5.0 mm.
Included species: T. eurydice only, known from locality Average length of m 1: 6. 8 (est.) mm.
NP16C only. Included species: B. denverensis (known from localities
Tinuviel? n. sp. is also reported from locality CP61A. CP61A, NP7B, NP7C, [NP17]); B. antiquus (locality
SB23B); B. ?n. sp. (locality SB39II).
CETUNGULATA IRWIN AND WILSON, 1993
Ragnarok Van Valen, 1978
In their molecular study of mammalian phylogeny, Irwin and Wil-
Type species: Ragnarok harbichti Van Valen, 1978.
son (1993) used' the term "cetungulates" to include Cetacea, Ar-
Type specimen: AMNH 35983, right dentary with ml-3.
tiodactyla, and Perissodactyla. They argued that Proboscidea is not
Characteristics:
likely part of this clade, although they did not exclude the possibil-
ity that proboscideans and cetungulates form a monophyletic group. Molars moderately low-crowned and bulbous: upper molars
If Prothero, Manning, and Fischer (1988) are correct, Cetungulata and lingual lobe of P4 transverse. (Van Valen, 1978.)
would include Cete as the sister taxon with what are here termed
Taxeopoda. If the alignment suggested by Novacek (1992) is correct, As Sloan et al. (1986) noted, Middleton (unpublished
Cetungulata includes Cete (as used here) and Artiodactyla. Thewis- dissertation) has argued that this genus is synonymous with
sen (1994) discussed the various concepts of Cetungulata but did not the next listed genus, Baioconodon. Until such time as this
prefer one arrangement over the other. If one follows his cladistic is published by its author, the two genera should continue
analysis, however, Cetungulata would (among extant mammals) in- to be recognized.
clude Cete, Perissodactyla, and Artiodactyla, much as advocated by Average length of ml: 4.4-5.3 mm.
Irwin and Wilson (1993). Tethytheria (at least proboscideans and Included species: R. harbichti (known from localities NP l 5C,
sirenians, but possibly including hyracoids) were not included in NP16A); R. nordicum (localities CP12A, NP16A); R. wo-
Thewissen's analysis. Whatever the more specific relationship that vokae (locality CP12B); R. engdahli (locality NP16A).
is eventually determined, there is little doubt that Cete falls some- Ragnarok sp. has also been reported from locality NP7 A.
where among ungulates based on both molecular and anatomical
evidence. Stelocyon Gingerich, 1978
Type species: Stelocyon arctylos Gingerich, 1978.
CETE LINNAEUS, 1758 Type specimen: YPM-PU 17929, right dentary with p4, ml-2
and associated leftdentary with ml-2.
The taxa within the Cete include, according to Prothero, Manning, Characteristics:
and Fischer (1988), some genera traditionally (e.g., Van Valen, 1978)
placed in the archaic ungulate families Arctocyonidae, Hapalodec- Differs markedly from Colpoclaenus and other Arctocyoni-
tidae, and Mesonychidae, along with the cetaceans. Although this dae in having much narrower, higher-crowned lower molars.
 J. David Archibald 316

Resembles some triisodontine Mesonychidae (Triisodon, Eo- M3 reduced and Ml as large or larger than M2, instead of
conodon) but differs from them in the structure of the molar smaller. The third upper premolar is three-rooted without
trigonids, especially on m2. (Gingerich, 1978, p. 2.) well-developed deuterocone. The paraconid is perhaps some-
what smaller than in Triisodon, and the proto- and metaconid
Length of ml: 7.5 mm. higher and more equal in size. The position of the paraconid is
Included species: S. arctylos only, known from locality not entirely constant, but it is usually submedian on ml, inter-
CP13B only. nal on m2 as in Triisodon. The third lower molar is reduced.
A character observed in two specimens, and perhaps a con-
Oxyclaenus Cope, 1884a (synonym: Thangorodrim stant one, is the position of the mental foramen underneath
Van Valen, 1978) the second premolar instead of between the third and fourth.
Type species: Oxyclaenus cuspidatus (Cope, 1884a) (=Mio- This is associated with the short deep symphyseal part of
claenus cuspidatus). the jaw. The symphysis is ovate and widest behind, extend-
Type specimen: AMNH 3252, parts of maxillae with premo- ing back to beneath the third premolar, while in Triisodon
lars and molars; associated dentaries. it is widest anteriorly and pointed behind. (Matthew, 1897,
Characteristics: p. 282.)
Upper molars tritubercular with conic cusps of moderate Average length of ml: 8.0 (est.)-13.6 (est.) mm.
height, hypocones rudimentary or absent, transverse diam- Included species: G. levisanus (known from localities
eter considerably exceeding the antero-posterior, strong en- SB23E, SB23F, SB23G, SB23H, CP13B, CP14A); G. cra-
circling cingula except upon inner face of protocone. Exter- ssicuspis (locality SB23E and/or SB23H); G. hiawathae
nal face of protocone fiat, conules small but distinct. Lower (locality CPlB).
molars with trigonid higher than talonid, metaconid and pro- Goniacodon sp. has also been reported as possibly pre-
toconid subequal, well separated, paraconid smaller, nearly sent at locality CPIC.
internal in position, talonid (except on m3) as wide as trigo-
nid, deeply basined, the basin more open internally than in Eoconodon Matthew and Granger, 1921
Loxolophus. Cusps oflower molars not inflated, tending to be Type species: Eoconodon coryphaeus (Cope, 1882a) (=Tri-
conic. Hypoconulid small except on m3, which has a narrow, isodon heilprinianus).
more or less elongate talonid pinched up posteriorly into a Type specimen: AMNH 3224, a dentary fragment with a
large hypoconulid equaling the hypoconid in height. damaged molar. According to Van Valen (1978), the type
Fourth premolar above and below with high, acute, con- specimen of Triisodon heilprinianus Cope, l 882a is an un-
ical protocone; P4 with large inner cusp and small cusps at worn molar of the taeniodont Conoryctes comma. The next
outer angles; p4 with protocone somewhat compressed and available name is Sarcothraustes coryphaeus Cope, 1885.
prominent sharp anterior basal cusp and heel. Anterior teeth Type specimen: AMNH 3181, an incomplete skull. I follow
unknown. (Matthew, 1937, p. 39.) Van Valen's usage in accepting Sarcothraustes coryphaeus
Cope, 1885 as the subsequently designated type species.
Average length of ml: 4.7-5.5 (est.) mm.
Characteristics:
Included species: 0. cuspidatus (known from localities
SB23A, SB23B); 0. simplex (possibly= Carcinodonfilho- Dentition unreduced, I3/3, C 1/1, P4/4, M3/3. Molars typical-
lianus; see Carcinodon) (localities SB23A, SB23B, ly tritubercular, robust and massive. Upper molars wider than
[CP13A], NPl 7); 0. pearcei (locality CPlC); 0. corax long, subquadrate. Paracones and metacones round-conic,
(localities NP7B, NP7C); 0. pugnax (localities CPlA, protocone subcrescentic, conules and rudimentary hypocones
NP16C, NP19A). on Ml-2, heavy encircling cingula except on inner face of
Oxyclaenus sp. has also been reported from locality M3. M3 somewhat reduced, metacone smaller than paracone,
CPlB. metaconule vestigial or absent. Lower molars with trigonid
and heel of subequal size, heel somewhat wider, trigonid a
Comments: Robison (1986) argued that Thangorodrim
little higher, protoconid slightly higher than metaconid, para-
Van Valen, 1978, is a junior synonym of Oxyclaenus.
conid well developed but lower, the talon[id] broadly basined
with hypoconid largest, entoconid and hypoconulid distinct,
Goniacodon Cope, 1888
somewhat blunted, but not compressed. P3-4 three-rooted,
Type species: Goniacodon levisanus (Cope, 1883c) (= Tri-
with internal cusps, rudimentary on P3, large on P4, and
isodon levisanus).
small cusps at external angles, in addition to the principal
Type specimen: AMNH 3217, right dentary with broken p4,
cusp. Pl one-rooted simple and small, P2 and p2-4 two-
ml-2.
rooted, p4 much larger than the preceding teeth, with high
Characteristics:
protoconid, sharp basal anterior cusp and bicuspid heel. The
[U]pper molars . . . much less quadrate than in Triisodon posterior premolars, above and below, pitched backward, as
(read Eoconodon here and following-Matthew, 1937, p. 86), in the Mesonychidae. Canines large, stout, not compressed,
 Archaic ungulates ("Condylarthra") 317

moderately recurved and subequal. Lateral incisors fairly Hapalodectidae is also known from Asia (Russell and
large, medial and second pair small, crowded. (Matthew, Zhai, 1987; Ting and Li, 1987).
1937, p. 81.)
Hapalodectes Matthew, 1909
Average length of ml: 8.0 (est.)-12.1 (est.) mm.
Type species: Hapalodectes leptognathus (Osborn and
Included species: E. coryphaeus (known from localities
Wortman, 1892) (= Dissacus leptognathus).
SB23A, SB23B); E. gaudrianus (localities SB23A,
Type specimen: AMNH 78, dentary with m2 and broken
SB23B); E. copanus (localities [SB23A], CPI 2A); E. nid-
crowns of p4 and ml.
hoggi (locality NP16C).
Characteristics:
Eoconodon sp. has also been reported from localities
CP12A, ?NP7B, ?NP7C, NP16C. Dentition I?/?O, C?/l, P?/4, M?/3; jaw slender, teeth highly
compressed, metaconids small, paraconids well developed
Triisodon Cope, 1881 b
on ml-3, talonids high, sharp, trenchant. (Matthew, 1915a,
Type species: Triisodon quivirensis Cope, 188lb. p. 102.)
Type specimen: AMNH 3352, left dentary with C, dp4-m2,
and unerupted p3-4 and m3. Average length of ml: 5.5 mm.
Characteristics: Included species: H. leptognathus ( = H. compressus) (known
from localities CP20A, CP20B, CP20C, CP20D, CP25H,
Teeth tritubercular, with robust, massive, well-rounded cusps.
CP27B, CP27D, CP34B and/or C); H. anthracinus (local-
Upper molars somewhat wider than long, rounded trigonal ity CP20A).
with conic blunt-pointed cusps.
Hapalodectes sp. has also been reported from localities
Paracones and metacones somewhat connate, the meta- SB22C, CP20A, CP27D, CP27E.
cones reduced on M2-3. No hypocones. Ml decidedly smaller
than M2, larger than M3. Premolars very robust, canines large Comments: "Hapalodectines differ from mesonychines and
and massive. Lower molars with paraconids much reduced, andrewsarchines in their distinct! y smaller size, in the
metaconids connate with protoconids. Heels large, equaling greater constriction of the lower teeth, in the presence of a
or exceeding trigonids, with rather high subequal hypoconid hypocone on the upper molars, and in having deep vascu-
and metaconid, hypoconulids vestigial or absent. Jaw very larized embrasure pits between the lingual parts of the up-
massive. (Matthew, 1937, p. 77; see Figure 20.4B.) per cheek teeth. They differ from all known mesonychines
in having m3 the longest tooth of the lower molars" (Sza-
Average length of ml: 13.3 (est.) mm.
lay, 1969a, pp. 2-3). Ting and Li (1987) give the same
Included, species: T. quivirensis (known from localities
diagnosis for Hapalodectes as they do for Hapalodectidae
SB23C, SB23E, SB23F, SB23G, SB23GG); T. antiquus
(see earlier), although they do not include Microclaenodon
(localities SB23C, SB23E, SB23F).
(see Figure 20.4A).
Triisodon sp. has also been reported from locality SB23F.
Hapalodectes is also known from Asia (Szalay, l 969a;
Russell and Zhai, 1987; Ting and Li, 1987).
HAPALODECTIDAE (SZALAY AND GOULD, 1966)
TING AND LI, 1987
HAPALODECTIDAE?
Characteristics:
Microclaenodon Scott, 1892
Distinctly small size; skull moderately long and narrow, Type species: Microclaenodon assurgens (Cope, 1884a) (=
with lower sagittal and occiptial crests; orbit closed; lachri- Triisodon assurgens).
mal with nearly no facial expansion; braincase large; pari- Type specimen: AMNH 3215, part of dentary.
etal long with lateral expansion; basicranial region relatively Characteristics:
long; three arteries (promontory, stapedial, and medial inter-
nal carotid) probably present; stapedial fossa large; dental The lower molars are sharply distinguished from those of
formula: ?I3/3, Cl/1, P4/4, M3/3; upper molars with acute Arctocyonidae by their narrowness, agreeing in proportions
cusps, large hypocone, and no shear; vascularized embrasure with Dissacus but with narrow basined heels. The trigonids
pits between the lingual parts of the upper cheek teeth; lower are high, paired metaconid and protoconid, and the proto-
teeth greatly compressed transversely, with trenchant talonid conid [paraconid?] low but projecting forward. The jaw is
heel and re-entrant groove on the anterior surface. (Ting and long and rather slender anteriorly, the premolars compressed
Li, 1987, pp. 178-79; see Figure 20.4A.) and spaced, the symphysis loose, quite unlike the short deep
jaw with massive symphysis of the triisodontines, but agree-
Zhou et al. (1995) restricted the family to Hapalodectes. ing with Dissacus, to which it appears to be allied rather than
If this is borne out, Hapalodectidae would be a redundant to the Triisodontinae with which it has hitherto been asso-
taxon. ciated.
 J. David Archibald 318

The upper molars so far known are trigonal with somewhat Average length of ml: 16.7 mm.
reduced inner half, approaching Dissacus in this respect, but Included species: A. saurognathus only (known from locali-
an external cingulum and a small hypocone crest present, as ties SB23G, SB23GG, SB23H).
in Arctocyonidae. The skull is distinguished by a long slender
muzzle, and the femur has a long patellar trochlea; otherwise PachyaenaCope,1814
the characters of vertebrae and limb bones conform to those Type species: Pachyaena ossifraga Cope, 1874.
of Arctocyonidae of equal size. (Matthew, 1937, p. 99.) Type specimen: USNM 1096, a left Ml.
Characteristics:
Average length of ml: 7.0 (est.) mm.
Included species: M. assurgens only (known from localities Molars 3/3; metaconids vestigial: paraconids large on p4-m3;
SB23E, SB23F, CP13B). pollex much reduced, probably vestigial. (Matthew, 1915a,
p. 87.)

MESONYCHIDAE COPE, 1875B Average length of ml: 19.8-33.9 mm.

Characteristics: Mandibular condyle below level of cheek teeth, Included species: P. ossifraga (known from localities SB24,
P3 protocone moved posteriorly (Prothero, Manning, and Fischer, CPI 9, CP20A, CP25A, CP25B, CP25C, CP63); P. gi-
1988; see Figures 20. lA, 20.2E, 20.5E). Thewissen (1994) and Zhou gantea (locality CP20A); P. gracilis (localities [CP20A],
et al. (1995) also present analyses of Mesonychidae. CP25D).
Mesonychidae is also reported from Europe (Russell, 1964) and Pachyaena sp. has also been reported from localities
Asia (Szalay and Gould, 1966; Russell and Zhai, 1987). CP20A,CP25H,CP31C,CP34Band/orC,CP64A,CP64B.

Dissacus Cope, 1881f Synoplotherium Cope, 1872b (synonym: Dromocyon

Type species: Dissacus navajovius (Cope, 188ld) (= Meso- Marsh, 1876)
nyx navajovius). Type species: Synoplotherium canius Cope, 1872b.
Type specimen: AMNH 3356, dentary fragments with right Type specimen: AMNH 5022, dentary, skull, skeletal frag-
p3-4, ml-3; left p4, ml-3. ments.
Characteristics: Characteristics: According to Matthew (1937), Synoploth-
erium "represents a cynoid specialization," but unlike
Metaconids distinct on ml-3; paraconids weaker especially
Mesonyx has M3 present. From Matthew's key (1937,
on m3; pollex (?) complete; humerus with entepicondylar
p. 91) this genus has "heels of the lower molars crested
foramen. (Matthew, 1915a, p. 85.)
... metaconids vestigial or absent on all molars ... M3
Average length of ml: 11.0-14.3 mm. reduced ... size medium, deep astragalar trochlea, pollex
Included species: D. navajovius (known from localities and hallux absent. Cursorial."
[CC50], SB23E, SB23GG, SB23H, CP13B, CP13G, Average length of ml: 20.0 (est.)-27.5 (est.) mm.
CP14A, [CP62B]); D. praenuntis (localities CP13G, Included species: S. canius (known from locality CP38A); S.
CP13H, ?CP17A, CP17B, ?CP18A, CP18B, CP19); D. vorax (localities CP34B and/or C, CP34D).
argenteus (locality CP13G); D. willwoodensis (locality Synoplotherium sp. has also been reported from locality
CP20A); D. serior (locality CP20D). CP38B.
Dissacus sp. has also been reported from localities
Comments: There is some taxonomic confusion regarding
CP13B, CP13E, CP13H, CP14D, CP16A, CP16B, CPl 7 A,
Synoplotherium. In his original description, Cope (1872b)
CPI 7B, CPI SA, CP20A, CP20B, CP24A, CP24B, CP26C,
applied the specific epithet canius, yet in later publica-
NP19C, NP19IIC.
tions he (e.g., 1884c) and others refer to it as Lanius. Cope
Comments: Dissacus is also reported from Europe (Russell, ( 1884c) also later argued that Synoplotherium was a ju-
1964). nior synonym of Mesonyx and thus he now called this
species Mesonyx Lanius. According to Matthew (1909),
Ankalagon Van Valen, 1978 Scott (1888) thought that Synoplotherium was distinct.
Type species: Ankalagon saurognathus (Wortman in
Matthew, 1897) (= Dissacus saurognathus). Harpagolestes Wortman, 1901
Type specimen: AMNH 2454, left dentary with canine, p 1-4, Type species: Harpagolestes macrocephalus Wortman, 1901.
ml-3. Type specimen: YPM 11901, incomplete skull and jaws, and
Characteristics: a right humerus.
Characteristics: Large to medium-sized mesonychids with a
Canine and molar cusps massive, ml without metaconid, m2- somewhat arcuate and deep lower jaw. M2 is considerably
3 with strong metaconid, protocone lobes P4-M3 longer than wider than long, with a paracone subequal to the protocone.
wide, pollex complete. (Van Valen, 1978, p. 65.) Average length of ml: 30.0-37+ mm.
 Archaic ungulates ("Condylarthra") 319

Included species: H. macrocephalus (known from locality According to Matthew's key for mesonychids (1937, p.
CP34B and/or C); H. brevipes (locality CP6A orCP6B); H. 91 ), this genus has "heels of lower molars crested ... meta-
uintensis (locality [SB44A]); H. leotensis (locality CP6B); conids vestigial or absent on all molars ... M3 absent ... size
H. immanis (locality CP38C). medium, astragalus with deep trochlea, pollex and hallux ab-
Harpagolestes sp. has also been reported from localities sent" (see Figures 20.lA, 20.5E).
CC4, (SB42B), SB43C, CP38A.
Average length of ml: 20+ (est.) mm.
Comments: Included species: M. obtusidens only (known from localities
Harpagolestes differs from Dissacus and Pachyaena in lack- SB22C, [CP31E], CP34B and/or C).
ing M3 and cingula on the upper teeth and in the absence Mesonyx sp. has also been reported from localities CP6B,
of a metaconid on the trigonids. Harpagolestes differs from CP31C.
Mesonyx in having a relatively deeper and wider skull.
Mesonyx retains a metacone on the M2, while in CETE?
Harpagolestes it is almost completely vestigial or reduced.
Homologous teeth are relatively wider in Harpagolestes than Platymastus Van Valen, 1978
in Mesonyx. Teeth of HarpagoLestes are often heavily worn, Type species: PLatymastus palantir Van Valen, 1978.
while those of Mes onyx rarely exhibit extreme wear. M esonyx Type specimen: AMNH 58034, right M2.
and Synoplotherium possessed highly cursorial feet. When Characteristics:
the size of most of the species belonging to Harpagolestes
Occlusal surfaces rather fiat; molar protocone central, broad,
is considered, it is unlikely that these larger species were apt
and massive, expanded lingually, paracone and metacone
runners. (Szalay and Gould, 1966; see Figure 20.2E.)
small; molar paraconid small, central, proximal. (Van Valen,
Harpagolestes is also known from Asia (Szalay and Gould,
1978, p. 56.)
1966; Russell and Zhai, 1987).
Average length of ml: unknown. Length of M2: 4.2 mm.
Mesonyx Cope, 1872a Included species: P. palantir only, known from locality
Type species: Mesonyx obtusidens Cope, 1872a. SB23A only.
Type specimen: AMNH 5021, teeth and numerous fragments
Comments: Van Valen (1978) allocated two species to Platy-
of the skeleton according to Matthew (1909).
mastus, P. palantir and P. mellon. Gingerich (1983) trans-
Characteristics:
ferred P. mellon to Aletodon mellon.
Dentition: 13/3; Cl/l; P4/4; M?/3 [M3 is apparently absent].
The space between the inferior canines is so narrow that it is Wyolestes Gingerich, 1981
probable that the incisors were wanting or reduced in number. Type species: ifyolestes apheles Gingerich, 1981.
The external superior incisor conical and larger than the oth- Type specimen: UM 67517, left maxilla with roots for P4 and
ers. Superior canines vertical; inferior canines subhorizontal. Ml, and crowns ofM2-3 intact.
First inferior premolar one-rooted. Premolars from the third Characteristics:
to the molars, consisting of a posterior median blade, an an-
teromedian conic cusp, and an anterior basal tubercle. The Differs from Mongoloryctes in being smaller and having rel-
true molars diminish in size posteriorly, but their composi- atively narrower upper molars, with slightly better developed
tion is identical with that of the first, so far as determinable. conules and stylar cusps, and a weak ectocingulum. Differs
(Cope, 1884c, p. 348.) from Didymoconus, Ardynictis, Hunanictis, and Archaeo-
The inferior canine is stout, especially in the root, which ryctes as outlined in the subfamilial diagnosis. (Gingerich,
is a fiat oval in section. The crown is but slightly curved, 1981, p. 529.)
slightly compressed, and without edge or groove. The pre-
Average length of ml: 6.8 mm.
molars graduate into the molars, so that the line of distinction
Included species: W apheLes (known from locality CP20A);
is not easily drawn. The first premolar has a single root; the
W iglesius (locality CC50).
crown is slightly conic, with a small tubercle at the base be-
hind. This tubercle increases in size on the premolars 2 and 3, Comments: ifyolestes is included here, although I seriously
and becomes on the true molars a longitudinal cutting edge doubt its affinities with mesonychids. According to
extending along the axis "of the crown, not much elevated Gingerich (1981), this taxon is the only North Ameri-
above a wide base. It occupies half the length of the crown in can representative of the otherwise Asian family Didymo-
the larger molars, and is preceded by an elevated conic cusp. conidae. Gingerich (1981) included this family along with
In front of the base of this, a small conic tubercle projects the Mesonychidae within the superfamily Mesonychoidea
forwards, which appeared as a rudiment on the third premo- Osborn, 1910. As noted by Gingerich, the Didymoconidae
lar. The number of mandibular teeth would appear to be, Pm. have been suggested to be closely related to as many as
4, M. 3. (Cope, 1884c, p. 356.) eight families within six or seven orders. Meng (1990)
 J. David Archibald 320

rejected the inclusion of ifyolestes in Didymoconidae and Type specimen: Type of Mioclaenus fioverianus was not given
did not follow Gingerich's (1981) placement of this family by Matthew, 1937, but Cope, 1888, describes it as "parts
in the Mesonychoidea. of two mandibular rami which display four of the mo-
N ovaceket al. ( 1991) described a new species of ifyoles- lar teeth, with which were found the following bones of
tes, W iglesius, from the early Eocene of Baja Califor- the skeleton: several vertebrae, parts of both humeri, dis-
nia. These authors, although persuaded by Gingerich's ar- tal end of radius, greater part of ilium." Type specimen of
gument of possible didymoconid affinities, felt that rela- Phenacodus puercensis: AMNH 3832, parts of maxillae
tionships with other early Tertiary mammals could not be and dentaries and fragments of skull and skeleton.
rejected (e.g., hyaenodontids). They treated ifyolestes as Characteristics:
Eutheria incertae sedis.
A third view (McKenna, written comm., 1989) suggests Tetraclaenodon differs from Phenacodus in lacking a meso-
that ifyolestes is not a didymoconid, but rather a cimo- style on upper molars. Tetraclaenodon is more bunodont than
lestid derivative. After viewing illustrations in Novacek Ectocion, its Pl is single rooted, and the paracristid ends high
et al. (1991 ), I am inclined toward this assessment. This on the metaconid. Tetraclaenodon differs from Copecion in
does not preclude possible hyaenodontid affinities because having short premolars and lacking the mesostyle. (Thewis-
creodonts appear to be cimolestid derivatives (Lillegraven, sen, 1990, p. 21; see Figure 20.5D.)
1969). Comparisons with such taxa as Cimolestes magnus
Average length of ml: 7.7-7.8 mm (T. puercensis only).
might prove worthwhile.
Included species: T. puercensis(knownfromlocalities [NB 1],
SB23C, SB23DD, SB23E, SB23F, SB23G, SB23GG,
TAXEOPODA COPE, l882F SB23H, CP13B, CP13C, CPl4A, CP14B, NPIB, NP19C,
[NP47 A]); T. septentrionalis (known from locality NPl 9C
In naming Taxeopoda, Cope ( 1882f) included Condylarthra and Pro-
and others).
boscidea. He also thought it probable that toxodonts and hyracoids
belong in Taxeopoda. Later the same year (1882b) he more fully
explained his views on ungulate classification and phylogeny. In PHENACODONTIDAE COPE, 1881E
this later paper, he separated proboscideans from taxeopods as a
Characteristics and comments:
separate order. In the diagram in the same paper he placed tax-
eopods as ancestral to both known extinct and extant ungulates as Bunodont to bunolophodont condylarths. [Lower fourth pre-
well as speculating on their ancestral position for several hypothet- molar] with talonid basin and subequal protoconid and meta-
ical ungulate groups. Cope's Taxeopoda is essentially what is here conid. Paraconid of lower molars absent or small. P3 with
regarded as Ungulata, with the exception of the absence of some small protocone, P4 with paracone, metacone, and protocone.
archaic ungulates and cetaceans and the inclusion of pantodonts Ml-2 with hypocone and often mesostyle. Dental formula;
(along with dinoceratans in Amblypoda). I use Taxeopoda with I3/3, Cl/I, P4/4, M3/3. Humerus with weak deltopectoral
only a few modifications from that of Cope. The phenacodontids crest and supracondylar foramen. Ulna strong with anterior
("Phenacodonta"), which were central to Cope's concept of Condy- process distally. Five digits in hand and foot. Femur with third
larthra, as well as all fossil and living paenungulates are included. trochanter. Fibula complete. Astragalus with convex head.
The only major taxa that Cope included as being descended from Arctocyonidae differ from Phenacodontidae in having less
Taxeopoda, but are excluded here, are Artiodactyla, Periptychidae, inflated molar cusps and unmolarized premolars. Hyopsodon-
and various other archaic ungulates now assigned to the Arctocy- tidae differ from Phenacodontidae in having higher trigonids,
onidae, Hyopsodontidae, and Mioclaenidae. Thewissen and Damn- no metacone on P4, and smaller hypocones. Meniscotheri-
ing ( 1992) proposed to expand the term Paenungulata to include not idae differ from Phenacodontidae in having more lophodont,
only most of the usual members of the Paenungulata (Proboscidea, hypsodont molars. Didolodontids differ from phenacodon-
Desmostylia, Sirenia, Hyracoidea, and Perissodactyla), but also the tids in absence of talonid basin on p4, and the more molar-
Phenacodontidae. (Embrithopoda was not included by these authors ized P3. Periptychidae differ from Phenacodontidae in labial
in Paenungulata.) Cope's Taxepoda exists for essentially the same position of the protocone and inflated posterior premolars
taxon, and thus I prefer to retain Paenungulata for the above taxa (Archibald, Schoch, and Rigby, 1983). Mesonychidae differ
that are usually placed within it (see Figure 15.1). from phenacodontids in the reduction of the trigon, trigo-
nid, and talonid basins, and the inflation of individual cusps.
"PHENACODONTA" MCKENNA, 1975 (Thewissen, 1990, p. 19.)

Tetraclaenodon Scott, 1892 (synonyms: Protogonia Cope, Phenacodontids were central in Cope's original conception
188ld; Euprotogonia Cope, 1893, footnote in Earle, 1893; and naming of the Condylarthra in 1881 (Cope, 188ld). Mem-
see discussion in Thewissen, 1990) bership in Phenacodontidae has remained relatively stable since
Type species: Tetraclaenodon puercensis (Cope, 1888) (= it was recognized by Cope in 1881 (Cope, 1881e). The synop-
Mioclaenus fioverianus, = Phenacodus puercensis Cope, sis of phenacodontids used here is after Thewissen ( 1990) with
1881d, fide Matthew, 1937). two exceptions. Meniscotherium is included and Tetraclaenodon is
 Archaic ungulates ("Condylarthra") 321

excluded from Phenacodontidae. Williamson and Lucas ( 1992) re- CP64C, [NP3F], NP7D, NP47B); P. matthewi (= P. gi-
viewed the relationships of Meniscotherium and also included it dleyi) (localities CP63, CP64A, SB20A, SB39A, SB39B,
in Phenacodontidae. They, however, also included Tetraclaenodon. SB39C, [SB39E]); P. bisonensis (localities CP13B,
Meniscotherium possesses most, if not all, characteristics of other CP14D, CP16A, CP16B); P. grangeri (localities SB20A,
phenacodontids. Its differences from other phenacodontids, notably SB39A, SB39B, SB39C, [SB39D], SB39E, CP13E,
increased lophodonty and hypsodonty, are autapomorphies that do CP22A, CP24A, CP24B, CP26A, CP24B); P. vortmani
not preclude it from membership in Phenacodontidae. The begin- (= P. almiensis) (localities SB22A, SB22B, SB24, CP13G,
nings of lophodonty seen in Ectocion suggest Meniscotherium and CP13H, CP14E, CP17 A, CPI 7B, CP18A, CP18B, CP20A,
Ectocion are sister taxa. Prothero, Manning, and Fischer (1988) CP20B, CP20C, CP20D, CP22C, CP24B, CP25B,
did not unite Tetraclaenodon with the other phenacodontids. They [CP25D], CP25H, CP26B, CP26C, CP27 A, CP27B,
treated it as the sister taxon to phenacodontids plus all remaining CP27D, CP27E, CP28A, [CP62B], CP62C, CP64B,
paenungulates. This usage is followed here. CC50); P. intermedius (localities SB21, [CP19], CP20A,
Taxa previously referred to Phenacodontidae that are now ex- CP20C, CP20D, CP63); P. trilobatus (localities SB24,
cluded or are synonymized with others include "Protogonodon" CP20, CP25, CP27, CP63, CP64; P. magnus (locality
(= Loxolophus), removed from the family by Granger in 1915; CPl3F).
Prosthecion Patterson and West, 1973, synonymized with Ectocion Phenacodus sp. has also been reported from localities
by Thewissen, 1990; several European taxa, especially Pleuraspi- SB39D, CP13G, CP13H, CPl5B, CP15C, CP16B, CPI 6C,
dotherium, have been linked to or even placed with the phenacodon- CP20F, CP22A, CP25A, CP28B, CP28C, NP3C, NP3D,
tids (Prothero, Manning, and Fischer, 1988), but Thewissen ( 1990) NP47C.
rejected this assignation (see Figures 20.IB, 20.2I, J, 20.4C,
20.5D, F). Comments: Phenacodus is also known from Europe (Thewis-
Phenacodontidae is also known from Europe (Thewissen, 1990) sen, 1990).
and questionably from South America (Muizon and Marshall, 1987b).
Copecion Gingerich, 1989
Phenacodus Cope, 1873
Type species: Copecion davisi Gingerich, 1989.
Type species: Phenacodus primaevus Cope, 1873.
Type specimen: UM 86002, right dentary with one root each
Type specimen: AMNH 4408, "somewhat worn and weath-
for canine and pl, two roots for p2, and teeth P3/p3-4,
ered last lower molar" (Granger, 1915, p. 335). Thewissen
ml-3.
(1990) restricted the name Phenacodus primaevus to the
Characteristics:
holotype. He noted that the holotype is intermediate in
size between P. intermedius and P. trilobatus. Following Differs from all other phenacodontids in the elongate p3 and
his argument, all the localities listed here for P. primaevus p4, with long trigonid basin, and elongate P3 and P4. Dif-
(except possibly CP25K, which is the type locality of P. fers from Ectocion by being more bunodont. (Lower third
primaevus) refer to either P. intermedius, P. trilobatus, or premolar) without metaconid and weak hypoconid, unlike
both. Ectocion. Paracristid usually ends high on the lingual side of
Characteristics: the molars and paraconid absent, unlike Ectocion. Parastyle
Differs from Ectocion in having p3 with large metaconid, and mesostyle of upper molars weak, unlike Ectocion. M3
hypoconid weak or absent. [Lower fourth premolar] short hypocone weak or absent, unlike Phenacodus. Mesostyle
unlike Copecion. Differs from Ectocion by bunodont mo- present, unlike Tetraclaenodon. (Thewissen, 1990, p. 66.)
lars, and paracristid ending usually high on the lingual side
Average length of ml: 6.7-6.8 mm.
of the molars. Pl single rooted, unlike Ectocion and mo-
Included species: C. davisi (known from localities CP19,
lar mesostyle smaller. Differs from Tetraclaenodon in large
CP25B); C brachyptemus (localities SB24, CP20B,
metacone on P4. Parastyle and mesostyle of upper molars
CP20C, CP25B, CP25C, CP27 A, CP27B, CP63, CP64A,
usually weaker than Ectocion, but stronger than Tetraclaen-
CP64B).
odon. Hypocone strong in Ml-2 and usually present in M3
Copecion sp. has also been reported from locality
unlike Ectocion. (Thewissen, 1990, p. 45; see Figures 20.2I,
CP25D.
20.5D.)

Average length of ml: 8.3-14.2 mm.

MENISCOTHERIINAE (COPE, 1882D)
Included species: P. primaevus (known from localities SB24,
VAN VALEN, 1978
SB40, CP13E, CP13G, CP13H, CP14C, CP14E, CPl5B,
CPI 5C, CPI 7A, CPI 7B, CP 18A, CPI 8B, CP20A, CP20B, Ectocion Cope, 1882e (synonyms: Oligotomus [in part],
CP20C, CP21B, CP22C, CP24B, CP25A, [CP25B], Cope, 1882a; Gidleyina Simpson, 1935; Prosthecion
[CP25C], CP25D, [CP25E], CP251, CP26B, CP26C, Patterson and West, 1973)
CP27A,CP27B,CP27D,CP27E,CP 28B,CP29A,CP32B, Type species: Ectocion osbornianus (Cope, 1882a) (= Olig-
CP34D, CP62A, CP62B, CP62C, CP63, CP64A, CP64B, otomus osbornianus).
 J. David Archibald 322

Type specimen: AMNH 4409, maxillary and dentary frag- metaconule), and a paraconule that parallels the paracone
ments. and is separated from the protocone by deep valley. The
Characteristics: lower cheek teeth of differ from those of Pleuraspidotherium
and Orthaspidotherium in possession of a cristid obIi qua that
Most lophodont phenacodontid. Differs from all other phena- meets the metaconid near the lingual edge of the tooth and
codontids in the large parastyle and mesostyle of the upper a more prominent metastylid. (Williamson and Lucas, 1992,
molars. Pl double rooted or with one elongate root and M3 pp. 35-36; see Figures 20. lB, 20.2J, 20.5F.)
without hypocone, unlike Tetraclaenodon and Phenacodus.
Differs from Copecion in the short p4, P3 and P4. (Thewissen, Average length of ml: 5.2-9.0 (est.) mm.
Included species: M. chamense (= M. robustus) (localities
1990, p. 25; see Figure 20.4C.)
?CP25D, ?CP25E, CP25G, CP25H); M. terraerubrae (lo-
Length of ml: 6.0-8.9 mm. calities SB24, CP25A, [CP25C], CP25H, CP27D, CP34B
Included species: E. osbornianus (known from localities and/or C, CP64A, CP64B); M. tapiacitum (= M. priscum)
CP13G, CP13H, CP14E, [CP15C], CPI 7 A, CPI 7B, (localities CP18B, CP19, CP25A, [CP25C], [CC50]); M.
CP18A, CP18B, CP19, CP24B, CP25B, CP26C, CP26D, chamense (localities SB24, [CP23A], CP27B, CP64B).
CP62B, CP63, CP64B, NP3F, NP4, NP7D); E. collinus Meniscotherium sp. has been reported from localities
(= E. montanensis from below locality NP19IIA and E. CP27C, CCL
silberlingi) (localities SB39B, SB39C, [SB39E], NPlC,
Comments: Species synonymy follows Williamson and
NPl 9IIA, [NP20B]); E. wyomingensis (localities CP13E,
Lucas ( 1992), although locality data is given for better
CP14B, CP14C, CP14D, CP16A, CP16B, CP16C,
known junior synonyms listed by these authors.
[CP21B], CP22A, CP24A, NP3B, NP3C, NP3D, NP3E,
NP20D, NP47B); E. parvus (localities CP18B, CP19,
CP26C, CP62B, ?CP64B); E. superstes (localities
BIOLOGY AND EVOLUTIONARY PATTERNS
[CP20E], CP27D, CP27E); E. major (locality CP62B);
E. cedrus (localities CP13F, CP26A, NP3D, NP20C); E.
Arctocyonidae have been viewed as the central stock from which
mediotuber (localities CP13E, CP13G, CP14C, CP15B,
other archaic ungulates arose (Van Valen, 1978). If one follows
CP24A or CP24B, NP4); E. ignotum (locality CC50).
the interpretation of Prothero, Manning, and Fischer (1988), Arc-
Ectocion sp. has also been reported from localities
tocyonidae are not a monophyletic taxon, but rather comprise two
GC2111, CP15A, CP15B, CP26A, (CP28C).
distinct lineages. One lineage forms the nucleus of a restricted Arc-
Comments: If Meniscotherium is included in Phenacodonti- tocyonidae that is the sistertaxon to all ungulates except artiodactyls,
dae (contra Thewissen, 1990) as I do here, then it is the and the other lineage is allied with the hapalodectids, mesonychids,
most lophodont phenacodontid rather than Ectocion. and cetaceans in Cete. The second group includes Triisodon and
other taxa that are traditionally referred to as the triisodontines, al-
though strictly they are a paraphyletic group.
Meniscotherium Cope, 1874
Members of Arctocyonidae (sensu stricto) do tend to maintain
Type species: Meniscotherium chamense Cope, 1874.
more primitive character states, especially in the better known
Type specimen: USNM 1093, right maxilla with Ml-3.
aspects of the dentition such as a more complete eutherian dental for-
Characteristics:
mula, relatively little molarization of posterior premolars, a hypo-
Meniscotherium is a small- to medium-sized phenacodontid cone that may still be small, and the stylar region on the upper
(ml = 5.2-9.1 mm) with molarized posterior premolars and molars that may retain more of the parastylar and metastylar lobes
lophodont-selenodont cheek teeth. Meniscotherium is distin- (Figures 20.2F, G). These plus other character states are retained
guished from Ectocion, which it most closely resembles, by from Protungulatum** donnae* (Figures 20.2A, 20.3A). Postcra-
its more lophodont dentition. Meniscotherium possesses a nially there are few, if any, modifications of the limbs toward the
metaloph formed from the metaconule and hypocone, and more generally conceived idea of an ungulate, such as distal limb
the paraconule is separated from the protocone by a val- elongation, reduction of side digits, fusion of limb bones, and the
ley, whereas in Ectocion these cusps are connected by a development of hoofed unguals. None of these states occurs in arc-
small ridge. Unlike Ectocion, Meniscotherium lacks a dis- tocyonids (sensu stricto ).
tinct hypoconid, has a cristid obliqua that meets the metacone The possession of hoofed unguals would seem prerequisite for
[sic] at the lingual edge of the tooth. Also, the p4 paracristid an ungulate; however, as used here the term "Ungulata" refers to
of Meniscotherium extends to the lingual edge of the tooth, a clade of mammals rather than to a distinctive aspect of morphol-
whereas in Ectocion it proceeds anteriorly. Meniscotherium ogy such as hooves. Hence it is not surprising to find that a number
differs from Pleuraspidotherium and Orthaspidotherium in of archaic ungulates, including some arctocyonids (sensu stricto)
its more lophodont cheek teeth, lack of a diastema between P2 such as Chriacus, retain claws. Although one cannot generalize
and P3, possession of a true hypocone (the pseudohypocone from the early Eocene Chriacus to other arctocyonids (sensu stricto)
of Pleuraspidotherium and Orthaspidotherium is a displaced (see also comments under this family), this creature is sufficiently
 Archaic ungulates ("Condylarthra") 323

preserved to provide some indication of habitat for one member of developed both on molars and premolars (Figure 20.3C). This den-
this clade. The skeleton of this animal (Rose, 1987) indicates a scan- tal morphology suggests that tough, fibrous plant material was first
sorial mammal that was probably equally adept in the trees or on the shredded by the bulbous, sometimes fluted premolars and then was
ground, somewhat like extant procyonids and viverrids. According pulverized by the molars. This scenario is probably most applica-
to Rose ( 1987), Chriacus had powerful limb musculature, very mo- ble to the larger sheep-sized periptychids. Some of the smaller taxa
bile joints, and flexible plantigrade pentadactyl feet with claws and show somewhat similar dental specializations, but by virtue of their
a slightly divergent hallux (Figures 20.lC, 20.5C). In addition, the size these smaller squirrel-sized animals probably ate higher-energy
tail was long and robust, showing some indication of prehensibility. diets, which may have included fruits, nuts, or even insects.
Larger arctocyonids (sensu stricto) such as Arctocyon (wolf- Postcranially, periptychids are best known from the squirrel-
sized) appear also to have retained many of the primitive eutherian sized Gillisonchus (Rigby, 1981) and the sheep-sized Ectoconus
postcranial features found in the much smaller Chriacus (Matthew, 1937). Matthew (1937) suggested that in size and gen-
(coati-sized). According to Matthew ( 1937), Arctocyon (including eral body proportions, although not in head, tail, or pedal struc-
Claenodon) was terrestrial. It had an extremely long and powerful ture, Ectoconus resembled the aardvark, Orycteropus, among extant
tail, a relatively reduced and inflexible lumbar region, limbs that mammals. In most parts of the skeleton, Ectoconus and Pe riptychus
were short and stout, a plantigrade manus and pes, and an oppos- are similar to larger arctocyonids (sensu stricto) such as Arctocyon
able hallux. The dentitions of all arctocyonids (sensu stricto) were in being quite massively built with no reduction of the distal ulna
somewhat similar with brachydont and bunodont cheek teeth and or the fibula, no elongation of limbs, and in having a pentadacty-
moderately to well-developed canines. Matthew (1937) suggested lous manus and pes. At least Ectoconus among periptychids differs
a bearlike diet of fruits, nuts, seeds, insects, and small animal prey. from the plantigrade arctocyonids such as Arctocyon and Chriacus
An omnivorous diet is also suggested for Chriacus. This taxon also in possessing a subplantigrade manus and pes that was short and
had a tooth comb composed of lower incisors that is believed to wide with considerable padding below, wide-spreading digits, and
have been utilized in grooming (Rose, 1987). Arctocyonidae (sensu small tapiroid hooves (Matthew, 1937).
stricto) reached their acme of diversity in the Torrejonian and earlier Periptychidae were the taxonomically most diverse archaic ungu-
Tiffanian (mid- to late Paleocene), finally disappearing from North lates in the later Puercan (early Paleocene), but they rapidly declined
America at the end of the Wasatchian (early Eocene) (Figure 20. 7 A). at the end of the Puercan. Only the paraphyletic "Anisonchus" was
Hyopsodontidae (without Mioclaenidae) were, by contrast to taxonomically diverse during the Torrejonian (mid-Paleocene). The
Arctocyonidae, generally smaller animals. If the ubiquitous early last of the family, represented by its namesake Periptychus, disap-
Eocene taxon Hyopsodus (rat-sized) is any indication, these animals peared during the mid-Tiffanian (late Paleocene).
were quite gracile and possessed shortened limbs (Figure 20.5B) The "triisodontines," hapalodectids, and mesonychids constitute
(Gazin, 1968). Their dentition, although not markedly different from the earlier lineages of Cete that eventually culminated in the Cetacea.
that of arctocyonids (sensu stricto ), shows the beginnings of molar- Unlike all other clades of ungulates, this clade tended away from
ization of P4/p4, amplification of the para- and metaconule giving herbivory and toward carnivory. In the palaeoryctid ancestry of
the appearance of a sexitubercular upper molar, and a hint of crest de- Carnivora and Creodonta, and possibly Ungulata, both a prevallid-
velopment of the molar cusps (Figures 20.3D, E). Hyopsodontidae postvallum (between the front of the lower and back of the upper
reached their taxonomic zenith in the mid-Tiffanian (late Paleocene), molar) and a postvallid-prevallum (between back of lower and front
but the mostly Eocene Hyopsodus extended into the Duchesnean. of upper) shear was present and sometimes amplified. In Creodonta
Hyopsodus and the mesonychid Harpagolestes are the last two taxa and Carnivora the prevallid-postvallum shear was emphasized to
of archaic ungulates (in the older sense of "Condylarthra") in the form the carnassials of these creatures. In the early Cete such shear
North American fossil record (Figure 20.7A,B). was absent because it had disappeared with development of the
The mioclaenids, sometimes placed within Hyopsodontidae, are brachydont, bunodont crushing dentition of early archaic ungulates.
in fact quite different in their dental architecture. Their tendency to- The postvallid-prevallum shear, however, was retained but not ac-
ward simplification of the occlusal surface and the slight to consid- centuated in the early ungulates.
erable inflation of the premolars (Figure 20.4D) suggests that these Szalay (1969b) has documented the development of the different
animals ate tougher, more fibrous plants. Less has been written on manner of shear seen in the molars of Cete. In the lower molars of
their postcranial anatomy than on that of other archaic ungulates, but Cete a postvallid notch analogous to the prevallid notch between
they seem in general to have been slightly larger (up to hare-sized) the protoconid and paraconid of Carnivora developed between the
than hyopsodontids, but similar to the smaller arctocyonids. The protoconid and the cristid obliqua. Concomitantly, the metaconid
mioclaenids are most diverse in the Torrejonian (mid-Paleocene) and was reduced and/or lost, the talonid was gradually reduced to an
disappeared from North America by mid-Tiffanian (late Paleocene) anteroposteriorly oriented, sectorial cristid obliqua, and the molar
or may have extended into the Wasatchian (early Eocene) if the became transversely narrow. In the upper molars, the conules were
apheliscines are referable to Mioclaenidae (Figure 20.7B). reduced, the hypocone was generally lost, the para- and metastylar
Periptychidae show even more premolar inflation than do the regions became aligned with the paracone and metacone, and a V-
probably related mioclaenids. In several taxa, especially Peripty- shaped trough developed between the protocone and the labial cusp.
chus, the quite bulbous premolars are distinctly grooved. Some The "carnassial" developed on the lowers occluded in this trough.
of the cusps of the molars tend to be accentuated, and crests are These dental changes, which are opposite those for ungulates as
 J. David Archibald 324

PALEOCENE EOCENE OLIGOCENE

m r- m ;;:: r- m r-

~~ ~"2"2"2~
.,
I";::1;::1;::1;::1::1;::1~~&!
; r

i1 ~I 0 ... N w :....
'Cl'(}
N w " N w ... "' 0:...Noi.i
ri
~
rn
~ "~ ~ I
!"
!!' "'
!!'
r
~
m
~
r

~
l? !"

I~ ~ ~
!:/ ~ '~" "'~
"D > •
:i::::1.
(; ~
&m::f:~!lgta=!:!l~Sn£x:p:::::
~ WC;.., bi WO NOG;cn\.n 0
!...OU. 00. c e ~ :b i:
" e c 11:
"'
E ~ c; :..t:
"' : !:!
"
~

~ !t
· ~ Protungulatum I I I I I I I
:i I I I I Ii

~ I ~rcinodonl I 111 I I I
f Oxl}omodon 111 I I I
::> Desmatoclaenus I I I
0
:"
II 111 I 11 I I I Princetonial I t
0
it I I
... I Prothryptacodon I I I
·"'0 I Chriacus I I I
?'
II
I Deuterogonodon I I
~ ...
>
Q,
a
"'
iii"
·->--··->
ILoxolophus I 111 I
IMimotricentes 11 I
I
I
I
I
-
(')
0
(')
'<
0
·"'c :I
c
ii I I I I I I Lambertocyon I I I a:
Ill
c ·->--I CD
c
":r I I I I I I Mentoc/aenodon I I I
m
"'z III I II 1A.n,acodo~ I I I
Arctocyon
·"'c: 111 I I I I I I I I I II I I I
? Colpoc/aenus
II
c:
I I 1-1+-1+-I11-111--1,. ,1 Ii II Ii I I
Thryptacodon
~ I II I 1-+-+-..,...,.:;~...,...,.---r---1--1---1
·"'m
lp1~ty,ri~~t~~I 1111 I I
1
:"
II

~ rs~i~c o'n~d~~ 111 I I I I

1

t. lTi I I I I I II 111 I Wyolestes

·"' ~~g,n~~~kl 11 111 I
~
1l~~e/~~~n,111111
11

~g: I l~?.c/aenus
I I I i 1 I 111
I
g. Goniacodon
iii" llr-r-r-ri 1 11lllll
·"' Eoconodon
F. 11 fiu~ddo~ I I I 111 I 0
:::c m
~ 111n-n11111111 Ill -I
m
~
"tJ
11 I I ~icroclaenodon I Ill
a: 0
a.
?
;:I
11
11 I I I I I I I 11 111 I Hapalodectes
I I I I I IDissacus I I I I I I
Ja a.
Ill

~~-
CD
1111I11II11 1111 I
Ankalagon 3:
~
!>
1111 I 11111111 n IPachyaena I CD
Ill
0
:I
~ 111111111111111 ISynoplotherium '<
(')
g ::r
~. 111111111111111 IHarpagolestes a:
0 Ill
"'iii" 1111 I 11 I I 11 1111 1Mesonyx CD

::>
111111111111111
A
Figure 20.7. Temporal ranges of genera of North American archaic ungulates. A More carnivorous groups (Arctocyonidae and Cete).
B. More herbivorous groups (Hyopsodontidae, Bulbulodentata, Phenacodonta).
 Archaic ungulates ("Condylarthra") 325

PALEOCENE EOCENE OLIGOCENE

m r- m r- m r-
!" !; ;

I !!' "'
!!' " 'c:"
?
c:
?

~ g i:
" e !:
"
:::c
'<
0
"tJ
Ill
0
a.

-
0
:I
a:
Ill
CD

I 1£!!oeroc/aenus I I
I 1Eilipsodon 11 111 I
I 1Lita!etes ~I I I I I
Ifc 11~~~~81 I 1111
c
1IPromioclaenus11
I"' I 1Protoselene I 111 I
z
·"'c: I '.!!!natzinia I I t 11 I .I I DJ
? Aphe/1scus c:
11 llll llll Ill lillli I I r-
~ Phenacodaptes DJ
c:
~ lllllllllITTm I r-
0
m
~ l':_Bcatepos I 11 111 cm
z
~ tyaioranal I 11 111 -I
>
"'~- ITinuviel I I II 111 ~
·f
~!" M~atutal I 11 111
"Cl
~ If!!..mithlaeus I 111
-c·...
CD

~ l'::!oconus1 11 111 ~
(')

~ ICarsioptychus I I I ::r
...::i ~1111111 a:
Ill
o Periptychus CD
~
11
I 11IIIIII11111
0 Gillisonchus I I I I
... 1 r-Mllllll
2S: "Anisonchus"
g j I I I I I I I I 11111
~ Hap/oconus
·"' 11111111111111
;:I Conacodon
11
hTI I 11I111111
'iii" I~~cf?f~·I I 111 "Cl "Cl
."'"f} I I I 1-1"'""1-1"-
1 1 1 1Tetrac/aenodon ::r :::c
I 11 10 I lj I I CD m
I I I I I Phenacodus z

u
11
>
l~ Copecion 0
0
1111 I I l~ct~~;~~""++'.,.....,___--1 c
"'~· 11 I I I I I 11 I I I Merliscotheri'um 0
z
-I
I I I I I I I 111 I I I I >
B
Figure 20.7. (Cont.)
 J. David Archibald 326

a whole, have aptly been referred to as the premolarization of the ACKNOWLEDGMENTS

molars.
In the earliest "triisodonts," only the suggestion of these later Thanks to Christine Janis for her editorial and organization skills in
dental trends can be seen. Szalay and Gould (1966) suggested that helping to complete this chapter, Malcolm C. McKenna and Don-
in some Cete such as Synoplotherium and Harpagolestes, excessive ald Prothero for reading and providing substantive comments, and
wear on molars and incisors, massiveness of some of the cusps, Philip Unitt for stylistic comments. Preparation of this manuscript
and large canines all point to carrion feeding. Dissacus, Mesonyx, was partially supported by NSF Grant EAR-8407507.
and Hapalodectes lacked these specializations, but probably were
carnivorous, as the slender-toothed Hapalodectes certainly was. In
size, these animals ranged from the raccoon-sized Microclaenodon
REFERENCES
and coyote-sized Dissacus to the wolf-sized Mesonyx and bear-sized
Harpagolestes and Pachyaena.
Ameghino, F. (1901). Notices preliminaires sur des ongules nouveaux des
From what is known of the postcrania, especially as based on terrains cretaces de Patagone. Boletfn da la Academia nacional Cien-
Mesonyx, the skeleton of cetans developed more cursorial adapta- cias Cordoba (RepublicaArgentina), 16, 349-426.
tions than did that of arctocyonids or periptychids (Figures 20. lA, Archibald, J. D. (1982). A study of Mammalia and geology across the
20.5E). Cetan skeletons tended to be somewhat more gracile, and Cretaceous-Tertiary boundary in Garfield County, Montana. Univer-
sity of California Publications in Geological Sciences, 122, 1-286.
in some, such as Mesonyx, the manus and pes were digitigrade
( 1983 ). Structure of the K-T mammal radiation in North America: specu-
with closely aligned metapodials rather than the more sprawling lations on turnover rates and trophic structure. Acta Palaeontologica
digits of Ectoconus. Some of the earlier taxa such as Dissacus re- Polonica, 28, 7-17.
tained the more primitive plantigrade posture, but in both this taxon (1993). The importance of phylogenetic analysis for the assessment of
and Mesonyx hoofed unguals were retained, rather than reverting to species turnover: a case history of Paleocene mammals in North
America. Paleobiology, 19, 1-27.
claws as in some arctocyonids.
(1994). Metataxon concepts and assessing possible ancestry using phy-
The earliest cetans were the triisodonts and their immediate an- logenetic systematics. Systematic Biology, 43, 27-40.
cestors, which were moderately diverse in the Puercan and Torrejo- (1996a). Dinosaur Extinction and the End of an Era: What the Fossils
nian (early and mid-Paleocene). In North America during all of the Say. New York: Columbia University Press.
Tiffanian and Clarkforkian (late Paleocene) the only known cetan ( l 996b). Fossil evidence for a Late Cretaceous origin of "hoofed" mam-
mals. Science, 272, 1150-3.
is Dissacus. Then in the Wasatchian through Duchesneau (Eocene),
Archibald, J. D., Clemens, W. A., Gingerich, P. D., Krause, D. W., Lind-
another moderate radiation (possibly via invasion from Asia) oc- say, E. H., & Rose, K. D. (1987). First North American Land
curred before cetans disappeared from the fossil record in North Mammal Ages of the Cenozoic Era. In Cenozoic Mammals of
America. North America, ed. M. 0. Woodburne, pp. 24-76. Los Angeles:
The phenacodontids, although quite numerous at some sites in University of California Press.
Archibald, J. D., Rigby, J. K., Jr., & Robison, S. F. (1983). Systematic revision
North America, are taxonomically the least diverse of the archaic
of Oxyacodon (Condylarthra, Periptychidae) and a description of 0.
ungulates. Both dentally and postcranially, however, they show sev- ferronensis n. sp. Journal of Paleontology, 57, 53-72.
eral trends away from the arctocyonids. Starting with Phenacodus, Archibald, J. D., Schoch, R. M., & Rigby, J. K., Jr. (1983). A new subfamily,
whose teeth are quite low crowned with distinctive cusps, including Conacodontinae, and new species, Conacodon kohlbergeri, of the
the conules in the upper molars, there was a trend toward lophs and Periptychidae (Condylarthra, Mammalia). Postilla, 191, 1-24.
Blainville, de, H. M. D. (1841). Osteographie et description iconographique
crests on both upper and lower molars. Phenacodontids also tended
des Mammiferes recents et fossiles (Carnivores), vols. I et II.
toward molarization of the posterior premolars more than did any Paris.
other North American archaic ungulates. The molarization of the Bown, T. M. (1980). Geology and mammalian paleontology of the Sand
premolars and the development of lophs and selenes reached its Creek facies, lower Willwood Formation (Lower Eocene), Washakie
extreme in Meniscotherium, which in the past has been placed in County, Wyoming. Geological Survey of Wyoming Memoir, 2,
1-151.
its own family with some European forms because of these dental
Carroll, R. L. (1988). Vertebrate Paleontology and Evolution. New York:
extremes. Freeman.
Postcranially, the best known representative of the family in North Cifelli, R. L. (1983). The origin and affinities of the South American Condy-
America is the sheep-sized Phenacodus (Figure 20.5D). Its limbs larthra and early Tertiary Litopterna (Mammalia). American Museum
were somewhat elongated, but in flexibility they were still similar Novitates, 2772, 1-48.
Cope, E. D. (1872a). Description of some new Vertebrata from the Bridger
to arctocyonids. The one exception may be in the carpus and tarsus,
Group of the Eocene. Proceedings of the American Philosophical
where mobility may have been reduced. Also, although the pes and Society, 12, 460-65.
manus are pentadactyl, digits I and V are reduced (Carroll, 1987). (l 872b ). Notices of new Vertebrata from the upper waters ofBitterCreek,
Phenacodontidae, in the sense of Prothero, Manning, and Fischer Wyoming Territory. Proceedings of the American Philosophical So-
(1988), who excluded Tetraclaenodon, first appeared in the Tiffa- ciety, 12, 483-86. (Published first as Paleo. Bull., No. 6, Aug. 20,
1872, reprinted Jan. 1873, in PAPS.)
nian (late Paleocene) and were never taxonomically diverse (al-
(1873). Fourth notice of extinct Vertebrata from the Bridger and Green
though sometimes numerically abundant). The taxon disappeared River Territories. Paleontological Bulletin, 17, 1-4.
from the North American fossil record in the Bridgerian (early mid- (1874). Report on vertebrate fossils discovered in New Mexico, with
dle Eocene). description of new species. Geographical Explorations and Surveys
 Archaic ungulates ("Condylarthra") 327

West of the JOOth Meridian, Appendix FF, Annual Report of the Chief tion near the Cretaceous-Tertiary boundary, Saskatchewan, Canada.
of Engineers for 1874, pp. 1-18. Palaeontographica A, 208, 11-59.
(1875a). Systematic catalogue of Vertebrata of the Eocene of New Gauthier, J. A., Estes, R. D., & de Queiroz, K. (1988). A phylogenetic
Mexico by parties of the expedition of 1874, Chapter 12: Fossils of analysis of Lepidosauromorpha. In Phylogenetic Relationships of
the Eocene period. Geographical Surveys west of the JOOth Merid- the Lizard Families. Essays Commemorating Charles L. Camp, eds.
ian, G. M. Wheeler, Corps of Engineers, U.S. army, Washington, 4, R. D. Estes & G. Pregill, pp. 15-98. Palo Alto: Stanford University
37-282. Press.
(1875b). On the supposed Carnivora of the Eocene of the Rocky Moun- Gazin, C. L. (1939). A further contribution to the Dragon Paleocene fauna
tains. Paleontological Bulletin, 20, 1-4. of central Utah. Journal of the Washington Academy of Sciences, 29,
(1880). The northern Wasatch fauna. American Naturalist, 14, 908. 273-86.
(188la). Mammalia of the lower Eocene beds. American Naturalist, 15, (194la). Paleocene mammals from the Denver Basin, Colorado. Journal
337-8. of the Washington Academy of Sciences, 31, 289-95.
(1881 b). The temporary dentition of a new creodont. American Natural- (1941b). The mammalian faunas of the Paleocene of central Utah, with
ist, 15, 667-9. notes on the geology. Proceedings of the United States National
(188lc). Mammalia of the lower Eocene. American Naturalist, 15, Museum, 91, 1-53.
829-31. (1956). Paleocene mammalian faunas of the Bison Basin in south-central
(188ld). On some Mammalia of the lowest Eocene beds of New Mexico. Wyoming. Smithsonian Miscellaneous Collections, 131, 1-57.
Proceedings of the American Philosophical Society, 19, 484-95. (1959). Early Tertiary Apheliscus and Phenacodaptes as pantolestid in-
(188le). A new type of Perissodactyla. American Naturalist, 15, sectivores. Smithsonian Miscellaneous Collections, 139, 1-7.
1017-18. (1965). A study of the early Tertiary condylarthran mammal Menis-
(188lf). Notes on Creodonta. American Naturalist, 15, 1018-20. cotherium. Smithsonian Miscellaneous Collections, 149, 1-111.
(1882a). Contributions to the history of the Vertebrata of the lower (1968). A study of the Eocene condylarthran mammal Hyopsodus. Smith-
Eocene of Wyoming and New Mexico, made during 1881. Proceed- sonian Miscellaneous Collections, 153, 1-90.
ings of the American Philosophical Society, 20, 139-97. Gidley, J. W. (1919). New species of claenodonts from the Fort Union (basal
(1882b). A classification of the ungulate Mammalia. Proceedings of the Eocene) of Montana. Bulletin of the American Museum of Natural
American Philosophical Society, 20, 438-47. History, 41, 541-56.
(1882c). Synopsis of the Vertebrata of Puerco epoch. Proceedings of the Giebel, C. G. (1855). Die Saugethiere in zoologischer, anatomischer und
American Philosophical Society, 20, 461-71. palaeontologischer Beziehung umfassend dargestellt. Leipzig.
(1882d). New characters of the Perissodactyla Condylarthra. American Gingerich, P. D. (1976). Paleontology and phylogeny: patterns of evolution
Naturalist, 16, 334. at the species level in early Tertiary mammals. American Journal of
(1882e). Notes on Eocene Mammalia. American Naturalist, 16, 522. Science, 276, 1-28.
(1882f). On the Taxeopoda, a new order of Mammalia. American Natu- (1977). Aletodon gunnelli, a new Clarkforkian hyopsodontid (Mam-
ralist, 16, 522-3. malia, Condylarthra) from the early Eocene of Wyoming. Contribu-
(1882g). Two new genera of the Puerco Eocene. American Naturalist, tions from the Museum of Paleontology, the University of Michigan,
16, 417-18. 24, 237-44.
(I 882h). New marsupials from the Puerco Eocene. American Naturalist, (1978). New Condylarthra (Mammalia) from the Paleocene and early
16, 684-6. Eocene of North America. Contributions from the Museum of Pale-
(1882i). The Periptychidae. American Naturalist, 16, 832-3. ontology, the University of Michigan, 25, 1-9.
(I 882j). Some new forms from the Puerco Eocene. American Naturalist, (1979). Lambertocyon eximius, a new arctocyonid (Mammalia, Condy-
16, 833-4. larthra) from the late Paleocene of western North America. Journal
(1883a). On the mutual relationships of the bunotherian Mammalia. of Paleontology, 53, 524-9.
Proceedings of the Academy of Natural Sciences Philadelphia, 35, (1981). Radiation of early Cenozoic Didymoconidae (Condylarthra,
77-83. Mesonychia) in Asia, with a new genus from the early Eocene of
(1883b). On some fossils of the Puerco Formation. Proceedings of the western North America. Journal of Mammalogy, 62, 526-38.
Academy of Natural Sciences Philadelphia, 35, 168-70. (1983). New Adapisoricidae, Pentacodontidae, and Hyopsodontidae
(1883c). First addition to the fauna of the Puerco Eocene. Proceedings (Mammalia, Insectivora and Condylarthra) from the late Paleocene
of the American Philosophical Society, 20, 545-63. of Wyoming and Colorado. Contributions from the Museum of Pale-
(1884a). Second addition to the knowledge of the fauna of Puerco epoch. ontology, the University of Michigan, 26, 227-55.
Proceedings of the American Philosophical Society, 21, 309-24. (1989). New earliest Wasatchian mammalian fauna from the Eocene
(1884b). The Condylarthra. American Naturalist, 18, 790-805. of northwestern Wyoming: Composition and diversity in a rarely
(1884c). The Vertebrata of the Tertiary formations of the West, Book 1. sampled high-floodplain assemblage. University ofMichigan Papers
Report of the United States Geological Survey of the Territories, 3, on Paleontology, 28, 1-97.
1-1009. Gingerich, P. D., Damning, D. P., Blane, C. E., & Uden, M. D. (1994).
(1885). The oldest Tertiary Mammalia.American Naturalist, 19, 385-86. Cranial morphology of Protosirenfraasi (Mammalia, Sirenia) from
( 1888). Synopsis of the vertebrate fauna of the Puerco series. Transac- the middle Eocene of Egypt: a new study using computed tomog-
tions of the American Philosophical Society, 16, 298-361. raphy. Contributions of the Museum of Paleontology, University of
Dorr, J. A., Jr. (1952). Early Cenozoic stratigraphy and vertebrate paleontol- Michigan, 29, 41-67.
ogy of the Hoback Basin, Wyoming. Geological Society of America Granger, W. (1915). A revision of the lower Wasatch and Wind River faunas,
Bulletin, 63, 59-94. Part III: Order Condylarthra, families Phenacodontidae and Menis-
Earle, C. (1893). On the systematic position of the genus Protogonodon. cotheriidae. Bulletin of the American Museum of Natural History,
American Naturalist, 27, 377-9. 34, 329-61.
Fischer, M. S. (1989). Hyracoids, the sister-group of perissodactyls. In The Gregory, W. K. ( 1951 ). Evolution Emerging. A Survey of Changing Patterns
Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch, from Primeval Life to Man, Vol. II. New York: Macmillian.
pp. 37-56. Oxford: Clarendon. Irwin, D. M., & Wilson, A. C. (1993). Limitations of molecular methods
Fox, R. C. (1989). The Wounded Knee local fauna and mammalian evolu- for establishing the phylogeny of mammals, with special reference
 J. David Archibald 328

to the position of elephants. In Mammal Phylogeny: Placentals, eds. fiict or Compromise?, ed. C. Patterson, pp. 55-93. Cambridge:
F. S. Szalay, M. J. Novacek, & M. C. McKenna, pp. 257-67. New Cambridge University Press.
York: Springer-Verlag. Meng, J. (1990). Morphology, systematics, and distribution of Didymo-
Jepsen, G. L. (1930). New vertebrate fossils from the lower Eocene of the conidae, a mammalian group from early Tertiary of Asia. Journal
Bighorn Basin, Wyoming. Proceedings of the American Philosoph- of Vertebrate Paleontology. Abstracts of Papers, 10 (suppl. to no. 3),
ical Society, 69, 117-31. 35A.
Johnston, P. A., & Fox, R. C. (1984). Paleocene and late Cretaceous Muizon, de, C., & Marshall, L. G. (1987a). Le plus ancien condylarthre
mammals from Saskatchewan, Canada. Palaeontographica A, 186, (Mammalia) sud-americain (Cretace superieur, Bolivie). Comptes
166-222. Rendus de l'Academie des Sciences, Serie II, 304, 771-4.
Kretzoi, M. (1943). Kochictis centenii n. g. n. sp., ein altertiimlicher Cre- ( l 987b ). Deux nouveaux condylarthres (Mammalia) du Maastrichtien de
odonte aus dem oberoligozan Siebenbiirgens. Foldtani Kozlony, 73, Tiupampa (Bolivie). Comptes Rendus de l'Academie des Sciences,
190--5. Serie II, 304, 947-50.
Krishtalka, L. (1976). Early Tertiary Adapisoricidae and Erinaceidae (Mam- (1991). Nouveaux condylarthres du Paleocene inferieur de Tiupampa
malia, Insectivora) of North America. Bulletin of Carnegie Museum (Bolivie). Bulletin du Museum National d'Histoire Nature Ile, section
of Natural History, 1, 1-40. C, 4' Serie, 13, 201-27.
Leidy, J. ( 1870). Remarks on a collection of fossils from the western territo- Murray, A. (1866). The Geographic Distribution ofMammals. London: Day
ries. Proceedings of the Academy of Natural Sciences Philadelphia, and Son.
22, 109-10. Nessov, L. A. (1985). New mammals from the Cretaceous of Kyzylkum.
Lemoine, V. (1891). Etude d'ensemble sur Jes dents mammirere fossiles Vestnik Leningradoskogo Universiteta Seriya 7, 17, 8-18 (in Russian
des environs de Reims. Bulletin de la Societe geologique de France, with English summary).
ser. 3, 24, 333-44. (1990). Small ichthyornithiform bird and other bird remains from the Bis-
Lillegraven, J. A. (1969). Latest Cretaceous mammals of upper part of Ed- sekty Formation (Upper Cretaceous) of Central Kyzylkum Desert.
monton Formation of Alberta, Canada, and review of marsupial- Trudy Zoologicheskogo Jnsituta Akademiya Nauk, SSR, 21, 59-62.
placental dichotomy in mammalian evolution. University of Kansas (in Russian with English summary).
Paleontological Contributions, 50, 1-122. Nessov, L.A., Archibald, J. D., & Kielan-Jaworowska, Z. (1997). Ungulate-
Linnaeus, C. ( 1758). Systema naturae regna tria naturae, secundum classes, like mammals from the Late Cretaceous of Uzbekistan and a phy-
ordines, genera, species cum characteribus, differentiis, synonymis, logenetic analysis of the Ungulatomorpha. Bulletin of the Carnegie
locis. Editio decima, reformata. Stockholm: Laurentii Salvii. Museum of Natural History (in press).
(1766). Systema naturae regna tria naturae, secundum classes, ordines, Nessov, L.A., & Kielan-Jaworowska, Z. ( 1991 ). Evolution of the Cretaceous
genera, species cum characteribus, differentiis, synonymis, locis. Asian therian mammals. In Fifth Symposium on Mesozoic Terrestrial
Editio duodecima, reformata. Stockholm: Laurentii Salvii. Ecosystems and Biota. Extended Abstracts. Contributions from the
Lofgren, D. L. (1995). The Bug Creek problem and the Cretaceous-Tertiary Paleontological Museum, University of Oslo, 364, 51-2.
transition at McGuire Creek, Montana. University ofCalifornia Pub- Nessov, L.A., Sigogneau-Russell, D., & Russell, D. E. (1994). A survey of
lications in Geological Sciences, 140, 1-185. Cretaceous tribosphenic mammals from middle Asia (Uzbekistan,
Luo, Z. (1991). Variability of dental morphology and the relationships of the Kazakhstan and Tajikistan), of their geological setting, age and fauna!
earliest arctocyonid species. Journal of Vertebrate Paleontology, 11, environment. Palaeovertebrata, 23, 51-92.
452-71. Novacek, M. J. (1982). Information for molecular studies from anatomi-
Marsh, 0. C. (1876). Notice of new Tertiary mammals, V. American Journal cal and fossil evidence on higher eutherian phylogeny. In Macro-
of Science, 112, 401-4. molecular Sequences in Systematic and Evolutionary Biology, ed.
Matthew, W. D. (1897). A revision of the Puerco fauna. Bulletin of the M. Goodman, pp. 3-41. New York: Plenum.
American Museum of Natural History, 9, 259-323. (1986). The skull of leptictid insectivorans and the higher-level classifi-
(1909). The Carnivora and Insectivora of the Bridger Basin, middle cation of eutherian mammals. Bulletin of the American Museum of
Eocene. Memoirs of the American Museum of Natural History, 9, Natural History, 183, I-ill.
289-567. (1992). Mammalian phylogeny: shaking the tree. Nature, 356, 121-5.
(1915a). A revision of the lower Eocene Wasatch and Wind River faunas, Novacek, M. J., Ferrusqufa-V. I., Flynn, J. J., Wyss, A. R., & Norell,
Part I: Order Ferae (Carnivora). Suborder Creodonta. Bulletin of the M. (1991). Wasatchian (early Eocene) mammals and other verte-
American Museum of Natural History, 34, 4-103. brates from Baja California, Mexico: the Lomas las Tetas de Cahra
(1915b). A revision of the lower Eocene Wasatch and Wind River faunas, fauna, Bulletin of the American Museum of Natural History, 208,
Part II: Order Condylarthra, family Hyopsodontidae. Bulletin of the 1-88.
American Museum of Natural History, 34, 311-28. Novacek, M. J., & Wyss, A. R. (1986). Higher-level relationships of
(1918). A revision of the lower Eocene Wasatch and Wind River fau- the recent eutherian orders: morphological evidence. Cladistics, 2,
nas, Part V: Insectivora (continued), Glires, Edentata. Bulletin of the 257-87.
American Museum of Natural History, 38, 565-657. Osborn, H. F. (1910). The Age of Mammals in Europe, Asia, and North
(1937). Paleocene faunas of the San Juan Basin, New Mexico. Transac- America. New York: Macmillan.
tions of the American Philosophical Society, n.s., 30, 1-510. Osborn, H. F., & Earle, C. (1895). Fossil mammals of the Puerco beds.
Matthew, W. D., & Granger, W. (1921). New genera of Paleocene mammals. Collection of 1892. Bulletin of the American Museum of Natural
American Museum Novitates, 13, 1-7. History, 7, 1-70.
McKenna, M. C. (1960). Fossil Mammalia from the early Wasatchian Four Osborn, H. F., & Wortman, J. L. (1892). Fossil mammals of the Wasatch
Mile Fauna, Eocene of northwest Colorado. University of California and Wind River beds. Collections of 1891. Bulletin of the American
Publications in Geological Sciences, 37, 1-130. Museum of Natural History, 4, 6-147.
(1975). Towards a phylogenetic classification of the Mammalia. In Patterson, B., & McGrew, P. 0. (1962). A new arctocyonid from the Pale-
Phylogeny of the Primates, eds. W. P. Luckett & F. S. Szalay, pp. ocene of Wyoming. Breviora, 174, 1-10.
21-46. New York: Plenum. Patterson, B., & West, R. M. (1973). A new late Paleocene phenacodont
(1987). Molecular and morphological analysis of high-level mammalian (Mammalia, Condylarthra) from western Colorado. Breviora, 403,
interrelationships. In Molecules and Morphology in Evolution: Con- 1-7.
 Archaic ungulates ("Condylarthra") 329

Prothero, D.R. (1993). Ungulate phylogeny: molecular versus morpholog- Szalay, F. S. (l 969a). The Hapalodectinae and a phylogeny of the Mesonychi-
ical evidence. In Mammal Phylogeny, vol. 2: Placentals, eds. F. S. dae (Mammalia, Condylarthra). American Museum Novitates, 2361,
Szalay, M. J. Novacek, & M. C. McKenna, pp. 173-81. New York: 1-26.
Springer-Verlag. (l 969b ). Origin and evolution of function of the mesonychid condylarth
Prothero, D.R., Manning, E. M., & Fischer, M. (1988). The phylogeny of the feeding mechanism. Evolution, 23, 703-20.
ungulates. In The Phylogeny and Classification ofthe Tetrapods, Vol. (1977). Phylogenetic relationships and a classification of the eutherian
II: Mammals, ed. M. J. Benton, pp. 201-34. Systematic Association Mammalia. NATO Advanced Study Institute, Series A, 14, 315-74.
Special Vol. no. 35B. Oxford: Clarendon. Szalay, F. S., & Gould, S. J. (1966). Asiatic Mesonychidae (Mammalia,
Reynolds, T. E. (1936). Two new insectivores from the lower Paleocene of Condylarthra). Bulletin of the American Museum of Natural History,
New Mexico. Journal of Paleontology, 10, 202-9. 132, 131-73.
Rigby, J. K., Jr. (1980). Swain Quarry of the Fort Union Formation, middle Thewissen, J. G. M. (1990). Evolution of Paleocene andEocenePhenacodon-
Paleocene (Torrejonian), Carbon County, Wyoming: geologic setting tidae (Mammalia, Condylarthra). University of Michigan Papers on
and mammalian fauna. Evolutionary Monographs, 3, 1-179. Paleontology, 29, 1-107.
(1981). A skeleton of Gillisonchus gillianus (Mammalia; Condylarthra) (1994). Phylogenetic aspects of cetacean origins: a morphological per-
from the early Paleocene (Puercan) Ojo Alamo Sandstone, San Juan spective. Journal of Mammalian Evolution, 2, 157-84.
Basin, New Mexico, with comments on the local stratigraphy of Be- Thewissen, J. G. M., & Domning, D. P. (1992). The role of phenacodontids
tonnie Tsosie Wash. In Advances in San Juan Basin Paleontology, in the origin of the modern orders of ungulate mammals. Journal of
eds. S. G. Lucas, J. K. Rigby, Jr., & B. S. Kues, pp. 89-126. Albu- Vertebrate Paleontology, 12, 494-504.
querque: University of New Mexico Press. Ting, S., & Li, C. (1987). The skull of Hapalodectes (?Acreodi, Mammalia),
Robison, S. F. (1986). Paleocene (Puercan-Torrejonian) mammalian faunas with notes on some Chinese Paleocene mesonychids. Vertebrata
of the North Horn Formation, central Utah. Brigham Young Unive r- Palasiatica, 25, 161-86.
sity Geology Studies, 33, 87-133. Trouessart, E.-L. (1879). Catalogue des mammireres vivants et fossiles. Re-
Rose, K. D. (1981). The Clarkforkian land-mammal age and mammalian vue et magazine de zoologie pure et applique, sfr 3, 7, 219-85.
fauna! composition across the Paleocene-Eocene boundary. Univer- (1904). Catalogus mammalium tam viventium quam fossilium. Quin-
sity of Michigan Papers on Paleontology, 26, 1-197. quennale supplementum. Berlin: Berolini, R. Friedlander iun Sohn
(1987). Climbing adaptations in the early Eocene mammal Chriacus and (vol. 1:1-546, 1904; vol. 2:547-929, 1905).
the origin of Artiodactyla. Science, 236, 314-6. Van Valen, L. (1966). Deltatheridia, a new order of mammals. Bulletin of
Russell, D. E. (1964). Les mammireres paleocenes d'Europe. Memoires du the American Museum of Natural History, 132, 1-126.
Museum National d'Histoire Naturelle, Serie C, 8, 1-324. (1978). The beginning of the age of mammals. Evolutionary Theory, 4,
(1980). Sur Jes condylarthres Cernaysiens Tricuspiodon et Landenodon 45-80.
(Paleocene superieur de France). Palaeovertebrata Mem. Jubil. R. Weigelt, J. (1960). Die Arctocyoniden von Walbeck. Freiberger Forschung-
Lavocat, pp. 127-66. shefte, 77, 1-241.
Russell, D. E., & Zhai, R. (1987). The Paleogene of Asia: mammals and West, R. M. (1976). The North American Phenacodontidae (Mammalia,
stratigraphy. Memoires du Museum National d'Histoire Naturelle, Condylarthra). Contributions in Biology and Geology, Milwaukee
Serie C, 52, 1-488. Public Museum, 6, 1-78.
Russell, L. S. (1974). Fauna and correlation of the Ravenscrag Formation Wible, J. R. (1986). Transformations in the extracranial course of the inter-
(Paleocene) of southwestern Saskatchewan. Royal Ontario Museum nal carotid artery in mammalian phylogeny. Journal of Vertebrate
Life Science Contributions, 102, 1-53. Paleontology, 6, 313-25.
Scott, W. B. (1888). On some new and little known creodonts. Journal of (1987). The eutherian stapedial artery: character analysis and implica-
the Academy of Natural Sciences Philadelphia, 9, 155-85. tions for superordinal relationships. Zoological Journal of the Lin-
( 1892). A revision of the North American Creodonta, with notes on some nean Society, 91, 107-35.
genera which have been referred to that group. Proceedings of the Wiley, E. 0. (1981). Phyla genetics. The Theory and Practice ofPhylogenetic
Academy of Natural Sciences Philadelphia, 44, 291-323. Systematics. New York: Wiley.
Simpson, G. G. (1935). New mammals from the Fort Union of Montana. Williamson, T. E., & Lucas, S. G. (1992). Meniscotherium (Mammalia,
Proceedings of the United States National Museum, 83, 221-4. "Condylarthra") from the Paleocene-Eocene of western North Amer-
(1936a). Additions to the Puerco fauna, lower Paleocene. American Mu- ica. Bulletin of the New Mexico Museum of Natural History and
seum Novitates, 849, 1-11. Science, 1, 1-75.
(1936b). Carsioptychus, '!new name for Plagioptychus Matthew, nee Wilson, R. W. (1952). Paleocene mammalian genus Ellipsodon. Geological
Matheron. American Journal of Science, 232, 234. Society of America Bulletin, 63, 1315-6.
(1937). The Fort Union of the Crazy Mountain Field, Montana and its (1956). The condylarth genus Ellipsodon. University of Kansas Publi-
mammalian faunas. United States National Museum Bulletin, 169, cation, Museum of Natural History, 9, 105-16.
1-287. Wortman, J. L. ( 1901 ). Studies of Eocene Mammalia in the Marsh collection,
(1945). The principles of classification and a classification of mammals. Peabody Museum. American Journal of Science, 162, 281-96.
Bulletin of the American Museum of Natural History, 85, 1-350. Wyss, A. R., Novacek, M. J., & McKenna, M. C. (1987). Amino acid se-
Sloan, R.E., Rigby,J. K.,Jr., Van Valen,L. M.,&Gabriel,D. (1986). Gradual quence versus morphological data and the interordinal relationships
dinosaur extinction and simultaneous ungulate radiation in the Hell of mammals. Molecular Biology and Evolution, 4, 99-116.
Creek Formation. Science, 232, 629-33. Zhou, X., Zhai, R., Gingerich, P. D., & Chen, L. (1995). Skull of a new
Sloan, R. E., & Van Valen, L. (1965). Late Cretaceous mammals from Mon- mesonychid (Mammalia, Mesonychia) from the Late Paleocene of
tana. Science, 148, 220--7. China. Journal of Vertebrate Paleontology, 15, 387-400.
 Appendix:

This classification is intended to reflect as much as possible the cladogram in Figure 20.6, but the latter should be regarded as depicting
most closely the views of relationship expressed in this chapter. Conventions used in this classification are as follows:
(1) Linnean ranks or categories are not used; rather, relative rank is signified by successive levels of indentation and by the sequential
numbering of taxa within Ungulatomorpha (i.e., 1.0, 1.1, 1.2, etc.). (2) Sequentially numbered taxa at the same level of indentation
represent successive dichotomous branchings such that each taxon is the sister taxon of all subsequent taxa in the list (Convention 3 of
Wiley, 1981) and in the cladogram in Figure 20.6. (3) Indentation signifies hierarchy and decreasing taxonomic level toward the right of
the page. (4) Identical numbers at the same level of indentation indicate trichotomous or polytomous branchings of the tax on in question.
(5) A taxon or number followed by a question mark indicates an uncertain reference to the more inclusive named taxon. Such
questionable taxa were not considered in the review of character-state changes. (6) Taxa followed by two asterisks are mixotaxa. (It
cannot be determined whether they are or are not monophyletic.) (7) Taxa enclosed by quotation marks are nonmonophyletic, indicating
that some taxa included within the nonmonophyletic taxon share a more recent common ancestor with the taxon immediately following
the nonmonophyletic taxon. See text for further comment.

330
 Appendix 331

UNGULATOMORPHA Archibald, 1996b 3.0 PERIPTYCHIDAE Cope, 1882i

1.0 "ZHELESTIDAE" Nessov (1985), 1990 4.0 PERIPTYCHIDAE?: Escatepos Reynolds, 1936
2.0 Alostera Fox, 1989 4.0 PERIPTYCHIDAE?: Maiorana Van Valen,
1.1 UNGULATA Linnaeus, 1766 1978
2.0 ARTIODACTYLA Owen, 1848 4.0 PERIPTYCHIDAE?: Tinuviel Van Valen, 1978
2.1 Protungulatum* Sloan and Van Valen, 1965 4.0 "Mimatuta" Van Valen, 1978
2. 1 ARCTOCYONIDAE (Giebel, 1855) Murray, 4.0 Unnamed taxon: PERIPTYCHINAE +
1866 ANISONCHINAE + CONACODONTINAE
3.0 ARCTOCYONIDAE?: Carcinodon Scott, 5.0 PERIPTYCHINAE Osborn and Earle, 1895
1892 6.0 Hemithlaeus Cope, 1888I
3.0 ARCTOCYONIDAE?: Oxytomodon Gazin, 6.1 Ectoconus Cope, 1884b
194lb 6.2 Carsioptychus Simpson, 1936b
3.0 Desmatoclaenus Gazin, 194lb 6.2 Periptychus Cope, 188la
3.0 Princtonia Gingerich, 1989 5.0 ANISONCHINAE
3.0 Prothryptacodon Simpson, 1935 6.0 Gillisonchus Rigby, 1981
3.0 Unnamed taxon: Chriacus 6.1 "Anisonchus" Cope, 188 lc
+ Deuterogonodon + Loxolophus 6.1 Haploconus Cope, 1882g
·+ Mimotricetes + Lambertocyon 5.0 CONACONDONTINAE
+ Mentoclaenodon + Arctocyoninae 6.0 Conacodon Matthew, 1897
4.0 Chriacus Cope, 1883a 6.0 Oxyacodon** Osborn and Earle, 1895
4.1 Unnamed taxon: Deuterogonodon 2.3 CETUNGULATA Irwin and Wilson, 1993
+ Loxolophus 3.0 CETE Linnaeus, 1758
4.2 Mimotricentes Simpson, 1937 4.0 CETE?: Platymastus Van Valen, 1978
4.3 Lambertocyon Gingerich, 1979 4.0 CETE?: lfyolestes Gingerich, 1981
4.3 Mentoclaenodon Weigelt, 1960 4.0 Baioconodon Gazin, 1941a
4.3 ARCTOCYONINAE Giebel, 1855 4.1 Ragnarok Van Valen, 1978
5.0 Anacodon Cope, 1882a 4.1 Stelocyon Gingerich, 1978
5.0 Arctocyon Blainville, 1841 4.1 Oxyclaenus Cope, 1884a
5.0 Colpoclaenus Patterson and 4.2 Goniaconodon Cope, 1888
McGrew, 1962 4.3 Eoconodon Matthew and Granger, 1921
5.0 Thryptacodon Matthew, 1915a 4.4 Triisodon Cope, 1881 b
2.2 HYOPSODONTIDAE Trouessart, 1879 4.5 ?Microclaenodon Scott, 1892
3.0Aletodon Gingerich, 1977 4.5 HAPALODECTIDAE (Gould and Szalay, 1966)
'3.0 Dorraletes Gingerich, 1983 Ting and Li, 1987, including only Hapalodectes
3.0 Haplaletes Simpson, 1935 Matthew, 1909
3.0 Haplomylus Matthew, 1915b 4.6 MESONYCHIDAE Cope, 1875b
3.0 Hyopsodus Leidy, 1870 5.0 Dissacus Cope, 1881f
3.0 Litomylus Simpson, 1935 5.1 Ankalagon Van Valen, 1978
3.0 Oxyprimnus Van Valen, 1978 5.2 Pachyaena Cope, 1874
3.0 Utemylus Gingerich, 1983 5.3 Synoplotherium Cope, 1872b
2.2 BULBULODENTATA new 5.4 Harpagolestes Wortman, 1901
3.0 MIOCLAENIDAE Osborn and Earle, 1895 5.5 Mesonyx Cope, 1872a
4.0 Bubogonia Johnston and Fox, 1984 4.6 CETACEA Brisson, 1762
4.0 Choeroclaenus Simpson, 1937 3.0 TAXEOPODA Cope, 1882f
4.0 Ellipsodon Scott, 1892 4.0 PHENACODONTA McKenna, 1975
4.0 Litaletes Simpson, 1935 5.0 Tetraclaenodon Scott, 1892
4.0 Mioclaenus Cope, 188lc 5.0 PHENACODONTIDAE Cope, 1881e
4.0 Promioclaenus Trouessart, 1904 6.0 Phenacodus Cope, 1873
4.0 Protoselene Matthew, 1897 6.0 Copecion Gingerich, 1989
4.0 Iiznatzinia Simpson, 1936a 6.0 Meniscotheriinae (Cope, 1882d)
4.0 MIOCLAENIDAE?: APHELISCINAE Van Valen, 1978
(Matthew, 1918) 7 .0 Ectocion Cope, 1882e
5.0 Apheliscus Cope, 1875a 7.0 Meniscotherium Cope, 1874
5.0 Phenacodaptes Jepsen, 1930 4.1 PAENUNGULATA Simpson, 1945
21 Arctostylopida
RICHARD L. CIFELLI and CHARLES R. SCHAFF

INTRODUCTION

The Arctostylopida constitute a small, enigmatic group of early Ter-

tiary ungulates, predominantly Asiatic in distribution but with one
species occurring in North America. The group was long placed in
the great, extinct neotropical order Notoungulata because of strik-
ingly similar modifications of the dentition. Because of their Asiatic-
North American distribution and resemblance to notoungulates, arc-
tostylopids have figured prominently in the development of faunal
correlation with Asia and in discussions regarding the origin of the
South American land mammal fauna. However, recent study indi-
cates notoungulate similarities to have been acquired independently
by arctostylopids. The group, consisting of the single family Arc-
tostylopidae, is currently placed in its own order, to which higher
relationships are uncertain.
Very little other than the dental anatomy is known of the Arctosty-
lopidae. The generally unreduced dentition is characterized by the Figure 21.1. Restoration of Arctostylops steini (by Brian Regal).
development of salient lophs with anteroposteriorly oriented vertical
shearing surfaces on the posterior cheek teeth. This, coupled with centrocrista, becoming a salient, straight ectoloph in advanced gen-
the small size of all known taxa, suggests that arctostylopids were era; parastyle usually prominent. Pre- and postprotocristae of upper
advanced, rabbit-sized herbivores adaptively similar to the hyraxes molars strong, conules lacking; upper molars primitively triangular
of Old World Tertiary and Recent faunas (Figure 21.1). Arctosty- but M 1-2 becoming quadrate in advanced species by the addition of
lopids appear to have originated in Asia, where they are represented a pseudohypocone (i.e., a cusp in the position of a hypocone but not
by eight genera. Most of these are of late Paleocene age, but the derived from the postcingulum; cf. Simpson, 1929). Anterior lower
family apparently lingered at least until the early or middle Eocene premolars serially tricuspid, with strong shearing surfaces; lower
on that continent. The single North American species is represented molars primitively with bicrescentic lophs; the paracristid is lost and
by a handful of specimens from Tiffanian and Clarkforkian horizons various accessory trigonid structures are acquired in advanced taxa.
in the central Rocky Mountain region of the United States. Lower molar hypoconid indistinct; entoconid transversely expanded
and, in advanced species, developed into an oblique entolophid. (see
Figure 21.2.)
DEFINING FEATURES OF THE ORDER
ARCTOSTYLOPIDA AND THE
FAMILY ARCTOSTYLOPIDAE SYSTEMATICS

Upper and lower dentitions forming an evenly graded series; canines

SUPRAFAMILY
poorly or not differentiated and without diastemata separating them
from adjacent teeth. Posteriorupper premolars somewhat molarized; The first known member of the family, Arctostylops steini, was de-
P4, at least, with a metacone. Upper molars with well-developed scribed by Matthew (1915) from the "lower Gray Bull beds, Clark
332
 Arctostylopida 333

A -- ---::;:--- .. -- ----- - ---=-===--=-:=---------- the most primitive family of the order, with the Arctostylopidae
constituting an aberrant early side branch derived from southern
ancestors.
Vigorous collecting programs in Asia have led to the addition
of six genera to the family in recent years (Tang and Yan, 1976;
Zhai, 1978; Zheng, 1979; Zheng and Huang, 1986; Nessov, 1987),
indicating that arctostylopids underwent a modest radiation on that
continent in the late Paleocene and early Eocene. Review of morpho-
logical diversity within the Arctostylopidae, coupled with revision
of the early Tertiary South American Notoungulata (Simpson, 1948,
1967), has led to a substantially different view of arctostylopid rela-
tionships than that based solely on Arctostylops and Palaeostylops
(Cifelli, Schaff, and McKenna, 1989). Virtually all resemblances
cm (including, for instance, reduction of lower molar trigonids, addi-
tion of accessory structures to those teeth, development of a talonid
B
on p4, presence of a high, smooth ectoloph on upper molars) of
these advanced genera to bona fide Notoungulata appear to have
been acquired independently within both groups. In certain cases,
such as with the posterointemal cusp ofupper molars, structural sim-
ilarities are evidently nonhomologous (Cifelli, 1983; Schaff, 1985).
When morphotypes for the two groups are compared, the single (and
somewhat questionable) remaining derived resemblance is the in-
cipient presence of an entolophid on the lower molars, a feature that
has evolved independently among several other mammalian groups.
Specializations of the arctostylopid ankle are similar to those seen
in Pseudictops, which, in tum, has been suggested as a relative of
Lagomorpha (Szalay and McKenna, 1971). Saltatory adaptations
seen lagomorphs and suspected allies are lacking in arctostylopids,
1 cm however, and at present, there is no unambiguous evidence as to the
higher relationships of the group.
Figure 21.2. Representative arctostylopid cranial and dental morphology. Notoungulata deserve further passing mention as a "near
A. Gashatostylops macrodon (restoration based on several specimens). miss" among North American Tertiary mammals. With the for-
B. Arctostylops steini (MCZ 2004).
mation of the Panamanian land bridge about 3 million years ago
(Marshall et al., 1979), a comprehensive intermingling ofland mam-
Fork Basin, Wyoming." Matthew astutely noted the striking sim- mal faunas occurred between North and South America. Of the
ilarity of the species to members of the Notoungulata, an extinct many indigenous South American ungulate groups, only notoun-
order otherwise restricted to South America, and placed A. steini gulates are known to have dispersed northward to North America
in the early Tertiary family Isotemnidae (a group then, as now, (as defined tectonically). Both early and late Pleistocene faunas of
conceived as a wastebasket of generally primitive notoungulates). El Salvador and Honduras include at least one notoungulate, Mixo-
Because the South American Tertiary mammal fauna had already toxodon (Van Frank, 1957; Webb and Perrigo, 1984). This genus
been recognized as insular and therefore endemic (Gaudry, 1906), is referred to the Toxodontidae, including mainly large, hippolike
there was initially some doubt that the single specimen of Arc- species, whose taxonomy ranks among the most confused of all
tostylops steini actually was collected in Wyoming (Matthew, 1915; mammals.
Jepsen and Woodbume, 1969; Gingerich, 1985). The distinctive-
ness of the North American species was recognized by Schlosser
INFRAFAMILY
(1923), who erected the family Arctostylopidae to contain it. This
view was substantiated by the discovery of two additional species, Most hypotheses of relationships within the family have been re-
both then placed in the genus Palaeostylops, from the Gashato fauna stricted to consideration of a few species only. Matthew and
of Mongolia (Matthew and Granger, 1925; Matthew, Granger, and Granger (1925) regarded the Mongolian species Palaeostylops itu-
Simpson, 1929). Matthew and co-authors, impressed by the relative rus as more primitive than, and therefore ancestral to, the North
primitiveness of the Asian genus, regarded it as ancestral to North American Arctostylops steini. This suggestion was based on the
American Arctostylops and, in a general sense at least, to southern "simple premolars" of P. iturus. Dashzeveg (1982), however, con-
Notoungulata. Simpson (1934) redefined the higher categories of sidered these and two other Mongolian species as congeneric, pos-
the order and erected a primitive suborder Notioprogonia to contain tulating A. steini to be the sister tax on of the Asian taxa. Dashzeveg's
the Arctostylopidae and two South American families. In Simp- arrangement was based on shared primitiveness of A. steini and P.
son's view (e.g., 1948), the neotropical Henricosbomiidae represent iturus with respect to Gashatostylops macrodon (all three species
 Richard L. Cifelli and Charles R. Schaff 334

NORTH
ASIAN AMERICAN

~
Cl)
.gc::
~ ~
Cl)
Cl)
0
e Cl)
.:2
.,g Cl)

-~
Cl)
Ol <ll
.!:Q
e0.. 2 "§ :s
g E Q)
Cll
~ l Cl) Cl)
·-.d Cl)
g.
~
Cl)
g.
"O
.s ~
Cl)

g. g. ~ ci. Cl)
Cl)
g. Cl)
g. ~ ~
S!
~
~ ~ CIJ g.
~ ~
"O
~ ~
Cl)
.g .g Cl)

2 ~
Cl)
S!
.9
c::
ell Cl)
.9
-~ ~ -s0 -s0 ~ Cl) Cll
c:
c::Q) ~
~ ~ ~
Cl)
<:( Ill Cl <:(
Ill <:(

Figure 21.3. Continental distribution and hypothesized Telationships of the Arctostylopidae. Left upper and right lower second molars. representing
major morphological types among the family, are of the following taxa (reversed where appropriate): base of cladogram, Asiostylops spanios; node 1,
Bothriostylops progressus; node 4, Arctostylops steini; node 7, Anatolostylops dubius. Characters at nodes: (1) Metaconid added to p3, pseudohypocone on
at least one upper molar, upper molar paracone fold lost, ectocingulid developed on lower molars, ectocingulid developed on p4, p4 talonid curved, upper
molar parastyle enlarged. (2) Lower molar cristid obliqua attaches lingually to rear of trigonid. (3) m3 elongate? (condition unknown in S. promissus). (4)
Lower molar entolophid well developed, lower molar cristid obliqua attaches labially to rear of trigonid, lower molar ectocingulid strong and pillar like, p4
paraconid shifted labially, lower molar paracristid lost. (5) Shearing notch developed on p4 talonid, P3-4 paracone fold lost, canines less differentiated. (6)
Second molars enlarged, pseudohypocone lost on Ml(?). (7) Upper molar protocristae salient, M2 pseudohypocone reduced(?), parastyle fold on ectoloph
of posterior upper cheek teeth lost, ectoloph of upper molars anteroposteriorly elongate, cheek teeth very high crowned.

were included in the genusArctostylops by Dashzeveg, 1982). Zheng advanced genera (including North American Arctostylops) form a
(1979) recognized Asiostylops, from the Paleocene of China, as the sister group (Figure 21.3). A number of shared specializations
most primitive member of the group. (mainly associated with the strong development of vertical shearing
Detailed comparison of all arctostylopids, with reference to an surfaces and corresponding distinctive coronal patterns) character-
"ungulate" morphotype as represented by the earliest "condylarth," ize the upper and lower dentitions of these remaining species, of
Protungulatum (from the Late Cretaceous), substantiates the posi- whichArctostylops steini appears to be the most primitive. Thus the
tion of Asiostylops as sister taxon of other known genera (Cifelli, only North American representative of the Arctostylopidae appears
Schaff, and McKenna, 1989). Several slightly more derived taxa, to lie phylogenetically nested within an otherwise Asiatic radiation
also from Asia, appear to form a monophyletic clade, to which (Figure 21.3). As developed here, this hypothesis of relationships
 Arctostylopida 335

has implications for both the direction and timing of the zoogeo- BIOLOGY AND EVOLUTIONARY PATTERNS
graphic deployment of the group.
The Arctostylopidae were an insignificant element in North Amer-
ica's Tertiary land mammal fauna. Arctostylops steini ranged from
INCLUDED NORTH AMERICAN GENERA IN THE late Tiffanian through Clarkforkian time, a wide temporal range for
ORDER ARCTOSTYLOPIDA mammalian species of that age. Nonetheless, the species was never
abundant or widely distributed. Climate in the Western Interior be-
The locality numbers listed for each genus refer to the list of unified came more equable, and diversity of the macrofiora apparently in-
localities in Appendix I. Question marks in front of the locality (e.g., creased during this time (Rose, 1981 ), but the significance of this
?CP101) mean the taxon is questionably known from that locality. trend for North American Arctostylopidae is unknown.
The acronyms for museum collections are listed in Appendix III. Pre-Wasatchian equivalent faunas of Asia are difficult to correlate
outside that continent because of their apparent endemism (Szalay
Arctostylops Matthew, 1915 and McKenna, 1971). Arctostylopidae are important in this respect
Type species: Arctostylops steini Matthew, 1915. because the similarity of North American Arctostylops to Asian
Type specimen: AMNH 16830. Palaeostylops and Gashatostylops provides some of the earliest,
Characteristics: Large arctostylopid with a salient lingual rib most compelling evidence of Paleogene interchange between the
on the lower canine; p4 with strongly curved talonid cres- two continents. The presence of Arctostylops (MCZ 2004, YPM-
cent and with ectocingulid. Distinguished from primitive PU 20397) in the Tiffanian of Wyoming and the relationship of
genera (Asiostylops, Bothriostylops) in having quadrate that genus to newly described Asian taxa indicates a late Paleocene
upper molars, with a lingual sulcus separating two internal (Szalay and McKenna, 1971), rather than early Eocene (i.e., Clark-
cusps on Ml-2; lower molars distinct from those taxa in forkian equivalent; Gingerich and Rose, 1977) immigration of the
having paracristid reduced, labially attached cristid obli- family between continents. Dispersal from Asia to North America,
qua, and pillarlike ectocingulid. Distinct from advanced rather than vice versa, is more likely (Cifelli, Schaff, and McKenna,
Asian genera such as Palaeostylops and Gashatostylops 1989).
in lacking a p4 talonid shearing notch and in having a The biochronology of Asiatic Arctostylopidae is insufficient to
better differentiated canine and paracone folds on evaluate morphological trends through time within the family. Pro-
P3-4. gressive taxa (Arctostylops and more derived genera) are charac-
Range and mean ofm2 length: 3.4-4.7; 4.1 mm. terized by strongly developed, anteroposteriorly oriented, vertical
Included species: A. steini only (known from localities CPI 3F, shearing crests on the cheek teeth (Figure 21.2). These are formed
CPI 7B, CP18B, ?CP20A, CP26C, and, possibly, CP62C). by the ectoloph and cristid obliqua of upper and lower teeth, re-
spectively, which are hypertrophied in Eocene Anatolostylops and
Comments: All genera but one are Asiatic in occurrence. an undescribed genus. The form and function of these shearing
These genera, not directly relevant in the present con- crests in advanced arctostylopids are analogous to those of car-
text, are Asiostylops Zheng, 1979 (late Paleocene); Sinos- nassial blades. Upper and lower crests are curved in two planes,
tylops Tang and Yan, 197 6 (late Paleocene); Bothriostylops so that food material is trapped and sheared between them by ver-
Zheng and Huang, 1986 (late Paleocene); Palaeostylops tical and transverse jaw movement (Crompton and Sita-Lumsden,
Matthew and Granger, 1925 (late Paleocene); Gashatosty- 1970). Emphasis on these primary shearing crests (ectoloph, cristid
lops Cifelli, Schaff, and McKenna, 1989 (late Paleocene); obliqua) in advanced arctostylopids resulted in two notable aspects
Anatolostylops Zheng, 1979 (early or middle Eocene); of occlusion. First, the upper and lower molars, primitively in dou-
Kazachostylops Nessov, 1987 (late Paleocene); and an un- ble opposition, became vertically aligned and thtis opposed each
named genus (Cifelli, Schaff, and McKenna, 1989). Sub- other singly. Second, "Phase II" jaw movement and occlusion, as-
familial groups within the family have been proposed by sociated with shredding and grinding in primitive mammals (Kay
Zheng (1979) and Nessov (1987); we follow our earlier re- and Hiiemae, 1974), was negligible, with a corresponding reduc-
vision of the group (Cifelli, Schaff, and McKenna, 1989) tion of lingual cusps on upper molars. Unlike primates (Kay and
in refraining from formally subdividing the family until its Hiiemae, 1974), similar relationships among living ungulates are
constituents are better known. indicative of a shift to dependence on more fibrous vegetation (Janis,
The stratigraphic and geographic distribution of Arc- 1979).
tostylops steini is late Tiffanian (Tiffanian 5) through Clark- Little is known of arctostylopid cranial morphology, but a par-
forkian, central Rocky Mountain region, United States. tial skull of Gashatostylops macrodon indicates that in this derived
Although Matthew (1915) initially reported the species genus, at least, the rostrum was short, flaring broadly posteriad
A. steini from the "lower Gray Bull beds," this early (Figure 21.2). The zygomatic arch was deep and bore a small de-
Wasatchian record has not been confirmed and is likely to scending process, indicative of powerfully developed masseteric
been have incorrect, probably representing a Clarkforkian musculature. The infraorbital foramen was relatively small, sug-
occurrence (cf. Rose, 1981). gesting that vibrissae may have been poorly developed (Kay and
 Richard L. Cifelli and Charles R. Schaff 336

Cartmill, 1977). Beyond the observation that these animals were, Matthew, W. D., Granger, W., & Simpson, G. G. (1929). Additions to the
for their time, relatively specialized herbivores, little more can be fauna of the Gashato Formation of Mongolia. American Museum
Novitates, 376, 1-12.
said regarding their mode of life. Nessov, L.A. ( 1987). Results of searches and investigations in the mammal-
bearing Cretaceous and early Paleogene in the territory of the USSR.
Annual of the All-Union Paleontological Society, 30, 199-218 [in
REFERENCES Russian].
Rose, K. D. (1981). The Clarkforkian Land-Mammal Age and mammalian
Cifelli, R. L. ( 1983). The origin and affinities of the South American Condy- fauna! composition across the Paleocene-Eocene boundary. Univer-
larthra and early Tertiary Litopterna (Mammalia). American Museum sity of Michigan Papers in Paleontology, 26, 1-197.
Novitates, 2772, 1--49. Schaff, C. R. (1985). Paleocene mammals from the Beartooth Region of
Cifelli, R. L., Schaff, C. R., & McKenna, M. C. (1989). The relationships Wyoming and Montana. National Geographic Research Reports,
of the Arctostylopidae (Mammalia): new data and interpretation. 1979, 589-95.
Bulletin of the Museum of Comparative Zoology, 152, 1--44. Schlosser, M. (1923). Saugetiere. In Grundziige der Palaontologie,
Crompton, A. W., & Sita-Lumsden, A. (1970). Functional significance of neubeart., eds. F. Broili & M. Schlosser, pp. 402-689. Munich:
the therian molar pattern. Nature, 227, 197-9. R. Oldenbourg.
Dashzeveg, D. (1982). La faune de Mammireres du Paleogene inferieur Simpson, G. G. (1929). Paleocene and lower Eocene mammals of Europe.
de Naran Bulak (Asie centrale) et ses correlations avec !'Europe et American Museum Novitates, 354, 1-17.
l' Amerique du Nord. Bulletin de la Societe Geologique de France, (1934). Provisional classification of extinct South American hoofed
24, 275-81. mammals. American Museum Novitates, 750, 1-21.
Gaudry, A. (1906). Fossiles de Patagonie. Etude sur un portion du monde (1948). The beginning of the Age of Mammals in South America, Part 1.
antartique. Annales de Paleontologie (Paris), 1, 1--42. Bulletin of the American Museum of Natural History, 91, 1-232.
Gingerich, P. D. ( 1985). South American mammals in the Paleocene of North (1967). The beginning of the Age of Mammals in South America,
America. In The Great American Biotic Interchange, eds. F. G. Stehli Part 2. Bulletin of the American Museum of Natural History, 137,
& S. D. Webb, pp. 123-37. New York: Plenum. 1-259.
Gingerich, P. D., & Rose, K. D. (1977). Preliminary report on the American Szalay, F. S., & McKenna, M. C. (1971). Beginning of the Age of Mammals
Clark Fork mammalian fauna, and its correlation with similar faunas in Asia: the late Paleocene Gashato fauna, Mongolia. Bulletin of the
in Europe and Asia. Geobios Memoir Special, 1, 39--45. American Museum of Natural History, 144, 269-318.
Janis, C. M. (1979). Mastication in the hyrax and its relevance to ungulate Tang Ying-jun, & Yan De-fa. (1976). Notes on some mammalian fossils
dental evolution. Paleobiology, 5, 50--9. from the Paleocene of Qianshan and Xuanchen, Anhui. Vertebrata
Jepsen, G. L., & Woodburne, M. D. (1969). Paleocene hyracothere from Palasiatica, 14, 91-9.
Polecat Bench Formation, Wyoming. Science, 164, 543-7. Van Frank, R. (1957). A fossil collection from northern Venezuela, 1: Tox-
Kay, R. F., & Cartmill, M. C. (1977). Cranial morphology and adaptations odontidae (Mammalia, Notoungulata). American Museum Novitates,
of Palaechthon nacimienti and other Paromomyidae (Plesiadapoidea, 1850, 1-38.
?Primates), with a description of a new genus and species. Journal Webb, S. D., & Perrigo, S. C. (1984). Late Cenozoic vertebrates from
of Human Evolution, 6, 19-53. Honduras and El Salvador. Journal of Vertebrate Paleontology, 4,
Kay, R. F., & Hiiemae, K. (1974). Jaw movement and tooth use in recent 237-54.
and fossil primates. American Journal of Physical Anthropology, 40, Zhai Ren-jie. (1978). More fossil evidences favouring an early Eocene con-
227-56. nection between Asia and Neoarctic. Memoirs of the Institute of
Marshall, L. G., Butler, R. F., Drake, R. E., Curtis, G. H., & Tedford, R.H. Vertebrate Paleontology and Paleoanthropology, Academica Sinica,
(1979). Calibration of the great American interchange. Science, 204, 13, 107-15.
272-9. Zheng Jia-jian. (1979). The Paleocene notoungulates of Jiang-xi. In A Sym-
Matthew, W. D. (1915). A revision of the lower Eocene Wasatch and Wind posium on Cretaceous and Early Tertiary Red Beds of South China,
River faunas, Part IV: Entelonychia, Primates, Insectivora (part). Bul- pp. 387-94. Beijing: Science Press, Academica Sinica.
letin of the American Museum of Natural History, 34, 429-83. Zheng Jia-jinan, & Huang Xueshi. (1986). New arctostylopids (Notoun-
Matthew, W. D., & Granger, W. (1925). Fauna and correlation of the Gashato gulata, Mammalia) from the late Paleocene of Jiangxi. Vertebrata
Formation of Mongolia. American Museum Novitates, 189, 1-12. Palasiatica, 24, 121-8.
 Part IV: Artiodactyla

22 Artiodactyla

CHRISTINE M. JANIS, JAMES A. EFFINGER, JESSICA A. HARRISON, JAMES G. HONEY, DONALD

G. KRON, BRUCE LANDER, EARL MANNING, DONALD R. PROTHERO, MARGARET S. STEVENS,
RICHARD K. STUCKY, S. DAVID WEBB, and DAVID B. WRIGHT

INTRODUCTION caries) may represent an endemic North American radiation. Al-

though "tayassuids" are reported in Old World sites ranging from
Artiodactyls, especially ruminant artiodactyls, are commonly con- late Eocene to early Pliocene (Savage and Russell, 1983), these taxa
sidered as the evolutionary success story among ungulates. As well may not belong in the family Tayassuidae (Wright, this volume,
as their present-day species diversity and biogeographic range, they Chapter 26, but see also Pickford, 1993). Among a diversity of mid-
have also played an important role in human evolution and civiliza- Tertiary suiform families of Old World origin (e.g., Helohyinae and
tion. As noted by Prothero (1994), most of our domestic animals Cebochoeridae), only the Entelodontidae and Anthracotheriidae are
(if not our household pets), which provide our milk, our wool, and known from North America. In the case of the Anthracotheriidae, at
most of our meat, are ruminants. Moreover, it should be appreciated least, both the generic diversity and the range (late Eocene through
that the majority of species that make up the present-day artiodactyl early Miocene) is considerably less in North America than in the
diversity is the result of the late Neogene speciation of African Old World. The family Suidae, the surviving diverse suiform fam-
bovids. Texts prominently feature bovids as examples of ungulate ily in the later Cenozoic of the Old World, is unknown in the New
evolution, yet Qther groups, which were equally diverse in the Ter- World except as the result of human introduction.
tiary but are oflimited diversity today, such as came lids, are ignored Although camelids are the only surviving tylopods, there was
(see Prothero, 1994). an extensive late Eocene radiation of families in this suborder (al-
There has been a long held evolutionary truism, commencing though see later for a discussion of the problems with the "sub-
with Kowalewsky (1883) and upheld in both technical and semipop- order Tylopoda"). Tylopods were either exclusively Old World in
ular works on evolution for most of this century (e.g., Simpson, their distribution (e.g., Xiphodontidae, Anoplotheriidae, Cainotheri-
1953), that ruminant artiodactyls were competitively superior to idae) or exclusively North American (oreodonts [families Agrio-
perissodactyls because of their foregut mode of cellulose diges- choeridae and Merycoidodontidae], Oromerycidae, Protoceratidae,
tion. This truism was used to explain the explosive radiation of and Camelidae [excluding the Plio-Pleistocene dispersal of derived
artiodactyls during the later Tertiary at the apparent expense of the camelids to the Old World and to South America]). It is commonly
perissodactyls. Yet the actual patterns of artiodactyl radiation and agreed that the majority of North American tylopods, first appearing
perissodactyl decline do not support a model of competitive in- in the Uintan (late middle Eocene), represent an immigration event
teraction (Cifelli, 1982). The diversification of ruminants, which from Asia, although the picture is not so clear for the oreodonts,
must feed more selectively than hindgut fermenters such as peris- with a possible earlier first appearance in the Bridgerian (see Janis,
sodactyls, may have resulted from late Tertiary climatic and veg- 1993, and references therein).
etational changes (Janis, 1989). Additionally, the "success story" Traguloid ruminants are well represented in the late Eocene
bovids have always been a primarily Old World group, and in the through early Miocene of North America by the families Hyper-
late Miocene their diversification was paralleled by equids in North tragulidae and Leptomerycidae, the latter also having extensive
America (see discussion in Janis, Gordon, and Illius, 1994). representation in the Old World. However, the North American
The pattern of artiodactyl evolution in North America contrasts Miocene diversity of pecoran ruminants represents an independent
with that seen in the Old World. North American Tertiary artio- endemic radiation, again from immigrations from originally Old
dactyls represent an eclectic mix of endemics that are not found World stocks. The hornless blastomerycines are probably related
in the Old World (although they may be of ultimate Old World to Moschus, the living musk deer (not a true cervid; see discus-
ancestry) and occasional immigrations of taxa with a primarily sion later) (Webb and Taylor, 1980). Nevertheless, blastomerycines
Old World distribution. Among suiformes, the Tayassuidae (pee- themselves are uniquely North American. Dromomerycids (which

337
 Christine M. Janis et. al 338

may or may not be separate at the family level from Old World HISTORY OF IDEAS OF
palaeomerycids) are also apparently uniquely North American, al- ARTIODACTYL INTERRELATIONSHIPS
though some Old World representatives have been proposed (see A fundamental division within artiodactyls has long been perceived.
discussion in Janis and Manning, this volume, Chapter 32). Antilo- Based on living taxa, there is a clear split between the more omnivo-
caprids, the only members of this Miocene radiation to survive to rous, bunodont, four-toed, simple-stomached piglike forms (suines)
the present, lack any clearly defined sister-group affinities with any like the Non-Ruminantia or Bunodontia, and the more folivorous,
Old World Tertiary taxon (see discussion in Janis and Scott, 1987). selenodont, usually two-toed, forms that have a ruminant type of
The surviving pecoran families that were predominant in the Old forestomach with several chambers (camels and ruminants) like the
World Tertiary played little or no role in the ruminant radiations Ruminantia or Selenodontia (e.g., Flower, 1883; Matthew, 1929).
of the North American Tertiary. Giraffids are unknown from the This grouping broadly corresponds to the now discredited notion of
New World. Despite a Pleistocene to Recent radiation and diversi- "amastoid" and "mastoid" artiodactyls, respectively (e.g., Romer,
fication in both North and South America, cervids did not appear 1966). However, division between these categories is not so clear
in North America until the Pliocene. North American bovids are cut with the inclusion of fossil groups. Certain fossil taxa, such as
a purely Pleistocene to Recent phenomenon, with the exception of anthracotheres, were placed by some workers in a third grouping:
the enigmatic late Miocene to early Pliocene genus Neotragoceras.
Bunoselenodontia (e.g., Zittel, 1925).
However, the horn cores of this taxon also resemble those of the Matthew (1929) recognized five primary divisions to encompass
Old World "homed" cervoid, Hoplitomeryx Leinders (1983). It is seventeen living and fossil families (although he did not attempt
the opinion of the senior author, at least, that the "bovid" affinities to show how these divisions were interrelated): The Palaeodonta
of Neotragoceras are far from conclusive. comprised the Dichobunidae and the Entelodontidae; the Hyodonta
comprised the Suidae, Tayassuidae, and Hippopotamidae; the An-
codonta comprised the Anthracotheriidae, Anoplotheriidae, Cain-
SYSTEMATICS otheriidae, and Merycoidodontidae (= Oreodontoidea as used in this
volume); the Tylopoda comprised the Xiphodontidae and Cameli-
The Order Artiodactyla was formally named by Owen (1848), who dae; and the Pecora comprised the Amphimerycidae, Tragulidae
recognized the uniqueness of the paraxonic foot symmetry. The or- (including the protoceratids), Cervidae, Giraffidae, Bovidae, and
der can be defined by a number of additional synapomorphies: lower
Antilocapridae.
molar trigonids very narrow, with paraconid and metaconid closely Simpson (1945) reduced this classification to three suborders:
appressed; formation of a partial double mesocylix in the distal Suiformes (including Matthew's Palaeodonta, Hyodonta, and An-
deciduous premolars (except for P4); enlarged orbitosphenoid, sep- codonta), Tylopoda, and Ruminantia (equivalent to Matthew's
arating the frontal from the alisphenoid; "double pulley" type of as- Pecora). The essence of this classification is retained by most work-
tragalus in which both proximal and distal ends are trochleated; and ers today (e.g., Romer, 1966; Carroll, 1987), and is reflected in
symmetrical reduction of the second and fifth metapodials around Figure 22.1 in the inclusion of the living families and the extinct
the enlarged third and fourth metapodials (Schaeffer, 1948; Gentry families with North American representatives. Figure 22.l reflects
and Hooker, 1988; Prothero, Manning, and Fischer, 1988). a synthesis of current information of the higher interrelationships
A historical account of the ideas behind the various classifica- of the Artiodactyla: It seems fairly certain that the Ruminantia is
tory schemes of the artiodactyls is discussed in Gentry and Hooker monophyletic, as is an association ofRuminantia and Tylopoda, but
( 1988), and the our summary here is largely derived from their con- both the Tylopoda and Suiformes are probably paraphyletic (contra
tribution. Note that, as already described, there are a large number
Gentry and Hooker, 1988).
of extinct Old World families whose taxonomic affinities are un-
certain and controversial. Because this volume is designed to deal
with North American taxa, the controversial Old World taxa are PRIMITIVE ARTIODACTYLA
largely ignored. Much attention has been paid in recent years to the
interrelationships of the Ruminantia (e.g., Webb and Taylor, 1980; The concept of three artiodactyl suborders does not usually in-
Janis and Scott, 1987; Miyamoto et al., 1993; Scott and Janis, 1993), clude the early primitive, bunodont members lumped together as the
but with the exception of Gentry and Hooker (1988) there has been Palaeodonta or Dichobunidae (excluding entelodonts from the con-
no attempt at a cladistic analysis of nonruminant artiodactyls. The cept of the Palaeodonta as originally expressed by Matthew, 1929).
conclusions of Gentry and Hooker (1988) on suiform and tylopod For example, Romer (1966) depicts a basal "Diacodectidae" giving
interrelationships are at variance with many of the ideas expressed rise not only to the three suborders but also to a paraphyletic as-
by the individual authors in this volume: The problems with Gentry semblage of palaeodonts, including the families Achaenodontidae,
and Hooker's (1988) proposed interrelationships of the ruminants Leptochoeridae, Homacodontidae, and Dichobunidae.
have been addressed elsewhere (Scott and Janis, 1993; see also dis- Gazin (1955) depicts a basal "bushy" assemblage of North Amer-
cussion later). It is clear that the interrelationships of nonruminant ican primitive artiodactyls ("Dichobunids") giving rise to more de-
(in a cladistic sense) artiodactyls are in need of an extensive revision rived North American artiodactyls in the following fashion: A rather
that this chapter is not designed to address. loose assemblage of selenodonts (excluding hypertragulids, which
 Artiodactyla 339

BUNODONTIA or
SUI FORM ES SELENODONTIA

NEOSELENODONTIA

SUINA TYLOPODA RUMINANTIA

SUOIDEA "TRAGULINA" PECORA

CERVOIDEA

Figure 22.1. Interrelationships among artiodactyls. * indicates taxon not found in North America.

he links with the Homacodontinae) appear to be derived from the The homacodonts are a paraphyletic assemblage of more selenodont
vicinity of Antiacodon; entelodontids are derived from the forms, with some genera perhaps more closely related to the Neose-
Helohyinae; leptochoerids are derived from the Diacodexinae; the lenodontia and others to the Tylopoda.
tayassuids appear to be derived from nowhere, perhaps somewhere
in between the Homacodontinae and the Helohyinae.
SUIFORMES
The analysis of Gentry and Hooker (1988) dispenses with any
notion of basal artiodactyl paraphyly, but renders the "Palaeodonta" Suids and tayassuids have long been recognized as sister taxa,
paraphyletic by virtue of placing them within the subordinal group- grouped together in the Suoidea. Although hippos are recognized
ings. As far as the North American representatives are concerned, as the closest living relative to the suoids, and sometimes included
their preferred cladogram (p. 257) shows the following pattern: Dia- within the Suoidea (usually as the sister taxon to suids plus tayas-
codexeidae (with the exception of Diacodexis pakistanensis, which suids, but see later), problems arise when the fossil taxa are included.
they place as the basal taxon within the Tylopoda) are linked with Matthew (1929) recognized an association of hippos with suoids
Leptochoeridae and are placed together in an unresolved trichotomy (as the Hyodonta), but placed anthracotheres in the Ancodonta, on
with Tylopoda and Ruminantia; "Bunomerycidae" (presumably = the basis of their more selenodont cheek teeth. Colbert (1935) pro-
Homacodontidae) is placed as the next basal taxon of the Tylopoda; posed that anthracotheres were ancestral to hippos, including the two
and Helohyidae and Antiacodontidae are placed as basal taxa within groups in the Anthracotheroidea or Hippopotamoidea, a view almost
the Bunodontia (= Suiformes). universally followed for much of the twentieth century (e.g., Simp-
Stucky (this volume, Chapter 23) recognizes only leptochoerids, son, 1945; Romer, 1966; Carroll, 1987). Pickford (1983) suggested
antiacodontids, and (probably) helohyids as monophyletic groups. that hippos were descended from tayassuids, rather than from anthra-
Diacodexis and related taxa are seen as a basal paraphyletic as- cotheres, but Wright (this volume, Chapter 26) doubts the tayassuid
semblage, with the genus Diacodexis itself also being paraphyletic. affinities of any Eurasian suoid (but see also Pickford, 1993).
 Christine M. Janis et. al 340

Whereas the cladistic analysis of Gentry and Hooker ( 1988) links ferred position was in the Ruminantia with the hypertragulids, or
anthracotheres and hippos as separate within the suines from suids as traguloids of some sort (Colbert, 1941; Stirton, 1944). This view
and tayassuids, Kron and Manning (this volume, Chapter 25) note was upheld by most later workers (e.g., Romer, 1966) and is still
that the derived nature of the anthracothere P4 removes them from often portrayed in popular texts today (see Prothero, 1994), despite
the traditional relationship with hippos, leaving them as the sister the fact that Stirton (1967) himselflater revised his previous views
taxon to the suoids plus hippos (the "expanded Suoidea" of Kron and linked protoceratids with tylopods.
and Manning). Patton and Taylor (1973) extensively revised the protoceratids
Although there may be some debate about the exact taxonomic and showed they had to be excluded from the Ruminantia on the
placement of the anthracotheres, made even more complex if other basis of a number of primitive features, including the lack of cubo-
Old World fossil families are included, it is still clear that they navicular fusion. However, many of the features they used to unite
are allied with living suiforms. The taxonomic position of the en- them with camelids in the Tylopoda are symplesiomorphies (e.g.,
telodonts is much more problematical. Matthew (1929) placed en- unfused metacarpals, lack of complete distal metapodial keels).
telodonts outside of other suiforms in the Palaeodonta, following Webb and Taylor (1980) noted that a distinct synapmorphy could
Stehlin (1910) in recognizing the primitive nature of the entelodont be used to unite camelids and protoceratids, that of the passage of
molars where the hypocone has not been replaced by the meta- the vertebrarterial canal through the neural arch pedicels rather than
conule. However, Scott (1940) criticized Matthew's grouping of through the transverse processes of the cervical vertebrae, as orig-
entelodonts and dichobunids (the "Hypoconifera") as a so-called inally noted for camelids by Flower (1885). Note, however, that a
wastebasket association. Romer (1966) grouped the entelodonts recent publication (Joeckel and Stavas, 1996) holds that the basi-
within the Suina, although with an apparently separate ancestry cranial evidence linking protoceratids and came!ids may not be as
from the "basal Diacodectidae" to the other suines, in a superfamily robust as previously thought.
Entelodontoidea that also included the cebochoerids and the cho- Webb and Taylor (1980) restricted the Tylopoda to camelids
eropotamids. This convention was also followed by Carroll (1987), (including oromerycids), protoceratids, xiphodontids, and amphi-
who included the leptochoerids within this grouping. The analy- merycids, and linked this group as the sister taxon to the Rumi-
sis of Gentry and Hooker (1988) places entelodonts as the sister nantia. They termed this combined clade (restricted Tylopoda plus
taxon to the suids plus tayassuids. However, Effinger (this volume, Ruminantia) the Neoselenodontia, grouped on the reduction or loss
Chapter 24) and Kron and Manning (this volume, Chapter25) are of of the upper incisors, the fusion of the ectocuneiform with the
the opinion that entelodonts are more primitive than other suiformes. mesocuneiform, the lack of a molar paraconule, and the presence
Figure 22.1 reflects our current lack of knowledge of the phyloge- of a ruminating stomach with at least three chambers (inferred
netic position of the entelodontids, including the possibility that from the condition in living camelids in the case of the other
they might be more primitive than all other artiodactyls (following tylopods).
Stehlin, 1910), thus rendering the Suiformes paraphyletic. Since the original creation of the Neoselenodontia, Gentry and
Hooker (1988) have reassigned amphimerycids to the Ruminantia,
based on the possession of a fused cuboid and navicular, and oromery-
TYLOPODA cids once again have been removed from the camelids proper as
Matthew (1929) constructed the Tylopoda to include the their sister taxon (see Prothero, this volume, Chapter 28). This clade
Camelidae and the Xiphodontidae: He grouped other nonruminant of North American tylopods (camelids, oromerycids, and protocer-
(nonpecoran, in his terms) taxa with selenodont cheek teeth in the atids ), with xiphodontids as a sister tax on, was also upheld by Gentry
Ancodonta, a grouping he himself recognized as probably unnatural. and Hooker (1988).
Scott (1940) included other selenodonts, such as oreodonts, proto- However, despite the fact that Gentry and Hooker (1988) also
ceratids, and hypertragulids, in the Tylopoda. Matthew's restricted upheld the monophyly of the Tylopoda as a whole (in the expanded
notion of the Tylopoda was followed by Colbert ( 1941) and Simp- sense of Romer, 1966), placing the oreodonts as one of the more
son (1945) (although note that Simpson placed oreodonts within primitive branches, it is not clear that oreodonts should necessar-
the Suiformes). However, later workers (e.g., Romer, 1966; Carroll, ily be united with the more restricted notion of tylopods (as also
1987) employed Scott's (1940) usage of the Tylopoda to group all inferred by Webb and Taylor, 1980). Both Matthew (1929) and
selenodont artiodactyls that are not clearly members of the Rumi- Simpson (1945) included oreodonts as suiforms of some descrip-
nantia (although they did not follow his tylopod assignations of the tion. As mentioned earlier, oreodonts are a problem not only because
Hypertragulidae ). Thus our current notion of the Tylopoda itself is of of the lack of clear synapomorphies linking them with the camelid
a paraphyletic assemblage. The picture is confounded by a consider- clustering, but also because of their very early appearance in North
ation of the taxa classed as tylopods: Camelids are the only surviving America, predating the immigration of other tylopods, and their
members, and the other families are either exclusively Eurasian or probable endemic origin from homacodontid "dichobunids" (see
exclusively North American, making comparisons difficult. Lander, this volume, Chapter 27). Figure 22.1 depicts the currently
A further complication has been the placement within the unresolved position of the oreodonts, suggesting a paraphyletic na-
Artiodactyla of the protoceratids. Although they had been assigned ture of the term Tylopoda as used by Romer (1966) and Gentry·and
to the Tylopoda by some early workers (e.g., Scott, 1940) a pre- Hooker (1988).
 Artiodactyla 341

RUMINANTIA tax on to the Pecora plus Leptomeryx (but see later discussion on the
cladogram presented by Gentry and Hooker).
Under current usage, the term Ruminantia is equivalent to Matthew's The major families ofliving pecorans, Cervidae, Bovidae, and Gi-
(1929) "division Pecora." Flower (1883) had divided artiodactyls raffidae, have been considered to be interrelated in a variety of ways.
into four suborders, recognizing the "Tragulina" (tragulids and their Cervoidea (including palaeomerycids), Giraffoidea, and Bovoidea
fossil relatives) as separate from the higher (homed) ruminants. He (including antilocaprids) were accorded separate status by Simpson
restricted the term Pecora to the higher ruminants, grouping together (1945), but he mentions a close association between cervoids and
Tragulina and Pecora as the Ruminantia, a terminology adopted giraffoids. Cervids have long been considered as primitive peco-
by Colbert (1941) and Simpson (1945) and recently upheld and rans, based largely on primitive characters such as brachydont cheek
advocated by Webb and Taylor (1980). However, Scott's (1940) use teeth and retention of the lateral digits, and also on the mistaken
of the term "Pecora" to include all members of the Ruminantia, assumption that the Palaeomeryx fold on the lower molars was a
where he uses the term "Ruminantia" to group his "Pecora" with primitive ruminant character (see Janis and Scott, 1987, p. 31 ). How-
the Tylopoda, has been upheld as recently as Romer (1966). ever, Leinders (1979) and Leinders and Heintz (1980) have shown
Stirton ( 1944) divided the "suborder Pecora" into three superfam- that cervids possess derived features such as a double lacrimal orifice
ilies: Traguloidea (= Tragulina of Flower [1883], including Hyper- and a closed metatarsal gully. No morphological synapomorphies
tragulidae, Protoceratidae, Tragulidae, and Gelocidae); Cervoidea unite cervids with giraffids, although such a grouping might be sup-
(Cervidae [including blastomerycines], Giraffidae, and Palaeo- ported by molecular evidence (see Miyamoto et al., 1993). Various
merycidae [including Dromomerycidae]); and Bovoidea (Antilo- authors have grouped giraffids with bovids in the Bovoidea (e.g.,
capridae and Bovidae). However, Stirton (1944) recognized that Hamilton, 1978; Bubenik, 1982; Ginsburg, 1985; Gentry, 1994),
Cervoidea plus Bovidae formed arank of equal status to Traguloidea, based on the observation that some living bovids possess a similar
as the Pecora of current usage (see also Matthew, 1934). The trag- os comu in their horn development to the condition in living giraf-
uloids were mainly defined on primitive characters (e.g., unfused fids (but see discussion in Janis and Scott, pp. 10-12, where they
metapodials, trenchant premolars). Cervoids were defined as hav- uphold that this supposed homology is mistaken).
ing solid horns (in contrast to the hollow horns of bovoids), but Figure 22.1 shows an unresolved trichotomy among cervids,
were otherwise grouped together on primitive characters, such as bovids, and giraffids. Janis and Scott (1987) favor a sister-group
the possession of brachydont cheek teeth. Bovoids were united by relationship between bovids and cervids, but note that this group-
the presence of hypsodont cheek teeth. ing is not supported by strong characters (see also Scott and Janis,
An account behind the concepts of the Tragulina is reviewed by 1993). Note also that molecular evidence does not strongly favor a
Webb and Taylor (1980). They showed that hypertragulids are the monophyletic Bovidae (see Miyamoto et al., 1993).
most primitive ruminants (they excluded the amphimerycids from The position of the Antilocapridae is more controversial. The
the Ruminanti11), and upheld the separation of the Leptomeryci- traditional grouping of antilocaprids with bovids in the Bovoidea
dae from the Hypertragulidae (see also Gazin, 1955; Taylor and is based either on probable homoplasies (e.g., loss of side toes,
Webb, 1976), placing them as more derived than the Tragulidae hypsodont cheek teeth) (e.g., Matthew, 1904; Pilgrim, 1941; Simp-
(see Figure 22.1). They also argued for the pecoran status of the son, 1945), on symplesiomorphies (retention of the gallbladder), or
genus Gelocus, previously considered a traguloid, on the basis of on the purported synapmorphy of horns with akeratinous cover (e.g.,
derived features of the postcranial skeleton such as a parallel-sided O'Gara and Matson, 1975). However, the similarity of horn exfolia-
astragalus, and discussed the features that unite the Pecora as a tion in bovids and antilocaprids may be more apparent than real, and
more derived clade within the Ruminantia. The late Tertiary Pseu- the cranial appendages of the more primitive merycodontine antilo-
doceras, commonly considered to be a late surviving "gelocid" (see caprids cannot so easily be homologized with those of bovids (see
Webb, this volume Chapter 31), is the only North American repre- discussion in Janis and Scott, 1987, pp. 13, 20). The derived cervid
sentative of this group. However, there is controversy as to whether morphological features discussed by Leinders (1979) and Leinders
the taxa commonly united in the "Gelocidae" represent a discrete and Heintz (1980) are also found in the living pronghorn, Antilo-
family, comprising a sister group to the other pecorans (Webb and capra. Janis and Scott ( 1987) supported the views of these authors
Taylor, 1980), or represent a paraphyletic assemblage of derived in linking antilocaprids with cervids in the Cervoidea and noted that
traguloids and basal pecorans (Janis, 1987). In Figure 22.1 Pseu- these features can also be found in primitive merycodontines.
doceras is placed as a basal pecoran taxon, thus sidestepping the Moschids are another problematical group. The living genus
issue. Moschus (the musk deer) is usually treated as a cervid (e.g., Nowak,
Note that Gentry and Hooker (1988) show a different phylogeny 1991): However, numerous features of the morphology, includ-
of the Tragulina. They retain hypertragulids as the most primitive ing features of the soft anatomy noted in the last century, exclude
family, placing them as more primitive than the amphimerycids. The Moschus from the Cervidae (see Groves and Grubb, 1987; Janis and
Old World taxonArchaeomer yx is removed from the Leptomeryci- Scott, 1987), and Gray (1821) erected the separate family Moschi-
dae (contra Webb and Taylor, 1980) and placed as the sister group dae for this taxon. Leinders (1979) and Leinders and Heintz (1980)
to all other ruminants. Tragulids form the next most derived clade, discuss the fact that Moschus possesses the derived cervoid feature
with Gelocus stripped of its pecoran status and placed as the sister of a closed metatarsal gully, but lacks the double lacrimal orifice,
 Christine M. Janis et. al 342

placing the taxon within the Cervoidea but as less derived than the EVOLUTIONARY AND
Antilocapridae (a view also followed by Groves and Grubb, 1987, BIOGEOGRAPHICAL PATTERNS
and Janis and Scott, 1987, as shown in Figure 22.1).
Webb and Taylor (1980) included a number of extinct horn-
EOCENE
less ruminants with Moschus in the family Moschidae, including
European taxa such as Dremotherium, and the North American blas- Artiodactyls made their first appearance at the start of the Eocene,
tomerycines, previously considered to be cervoid palaeomerycids with the appearance of Diacodexis in the earliest Wasatchian. There
(e.g., Stirton, 1944; Simpson, 1945) orcervids (e.g., Matthew, 1908; was modest diversity of primitive artiodactyls such as antiacodonts
Ginsburg and Heintz, 1966), and Webb and Taylor (1980) renewed and "homacodonts" in the early Eocene. Some change was apparent
usage of this family. Note that Webb and Taylor (1980) consider in the early middle Eocene: By the early Bridgerian, Diacodexis
moschids to be primitive pecorans: They unite them with the ge- and related taxa were on the decline (making their probable last
locids in their "Division Moschina," grouping other pecorans in the appearance in this time interval), and helohyines and leptochoerines
"Division Eupecora" on the basis of the possession of cranial ap- were now in evidence (see Figure 22.2). Agriochoerid oreodonts
pendages: However, Janis and Scott (1987) argue strongly for the may also be known from as early as the middle Bridgerian. There
nonhomology of the various cranial appendages (horns, antlers, os- are no obvious biogeographical patterns during the earlier part of
sicones, etc.) among the Pecora. Note that Moschus has not been the Eocene, although "homacodonts" are virtually unknown outside
included to date in any molecular studies on ruminant relationships. of the central Great Plains.
The final North American ruminant group under consideration is Agriochoerids were definitely established by the early Uintan,
the dromomerycids. Dromomerycids have been variously grouped a time interval when a number of other taxa made their first ap-
with cervids or giraffids: The history of ideas about their pearance: the suiform Brachyhyops and the earliest members of
phylogenetic affinities and the reason for placing them as cervoids the oromerycids, protoceratids, and camelids. Among the primitive
related to the Old World palaeomerycids (see Figure 22.1) is cov- artiodactyls, the antiacodontids declined while the "homacodonts"
ered extensively by Janis and Manning (this volume, Chapter 32) diversified. The change to a later middle Eocene artiodactyl as-
and thus not repeated here. semblage became more profound in the late Uintan, with the first
The phylogeny of the Ruminantia depicted in Figure 22.1, largely appearance of hypertragulids, leptomerycids, and anthracotheriids,
based on Janis and Scott (1987), is the phylogeny that has been and an even greater diversification of the "homacodonts." In the
adopted by many subsequent authors (e.g., Prothero, 1994). Note, Duchesneau, agriochoerids continued to diversify, merycoidodon-
however, that other schemes of interrelationship have been subse- tids and entelodontids made their first appearance, and all primitive
quently proposed. For example, in addition to the alternative "dichobunid" artiodactyl families, with the exception of the lepto-
schemes just discussed, Ahearn (1994) groups together antilocaprids choerines, appeared for the last time (see Figures 22.2, 22.3, 22.4).
and dromomerycids, which together form the sister group to the gi- Several late middle Eocene biogeographic patterns are discerni-
raffids. Gentry and Hooker (1988) support a sister-group relation- ble. The California Coast region contained only the antiacodontid
ship between antilocaprids and bovids, with giraffids as the sister Tapochoerus, the agriochoerid Protoreodon, and the hypertragulid
tax on to this grouping (forming the classical notion of the Bovoidea). Simimeryx (although the latter two tax a were also known from else-
Cervids (excluding the antlerless Chinese water deer, Hydropotes, where); and the earliest entelodontids had an exclusively southern
which forms the sister taxon to the bovoids) are the most primitive distribution.
of the living pecorans. A paraphyletic Moschidae places Dremoth-
erium and Moschus between cervids and Hydropotes, and dromo-
WHITE RIVER CHRONOFAUNA
merycids and Palaeomeryx cluster together as the sister group to the
other pecorans. The early Chadronian marked the inception of the "White River
The cladogram of the Ruminantia presented by Gentry and chronofauna," which lasted through the Oligocene (see summary
Hooker (1988) was criticized by Scott and Janis (1990) on the basis chapter on carnivorous mammals, Chapter 4, for discussion and
of including characters likely to be homoplastic (e.g., hypsodonty, nomenclature of chronofaunas). There do not appear to be any pro-
loss of side toes, possession of cranial appendages). Scott and Janis nounced biogeographical variations during the White River chrono-
(1990) demonstrated that, by the inclusion of these homoplastic fauna.
characters in their own character matrix, they generated a cladogram Among the primitive artiodactyls, leptochoerids retained a mod-
very unlike the one in Figure 22.1, and with a strong resemblance to erate diversity until the Arikareean, when only Leptochoerus re-
that of Gentry and Hooker. They concluded that character analysis mained. Among the suiforms, tayassuids first appeared in the
is important prior to cladistic analysis in order to avoid instances of Chadronian, and maintained, together with anthracotheriids and en-
"Paup Fiction." However, as Kraus and Miyamoto (1991) point out telodontids, a moderate diversity through the Oligocene. More de-
(see also Miyamoto et al., 1993), both molecular and morphological rived genera appeared in the early Oligocene, including the tayas-
evidence point to a very rapid diversification at the base of the ru- suids Perchoerus and Thinohyus, the anthracotheriid Elomeryx (with
minant clade, effectively muddling and obliterating characters that the disappearance of the more primitive Aepinacodon), and the en-
might be useful in resolving ruminant interrelationships. telodontids Megachoerus and Choerodon (see Figure 22.3).
 Artiodactyla 343

0
~
'Iii(II
0
0
ui
z

Figure 22.2. Biogeographic ranges

of primitive artiodactyls. A "box" (for
a particular time period in a par-
ticular biogeographic region) with a
cross through it means no fossil lo-
calities are known for that time pe-
riod from that area; a single dashed
line through the box means that only
scant fossil information is available
(usually only a single, small, locality).
Key: Diacodexis and related taxa are
in roman type: Bp = Bunophorous,
Dx = Diacodexis, Si = Simpson-
odus. Leptochoerinae are in roman
type underlined: Dw = "Diacodexis"
woltonensis, !Q = lbarus, La =
Laredochoerus; Le = Leptochoerus,
Ln = new leptochoerine genus, ~ =
Stibarus, L? = Leptochoerine indet.
Antiacodontinae are in boldface: An
= Antiacodon, Au = Auxontodon,
Nd = Neodiacodexis, Sa = Sar-
colemur, Ta = Tapochoerus. "Homa-
codonts" are in boldface italics: Ap
= Apriculus, Bm = Bunomeryx, Ha
= new genus a, Hb = new genus
b, Ho = Homacodon, Hy = Hy-
l,,_IJ, lomeryx, Hx = Hexacodus, Me =
?Au
Pe, De Mesomeryx, Mi = Microsus, My =
Mytonomeryx, Pe = Pentacemylus,
c: Au, Te = Texodon. Helohyinae are in ital-
19 !Q,
Q)
-!\l ·-
c: Ap, Bm, Hy, ics: Ah =Achaenodon, De= Discrito-
::> Au, My
_J Me, My, Pe choerus, Hn =new genus, He= Helo-
hyus, Pa = Parahyus.

Oreodonts were exceptionally diverse in the White River chrono- Oromerycids and basal protoceratids were moderately diverse in
fauna. Merycoidodontids diversified in the early Chadronian, which the Chadronian, but did not survive the end of the Eocene. Proto-
was also the time of the greatest diversity of agriochoerids. By the ceratines, first appearing in the Duchesneau as Pseudoprotoceras,
start of the Oligocene, Agriochoerus was the only surviving agri- persisted through the Eocene as the homed genus Protoceras (see
ochoerid, and primitive merycoidodontids such as Bathygenys and Figure 22.5). Chadronian camelids were mainly represented by
Aclistomycter were extinct, but the diversity of merycoidodontids the ubiquitous basal Poebrotherium, although the fioridatraguline
persisted through the Oligocene with the addition of more derived ?Poebrotheriumfranki is known from Trans-Pecos, Texas (southern
genera such as Eporeodon, Merycoides, and Merycochoerus. Lep- Great Basin). By the late early Oligocene (Whitneyan), stenomy-
taucheniinines also first appeared in the Oligocene, with the first lines were in evidence, with the first appearance of the peculiarly
appearance of Leptauchenia in the Orellari, and of Sespia in the specialized Stenomylus in the late early Arikareean (see
earliest Arikareean (see Figure 22.4). Figure 22.6).
 Christine M. Janis et. al 344

iii iii ii:' ii:' ii:'

ec !!?. ~ e z e0
0
....
_o
111~ c 'iii c·-
c
!!?.
.. Ill
Ill

.. -
c.5
Ill
c
iii 'iii
c.5
.. Ill
Ill
..
e:..
Ill
'Iii
:; .. :; ..
.. Ill Ill

=.c;:
.. Ill .. Ill GI Ill u
GI -
c Ill j!m .cm GI -
.ca. .. a. .ca. 0
0
GI 0 c .. t:: 1ii u
Oo
~oil
t:: 1ii
z0 ..GI
0
CJ)~
om
Ill
GI 111
o~ z0 ..GI Ill
a.
t::
0
w
G G
CJ) ..
G G G z z
Bm,Hb, /
Hy, /
Ah, Hn, Pa

/
/

/ I
~.An,
/
?Ho, Mi, I
/
Ah, He

/ I I
/
I I

Figure 22.2. (Cont.)

Among the ruminants, both hypertragulids and leptomerycids reean and the early Hemingfordian. A similarfaunal assemblage was
continued at moderate diversity through the latest Eocene and the apparent in the artiodactyls, the Runningwater chronofauna of Webb
Oligocene. Some primitive taxa were extinct by the Oligocene, such and Opdyke (1995). The leptochoerids were now extinct. Primitive
as the hypertragulid Simimeryx, and other more derived taxa ap- tayassuids were replaced by more derived taxa such as Hesperhys
peared, such as the hypertragulid Nanotragulus and the leptomerycid and Dyseohyus. Anthracotheriids and entelodontids persisted, each
Pronodens. now represented only by a single large-sized, specialized genus, Ar-
retotherium and Dinohyus, respectively, that did not survive past the
early Hemingfordian (see Figure 22.3).
RUNNINGWATER CHRONOFAUNA Agriochoerids were extinct by the early Miocene, but merycoido-
The start of the Miocene heralded the appearance of the "Cat Gap dontids remained diverse. Although some of the more primitive
fauna" among the carnivorous mammals, spanning the late Arika- merycoidodont subfamilies, such as the Prodesmatochoerinae, were
 Artiodactyla 345

iii iii ii:' ii:' ii:'

6
6 ec !!?. !!?. ~ e z 6e
< ~ _o Ill e:..
~ 'Iii ....
111~ c 'iii
c
E ·u; c.5
.. Ill
Ill

.
Ill
c
iii 'iii
.. -a.
c.5
.. Ill
u li 'Iii
Ill

:; .. =;:
Ill .. Ill
iii -~
.. Ill GI Ill GI - GI -
.ca. 0

- c Ill j!m .cm .ca.

0
E c .. 0
!·c 0 t:: 1ii u .c
'ti E GI . . GI
MA r::: Ill ti) ii E
.!! ::ii: oq: O<( '3
GI 0
Oo
~oil
t:: ..
0
z~
Ill
:; ii
0 GI
CJ) ..
om
CJ) ..
GI 111
o~ z0 ..GI Ill t::
a. 0 w
z
G G G G G G z
1.8 I

IX
c I
Ctl
0
I My, .El .El /
?El .El E1 E.! E1 Figure 22.3. Biogeographic ranges
Q) c I
Ctl
-Ctl- / of suiformes. A "box" (for a particular

x
w 2.4 ...J Ill I I/ 1,

IX x
time period in a particular biogeo-
z c I
w >-
Ctl graphic region) with a cross through
()
"!:@
0
I E1 E1 E1 E1 E1 it means no fossil localities are known
0 Ctl-

xx x x
Will for that time period from that area;
:::i 4.5
a. ' ca. a single dashed line through the box
(;) a.~ Ca, Ca, /
.'!l ~= I
MY EL .El E.! / means that only scant fossil informa-
j :c '.E Pr , tion is available (usually only a single,
5.2 I
/ I small, locality). Key: Tayassuidae are

t><
c. c I Ty?/ .Er. / in unitalicized, non-boldface roman
.'!l~~ I ?Tu / Ca .El El / El type. Basal tayassuids are in plain ro-

x
j :c '.E
I/ / man type: Cs= "Cynorca" sociale, He
6.0 /

IX
= Hesperhys, Fl = Floridachoerus,

~
~.
~ c. c / ?.El, My,
Ca,
~~~ Ty? Ty? Ty?
/ er .Er E1
Pe = Perchoerus, Th = Thinohyus,
.J :c :.c I/
Ts = "Thinohyus" siouxensis, Tc?
7.0
>- / / / , = Tayassuid indet. (?Hesperhys-

~
~ a. c / / / ?El./ "Cynorca" sociale clade), Ty? =
w~~ er My / / /
Ty? .Er er other tayassuid indet. Tayassuinae
u..i :c :.c / Pr
/ / I/ 11 - are in roman type underlined: Ca =

x
8.0 /
c:
a> ~ _g!
c I / Mg, /
/
Ty? I Catagonus, Co = "Cynorca" occi-
II Th Ty? Ty? dentale, Dy = Dyseohyus, Mg =
-Ctl-Ctl c I Tm / I

x
w 0 /
...J 0 'O , Macrogenis, My = Mylohyus, Pc
z / I/ I/
w 9.5 / = Perchoerus, B =Platygonus, Pn
()
~ ~ _g!
c: c
Mg, / I I "Prosthennops" niobrarensis,
0 c Ty? Ty? Ty? Mg Ty? I
~ Ctl
Pn, .Er / I ft = Prosthennops sensu stricto,

x
:E Ctl-
w0 'O
0
I/ I/ I
Px = "Prosthennops" xiphidonticus,
11.0 / / I I

x
I ?Ta = ?Tayassu, Tb = Black Hawk
cu .9
Q)

...J ~
'
,
c ?He
Ty?/ / /
Pn,Px
Ty?/ !Ty? I Ranch species, TI = Love Bone
I I
.J ~ ·~ / / / /
I/ I/ i/ I/ [/ 11 Bed species, Tm = Machaerodus
12.5 / Quarry species. Entelodontidae are in

~
Q)
co .9 '
/ He I
boldface: Ac = Archaeotherium, Ch
...J ~ c Ty? Ty? Qy Ty?
XJ/
/
I = Choerodon, Di = Dinohyus, Mg
u..i~·~ /
11
14.0 = Megachoerus, E?= entelodontid
,
6 He,

x
I Pn, indet. Anthracotheriidae are in roman
~enc Px He,
«i «i co I
Co, He, Px He Ty? I type italics: Ae = Aepinacodon, Ar

x
Will'> Co Qy , Px = Arretotherium, Bo = Bothriodon,
15.8 / El = Elomeryx, Hp = Heptacodon,
bi c

x
c I I
·EQ) 15~
Ctl /
Ty? Cs,
Ty? He, Co
Ty?, Ku = Kukusepasutanka. Suiformes
Q)
/ He I I indet. are in boldface underlined:
-
j:c.E E?
17.5 / I/ It Br = Brachyhyops.

extinct, others, such as the Leptaucheniinae, persisted. The earliest late Arikareean, camelid diversity was increased by the addition of
member of the most derived merycoidodontid family Ticholeptinae, protolabines, miolabines, and a greater diversity of fioridatragulines
Merychyus, first appeared in the early Miocene. Ticholeptines were (see Figure 22.6).
the only oreodonts to survive past the early Hemingfordian (see Pecoran ruminants also made their first appearance in the early
Figure 22.4). Miocene: Blastomerycines first appeared in the early late Arika-
Although the protoceratines continued into the late Miocene, reean, and the cranioceratine dromomerycid Barbouromeryx first
primarily in the form of the bizarrely specialized Paratoceras, in appeared in the late late Arikareean. Merycodontine antilocaprids
the earliest Miocene the more derived synthetoceratine protocer- may be known from as early as the latest Arikareean, but Para-
atids first appeared (see Figure 22.5). Camelid diversity increased cosoryx was firmly established in the early Hemingfordian. By the
markedly: Although basal camelids were now extinct, and Steno- early Hemingfordian there was also a diversity of blastomerycines
mylus was the only Oligocene stenomyline to survive past the early and the presence of the aletomerycine dromomerycid Aletomeryx.
 Christine M. Janis et. al 346

ii:' 6
e Ill
e
E·i
GI -
~0
.c 0.. (..)
t::-
0 Ill u.i
z I!!
Cl :i

I Cs, Fl, / He, Ar /

w Ar, / /
z Di Ar, Di
w
(,) 18.8
0 /
Si /

39.5

41.3

Figure 22.3. (Cont.)

In contrast, the traguloid ruminants were in decline. Hypertragulids momerycids, and antilocaprids, are absent. Gulf Coast taxa included
probably did not survive the Arikareean, and although the peculiar the tayassuid Floridachoerus (the only known occurrence), the pro-
leptomerycid Pseudoblastomeryx survived into the middle Miocene, toceratid Prosynthetoceras, the camelids Floridatragulus, Nothoke-
!eptomerycids essentially followed the same pattern as the hyper- mas, and Australocamelus (again, the only known occurrence).
tragulids (see Figure 22.7, 22.8).
Regional biogeographic variation was developing in the earli-
CLARENDONI AN CHRONOFAUNA
est Miocene. During the early Hemingfordian, the Gulf Coast had
an endemic and unusual component of artiodactyl fauna, where The late Hemingfordian (late early Miocene) marked the incep-
typical Runningwater chronofauna taxa, such as nonticholeptine tion of the Clarendonian chronofauna, terminating at the end of
merycoidodontids, the tayassuid Hesperhys, the protoceratids Lamb- the Clarendonian. A rather dramatic change in artiodactyl fauna!
doceras and Syndyoceras, stenomyline and miolabine camelids, dro- assemblages occurred in the late Hemingfordian. Although all the
 Artiodactyla 347

6 ID ID ii:' CL CL
ec: ~ ~ ~ e z e6
~
C(
~
~
1ii .....
_(..)
111~
6
c ·c;;
c:
c: ·-
... Ill
Ill
·=
c: Ill
...
-
1ii"lii
...
Ill
c:
c: ·=
Ill
... lll
..
e:..
1ii
lll lll ... lll GI lll
= lll

-
GI -
0 -c: Ill ~ IQ GI - 0
E ......
1ii -~ (..)
Ill
GI 0
.c IQ
m
.c 0..
caGI
-o.. .c 0.. !E
0 ;:: (..)
MA "g E GI .. GI uu t::- 'S '5 c: -lll
GI t:: 1ii 0 .c
~I!!
0 lll t:: u.i
~~~ (..)ii <(E (..)I!! lll
"5 z I!! 0 GI 0
Cl ~oil Cl
CJ) ...
Cl
CJ) ...
Cl Cl Cl
0.. 0
z z

x
1.8
c
lll
u I / I
Q) c
1ii .Ill I / I

x
_,CD
~ 2.4 / I

x
c I
w lll
/

C><
0 >- u
- c I
0 ca .£2 /
:; 4.5
UJ CD I/

~ D< x
0..
t) 6. c

IX
I /
2~,jg
..'.'.!IE /
I ,

x
5.2

x
6. c: I Figure 22.4. Biogeographic ranges
E / /
<ll t11
I My of oreodontoids. A "box" (for a partic-
j~~ / /
ular time period in a particular biogeo-

x x
6.0 ii I/ /
graphic region) with a cross through it
2!- ' c
... a. I
means no fossil localities are known
U'.l ~~ I My
for that time period from that area;
_j I:C
7.0 i/
a single dashed line through the box

8.8
-E'
ctl a. c
w E -~
•
UJI:C
Q)
I

=
/
/
/

/
My /
/
I/
/
/
My
x means that only scant fossil informa-
tion is available (usually only a sin-
gle, small, locality). Key: Agriocho-

x
c c I eridae are in roman type italics: Ae
<D ~ -~ / / My/
w ca J2 g
_, () -0
I My My / My My / /
= unnamed genus E, Af = unnamed

x
ffi 9.5
I I/ I/ I/ genus F, Ag = Agriochoerus, Pa = un-
named genus A, Pb = unnamed genus
0 cc My/ I
~~-~ /
0 ... ctl c B, Pc = unnamed genus C, Pd = un-
~ ctl- 0 My My My My My My / / I

x
UJ () -0 named genus D, Pr = Protoreodon.
11.0 I/ I/ 11
Merycoidodontidae are in unitalicized

x
Q) '
I
«i ~ c My
/ My / My /
My
/ My I roman type. Leptaucheniinae are in
...J ~ co I
_j CD·;; / / / / / roman type underlined: Le = Lep-

x
12.5 '/ / I/ I/ I/ 11
tauchenia, Se = Sespia. Ticholepti-

x
Q) '
«i
...J ~ c
.s Br, Me,
My / I nae are in boldface: Br = Brachy-
My My / My Ti
• ctl ctl
UJ CD·;; My /
I crus, Me = Mediochoerus, My =
14.0 I~ Ii Merychyus, Ti = Ticholeptus. Other

15.8
>--
"'i:.
6
~
ctl ctl ctl
UJ CD·;;
c:
My

I
I
I
Tl Br Br, My,
Ti
Br, My
x Br, Me, My, Tl Br, My,
Tl
Br,
My,
Tl Ii
I
I merycoidodontines are in plain ro-
man type: Ac = Aclistomycter, Ba =
Bathygenys, Bl = Blickohyus, Ep =

~v x
Ci c I Eporeodon, Eu = Eucrotaphus, Hy
-~
co / Br, My, My/
<ll E i5
Br, My, = Hypsiops, Mi = Merycoides, Mo
- Q) ... / Ti Ti
My Br, My, Ti Br, My / I
j I .E = Merycochoerus, Od = Oreodon-
17.5 I/ II
Ci toides, On= Oreonetes, Po= Parore-

x
cc Mo I ?My/ My/ Me, My, Ti, bll I
>."- co: / My, odon, Pm = Prodesmatochoerus, Ph
My
~~E I / / Hy, Mi, Mo / Mi?, Mo
I = Phenacocoelus, O? = oreodontoid
UJ I . 8 I/ I/ I/
18.8 I/ II
indet.

taxa that typify the C!arendonian chronofauna were survivors of The stenomyline camelids Blickomylus and Stenomylus were ex-
earlier North American lineages, many of the earlier lineages were tinct by the early Barstovian, and the cameline Procamelus made
now extinct, including anthracotheriids, entelodontids, nonticholep- its first appearance. The dromomerycids Subdromomeryx and Bar-
tine merycoidodontids, hypertragulids, and leptomerycids (with the bouromeryx were extinct by the early Barstovian, and the meryco-
exception of the pecoranlike Pseudoparablastomeryx). dontine antilocaprid Ramoceros made its first appearance.
Webb and Opdyke (1995) perceived a separate late Hemingfor- The first tayassuine is known from the late Hemingfordian, and
dian "Sheep Creek chronofauna." However, there is little evidence tayassuids were moderately diverse during the Clarendonian
for this at the generic level of artiodactyl diversity. No change is chronofauna. More primitive tayassuids, such as Hesperhys, sur-
apparent among suoids, oreodonts, protoceratids, or hornless rumi- vived until the end of the Barstovian. More derived tayassuines,
nants. Only slight changes are apparent among camelids and homed such asMacrogenis, appeared in the Clarendonian (see Figure 22.3).
ruminants between the late Hemingfordian and early Barstovian. Merycoidodontids were moderately diverse during the late
 Christine M. Janis et. al 348

iii iii ii' ii' ii'

0
ec: ~ ~ ~ e z 0e
~ 0 c: Ill Ill Ill ~
_(.) c: c: .5 ti
ti
l'?= c: 'iii c: ·-
~ Ill
c: .5
~ lll
~ :! ~ lll
oi lll

-- --
lll ~ lll Cl> lll Cl>- Cl>- 0
0
(.)
-c: Ill
lll
Cl> 0
11 IQ .c IQ .cO.. -o..
c:
.c 0.. !E ;::
o.c
(.)
(.) () t: 1ii ::I lll ::I lll Cl> -lll t:
0 -lll lll t: u.i
!!:: 0 Cl>
en0 Cl>
en0 Cl> (.)I!! z I!! 0.. 0
::I
Cl C3 oil z~
Cl
~
Cl
~
Cl Cl Cl z z
?My/ Me / My, !.& My, ?My,/
Ep, Hy, /Hy,
/ /
Mi, Mo, Od Mi

/ / !.& My,
Ep, Hy, Mi,
Mo, Od, Ph

Ag, !.& Se,

!.& fie. Ep, Mi,
Ep, Mi, Mo
Mo, ?Od
Ag, !.& fu!.
Bl, Ep, !.&~.
Mi, Mo, Od, ?Ep, Mo
On, Pm
Ag,
Bl, Ep,
!.& Bl, Eu, On

Ag, Ae, Ag,

!£Ba,
Ba, On, Pm
On, Pm
35·5 1---,-c---l~-~,...----,'°*--~;---~f----t--i-~A~e.~A~g-,-t-A-g,-P-c--i
~ al .!!!
w
z
~ c3 e Pb, Pc, Pr,
1& ?Ac,
Ba, Pm, On
!&
Ba, On,
Pm
~37.1 Ag, Pc,
0 ?Ag, Ae, Pr,
w ?Pb, Pd, Pr !&
Pm, On

c Ag, Pa, ?Ag, Pb,

Cll .l!! Pb, Pr Pc, Pr
-lll ·-
c
_J ::::>

41.3 ~-c--4---.lJ.L---'ll~---k----4----f.-+-----+----,,.,--li'---~---'11
-E"~ Ag, Pr /
"'·-
45.9 1--=LU=--:::>:.__¥---¥--~
/
?Pr / I
I
/
/ I
Figure 22.4. (Cont.)

Hemingfordian and early Barstovian. After the early late Barstovian split into Lamini and Camelini sublineages), became common dur-
they were reduced to a single genus, Merychyus (see Figure 22.4). ing the middle Miocene. Although both Aepycamelus and Pro-
Protoceratids, although individually rare as fossils, persisted camelus survived until nearly the end of the Miocene, by the late
through this interval. By the early Clarendonian the more primi- late Barstovian more derived camelines, such as Hemiauchenia and
tive synthetoceratines were replaced by the larger, more hypsodont, Megatylopus, were in evidence. The highly specialized stenomyline
Synthetoceras (see Figure 22.5). Camelids were very abundant and Rakomylus persisted through the early Barstovian. Floridatragulus
diverse. Camelid generic diversity reached its zenith in the early was the only ftoridatraguline to survive through the middle Miocene.
Barstovian, with 13-14 sympatric genera. The first undoubted came- Miolabines and protolabines persisted through the Clarendonian
line is known from the late Hemingfordian, and camelines, mainly chronofauna interval, but their diversity was reduced after the mid-
by species of Aepycamelus and Procamelus (representing an early dle Miocene (see Figure 22.6).
 Artiodactyla 349

~ ~ ~
u
u
iii
ec:
iii'
~ ~ ~ ~ z u~
< ~
... _u c: Ul Ul Ul ~
~ Ul
.,~

~-
c: 'iii c:
~ ·-
Ul c: lll
~ ·= c:
iii 'iii E·~ 'IiiCl> 'Iii
lll

~
Ul ... Ul ~ lll Cl> lll Cl>- u
i:i e Cl>
-~
"'·-
~
... Cl>
~
0
(.)
!!::
c: lll
Cl> 0
(.) (.)
11t:: IQ...
0 lll
......
.c IQ
::I lll
Cl>-
.c
:; 1)
0.. ~-
'!: 0..
Cl> 1i
.c
0
0..
t:: ...
lll
!E ;::
u .c
lll t::
0
(.)

ui
MA ~ lll tl) :ii E z I!! en0 Cl>
en0 Cl> (.)I!! z I!! 0.. 0
-.I::!!:":( (.) <( Cl
::I
(j oil Cl
~
Cl
~
Cl Cl Cl z z
1. 8 . ;
c:

w
z 2.4
Cl> c:
-l l llll
_J
-
co
lll
()

I
I
I
I/
/
/
[X I
I
I

[)\
c:
w
(,)
0 ~~
>- ()
lll

11
I
[X ~
rx[X
LU CO
:J

x
Cl.
4.5

5.2
1ii
2~~
j::c:~
a. c
I
I
'
Ky
[X .
/
/

[)\
I I
6. c: I
/ /
2 ~ ,gJ I
j::c:~I/
I/
/
I/
/
IX Figure 22.5. Biogeographic ranges

x
6.0
of oromerycids and protoceratids. A

[X
~.
~ a. c: I
~~~ I
"box" (for a particular time period in a
_j ::c: :.c: 11 particular biogeographic region) with

x
7.0 I / I /
a cross through it means no fossil lo-
>-
~ O..c / / / / calities are known for that time period
LU E -~ St from that area; a single dashed line
• Cl>= / / / /
LU ::C: :.C:
11 I/ I/ I/ through the box means that only scant

x
8.8 / /
C:: c fossil information is available (usually
I /
Q) ! .~ / / only a single, small, locality). Key:
- c:
-l l llll I

x
Ii.I 0 / / /
_J (..) "O Oromerycidae are in boldface: Eo =
zw 9.5
I/ I/ I/ 11
Eotylopus, Ma= Malaquiferus, Me=
I
(,) C:: c: Pa, / I Merycobunodon, Mo= Montanatylo-
~~-~ /
0 ~ lll c:
St
I
Si lll- 0
LU (..) "O
St / /
,, pus, Or = Oromeryx, Py = Protylo-

x
11.0 ; ; ;
v II pus, O? = oromerycid indet. Protocer-
I

x
2 .8 { I atidae are in roman type. Basal proto-
j~c
Pa, / / /
La or Ps / /
I ceratids are in roman type italics: He
_j ~ -~ La, Ps / / / / /
= Heteromeryx, Lo = Leptoreodon,
1, I/ I/ I/ II 11
12.5 Lp = "Leptotragulus" profectus, Lt=

~ x
/
Cl>
1ii .8
'
/ I
Leptotragulus, Po= Poabromylus, To
_J ~ c Ps Ps /
I
ui ~ -~ 1....
/
II
= Toromeryx. "Protoceratinae" are in
14.0 roman type underlined: Pa = Parato-
0 P?/ I ceras, Pp = Pseudoprotoceras, Pt =
>--

15.8
'i:: ~
lll lll lll
LU Cll ·::; Ii
c
I
P?
;
IX La
II
I Protoceras. Synthetoceratinae ar--;;- in
plain roman type: Ky = Kyptoceras,
La = Lambdoceras, Ps = Prosyn-

~ x
Ci c:
c: lll Ps / I
/ thetoceras, Sd = Syndyoceras, Sy =
Cl> .E 'E La P? I
m::c: -o
Q)
I/
/ / Synthetoceras. P? = protoceratid in-
17.5
_J 11 11
det.

Among the ruminants, blastomerycines were less diverse than in cines and the more derived cranioceratines such as Bouromeryx
the earlier Miocene, but Blastomeryx and Parablastomeryx survived first appeared. Dromomerycid diversity declined through the mid-
until the end of the Clarendonian, joined in the late late Barsto- dle Miocene, and by the early Clarendonian the cranioceratines
vian by the more derived Longirostromeryx (see Figure 22.7). Dro- were the only surviving lineage (with the exception of a couple
momerycids and merycodontine antilocaprids greatly diversified in of possible Clarendonian specimens of Dromomeryx). The larger,
the late Hemingfordian and early Barstovian: Dromomerycids had more derived Pediomeryx (Yumaceras) made its first appearance
their zenith of seven sympatric genera in the early Barstovian, and in the late Clarendonian. Merycodontines remained fairly diverse
merycodontines were most diverse with six genera in the late late through the early Clarendonian, but were reduced to three gen-
Barstovian (see Figure 22.8). era (Cosoryx, Ramoceros, and Merycodus) by the late Clarendo-
In the late Hemingfordian, the aletomerycine dromomerycid nian. The first antilocaprine, Plioceros, was in evidence in the late
Aletomeryx was replaced by Sinclairomeryx, and the dromomery- late Barstovian, joined by the more derived genera Ceratomeryx
 Christine M. Janis et. al 350

6 m m CL
!!!.
ii:' ii:'
~ ze:. 6
~ 6 ~ !!!. £ ~
c: Ul Ul
_(..) c: Ul
c: c: .E 'lii
c: ·- c: .E 'lii
'lii c: 'iii
; = ......
.,~

lii "iii .... Ill

lll
0
.::= 1ii
c: Ill
.... lll
_2!m .cm
.... Ill
GI -
.c D.. ....
... -
o.. GI -
.c D..
GI
:=U .c;:: 0
lll
(..)
(..) GI 0 t:: ... c: ... t:: ... (J
'S 1ii :::J lll 0 Ill t:: ui
z0 l....B
:t:: (..) (..) GI Ill lll
(..) ! z!
:::J
Cl (j oil Cl "' .... "' ....
0 GI
Cl
0 GI
Cl Cl Cl
D.. 0
z z
/
Psi
/ /
Ps La
I
/ /

/ / /
Sd
/ /

/ et, /
Sd /

Eo,
He, Lp, Po, Lp
J:Q.

Eo, Eo,
Ell Lp

Eo, ?Mo, Mo,

He, Lp, Lt, Lp,
Po, !:p ~

?Lt, Po

Figure 22.5. (Cont.)

and Sphenophalos by tbe late Clarendonian. The immigrant gelocid few other camels are found there, at least until the late Clarendo-
Pseudoceras also made its first appearance in the Clarendonian (see nian. Antilocaprids were absent from this region prior to the late
Figure 22.8). Clarendonian.
Patterns ofbiogeographic variation are apparent during the Claren- Some additional patterns may also exist. Dromomerycids and
donian chronofauna, as they were in the Runningwater chronofauna. antilocaprids are sparsely represented in the Pacific Northwest and
The Gulf Coast again appeared to have a partially endemic fauna: northern Great Plains regions, especially past the early late Barsto-
Merychyus was the only merycoidodontid in evidence; the protocer- vian, and cranioceratine dromomerycids are unknown. Antiloca-
atid Prosynthetoceras mainly occurs in this region (whereas Lamb- prines are well represented in tbe northern and southern Great Basi.ns;
doceras was primarily a central Great Plains taxon). The camelid dromomerycids are rare. In contrast, dromomerycids are fairly well
Floridatragulus also is known primarily from the Gulf Coast, and represented in tbe California Coast Region, at least until the late
 Artiodactyla 351

6 m m CL ii' ii:'
z 6
~ !!!. !!!. £ ~
< ~ 6
_(..) c: c: Ul Ul Ul e:. ~
£ ·u;Ill c: .E c: 'lii
'lii ....... c: c: ·- E ·i
:=ui.c;::
111~
.... Ul iii "iii
Ci Ill lll ... Ul ....
1!t:: m... GI Ill
.... Ill
GI - -
.... D.. GI - 0
Ill

- c: Ill .c
0 .c m .c D.. ... D..
E lii .2
........ c: ... (..)

MA ~
i::i E GI +- GI
(..) GI 0
(..) (..) 0 Ill 'S 1ii 'S0 m GI Ill
t:: ...
0 Ill
(J
Ill t:: ui
Ill tl) 5i E 3 (j z f "'
0 GI
"' ....GI (..) f z ! D.. 0
oil .... z

xx
..... ::: '<( (..) <( Cl Cl Cl Cl Cl Cl z Figure 22.6. Biogeographic ranges
1.8 I
c I of camelids. A "box" (for a particular
lll ?Ca,/ Ca, Hx,
?Be, Ca,
Be, Ca, Ca, Gi, Gi,
I
GI C:
u I Hx Gi, Hx, time period in a particular biogeo-
-l l llll
-
/ Ti Gi, Hx Hx, Ti Hx I graphic region) with a cross through
co I ?Hx ?Mg, Ti

x[X
w 2.4 _J [/ I/
it means that no fossil localities are

x
z c: I
w >- u
lll
Ca, Hx, Ca, Hx, Ca,Gi, Ca, known for that time period from that
0 Ca,Hx
0 ~~ I Mg, Ti Mg Hx, Mg Hx, area; a single dashed line through
wen II Mg

rx
:::i 4.5 the box means that only scant fossil

x
Cl..
6. c:: ' information is available (usually only

IX
1i) Hx, Mg, ?Mg/
.2! ~~ I Hx Hx, Mg Hx, Mc
/ a single, small, locality). Key: Basal
..'.3::c:E Pa
I camelids are in roman type italics
5.2

~ x
I Hx,
I / underlined: Pa = Paralabis, Pb =
a. c Hx,/
Al, Hx, Al, Hx, Al, Hx,
?Mg/ ?Hx, Poebrotherium, Pd = Poebrodon, fl
.$ ~ ~ I Mc,
..'.3::c:E Mc, Mg Mg Mg / Mg = Paratylopus, "Pt"= "Paratylopus"
I
?Mg , /Mg
6.0
I
cameloides. Stenomylinae are in

7.0
~ 6.
Jl ~ ~
_j ::c: E
c:

t><
Ae, Hx,
Pr
C? Hx, Mg Hx, Mg
Hx, Mg, Al, Hx,
Pr Mg, Pr [>\ ,
Hx,
Mg,
?Pr
x boldface: Bk = Blickomylus, Mt
= Miotylopus, Ps = Pseudolabis,
Rk = Rakomylus, St = Stenomylus.

x
>- I
Al 7 Mi, I Floridatragulinae are in boldface
~
w E.!!!
• =
O..c
Q)
Pl,
Pr
Ae, Hx,
Mg
C? /
/
H~ /
) Mg
Mg,
/
:i '
Ae, Hx, ?Al, Hx,
Mg, Pr Mg, Pr
Mg/
/
Mg, underlined: Ag = Aguascalientia, fil
= Floridatragulus, ff= ?Poebroth-
w::c::c I/ 11 1,
Pr Pr
8.8 I/ erium franki. "Nothokematinae" are
c. ?Mi,?PI, Hx,/ I

x
Ml, Nt, I
c: I Mg,/ in boldface italics underlined: Ge =
<D e:? -~ Ae, Hx, Pl,
Ae, He, Ae, Mg, Pl, Ae, Mg/
-Ctl -lll c: I ?Mg, Pr Hx, Mg, Gentilicamelus, Nk = Nothokemas.
w 0 Pr

xx
(..) -c / /p
_J , ?Mg, Pr ,, /?M
. g Pr , r
ffi 9.5
Pr, Px 11 Miolabinae are in roman type ital-
ics: Cu = Cuyamacamelus, Ml =
0 >-
c. c: Nt, Pl,
Pl,
Mi, Pl, ?Mi, Pl, Nt, Mi, Pl,
Ae, C? /
I
I I
0 GI lll
-.:: (Tj ·c: Ae, Mg, Ae,?Mg, Hx, Mg, Pl, Miolabis, Ms = Miolabine sp., Pm
Hx, Mg, I
Si (..) -c0
lll -
w Pr
Pr ?Pr ?Pr Ae, Pr Pr, Px II
/ / = Paramiolabis, Nt = Nothotylopus.
11.0 11 II Protolabinae are in boldface italics:
1

x
I I
_o
Q) '
.El, C? / /
Mi, E!, Ml,Ms,Nt, C? I
C?; I Mi= Michenia, Pl= Protolabis, Ta=
ctl t) CI Pl, / '/
~ ctl Mi, Pl, Ae, I
_J
_j co ·:;; Ae / / / Ae / / Tanymykter. Camelinae incertae sedis
I/ Hx, ?Mg, Pr 11
12.5 I/ 11 11 II are in roman type underlined: Au =
/
Australocamelus, Ox = Oxydactylus,
.e
[>< Ml, Pm, Ml, Ml, Nt, Pm, ?Ae
Q) '
(ij Ml, / Ml,
_J ~ c Pl, Pl, Mi, Pl, / C? Pc= Priscocamelus. Camelinae are in
. lll lll
Mi, Pl Ae, Pr I
w co"> Ae Ae, ?Pr Au, Ae, Pr I/
/
I/ plain roman type: Ae = Aepycamelus,
14.0 Al = Alforjas, Ap = "Aepycamelus"
6 ?Ox
1 .El, Cu, Ml, Ml, Pm, Rk,Ml,Nt, Ml, Ms, Pm, I
>--
-.:: ~ c Pm.Mi, Pl, Ae, Pr
priscus, Be = Blancocamelus, Ca
Ms, Pm, Pl, Mi, Pl, ?Mi, Pl, Ml
wen-:;; 1 I 8Jd., Ae, I = Camelops, Gi = Gigantocamelus,
lll lll lll
Ae Ae,Hc Ae,Hc Hc,?Pr, Px Ae, ?He, Pr
15.8 II He = Hesperocamelus, Hx = Hemi-

~
Ci c: / Bk, I
St, Ml, Pm, Bk, St, ?Ml, I auchenia, Mc = Megacamelus, Mg=
Q) ·~ ~ / Ag, Ml, Ml, Ms, Pm, I
Ml, Pl, Mi, Pl, Mi, Megatylopus, Pa = Paleolama, Pr =
j ~ .E /
Ae Ae
Pm, Nt, Mi, Pl,
Ae
I I
Procamelus, Px = Pliauchenia, Ti =
17.5 I/ Mi, Pl, Ae Ox, Ap I I
Titanotylopus. C? = camelid indet.

late Barstovian; antilocaprids are rare during this time (see Fig- in the California Coast region than they were in the central Great
ure 22.8). Because antilocaprids and dromomerycids are clearly Plains. Additionally, they are rare in the Gulf Coast faunas (see
specialized for different habitats (open savanna, and woodland or Figure 22.6).
forest, respectively) the differences in their distributions may serve
to identify different paleoenvironments, much like the separate dis-
tributions of pronghorn and deer today.
MIO-PLIOCENE CHRONOFAUNA
A similar biogeographical pattern may be evidenced by the The Hemphillian (late late Miocene) marked the inception of the
camelids. The Clarendonian chronofauna miolabines and proto- Mio-Pliocene chronofauna. Tayassuids remained diverse throughout
labines were generally absent from the northern Great Plains and the chronofauna. The more derived tayassuines Mylohyus, Platy-
Pacific Coast regions, but prominent in the northern and southern gonus, and Catagonus appeared, although Catagonus did not sur-
Great Basins. Also, these animals were probably more common vive into the Blancan (see Figure 22.3). A few specimens of the
 Christine M. Janis et. al 352

m m CL
e
ii:' CL
~ ze:.. u~
~ e 8.
c:
c: 'iii
c:
c: ·-
... Ul
Ul
c: .E
Ul
c:
Ul
... ...
E·a u :II Ul
lll
... lll
~m
GI lll
.cm
... lll
GI -
.c 0..
!1!
... 0..
GI -
.c !E ;::
0.. 0
(..)
'Iii u.c
t:: ... '5 'Iii '5 'Iii i 1a t:: lll t:: ui
0 lll
(..) ! z0 GI...
z !
Cl "' ... "' ...
0 GI
Cl
0 GI
Cl Cl Cl
0.. 0
z z
?He/ Bk, "Pt", I
Ms,
/
Pl

Mi,/ St,
/Pl ~
Pc C?
St / /
/

Ge

Ps

Pb,El Pb

Pb,E!l Ell

Ell Pb

Ell

Figure 22.6. (Cont.)

merycoidodontid Merychyus persisted until the end of the Miocene southern distribution of the group. The Hemphillian camelid diver-
(see Figure 22.4). The apparent persistence of this taxon in the Cen- sity was enriched by the addition of derived taxa such as Alfor-
tral Great Plains alone may be real or may represent a sampling jas and Megacamelus. The majority of the Clarendonian camelid
genera survived through the early Hemphillian. Hemiauchenia, and
artifact.
Only two protoceratids, the large and bizarrely homed Syntheto- perhaps Megatylopus, survived through the Pliocene. The Blancan
ceras and Kyptoceras, are known from the Mio-Pliocene Gulf Coast was a time of bizarre cameline specialization, with the evolution
faunas. As was the situation with oreodonts, protoceratids barely of giants (giraffe-sized but with greater bulk) such as Titanotylopus
survived through the early part of the Mio-Pliocene chronofauna. and Gigantocamelus. The more moderately sized Came/ops (around
Camelids were represented only by camelines, with the exception the size of present-day Came/us) was also a common element in
of two rare Hemphillian protolabines that may represent a relict Pliocene faunas (see Figure 22.6).
 Artiodactyla 353

u m m e
CL CL CL
z u~
< ~ u ~ ec: 8. ~
8. ...:(!
_u c: Ul Ul
c:
Ul e:..
CQ lll
f! ::-
... Ul
c: 'iii
... lll
c: ·-
; = .c... 0..... c: -=
... lll
GI- f!... :iii E·i
GI - u'i '1lllii
E -u 0 c: lll ~m .cm 0.. .c 0.. !E ;:: 0

MA "g E GI
f! ·c:
.,. GI ...'3
(..) GI 0
(..) (..) t:: 'Iii '5 1a :::J lll
c: ..
GI lll
t:: ...
0 lll
u.c
lll t::
(..)

ui
~~~ (..):ii <(E "' ... "' ...
z0 ...GI 0 GI 0 GI (..) ~ z!
(j oil 0.. 0
z

xx
Cl Cl Cl Cl Cl Cl z
1.8
c: /
lll
u / I
Q) c: I
I

x~~
Cil .!!l /
w --'co I I/ 1,

x
z 2.4 c
w lll
I /

0 >-"' /
I
0 ~~ /

x
LU CO I
:J 1,

x
4.5
Cl. ..... 6.. c:: I

*~~
/
I /
--'I .c
5.2 I

[>< x
/ /
6. c I
/ /
2~~ I

x
/ /

x
jI'.E Figure 22.7. Biogeographic ranges
' I/ /

x
6.0 of hornless ruminants and Rumi-
.:2:-'
~ a. c:
nantia incertae sedis. A "box" (for a
i1l ~Ji! Ps Ps Ps
particular time period in a particular
..JI:C
7.0 biogeographic region) with a cross

x
>- / / / /
~ 6..c through it means no fossil localities
/ / / /
LUE.~ Ps Ps are known for that time period from
/ / /
u.d£ I/ '/ / I/
/
that area; a single dashed line through

x
8.8
c c: I ?Bl/
/
Lo, Pb,
/ the box means that only scant fossil
<D ~-~ / / information is available (usually
-l l llll
- 0 c: I Ps I

x
w --' (..) 'O , ,
/ Ps / only a single, small, locality). Key:
ifi 9.5
1,, /
Hypertragulidae are in plain roman
0 c c: Bl, Bl, Lo, ?Bl Bl, Lo, Pb,
/
I I type: Ht = Hypertragulus, Hs =
0 2:-- ~ -~ /
~ lll c

:E Cll- 0 ?Ps Ps I I Hypisodus, Nt = Nanotragulus, Pt=

x
LU(..) 'O Ps /
,
11.0 I/ I Parvitragulus, Sx = Simimeryx, H? =
/ 7 / hypertragulid indet. Leptomerycidae
JB 6 I
~ ~ c:
..J a3 ·~
Bl, Lo
/
/
/
/

x Pu,
/
/
I are in roman type italics: He =

x
Bl, Lo Hendryomeryx, Le = Leptomeryx,
12.5 / / I/ v
/ Pd= Pronodens, Pu = Pseudopara-
.,- c::
x
Q) '
-o
Pu,
Pu, / I
...J ~ Bl blastomeryx. Blastomerycinae are
. lll Cll Bl, Lo / Bl I
LU C:C ·:;; Bl / in boldface: Bl = Blastomeryx,
14.0 / v Lo = Longirostromeryx, Ma =

x
0
>-- I I Machaeromeryx, Pb = Parablas-
i:: ~ c:: Pu, Bl or Bl Pu, Bl Pr
lll lll Cll I Pb I tomeryx, Po = Pseudoblastomeryx,
LU C:C '$ Bl Bl
I , Pr = Problastomeryx, B?= blas-
15.8

~
tomerycine indet. Gelocidae are in
Ci c
c lll
Bl I Pu,
Pr I ' roman type underlined: ~=Pseudo­

17.5
.E 'ti
Q)

3 ~ .E
[/
I
I Pb Bl Bl
x Bl, Pr
Bl, Pb

11
I
I

1,
I
I
ceras. Ruminantia incertae sedis are
in roman type italics underlined: De =
Delahomeryx, Hi = Hidrosotherium.

There was a great diversification of antilocaprine antilocaprids in cids were virtually extinct. The great diversity of antilocaprines that
the later Miocene, with the persistence of only a few merycodontines. characterized the late Miocene survived only in the latest
Blastomerycines were extinct by the Hemphillian, although the sim- Hemphillian of Florida. Most of the Pliocene antilocaprines are
ilarly appearing immigrant Pseudoceras persisted through the early represented by the specialized tiny Capromeryx and stocky Texo-
Hemphillian. Cranioceratine dromomerycids persisted at fairly low ceros. The first true cervids appeared in North America in the ear-
diversity, with the larger (eland-sized) Pediomeryx (Yumaceras) re- liest Pliocene. They persist to the present, although cervid Blancan
placed by the smaller, apparently rare (hartebeest-sized) Pediomeryx diversity was still fairly limited (see Figure 22.8).
(Pediomeryx) in the late Hemphillian. The enigmatic, poorly known, Several biogeographic patterns are apparent during the Mio-
supposed bovid Neotragoceras was also an immigrant to the Mio- Pliocene chronofauna. On the one hand, antilocaprids continued
Pliocene faunas (see Figures 22.7, 22.8). to be more diverse in southern and western regions, especially in
By the Pliocene, merycodontines were extinct and dromomery- the Miocene and earliest Pliocene, than elsewhere, and are virtually
 Christine M. Janis et. al 354

~
u ~ z
_u "' e:.
f! :::- E·~ U GI
'Iii
GI -
~"'
c: Ill .c: 11. ;;:: ;::
GI 0 t::~ ·c:; .c:
(..) (..) 0 Ill Ill t::
MA z !!! 11. 0
~oil CJ z

Nt, Le Pd

ch
Q) c
ii::J Ctl He, Le
0
Q)
c: He, Le

Sx, H?,
He ?Le
Figure 22.7. (Cont.)

unknown from the northern Great Plains. Camelids, on the other Bridgerian) is dominated by the radiation of primitive artiodactyls.
hand, were not especially diverse in the Gulf Coast and California During the Bridgerian, diacodexids declined while helohyines in-
Coast regions, but were most prominent in the southern Great Basin creased in diversity, and the first agriochoerid oreodonts may have
and southern and central Great Plains. appeared during this time.
A late middle Eocene assemblage (Uintan to Duchesnean) was
characterized by the appearance ofnew forms, probably immigrants:
SUMMARY Suiforms (anthracotheres and entelodonts), tylopods (protoceratids,
The North American Tertiary artiodactyl radiations follow a succes- oromerycids, and camelids), and traguloid ruminants (hypertrag-
sion of faunal assemblages, or chronofaunas, that are coincident with ulids and leptomerycids ). Among the artiodactyls surviving from the
those observed in carnivorous mammals (see Chapter 4, this vol- earlier Eocene, there was a moderate diversification among the anti-
ume). An early to early middle Eocene assemblage (Wasatchian to acodonts while the diversity of other primitive artiodactyls declined,
 Artiodactyla 355

< ~ u ~ ~ ~ ~ ~ z-u Figure 22.8. Biogeographic ranges

~ t; 'i6~ ~ ~
i~
of homed ruminants. A "box" (for a
c'1115 !:,
~·= m~
C UI
iii m .=t; .... E·m E·m __uu;a> particular time period in a particular
1le "iii~ u0 c:., _!~ .c:m .!ii: J=ii: .!ii: :=;:: o
-a>~ ~8

x
biogeographic region) with a cross
MA c: E
t11lll0!
!II Eli; ::
:::i <C
t:: a;
~I!!
:;Oa>
a; :;Oa>
a; 5i a;
UI!!
za Oil!! u., S
._
uw· through it means no fossil localities
uc:c
...i::i!lq:
CJ uo!I CJ enc; a.~ z
enc; CJ CJ
.8r--~c~r-"-f'----T---.,. __...,____+-...;;,-+--;;;_"-..;;...,1.._:.,i.:;.J are known for that time period from

a> c:~ I Cx, A?, / Cx,Tr, cx, cx, Tr, cx, OdA?, II
that area; a single dashed line through
the box means that only scant fossil
«i,g! I

x
Od /C? ?Od Od Od C?
information is available (usually

[X
_J ~ I/ ..
~ 2.4
[X
11

w
0 >- ~ / A?, C? / Cx, Hb, Cx, Cr, Cx,
only a single, small, locality). Key:
Dromomerycidae are in boldface.

l\./x x/// x
0 tfl-.:::~
iii ,, Od /
Od Od
Od BO
r, d,
Aletomerycinae are in boldface
:::::; 4.51:"-:;:-~t"~--r'--~->jo~~~/~~71-~~-J-~~-f.~~-k'...._~4!-~.!Y!L~_jl~
t~
11. underlined: Al Aletomeryx,
j I HD\ Hb,11, Sc Sinclairomeryx. Dromo-

ex:
Pp,
.'.3 :i: E C? / ~d a.I~ ?Od ?Pp A?, C?
i----- ~~..,f--+--"'*---+----+---~"'-.....;~-~~
merycini are in plain boldface: Dd =
5.2

2
E 1,
6. c
§ :ll1 I
I
~:· Tx, / ?Pp,
Pp,
01,
?Hb, / • Pp,
11/ II,
Pp, tyg / IX Drepanomeryx (Drepanomeryx), Dm
= Drepanomeryx (Matthomeryx), Do
j J: Sp, Tx ll/g. A? Tx Sp, Tx = Dromomeryx (Dromomeryx) Ds
6.0r-;:::--""""11[)<--;f-~-t~~1l/~~-t-~--1L--~-+--=:.__~l/:__~.J.-~-¥-~
~ ~~
..J :i: :2
Pp,
Hm, Tx
Py,
Tx
II, Sp II, Os,
Pl
Py, ?Cc, Py,
Os, Pl, Tx Sp, Tx,
[X II, Sp I\/
IJ\
= Dromomeryx (Subdromomeryx),
Rk Rakomeryx. Craniocera-
tini are in boldface italics: Bb =
7.01;:,.,::--~-r-~-'!-~~-t~~-.t~~--.1~~--,l--~~+-~tyg~~c.....~~~~-¥.~
(ij 6. c
wE~
u.i ~ E
~
A?
Py, Py
/
/ II, /
' • / Pl I/ / Pl
Os, / ?Cc, Py, Pl /
~
1x Barbouromeryx

bouromeryx
(Barbouromeryx),
Bo = Bouromeryx, Bp = Bar·
(Probarbouromeryx),

I
Py ?Pl, Tx, / Sp
8.8J-~-:-t---::t-~~il/,:_~--1''/~::--f~~-,f-~~-+_::l'::!Jl::_-J<l/'...._~...J..~~!-~

95
C:c
.'!l ~ ·~
j (3 .g
I
I/
Py,
A?
Cc,
QQ, ex
?Do,
C
p~ /
/ Cc, Py,
QQ
Cc,
Co, /
/ D/?, Ix Bt = Barbouromeryx (Protobar-
bouromeryx), Cc = Cranioceras, Pc
= Procranioceras, Pp= Pediomeryx

w . r---~~.-~-;;x~71-~c-c~t--c~---t-10_0_.-c-~-f-c-c-.-Q_-o.-+--.~1-o~o-.~cc~.~-JL~-/--Jlt-.
I/ MY / Pl (Pediomeryx), Py = Pediomeryx
.:::'.1~~.
I/

(Yumaceras). D? = Dromomerycid

x /
z ~ (ij ·c; ' c, Co. Mv Ml;, .My, Cc
W tfl (3 .g Rm QQ ~.Pi E& Rm, QQ, Rm, / / / indet. Antilocapridae are in non-
g11.or-----:-~t---:7f'-~~+-~~t--~~~S~m~,P~l-*--.l~~P~a~,~Pl~~l/'...._~-¥l/'...._~~V_J
;;;;
.:: j2 B ~ ~
..J co ·;;:
Bo, ?Cc ~
/ R
Ex_ /
/
Cc,/
/ A?
Do, Cc, Pc,
QQ, Ml;, MY /
/ Do,
MY. ex I
I x boldface roman type. Merycodontinae
are in roman type underlined: Co =
Cosoryx, Mc = Meryceros, Mm =

~ .9
-' !!? c
u.i 01l ·~'
[X
I/ __!!! I/
1 2.51--:::--:-~t:'"~7t"~~-f'--~--f' / . Rm S Pl I/
--~--jL-~~f-.lJ-~
Cc
C MY
0
''
~·~m~·..'.:..'....-.,.IC.....~~~~-ll"__j
Do, Bo, _Mc,oM"
"'-'
Cc,
M c~
__i;, u!],
IX Do, Bo,
/.
/
/ ?Ji!1, Do, D? I
Rk, Bo, Do, Rk, /
gg, Mc, Pc, QQ, Mv Px
Merriamoceros, My = Merycodus,
Px = Paracosoryx, Rm = Ramoceros,
Sm= Submeryceros. Antilocaprinae
- "
1--:---r---r----1"---t===-t-=::__+-*~~~~~~---l"L....
c, Px, Rm Sm, Pl I~ MY. ex, Sm MY. ex,~ =· are in plain roman type: Cx =
14.0 ..l
'X
x
o / Do, Rk, Do Rk
SQ. Dd, Dm, Do, ?"- Do, Do Rk /
Capromeryx, Cr = Ceratomeryx,
~1D
- - c I Dm, Ds, '>C' ' ' ' QQ, Rk, Bo, ?Cc Rk,
..!ll!u ?Bo, Do, Rk, I Hb = Hexobelomeryx, Hm = Hex-
coco ·->"' ,
lll Rk, Bo ...~·
·= Ml;,PxMm, __i;,
M M ?C M Jl!!Y,
M

x
_m, 12
UJ . .Ji, -1!!. QQ, Mm, ameryx, II = llingoceros, Os =
15.8 \"-;-~it-~71'-~~-f-==---t__::=..--j~MY~.£Rgm~--4-S~m~,~P~x~,R~m:ll_...j_~~&Jex~~~_J'.V__j

'* t~
Osbomoceros, Ot = Ottoceros, Pl =
Rk / Do, Rk, Ds, Bo, ?A!, Si;, Do, Os, ?!!!1, Do, Rk / I
§ ~ / Plioceros, Pa= Proantilocapra, Sa=
Boift_Cc, Ml;. Sm Rk, Bb, Bo, Bt, Ds, Rk, / /
_,:i:- / ?~]2 Mm,.My,Sm,_E'x?Bo, A?I/ V Subantilocapra, Sp = Sphenophalos,
17·5 1~rn~c:f~~/f-~-t----i-~-f--~~'-4-~_::_.::j._.:..__jL_-¥..._J Tx = Texoceros, Tr = Tetrameryx.
~-~ ~
co a> ..
w:i:.E
/
/ ?Bo I / / Al
/ / MY IX Al. Bb,?Bo, Bp, D? / A? = antilocaprid indet. Cervidae are
in roman type italics: Br = Bretzia,
I/ I/ MY.ex ••
18.8!2a>-'L--j~-;t-~--i'-~7J<=----~:t-~~-f--*-~~~~~":__~-1-~-.Ji'~
Ol:ijc1X
-'-"'Ctl
..J ~ :Jl I.;
/ ,., I/
/
, . / D?
,., X Bb ,., [X Od = Odocoileus, C? = cervid indet.
Bovidae (?) are in italics underlined:
19.2 --=-~"-"- _ _,.__ _.."'----""--~--..J[_JL_ _ _J.,,.:::._--=--:....L_J L~ Ng= Neotragoceras.

and merycoidodontid oreodonts made their first appearance in the The start of the Miocene marked the turnover from the White
late Duchesnean. River chronofauna to the Runningwater chronofauna. By this time
A distinctive change is apparent by the Chadronian (late Eocene), leptochoerids and agriochoerid oreodonts were extinct, although
with the inception of the White Riverchronofauna that lasted through the merycoidodontids remained diverse. Entelodonts and anthra-
the Oligocene, marked by the extinction of most of the primi- cotheres were each reduced to a single, specialized genus. Tragu-
tive artiodactyl families (with the exception of the leptochoerids), loid diversity decreased while camelid and protoceratid diversity in-
the diversification of the earlier late Eocene immigrants (although creased. The early Miocene was also the time of the first appearance
oromerycids did not survive into the Oligocene), and the great pro- of immigrant pecoran ruminants: blastomerycines, dromomerycids,
liferation of the merycoidodontid oreodonts. Tayassuids were the and merycodontine antilocaprids.
only new immigrants. The Clarendonian chronofauna of the mid-Miocene began in the
 Christine M. Janis et. al 356

late Hemingfordian, at the end of the early Miocene. By this time, (1885). Osteology of the Mammalia (3rd ed.). London: Macmillan.
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North
entelodonts, anthracotheres, and traguloids (with the exception of a
America. Smithsonian Miscellaneous Collections, 128, 1-96.
single genus) were extinct. The diversity of merycoidodontids and
Gentry, A. W. ( 1994). The Miocene differentiation Old World Pecora (Mam-
blastomerycines was reduced. Tayassuids and protoceratids both malia). Historical Biology, 7, 115-58.
sustained a modest diversity, and there was a marked increase in the Gentry, A. W., & Hooker, J. (1988). The phylogeny of the Artiodactyla. In
diversity of camelids, dromomerycids, and merycodontines, espe- The Phylogeny and Classification of the Tetrapods, Vol. II: Mam-
cially during the late early to early middle Miocene (late Hem- mals, ed. M. J. Benton, pp. 235-72. Systematics Association Special
Volume no. 35B. Oxford: Clarendon.
ingfordian to early Barstovian). Antilocaprine antilocaprids and
Ginsburg, L. (1985). Essai de phylogenie des Eupecora (Ruminantia, Ar-
the immigrant Pseudoceras were in evidence by the Clarendonian tiodactyla, Mammalia). Compte Rendu Hebdomadaire Seances de
(early late Miocene), by which time there was a decline in the l'Academie des Sciences, Paris, series D, 262, 979-82.
diversity of protoceratids, noncameline camelids, dromomerycids, Ginsburg, L., & Heintz, E. (1966). Sur !es affinites du genre Palaeomeryx
(Ruminant du Miocene europeen). Compte Rendu Hebdomadaire
and merycodontines. Nontayassuine tayassuids were extinct by the
Seances de l'Academie des Sciences, Paris, 301, 1255-7.
end of the Barstovian, and blastomerycines were extinct by the end
Gray, J. E. (1821 ). On the natural arrangement of vertebrose animals. London
of the Clarendonian. Medical Repository, 15, 296-310.
By the latest Miocene (Hemphillian), the fauna of North America Groves, C. P., & Grubb, P. (1987). Relationships of living deer. In The
took on a more modem aspect, with a reduction in diversity in most Biology andManagment of the Cervidae, ed. C. Wemmer, pp. 21-59.
groups, especially among the browsers. In the early part of this Washington, D.C.: Smithsonian Institution Press.
Hamilton, W.R. (1978). Fossil giraffes from the Miocene of Africa, and a
Mio-Pliocene chronofauna, there was continued diversity among
revision of the phylogeny of the Giraffoidea. Philosophical Trans-
tayassuine tayassuids, cameline camelids, and antilocaprine antilo- actions of the Royal Society (London), Series B, 282, 165-229.
caprids, with the appearance of the immigrant bovid Neotragoceras. Janis, C. M. (1987). Grades and clades in hornless ruminant evolution: the
By the end of the Hemphillian (early Pliocene), the following taxa reality of the Gelocidae and the systematic position of Lophiomeryx
were extinct, or made their final appearance: merycoidodontids, pro- and Bachitherium. Journal of Vertebrate Paleontology, 7, 200-16.
(1989). A climatic explanation for patterns of evolutionary diversity in
toceratids, noncameline camelids, dromomerycids, merycodontine
ungulate mammals. Palaeontology, 32, 463-81.
antilocaprids, and Pseudoceras. Blancan faunas are characterized (1993). Tertiary mammal evolution in the context of changing climates,
by a reduced diversity among the antilocaprids and tayassuids, the vegetation, and tectonic events. Annual Review of Ecology and Sys-
appearance of large, specialized camelids, and a modest radiation tematics, 24, 467-500.
of immigrant cervids. Janis, C. M., Gordon, I. J., & Illius, A. W. (1994). Modelling equid/ruminant
competition in the fossil record. Historical Biology, 8, 15-29.
Faunas from the Gulf Coast were virtually unknown before the
Janis, C. M., & Scott, K. M. (1987). The interrelationships of higher ruminant
Miocene, but from their initial appearance appear to represent areas families with special emphasis on the members of the Cervoidea.
of endemism. Later Miocene Gulf Coast faunas in part may represent American Museum Novitates, 2893, 1-85.
a refuge for taxa previously having a more northern distribution Joeckel, R. M., & Stavas, J.M. (1996). Basicranial anatomy of Syndyoceras
(for example, the retained diversity of antilocaprids in the latest cooki (Artiodactyla, Protoceratidae) and the need for a reappraisal
of tylopod relationships. Journal of Vertebrate Paleontology, 16,
Hemphillian), but they persistently contain an endemic component
320-7.
throughout the Neogene. Kowalewsky, W. (1883). Monographie der Gattung Anchitherium Cov.
und Versuch einer naturlichen Classification der fossilen Hufthiere.
Palaeontolographica, n. s. II, 3, 1873-1874, 133-285.
REFERENCES Kraus, F., & Miyamoto, M. M. (1991). Rapid cladogenesis among the pec-
oran ruminants: evidence from mitochondrial DNA sequences. Sys-
Ahearn, M. E. (1994). Cranial morphology and evolutionary relationships tematic Zoology, 40, 117-30.
among antilocaprids and other ruminants. Journal of Morphology, Leinders, J. J. M. (1979). On the osteology and function of the digits
220, supplement to ICVM-4 abstracts. in some ruminants and their bearing on taxonomy. Zeitschrift fiir
Bubenik, A. B. (1982). Taxonomy of Pecora in relation to morphophy- Siiugertierkunde, 4, 305-18.
siology of their cranial appendages. In Antler Development in the (1983). Hoplitomerycidae fam. nov. (Ruminantia, Mammalia), from
Cervidae, ed. R. D. Brown, pp. 163-86. Kingsville, Tex.: Caesar Neogene fissure fillings in Gargano (Italy). Scripta Geologica, 70,
Kleberg Wildlife Research Institute. 1-51.
Carroll, R. L. (1987). Vertebrate Paleontology and Evolution. New York: Leinders, J. J. M., & Heintz, E. (1980). The configuration of the lacrimal
Freeman. orifice in pecorans and tragulids (Artiodactyla; Mammalia) and its
Cifelli (1982). Patterns of evolution among Artiodactyla and Perissodactyla significance for the distinction between Bovidae and Cervidae. Beau-
(Mammalia). Evolution, 35, 433-40. fortia, 30, 155-60.
Colbert, E. H. (1935). Distributional and phylogenetic studies on Indian Matthew, W. D. (1904). A complete skeleton of Merycodus. Bulletin of the
fossil mammals, IV: The phylogeny of the Indian Suidae and the American Museum of Natural History, 20, 101-29.
origin of the Hippopotamidae. American Museum Novitates, 799, (1908). Osteology of Blastomeryx and phylogeny of the American
1-24. Cervidae. Bulletin of the American Museum of Natural History, 24,
(1941). The osteology and relationships of Archaeomeryx, an ancestral 535-62.
ruminant. American Museum Novitates, 1135, 1-24. (1929). Reclassification of the artiodactyl families. Bulletin of the Geo-
Flower, W. H. (1883). On the arrangement of the orders and families of ex- logical Society of America, 40, 403-8.
isting Mammalia. Proceedings of the Zoological Society of London, (1934). A phylogenetic chart of the Artiodactyla. Journal ofMammalogy,
1883, 178-86. 15, 207-9.
 Artiodactyla 357

Miyamoto, M. M., Kraus, F., Laipis, P. J., Tanhauser, S. M., & Webb, Savage, D. E., & Russell, D. E. (1983). Mammalian Paleofaunas of the
S. D. (1993). Mitochondrial DNA phylogenies within Artiodactyla. World. Reading, Mass.: Addison-Wesley.
In Mammal Phylogeny: Placentals, eds. F. S. Szalay, M. J. No- Schaeffer, B. (1948). The origin of a mammalian ordinal character. Evolu-
vacek, & M. C. McKenna, pp. 268-82. New York: Springer- tion, 2, 163-75.
Verlag. Scott, K. M., & Janis, C. M. (1993). Relationships of the Ruminantia
Nowak, R. M. (1991). Walker's Mammals of the World. Baltimore: Johns (Artiodactyla) and an analysis of the characters used in ruminant
Hopkins University Press. taxonomy. In Mammal Phylogeny: Placentals, eds. F. S. Szalay,
O'Gara, B. W., & Matson, G. (1975). Growth and casting of horns by M. J. Novacek, & M. C. McKenna, pp. 282-302. New York: Springer-
pronghorns and exfoliation of horns by bovids. Journal of Mam- Verlag.
malogy, 56, 829-46. Scott, W. B. (1940). The mammalian fauna of the White River Oligocene,
Owen, R. (1848). Description of teeth and portions of jaws of two extinct An- Part 4: Artiodactyla. Transactions of the American Philosophical
thracotherioid quadrupeds (Hyopotamys vectianus and Hyop. bovi- Society, 28, 363-746.
nus) discovered by the Marchioness of Hastings in the Eocene de- Simpson, G. G. (1945). The principles of classification and a classification of
posits on the N.W. coast of the Isle of Wight: with an attempt to the mammals. Bulletin of the American Museum of Natural History,
develope Cuvier's idea of the classification of pachyderms by the 85, 1-350.
number of their toes. Quarterly Journal of the Geological Society of ( 1953). The Major Features of Evolution. New York: Columbia Univer-
London,4, 103-41. sity Press.
Patton, T. H., & Taylor, B. E. (1973). The Protoceratinae (Mammalia, Stehlin, H. G. (1910). Die Saugetiere des schweizerischen Eocaens. Abhand-
TY lopoda, Protoceratidae) and the systematics of the Protocerati- lungen Schweizerischen Paliieontologischen Gesellschaft, Basal, 36,
dae. Bulletin of the American Museum of Natural History, 150, 839-1164.
347-414. Stirton, R. A. (1944). Comments on the relationships of the cervoid family
Pickford, M. (1983). On the origins of the Hippopotamidae together with a Palaeomerycidae. American Journal of Science, 242, 633-55.
description of two new species, a new genus and a new subfamily (1967). Relationships of the protoceratid artiodactyls, and a description
from the Miocene of Kenya. Geobios, 16, 133-54. of a new genus. University of California Publications in Geological
( 1993). Old World suiod systematics, phylogeny, biogeography and bios- Science, 72, 1-28.
tratigraphy. Paleontologia i Evoluci6, 26-27, 237-69. Taylor, B. E., & Webb, S. D. (1976). Miocene Leptomerycidae (Artiodactyla,
Pilgrim, G. E. (1941). The dispersal of the Artiodactyla. Biological Reviews, Ruminantia) and their relationships. American Museum Novitates,
16, 155-8. 2596, 1-22.
Prothero, D.R. (1994). Mammal Evolution. In Major Features of Vertebrate Webb, S. D., & Opdyke, N. D. (1995). Global climatic influences on Ceno-
Evolution, eds. D. R. Prothero & R. M. Schoch, pp. 238-70. Short zoic land mammal faunas. In Effects of Past Global Change on Life,
Courses in Paleontology, no. 7. Knoxville: The University of Ten- pp. 184-208. Board of Earth Sciences and Resources Commission
nessee and the Paleontological Society. on Geosciences, Environment, and Resources. National Research
Prothero, D.R., Manning, E. M., & Fischer, M. (1988). The phylogeny of the Council. Washington, D.C.: National Academy Press.
ungulates. In The Phylogeny and Classification of the Tetrapods, Vol. Webb, S. D., & Taylor, B. E. (1980). The phylogeny of hornless ruminants
II: Mammals, ed. M. J. Benton, pp. 201-34. Systematics Association and a description of the cranium of Archaeomeryx. Bulletin of the
Special Volume no. 35B. Oxford: Clarendon. American Museum of Natural History, 167, 117-58.
Romer, A. S. (1966). Vertebrate Paleontology (3rd ed.). Chicago: University Zittel, K. A. (1925). Textbook of Palaeontology, Vol. 3: Mammalia, trans.
of Chicago Press. C. R. Eastman & A. S. Woodward. London: Macmillan.
23 Eocene bunodont and bunoselenodont Artiodactyla ("dichobunids")

RICHARD K. STUCKY

INTRODUCTION

The Eocene bunodont artiodactyls of North America have classi-

cally been placed in the family Dichobunidae, which is based on
the European genus Dichobune. Inclusion of the North American
taxa with these European bunodont artiodactyls results in a para-
phyletic family with genera that are more closely related to other
families than they are to taxa within the "dichobunids." Gentry and
Hooker (1988) most recently have analyzed the phylogenetic re-
lationships of the artiodactyls as a whole and examined a partial
subset of the North American bunodont and bunoselenodont forms
considered here. The present analysis differs from that of Gentry and
Hooker (1988) in that no special relationship is recognized between
the helohyines and the anthracotheres and Gobiohyus, and their
family Bunomerycidae is considered paraphyletic with some taxa
probably more closely allied to the tylopods (Bunomeryx, Hylom-
eryx, and Mytonomeryx) and others to the ruminants (Mesomeryx
and Pentacemylus). Much research and comparison is still required
to establish the basic relationships of North American, Asian, and
European taxa of bunodont artiodactyls.
The phylogenetic relationships discussed here are based princi-
pally on the dentition, but all taxa discussed are placed within the
Artiodactyla because of the possession of a double pulley astra- Figure 23.1. Restoration of Diacodexis, by Brian Regal.
galus. Analysis of the postcranial (Rose, 1982, 1985) and basicra-
nial (Coombs and Coombs, 1982) morphology will greatly assist
in clarifying the relationships as more complete specimens become
available.
The artiodactyls immigrated to North America at the onset of
the Eocene, and indeed their first appearance marks the begin-
ning of the Wasatchian Land Mammal age. Potential migration of
three clades, Diacodexis, Bunophorus, and homacodonts, occurred
between North America and Europe in the early Wasatchian. It
has long been hypothesized that after the early Wasatchian, fau-
nas from the two continents evolved endemically. However, the
degree of similarity among North American homacodonts and Eu- Figure 23.2. Skull of Homacodon vagans (modified from Sinclair, 1914).
ropean dichobunids appears high, and either represents remarkable Scale= 1 cm. (Drawing by Brian Regal.)

358
 Eocene bunodont and bunoselenodont Artiodactyla 359

degrees of convergent evolution and homoplasy or several episodes The forms considered here have witnessed a complex history.
of immigration between Europe and North America that is not re- The first North American form to be described was Microsus cus-
flected in other early Eocene mammals. Migration between Asia pidatus from the Bridger Formation of Wyoming (Leidy, 1870).
and North America is more continuous during the Eocene, based Critical references include Marsh (1894), Scott (1898, 1899, 1940),
especially on the first appearances of relatively derived groups in Wortman (1898), Sinclair (1914), Peterson (1919), Gazin (1955,
North America such as the rodents, rabbits, tillodonts, and perisso- 1956), McKenna(l959), Edwards (1976), Golz(l976), Coombs and
dactyls. There do not appear to be any artiodactyls considered here Coombs (1977a, 1977b), and the more recent treatments by Storer
that were part of the immigration to North America. (1984), Krishtalka and Stucky (1985, 1986), Gentry and Hooker
These early artiodactyls were generalized in habit. Most focused (1988), Stucky and Krishtalka (1990), and Westgate (1994).
on the utilization of plant resources from fruits and seeds to leafy
vegetation. Some of the smaller forms may have included some
animal protein (insects) in their diet. During the Wasatchian and DEFINING FEATURES OF BUNODONT AND
Bridgerian, the artiodactyls were represented by only a small number BUNOSELENODONT ARTIODACTYLA
of species in local communities that were usually not particularly
abundant. During the Bridgerian, the artiodactyls were moderately Features that characterize the basal artiodactyls and are thus primi-
diverse but were scarce in abundance. The homacodonts probably tive for the order Artiodactyla as a whole include the foll-
include the basal members of both tylopods and ruminants, which by owing. On the lower molars, the trigonid is open with the para-
the beginning of the Uintan were the most abundant medium-sized conid separated from the metaconid, becoming progressively less
mammals in those faunas. separated from ml-3. The paraconid is slightly smaller than the

-
E '
, .
.

Figure 23.3. Dentition of Eocene bunodont and bunoselenodont artiodactyls. A. Diacodexis secans (modified from Sinclair, 1914). B. Lepto-
choerus sp. (modified from Marsh, 1894). C. Antiacodon pygmaeus (modified from Sinclair, 1914). D. Homacodon vagans (modified from
Marsh, 1894). E. Helohyus (uppers, H. milleri, lowers, H. plicodon) (modified from Sinclair, 1914). F. Hexacodus pelodes (drawing after A. Redline). Scale
= 5 mm for A-E; I mm for F. (Drawings by Brian Regal.)
 Richard K. Stucky 360

metaconid, and the metaconid shows slight basal inflation. The is square on Ml and M2 with a posterior cingulum that may bear
hypoconulid is positioned on the posterior talonid rim on ml-2, and an incipient hypocone or slightly elevated cingular ridge. A double
m3 bears a simple conical hypoconulid on the posterior lobe. The pulley astragalus is present.
cristid obliqua strikes the postvallid low on the lingual side of the
protoconid buccal to the notch between protoconid and metaconid.
The postcingulum is a small ridge midway between the talonid cusp SYSTEMATICS
apices and the dento-enamel boundary; there are no crests connect-
ing from the cingulum to the hypoconulid or other talonid cusps. The
SUPRAORDINAL
p4 is simple, not robust, with a paraconid, protoconid, and a talonid
ridge. The p3 is simple and slightly longer or equal in length to p4. The interrelationships of the order Artiodactyla within other ungu-
No diastemata separate the premolars. Both upper and lower canines late mammals is discussed in Chapter 22, the summary chapter on
are caniniform. Upper and lower molars show a basic tribosphenic Artiodactyla.
pattern. The upper molars are symmetrical with a central protocone
and equal-sized paracone and metacone. Conules on upper molars
are very distinct, with well-developed pre- and postcristae. An in-
INFRAORDINAL
cipient mesostyle may be present and the lingual margin of the tooth The North American Eocene bunodont and bunoseleno-dont artio-
dactyls are placed into two monophyletic groups, Leptochoerinae
and Antiacodontinae; one probable monophyletic group,
Helohyinae; and two paraphyletic groups, the homacodonts and
Diacodexis (in part). The Linnean subfamily rank of the three mono-
phyletic clades is for convenience only and does not imply any spe-
cial phylogenetic level of significance.
The paraphyletic grouping that includes Diacodexis also includes
the genera Simpsonodus and Bunophorus ( = Wasatchia). Diacodexis
itself is paraphyletic, especially the species D. secans, and has been
thoroughly discussed by Krishtalka and Stucky (1985); future work
will require this genus to be subdivided. The features character-
Figure 23.4. Skeleton of Diacodexis metsiacus (modified from Rose, 1982). izing artiodactyls in general are the ones that characterize "Dia-
Scale = 5 cm. (Drawing by Brian Regal.) codexis" secans-primus, the earliest known lineage segment in the

Figure 23.5. Cladogram showing the

basic relationships of principal groups
of early artiodactyls. Key to charac-
ters at nodes: (1) Double pulley as-
tragalus, paraconid progressively less
separated from metaconid, symmetri-
cal upper molars with distinct conules.
(2) Molar cusps inflated, trend to-
ward hyperinflation of P4/p4 and
Ml/ml. (3) Upper and lower molar
crests well developed, hypoconulid on
postcingulum but connected to crest
from hyperlophid. (4) Cristid obli-
qua rises and connects with apex of
metaconid. (5) Isolated hypoconulid
on postcingulum isolated entoconid,
metaconule expanded and proto-
cone anterior of midline on Ml-2.
(6) Hypocone developed on Ml.
(7) Loss of hypocone and grea-
tly expanded metaconule on Ml,
well-developed centrocristae with-
out mesostyle. (8) Robust molar
crowns. Leptochoerinae includes Lep-
tochoerus, lbarus, Stibarus, Laredo-
choerus, and unnamed taxa. Antiacodontinae includes Antiacodon, Sarcolemur, Auxontodon, Neodiacodexis, Tapochoerus, and possibly Texodon (see
text for discussion). Helohyinae includes Helohyus and a new genus, and tentatively Achaenodon, Discritochoerus, and Parahyus. "Homacodonts"
include Hexacodus, Homacodon, Microsus, Texodon, Hylomeyx, Mytonomeryx, Bunomeryx, Apriculus, Pentacemylus, and Mesomeryx. Position of
the Suina is not shown because of the probable paraphyly (or polyphyly) of this taxon.
 Eocene bunodont and bunoselenodont Artiodactyla 361

species Diacodexis secans (Krishtalka and Stucky, 1985), and "Di- Characteristics: Metaconid moderately inflated; paraconid
acodexis" ilicus (Gingerich, 1989) among North American Eocene generally distinct but reduced especially on m2-3; hypo-
Artiodactyla. conulid on posterior ridge of talonid between entoconid
The homacodonts are essentially equivalent to the Bunomeryci- and hypoconid and not on postcingulum; upper molars tri-
dae (Gentry and Hooker, 1988), plus basal members of the Bun- angular with a central protocone, generally lacking hypo-
odontia (Gentry and Hooker, 1988) and the Neoselenodontia (Webb cone and mesostyle; paraconule, metaconule, and conular
and Taylor, 1980; Janis and Scott, 1987). The homacodonts of the wings generally well developed; protocone, paracone, and
Wasatchian (Hexacodus), Bridgerian (Microsus, Homacodon), and metacone bulbous and distinct. Trend toward elongation of
Uintan (Hylomeryx, Mytonomeryx, Mesomeryx, Bunomeryx, Pen- P3/3 relative to P4/4, bunodonty of molars and premolars,
tacemylus, Apriculus), as well as the members of the Helohyinae and inflation of metaconid.
(Helohyus, Parahyus, and Achaenodon), are all part of the basal ra- Average length ofm2: 3.7-5.3 mm.
diation of the selenodont artiodactyls. Texodon of the Texas Uintan Included species: D. secans (known from localities SB22A,
is tentatively included here, although it may prove to be classified CP5A,CP20A,CP208,CP20C,CP20D,CP20E,CP23A,
within the antiacodonts with the recovery of more complete mate- CP24C,CP258,CP25C,CP25D,CP25F,CP25G,CP25H,
rial, especially the lower dentition. The principal features uniting CP26D, CP27 A, CP27B, CP27C, CP27D, CP27E, CP28A,
these taxa are an expanded, posterolingually displaced metaconule, CP28C, CP63, NP49); D. gracilis (localities CC50,
anterior protocone, and the location of the hypoconulid on the pos- CP20A); D. minutus (localities SB228, CP27C, CP27D,
terior cingulum. In the early selenodont forms, the entoconid is CP27E); D. ilicus (locality CP19); D. sp. A (Krishtalka
isolated and enlarged. The homacodonts are paraphyletic and it is and Stucky, 1986) (localities SB24, CP64A, CP64B); D.
quite likely that different genera of homacodonts will prove to be sp. 8. (Krishtalka and Stucky, 1986) (locality SB24).
more closely allied to different taxa of more typical selenodont artio-
dactyls such as the agriochoerids, oromerycids, hypertragulids, and Comments: Krishtalka and Stucky (1985, 1986) have most
protoceratids (Webb and Taylor, 1980). Among the homacodonts, recently reviewed Diacodexis. They examined in detail
Mesomeryx and Pentacemylus appear to be related to the ruminants the morphological variation apparent in the genus and de-
(lack of a mesostyle and hypocone, greatly expanded metaconule), termined that Diacodexis is the stem artiodactyl. As cur-
whereas others (Hylomeryx, Mytonomeryx, and Bunomeryx) may rently defined, both the genus and the type species, D.
be more closely related to the tylopods (presence of mesostyle and secans, are paraphyletic. The difficulty arises in determin-
stylar cusps connecting to the paracone and metacone). ing the limits of variation for individual species. Samples
of Diacodexis from the same horizon often exhibit vari-
able characteristics that are suggestive of different, later,
INCLUDED NORTH AMERICAN species, and even different monophyletic groups of artio-
GENERA OF BUNODONT AND dactyls, such as the antiacodonts, while still seemingly part
BUNOSELENODONT ARTIODACTYLA of the same lithosympatric population from a single fos-
siliferous horizon.
The locality numbers listed for each genus refer to the list of unified Diacodexis secans is considered to be a lineage that ex-
localities in Appendix I. The acronyms for musuem collections are isted in western North America from the early Wasatchian
listed in Appendix III. to the earliest Bridgerian (Gardnerbuttean). Members of
The locality numbers may be listed in a couple of alternative ways. the early populations of this lineage have relatively gracile
Parentheses around the locality (e.g., [CPlOl]) mean the taxon in lower molars that possess prominent paraconids and meta-
question at that locality is cited as an "aff." or "cf." the taxon in conids, a p3 that is slightly longer than or equal in length
question. Parentheses are usually used for individual species, thus to p4, and upper molars with a central protocone, strong
implying the genus is firmly known from the locality, but the actual conules, and well-developed postcingulum. The trend in
species identification may be questionable. Question marks in front this lineage is toward elongation of p3, more robust cusps,
of the locality (e.g., ?CPlOl) mean the taxon is questionably known metaconid inflation, and more bunodont molars. Krishtalka
from that locality, thus implying some doubt that the tax on is actually and Stucky (1985) divided D. secans into four lineage seg-
present at that locality, either at the genus or the species level. ments to highlight the "average" morphologies through
time in this apparently continuous lineage. D. secans-met-
siacus, D. secans-kelleyi, and D. secans-secans show elon-
ARTIODACTYLA INCERTAE SEDIS gation of p3 and greater bunodonty and robusticity of the
molars overall that indicate affinities with the leptocho-
Diacodexis Cope, 1882 (including Trigonolestes
erines. "Diacodexis" woltonensis shows this pattern with
Cope, 1894)
greater inflation of the p4-m3 and metaconid on ml-3 that
Type species: Diacodexis secans (Cope, 1881) ( = Pantolestes warrants this species to be referred to a new genus within
secans). the Leptochoerinae, a taxonomic necessity that is beyond
Type specimen: AMNH 4899. the scope of this chapter (see later).
 Richard K. Stucky 362

Several specimens within D. secans-kelleyi display a CP20A, CP20B, CP64A, CP64B, CP64C).
cristid obliqua that rises to and connects with the meta- Simpsonodus sp. is also known from localities CP20A,
conid, a feature diagnostic of antiacodonts. It should be CP20B, CP64B.
stressed, however, that the sample from which these spec- Comments: Simpsonodus has long been confused with Dia-
imens come displays variation from this character to that codexis. In penecontemporaneous samples, Simpsonodus
more typical of Diacodexis where the cristid obliqua strikes may be distinguished by its larger size, lesser inflation of
the postvallid low and medially. Penecontemporaneous the molarmetaconids, and square upper molar construction.
samples of Diacodexis from the Piceance Creek and San
Juan basins display a closely appressed and elevated para- Bunophorus Sinclair, 1914 (including Wasatchia
conid and metaconid, a relatively strong postcingulum, a Sinclair, 1914)
very slightly expanded metaconule, and presence of an
Type species: Bunophorus etsagicus (Cope 1882) (=Mio-
incipient mesostyle, features typical of Hexacodus, the
claenus etsagicus).
stem selenodont artiodactyl. Within Diacodexis minutus,
Type specimen: AMNH 4698 (lp3-m3, rp3-m3).
the presence or absence of a metaconid on p4 is variable,
Characteristics:
as is the relative expression of the paraconid on the mo-
lars. It appears that features which define higher-level taxa P3/3 more robust than in [Diacodexis and Simpsonodus],
later in time are quite variable within different samples with proportionately more inflated P4/4 and molar cusps and
(populations) of Diacodexis at the point of the feature's conules, p4 paraconid more reduced, P4 lacking metacone,
first appearance. Detailed study of these features and their Ml-2 more nearly square and conular wings reduced or ab-
variation is at the heart of understanding the evolutionary sent; included species larger than [species of Diacodexis and
process. Simpsonodus chacensis] except Simpsonodus sp. (Krishtalka
Diacodexis is also known from Europe and has been and Stucky, 1986) and Diacodexis secans (only some speci-
described from Asia, including the species D. gazini mens in lineage segment Diacodexis s. ? -secans; Krishtalka
(Godinot, 1978; Europe), D. varleti (Sudre et al., 1983; and Stucky, 1985). (Description of subfamily Bunophorinae,
Europe), D. antunesi (Estravis and Russell, 1989; Europe), Stucky and Krishtalka, 1990, pp. 150-51.)
and "D." pakistanensis (Thewissen et al., 1983; Asia).
Average length of m2: 5.8-9.0 mm.
Given the large number of differences between North
Included species: B. etsagicus (known from localities CP20B,
American Diacodexis and "D." pakistanensis (cited by
CP20C, CP27B, CP28A); B. macropternus (localities
Thewissen et al., 1983), inclusion of the latter species
CP20C, CP20D, CP25E, CP27B); B. sinclairi (localities
within Diacodexis is not acceptable. "Homacodon" priscus
CP20E, CP25F, CP25G, CP27D, CP27E, CP64C); B.
Marsh, 1894 and "Homacodon" pucillus Marsh, 1894, both
grangeri (including Antiacodon crassus; Wasatchia
represented by astragali from the San Jose Formation, San
dorseyana; Wasatchia lysitensis) (localities SB24, CP20B,
Juan Basin, New Mexico, probably represent Diacodexis
CP20C, CP27B, CP64A); B. pattersoni (localities SB22II,
based on their size (Lucas, 1983).
CP64B, CP64C); B. robustus (localities CP20A, CP63).

Simpsonodus Krishtalka and Stucky, 1986 Comments: The principal features of Bunophorus are the bun-
odont molars and the simple, but inflated p4, some spec-
Type species: Simpsonodus chacensis (Cope, 1875b) (=Pan-
imens of which may bear a metaconid (in B. sinclairi).
tolestes chacensis).
Enigmatically, the p4 of Hexacodus pelodes, Hylomeryx,
Type specimen: Original holotype lost (USNM collections,
and Bunophorus robustus display an inflated p4. It is likely
fide Matthew, 1899); neotype, AMNH 48694 (Krishtalka
that p4 inflation and the presence of a metaconid are con-
and Stucky, 1986).
vergent.
Characteristics:
B. sinclairi has been divided into two subspecies: B. s.
Differs from penecontemporaneous Diacodexis in its larger sinclairi (including B. gazini) and B. s. robinsoni (Stucky
size, slightly elevated molartrigonid, and proportionately less and Krishtalka, 1990). B. cappettai from Europe includes
expanded metaconid (except Diacodexis cf. Diacodexis se- what probably constitutes several taxa among which the
cans [from the San Juan Basin, New Mexico]), relatively genus Bunophorus is probably represented (see Stucky and
longer talonid, and more quadrate upper molars. Differs from Krishtalka, 1990).
Bunophorus in having more gracile p4, larger p4 talonid
basin, non-bulbous molar cusps and crowns, more elevated LEPTOCHOERINAE (INCLUDING
molar trigonids, and deeper talonid basins. (Stucky and DIACODEXEINAE, IN PART)
Krishtalka, 1986, p. 186.)
Characteristics: molar cusps inflated, and P3/3-M3/3 robust. Trends
Average length of m2: 4.8-6.1 mm. toward inflation of the p4 and ml, and decreasing size of molars from
Included species: S. chacensis (known from localities SB24, ml to m3.
 Eocene bunodont and bunoselenodont Artiodactyla 363

Included genera: Leptochoerus, Stibarus, lbarus, Laredochoerus, Leptochoerus Leidy, 1856 (including Laopithecus
and two unnamed genera. Marsh, 1875)
Comments: Until recently, specimens of leptochoerines from the
Type species: Leptochoerus spectabilis Leidy, 1856.
middle and late Eocene (Bridgerian to Duchesneau) were unknown
Type specimen: ANSP 15593.
(Black, 1978; Storer, 1984). They are extremely rare during this
Characteristics:
time interval, and much more information is needed to clarify their
phylogenetic history. I follow Edwards (1976), Storer (1984), and
Skull with short rostrum. More massive, bunodont lower mo-
Westgate (1994) in their recent revisions ofleptochoerine taxa with lars than Stibarus, which do not have a subcrescentic appear-
data on distribution kindly provided by Christine Janis. The Lep- ance when little worn; paraconids lacking, p2, P2, and p3
tochoerinae are a monophyletic group. Part of the species Dia-
may be tall, single-cusped, or low tri-cusped, elongate teeth.
codexis secans should be included within this subfamily as discussed Ml-3 triangular but lingual edge is blunted-off. (Edwards,
earlier.
1976, p. 103.)

lbarus Storer, 1984 Average length of m2: unavailable. Average length of ml:
5.6-6.2mm.
Type species: lbarus ignotus Storer, 1984. Included species: L spectabilis (known from localities
Type specimen: SMNH Pl589.12. CP41B, CP68B, CP68C, CP68D, CP84A, CP84B); L.
Characteristics:
elegans (locality CP84A); L supremus (locality CP84B);
L emilyae (locality CP68C).
Cheek teeth low-crowned; P3/3-p4 elongate, narrow relative
Leptochoerus sp. is also known from localities CP40B,
to molars; molar series not massive as in Leptochoerus; M3/3
CP41A, CP41C, CP85A, CP85C, CP86B, CP99A, NP24D,
relatively unreduced; Ml-3 conules bulbous, not strongly
NP29D, NP50C.
crescentic; ml-3 hypoconulid, accessory cusp, entoconid
subequal, set in arc; m3 talonid shorter than in Oligocene Comments: Edwards's characterization of Leptochoerus is
taxa. (Storer, 1984, p. 76.) followed here with information on distribution provided
by Christine Janis.
Average length of m2: 4.4-4.6 mm.
Included species: /. ignotus only, known from locality NP8
"Diacodexis" woltonensis Krishtalka and Stucky, 1985
only.
Type species: Diacodexis woltonensis Krishtalka and Stucky,
Stibarus Cope, ,1873a (including Menotherium Cope, 1873b, 1985.
and Nanochoerus MacDonald, 1955) Type specimen: CM 43478; Paratypes: CM 22592, 40783,
42090-42092, 43474; AMNH 92871; UCM44423, 44696,
Type species: Stibarus obtusilobus Cope, l 873a. 46810.
Type specimen: AMNH 6784. Characteristics: Small genus of leptochoerine; differs from
Characteristics: other leptochoerine genera in having ml smaller than m2,
less inflated p4-m2, and relatively prominent paraconids
Skull with pinched rostrum. Massive, bunodont lower molars on the lower molars. Differs from Diacodexis in the nearly
which appear subcrescentic when little worn, with reduced square ml-2, inflation of p4-m2, hyperinflation of the
paraconids. The p2, P2, and p3 are low and tricuspid, the metaconid, and inflated protoconid and hypoconid; small
medial cusp being largest. Ml-3 are triangular with a sharp talonid basin; talonid notch well developed and trigonid
lingual apex. (Edwards, 1976, p. 100.) not compressed.
Average length of m2: 4.1-4.3 mm.
Average length of m2: unavailable. Average length of ml:
Included species: "D." woltonensis only (known from locali-
3.1-4.0 mm.
ties [SB22C], [SB43A], CP27D, CP27E, [CP32B],
Included species: S. obtusilobus (known from localities
[CP34B], [CP34C]).
CP41B, CP68B, CP68C, CP84A, CP98A, CP99A); S.
quadricuspis (localities CP40A, CP68C); ?S. montanus Comments: The specimens from the Huerfano Basin, Col-
(locality NP24C); S. yoderensis (locality CP42A). orado, and Bridger Formation of the New Fork-Big Sandy
Stibarus sp. is also known from localities CP39B, CP39C, area, Wyoming, probably represent a new species referable
CP39E, CP40B, CP41A, CP83B, CP83C, CP84A, CP84B, to the same genus as "D." woltonensis.
CP98C, CP99A, NP24D, NP50B, NP50C.
Laredochoerus Westgate, 1994
Comments: Edwards (1976) analysis of Stibarus is used here
with information on distribution provided by Christine Type species: Laredochoerus edwardsi Westgate, 1994.
Janis. Type specimen: TMM 42486-1130.
 Richard K. Stucky 364

Characteristics: Very robust and bunodont dentition. The p4 is A. vanvaleni (localities SB22C, CP20D, CP27D, CP27E);
robust, and lower dentition with progressive size reduction A. pygmaeus (including Nanomeryx caudatus) (localities
from p4 to m3. The m3 is about two thirds the length of SB22C, CP5A, CP25J, CP31E); Antiacodon undescribed
m 1 or m2. Small paraconid present on lower first or second species (Stucky and Krishtalka, in preparation) (CP20D,
molar. Talonid nearly equal to trigonid in length. M3 lacks ?CP27B).
expanded metaconule and is relatively bunodont.
Comments: Robinson ( 1966) has subdividedAntiacadon pyg-
Average length ofm2: 5.5 mm.
maeus into two subspecies: A. p. pygmaeus (Bridger For-
Included species: L. edwardsi only, known SB42A only.
mation, Wyoming) and A. p. huer:fanensis (Huerfano For-
mation, Colorado). As with other early artiodactyl species,
Indeterminate leptochoerines
specimens of Antiacodon pygmaeus are quite variable in
A taxon Leptochoerinae, gen. et sp. nov., is known from locality morphology. Of special note is the variation in the strength
CP29D. Indeterminate leptochoerines are known from localities of the hypocone, cuspule anterior to the protocone, and the
SB42, CP34B, CP34C, CP36B, NP9A. relative expression of the crests on the upper molars. Some
Comments: An undescribed genus ofleptochoerine is represented specimens show a diastema between p 1 and p2 foreshad-
by an isolated lower m2 and m3 from the Wood locality, Wagon owing the condition in Auxontodon. Samples of Bridgerian
Bed Formation, Wind River Basin, Wyoming (CP29D) (unpublished Antiacodon are relatively small with no more than a few
data; Black, 1978). Other specimens have not been examined by me specimens from each horizon represented. Future work
and are not referred to a specific tax on of Leptochoerinae. may necessitate the division of Antiacodon pygmaeus into
more than one species.
Comparison of the type dentary of Nanomeryx cauda-
ANTIACODONTINAE
tus with referred material of Antiacodon pygmaeus reveals
Characteristics: On the lower molars the cristid obliqua rises up the that both are virtually identical in size and in comparable
posterior wall of the postvallid and connects with the apex of the elements of the skeleton, most notably in the possession of
metaconid; temporal trend toward paraconid larger than metaconid; a diastema between pl and p2 on the dentary. Lucas (1983)
cl and pl approximately equal in size; p3 slightly longer or equal concluded that Nanomeryx caudatus was a valid species
in length to p4. Cusps on upper molars are generally conical, pro- of Homacodon, citing only the difference in size between
tocone central with tendency for strong protocone fold posteriorly, N. caudatus and H. vagans as a diagnostic feature. Homa-
strong well-developed conular wings. Hypocone developed on cin- codon species lack any diastema between p 1 and p2 and
gulum (incipient in Antiacodon vanvaleni), and an accessory cusp there appear to be no differences in the postcranial skele-
developed on cingulum anterior to protocone. tons between antiacodonts and Bridgerian homacodonts,
Included genera: Antiacodon, Sarcolemur, Tapochoerus, Aux- although the latter are poorly known (West, 1984; Rose,
ontodon, Neodiacodexis, and possibly Texodon. 1985). Nanomeryx is considered a junior synonym of An-
Comments: The characteristics cited here, especially the rise of tiacodon.
the cristid ob Ii qua to meet the apex of the metaconid, serve to define An undescribed species of Antiacodon from the Lost-
the antiacodontines as a monophyletic group. cabinian of the Bighorn Basin, Wyoming, represents the
earliest appearance of this genus. As discussed earlier,
Antiacodon Marsh, 1872b (including Nanomeryx Lysitean Diacodexis secans-kelleyi samples from the Wind
Marsh, 1894) River Basin, Wyoming, show the variable expression of the
cristid obliqua rising to meet the metaconid, a feature di-
Type species: Antiacodon venustus Marsh, 1872b.
agnostic of all antiacodonts.
Type specimen: YPM 11765.
Characteristics: Bridgerian Antiacodon are characterized by
"upper molars with well-developed hypocone on Ml and Tapochoerus McKenna, 1959
M2 which may be double, a small cusp anterior to the Type species: Tapochoerus egressus (Stock, 1934) (= Hyop-
protocone, and a small but distinct mesostyle. Conules are sodus egressus).
prominent. The p4 has a metaconid. Lower molars with Type specimen: LACM (CIT) 1590 (p4-m3).
metaconid reduced and paraconid and metaconid closely Characteristics:
appressed but nonetheless distinct," according to West
(1984, p. 6). In addition, on Bridgerian forms the para- Primitive dichobunid artiodactyl in which the teeth are more
conid is larger than the metaconid. In Lostcabinian and bunodont than selenodont; the hypocone is retained on Ml
Gardnerbuttean Antiacodon the paraconid is equal in size and/or M2, the metaconules are large but are neither hyper-
to the metaconid, and the mesostyle and hypocone are in- trophied nor markedly displaced and, when unworn, possess
cipient to absent. five posterior crests; there is neither a mesostyle nor ectoloph
Average length of m2: 4.1-5.5 mm. ribbing on Ml or M2, strong diastemata isolate the double
Included species: A. venustus (known from locality CP34D); rooted p2; p3 is trenchant and lacks a metaconid; p4 is also
 Eocene bunodont and bunoselenodont Artiodactyla 365

trenchant but possesses a distinct entoconid and a metaconid located on the posterior cingulum. The referred upper mo-
which separates from the protoconid high on the crown; ml lars have a continuous cingulum and the cusps are cres-
has a strong paraconid separated from the metaconid but centic and nearly equal in size and height. The proto-
not widely separated as in Antiacodon and Auxontodon, m2 cone is central, and there are small anterior and posterior
has a variably separated paraconid, and m3 has a fused sin- cuspules on cingulum. The labial margin of the tooth
gle lingual trigonid cusp; all the lower molars are elongate, has strong stylar ridges and the mesostyle is high and
with hypoconulids closely similar to those of Hexacodus. strongly connected to the paracone and metacone. (Black,
(McKenna, 1959, p. 127.) 1978.)
Average length ofm2: 6.7-7.l mm.
Average length ofm2: 7.4-8.0 mm.
Included species: A. pattersoni (known from localities CP6B,
Included species: T. egressus only, known from locality CC9A
CP29C); A. processus Storer, 1984 (locality NP8).
only.
Questionable specimens of Auxontodon sp. are also known
Comments: McKenna (1959) allied Tapochoerus with Mi- from localities SB43B, NP13.
crosus. Additional material recovered since McKenna's
Comments: A specimen from the Devil's Graveyard For-
(1959) excellent research on Hexacodus, Microsus, and
mation, Brewster County, Texas (locality SB43B), was
antiacodonts strongly suggests that Tapochoerus should
originally referred to Texodon meridianus (West, 1984).
be allied with the latter group. The distinctive feature of
The well-defined metaconule and paraconule and strong
all antiacodonts is the cristid obliqua, which rises to the
mesostyle, as well as a central protocone on this DP4, sug-
apex of the metaconid. The larger paraconid than meta-
gest affinities with antiacodonts and probably Auxontodon.
conid, close appression of these two cusps, and presence
A specimen of an upper molar from the Wagon Bed For-
of a metaconid on p4 suggests affinities of Tapochoerus
mation, Wyoming (Black, 1978), shows strong convergent
with Bridgerian Antiacodon. Tapochoerus is the largest
crescentic cusps and conules that are reminiscent of those
and latest known antiacodont.
in Pentacemylus and other derived "homacodonts." How-
ever, the lack of anteriorly placed protocone and enlarged
Sarcolemur Cope, 1875a
metaconule precludes reference to that group.
Type species: Sarcolemur furcatus Cope, l 875a.
Type specimen: AMNH 5008 (p4-m3). Neodiacodexis Atkins, 1970 (in West and Atkins, 1970)
Characteristics: Size identical to Antiacodon pygmaeus, but
the hypoconulid is developed on the postcingulum, with a Type species: Neodiacodexis emryi West and Atkins, 1970.
strong,crest between the hypoconid and entoconid, conver- Type specimen: AMNH 56054 (P4-M2).
gent with these features in homacodonts. The paraconid is Characteristics:
larger than the metaconid and closely appressed to it, and
P4 with two large sharp cusps; Ml and M2 tritubercular;
the cristid obliqua rises to the apex of the metaconid as in
paracone and metacone equal in size, conical sharp and high;
Antiacodon and other antiacodonts.
conules well developed and relatively independent of ma-
Length of m2: 4.9 mm.
jor cusps; metaconule not hypertrophied; mesostyle present.
Included species: S. furcatus only, known from locality
(West, 1984, pp. 39-40.)
CP34D only.

Comments: Except for the construction of the talonid and The protocone is central and the conular wings are strong. The
hypoconulid, Sarcolemur is virtually identical in morphol- hypocone and anterior accessory cuspule are developed.
ogy to Antiacodon, suggesting a close relationship. Only Length of M2: approximately 6.0 mm.
the type specimen is known. Included species: N. emryi only, known from locality CP34D
only.
Auxontodon Gazin, 1958
Comments: Neodiacodexis emryi is known only from upper
Type species: Auxontodon pattersoni Gazin, 1958. molars and is among the largest of the antiacodonts, only
Type specimen: MCZ 9316. somewhat smaller in size than Tapochoerus. Storer (1984)
Characteristics: Small incisorlike canine. Much enlarged and has convincing! y shown that N eodiacodexis is more closely
probably caniniform pl. Second to fourth premolars allied with the antiacodonts than the homacodonts as orig-
antero-posteriorly elongate. Marked diastema between pl inally proposed by West and Atkins (1970). The cusps
and p2. Cheek teeth resembling those of Antiacodon. In- show a tendency toward being crescentic, foreshadowing
ferior margin of lower jaw strongly convex in anteropos- the condition in Auxontodon. Two upper molars identi-
terior profile (Gazin, 1958, p. 2). The paraconid is larger fied as Neodiacodexis sp., from the early Bridgerian site
than the metaconid and the cristid obliqua rises and in- of Powder Wash, Utah (locality CP5A) (West, 1984), rep-
terconnects with the metaconid. The hypoconulid is resent upper molars of Antiacodon pygmaeus (Krishtalka
 Richard K. Stucky 366

and Stucky, 1984). The single known specimen of N. emryi midline of the tooth. The preparaconule crista and prepro-
may represent the upper dentition of Antiacodon venustus. tocrista form a distinct shearing crest. The hypocone is very
small and developed as an elevated ridge on cingulum. The
mesostyle on the upper molars is weak to absent.
SELENODONTIA INCERTAE SEDIS:
Average length of m2: 4.4--4.8 mm.
HOMACODONTS (HOMACODONTIDAE MARSH,
Included species: H. pelodes (known from localities CP25E,
1894; BUNOMERYCIDAE GENTRY AND
CP25~CP25G,CP25H,CP27B,CP27C,CP27D,CP64A
HOOKER, 1988)
or B); H. uintensis Gazin, 1952 (locality CP25E).
Characteristics: Small bunoselenodont artiodactyls that possess an Hexacodus sp. is also known from locality CP25E.
anterior protocone, posterolingually enlarged metaconule, and con-
comitant small protoconule. The preprotocrista and preprotoconule- Comments: Hexacodus is the earliest known selenodont ar-
tiodactyl displaying an enlarged metaconule, anterior pro-
crista are often united into an anterior cross crest. The hypocone is
tocone, and the well-developed posthypocristid and posi-
absent on M3 and variably present on Ml and M2. A mesostyle
tioning of the hypoconulid on a relatively broad posterior
is variably present or absent on Ml-3. The cl and pl are approx-
imately equal in size. Lower molar cusps are pyramidal and the cingulum. These features unite all selenodont artiodactyls
into a monophyletic clade, distinguishing them from lep-
paraconid is reduced. The talonid is equal in length to trigonid and
tochoerines and antiacodonts. The genus may, however, be
the hypoconulid is located on a broad postcingulum. The entoconid
paraphyletic. The elevated twinning of the paraconid and
is tall and often isolated in early forms; in later forms a crest may
form that is continuous with a crest from the hypoconid (Scott, 1891; metaconid in H. pelodes suggests that this species may be
more closely related to Microsus and Homacodon than it
Peterson, 1919, pp. 66-67; West, 1984, p. 4).
is to H. uintensis, which bears a separate paraconid and
Texodon West, 1982 metaconid, resembling Helohyus despite its much smaller
size. Among early and middle Eocene "homacodonts," H.
Type species: Texodon meridianus West, 1982. pelodes uniquely possesses a very robust P3 and P4/p4.
Type specimen: TMM 41672-62. These teeth are unknown in H. uintensis.
Characteristics: Medium-sized bunodont artiodactyl known There is no compelling reason to synonymize Hexaco-
only from M2-3. M2 with strong hypocone, M2 meta- dus with European Protodichobune as suggested by Van
conule small and slightly larger in size than paraconule. Valen (1971 ). A number of features clearly distinguish Pro-
Parastyles small, no mesostyle, and strong cingula. Proto- todichobune, which is more derived in having the buccal
cone on M3 central, slightly anterior on M2. Preprotocrista wall of the protoconid and hypoconid sloped, inflated mo-
and preparaconule crista appear to be continuous. lars, an accessory cuspule in the ectoflexid (ectostylid),
Length of M2: 5.0 mm. a relatively larger hypoconid, a reduced parastyle, and a
Included species: T. meridianus only, known from locality specialized mesostyle that is developed on the buccal rim
SB43B only. of the stylar shelf (Sudre et al., 1983). Hexacodus is more
Comments: The phylogenetic position of Texodon is enig- derived in having a more expanded metaconule.
matic. The slightly expanded metaconule and anterior M2
New homacodont genus A
protocone would suggest affinities with "homacodonts,"
but the differences are slight and the teeth resemble those Characteristics: This new genus, from the late Greybullian
of antiacodonts equally well. part of the Wasatchian in the Bitter Creek Area, Wyoming
(locality CP25A), is currently under study by David Hobbs
Hexacodus Gazin, 1952 and Bert Covert of the University of Colorado at Boulder.
Type species: Hexacodus pelodes Gazin, 1952. The genus displays a number of unique features includ-
Type specimen: USNM 19215 (p4--m2). ing buccally exodaenodont lower molars with extreme de-
Characteristics: P3 and P4/p4 are robust and broad relative velopment of the buccal cingulum and the presence of a
to length; P3 is short, and slightly longer in length than neomorphic cusp buccal to the paracone that is neither a
P4/p4. The p4 is simple with the metaconid very weak or mesostyle or parastyle. The genus is similar to Hexacodus
absent. The m2-3 paraconid and metaconid are elevated, and other homacodonts in the possession of an expanded
higher than the protoconid and joined or closely appressed metaconule and the location of the hypoconulid on the
to one another. The paraconid is smaller than the meta- postcingulum. It resembles European Protodichobune and
conid (equal on ml in H. uintensis) and the metaconid is Aumelasia in the development of accessory cusps and the
not inflated. The hypoconulid is located on the posterior inflation of the tooth bases. It is the size of Hexacodus
cingulum on ml-2 and connected to the posthypocristid. pelodes.
The entoconid is tall, often spirelike, and lies lingual to
Homacodon Marsh, 1872a
the hypoconid. On Ml-2 the metaconule is expanded pos-
terolingually, and the protocone apex lies anterior to the Type species: Homacodon vagans Marsh, 1872a.
 Eocene bunodont and bunoselenodont Artiodactyla 367

Type specimen: YPM 13129. The p4 has a metaconid. The paraconid on m2-3 is either
Characteristics: Similar to Hexacodus pelodes, but differs in vestigial or absent resulting in a compressed trigonid, but
the following characteristics: m 1-3 paraconid very small or ml may retain a relatively well-separated paraconid and
incipient (absent in worn specimens) and closely appressed metaconid of equal size. A hypocone is present on Ml-2,
to metaconid; P3/p3 and P4/p4 gracile, p4 with a dou- and the mesostyle may be incipient or absent.
ble crest descending the posterior wall of the protoconid; Average length of m2: 4.0-4.9 mm.
crests on lower molars better developed and higher than Included species: M. cuspidatus only (known from localities
in Hexacodus; posthypocristid relatively well developed, CP34A, CP34B, CP34C, CP34D).
descending at a 45° angle to the hypoconulid, which is Microsus sp. is also known from localities CP5A, CP34C.
located medially on the postcingulum. The cusps of the
upper molars more crescentic than in Hexacodus and there Comments: Microsus is distinguished from Homacodon by
is no mesostyle. The cingular hypocone is well developed the presence of a metaconid on p4 and by, overall, more
onMl-2. crescentic and taller cusps. Microsus sp. represents a prob-
Average length of m2: 4.0-5.8 mm. able new species but is known from relatively poor mate-
Included species: H. vagans (known from locality CP34D); rials. M. sp. (USNM 364914 and other USNM specimens)
?H. n. sp. A (locality CP34D);H. n. sp. B (locality CP34B); differs from M. cuspidatus in having lower molars with
?H. n. sp. C (locality CP25K). a hypolophid that connects the hypoconid and entoconid
directly and a strongly developed cristid that goes anteri-
Comments: Specimens of small selenodont artiodactyls are orly from the hypoconulid, intersecting the posthypocristid
extremely rare in the Bridger Formation of the Green at a 90° angle. In M. cuspidatus the posthypocristid and
River Basin. In the past, the selenodont forms have been the crest from the hypoconulid are continuous as the hy-
placed into Microsus cuspidatus if they were small and polophid. Upper molars from Powder Wash, Utah (local-
into Homacodon vagans if they were larger. It is beyond ity CP5A; CM collections), display relatively gracile, tall
the scope of this chapter to revise these taxa, but a num- cusps compared to material referred to M. cuspidatus. A
ber of morphologies are apparent that represent at least six specimen referred to Microsus by West (1973) from the
species between the two genera. The four species assigned Bridger Formation of the Big Sandy-New Fork area (lo-
to Homacodon differ principally in size, with H. n. sp. A cality CP34B) is an m3 of Hyopsodus.
> H. vagans > H. n. sp. B > H. n. sp. C. Homacodon n.
sp. A and C may represent two additional genera of middle Hylomeryx Peterson, 1919 (including Sphenomeryx
Eocene Artiodactyla. The H. n. sp. A (USNM field number Peterson, 1919)
41-23, collected by C. L. Gazin, basicranium and associ-
ated dentition) is more robust with a deep mandible and rel- Type species: Hylomeryx annectans Peterson, 1919.
atively wider lower dentition and has a relatively longer ba- Type specimen: CM 2335.
sicranium than H. vagans. H. n. sp. C (DMNH 8625) and H. Average length of m2: 5.8 mm.
n. sp. D (DMNH 15042) are smaller; they are also the first Characteristics: The parastyle, mesostyle, and metastyle are
records of Homacodon from Bridger B (locality CP34B) relatively more developed than in Homacodon and Micro-
and the Cathedral Bluffs Tongue (locality CP25K) of the sus but less so than in Bunomeryx. The paracone and meta-
Wasatch Formation. All species assigned to Homacodon cone are conical and bunodont and show a trend toward
have a relatively gracile p4 that lacks a metaconid but has labial ribbing that is typical of Bunomeryx, Hypertragu-
a doubled ridge which descends the protoconid down to a lus, and North American tylopods. Ml-2 retain ahypocone
simple lateral ridge on the talonid. Homacodon vagans is like those in Homacodon. The protoconule is incipient on
represented by rather exquisite skull material (YPM 13129, the protoloph ofM2 and M3, but more distinctive on Ml.
type, and AMNH 12695). Regrettably, the premolars have P4/p4 are robust, and p4 retains a strong metastylid. The
been lost from AMNH 12695 since its original description posthypocristid of the lower molars is directed toward the
(Sinclair, 1914). A rather poor nineteenth-century recon- entoconid as in Homacodon and Micros us but not like that
struction of the type (YPM 13129) has placed the maxil- in Bunomeryx where the hypolophid is directed at an angle
laries much more posteriorly than actually occurred in life. behind the entoconid.
A lower molar referred to Homacodon from Texas (West, Average length of m2: 4.7-5.8 mm.
1982) is an ml or m2 of the primate Macrotarsius. Included species: H. annectans (known from localities CP6A,
CP38C); H. quadricuspis (localities ?CP6A, CP6B, CP65).
Microsus Leidy, 1870 A questionable Hylomeryx sp. is also known from lo-
cality SB43A.
Type species: Microsus cuspidatus Leidy, 1870.
Type specimen: USNM 1178 (formerly ANS 10260). Comments: An ml or m2 (TMM 41372-245) is provision-
Characteristics: Closely resembles Homacodon, but all cusps ally identified as Hylomeryx (referred to Microsus; West,
are relatively higher, less bunodont, and more crescentic. 1982), based on the development of the hypolophid and
 Richard K. Stucky 368

size. The specimen of Hylomeryx from the Washakie Basin continuous with the preparaconule crista. The labial mar-
(CM collections), if correctly referred, has a greatly ex- gins of the paracone and metacone show ribbing. The meta-
panded and rounded posteroventral margin of the dentary conule is crescentic. The premolars are unspecialized. P3
as in agriochoerids and in type material of Hylomeryx as is shortened anteroposteriorly and the deuterocone is for-
inferred by the structure of the posterior part of the dentary. ward in position. Size is similar to Microsus.
This specimen also has relatively well-developed cingula Average length ofm2: 4.5-5.0 mm (est.).
and paraconids on the lower molars, suggesting that it may Included species: M. grangeri only, known from locality
represent a distinct genus. CP6B only.

Comments: Mesomeryx may lie near an ancestral position to

Bunomeryx Wortman, 1898 Simimeryx and Hypertragulus (Gazin, 1955).
Type species: Bunomeryx montanus Wortman, 1898.
Type specimen: AMNH 2071. Pentacemylus Peterson, 1931
Characteristics: Cusps on the lower molars are more cres- Type species: Pentacemylus progressus Peterson, 1931.
centic than in Hylomeryx and the p4 is more gracile. The
Type specimen: CM 11865.
posthypocristid is directed posteriori y at an angle to behind Characteristics: Related to Bunomeryx but more progressive
the entoconid. A well-developed hypocone is preserved on with much more crescentic cusps and a complete absence
Ml, but only incipiently present and much reduced on M2 of hypocone on Ml (vestige lost on M2). P4 cusps are
over the condition in Hylomeryx. The parastyle, mesostyle more crescentic, especially that of the deuterocone. Lower
and metastyle are all well developed and the paracone and molars more "hypsodont" and the principal cusps are more
metacone are ribbed. Overall, the cusps present a more crescentic. The paraconid is better defined on p3 and dis-
crescentic appearance than in Hylomeryx. tinctly larger on p4 compared to Bunomeryx.
Average length ofm2: 5.2-5.4 mm. Average length ofm2: 5.7-6.7 mm.
Included species: B. montanus only (= B. elegans Wortman, Included species: P. progressus (known from localities CP7 A,
1898, fide Gazin, 1955; personal observations) (known
CP7C); P. leotensis (locality CP6B).
from localities CP6A, CP6B).
New "homacodont" genus B
Mytonomeryx Gazin, 1955
Characteristics: A new genus ofbunoselenodont artiodactyl is
Type species: Mytonomeryx scotti Gazin, 1955. represented by a specimen in the Denver Museum of Natu-
Type specimen: USNM 20401. ral History collections preserving ml-2 from the Washakie
Characteristics: Elongate rostrum with marked diastemata Formation of the Sand Wash Basin, Colorado (locality
separating the first and second premolars, above and be- CP65). The metaconid and entoconid are somewhat coni-
low, from adjacent teeth. Ml and M2 noticeably quadrate cal, but both are canted anteriorly. The anterior cingulum is
and both with a well-developed hypocone. Cusps appar- very strong and there is a crest that descends down the ante-
ently more crescentic than in Bunomeryx. Upper molars rior part of the paraconid and swings lingually but does not
moderately selenodont with prominent mesostyle (Gazin, intersect with the metaconid. There is a small ectostylid,
1955). and both the protoconid and hypoconid are crescentic. A
Average length of m2: 6.0 mm. small metastylid is also present. The cristid obliqua strikes
Included species: M. scotti only (known from localities CP6B, the postvallid low and medially.
NP8). Average length ofm2: 6.0 mm.
Comments: This taxon appears to be intermediate between
Comments: Dental morphology of Mytonomeryx closely re- Pentacemylus and Protoreodon. The stylids on the lingual
sembles that of Bunomeryx, which lacks the elongate snout
part of the trigonid are incipient, more developed than in
typical of Mytonomeryx. Pentacemylus but less developed than in Protoreodon and
other tylopods.
Mesomeryx Peterson, 1919
Type species: Mesomeryx grangeri Peterson, 1919. Apriculus Gazin, 1956
Type specimen: CM 3189. Type species: Apriculus praeteritus Gazin, 1956.
Characteristics: Mesomeryx has simple bunoselenodont teeth Type specimen: USNM 21100.
distinguished by the absence of both a mesostyle and hypo- Characteristics:
cone on the upper molars. Although Mesomeryx lacks a
hypocone on Ml and M2, there is flexure on the cingulum Simple conical cusps on upper molars as in Helohyus, but
where a hypocone would be developed on Ml. The para- these teeth more nearly quadrate with large lingually placed
conule is absent or barely apparent with the preprotocrista metaconule on all three. Cingulum continuous around molars
 Eocene bunodont and bunoselenodont Artiodactyla 369

without external styles and without evidence of a hypocone. iohyus and the anthracotheres (Coombs and Coombs, l 977a, 1982).
Protoconule slightly better defined than in Helohyus. P4 with As indicated later, characters used to ally these taxa define the se-
single primary cusp and deuterocone. (Gazin, 1956, p. 23.) lenodont artiodactyls as a whole and thus have no bearing on the
relations of helohyines to other selenodont artiodactyls. Although
These upper molars are square in plane view and the parastyle
somewhat smaller, Hexacodus uintensis is virtually identical in pre-
is reduced. The preprotocone crista is continuous with the
serve<:: features ofml-2 of Helohyus.
preparaconule crista as in other homacodonts. On P4, the
pre- and postcrista of the primary cusp are ridgelike, resem- Helohyus Marsh, 1872b (= Thinotherium Marsh, 1872b
bling those of more derived homacodonts than helohyines
[preoccupied by Thinotherium Cope, 1870, a tayassuid],
(in which the primary cusp is more bulbous). The size of
Lophiohyus Sinclair, 1914)
molars increases from Ml-3. The posthypocristid on the
lower molars is directed toward the entoconid, on which Type species: Helohyus plicodon Marsh, 1872b.
the apex is slightly anterior to the hypoconid. Type specimen: YPM 11893 (M3).
Average length of m2: 8.0 mm (est.). Characteristics: Helohyus is moderately large in size with
Included species: A. praeteritus only, known from locality bunodont molars and generally crenulate enamel, espe-
CP29C only. cially on the upper molars. Compared to homacodonts,
the upper molars have the parastyle reduced and the para-
Comments: A lower molar from Badwater SA (locality conule more isolated and distinct. Although the metaconule
CP29C) is tentatively referred here. The posthypocristid is expanded, it is only slightly larger than the paraconule.
is directed toward the entoconid and the hypoconulid is The upper molars bear a very small cingular hypocone.
reduced to a small swelling on the postcingulum. Apricu- The paraconid is more prominent on the lower molars than
lus differs from Helohyus in the lack of any vestige of a in homacodonts but reduced on the premolars. There is
hypocone and the greatly enlarged metaconule. It resem- no metaconid on the premolars, which are simple in mor-
bles Helohyus in its bunodont and low cusps and reduced phology. The p3 is taller than p4 but equal in length. A
parastyle. neomorphic cusp intermediate in position between the en-
Gazin (1955) and Coombs and Coombs (1977a, 1977b) toconid and hypoconid may be present on the hypolophid.
have suggested a close relationship with Helohyus, and As in the homacodonts, the hypoconulid is located on the
Coombs and Coombs have suggested that Apriculus is posterior cingulum and is connected to the hypolophid by
a basal anthracothere. Characters cited for a relationship a crest that runs directly to the hypoconid (H. plicodon) or
with Helohyus are typical of homacodonts as a whole, as to the neomorphic cusp (H. lentus).
are those that ally Helohyus with Gobiohyus and anthra- Average length of m2: 8.4-14.3 mm.
cotheres. Apriculus may be more closely related to such Included species: H. plicodon (including H. validus) (known
homacodonts as Mesomeryx and Pentacemylus by virtue from localities CP34C, CP34D); H. milleri (including
of the loss of the hypocone, presence of a complete cin- Lophiohyus alticeps) (localities SB43A, ?CP32B, CP34D);
gulum, and generally square molars due to the large size H. lentus (including Parahyus aberrans and perhaps Ithyo-
of the metaconule. There is some resemblance to tayas- grammodon cameloides) (localities CP31C, CP34D).
suids, but the presence of a complete lingual cingulum on
Apriculus and the lingual fissure between the protocone Comments: Helohyus has been allied with the anthracotheres
and metaconule on tayassuids would appear to negate any and Asian Gobiohyus (Coombs and Coombs, 1977a,
phylogenetic relationship. 1977b). This resemblance is considered superficial, based
on convergence due to molar bunodonty and inflation. In-
deed, all features cited by them as derived are also shared
HELOHYINAE
withHomacodon. Helohyus has much more bulbous cusps,
Characteristics: Bunodont molars; metaconule expanded; hypo- with a more pronounced cingular hypocone and more
conulid on postcingulum oflower molars; the lower canine is canini- crenulate enamel than Gobiohyus. Gobiohyus bears a num-
form; and the p4 lacks a metaconid. The ml and m2 have a weak ber of derived characters including more crescentic pro-
hypoconulid located on the postcingulum. A neomorphic cusp may tocone, paraconule and metaconule, ribbed paracone and
be present on the talonid rim of m 1-2 directly above the true hypo- metacone, large parastyle, entoconid anterior in position
conulid. On the upper molars the protocone is anterior and coni- to the hypoconid, relatively taller cusps on the molars, and
cal, similar to the condition in Hexacodus, but the lingual margin a prominent metaconid on p4. The relationships of Helo-
of the upper molars is rounded and not square. Upper molars gen- hyus to the entelodonts and achaenodonts requires much
erally have a small hypocone developed on the postcingulum. The more study. In the entelodonts, the snout is elongate and
mesostyle is absent. there are diastemata between the premolars, whereas in
Included genera: Helohyus, helohyine gen. nov., and possibly Dis- achaenodonts the premolars are crowded with the loss of
critochoerus, tentatively also including Parahyus and Achaenodon. p 1. H. plicodon lacks diastemata between the premolars,
Comments: The helohyines have been variously allied with Gob- but H. milleri has relatively well-developed diastemata
 Richard K. Stucky 370

as in the new genus discussed later. An unworn speci- Achaenodon Cope, 1873c (including Protelotherium
men of Achaenodon shows a talonid structure very simi- Osborn, 1895)
lar to that in H. lentus, with a neomorphic cusp and crest
Type species: Achaenodon insolens Cope, l 873c.
which descends from the hypolophid to intersect with the
Type specimen: AMNH 5143.
hypoconulid. Ithyogrammodon cameloides, represented by
Characteristics: Large bunoselenodont artiodactyls; skull
the anterior maxilla bearing the canine and alveoli and
broad with orbits not enclosed posteriorly. The limbs are
roots for the incisors in the premaxilla, is probably refer-
short, and the feet are tetradactyl. The pl is lost and the
able to Helohyus. Size of the specimen is near that of H.
molars are crowded together. The hypoconulid lies on the
lentus and there is a marked resemblance in overall mor-
postcingulum and there is a neomorphic cusp on the talonid
phology to the only known snout of Helohyus (type of
rim on ml. The hypoconulid is lobate on m3. The up-
Lophiohyus alticeps; Sinclair, 1914). The size and general
per molars are four cusped with a greatly expanded meta-
similarity of the specimen suggest that it may be a junior
conule.
synonym of H. lentus. lthyogrammodon essentially con-
Average length of m2: 24.2-26.7 mm.
sists of material that is not comparable to other helohyines
Included species: A. insolens (known from localities CP6A,
and is here considered a nomen dubium.
CP38C); A. robustus (locality CP38C); A. uintensis (local-
ity CP6A).
New helohyine genus
Achaenodon sp. is also known from localities CC4,
Characteristics: Size equal to Helohyus lentus. Molars more CP36B.
bunodont than in Helohyus, and marked diastemata be-
Comments: Achaenodon, despite the crowded premolars and
tween premolars suggesting an elongate snout. The pl is
loss of pl, would appear to be related to the helohyids
single rooted. Skull relatively broad with a marked sagittal
based on the similarity of the hypoconulid construction
crest compared to Helohyus. The centrocrista of the upper
on ml.
molars appears to be higher in relation to the stylar shelf.
Length of m2: 14.6 mm. Parahyus Marsh, 1876
Comments: This enigmatic taxon, known only from locality
CP65, has been referred to Parahyus by West and Dawson Type species: Parahyus vagus Marsh, 1876.
(1975), but differs from that genus in the possession of Type specimen: YPM 10972.
diastemata between the premolars and the presence of p 1. Characteristics: Size intermediate between H. lentus and
It is convergent on entelodonts in the possession of an Achaenodon. The p 1 is lost and the anterior premolars are
elongate snout rather than closely related to them (contra crowded. The ml retains a paraconid and has an entostylid.
Stucky in Stucky and McKenna, 1993; see also Snyder, The trigonids of m2-3 are broader than talonid. The p4 is
1993 ). In entelodonts, the p 1 is double rooted and the third longer and taller than p3.
lower molar bears a much reduced talonid. Average length of m2: 20.8 mm.
Included species: P. vagrans only, probably known from lo-
Discritochoerus Gazin, 1956 cality CP38C and/or D (see Gazin, 1955, for discussion).

Type species: Discritochoerus lapointensis Gazin, 1956.

Type specimen: CM 11912 (type and only known specimen INDETERMINATE ARTIODACTYLA
lost). West (1973) identified a p4 from the lower Bridger Formation of
Characteristics: p4 with parastylid; ml paraconid far for-
the New Fork-Big Sandy area (locality CP34B). The presence of
ward; crest extending posteriorly from the protoconid or
a strong metaconid on this specimen suggests affinities with either
the triangular-shaped basin formed between this crest and
antiacodonts or derived homacodonts.
cristid obliqua; size near Helohyus lentus (Gazin, 1956).
Average length of m3: unavailable. Length of M 1: 13 .0 mm.
Included species: D. lapointensis only, known from locality BIOLOGY AND EVOLUTIONARY PATTERNS
CP7C only.
Members of all of the groups considered here were generally small,
Comments: Gazin (1956) named this taxon on the basis of ranging in size from a rabbit to that of a peccary. Exceptions are
the original description of Peterson ( 1934 ), despite the fact Parahyus and Achaenodon (Helohyinae), which approach the size
that the specimen could not be located in the Carnegie Mu- of a prize pig.
seum collections at that time. The specimen is still lost. I The tooth morphology is quite primitive in Diacodexis, Hexa-
agree with Gazin that the specimen represents a distinctive codus, and the leptochoerines, suggesting they were omnivorous
genus, but given that the type is lost, relationships of Dis- herbivores relying on fruit, seeds, and leafy vegetation. Some of the
critochoerus will remain obscure until more material from early forms, such as Diacodexis and Hexacodus, may have included
the type area is discovered. animal protein (insects) in their diets. This applies equally well to
 Eocene bunodont and bunoselenodont Artiodactyla 371

PALEOCENE EOCENE OLIGOCENE

m r- m iii: r- m r-
~ ~ r~-~-~-~-6~6-6Tl?-?-~~~~-?-~-n-~T~~~~-~~-r~.-,m~-~~--=~~~~~~-c-~1m~~~r...-o_:_~~.~l~-mm.~_:_-r~~
11 11 o :..N w:... N ~ ... ...,c,, ~ui m:...~w rn :s ~ ~ I ~ ~ ~ ~ i c o o g ~ == -~
~~ i-..,.-:---,,,..-~'_,.~-,,,,,,.,,-.,,~~'"':-~~~~~~~~~~~~~~~~-'-'-'~'~·..__~~-'-'--'-~~~-'--·--t
~ I .~-~-~-~-~-~-~-~~:-~-~-~-~-~-~-~T~~~-=-=t~-~-=--=--~~-e~~e~-e~~~~~-!=--~~~~~;._:;~-r:~~;~~:=--~~E~~__:g:...i
? IE IDiacodexisl I I I I I I I I I I I
.:.
~ ISimpsonodus I I I I I I I I I I I
~ lsunophorus I I I I
I I I I I I I
g - I I I I I I I lb~rus I I I I I I I
~ !:;; I I I I I I I r--1 I I I I I I
~ I I I I 1 , , , Stibarus
~
~
a
"Cl

C"> I I
1
I I I I
1 1
I I I L 1 Leptochoerus
1

~ 5m I 1"Diacodexis w'?!!.o!ef!!iS I
II I r.,, I I I I
~ ::o I I I I I I ._, I I I I I I I I
fi ~ I I I I I 1Laredoch~ru~ I I I I I
~ m I 1 I I I I I 1New leptochoerine genus 1
c.
0 ;:::: I I I I I I I I I I I I I I
~ ~ L _ 1Antiacodon 1 1 I I I I I I I I
~ ~ I ..,, I I I ITapochoe_r_us_I I I I I I I
~ O I I I I ISa;coled.ur I I I I I I I
g. 8 I I I I t--i I I I I I I I I
IE~ z
:::! I I I I 1Auxontodon I I I I I I I
~ ~ I I I I I INeodiacodexis I I I I I I
;, m - I I I I I II I I I 11 I I I
~ - I I I I I I I 1Texodon1 I I I I I
_g I Hexacodus I I I I I I I I I I I
~ I I I I I I I I I I I I I I I
11
m
,New homacont genus A
f::!

:i
-

I I
I I I I

1Homacodon1
I 1
I
I
I
I
I
II
I I
I
I
I
I
I
I
:> o I I IMicrosus I I I I I I I I I
~ ~ I I I I I I I 1 I I I I I I
11
C"> Hy/omeryx
~ 0 I I I I I I 1 1 I I I I I I
~ ~ I I I I I 1Bunomeryx I I I I I I
,,. -1 I I I I I IMytonomeryx I I I I I
~- CJ? I I I I I I II I II I I
o Mesomeryx
~ I I I I I I r--i I I I I I
~ I I I I I 1Pentacemylus I I I I I
~ I I I I I INew homacodont genus B I I
} I I I I I II I I I I I I I
Apricu/us
;:!

~
II

:.
= I
I I
I I
I
I
I I

I
I
IHel~hyu~
I
I

I
I
I
I
I
I
r--i

I
I
I I I
I I I
I I I
I
I
I
I
I
I
I
I
I
iii" :::c
:> m I I I I I 1 I 1Disc!!!!!!!:,hoerus 1 I I
~II 0:::C I I I 1Achaenodon I I I II I I I
~ < I I I I I I I I I I I I I I I
j_
g
~m I I I 1 1 1 1
New helohyine genus
1 I I
-~· I I I 1Parahyus1 I I I I
- I
I
I
I
.1
I I I I
Figure 23.6. Temporal ranges of genera of Eocene bunodont and bunoselenodont artiodactyls.
I I I I I I I
I
I
I
I
 Richard K. Stucky 372

the helohyines, antiacodonts, and the more primitive members of the ACKNOWLEDGMENTS
homacodonts, which can all be considered as broad-spectrum her-
bivores. However, the more selenodont and higher crowned homa- Special thanks are due to the following people for access to spec-
codonts probably relied more heavily on browse. Some of the more imens in their care: M. McKenna, R. Tedford, and J. Alexander
derived antiacodonts (Auxontodon) may also have relied more on (AMNH); M. Dawson (CM); R. Emry and R. Purdy (NMNH); D.
browse than earlier taxa. Rasmussen (Washington University, St. Louis); P. Robinson (UCM);
These early artiodactyls probably lived in the dense undergrowth J. Ostrom and M. Turner (Yale); J. Wilson (TMM); R. West (ISE);
and thickets of paratropical and subtemperate woodlands at all eleva- J. Storer (SMNH); K. Rose (JHU); W. Turnbull (FMNH), T. Bown
tions. They were quite abundant and diverse during the Wasatchian (formerly of USGS); P. Gingerich and G. Gunnell (UMMP); D.
(early Eocene) but were substantially reduced during the Bridgerian Savage (UCMP); D. Russell (MNHN); J. Hooker (BMNH); J. Lil-
(Gazin, 1976). During the Uintan (middle Eocene) the artiodactyls legraven and B. Breithaupt (UW); M. Coombs (ACM); E. Lindsay
again flourished, at which time homacodonts and other selenodont (UA); D. Baird (PU); J. Harris (LACM). Thanks are due to Leonard
forms (protoceratids, hypertragulids, camelids, oromerycids, agri- Krishtalka of the University of Kansas Museum of Natural History
ochoerids) suddenly first appeared across much of the western half for codeveloping many of the ideas presented here. Cindy Grubb
of the North American continent. provided word-processing assistance. Connery Calhoun, Andrew
The first appearance of artiodactyls (i.e., Diacodexis ilicus and D. Redline, and Ken Carpenter prepared many of the original figures,
secans-primus) marks the beginning of the Wasatchian Land Mam- which were redrawn by Brian Regal. I especially thank Christine
mal age in North America (late Paleocene to earliest Eocene). Dia- Janis for her herculean efforts in pushing this manuscript forward:
codexis, Simpsonodus, and Bunophorus, as well as a homacodont, Without her persistence it may never have appeared. Support for
may be represented in the European late Paleocene and early Eocene. this research was provided by the National Science Foundation,
All other taxa considered here are restricted to North America. NASA, Carnegie Museum of Natural History, Denver Museum of
Diacodexis and related leptochoerines ranged from the late Pa- Natural History, and the Scientific and Cultural Facilities District of
leocene or earliest Eocene into the late Oligocene. Simpsonodus Colorado.
is restricted to the Wasatchian (early Eocene), and Bunophorus
is known from the Wasatchian into the early Bridgerian (early to
middle Eocene). The antiacodonts range from the late Wasatchian REFERENCES
through the Uintan (late early Eocene to middle Eocene). The homa-
codonts range from the middle Wasatchian into the Duchesnean Black, C. C. ( 1978). Paleontology and geology of the Badwater Creek area,
(early to late middle Eocene), and the helohyines range from the central Wyoming, Part 14: The artiodactyls. Annals of the Carnegie
Bridgerian to the Duchesnean (middle Eocene). Museum, 47, 223-59.
Coombs, M. C., & Coombs, W. P., Jr. (1977a). Dentition of Gobiohyus and a
The earliest selenodont artiodactyl is the genus Hexacodus, which
reevaluation of the Helohyidae (Artiodactyla). Journal of Mammal-
bears an anterior protocone, an expanded metaconule, a reduced ogy, 58, 291-308.
paraconid, an isolated entoconid, and a hypoconulid that lies on Coombs, W. P., Jr., & Coombs, M. C. ( l 977b ). The origin of anthracotheres.
the postcingulum rather than on the posterior margin of the talonid Neues Jahrbuchfuer Geologie und Palaeontologie, Abhandlungen,
basin. The homacodonts are extremely rare in most assemblages dur- 10, 584-99.
Coombs, M. C., & Coombs, W. P., Jr. (1982). Anatomy of the ear region
ing the Wasatchian and Bridgerian (early to early middle Eocene),
of four Eocene artiodactyls: Gobiohyus, ?Helohyus, Diacodexis and
and it is quite likely that current taxonomy is quite conservative. Homacodon. Journal of Vertebrate Paleontology, 2, 219-36.
Given that the different Uintan homacodonts can be allied to ei- Cope, E. D. (1870). Fourth contribution to the fauna of the Miocene and
ther the tylopods or ruminants, it is quite likely that a large part Eocene periods of the United States. Proceedings of the American
of the early radiation of the selenodont forms remains masked by Philosophical Society, 11, 285-94.
( 1873a). Third notice of extinct Vertebrata from the Tertiary of the Plains.
our limited knowledge of the Bridgerian forms. It is probable that
Paleontological Bulletin, 16, 1-8.
both the ruminants and tylopods had their origin in North America, (l 873b ). (On Menotherium lemurinum, Hypisodus minimus, Hypertrag-
migrating later to the Old World. ulus calcaratus, H. tricostatus, Protohippus, and Procamelus occi-
The abundance patterns of the homacodonts and selenodont dentalis). Proceedings ofthe Academy ofNatural Sciences, Philadel-
families reflects an inverse relationship to primate diversity in phia, 1873, 419-20.
(l 873c ). Fourth notice of extinct Vertebrata from the Bridger and the
North America (Stucky, 1990, 1992), suggesting a pattern coin-
Green River Tertiaries. Paleontological Bulletin, 17, 1--4.
cident with the opening up of habitats or savanna environments. In (1875a). On fossil lemurs and dogs. Proceedings of the Natural Academy
North America, there is also an inverse relation between selen- of Sciences, Philadelphia, 1875, 255-6.
odont artiodactyl richness and perissodactyl richness after the Uin- (1875b). Systematic catalogue of Vertebrata of the Eocene of New Mex-
tan, especially as the body size range increased among the ar- ico, collected in 1874. Washington, D.C.: Report of the Engineers
Department, U.S. Army, in charge of Lieut. George M. Wheeler,
tiodactyls to overlap that of the medium to larger sized perisso-
pp. 5-37.
dactyls. Since the Uintan, the artiodactyls have been a dominant (1881). On the Vertebrata of the Wind River Eocene beds of Wyoming.
element of the herbivore guild in ecological communities in North Bulletin of the United States Geological and Geographical Survey
America. of the Territories, 6, 165-81.
 Eocene bunodont and bunoselenodont Artiodactyla 373

(1882). Two new genera of Mammalia from the Wasatch Eocene. Amer- (1894). Description of Tertiary artiodactyls. American Journal of Sci-
ican Naturalist, 16, 1029. ence, 48, 259-74.
(1894). Marsh on Tertiary Artiodactyla. American Naturalist, 28, 867-9. Matthew, W. D. (1899). A provisional classification of the Freshwater Ter-
Edwards, P. (1976). The subfamily Leptochoerinae (Artiodactyla, Di- tiary of the West. Bulletin of the American Museum of Natural His-
chobunidae) of North America (Oligocene). Contributions to Ge- tory, 12, 19-75.
ology, University of Wyoming, 14, 99-114. McKenna, M. C. (1959). Tapochoerus, a Uintan dichobunid artiodactyl from
Estravis, C., & Russell, D. E. (1989). Decouverte d'un nouveau Diacodexis the Sespe Formation of California. Bulletin of the Southern Califor-
(Artiodactyla, Mammalia) dans !'Eocene inferieur de Silveirinha, nia Academy of Sciences, 58, 125-32.
Portugal. Palaeovertebrata, 19, 29--44. Osborn, H. F. ( 1895). Fossil mammals of the Uinta Basin. Bulletin of the
Gazin, C. L. (1952). The lower Eocene Knight Formation of western American Museum of Natural History, 7, 71-105.
Wyoming and its mammalian faunas. Smithsonian Miscellaneous Peterson, 0. A. (1919). Report upon the material discovered in the upper
Collections, 117, 1-82. Eocene of the Uinta Basin by Earl Douglas in the years 1908-1909,
(1955). A review of the upper Eocene Artiodactyla of North America. and by 0. A. Peterson in 1912. Annals of the Carnegie Museum, 12,
Smithsonian Miscellaneous Collections, 128, 1-96. 40-168.
( 1956). The geology and vertebrate paleontology of upper Eocene strata (1931). New species from the Oligocene of the Uinta. Annals of the
in the northeastern part of the Wind River Basin, Wyoming, Part 2: Carnegie Museum, 21, 51-78.
The mammalian fauna of the Badwater area. Smithsonian Miscella- ( 1934). List of species and description ofnew material from the Duchesne
neous Collections, 131, 1-35. River Oligocene, Uinta Basin, Utah. Annals ofthe Carnegie Museum,
(1958). A new dichobunid artiodactyl from the Uinta Eocene. Breviora 23, 373-89.
(Museum of Comparative Zoology), 96, 1-6. Robinson, P. (1966). Fossil Mammalia of the Huerfano Formation, Eocene,
(1976). Mammalian fauna! zones of the Bridger middle Eocene. Smith- Colorado. Bulletin of the Peabody Museum of Natural History, 21,
sonian Contributions in Paleobiology, 26, 1-25. 1-95.
Gentry, A. W., & Hooker, J. J. (1988). The phylogeny of the Artiodactyla. Rose, K. D. (1982). Skeleton of Diacodexis, oldest known artiodactyl.
In The Phylogeny and Classification of the Tetrapods, vol. 2: Mam- Science, 216, 621-3.
mals, ed. M. J. Benton, pp. 235-72. Oxford: Systematics Association. (1985). Comparative osteology of North American dichobunid artio-
Special Volume no. 35B, Clarendon Press. dactyls. Journal of Paleontology, 59, 1203-26.
Gingerich, P. D. (1989). New earliest Wasatchian mammalian fauna from Scott, W. B. (1891). Osteology of Poebrotherium: a contribution to the phy-
the Eocene of Northwest Wyoming: composition and diversity in a logeny of the Tylopoda. Journal of Morphology, 5, 1-74.
rarely sampled high-floodplain assemblage. University of Michigan (1898). Preliminary note on the selenodont artiodactyls of the Uinta
Papers on Paleontology, 28, 1-97. Formation. Proceedings of the American Philosophical Society, 157,
Godinot, M. (1978). Diagnose de trois nouvelles especes de mammireres du 1-9.
Sparnacien de Provence. C.r. somm. Seane. Soc. geol. Fr., 6, 286-8. (1899). The selenodont artiodactyls of the Uinta Eocene. Transactions
Golz, D. J. (1976). Eocene Artiodactyla of southern California. Natural of the Wagner Free Institute, 6, 1-121.
History Museum of Los Angeles County Science Bulletin, 26, 1-85. ( 1940). Artiodactyla. Transactions of the American Philosophical Soci-
Janis, C. M., & Scott, K. M. (1987). The interrelationships of higher ruminant ety, 28, 363-746.
families, with special emphasis on the members of the Cervoidea. Sinclair, W. J. (1914). A revision of the bunodont Artiodactyla of the mid-
American' Museum Novitates, 2893, 1-85. dle and lower Eocene of North America. Bulletin of the American
Krishtalka, L., & Stucky, R. K. (1984). Middle Eocene marsupials from Museum of Natural History, 33, 267-95.
northeastern Utah and the mammalian fauna from Powder Wash. Snyder, J. R. (1993). A new genus ofHelohyidae (Mammalia, Artiodactyla)
Annals of the Carnegie Museum, 53, 31--45. from the Sand Wash Basin, Colorado (Washakie Formation, Eocene,
(1985). Revision of the Wind River faunas, early Eocene of central earliest Uintan). Journal of Vertebrate Paleontology, 13 (suppl. to
Wyoming, Part 7: Revision of Diacodexis (Mammalia, Artiodactyla ). issue no. 3).
Annals of the Carnegie Museum, 54, 413-86. Stock, C. (1934). Microsyopinae and Hyopsodontidae in the Sespe upper
(1986). Early Eocene artiodactyls from the San Juan Basin, New Mex- Eocene, California. Proceedings of the National Academy of Sci-
ico, and the Piceance Basin, Colorado. Contributions to Geology, ences, 20, 349-54.
University of Wyoming, Special Paper 3, 183-96. Storer, J. E. (1984). Mammals of the Swift Current Creek Local Fauna
Leidy, J. (1856). Notices of remains of extinct Mammalia, discovered by (Eocene: Uintan, Saskatchewan). Natural History Contributions,
Dr. F. V. Hayden in Nebraska Territory. Proceedings of the Academy Museum of Natural History, Regina, 7, 1-158.
of Natural Sciences, Philadelphia, 8, 88-90. Stucky, R. K. (1990). Evolution ofland mammal diversity in North America
(1870). (Descriptions of Palaeosyops paludosus, Microsus cuspidatus, during the Cenozoic. Current Mammalogy, 2, 375--432.
and Notharctus tenebrosus). Proceedings of the Academy of Natural (1992). Mammalian faunas in North America of the Bridgerian to early
Sciences, Philadelphia, 111-14. Arikareean "age" (Eocene and Oligocene). In Eocene-Oligocene
Lucas, S. G. (1983). Comments on two species of the Eocene artio- Climatic and Biotic Events, eds. D. R. Prothero & W. Berggren,
dactyl H omacodon and the taxonomic status of Nanomeryx caudatus pp. 464-93. Princeton: Princeton University Press.
Marsh, 1894. New Mexico Journal of Science, 23, 48-56. Stucky, R. K., & Krishtalka, L. (1990). Revision of the Wind River
Macdonald, J. R. (1955). The Leptochoeridae. Journal of Paleontology, 29, faunas, early Eocene of central Wyoming, Part 10: Bunophorus
439-59. (Mammalia, Artiodactyla). Annals of the Carnegie Museum, 59,
Marsh, 0. C. (1872a). Preliminary description of new Tertiary mammals, 149-71.
Part I. American Journal of Science, 3, 122-8. Stucky, R. K., & McKenna, M. C. ( 1993 ). Mammalia. In The Fossil Record 2,
(1872b). Preliminary description of new Tertiary mammals, Part II. ed. M. J. Benton, 739-71. London: Chapman and Hall.
American Journal of Science, 3, 202-24. Sudre, J., Russell D. E., Louis P., & Savage, D. E. (1983). Les artio-
(1875). Notice of new Tertiary mammals, Part IV. American Journal of dactyles de !'Eocene inferieur d'Europe. Bulletin du Museum Na-
Science, 3, 239-50. . tional d'Histoire Naturelle, Sec. C, 281-333.
(1876). Notice of new Tertiary mammals, Part V. American Journal of Thewissen, J. G. M., Russell, D. E., Gingerich, P. D., & Hussain, S. T. (1983).
Science, 9, 239-50. A new dichobunid artiodactyl (Mammalia) from the Eocene of
 Richard K. Stucky 374

northwest Pakistan: dentition and classification. Paleontology. Pro- Green River Basin, southwestern Wyoming, Part 7: Survey ofBridge-
ceedings, Koninklijke Nederlandse Akademie van Wetenschappen, rian Artiodactyla, including description of a skull and partial skeleton
Ser. B, 86, 158-80. of Antiacodon pygmaeus. Contributions to Biology and Geology of
Van Valen, L. (1971). Toward the origin of artiodactyls. Evolution, 25, the Milwaukee Public Museum, 56, 1-47.
523-29. West, R. M., & Atkins, E.G. (1970). Additional middle Eocene (Bridge-
Webb, S. D., & Taylor, B. E. (1980). The phylogeny of higher ruminants rian) mammals from Tabernacle Butte, Sublette County, Wyoming.
and a description of the cranium of Archaeomeryx. Bulletin of the American Museum Novitates, 2404, 1-26.
American Museum of Natural History, 167, 117-58. West, R. M., & Dawson, M. R. (1975). Eocene fossil Mammalia from the
West, R. M. ( 1973 ). Geology and Mammalian paleontology of the New Fork- Sand Wash Basin, northwestern Moffet County, Colorado. Annals of
Big Sandy area, Sublette County, Wyoming. Fieldiana (Geology), 29, the Carnegie Museum, 45, 231-53.
1-193. Westgate, J. W. ( 1994). A new leptochoerid from middle Eocene (Uintan) de-
(1982). Fossil mammals from the Lower Buck Hill Group, Eocene posits of the Texas coastal plain. Journal of Vertebrate Paleontology,
of Trans-Pecos Texas: Marsupicamivora, Primates, Taeniodonta, 14, 296-9.
Condylarthra, bunodont Artiodactyla, and Dinocerata. The Pearce- Wortman, J. L. (1898). The extinct Camelidae of North America and some
Sellards Series, Texas Memorial Museum, 35, 1-20. related forms. Bulletin of the American Museum of Natural History,
(1984). Paleontology and geology of the Bridger Formation, southern 10, 93-142.
24 Entelodontidae

JAMES A. EFFINGER

INTRODUCTION

The Entelodontidae were bunodont artiodactyls known from Eu-

rope, Asia, and North America. In North America, the family ranged
from late Eocene (Duchesnean) to early Miocene (Hemingfordian),
and they were most common in floodplain environments of the north-
ern Great Plains region. Adult entelodonts ranged in size from about
3 to 6 feet at the shoulder and are distinguished by jugals with ven-
trolaterally expanded processes and paired mental and mandibular
tubercles. They were probably omnivorous animals that fed at least
as much on flesh as on vegetation. The Entelodontidae left no de-
scendants, and their ancestry is uncertain because they were already
relatively derived animals at their earliest appearance. Their great-
est diversity and abundance (number of fossil remains) was in the
Orellan and Whitneyan (early Oligocene).
The entelodonts share some distinct and unusual features that
readily set them apart from other bunodont artiodactyls. In contrast, Figure 24.1. Restoration of Archaeotherium, by Brian Regal.
Brachyhyops shares few of these distinctive characteristics and yet
has generally been considered an entelodont since after its origi-
nal description by Colbert (1938). Brachyhyops is included in this skull processes take the form of one to two pairs of mandibular
chapter because specimens referred to it are entelodontoid in form, bosses, or tubercles; one pair of these is located below p3-p4 on the
but this author is not convinced that it is an entelodont, so this genus ventrolateral edge of the ramus, and there is almost always another
is considered in this chapter as "Suina indet." pair on the anterior end of the jaw below c 1-p 1.
The occiput is relatively broad due to the lateral expansion of the
squamosals, which form the high posterior walls of the large tempo-
DEFINING FEATURES OF THE ral openings. The occipital condyles and the foramen magnum are
FAMILY ENTELODONTIDAE distinctly small for the size of the skull, as is the braincase. The large
antorbital foramen is located above P3 and is just posterior to the
marked constriction of the snout, which is always situated at the level
CRANIAL
of P2. The glenoid fossa is relatively shallow and laterally elongate,
The preorbital skull length is distinctly greater than the postorbital as is the corresponding mandibular condyle. On the mandible, the
length, and the skull is distinguished by the previously mentioned masseteric fossa is deep and well defined, the coronoid process is
processes that are unique to this family. The jugals are expanded into low and well separated from the condyle, and the mandibular angle
relatively flat and broad alar processes, or flanges, that are directed is rugose and slightly everted. The ram us is of fairly uniform depth
ventrolaterally and vary both in morphology and anterior-posterior throughout and the dentaries distinctly diverge posteriorly.
orientation. Certain edentates also possess ju gal processes, but these Regarding Brachyhyops, the only skull known for this genus is
are ventrally directed and are rodlike rather than broad. The other that of the type specimen. In this specimen, the postorbital skull

375
 James A. Effinger 376

Figure 24.2. Entelodont osteology, all adapted from Peterson (1909). A. Skeleton of Dinohyus hollandi (scale bar= approx. 18 cm). B. Skull of Ar-
chaeotherium mortoni (scale bar= approx. 30 mm). C. Typical entelodont dentition illustrated by Entelodon magnum (scale bar= approx. 17 mm).

length slightly exceeds the preorbital length, which is in contrast to The upper molars are brachydont and six cusped, except for M3,
typical entelodonts. The Brachyhyops type specimen represents a which is usually reduced posteriorly in cusp size and/or number.
very mature individual and still has only an incipient jugal process Ml and M2 are subquadrate in outline and are characterized by
that is no larger than those of some of the most juvenile entelodont low, rounded to subconical cusps. The M3 shares with the Ml-M2
genera. Also, all of the jaws attributed to Brachyhyops are referred only the morphology and placement of the anterior cusps, is usually
specimens because there was no mandible associated with the type subtriangular to subrhomboid in outline, and is smaller than the
skull. In all of these referred specimens, the mandibles are short and other molars.
do not have any tubercles. The lower molars are morphologically very similar to each other
and vary in size (increase posteriorly), in wear (decrease poste-
riorly), and in "hypoconulid" development (increase posteriorly).
DENTAL
The crowns are normally quadritubercular, except in the unworn
The incisors are pointed, round to subtriangular in horizontal sec- state when the para- and metaconids are usually distinct. Dinohyus
tion, and usually possess cingula on their posterior sides. Incisor size presents the only exception to this feature with indistinguishable
more or less regularly increases from the Il/i 1 to the I3/i3; the third paraconids and metaconids in the unworn state. Also, the trigonid is
incisor may be up to one half the height of the canines. The canines higher than the talonid and the crown is subrectangular in horizontal
are large, pointed, recurved, and serrated on the posterolateral and section.
posteromedial edges (in the unworn stages). The premolars are sim- The elongate face of entelodonts is accompanied by diastemata,
ple, laterally compressed conical teeth, especially the anterior ones. principally between the first and second premolars and the second
The Pl is usually situated at an angle to the sagittal plane of the and third premolars. The pointed third incisors, canines, and anterior
skull. The P2/p2 and P3/p3 are essentially enlarged versions of the premolars impart a carnivorous image to the skull, and none of the
Pl/pl, except that the P2/p2 have weak cingula and weakly serrate genera depart from the primitive eutherian dental formula.
ridges on the posterior sides, and the P3/p3 are usually distinctly The type specimen of Brachyhyops is so mature that it has no pre-
more robust. The p3 is the second tallest tooth in the lower arcade served crown morphology on the remaining teeth, but the upper and
and, along with p4, has a distinct heel. The P4 is subquadrate in lower cheek teeth may have been similar to those of Archaeotherium
horizontal section and is transversely bicuspate. based on referred upper cheek teeth from Utah and the referred
 Entelodontidae 377

mandible from Saskatchewan. Also, because of the shorter face of "Bunodontia" (bunodont artiodactyls) including helohyids, anthra-
Brachyhyops, there are no dental diastemata. cotheres, and hippopotami. Despite their claim for a close cladistic
grouping of the entelodonts with the suoids, a number offeatures of
entelodonts distinguish them from the suoids, including relatively
POSTCRANIAL
simpler cheek tooth crest patterns (especially compared to the suids ),
The neck is of moderate length and appears lightly structured con- more elongate limbs, functionally didactyl feet, and a dentition and
sidering the size of the skull it supported. Support was apparently by masticatory apparatus that is much more camivorelike in form and
large muscles and tendons attached, among other places, to the high function. The Entelodontidae left no known descendants, and their
neural processes of the anterior thoracic vertebrae, similar to those ancestry still remains obscure.
of present-day Bison (American bison) or Ceratotherium (white The origin of the Entelodontidae appears to be in the early Ter-
rhinoceros). The tail is also of moderate length and consists of about tiary of eastern Asia. Eoentelodon, from the late Eocene of this area,
fifteen vertebrae. The limbs are elongate and slender relative to the is the oldest and most plesiomorphic of the undisputed entelodont
size of the body. The feet are didactyl and the podials are unfused. species and is regarded as ancestral to later North American and
The radius and ulna are co-ossified, whereas the tibia and fibula vary, European species. With its short face, absence of tubercles, and
being co-ossified in Dinohyus, but not in Archaeotherium. The ster- very undeveloped jugal processes, Brachyhyops is plesiomorphic
num consists of six stemebrae and is unusual in the thinness of the to typical entelodonts (even Eoentelodon), but its contemporaneity
segments and the relatively great vertical dimension of the anterior withArchaeotherium in the Duchesnean would lead one to Wilson's
ones, especially the prestemum (Scott, 1940). The postcranial skele- (1971) implication that Brachyhyops and Archaeotherium were di-
ton of Brachyhyops is unknown, so the similarities and differences vergent forms from a common ancestor rather than the former being
between this genus and mainstream entelodonts in regard to fusion, ancestral to the latter.
diminution, and relative proportions of tarsal and limb elements are Achaenodon was selected as the sister group to the Entelodon-
also unknown. tidae in Figure 24.3 because in certain ways it resembles the en-
telodonts more in overall morphology than does Brachyhyops.
However, Achaenodon is too derived in other features (reduced pre-
SYSTEMATICS molar count, hypoconulid on m3) to be ancestral to the entelodonts,
especially if one includes Brachyhyops. Peterson (1909) attributed
SUPRAFAMILY similarities between Achaenodon and the entelodonts to parallelism
rather than close relationship.
The relationship of entelodonts to other bunodont artiodactyls is
by no means agreed upon. Carroll (1987) presented the more tra-
INFRAFAMILY
ditional relationship in which the entelodonts are grouped with
the cebochoerids, choeropotamids, and leptochoerids in the super- The first publication on entelodonts was that of Pomel (1847), who
family Entelodontoidea, which in tum shares the suborder Suina described and named Elotherium magnum from France.
with the other superfamilies Suoidea (suids and tayassuids) and the Aymard (1848) described and named Entelodon magnum from the
Hippopotamoidea (hippopotamids, anthracotheriids, and haplobun- same area, but because Pomel's type specimen was not illustrated,
odontids ). Gentry and Hooker (1988) presented a different view was materially inadequate, and is now lost, Entelodon is the generic
based on cladistic analysis. Their results link the entelodonts most name used for European forms and the basis for the family name. The
closely with the Suoidea. This grouping of the entelodonts with the first North American form, Archaeotherium mortoni, was named
suoids is one third of a trichotomy represented also by a grouping and described by Leidy (1850). Seven genera and nine species were
of cebochoerids with choeropotamids and a grouping of all other named in North America before any were even informally placed

Achaenodon Brachyhyops Archaeotherium Dlnohyus Choerodon Megachoerus Figure 24.3. Interrelationships of the
North American entelodont genera.
Key to characters at the nodes. (1)
6 P4 transversely bicuspate; nonmolar-
ized premolars; p2 much shorter than
5
p3; Ml and M2 with cingulum around
4
base ofmetacone. (2) ENTELODON-
TIDAE: Posteriorly enclosed orbit;
2 skull broad due to lateral expansion of
zygomatic arches; pterygoid midline
2
synarthrosis. (3) Face elongate with
antorbital skull length greater than
postorbital; presence of jugal flanges;
one to two pairs of mandibular tubercules; orbit posterior to M3; p3 tallest tooth after canine; jaw condyle at level of tooth row; low coronoid process. (4)
M1 without distinctly labial notch; Ml and M2 distinctly squared off in outline; Ml and M2 cusps fairly low and blunt. (5) Ju gal flange large relative to
skull and bent (distal portion directed more posteriorly than proximal portion). (6) P4 without anterior notch.
 James A. Effinger 378

into a family. Alston (1876) placed Elotherium (apparently not rec- CP83A, CP83B); A. coarctatum (localities CP39C, CP39F,
ognizing the other genera) in the "Elotheriidae" along with Helo- CP83A, NPlOB); A. trippensis (locality CP87A); A. lem-
hyus and Parahyus. The family name used subsequently, however, leyi (locality CP84B); A. ramosum (localities CP68B,
is Lydekker's (1883) Entelodontidae. Peterson (1909) was the only CP68C); A. potens (localities CP68B, CP68C)
author to write formal revision of the family. Of the nine genera and Archaeotherium sp. is also known from localities GC22,
sixteen species that existed in the literature at that time, Peterson CP40A,CP83B,CP98B.
declared two species indeterminate and dropped two genera to sub-
genus rank. Since Peterson's work, the literature representation of Megachoerus Troxell, 1920 (synonym: ?Boochoerus)
entelodonts consists primarily of papers by Troxell, Sinclair, Scott, Type species: M egachoerus zygomaticus Troxell, 1920.
and Russell and of various descriptions and name assignments in Type specimen: YPM 10008.
the context of faunal lists. Characteristics: Jugal flange large relative to skull and bent
and/or with greatly swollen distal end; Ml-M2 subquadr-
ate, but with distinctly squared-off outline; weak to virtu-
INCLUDED NORTH AMERICAN GENERA ally absent labial notch on M 1; P4 without anterior notch;
IN THE FAMILY ENTELODONTIDAE overall adult size consistently large.
Average length of m2: 39.8 mm.
There is no single reference in which descriptions of all of the Included species: M. zygomaticus only (= M. praecursor
entelodont taxa may be found. However, principal works include ?Archaeotherium scotti), known from locality CP84B only.
those of Peterson ( 1909), Troxell (1920), Sinclair ( 1921 ), and Scott
Dinohyus Peterson, 1909 (synonyms: ?Daeodon,
(1940). Of the 12 genera and 32 species of entelodonts named to
?Ammodon)
date, only 4 genera (5 if one counts Brachyhyops) are recognized
Type species: Dinohyus hollandi Peterson, 1905.
in this chapter. Some of the genera and many of the species are Eu-
Type specimen: CM 1594.
ropean and Asian; others were based on inadequate fossil material
Characteristics: No distinct para- and metaconids on unworn
and have been combined into others. The listing of principally White
lower molars; jugal flange proportionally small; posterior
River taxa follows the conclusions of Effinger ( 1987), although that
end of jugal forms strong preglenoid process; mandibu-
research is incomplete. Conclusions regarding other taxa follow
lar tubercles small and anterior ones are smaller than the
those of their principal workers or authors unless otherwise stated.
posterior ones; relatively short alveolar border on pre-
It seems likely, for instance, that at this point Choerodon will eventu-
maxilla causing crowding of incisors; tibia and fibula co-
ally be synonymized into Megachoerus, but until the type specimen
ossified; trapezium absent and metatarsal V sometimes
is examined, Choerodon is here considered a separate genus based
absent (Peterson, 1909).
on Troxell's (1920) observations.
Average length ofm2: 47 mm.
The locality numbers listed for each genus refer to the list of
Included species: D. hollandi (known from localities ?GC2,
unified localities in Appendix I. Question marks in front of the
GC3A, GC3B, GC4A, GC24II, CP85C, CP88, CP104A,
locality (e.g., ?CPlOl) mean the taxon is questionably known from
CP104B, NClA); ?D. mento (locality GC25).
that locality. The acronyms for museum collections are listed in
Dinohyus sp. is also known from localities GC5, GC8C,
Appendix III.
CP103, CP105.

ENTELODONTIDAE Choerodon Troxell, 1920

Type species: Choerodon caninus Troxell, 1920.
Archaeotherium Leidy, 1850 (synonyms: Scaptohyus, Type specimen: YPM 11665.
Pelonax) Characteristics: Jugal flange large, bent with thickened end
Type species: Archaeotherium mortoni Leidy, 1850. and thin anterior edge; posterior end of jugal forms anterior
Type specimen: ANSP specimen (number uncertain). edge of glenoid fossa; supraorbital rim exceptionally thick;
Characteristics: Jugal flange relatively straight with taper- short diastemata; P4 with anterior notch; Ml-M2 as in
ing thickness to slightly swollen distal end; Ml-M2 sub- Archaeotherium, except for virtually absent notch on Ml.
quadrate; Ml with distinct labial notch between anterior Average length of m2: 36 mm.
and posterior halves; unworn lower molars with anterior Included species: C. caninus only, known from locality PN6C
cusps much high than posterior ones; trapezium present; only.
tibia and fibula not co-ossified. Choerodon sp. is also questionably known from locality
Average length ofm2: 27.5 mm. PN6E.
Included species: A. mortoni (=A. clavum, ?A. crassum)
(known from localities SB44B, SB44C, SB44E, CP41A,
"SUINA INDET."
CP42A, CP83A, CP83B, CP84A, CP98C, CP99A); A.
wanlessi (= ?A. marshi) (localities CP84A, CP84B, Brachyhyops Colbert, 1938
CP99A); A. latidens (= ?A. scotti) (localities CP84B, Type species: Brachyhyops wyomingensis Colbert, 1938.
 Entelodontidae 379

w
zw L L.81.
1.8
2.4
E.81
0
0 E Let. Hp.
4.5

:i L.Hp.
5.2
a. L.E.Hp.
6.0
7.0
E.E.Hp.
L 8.8
L.CI.
9.5

E.CI.
11.0
w
zw LL.Ba.
12.5
0 M
E.L.Ba.
0 4.0

:ill E.Ba.

5.8
L.Hm.
7.5
E.Hm.
8.8
E L.L.Ar.
19.2

E.L.Ar.
23.0
I
w
zw LE.Ar. e
.::s ?1
;::
0
L
- - - - - - - - - :S
Q) 2
0
CJ
:i
E.E.Ar.
27.7

j -i -Q)-
Q

0
Wt.
28.4
~
'o::t
-o
.5
Q
-1[
~
E 31.9
Or.
33.4
L.Ch.
34.5
L M.Ch.
35,5
E.Ch.
37.1

Ou I?
39.5
_J
1
L.Un.

w 41.3

zw E.Un.

0 45.9
0
w
--+-
L.Br.
M 46.7

M.Br.
ENTELODONTIDAE OR SUOIDEA INCERTAE SED/S
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
55.0
E.E.Wa.
55.5
Cl<.3 55.6
w Ck.2 55,B
zw Ck.1
Tl.6
Tl.6
56,D
56.2
L 57.0
0 Tl.4
Tl.3
57.3
0 Tl.2
w Tl.1
..J 60.9
To.3
<C
a. To.2
62.5
To.1
E Pu.3
Pu.2 64.5
Pu.1 65.0
Pu.O 65.1

Bl.=Blancan, Hp. = HemJ>hillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar.= Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un. = Uintan, Br.= Brldgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 24.4. Temporal ranges of North American entelodont genera.

 James A. Effinger 380

Type specimen: CM 12048. vational evidence for sociality among entelodonts, but they most
Characteristics: Broad cranium with widely separated pari- likely had at least an association between females and offspring as
etal crests; orbit closed posteriorly and situated above in present-day suines.
M2-M3; postorbital skull length somewhat exceeds preor- Based on the predominant stratigraphic occurrence of entelodont
bital length; zygomatic arch vertically expanded; glenoid remains, their habitat preference was floodplain. As generalized
fossa broad, shallow, and at the same level as the occlusal omnivores, diet would not have restricted them geographically nor
surface of the upper cheek teeth; paraoccipital processes certainly would their size; this may help to explain their wide distri-
short; pterygoids weak; posterior nares extend to above bution in North America as well as the rest of the Northern Hemi-
M3; arrangement ofbasicranial foramina similar to that of sphere. The cause of their extinction in the early Miocene is uncer-
the dichobunids. tain, but could be related to the conservative retention of a distinctly
Average length of m2: 18.6 mm. small brain in a time when large carnivorous predators, such as Am-
Included species: B. wyomingensis (known from localities phicyon and Hemicyon, were appearing and presenting pressures
SB26B., SB44B, CP7C, CP39IIA); B.(?) viensis (locality of competition and/or predation. The disappearance of entelodonts
NPIOA). may also be correlated in some unknown way with their large size
Brachyhyops sp. is also known from locality NP29A. or changes in climate and/or habitat in the early Miocene.

INDETERMINATE ENTELODONTIDS
REFERENCES
Fragmentary remains ascribed to entelodontids are also known from
localities CP52, CP84C, NPIOC, PN6B, PN6D. Alston, W. (1876). Zoological Record. (Use of"Elotheriidae" in systematic
listing of taxa.)
Aymard, A. (1848). Essai monographique sur un nouveau genre de mam-
mifere fossile trouve dans la Haute-Loire, et nomme Entelodon. An-
BIOLOGY AND EVOLUTIONARY PATTERNS na/es de la Societe d'Agriculture, Sciences, Arts et Commerce du
Puy 12 (for 1842-46), 227-67.
Entelodonts were most likely omnivores as evidenced by cranial and Carroll, R. L. (1987). Vertebrate Paleontology and Evolution. New York:
dental features. These include a laterally unrestricted jaw movement Freeman.
Colbert, E. H. (1938). Brachyhyops, a new bunodont artiodactyl from Beaver
in the glenoid fossa (as in most herbivores), a condyle at the same
Divide, Wyoming. Annals of the Carnegie Museum, 27, 87-108.
level as the tooth row (carnivores), pointed incisors, and pointed and Effinger, J. A. (1987). Systematics and Paleobiology of White River
serrate canines and premolars (carnivores). The diet of entelodonts Group Entelodontidae (Mammalia, Artiodactyla). Unpublished M.S.
may have been similar to those of most Recent pigs (suids) and pec- Thesis. Rapid City: South Dakota School of Mines and Technology.
caries (tayassuids), that is, primarily vegetarian diets supplemented Gentry, A. W., & Hooker, J. J. (1988). The phylogeny of the Artiodactyla.
by carrion and fresh "kill," such as reptiles, eggs, young birds, and In The Phylogeny and Classification of the Tetrapods, Vol. 2: Mam-
mals, ed. M. J. Benton. Systematics Association Special Volume, 35B,
insect larvae (Walker, 1964). There are, however, significant dif- pp. 235-72. Oxford: Clarendon.
ferences in the skulls and dentitions of these forms that indicate Joeckel, R. M. (1990). A functional interpretation of the masticatory system
differences in their dietary preferences. Based on analysis of tooth and paleoecology of entelodonts. Paleobiology, 16, 459-82.
wear and function of the masticatory apparatus, Joeckel (1990) con- Leidy, J. (1850). On two genera of fossil mammals, Eucrotaphus and Ar-
chaeotherium. Proceedings of the Academy of Natural Sciences in
cluded that entelodonts were suited to the ingestion of resistant and
Philadelphia, 5, 90-3.
possibly large food items, including bones. The more carnivorous Lydekker, R. (1883). Siwalik selenodont Suinae. Paleontologica Indica, Ser.
aspect of the jaws and teeth of entelodonts relative to the suoids may 10, 2, 146.
indicate that flesh made up a relatively greater portion of their diet. Peterson, 0. A. (1905). Preliminary note on a gigantic mammal from the
A number of entelodont specimens exhibit marks on the skull Loup Fork beds of Nebraska. Science, new series, 22, 211-12.
which have been interpreted (e.g., Sinclair, 1921) as "battle scars" (1909). A revision of the Entelodontidae. Carnegie Museum, Memoirs,
4, 41-158.
that attest to the belligerence of entelodonts. It is possible, though Pomel, A. (1847). Note surun nouveau pachyderme du bassin de la Gironde
improbable, that the marks were made by the canines of a predator, (Elotherium magnum). Bulletin Societe Geologique de France, Ser.
such as Hyaenodon. The canines of Hyaenodon are not the cor- 2, 1083-5.
rect shape, however, for the marks on the wounded specimens. The Scott, W. B. (1940). Artiodactyla. In The Mammalian Fauna of the White
River Oligocene, Part IV, eds. W. B. Scott and G. L. Jepsen. American
marks were most likely made by other entelodonts during mating
Philosophical Society, Transactions, 28, 363-746.
or territorial fights.
Sinclair, W. J. (1921). Entelodonts from the Big Badlands of South Dakota
Sexual dimorphism in entelodonts is, at this time, only inferable in the Geological Museum of Princeton University. American Philo-
and not demonstrable. It is likely that jugal flanges, mandibular tu- sophical Society, Proceedings, 60, 467-95.
bercles, and canines may prove to be sexually dimorphic features Troxell, E. L. (1920). Entelodonts in the Marsh Collection. American Journal
of Science, 50, 243-55, 361-86, 431-55.
when a greater number of entelodont specimens are examined than
Walker, E. F. ( 1964). Mammals of the World, V, II. Baltimore: Johns Hopkins
are currently available. These features were not functionally inte-
University Press.
grated into skull and jaw musculature and were thus most probably Wilson, J. A. (1971 ). Early Tertiary vertebrate faunas, Vieja Group, Trans-
used for sexual or dominance display. Similarly, there is no preser- Pecos Texas: Entelodontidae. The Pearce-Sellards Series, 17, 1-17.
25 Anthracotheriidae

DONALD G. KRON and EARL MANNING

INTRODUCTION

Anthracotheres were a persistently primitive group of browsing ar-

tiodactyls that achieved wide distribution across Eurasia and parts
of Africa during much of the Tertiary and Pleistocene. Their North
American record is more restricted: It ranges from the late mid-
dle Eocene to the mid-Miocene and is concentrated primarily in
South Dakota and adjacent states. They were rather large and heav-
ily built animals for the mid-Tertiary, and their low-crowned teeth
and frequent occurrence in paleochannel deposits suggest habits and
habitat similar to those of modem hippos. Their decline and even-
tual extinction in the Old World correlates with the later radiation
of hippos there'. Their earlier disappearance in North America may
have resulted from habitat loss and lack of diversity.
The fossil record of North American anthracotheres is neither
abundant nor very diverse except for the Whitneyean "Protoceras
Channels" in the Big Badlands of South Dakota, a circumstance
that probably reflects ecological and sedimentological bias. Two
subfamilies are recognized in North America, containing six genera
and fourteen species, plus one genus and species as Anthracotheri-
Figure 25.1. Restoration of Elomeryx armatus, by Brian Regal.
idae incertae sedis. Both subfamilies are immigrants with origins in
Eurasia.
directed extension from the palatines. The mandible (Heptacodon
is again the exception) has a strongly downtumed angle, and a broad,
DEFINING FEATURES OF THE long symphysis.
FAMILY ANTHRACOTHERIIDAE
DENTAL
CRANIAL
The family is characterized by an unreduced dental formula (loss of
Except for the more heavily built Heptacodon, the anthracothere Pl is an occasional exception), persistent brachydonty, progressive
skull is characteristically long and rather slender, the snout is variably development of diastemata between the anterior cheek teeth with an
elongate, the braincase is narrow, the saggital crest prominent, and accompanying elongation of the skull, and nonmolariform premo-
the occipital sharply underslung beneath the supraoccipital crest. Or- lars. Upper molars are large, nearly square, and are usually (excep-
bits usually do not extend forward of M3, ai,d the postorbital bar is tion: Arretotherium) five cusped with a prominent paraconule. They
incomplete. The mastoid is not exposed externally. Internal nares are combine pyramidal to moderately crescentic inner cusps with a W-
located slightly to well behind M3 and in some cases (Bothriodon shaped ectoloph that includes a prominent mesostyle and parastyle.
and Aepinacodon) are partially bisected by a· medial, posteriorly The protocone lacks a posterior crest, so a transverse valley extends

381
 Donald G. Kron and Earl Manning 382

Figure 25.4. Skeleton of Elomeryx armatus. Original about 1. 7 meters high

(from Scott, 1894).

Figure 25.2. Skull of E/omeryx armatus. (After Scott, 1940.)

longer. The manus is five toed, the pes has four, and metacarpals
and metatarsals are short and unfused. The astragalus is hippolike
and the patella unusually large.

SYSTEMATICS

SUPRAFAMILY
The first description of a North American anthracothere was by
Leidy in 1856 (as "Hyopotamus" [now Aepinacodon] americanus),
Figure 25.3. A. Upper dentition of Heptacodon curtus. Scale bar equals and some years later (1869) he proposed the family Anthracotheri-
5 cm (after Scott, 1940). B. Lower dentition of Heptacodon curtus. (After
idae defined by its generic content. Gill ( 1872) emended the spelling
Scott, 1940.)
and formally diagnosed the family, based on the upper dentition and
suiform body shape with semiunguligrade feet having reduced lat-
from the lingual margin to the mesostyle. There is a progressive eral digits. Anthracotheres are dentally quite distinctive and, by and
tendency for the ectoloph to "invade" the mesostyle, thus length- large, there has been little confusion about just what does, or does
ening the valley. Upper and lower canines are sometimes large and not, belong in the family. Higher relationships of the anthracotheres
tusklike, depending on genus and/or sex. Lower molars lack the have proven another matter: Because of their generally primitive
paraconid and increase in size posteriorly, with m3 by far the largest structure, anthracotheres have historically been shuttled in and out
in the series. Strong crests on p4 connect the protoconid with the of relationship with a variety of other artiodacty 1 families at a vari-
paraconid and metaconid, the latter often reduced. Lower incisors ety of taxonomic levels. Gill, for example (1872), grouped anthra-
are spatulate. cotheres, pigs, and hippos as superfamilies within the Omnivora, or
One of us (Manning) notes that previous characterizations of the nonruminants, but Osborn later (1910, pp. 547-548) grouped them
family have been just that: characterizations based on an array of apart from the Suina in the "Section Primitive Artiodactyls (families
primitive cranial and dental characters variably shared with other of more or less uncertain affinities)."
artiodactyl families (some, such as bunodonty, with the entire Ar- Matthew (1929) somewhat dubiously united the anthracotheres,
tiodactyla), but not necessarily present in all anthracotheres. Addi- oreodonts, anoplotheres, and caenotheres as the suborder Anco-
tion of the derived character of strong crests on p4 serves to unite donta, defined on the basis of bunoselenodont to selenodont teeth
the two subfamilies, and, by extension, to complete the diagnosis and tetradactyl (sic) feet. Citing their more specialized molars,
of the Anthracotheriidae as a monophyletic group apart from the Matthew placed pigs and hippos together as the Hyodonta. It is note-
Hippopotamidae and Suidae, with which anthracotheres have tradi- worthy that Matthew partially agreed with Stehlin (1910) and took
tionally been grouped at one taxonomic level or another. the uncommon position of separating dichobunids and entelodonts
(as the Palaeodonta, but not to be confused with Romer's (1945,
1966) category of the same name, which is philosophically more
POSTCRANIAL nearly akin to Osborn's (1910) "Section Primitive Artiodactyls")
The anthracothere skeleton has been most commonly compared to from all other artiodactyls because they lack the derived character
that of pigs, oreodonts, or hippos, the common thread being the of replacement of the upper molar hypocone by the metaconule.
robust, short-legged build. Absence of the graviportal specializa- Based on the work of Pearson (1927), Romer initially (1933)
tions seen in hippos and absence of the fully unguligrade stance of divided the Artiodactyla into the suborders Suina and Ruminan-
pigs means that overall proportions most closely resemble those of tia depending on whether the mastoid was hidden or exposed in
oreodonts. Cervical spines are shorter than in pigs, and the neck is external view. The Suina united anthracotheres with pigs, hippos,
 Anthracotheriidae 383

Figure 25.5. Cladogram showing hypothesized relationships of North American genera of anthracotheres. Key to characters on cladogram.
(1) ANTHRACOTHERITDAE: The p4 with crest connecting protoconid and hypoconid; p4 metaconid reduced, incorportated into crest connect-
ing metaconid and protoconid. (2) Face shortened, diastemata closed; zygoma deep; canines set laterally; mandibular angle reduced and symph-
ysis upturned, {orming "chin." (3) ANTHRACOTHERITNAE: Lateral styles of upper molars invaded (ectoloph fully W-shaped); P3 with strong
protocone; ml-m2 hypoconulids reduced. (4) Large size; tubular rostrum; M3 with large styles. (5) Orbits retracted, not extending anterior to
M3; Pl-P2 shortened; mandibular angle strongly downturned; lateral styles on upper molars reduced. (6) Posterior medial projection of palatines
absent; internal nares opening behind M3; orbits relatively more dorsally located. (7) BOTHRIODONTINAE: Mesodont; Ml-M2 mesostyle di-
vided, selenes separate and anteroposteriorly compressed; internal cingulum reduced on upper molars, lateral styles further reduced; auditory bul-
lae expanded, lambdoid crest reduced. (8) Facial length reduced; cheek teeth with relatively high, sharp cusps; Ml-M3 without protoconule.
(9) Cl-Pl and Pl-P2 diastemata lengthened (face lengthened); mandibular angle reduced; Pl, male canines, and lingual cingulum of upper molars
all reduced. (10) Rostrum very elongate with pronounced diastema; P3 protocone reduced. (Bothriodon americana only.)

entelodonts, achaenodonts, and cebochoerids, with the cebocho- equal rank. The Ancodonta was further refined by Scott (1940), who
erids shown as the implied ancestral group for anthracotheres in excluded oreodonts and agriochoerids on the grounds that similari-
an accompanying phylogenetic chart. Romer later (1945, p. 443) ties to anthracotheres involved only retained primitive characters.
favored the homacodonts (including diacodectids) as probable an- Simpson's (1945) influential classification rejected the "mastoid-
thracothere ancestors. Romer's 1966 classification united anthra- amastoid" condition as a fundamental division in the Artiodactyla
cotheres and hippos as the superfamily Hippopotamoidea, with pigs and, instead, modified and greatly extended Matthew's (1929) clas-
and peccaries separated as the Suoidea, and entelodonts and cebo- sification. The Ancodonta and Palaeodonta were reduced to infraor-
choerids plus others as the Entelodontoidea, all within the Suina. dinal rank within the Suiformes (essentially "nonruminants"), which
A restricted Diacodectidae was implied as ancestral to the anthra- also included the Suina and Oreodonta. The familial contents of the
cotheres (p. 274). Ancodonta reflected Colbert's (1935b) and Scott's (1940) modifi-
Colbert (1935a, 1935b; 1938a, 1938b) discussed anthracotheres cations, but anthracotheres, hippos, and anoplotheres were grouped
in several contexts during the 1930s. Comparing at length the teeth, as the Anthracotheroidea apart from the Cainotheroidea. Simpson
skulls, and postcrania of fossil pigs, hippos, and anthracotheres, (1945) commented that hippos "might, indeed, be called specialized
he concluded (1935a, 1935b) on the basis of overall similarity surviving anthracotheres" (p. 262).
that hippos most probably arose from anthracotheres. With this in Coombs and Coombs (1977) considered anthracothere relation-
mind ( 1935b), he transferred hippos to Matthew's ( 1929) Ancodonta ships and concluded that helohyids (including Gobiohyus, men-
as the superfamily Hippopotamoidea and raised anthracotheres to tioned by Colbert [1935a, 1935b] as the cebochoerid most similar to
 Donald G. Kron and Earl Manning 384

anthracotheres) were the sister group of anthracotheres and should of a fully invaded mesostyle (that is, ectoloph fully W-shaped) on
be included in an expanded Anthracotheroidea. This conclusion is the upper molars and reduced hypoconulids on ml-2. Heptacodon
rejected here on the grounds that it was based on mistaken character is transferred to the Anthracotheriinae as the sole North Ameri-
polarities and shared primitive characters. The present classifica- can representative of that subfamily. Kukusepasutanka is a bothri-
tion (Figure 25.5) uses the derived character in anthracotheres of odontine sensu stricto, despite the peculiar tubular snout. Apriculus,
crests on p4 to remove them from their "traditional" association sometimes considered as Anthracotheriidae incertae sedis, is consid-
with hippos. An expanded Suoidea includes pigs, peccaries, and ered by Stucky (this volume, Chapter 23) to belong with the "hom-
hippos as the sister group of a restricted Anthracotheroidea (An- codont" primitive bunoselenodont artiodactyls and is discussed in
thracotheriidae only), all within a restricted infraorder Suina, from that chapter.
which entelodonts and other noneuartiodacty ls (sensu Stehlin, 1910)
are excluded.
INCLUDED NORTH AMERICAN GENERA
IN THE FAMILY ANTHRACOTHERI IDAE
INFRAFAMILY
Gill (1872) divided the Anthracotheriidae into two subfamilies based The locality numbers listed for each genus refer to the list of unified
on the degree of molarization of P4. Hyopotamus and Bothriodon localities in Appendix I. The acronyms for museum collections are
(includingAncodus) formed the Hyopotaminae, defined as sharing a listed in Appendix III.
molariform P4, andAnthracotherium and Elotherium (a junior syn- The locality numbers may be listed in a couple of alternative ways.
onym of Entelodon) made up the Anthracotheriinae, with simpler Parentheses around the locality (e.g., [CPlOl]) mean the taxon in
premolars. Two late and derived genera from Asia, Merycopotamus question at that locality is cited as an "aff." or "cf." the taxon in
and Choeromeryx, were not recognized as anthracotheres and were question. Parentheses are usually used for individual species, thus
excluded. implying the genus is firmly known from the locality, but the actual
Troxell (1921) briefly reviewed some questions of priority in his species identification may be questionable. Question marks in front
paper on North American forms. He emphasized that Hyopotamus of the locality (e.g., ?CPlOl) mean the taxon is questionably known
was a junior synonym of Bothriodon, making abandonment of Gill's from that locality, thus implying some doubt that the tax on is actually
Hyopotaminae necessary. At the same time, he erectedAepinacodon present at that locality, either at the genus or the species level.
to include North American species previously referred to Hyopota-
mus andAncodus. Scott (1940), however, droppedAepinacodon and ANTHRACOTHERI INAE
reverted to Bothriodon because new material showed the former was
poorly diagnosed and closer to the latter than Troxell had thought. Characteristics: transverse valley in upper molars extending at most
Scott also proposed two subfamilies to the Anthracotheriidae: the only part way into mesostyle, so that mesostyle is cuspate to partially
exclusively Old World Anthracotheriinae with pyramidal cusps on cuspate in shape. Hypoconulids prominent on ml-2.
their upper molars, and the mostly North American Bothriodontinae
with subcrescentic cusps. Heptacodon Marsh, 1894a (synonyms: Octacodon;
Simpson ( 1945) rejected Scott's use of subfamilies on the grounds Anthracotherium)
that a twofold division based on a single dental characteristic was Type species: Heptacodon curtus Marsh, 1894a.
probably too simplistic to be applied to all anthracotheres, given the Type specimen: YPM 2232.
very wide range of morphology displayed by Old World forms in Characteristics: rostrum short; zygoma deep; anterior margin
particular. Macdonald (1956) also avoided using subfamilies in his of orbit extending almost to level of M 1. Mandibular angle
general review of North American forms, noting that Heptacodon reduced; symphysis upturned. Canines set far laterally; up-
(including Octacodon Marsh, a new synonymy) was misplaced in per and lower anterior premolars without diastemata; up-
the Bothriodontinae as originally defined, and that Kukusepasu- per molars with mesostyle not invaded by transverse valley,
tanka, new genus, also belonged outside Scott's groupings. mesosty le prominently connate in consequence; ml-2 with
Macdonald disagreed with Scott on the Bothriodon-Aepinacodon prominent hypoconulids.
question, and resurrected the latter with an expanded diagnosis. He Average length of m2: 20.0 mm. Average length of M2:
recognized three species of Aepinacodon. Russell (1978) revived 20.0-28.0 mm, except for Heptacodon (undescribed
Bothriodon for two North American species, one (B. advena) new species), which has M2 length of 15.0 mm.
and one (B. rostratus) previously assigned by Macdonald toAepina- Included species: H. curtus (known from localities CP84B,
codon. He noted that differences between the two genera are slight, CP99A, CP99B); H. gibbiceps (locality CP84B); H. occi-
but that the rostrum of Bothriodon is sufficiently more elongate that dentale (locality CP84A); H. quadratus (locality CP84A);
the distinction needed to be made. H. pellionis (localities NP9A, NP9B); H. undescribed
The present classification (Figure 25.5) recognizes two subfami- species (locality GC 1).
lies as proposed by Scott, but the Bothriodontinae are differentiated Heptacodon sp. is also known from localities CP42A,
from the Anthracotheriinae by possession of the derived characters CP68C, ?CP83A.
 Anthracotheriidae 385

BOTHRIODONTINA E Characteristics: facial region shorter than in Elomeryx, di-

astema almost absent. Upper and lower cheek teeth with
Characteristics: canine tusklike; upper molar mesostyles fully in-
high, sharp cusps; protoconule lost in upper molars.
vaded by transverse valley; mesostyle not forming a fully devel-
Average length of m2: 23.5 mm (one specimen of A. fricki).
oped cusp in consequence, and ectoloph completely W-shaped.
Average length of M2: 24.0-26.0 mm.
Hypoconulids reduced on ml-2.
Included species: A. acridens (known from locality NP33B);
A. fricki (localities CP88, CP105, [NPlOD]); A. leptodus
Elomeryx Marsh, 1894c (synonyms: Heptacodon, in part;
(locality CP86C).
Hyopotamus, in part)
Arretotherium sp. is also known from localities GC5,
Type species: Elomeryx armatus Marsh, 1894b.
(CP99C).
Type specimen: YPM 10176.
Characteristics: Rostrum shorter with smaller diastemata
Kukusepasutanka Macdonald, 1956
than in Aepinacodon; posterior medial projection of
Type species: Kukusepasutanka schultzi Macdonald, 1956.
palatines absent; internal nares opening far behind M3;
Type specimen: F:AM 49463.
orbits set rather high on skull. Male canines with ser-
Characteristics: size large; rostrum tubular in cross section
rate posterior edge; P2-3 with stronger protocones than
P2-3 with large protocones; M3 with large styles.
in Aepinacodon.
Length ofM2 (type only): 29.9 mm.
Average length of m2: 24.0-28.0 mm. Average length ofM2:
Included species: K. schultzi only, known from locality
24.0-27 .0 mm.
NP36A only.
Included species: E. armatus (known from localities CC9,
CP84B, CP99B, NPlOC); E. garbantii (localities CP84C,
CP85C).
BIOLOGY AND EVOLUTIONARY PATTERNS
Elomeryx sp. is also known from localities CPlOl,
NPlOBB, NPlOCC.
Studies of the Big Badlands paleosols in South Dakota indicate that
the late Eocene through the Oligocene was a period of gradually
Aepinacondon Troxell, 1921 (synonyms: Hyopotamus, increasing aridity, from humid and subtropical through subhumid
in part; Ancodus, Ancodon, Bothriodon in part) and warm temperate to semiarid or arid in the Arikareean (Retallack,
Type species: Aepinacodon deflectus Marsh, 1890. 1983 ). Associated with this are a series of inferred changes from
Type specimen: YPM 11802. closed canopy woodlands through open woodland to savannalike
Characteristics: Face shorter than in Bothriodon, especially shrublands with forests concentrated mainly along stream courses.
in region of canine-Pl diastema. P3 with more prominent Concentration of anthracothere remains in the Orellan "Metamyn-
protocone. odon Channels" (H eptacodon and Bothriodon) and especially in the
Average length ofm2: 24.0-26.0mm. Average length ofM2: Whitneyean "Protoceras Channels" (Heptacodon and Elomeryx) of
26.0-30.0 mm. the Big Badlands has led to a general assumption that anthracotheres
Included species: A. deflectus (known from locality CP83C); were hippolike in their habits, a conclusion reinforced by the abun-
A. americanus (localities CP83B, CP98B, CP98C). dance of Arretotherium in riparian deposits at a Miocene locality
Aepinacodon sp. is also known from locality CP68B. (locality CP88) in Nebraska (Martin, 1985). Evidence for a truly
semiaquatic lifestyle as favored by Clark and Beerbower (1967,
Bothriodon Aymard, 1846 pp. 58, 68) is, however, equivocal. Modem hippos are specialized in
Type species: Bothriodon velaunus von Meyer, 1832. their lifestyle, and anthracotheres are often associated with taxa not
Characteristics: rostrum more elongate than inAepinacodon; believed to be aquatic. Wilson ( 1975) took a conservative approach
diastema between canine and P 1 more pronounced. P3 pro- and assigned the Protoceras Channel fauna as a whole to a stream
tocone reduced relative to Aepinacodon. border habitat, and briefly noted that some taxa might originally
Average length ofm2: 21.0-25.0 mm. Average length ofM2: have been forest dwellers by preference, but by the late Oligocene
25.0-30.0 mm. forests had shrunk to a point where the distinction could no longer
Included species: B. rostratus (known from localities CP84A, be made.
CP99A); B. advena (locality NPlOB). Black ( 1978) remarked that African anthracotheres display a
Bothriodon sp. may also be known from locality CP39C, tremendous amount of tooth wear in a pattern very similar to hippos,
and either Bothriodon sp. or Aepinacodon sp. is known and Macdonald (1956) also noted that North American forms often
from locality CP83A. obliterate the pattern on Ml before M3 erupts. Hippos are bun-
odont/bilophodont and somewhat hypsodont, however (Janis and
Arretotherium Douglass, 1901 Fortelius, 1988), and are primarily grazers (Walker, 1975), which
Type species: Arretotherium acridens Douglass, 1901. makes them a rather poor analogue for anthracotheres. Chadronian
Type specimen: CM 704. titanotheres are perhaps a better one because their upper molars
 Donald G. Kron and Earl Manning 386

w
zw L L.81. 1.8
2.4
E.BI
(.)
4.5
0 E Lst. Hp.
5.2
::::i L.Hp.
6.0
a. LE.Hp.
1.0
E.E.Hp.
8.8
L L.CI.
9.5

E.CI.
11.0
w
zw LL.Ba.
12.6
(.) E.L.Ba.
M
0 14.0

:i E.Ba.

15.8
L.Hm.
17.5
E.Hm.
18.8
E L.L.Ar.
19.2

E.L.Ar.
23.0

w LE.Ar.
zw
L
(.) 27.7
0
CJ E.E.Ar.

::::i c:
0 29.4
.g c:
E
Wt.
31.9 -~
1.; g0 .g,0
~ ....____ .5 r=-=-~ 1---------------- --1
--..
Or.
33.4

ti==~
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5

L.Un.
41.3
w
zw M
E.Un.

(.)
0
45.9 ANTHRACOTHERllNAE BOTHRIODONTINAE
w L.Br.
46.7

M.Br.
49.0
E. Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0
55.5
-
Ck.3 55.6
w Ck.2 55.8
zw Ck.1
Tl.6
Tl.5
56.0
56.2
L Tl.4
57.0
(.) 57.3
TI.3
0 Tl.2
58.5
w Tl.1
...I
ca:
a.
To.3
To.2
62.5
To.1
E Pu.3
Pu.2 64.5
Pu.1 65.0
Pu.a 65.1

Bl.=Blancan, Hp.= Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
= = =
Wt. =Whitneyan, Or.= Orellan, Ch. Chadronian, Du.= Duchesnean, Un. Uintan, Br. = Bridgerlan, Wa. Wasatchian,
= = = =
Ck. Clarkforkian, Ti. Tiffanian, To. Torrejonian, Pu. Puercan.

Figure 25.6. Temporal ranges of North American anthracotheriid genera.

 Anthracotheriidae 387

feature a W-shaped ectoloph similar to that of anthracotheres and Riparian habitats were shrinking during much of the later Tertiary,
also wear sequentially although not to such an extreme. In both and the seemingly inefficient anthracothere dental battery (in which
cases, some sort of abrasive browse is indicated. An exception to the first molars are nearly obliterated before the third molars show
this may be Arretotherium, which alone among North American much wear) may not have been able to cope with concomitant veg-
forms had a skull with a broad, square muzzle reminiscent of hip- etational changes. Janis and Fortelius (1988), however, pointed out
pos (compare the reconstruction of Elomeryx in Figure 25 .1 with that this is not necessarily a disadvantage. A number of extant her-
that of Arretotherium in Martin [1985, Figure 25]). Among mod- bivores, including both browsing and grazing forms and certain
ern rhinoceroses, a similar adaptation is present in the only graz- species of hippos and suids, compensate for tooth wear by sequential
ing species (Ceratotherium, the White, or Square-Lipped, Rhino). or delayed tooth eruption. As the anterior teeth wear out, the emerg-
Other species all have narrower and more mobile lips that they use ing last molars (typically enlarged) take a progressively greater role
to browse in a more selective manner. in food comminution and the result is no net loss of feeding effi-
Most anthracotheres show a moderate amount of sexual dimor- ciency. One factor that almost certainly played a part in the earlier
phism in which individuals adjudged to have been male have larger extinction of anthracotheres in North America than in the Old World
canines than the females. Macdonald and Martin (1987) drew at- was loss of species diversity. The family (never particularly diverse
tention to some variation in the relative development of the cra- at any time) was reduced to one spottily distributed species by the
nial "shield" in Arretotherium specimens, suggesting that it, too, early Hemingfordian (Arretotheriumfricki), and was consequently
may have been somewhat more prominent in males. The relative vulnerable to events such as habitat change, disease, or competition
size differential of the canines does not approach that seen in ei- for resources from more diverse groups such as the contemporary
ther hippos or pigs, however, so there is some question as to their rhinoceroses. The early Hemingfordian-late Hemingfordian transi-
importance in display, intraspecific combat, or defense. Nor, using tion is marked by considerable fauna!, biologic, and environmental
the same groups as analogues, can reliable guesses be made con- alterations (Tedford et al., 1987), and it is quite possible that the
cerning either the social or reproductive habits of anthracotheres: anthracotheres were in some manner victims of the changing times.
Depending on the species, both hippos and pigs range from solitary
to gregarious and may bear from one to several young.
REFERENCES
The earlier (pre-Arikareean) evolutionary history of North Amer-
ican anthracotheres is a generally conservative one in which several
Aymard, A. (1846). Essai monographique sur un nouveau genre de mam-
species of the primitive anthracotheriine Heptacodon coexisted with mirere fossile dans la Haute-Loire, et nomme Entelodon. Anna/es
the more derived bothriodontines for most of the late Eocene and de la Societe d'agriculture, sciences, arts et commerce du Puy, 12,
Oligocene. The contemporary bothriodontine record is centered on 227-67.
South Dakota and consists of replacement at the generic level by Black, C. C. (1978). Anthracotheriidae. In Evolution of African Mammals,
intercontinental immigration. The oldest bothriodontine record def- eds. V. J. Maglio & H.B. S. Cooke, pp. 423-34. Cambridge: Harvard
University Press.
initely identified to genus is Bothriodon from the early Chadro-
Clark, J., & Beerbower, J. R. (1967). Geology, paleoecology, and paleo-
nian (late Eocene) of Saskatchewan (locality NPlOB). Clark and climatology of the Chadron Formation. In Oligocene Sedimenta-
Beerbower ( 1967) listed Bothriodon from the early Chadronian Ah- tion, Stratigraphy, Paleoecology and Paleoclimatology in the Big
ern member of the Chadron Formation in the Big Badlands (locality Badlands of South Dakota, eds. J. Clark, J. R. Beerbower, &
CP83A), but these specimens are probably not distinguishable from K. K. Kietzke. Fieldiana, Geology Memoirs, 6, 21-74.
Colbert, E. H. (l 935a). Distributional and phylogenetic studies on Indian
Aepinacodon, which first definitely appeared in the mid-Chadronian fossil mammals, IV: The phylogeny of the Indian Suidae and the
Crazy Johnson member of the Chadron Formation (locality CP83B). origin of the Hippopotamidae. American Museum Novitates, 799,
Aepinacodon survived through the late Chadronian in South Dakota 1-24.
and adjacent states before being replaced there by Bothriodon in the (1935b). Siwalik mammals in the American Museum of Natural His-
tory. Transactions of the American Philosophical Society (N.S.), 26,
Orellan, and Bothriodon was, in turn, displaced by two successive
1-401.
species of Elomeryx in the Whitneyean through very early Arika- (1938a). Brachyhyops, a new bunodont artiodactyl from Beaver Divide,
reean. Elomeryx is known from the late Eocene of France, and so Wyoming. Annals of the Carnegie Museum, 27, 87-108.
can be presumed to be an immigrant. (1938b). Fossil mammals from Burma in the American Museum of Nat-
The Arikareean of Montana records the introduction of two new ural History. Bulletin of the American Museum of Natural History,
74, 266-436.
forms: Kukusepasutanka, with its peculiar tubular muzzle, and the
Coombs, W. P., & Coombs, M. C. (1977). The origin of anthracotheres.
broad-lipped Arretotherium. Arretotherium survived into the early Neues Jahrbuchfuer Geologie und Palaeontologie, Abhandlungen,
Hemingfordian (late early Miocene) in Nebraska, South Dakota and 10, 584-99.
Texas, a southward move in its known distribution similar to that Douglass, E. (1901). Fossil Mammalia of the White River Beds of Montana.
displayed earlier by Bothriodon and Elomeryx. Transactions of the American Philosophical Society, 20, 237-79.
Gill, T. (1872). Arrangement of the families of mammals with analytical
As is usually the case, no definite cause can be pinpointed for
tables. Smithsonian Miscellaneous Collections, 11, 1-98.
the extinction of anthracotheres in North America. Macdonald and Janis, C. M., & Fortelius, M. (1988). On the means whereby mammals
Martin (1987) discussed the extinction briefly and suggested that achieve increased functional durability of their dentitions, with spe-
habitat preference and dental morphology may have been involved: cial reference to limiting factors. Biological Reviews, 63, 197-230.
 Donald G. Kron and Earl Manning 388

Leidy, J. (1856). Notice of some remains of extinct Mammalia, recently National Park, South Dakota. Geological Society ofAmerica Special
discovered by Dr. F. V. Hayden in the badlands of Nebraska. Paper, 183, 1-82.
Proceedings of the Philadelphia Academy of Natural Sciences, 8, Romer, A. S. (1933). Vertebrate Paleontology (1st ed.). Chicago: University
1-59. of Chicago Press.
(1869). The extinct mammalian fauna of Dakota and Nebraska. Journal (1945). Vertebrate Paleontology (2nd ed.). Chicago: University of
of the Philadelphia Academy of Natural Sciences, 7, 1-472. Chicago Press.
Macdonald, J. R. (1956). The North American anthracotheres. Journal of (1966). Vertebrate Paleontology (3rd ed.). Chicago: University of
Paleontology, 30, 1315-46. Chicago Press.
Macdonald, J. R., & Martin, J.E. (1987). Arretotheriumfricki (Artiodactyla, Russell, L. S. (1978). Tertiary mammals of Saskatchewan, Part IV:
Anthracotheriidae) from the Hemingfordian (Miocene) Flint Hill lo- The Oligocene anthracotheres. Life Sciences Contributions, Royal
cal fauna in South Dakota. In Papers in Vertebrate Paleontology Ontario Museum, 115, 1-113.
in honor of Morton Green, eds. J. E. Martin & G. E. Ostrander. Scott, W. B. (1940). Part IV: Artiodactyla. In The Mammalian Fauna of the
Dakoterra, 3, 57-62. White River Oligocene, eds. W. B. Scott& G. L. Jepsen. Transactions
Marsh, 0. C. (1890). Notice of new Tertiary mammals. American Journal of the American Philosophical Society (N.S.), 28, 363-746.
of Science (ser. 3), 39, 523-5. Simpson, G. G. (1945). The principles of classification and a classification
(1894a). A new Miocene mammal. American Journal ofScience (ser. 3), of mammals. Bulletin of the American Museum of Natural History,
47, p. 409. 86, 1-350.
(1894b). Eastern division of the Miohippus beds, with notes on some Stehlin, H. G. (1910). Die Saugetiere des schweizerischen Eocaens.
of the characteristic fossils. American Journal of Science (ser. 3), Abhandlung Schweizerischen Paleont. Gesellsch., Basel, 36,
48, 91-4. 839-1116.
(1894c). Miocene artiodactyls from the eastern Miohippus beds: Amer- Tedford, R. H., Skinner, M. F., Fields, R. W., Rensberger, J. M., Whistler,
ican Journal of Science (ser. 3), 48, 175-8. D. F., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb,
Martin, J. E. (1985). Geological and paleontological road log from S. D. (1987). Fauna! succession and biochronology of the Arika-
Rapid City, through the Oligocene White River badlands and reean through Hemphillian interval (late Oligocene through earliest
Miocene deposits, to Pine Ridge, South Dakota. In Fossiliferous Pliocene Epochs) in North America. In Cenozoic Mammals ofNorth
Cenozoic Deposits of Western South Dakota and Northwestern America, ed. M. 0. Woodburne, pp. 153-210. Berkeley: University
Nebraska, eds. J. E. Martin & G. E. Ostrander. Dakoterra, 2, of California Press.
13-59. Troxell, E. L. (1921). The American bothriodonts. American Journal of
Matthew, W. D. (1929). Reclassification of the artiodactyl families. Bulletin Science (ser. 5), 1, 325-39.
of the Geological Society of America, 40, 403-8. von Meyer, H. (1832). Palaeologica, zur Geschichte der Erde und ihrer
Osborn, H. F. (1910). The Age of Mammals in Europe, Asia, and North Geschopfe. Frankfurt am Main.
America. New York: Macmillan. Walker, E. P. (1975). Mammals of the World (3rd ed.). Baltimore: Johns
Pearson, H. S. (1927). On the skulls of early Tertiary Suidae, together with Hopkins University Press.
an account of the otic region in some other primitive Artiodactyla. Wilson, R. W. (1975). The National Geographic Society-South Dakota
Philosophical Transactions of the Royal Society, London (ser. B), School of Mines and Technology expedition into the Big Badlands of
215, 389-462. South Dakota, 1940. National Geographic Society Research Reports,
Retallack, G. J. (1983). Late Eocene and Oligocene paleosols from Badlands 1890-1954 Projects, 79-85.
26 Tayassuidae

DAVID B. WRIGHT

INTRODUCTION

The Tayassuidae, or peccaries, range in age from late Eocene

(Chadronian) to Recent in North America, and from early Pliocene
(Chapadmalalan) to Recent in South America. Several Old World
species have been regarded as tayassuids (e.g., Stehlin, 1899; Pear-
son, 1927; Colbert, 1933; Hendey, 1976), but their alliance with
North American Tayassuidae is not based on synapomorphy, and at
least some of the species in question share apparent apomorphies
with suids.
There are three tayassuid species: Dicotyles tajacu, Catagonus
wagneri, and Tayassu pecari. They average from 15-40 kg in mass;
most fossil tayassuids fall within this range as well. The geographic Figure 26.1. Restoration of Macrogenis crassigenis, by Brian Regal.
range of extant tayassuids extends from southern Texas and Arizona
to northern Argentina; ranges of individual species are also quite DEFINING FEATURES OF THE
broad: Dicotyles tajacu inhabits nearly all of this area, and Tayassu FAMILY TAYASSUIDAE
pecari is only slightly more restricted geographically. Some ex-
tinct tayassuids (e.g., Platygonus compressus) apparently had conti-
CRANIAL
nentwide geographic ranges. All three extant peccaries are sym-
patric in the Gran Chaco of Paraguay, and Dicotyles and Tayassu Occipital region is broad and subplanar; external auditory meatus
are sympatric over much of their ranges; there is no evidence of is conveyed within dorsoventrally elongate tympanic process of the
fossil tayassuids exceeding this level of sympatric diversity. squamosal; basicranial axis is moderately to steeply inclined; sagit-
The extant species are bunodont to zygodont (approaching bilo- tal crest is most often narrow and sharp; auditory bullae may be
phodont) and are primarily herbivorous; they eat mostly nuts, roots, hollow or filled with cancellous tissue; pterygoid bones may be
and cacti, but occasionally consume animals (Sowls, 1984). Fossil broad and strutlike or represented only by the hamular processes;
tayassuids have a similar range of dental morphologies and may anteroventral part of choanal fossa may be isolated from nasal cav-
have had a similar range of diets, but a clear correlation between ity by medially appressed palatine bones and pterygoid processes of
tooth morphology and diet has not been demonstrated in extant alisphenoid; glenoid fossa is transversely elongate and has a strong
tayassuids. Present-day peccaries are social and live in mixed-sex postglenoid process that lies slightly above the occlusal plane; max-
herds ranging in size from an average of four or five individuals in illopalatine suture evident in juveniles of Oligocene tayassuids, ab-
Catagonus to several hundred in Tayassu (Sowls, 1984). The extant sent in all Miocene and younger tayassuids; upper canine alveoli
species are only slightly dimorphic sexually in body size and ca- buttressed by massive lateral processes; nasal incision varies from
nine size (Woodburne, 1968; Mayer and Brandt, 1982), but Tertiary shallow to deep; supraorbital canals deeply excavated; zygomata
tayassuides are strongly dimorphic in canine size, and most late may be narrow and compact, or expanded into pneumatic, later-
Miocene and Pliocene species had large laterally flaring zygomatic ally flaring winglike structures; postorbital processes of frontal and
processes, which are dimorphic in some taxa (Wright, 1991, 1993). jugal robust; vomer may be pneumatic and developed into bilaterally

389
 David B. Wright 390

Figure 26.2. Skulls of tayassuids. A. cf. Perchoerus sp. AMNH 7394 (from Pearson, 1927). B. "Thinohyus" siouxensis, type, CM 1423 (from Peterson,
1905). C. Platygonus compressus (from Leidy, 1853). D. Macrogenis crassigenis, type, AMNH !0882 (from Gidley, 1904).

absent. Cheek teeth usually bunodont, sometimes zygodont or zy-

golophodont. Molars quadrate in outline, having four principal cusps.
Upper premolars may be relatively simple, dominated by a single
principal cusp, or complex and molariform. Lower premolars may
be relatively simple, having a single large trigonid cusp and small
talonid, or may be multicuspid, having talonid cusps subequal in size
to trigonid cusps. DPl, if present, having single, angular, anteropos-
teriorly elongate cusp, two or three roots; often absent. DP2 having
from one to four principal cusps, two to four roots. DP3 having three
or four principal cusps. DP4 having four principal cusps. P2, P3 hav-
ing a single, transversely narrow cusp, or having a single subconical
principal cusp and a robust posterolingual cingulum, or having up
to four principal cusps. P4 having two to four principal cusps.
Figure 26.3. Tayassuid cheek teeth: uppers = "Cynorca" sociale; lowers =
Lower dp 1, if present, having single transversely narrow cusp, or
"Cynorca" occidentale (modified from Woodburne, 1969). Scale bar= I cm. multicuspid, having up to four principal cusps. Lower dp2 having
from one to four principal cusps. Lower dp3 having from one to six
symmetrical chambers; bony paravomeronasal element may be present; principal cusps. Lower dp4 having six principal cusps. Lower p2,
palatine processes of maxillary bone may be pneumatic and parti- p3 having from one to four principal cusps. Lower p4 having two
tioned internally by bony laminae. principal cusps on trigonid; paraconid ranges from robust to entirely
absent; talonid ranges from having a single, subcentral cusp enclosed
by a subarcuate posterior cingulum, to having two principal cusps.
DENTAL
I1 large, anterior surface rounded, lingual surface more nearly flat,
POST CRANIAL
robust lingual cingulum; 12 smaller, angular, having sharp crests; I3
large and caniniform, or absent. Cl and cl large, sharply pointed; Neural arches of cervical vertebrae C3-C6 are perforated by foram-
anterior surface of Cl occludes with posterior surface of cl, pro- ina that transmit spinal nerves; radius and ulna may be fused; meta-
ducing subplanar wear facets. Lower il and i2 subspatulate. Lower carpals 3 and 4 robust, 2 and 5 gracile; metatarsals 3 and 4 robust,
i3 large and angular, having sharp crests, or small and rounded, or may be fused, 2 and 5 gracile or absent; caudal vertebrae reduced.
 Tayassuidae 391

SOFT ANATOMY transmit dorsal branches of cervical spinal nerves in Sus; Ghandi and
Getty, 1969); disklike rhinarium. Gentry and Hooker (1988) listed
Flattened rhinarium; large compound scent gland dorsal to lum-
additional dental and postcranial traits they regarded as synapomor-
bar vertebrae (Tyson, 1683; Werner, Dalquest, and Roberts, 1952);
phies uniting suids and tayassuids.
forestomach developed into fermentation chamber (Tyson, 1683;
Hippopotamidae are often included with the Suidae and Tayas-
Langer, 1978); liver lacking hepatic lobules (Steiner and Ratcliffe,
suidae in the suborder Suiformes (e.g., Simpson, 1945). Pearson
1968); gallbladder absent (Tyson, 1683); hairs having spongy inte-
(1927) included hippotamids, suids, and tayassuids (along with sev-
rior subdivided by radial ribs (Wetzel, 1977; Hess et al., 1985).
eral other taxa) in her "amastoid" group of artiodactyls. Gentry and
Hooker (1988) found some support for Pearson's view from dental
characters they regarded as synapomorphies. Pickford (1983) sug-
SYSTEMATICS gested that the Hippopotamidae evolved from Old World tayassuids.
The tayassuid characters Pickford used to support that arrangement,
however, occur only in late Miocene and younger tayassuids from
SUPRAFAMILY
the New World; these characters are either lacking in, or unknown
Tayassuids have long been recognized as being similar to Old World for, the putative Old World tayassuids. Moreover, this scheme does
suids (e.g., Tyson, 1683). At least a few synapomorphies diag- not account for apomorphies that are shared by suids and tayassuids
nose suids and tayassuids as sister groups: external auditory meatus but not by hippopotamids, nor for tayassuid synapomorphies that
enclosed by tympanic and posttympanic processes of squamosal; are absent in hippopotamids.
ethmoid exposed on orbital wall; neural arch pedicels of cervical Simpson (1945) erected the subfamily Doliochoerinae for Old
vertebrae 3 through 6 perforated by foramina (which are known to World taxa that were referred to the Tayassuidae by Stehlin (1899,

Figure 26.4. Interrelationships

among tayassuids. Synapomorphies
(relative to outgroup taxa Sus scrofa
and Antilocapra americana) diagnos-
ing numbered nodes are as follows:
(!) TAYASSUIDAE: Platelike post-
tympanic process of squamosal
having rounded lateral edge. (2)
PERCHOERUS-TAYASSU CLADE:
Tympanic process of squamosal
moderately deep dorsoventrally. (3)
HESPERHYS-TAYASSU CLADE:
Cancellous, anteriorly acuminate au-
ditory bulla; glenoid fossa condyloid,
lies ventral to level of basioccip-
ital; tympanic process very deep
dorsoventrally; maxillopalatine laby-
rinth present; premaxilla overlaps
incisive foramen dorsally; Ml,2 prin-
cipal cusps not separated by accessory
cusps; P2 having lingual cingulum.
(4) HESPERHYS-"CYNORCA"
SOC/ALE CLADE: Plesiochoanal
chamber present. (5) "THINO-
HYUS'' S/OUXENSIS-HESPERHYS
CLADE: Plesiochoanal chamber
transversely narrow; posthamular part
of pterygoid narrow. (6) TAYASSUINAE: Suborbital bulla present; maxillopalatine labyrinth having dorsal median sulcus; postdental process of palate
extends well posterior to M3; braincase wide; M3 principal cusps not separated by principal cusps; P4 having metaconule; p4 lacking paraconid, having
multicuspid talonid. (7) "PROSTHENNOPS'' XIPHIDONTICUS-TAYASSU CLADE: Atrium of maxillopalatine labyrinth present; fossa extends anteriorly
from infraorbital foramen; p2,3 having broad talonid; dp2 having basined talonid; DP3 having protocone; upper molars anteroposteriorly elongate. (8)
"PROSTHENNOPS" NIOBRARENSIS-TAYASSU CLADE: Foramen ovale confluent with median lacerate foramen; P2,3 having metacone and metaconule;
p3 having metaconid; 13 small or absent; dp3 having metaconid. (9) MACROGENIS-TAYASSU CLADE: Zygoma pneumatic; anterior palatine foramen lies
anterior to Ml; 13 absent; p3 lacking paraconid, having large talonid cusps. (10) BLACK HAWK RANCH SP.-TAYASSU CLADE: Facial crest of zygoma
present, extends anteriorly dorsal to P4. (11) LOVE BONE BED SP.-TAYASSU CLADE: Tectum of maxillopalatine labyrinth meets nasal septum well dorsal
to floor of nasal cavity; atrium of maxillopalatine labyrinth subangular in cross section; P4 having hypoconulid, entoconulid. (12) CATAGONUS-TAYASSU
CLADE: P2 protocone lingual to paracone; p2 having metaconid; dp2 having metaconid. (13) CATAGONUS-DICOTYLES CLADE: Pterygoid processes
of alisphenoid converge medially at choanal margin; posterior palatine foramen opens within nasal cavity, anterior to sphenopalatine foramen; zygodonty
present in some specimens (polymorphic); large suborbital bulla having sharp lateral crest. (14) PLATYGONUS-TAYASSU CLADE: Nasal incision deep,
posteriorly acuminate. (15) PROSTHENNOPS-PLATYGONUS CLADE: Atrium of maxillopalatine labyrinth having posterior aperture; DP2 having
protocone.
 David B. Wright 392

1900), Pearson (1927), Colbert (1933), and others. Simpson did not Day beds of Oregon; Matthew (1915) transferred the species to Per-
offer any morphological characters to diagnose the Doliochoerinae. choerus. Sinclair (1905) and Peterson (1905) suggested that Both-
It is not clear that "doliochoerines" share apomorphies with tayas- rolabis Cope and Chaenohyus Cope are synonyms of Thinohyus
suids, and at least some of them share apparent apomorphies with Marsh, and Matthew (1907) suggested that Thinohyus is a synonym
suids. For example, Doliochoerus quercyi has a rostral muscle fossa of Perchoerus Leidy. Pearson (1923) studied several specimens of
that is subdivided into distinct dorsal and ventral regions, a trait Oligocene peccaries and identified several primitive characters in
also present in Sus and other suids (but absent in tayassuids and ru- their skulls. Matthew (1924) placed Perchoerus, Thinohyus, and the
minants), and Taucanamo sansanense has an elongate, multicuspid other Oligocene taxa in the Paleochoerinae, which he regarded as a
hypoconulid of M3, an unusual trait shared with many suids (but subfamily of the Suidae, whereas he regarded Desmathyus (= Hes-
absent in tayassuids and ruminants) (Wright, 1991). perhys) as a member of the dicotyline subfamily of the Dicotylidae
(= Tayassuidae). The significance of this is obscure, however, as
he regarded Desmathyus as "an obvious connecting link with the
INFRAFAMILY
Oligocene genera." R. A Stirton began a review of these tayassuids
Among the North American species that constitute the Tayassuidae in the 1950s and was joined in that effort by M. 0. Woodburne in
sensu stricto, I recognize several taxa for which published names the 1960s; in 1965, they suggested in an abstract that Perchoerus
are unavailable and use three conventions in this chapter to discuss and Thinohyus should be regarded as distinct genera. My analysis
such taxa: (1) certain named species represent monotypic genera also indicates that Perchoerus and Thinohyus must be regarded as
for which no generic name is available; in these cases I use the cur- separate genera to avoid paraphyly.
rently assigned generic name, but place it within quotation marks Hesperhys, Desmathyus, Pediohyus, and Floridachoerus were
to indicate that the name is inappropriate (e.g., "Prosthennops" nio- named on the basis of specimens from the early and medial Miocene
brarensis is not a member of Prosthennops sensu stricto); (2) in of North America. These taxa differ from Thinohyus and Perchoerus
the case of an unnamed species, I designate the taxon by reference by having dorsoventrally deeper tympanic processes, modifications
to the name of a representative locality (e.g., "Platygonus Edson of the choanal region, and more complex premolars. Matthew ( 1907)
species" and "Machaerodus Quarry species"); and (3) in the case transferred Thinohyus siouxensis Peterson to his new genus Des-
of an unnamed higher taxon comprising multiple genera, I refer to mathyus. Simpson (1945) suggested that Desmathyus Matthew and
the clade by the names of two genera included in it, such that the Pediohyus Loomis are probably junior subjective synonyms of Hes-
clade so identified is the smallest monophyletic group which in- perhys Douglass. In my analysis, Matthew was correct in recogniz-
cludes both genera (e.g., the Hesperhys-Tayassu clade is the group ing the affinity of Thinohyus siouxensis and his Desmathyus, and
united at node 3 in Figure 26.4). Simpson was correct in recognizing that Hesperhys can be consid-
Several synapomorphies unite the Tayassuinae with its sister ered a subjective senior synonym of Desmathyus. These taxa, along
group, the Hesperhys-"Cynorca" sociale clade, as the Hesperhys- with Cynorca sociale (Marsh) Woodburne, Floridachoerus White,
Tayassu clade (Figure 26.4); this clade comprises approximately and two unnamed species, form a monophyletic group (here termed
forty species of Miocene and younger tayassuids. The Oligocene the Hesperhys-"Cynorca" sociale clade) that is the sister group of
tayassuids share with this clade the apomorphic trait of a large, lat- the Tayassuinae, a large clade which comprises Dyseohyus and the
erally rounded post-tympanic process of the squamosal. Among the living peccaries.
Oligocene peccaries, Perchoerus probus shares with the Hesperhys- Several tayassuine species from the late Miocene and Pliocene
Tayassu clade a dorsoventrally deeper tympanic process, an apomor- have submolariform premolars similar to those of the living pecca-
phy absent in Thinohyus lentus. Other cranial characters may rep- ries. Cope (1878) based Dicotyles serus on a lower jaw from the
resent synapomorphies that would unite Thinohyus and Perchoerus late Miocene of Kansas; he recognized its resemblance to extant
with the Hesperhys-Tayassu clade, but their status as such is uncer- Dicotyles (in particular, its similarly complex premolars). Gidley
tain and awaits more thorough study of both the Oligocene speci- (1904) pronounced Cope's species generically distinct and provided
mens in question and primitive suids. for it the new name Prosthennops (Cope had actually relabeled the
Perchoerus, Thinohyus, Chaenohyus, and Bothrolabis were· type specimen of D. serus with this name, but never published it
named on the basis of specimens from the Oligocene and earli- himself). Gidley also named a new species, Prosthennops crassige-
est Miocene of the northern Great Plains and Pacific Northwest. nis, the type being a skull with inflated, laterally flared zygomata
These taxa have relatively shallow tympanic processes, hollow au- from the late Miocene of South Dakota. Matthew (1924) referred
ditory bullae, narrow braincases, simple premolars, and transversely a partial skull from the late Miocene of Nebraska to Prosthennops
upper broad molars, all of which are plesiomorphic relative to char- serus. The zygomata of this specimen are distally rounded, whereas
acters of the living peccaries. Some of these taxa were originally re- those of Gidley's P. crassigenis are distally angular. Matthew re-
ferred to the Old World suid genera Hyotherium and Palaeochoerus. garded this difference as a subgeneric distinction, and proposed
Leidy (1856) assigned a partial dentary from the Mauvaises Terres as new taxa P. (Prosthennops) serus and P. (Macrogenis) crassi-
of South Dakota to Palaeochoerus; he later (1869) transferred the genis (because Gidley failed to designate a type species for Pros-
species in question to a new genus, Perchoerus. Cope (1879) named thennops, Matthew's action as first reviser designated P. serus the
Palaeochoerus subaequans on the basis of a skull from the John type species [Skinner, Skinner, and Gooris, 1977]). Few of the
 Tayassuidae 393

numerous specimens referred to Prosthennops actually represent prototype." He suggested that P. crassigenis Gidley is "obviously
that genus. The genotypic species, Prosthennops serus, has no too specialized" (apparently because of its large zygomatic pro-
known sister species; the genus is monotypic. cesses) to be ancestral to the Quaternary and Recent peccaries, and
Woodbume (1969) recognized Cynorca and Dyseohyus as early concluded that "Cynorca and Dyseohyus are the only known genera
members of a clade that includes "Prosthennops" (sensu Jato) and which are suitable ancestors for the Pliocene, Pleistocene, and Re-
other late Tertiary taxa as well as the Pleistocene and Recent species. cent peccaries." My interpretation is consistent with Woodburne's
My analysis agrees with Woodburne's in that Dyseohyus fricki is an in that Dicotyles tajacu and Tayassu pecari are members of long-
early member of this clade, but differs on other points. Cynorca is separated clades, but differs from his interpretation in that Dicotyles
a nomen dubium because the type specimen of Cynorca proterva is not particularly close to Platygonus (and Brasiliochoerus steno-
Cope, the genotypic species, is undiagnostic (it is an isolated canine cephalus actually represents Catagonus). Another departure is as-
tooth that could represent any of three tayassuid species known from sessment of the polarity of winglike zygomata, which in my analysis
the region of the type locality; Wright and Eshelman, 1987). Most of are plesiomorphic relative to the narrow, but pneumatic, zygomata
the specimens Woodburne referred to that species I regard as repre- of the extant taxa. These processes were reduced in size indepen-
senting "Prosthennops" xiphidonticus. "Cynorca" sociale is actually dently in Tayassu, Platygonus, Mylohyus, Catagonus, and Dicotyles
a member of the clade just mentioned that includes Hesperhys and (Wright, 1993).
its sister taxa.
Two tayassuid genera are common in the North American Pleis-
tocene: Platygonus and Mylohyus. These taxa, along with the INCLUDED NORTH AMERICAN GENERA IN THE
extant genera Tayassu, Catagonus, and Dicotyles, can be recogn- FAMILY TAYASSUIDAE
ized in the North American Tertiary (or can be inferred on the
basis of cladistic relationships to have existed then). Pleistocene The locality numbers listed for each genus refer to the list of unified
Platygonus species have zygolophodont cheek teeth, relatively sim- localities in Appendix I. The acronyms for museum collections are
ple premolars, and (in most cases) rounded zygomatic processes. listed in Appendix III.
Pleistocene Mylohyus has molariform premolars and a dolicho- The locality numbers may be listed in a couple of alternative ways.
cephalic skull. Matthew (1924) proposed Desmathyus (=Hesper- Parentheses around the locality (e.g., [CPlOl]) mean the taxon in
hys) vagrans as an ancestor of Platygonus because both taxa possess question at that locality is cited as an "aff." or "cf." the taxon in
zygodont cheek teeth and bicuspid upper premolars. The specimens question. Parentheses are usually used for individual species, thus
Matthew identified as Platygonus did not have laterally expanded implying the genus is firmly known from the locality, but the actual
zygomata. Matthew saw the submolariform premolars and wing- species identification may be questionable. Question marks in front
like zygomata of the skull he referred to Prosthennops serus as of the locality (e.g., ?CPlOI) mean the taxon is questionably known
specializations precluding that species from ancestry of Platygonus from that locality, thus implying some doubt that the tax on is actually
(the specimen in question, however, actually represents an early present at that locality, either at the genus or the species level.
species of Platygonus). The submolariform premolars of Prosthen-
nops suggested to Matthew a link with Mylohyus, and he predicted OLIGOCENE TAYASSUIDS
that the zygomata of Mylohyus, then unknown, would have large
Thinohyus Marsh, 1875
processes (corroborated, in part, by Mylohyus elmorei; Wright and
Webb, 1984). Matthew suggested Desmathyus as a suitable ancestor Type species: Thinohyus lentus Marsh, 1875.
for Dicotyles as well as Prosthennops and Platygonus. The dental Type specimen: YPM 11783.
similarity that Matthew (1924) saw between Hesperhys (= Des- Characteristics (based on the type specimen of T. lentus):
mathyus) and Platygonus is convergent in my analysis: The bicus- Postglenoid canal passes between post-tympanic process
pid upper premolars of Platygonus resulted from simplification of of squamosal and postglenoid process of jugal, ventral
multicuspid teeth, whereas the bicuspid upper premolars of Hes- and parallel to external auditory meatus; tympanic pro-
perhys resulted from elaboration of unicuspid teeth. Dalquest and cess of squamosal dorsoventrally shallow, external audi-
Mooser (1980) named Desmathyus brachydontus on the basis of a tory meatus angles laterally; articular surface of glenoid
large sample of teeth from late Hemphillian Rancho EI Ocote, Mex- fossa approximately in plane ofbasioccipital; choanal mar-
ico (locality SB58). D. brachydontus is represented by skulls from gin opens medial to M3 (may be influenced by young age of
other late Hemphillian localities; those specimens have apomorphic specimen); upper molar paracone and protocone separated
characters that link the species with Catagonus (Wright, 1989). by paraconule; Thinohyus also lacks the synapomorphies
Woodburne (1968) studied cranial osteology and myology of of the Perchoerus-Tayassu clade.
the extant peccaries, and on that basis he allied Dicotyles with Length of m2: 16.0 mm (from Marsh, 1875).
Platygonus and Tayassu with Mylohyus. He also suggested that the Included species: T. lentus only, known from "Miocene, John
South American Pleistocene species Brasiliochoerus stenocephalus Day River, Oregon" (Marsh, 1875, within locality PN6).
is closely related to Tayassu and Mylohyus, and allied Prosthennops Thinohyus sp. is also known from localities (CP84A),
niobrarensis with this clade as a broadly construed "ancestral (CP85C), (CP99B), (NP50B).
 David B. Wright 394

Perchoerus Leidy, 1869 HESPERHYS-TAYASSU CLADE

Type species: Perchoerus probus Leidy, 1869. Included taxa: Hesperhys-"Cynorca" sociale clade and Tayassuinae.
Type specimen: ANSP 10588, dentary with dp4, p4, ml, m2. Characteristics: Pneumatic labyrinth developed in maxillary and
Characteristics (based on SDSM 2838, referred specimen): palatine bones of floor of nasal cavity; paravomeronasal element
Tympanic process of squamosal dorsoventrally deep, ex- present; occipital crest continuous with tympanic crest; kidney-
ternal auditory meatus directed dorsally; articular surface shaped, condyloid articular surface of glenoid fossa; robust pre-
of glenoid fossa lies slightly ventral to basioccipital, but glenoid process of jugal; upper molar paraconule anteriorly dis-
well dorsal to occlusal plane; choanal margin opens ap- placed; P2 has robust lingual cingulum or separate cusps in addition
proximately 15 mm posterior to M3; palatine wall of to paracone.
choanal fossa medial to maxillary foramen inflated lat- Range: Late Arikareean to Recent in North America; Pliocene to
erally; robust occipital crest; upper molar paracone and Recent in South America.
protocone separated by paraconule; lacks synapomorphies Comments: This clade is the smallest monophyletic group that in-
of Hesperhys-Tayassu clade. cludes both Hesperhys and Tayassu; that is, it comprises essentially
Average length of m2: unknown. Length of M2: 14.0 mm all known tayassuid taxa except for Perchoerus, Thinohyus, and the
(from Pearson, 1923). other Oligocene taxa. In the Tayassuinae, the lateral parts of the
Included species: P. probus only, known from locality CP84 vomer are inflated into elongate chambers, the paraseptal cartilage
only. (which houses the vomeronasal organ) is ossified, and the paired,
Perchoerus sp. is also known from localities (CP84B), apparently neomorphic, paravomeronasal elements lie lateral to the
(NP50B). paraseptal cartilage. These structures are present in a specimen of
Hesperhys cf. H. pinensis, but are unknown for other members of the
Comments: For the purpose of comparison in this report, I
Hesperhys-"Cynorca" sociale clade and for Oligocene tayassuids.
tentatively accept Scott's (1940) reference of SDSM 2838
At least some Oligocene specimens, which represent taxa excluded
to this species. Should future work show that Scott was in
from the Hesperhys-Tayassu clade, have a structure on the floor
error, the status of Perchoerus probus might be different
of the nasal cavity that may be homologous to the maxillopalatine
than indicated here, but the comparison of SDSM 2838
labyrinth.
with Thinohyus would still be valid.

Other Oligocene taxa HESPERHYS-"CYNORCA" SOC/ALE CLADE

Several other taxa from the Oligocene have been named. I have Characteristics: Members of the Hesperhys-Tayassu clade having
not studied them comprehensively and do not discuss them here medially fused pterygoid bones, pterygoid processes of alisphenoid,
beyond a brief comparison of the type specimens with Thinohyus and vertical processes of palatine bones.
lentus and Perchoerus probus. The type specimens of Hyotherium Included taxa: Hesperhys (and Desmathyus, its junior subjective
americanum Scott and Osborn, 1887, Palaeochoerus subaequans synonym fide Simpson, 1945), "Thinohyus" siouxensis, Florida-
Cope, 1879, and Chaenohyus decedens Cope, 1879 are brachy- choerus, "Cynorca" sociale, and an unnamed genus from the early
cephalic and otherwise similar in morphology to the type of Thi- Miocene of Wyoming.
nohyus lentus; these specimens lack the synapomorphies of the Range: Late Arikareean to Barstovian of North America.
Perchoerus-Tayassu clade. The type specimens of Thinohyus os- Comments: The plesiomorphic condition of the choanal margin
monti Sinclair, 1905, and Bothrolabis rostratus Cope, 1888, are in mammals is to be bordered anteriorly by the palatines and lat-
both dolichocephalic, as is Perchoerus probus (referred); these spec- erally by the vertical processes of the palatines, the pterygoids, the
imens have the synapomorphies that Perchoerus shares with the orbitosphenoids, and the pterygoid processes of the alisphenoids.
Hesperhys-Tayassu clade, and also have a similar pneumatic expan- In members of the Hesperhys-Tayassu clade it is as though the dor-
sion of the palatine region medial to the maxillary foramen, which sal parts of elements forming the lateral walls of the choanal fossa
may be an autapomorphy of Perchoerus. Thinohyus nanus Marsh, have been squeezed together medially, isolating the ventral portion
1892, and Perchoerus minor Cook, 1922, are both based on den- of the choanal passage as a narrow, elongate chamber roofed by
taries with cheek teeth. These dentaries are distinctly smaller than an arch formed of palatine and pterygoid bones. The roof descends
those of most other Oligocene taxa, but the characters they possess anteriorly to meet the floor of the palate, isolating the plesiomor-
may be inadequate for analysis of phylogenetic relationships. Per- phic choanal fossa ("plesiochoanal fossa") from the nasal cavity.
choerus minor may be a junior subjective synonym of Thinohyus This rearrangement results in the dorsal and posterior displacement
nanus. Both the White River (locality CP68) and John Day (locality of the floor of the choanal fossa, the functional effect of which is
PN6) regions have produced specimens sharing apparent apomor- open to speculation. A superficially similar morphology occurs in
phies with Perchoerus probus (referred), and specimens similar to Catagonus, an extant tayassuine, but in that species only the most
Thinohyus lentus are also known from both regions. ventral and distal parts of the pterygoid processes of the alisphenoid
Oligocene tayassuids of indeterminate taxonomic status have are appressed medially, and the appression occurs right at the pos-
been reported from localities GC7, CP68B, CP68C, CP68D, CP83A, terior margin of the palate. The lateral walls of the choanal fossa
CP83C,CP98B,CP98C,PN6B,PN6C,PN6D. and the pterygoids thus retain a plesiomorphic morphology. In an
 Tayassuidae 395

unnamed member of the Hesperhys-"Cynorca" sociale clade from Characteristics: Very small member of the Hesperhys-"Cyno-
Van Tassel, Wyoming, the plesiochoanal fossa remains large and has rca" sociale clade having narrow plesiochoanal fossa, re-
a U-shaped anterior margin; in other taxa, the fossa is very narrow taining narrow lingual cingulum of P3.
and has a V-shaped anterior margin. Included species: "C." sociale only (known from localities
NB17, PN6G).
Hesperhys Douglass, 1903 (synonym: Desmathyus)
Other possible specimens of "C." sociale, or of tayas-
Type species: Hesperhys vagrans Douglass, 1903. suids of the Hesperhys-"Cynorca" sociale grade, are known
Type specimen: CM 748. from localities GC5, GC9C, CP103, CP104B, ?PN7, NC2.
Characteristics: Members of Hesperhys-"Cynorca" sociale
clade having very large orbitomaxillary fossa; plesiochoa-
TAYASSUINAE
nal fossa very narrow or reduced to subangular depression;
posthamular part of pterygoid very narrow; pterygoid pro- Included taxa: Dyseohyus, "Cynorca" occidentale, "Pros-
cesses of alisphenoid narrow, join postdental process of thennops" xiphidonticus, "Prosthennops" niobrarensis, Macroge-
palate at low angle. nis, Black Hawk Ranch species, Machaerodus Quarry species, Love
Average length of m2: 21.6 mm (H. vagrans); 20.9 mm Bone Bed species, Catagonus, Dicotyles, Prosthennops, Tayassu,
(H. pinensis). Mylohyus, Platygonus.
Included species: H. vagrans (= Desmathyus validus Characteristics: suborbital bulla present; maxillopalatine lab-
Matthew, 1932) (known from localities CC17D, NB6C, yrinth has dorsal median sulcus; palate extends posteriorly nearly
[NB 17], CPl l 0, NP42); H. pinensis ( = Desmathyus pinen- to level of glenoid fossa; infraorbital fossa has anterior accessory
sis Matthew, 1907) (localities NB17, CP86D, CP88). foramen; DPl absent; glenoid fossa lies below ventral surface of
Hesperhys sp. is also known from localities ?GC4E, auditory bulla; P4 having metaconule; p4 lacks paraconid, has mul-
?CP71, CP105, CP107, (CP108B), ?PN8B, (NCI). ticuspid talonid.
Comments: This clade includes at least thirty-five species. In or-
"Thinohyus" siouxensis (= Thinohyus [Bothrolabis] der to simplify description of generic characteristics here, I diagnose
siouxensis Peterson, 1905) five monophyletic subgroups, as follows:
Type specimen: CM 1423. "Prosthennops" niobrarensis-Tayassu clade. Tayassuines having
Characteristics: MemberoftheHesperhys-"Cynorca" sociale elongate fossa extending anteriorly from infraorbital foramen; DP2
clade having narrow plesiochoanal fossa, narrow postham- and DP3 with crescentic protocone; P2 and P3 with protocone,
ular part of pterygoid; retaining broad pterygoid processes metacone, and metaconule; P4 having four principal cusps, plus
of alisp~enoid that join the palate at a high angle, shal- hypoconule, metaconule. Includes "Prosthennops" niobrarensis and
low orbitomaxillary fossa, large maxillary tubercle, DPl, the Macrogenis-Tayassu clade.
narrow lingual cingulum on P3. Macrogenis-Tayassu clade. Members of "Prosthennops" niobra-
Average length ofm2: 19.1 mm. rensis-Tayassu clade having pneumatic zygoma; anterior palatine
Included species: "T." siouxensis only, known from locality foramen anterior to Ml; lacking 13; p3 lacking paraconid, having
CP104B only. large talonid; radius and ulna fused. Includes Macrogenis, unnamed
taxa from Black Hawk Ranch, Machaerodus Quarry, and Love Bone
Floridachoerus White, 1941 Bed, and the Catagonus-Tayassu clade.
Catagonus-Tayassu clade. Members ofMacrogenis-Tayassu clade
Type species: Floridachoerus olseni White, 1941.
having facial crest extending over rostral muscle fossa, tectum of
Type specimen: MCZ 4290.
maxillopalatine labyrinth meeting nasal septum well dorsal to floor
Average length of m2: 16.0 mm.
of nasal cavity, atrium of maxillopalatine labyrinth angular in cross
Characteristics: Member of the H esperhys-"Cynorca" sociale
section; p2 and dp2 with metaconid; p4 with hypoconid, meta-
clade, having narrow plesiochoanal fossa, P3 with large lin-
conid; metatarsals III and IV fused medially. Includes Love species,
gual crescent in protocone position, retaining broad ptery-
Catagonus-Dicotyles clade, and Platygonus-Tayassu clade.
goid processes of the alisphenoid that join the palate at a
Catagonus-Dicotyles clade. Members of Catagonus-Tayassu
high angle.
clade having pterygoid processes of alisphenoid converge medially
Included species: Floridachoerus olseni only, known from
at palate, forming V-shaped choanal margin; rounded dorsolateral
locality GC8D only.
edge of premaxilla; anteroventral aperture of vomeroethmoid cham-
Floridachoerus sp. is also known from locality GC5.
ber lies above P2; posterior palatine foramen opens inside nasal
"Cynorca" sociale (Marsh, 1875) Woodburne, 1969 (=
cavity, anterior to sphenopalatine foramen. Includes Catagonus and
Dicotyles.
Thinohyus [Bothrolabis] socialis Merriam and Sinclair,
1905; Palaeochoerus socialis Cope, 1879; Perchoerus Platygonus-Tayassu clade. Members of Catagonus-Tayassu clade
socialis Matthew, 1909) having posteriorly acuminate nasal incision. Includes Platygonus,
Prosthennops (sensu stricto), Mylohyus, and Tayassu.
Type specimen: YPM 11785. Platygonus-Prosthennops clade. Members of Platygonus-
Average length of m2: 11.3 mm. Tayassu clade having posterior aperture of atrium of maxillopalatine
 David B. Wright 396

labyrinth. Includes Platygonus, Mylohyus, and Prosthennops (sensu Characteristics: Tayassuine having synapomorphies of Mac-
stricto ). rogenis-Tayassu clade; distally angular (M. crassigenis) or
rounded (Macrogenis n. sp.) winglike zygomatic process
Dyseohyus Stock, 1937 distally angular, glenoid fossa extends laterally beyond lat-
eral border of tympanic process; transversely broad den-
Type species: Dyseohyusfricki Stock, 1937. tary and cheek teeth.
Type specimen: LACM (CIT) 2039. Average length of m2: 17.8 mm (M. crassigenis); 20.3 mm
Characteristics: Member of Tayassuinae retaining transver-
(M. n. sp.).
sely broad upper molars; larger than "Cynorca" occiden- Included species: Macrogenis crassigenis (known from lo-
tale. calities SB34A, CP90A, CPI ISA, CPI 16B); Macrogenis
Average length of m2: 13.1 mm. n. sp., an unnamed species (localities CP90A, CP116B).
Included species: D.fricki only (known from localities NB6C,
NB6E). Black Hawk Ranch species

"Cynorca" occidentale Woodburne, 1969 Characteristics: Member of Macrogenis-Tayassu clade hav-

ing distally angular winglike zygomatic process; facial
Type specimen: AMNH 73660. crest of zygoma that extends anteriorly over P4; lacking
Characteristics: Member of Tayassuinae having M3 very synapomorphies of Catagonus-Tayassu clade.
small relative to M2; retaining transversely broad upper Average length of m2: 17 .2 mm.
molars. Included species: Type only, known from locality CC26B
Average length of m2: 12.8 mm. only.
Included species: "C." occidentale only (known from locali-
ties CCI 7D, NB6C, CP108B). Machaerodus Quarry species

"Prosthennops" xiphidonticus Barbour, 1925 (= Dyseohyus

Characteristics: Member of Macrogenis-Tayassu clade hav-
ing distally angular winglike zygomatic processes; facial
stirtoni Woodburne, 1969; Cynorca proterva Gazin and
crest of zygoma that extends over P4; lacking synapomor-
Collins, 1950 [in part])
phies of Catagonus-Tayassu clade.
Type specimen: UNSM 2655. Average length of m2: I 6.4 mm.
Characteristics: Tayassuine peccary having maxillopalatine Included species: Type only, known from locality CP1 I6B
atrium; DP3 having crescentic protocone; upper molars only (from Machaerodus, Egger, Emry, Wade, and Xmas
relatively narrow transversely; retaining plesiomorphic Quarries).
traits of narrow, cancellous zygoma; p4 having postero-
labial process of metaconid; P2 and P3 lacking protocone, Love Bone Bed species
metacone, metaconule; P4 lacking hypocone; p3 having
Characteristics: Member of Catagonus-Tayassu clade having
single trigonid cusp.
distally angular winglike zygomatic processes; cheek teeth
Average length ofm2: 12.6-14.6 mm.
Included species: "P." xiphidonticus only (known from local- bunodont.
ities GC4C, CP75B, CPI 10, CPI 11, CP114A, CPI I4B, Average length ofm2: I6.0 mm.
Included species: Type only, known from locality GCI IA
CPI 14C, NC3B).
only.
"Prosthennops" niobrarensis Colbert, 1935
Catagonus Ameghino, 1904 (synonym: Platygonus, in part)
Type specimen: UNSM 2604.
Type species: Catagonus metropolitanus Ameghino, 1904.
Characteristics: Tayassuine having synapomorphies of "Pro-
'fype specimen: MACN 850.
sthennops" niobrarensis-Tayassu clade; retaining narrow,
Characteristics: Members of Catagonus-Dicotyles clade hav-
cancellous zygoma.
ing very large suborbital bulla (diameter> 30 mm), with
Average length of m2: I 7 .3 mm.
sharp lateral crest; pterygoid processes of alisphenoid meet
Included species: "P." niobrarensis only (known from local-
palate at very low angle (ca. 20°); cheek teeth zygodont or
ities CPI I4C, CP114D, CP116A, NC4).
bunodont.
Average length of m2: 20.6 mm (C. wagneri); 19.6 mm
Macrogenis Matthew 1924 (synonym: Prosthennops
(C. stenocephalus); 20.9 mm (C. brachydontus); 17.8 mm
[Macrogenis])
(? C. sp.).
Type species: Macrogenis crassigenis (Gidley, I904) (=Pro- Included species: C. metropolitanus (Ensenadan of Argen-
sthennops). tina); C. wagneri (Pleistocene-Recent, Argentina and
Type specimen: AMNH 10822. Paraguay) (Wetzel, 1977); C. stenocephalus (Pleistocene
 Tayassuidae 397

ofBrazil); C. brachydontus (known from localities GCI 3B, WM26, GC14, GC17, NB12B, (NB13C), ?SB13, SBI4A,
SB58A, SB62, SP4B). SBI4B, SB14C, SB14D, SBI4D, SB14E, SB14F, SB15,
Catagonus sp. is also questionably known from locality SB I SA, SB I SB, SB3 l, SBSO, SB58B, SB58C, SB60, SB66,
CPII6D. SB67, (SPlD), SPIF, ?SP4A, SPS, ?CP78, CP80, CPI 16F,
CP117A, CPII7B, CPll8, CP128B, CP128C, CP130A,
Tayassu Fischer, 1814 CP13I, CPI32B, (NP45), PN23C.

Type species: T. pecari Fischer, 1814 (extant). Mylohyus Cope, 1869 (synonym: Prosthennops, in part)
Characteristics: Member of Platygonus-Tayassu clade having
large paravomeronasal chamber; lacking atrium of max- Type species: M. fossilis (Leidy, 1860) (= Dicotyles).
illopalatine labyrinth; tectum of maxillopalatine labyrinth Type specimen: Unavailable.
originates over P4 or Ml; bunodont cheek teeth; zygomata Characteristics: Member of Prosthennops-Platygonus clade
narrow in T. pecari, unknown for ?Tayassu edensis. having very long postcanine diastema; molariform P2;
Average length of m2: 16.1 mm (T. pecari); 17 .2 mm (T. atrium of maxillopalatine labyrinth very narrow in
edensis). diameter.
Included species: T. pecari (Pleistocene of Argentina and Average length of m2: IS.0-16.7 (M.fossilis); I 7.1 mm (M.
Brazil; Recent, Mexico to Argentina). ?Tayassu edensis longirostris); I9.0 mm (M. elmori).
may represent this genus at locality CC40. Included species: M.fossilis (late Blancan through Ranchola-
brean of North America); M. longirostris (known from
Prosthennops Gidley, 1904 localities [GCllB], PN12); M. elmorei (= Prosthennops
elmorei White, 1942) (locality GCI3B).
Type species: Prosthennops serus (Cope, 187 8) (= Dicotyles). Mylohyus sp. is also known from localities GC 14, GCI 5,
Type specimen: AMNH 8511. GC17.
Characteristics: Member of Platygonus-Tayassu clade hav-
ing moderately large (ca. 10 mm transverse by 30 mm
INDETERMINA TE NEOGENE TAYASSUIDS
dorsoventral) posterior aperture of atrium of maxillopala-
tine labyrinth; distally angular winglike process of zy- Fragmentary material ascribed to tayassuids are known from locali-
goma; facial crest of zygoma extends anteriorly over pre- ties CA9, GC2A, GC2B, ?GC4D, GC6B, GCSA, GC9C,
molars; cheek teeth bunodont; P2, p2 not fully molariform. GC1211, GC28, CC14, CC21A, CC21C, CC23, CC25A, CC25B,
Average length ofm2: I8.9 mm. CC32A, CC32B, CC35A, CC38 (= "Prosthennops kernensis"),
Included species: P. serus only (known from localities CA4, CC39B, CC4I, NB7D, NBIO, NBll, NB23B, NB23C, NB27B,
SB58A, SB58B, [SPIC], SP3A, [CP78], CP116C, NB3I, SBl 7, SB35, SB55, SB65, SPIA, SPlB, SP2A, CP51A,
CPII6D, CPI23A, [NP45], PNll). CP93, CP94, CP104C, CP106, CPI ISB, CPI 15C, CPI 16E,
CP123D, CP126, NPll, NP34D, PN9A, PN9B, PNlO, PNI3.
Platygonus LeConte, 1848

Type species: Platygonus compressus Leconte, 1848. BIOLOGY AND EVOLUTIONA RY PATTERNS
Type specimen: Unavailable.
Characteristics: Members of Prosthennops-Platygonus clade Extant tayassuids inhabit a spectrum of habitats ranging from tem-
having moderately large to very large posterior aperture of perate desert (e.g., Arizona) to tropical rain forest (e.g., the Ama-
atrium of maxillopalatine labyrinth; moderate or complete zon). Indeed, a single species, Dicotyles tajacu, occupies this en-
reduction of posterior cusps of premolars; bunodont to zy- tire range of biomes. It therefore seems unwise to postulate precise
golophodont cheek teeth. habitat preferences for extinct tayassuids on the basis of simplistic
Average length of m2: 16.8 mm (P. compressus, Edson spec- taxonomic extrapolation.
ies); I9.2 mm (P. pearcei); 16.8 mm (?P. oregonensis); The cranial sutures of extant peccaries are unusual. All cranial
16.7 mm (?P. brachyrostris). sutures are completely fused (closed) in adults, and there is no sign
Included species: P. compressus (Rancholabrean of North of a maxillopalatine suture even in neonates (Herring, I974). This
America); Platygonus Edson sp. (known from localities is the case in all known members of the Hesperhys-Tayassu clade,
SP3B, SP4B, CP115D, CPI23D [Edson Quarry]); P. pea- and it is a synapomorphy of that clade because many Oligocene
rcei (localities PN3C, PN4, PN23A); P. cumberlandensis specimens (e.g., Perchoerus probus) have distinct sutures in adults,
(early lrvingtonian of North America). ?Platygonus ore- and the maxillopalatine suture is evident in juvenile specimens of
gonensis (locality PN12), and ?Platygonus brachyrostris these taxa. The adaptive significance, if any, of fusing cranial su-
(locality PNI4) may also represent_ this genus. tures is open to speculation. In the Tayassuidae, fusion of cranial
Platygonus sp. is known from the Pliocene and Pleis- sutures is associated with a change in the masticatory apparatus:
tocene of South America (at least two species), and also Members of the Hesperhys-Tayassu clade also have shorter tempo-
from the following North American· Tertiary localities: ralis outforce moment arms (relatively shorter coronoid processes)
 David B. Wright 398

w
Z
w
L L.81.

E.81
1.8
2.4
-~ Il
~ ..!!! ___ - cL
ci. Ill
a-
0
T""
1
?
111 -
~
111
~ Il
-

si=- t
(.)

- =~-~=·Ille:--=-~ =~-~---~E-=~&~~
4.5
0 E Lst. Hp.
5.2
:::i
a.
l.Hp.
6.0
-~ c ....
::::s -;: r.11

e
QI .._

-c-- . . -eu
LE.Hp.
7.0 ·ii - .c: -&- .Q I'll -0- -ci.- Ill-~- !11 -..\l! ::...
-~ -~
E.E.Hp.
a:l-gil -"'[ o _a.1?:t
'u Il a
L L.CI.
8.8
9.5 -~ - () i --c -~ Q.
_i;.;;_
_c-·_ - - - -
~
e _g I'll
-U-~
E.CI.

w 11.0 o- ::C-Cl>--
0 c:c: ~I'll
zw LL.Ba.
12.5 ·"as - ~ ~ :;! -~-
.~ ~ ~ m-:i:-
E.L.Ba.
(.) M
0
~ E.Ba.

L.Hm.
14,0

5.8 -
§.t e e
-~ ~ ~

E.Hm.
17.5 - -
18.8
E LL.Ar.
19.2

E.L.Ar.
23.0

w
z LE.Ar.

W L

Il
(.) 27.7
0
CJ
:::i
E.E.Ar.
g:
0 29.4 ~o t
Wt.
E 31.9 .s ii._o
Or.
~ if ___________________________ ________--:--1
33.4L!::.L-.!:!:...L
L.Ch.
34.5 .,..
L M.Ch.

E.Ch.
3
5.5 - · t- -HESPERHYS-"CYNORCA" TAYASSUINAE
37.1 - '- - SOC/ALE CLADE
Du

39.5

L.Un.
41.3
w
zw E.Un.

(.) 45.9
0 L.Br.
W M 46.7

M.Br.
49.0
E. Br.
50.4

L.Wa.

53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0
55.5
-
Ck.3 55.6
w Ck.2 55.8
z Ck.1
Ti.6
56.0
56.2
W L 11.5 57.0
(.) Tl.4 57.3
Tl,3
0 Tl.2
w Ti.1
...I
To.3
<C
a. To.2
62.5
To.1
E Pu.3
63.8
64,0
Pu.2 64.5
Pu.1 65.0
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean, .
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Bridgerian, Wa. = Wasatch1an,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.
Figure 26.5. Temporal ranges of North American tayassuid genera.
 Tayassuidae 399

and stronger estimated masseter bite forces than do Perchoerus and evident in "Prosthennops" xiphidonticus, "P." niobrarensis, Macro-
other Oligocene taxa (data of Kiltie, 1981, interpreted in the phylo- genis, Love Bone Bed species, Machaerodus Quarry species, other
genetic context of this chapter). Clarendonian tayassuines, Platygonus, and Mylohyus. This dimor-
Tayassuid cheek teeth vary from being bunodont, having low phism is most simply interpreted as sexual in nature, with male
rounded cusps with thick enamel (e.g., Mylohyus), to being zy- canines being larger than those of females; such a relationship is
godont, having more sharply pointed principal cusps that may be common among many groups of mammals, including suids, hip-
connected by accessory crests (e.g., Catagonus), to nearly lophodont pos, and many perissodactyls.
(e.g., Platygonus). Bunodonty is plesiomorphic for tayassuids. Zy- Comparing the distribution of strongly dimorphic canines among
godonty occurs as an apomorphy in at least three clades: Hesper- tayassuids with these progressively distant outgroups indicates that
hys, Platygonus, and Dicotyles-Catagonus. Among living tayas- strong sexual dimorphism in canine diameter is plesiomorphic for
suids, Tayassu pecari has bunodont cheek teeth, and it frequently the Tayassuidae (and probably for the Ungulata). The wide overlap
inhabits forests, where it consumes a variety of nuts (Kiltie, 1981; of male and female canine sizes in living peccaries is apomorphic.
Sowls, 1984); Catagonus wagneri has zygodont cheek teeth and is The extant peccaries belong to two distinct clades, each of which
restricted to the arid thorn forest of the Gran Chaco of Paraguay includes taxa having discrete canine dimorphism. In fact, Holocene
and Argentina, where it subsists on a diet heavy in cacti (Wetzel, (ca. 0.01 Ma) specimens from Lagoa Santa, Brazil, of Dicotyles
1977; Sow ls, 1984). One might be tempted to ascribe corresponding cf. D. tajacu and Tayassu cf. T. pecari have canines that are more
dietary preferences to Mylohyus (which is bunodont), Platygonus strongly dimorphic than those of extant populations of those species
(which is zygodont to zygolophodont), and other extinct tayassuids (Wright, in preparation). Late Pleistocene Platygonus andMylohyus
on the basis of these correlations, but the picture is clouded by vari- have canines that are more weakly dimorphic than those of Tertiary
ation in extant Dicotyles tajacu. Cheek teeth in D. tajacu range members of their clades.
from bunodont to zygodont, and its diet ranges from roots to nuts to Many late Tertiary tayassuines have zygomata that are later-
cacti (Woodburne, 1968; Kiltie, 1981; Sowls, 1984). The morpho- ally expanded into winglike processes. Suids such as the warthog
logical variation may actually be correlated with variation in diet, (Phacochoerus) and forest hog (Hylochoerus) have zygomatic
however. Desert populations of Dicotyles tajacu (which consume processes that are at least superficially similar to those of tayassuids.
large amounts of cactus) tend to be zygodont, whereas mesic popu- Darwin ( 1871) suggested that these processes function to protect the
lations (which have a diet more like that of Tayassu) tend to be bun- eyes during combat, and a similar role is plausible for the winglike
odont (but this observation has not been evaluated quantitatively). zygomata of tayassuids; it is more difficult to interpret such ap-
Tayassuids plesiomorphically have relatively simple premolars, pendages as modifications of the masticatory apparatus, for they
as in Thinohyus and Perchoerus. In both the Hesperhys-"Cynorca" extend well beyond the lateral margin of the clearly delimited site
sociale clade and the Tayassuinae, the premolars increase in com- of origin for the masseter musculature. In some tayassuid species,
plexity, but the' patterns of complication differ in the two clades these processes are sexually dimorphic in size; in others they are not.
(Wright, 1991). In the Tayassuinae, the premolars become more Sexual dimorphism can result from sexual selection (male-male
complicated by the progressive "addition" of cusps; that is, they be- competition or female choice, Darwin, 1871 ), or from natural se-
come "molarized" much as premolars in some equids become mo- lection if the sexes have different ecological roles (e.g., feeding;
larized (e.g., Butler, 1952). The deciduous premolars show a parallel cf. Slatkin, 1984). Cranial appendages such as horns and tusks are
increase in complexity. In the Hesperhys-"Cynorca" sociale clade, used in intraspecific fights between males and are reasonably inter-
deciduous premolars become progressively more molariform, as preted as resulting from sexual selection (Darwin, 1871). Lande's
they do in Tayassuinae, but the permanent upper premolars increase (1980) polygenic model for the evolution of dimorphic secondary
in complexit)l.by simply adding a single lingual cusp; they do not be- sexual characters via sexual selection predicts that characters se-
come molariform, and they do not resemble any intermediate stage lected for in males will initially be expressed in both sexes: Se-
in molarization of permanent (or deciduous) premolars observed in lection on a character in males produces a correlated response in
the Tayassuinae. This pattern of upper permanent premolar compli- females. Winglike zygomata are large in males and females (sex
cation - development of a "bicuspid" tooth - is similar to that of diagnosed by canine size) of M. crassigenis, and also in both sexes
ruminants, camelids, and most other selenodont artiodactyls, and it of the Machaerodus Quarry species. Sexually dimorphic winglike
is probably the plesiomorphic pattern of premolar complication for zygomata occur in some species of Platygonus. This distribution in-
artiodactyls. It is not clear that a model of dental morphogenesis dicates that monomorphism of winglike zygomatic processes is ple-
based purely on morphogenetic gradients (e.g., Butler, 1952) is ad- siomorphic relative to their dimorphism, in agreement with Lande's
equate to account for these distinct patterns of permanent premolar model. In addition, the presence of winglike zygomatic processes is
complication. plesiomorphic for members of the Macrogenis-Tayassu clade. The
The three extant tayassuid species show only weak sexual dimor- lack of such processes in late Pleistocene and Recent members of
phism in such characters as body size, zygoma width, and canine di- this clade is therefore apomorphic; that is, the processes were lost.
ameter (Woodburne, 1968; Wetzel, 1977; Mayer and Brandt, 1982). Loss of the process is in each case associated with a reduction in the
In contrast, many Tertiary tayassuids were strongly dimorphic in degree of sexual dimorphism in canine diameter, which is also apo-
canine size (discretely bimodal in diameter) and, in some cases, in morphic. In this context it is worth noting that sexual competition
zygoma size (Wright, 1991, 1993 ). Discrete canine dimorphism is between males in wild herds of extant Dicotyles tajacu (the only
 David B. Wright 400

species that has been studied in detail) is apparently rare (Byers, Gentry, A. W., & Hooker, J. J. (1988). The phylogeny of the Artiodactyla.
In The Phylogeny and Classification of the Tetrapods, Vol. 2: Mam-
1981; Byers and Bekoff, 1981; Sowls, 1984; but see Packard et al.,
mals, ed. M. J. Benton, pp. 235-72. Systematics Association Special
1991, on captive D. tajacu). Volume 35B. Oxford: Clarendon.
Tayassuids first appear in the fossil record of South America in the Ghandi, S. S., & Getty, R. (1969). Cutaneous nerves of the trunk of the
Pliocene (Chapadmalalan); Catagonus, Dicotyles, Platygonus, and domestic pig with special reference to the spinal nerves, Part I: Cu-
Tayassu are represented by diagnostic specimens in the Pliocene, taneous nerves of the cervical region. Iowa State Journal of Science,
43, 15-30.
Pleistocene, and Holocene of that continent. Wetzel (1977) sug-
Gidley, J. W. (1904). New or little known mammals from the Miocene of
gested that a species of Platygonus colonized South America and South Dakota; American Museum Expedition of 1903. (By W. D.
produced the other South American tayassuid lineages via adaptive Matthew and J. W. Gidley. Part III, Dicotylidae, by J. W. Gidley.)
radiation. The hypothesis of relationships outlined here, however, Bulletin of the American Museum of Natural History, 20, 241-68.
indicates that all tayassuid genera known from South America were Fischer, G. (1814). Zoognosia Tabulis Synopticus Illustrata. Moscow: Nico-
lai Sergeidis Vsevolozsky.
distinct in North America by the late Hemphillian. If that is correct,
Hendey, Q. B. (1976). Fossil peccary from the Pliocene of South Africa.
a minimum of four·colonization events is required to account for
Science, 192, 787-9.
tayassuid diversity in South America. Herring, S. W. (1974). A biometric study of fusion and skull growth in
peccaries. Anatomical Embryology, 146, 459-69.
Hess, W. M., Flinders, J. T., Pritchett, C. L., & Allen, J. V. (1985). Character-
ization of hair morphology in families Tayassuidae and Suidae with
ACKNOWLEDGMENTS
scanning electron microscopy. Journal ofMammalogy, 66, 75-84.
Kiltie, R. A. (1981 ). Stomach contents of rain forest peccaries (Tayassu
M. C. Coombs, W. E. Bemis, S. W. Herring, R. M. Hunt, Jr., D. tajacu and T. pecari). Biotropica, 13, 234-6.
J. Klingener, R. H. Tedford, M. R. Voorhies, S. D. Webb, R. M. ( 1981). The function of interlocking canines in rainforest peccaries. Jour-
Wetzel, and M. 0. Woodburne aided my work on fossil tayassuids nal ofMammalogy, 62, 459-69.
Lande, R. (1980). Sexual dimorphism, sexual selection, and adaptation in
in many ways. Financial support was provided by the University
polygenic characters. Evolution, 34, 292-305.
of Massachusetts and University of Nebraska, and by grants from Langer, P. (1978). Anatomy of the stomach of the collared peccary Dicotyles
Sigma Xi, the Theodore Roosevelt Fund of the AMNH, and the tajacu (L. 1758) (Mammalia: Artiodactyla). Mammalia, 43, 235-45.
National Science Foundation (BSR 8601682; BSR 9021490). LeConte, J. L. (1848). On Platygonus compressus: a new fossil pachyderm.
Memoirs of the American Academy ofArts and Sciences, new series,
3, 257-74.
Leidy, J. (1856). Notices of some remains of extinct Mammalia, recently
REFERENCES discovered by Dr. F. V. Hayden, in the bad lands of Nebraska.
Proceedings of the Academy of Natural Sciences of Philadelphia,
Ameghino, F. ( 1904). Nuevas especies de mamfferos cretaeos y terciarios 8, p. 59.
de la Republica Argentina (continuaci6n). Annals de la Sociedad (1860). Description of vertebrate fossils. In Post-Pliocene Fossils of
Cientifica de Argentina, 58, 1-188. South Carolina, ed. F. S. Holmes, pp. 99-122.
Barbour, E. H. (1925). Prosthennops xiphidonticus sp. nov., a new fossil ( 1869). The extinct mammalian fauna of Dakota and Nebraska. Journal
peccary from Nebraska. Bulletin of the University of Nebraska State of the Academy ofNatural Sciences, Philadelphia, 7, 1-472 (p. 194).
Museum, 1, 25-32. Marsh, 0. C. (1875). Notice of new Tertiary mammals, Part 4. American
Butler, P. M. (1952). Molarization of the premolars in the Perissodactyla. Journal of Science, 109, 234-50.
Proceedings of the Zoological Society of London, 121, 819-43. Matthew, W. D. (1907). A lower Miocene fauna from South Dakota. Bulletin
Byers, J. A. (1981). Peaceable peccaries. Natural History (June), 61-6. of the American Museum of Natural History, 23, 169-219.
Byers, J. A., & Bekoff, M. (1981). Social, spacing, and cooperative behavior (1915). Hitherto unpublished plates of Tertiary Mammalia and Permian
of the collared peccary, Tayassu tajacu. Journal of Mammalogy, 62, Vertebrata. United States Geological Survey and American Museum
767-85. of Natural History, Monograph Series, No. 2.
Colbert, E. H. (1933). An upper Tertiary peccary from India. American (1924). Third Contribution to the Snake Creek Fauna. Bulletin of the
Museum Novitates, 635, 1-9. American Museum of Natural History, 50, 59-210.
(1935). A new fossil peccary, Prosthennops niobrarensis, from Brown Mayer, J. J., &. Brandt, P. N. (1982). Identity, distribution, and natural his-
County, Nebraska. Bulletin of the Universtiy of Nebraska State Mu- tory of the peccaries, Tayassuidae. In Mammalian Biology in South
seum, 1, 419-30. America, eds. M. A. Mares & H. H. Genoways, pp. 433-55. Storrs,
Cope, E. D. (1869). The Artiodactyla. American Naturalist, 23, 111-36. Conn.: Special Publication 6, Pymatuning Laboratory of Ecology.
(1878). Descriptions of new extinct vertebrata from the upper Tertiary Packard, J.M., Babbit, K. J., Franchek, K. M., & Pierce, P. M. (1991). Sexual
and Dakota formations. Bulletin of the U.S. Geological Geographic competition in captive collared peccaries. Applied Animal Behaviour
Survey of the Territories, 4, 379-96. Science, 29, 319-26.
(1879). Second contribution to a knowledge of the Miocene fauna of Ore- Pearson, H. S. (1923). Some skulls of Perchoerus (Thinohyus) from the
gon. Proceedings of the American Philosophical Society, 18, 370-6. White River and John Day Formations. Bulletin of the American
Dalquest, W.W., & Mooser, 0. (1980). Late Hemphillian mammals of the Museum of Natural History, 48, 61-96.
Ocote local fauna, Guanajuato, Mexico. Texas Memorial Museum, (1927). On the skulls of early Tertiary Suidae, together with an ac-
Pearce-Sellards Series, 32, 1-25. count of the otic region in some other primitive Artiodactyla. Philo-
Darwin, C. (1871). The Descent of Man, and Selection in Relation to Sex. sophical Transactions of the Royal Society, London, series B, 215,
London: J. Murray. 389-462.
Douglass, E. (1903). New vertebrates from the Montana Territory. Annals Peterson, 0. A. (1905). New suilline remains from the Miocene of Nebraska.
of the Carnegie Museum, 2, 145-99. Memoirs of the Carnegie Museum, II, 305-25.
 Tayassuidae 401

Pickford, M. (1983). On the origins of Hippopotamidae together with de- Tyson, E. (1683). Anatomy of the Mexico musk hog. Philosophical Trans-
scriptions of two new species, a new genus and a new subfamily from actions, 13, 359-85.
the Miocene of Kenya. Geobios, 16, 193-217. Werner, H.J., Dalquest, W.W., & Roberts, J. H. (1952). Histology of the
Scott, W. B. (1940). The mammalian fauna of the White River Oligocene, scent gland of the peccaries. Anatomical Record, 113, 71-7.
Part 4: Artiodactyla. Transactions of the American Philosophical Wetzel, R. M. ( 1977). The Chacoan peccary, Catagonus wagneri (Rusconi).
Society, new series, 28, 363-746. Bulletin of the Carnegie Museum of Natural History, 3, 1-36.
Simpson, G. G. (1945). The principles of classification and a classification Wetzel, R. M., Dubois, R. E., Martin, R. L., & Myers, P. (1975). Catagonus,
of mammals. Bulletin of the American Museum of Natural History, an "extinct" peccary, alive in Paraguay. Science, 189, 379-81.
85, 1-350. White, T. E. (1941 ). Additions to the fauna of the Florida Pliocene. Proceed-
Sinclair, W. J. (1905). New or imperfectly known rodents and ungulates from ings of the New England Zoological Club, 18, 27-91.
the John Day series. University of California Publications, Bulletin Woodbume, M. 0. (1968). The cranial myology and osteology of Dicotyles
of the Department of Geological Sciences, 4, 125-43. tajacu, the collared peccary, and its bearing on classification. Mem-
Skinner, M. F., Skinner, S. M., & Gooris, R. J. (1977). Stratigraphy and oirs of the Southern California Academy of Science, 7, 1-48.
biostratigraphy of late Cenozoic deposits in central Sioux County, (1969). Systematics, biogeography, and evolution of Cynorca andDyseo-
Nebraska. Bulletin ofthe American Museum ofNatural History, 158, hyus (Tayassuidae). Bulletin of the American Museum of Natural
263-370. History, 141, 273-356.
Slatkin, M. (1984). Ecological causes of sexual dimorphism. Evolution, 38, Wright, D. B. (1989). Phylogenetic relationships of Catagonus wagneri: sis-
622-30. ter taxa from the Tertiary of North America. In Advances in Neotrop-
Sowls, L. K. (1984). The Peccaries. Tucson: University of Arizona Press. ical Mammalogy, eds. J. F. Eisenberg & K. H. Redford, pp. 281-308.
Stehlin, H. G. (1899). Ueber die Geschichte des Suiden-Gebisses, I. Ab- Gainesville: Sandhill Crane Press.
handlungen der Schweizerischen Paliiontologischen Gesellschaft, (1991). Cranial morphology, systematics, and evolution of Neogene
26, 1-336. Tayassuidae (Mammalia). Unpublished Ph.D. Thesis. Amherst: Uni-
(1900). Ueber die Geschichte des Suiden-Gebisses, II. Abhandlungen versity of Massachusetts.
der Schweizerischen Paliiontologischen Gesellschaft, 27, 337-527. (1993). Evolution of sexually dimorphic characters in peccaries (Mam-
Steiner, P. E., & Ratcliffe, H. L. (1968). The hepatic lobules of Suidae, malia, Tayassuidae). Paleobiology, 19, 52-70.
Tayassuidae, and Hippopotamidae. Anatomical Record, 160, 531-7. Wright, D. B., & Eshelman, R. E. (1987). Miocene Tayassuidae (Mammalia)
Stirton, R. A., &. Woodbume, M. 0. (1965). Revision of the Oligocene from the Chesapeake Group of the mid-Atlantic coast and their bear-
and Miocene peccaries. Special Paper of the Geological Society of ing on marine-nonmarine correlation. Journal of Paleontology, 61,
America, 82, 281. 604-18.
Stock, C. (1937). A peccary skull from the Barstow Miocene of Cal- Wright, D. B., & Webb, S. D. (1984). Primitive Mylohyus (Artiodactyla:
ifornia. Proceedings of the National Academy of Sciences, 23, Tayassuidae) from the Late Hemphillian Bone Valley of Florida.
398-404. .. Journal of Vertebrate Paleontology, 3, 152-9.
27 Oreodontoidea
BRUCE LANDER

INTRODUCTION

Oreodonts (superfamily Oreodontoidea) were small- to medium-

sized tylopod artiodactyls with a unique combination of primitive
piglike skeletal and advanced dental characters (notably enlarged
caniniform Cl and pl and incisiform cl). They are known ex-
clusively from North America and are one of a number of en-
demic artiodactyl groups, arising from bunoselenodont homacodon-
tid "dichobunids" and appearing during the early middle Eocene.
The Agriochoeridae, the earlier and, in most respects, less derived
members of the Oreodontoidea, retain many primitive characters
also found in their homacodontid ancestors and in their descen-
dants, the Merycoidodontidae (particularly the earliest, most prim-
itive merycoidodonts).
Agriochoerids, a morphologically homogeneous group recorded
from the latest Bridgerian or early Uintan to late early Arikareean
(early middle Eocene to late Oligocene), were taxonomically diverse
during the earlier part of their history, when up to six taxa existed
contemporaneously. However, all eight agriochoerid genera are al-
lied by marked parallels in development of numerous derived dental
and postcranial characters (especially molariform P3-4 and clawed
ungual phalanges in later forms; see Figures 27 .1, 27 .2) not seen in
merycoidodontids or homacodontids and rarely observed in other
artiodactyl groups. Only one genus (Agriochoerus) is recorded after
the Chadronian (late Eocene), and Agriochoerus is the only genus
with a definite record past the early Chadronian. Agriochoerids were
probably browsers that usually inhabited gallery forests. Clawed un-
gual phalanges in later forms, as well as other postcranial characters,
suggest some were partly arboreal.
The Merycoidodontidae, the more derived members of the Ore-
odontoidea, developed complexly crested Pl-3, more hypsodont,
fully selenodont molars, a lacrimal fossa, and complete postor-
bital bar, and lost the pollex (see Figure 27.3), although these fea-
tures did not characterize the earliest members of this family. Some
later merycoidodontids had tapirlike skulls, suggesting the pres-
ence of a mobile upper lip, or short proboscis. Unlike agriocho- Figure 27 .1. Restorations of oreodonts: A. Agriochoerus (Agriochoeridae).
erids, merycoidodontids did not develop molariform premolars B. Leptauchenia (Merycoidodontidae), by Brian Regal.

402
 Oreodontoidea 403

A A

Figure27.3. Osteology of Prodesmatochoerus periculorum: A. Skull (scale

= !Ocm) (modified from Scott, 1940). B. Dentition (scale= 3 cm) (modified
from Schultz and Falkenbach, 1968). C. Skeleton (scale= 20 cm) (modified
from Scott, 1940). (Drawings by Brian Regal.)

DEFINING FEATURES OF THE

SUPERFAMILY OREODONTOIDEA

Figure 27.2. Osteology of Agriochoerus antiquus (after Wortman, 1895). A.

CRANIAL
Skull (scale= 10 cm). B. Dentition. C. Skeleton (scale = 10 cm). D. Manus
The skull in agriochoerids and the earliest, most primitive mery-
(scale= 5 cm). (Drawings by Brian Regal.)
coidodontids lacks a lacrimal fossa and a complete postorbital bar.
The dentary is comparatively short and deep, the symphysis and
(although the P4 is fully selenodont) or clawed ungual phalanges. ii-cl steeply inclined. Primitively short diastemata are developed
They were also much more diverse taxonomically and in dental and only to accommodate the caniniform teeth, and the dentition is unre-
cranial morphology. Recorded from the early late Duchesnean to late duced.
Hemphillian (late middle Eocene to latest Miocene), merycoidodon- Agriochoeridae: rostral premaxillary suture very short. External
tids were taxonomically diverse throughout most of their history. Of narial opening unretracted. Facial/lacrimal fossa undeveloped. Pos-
the nineteen genera recognized, as many as ten may have existed torbital bar incomplete. Sagittal crest prominent. Braincase unin-
contemporaneous! yin the earliest Miocene. Merycoidodontids were ffated. Supraoccipital wing long, posteriorly projected.
probably browsers or mixed browser/grazers that lived in herds and Merycoidodontidae: in all but the most primitive forms, lacrimal
usually inhabited gallery forests. fossa developed and postorbital bar complete.
 Bruce Lander 404

DENTAL neck, limbs, and feet (although the foot stance is digitigrade, in con-
trast to the unguiligrade stance of present-day pigs). The postcranial
The anterior dentition is highly modified (Cl enlarged, D shaped skeleton, especially the tarsus, is particularly piglike in retaining
in cross section; cl reduced, incisiform, spatulate; pl enlarged, many primitive features: ulna and fibula unreduced, not fused with
caniniform, bladelike). In agriochoerids and the earliest, most prim- radius and tibia, respectively; distal astragular trochlea offset rela-
itive merycoidodontids, the molariform teeth are brachydont and tive to proximal, retaining carina separating unfused navicular and
bunoselodont (crests weak; fossettes and fossettids very shallow, cuboid; metapodials short, unfused, primitively unreduced in num-
enamel lined, with prominent ribs); the P4 is incipiently molariform ber; manus five toed in agriochoerids and earliest, most primitive
(parametacone lingual surface with shallow vertical groove; tiny merycoidodontids; pes five toed in earliest, most primitive agrio-
hypocone developed); the Ml-3 protocone, incompletely seleno-
choerids.
dont (postprotocrista developed, extending from protocone-fossette Agriochoeridae: pollex retained. Ungual phalanges hoofed in the
rib juncture toward metacone, forming incipiently bifurcated pro- Protoreodontinae, clawed in the later and larger Agriochoerinae.
tocone and extension of prefossette posterior to rib [postprefos- Merycoidodontidae: pollex absent in later forms. Ungual pha-
sette], protoconule ~etained); the metaconule, greatly enlarged; the
langes hoofed.
hypocone, absent; the mesostyle, prominently developed; the ml-3
paraconid absent; the ml-m3 hypoconid incompletely selenodont
(prehypocristid directed toward metaconid); and the metaconid and SYSTEMATICS
entaconid ribs and the metastylid are prominent. Many of these
characters are variously developed in other artiodactyl groups.
Agriochoeridae: Pl-2 and p2-3 simple. Molariform teeth brachy- SUPRA-SUPRAFAMILY
dont and bunoselenodont, with weak crests; fossettes and fossettids Oreodonts are distinguished from other artiodactyl groups by their
very shallow, enamel lined, with prominent ribs. P3 partially molari- unique combination of primitive skeletal and advanced dental fea-
form, with a distinct protocone, and separate paracone and metacone tures. This mosaic of primitive and derived characters has resulted
incipiently developed from parametacone. P4 fully molariform, with in numerous interpretations regarding oreodont affinities. The mod-
separate paracone and metacone; small hypocone developed by ified anterior dentition and incompletely selenodont molars suggest
union of isolated fossette cusp at metacone base and spur extend- oreodonts are tylopods intermediate between oromerycids and pro-
ing anteriorly from postcingulum. Ml-3 protocone incompletely toceratids (see Black, 1978). The piglike nature of the oreodont tar-
selenodont, with the postprotocrista extending from protocone-rib sus is a primitive character also found in oromerycids and primitive
juncture toward metacone, forming an incipiently bifurcated pro- protoceratids (Matthew, 1910; see Scott, 1899, 1940). Golz (1976)
tocone; paracone and metacone very recumbent, concave, highly has noted similarities between the molars of oromerycids and prim-
selenodont; ribs greatly reduced or lost; parastyle enlarged, bul- itive agriochoerids. An alternative viewpoint (Gentry and Hooker,
bous; mesostyle inflated, pocketed, invaded by transverse valley; 1988) places oreodonts as the sister group to all other tylopods, in-
labial cingulum reduced, absent on parastyle and mesostyle. The cluding the European taxa such as cainotheres, anoplotheres, and
ml-3 hypoconid incompletely selenodont, prehypocristid directed xiphodontids.
toward metaconid; metaconid and entoconid comparatively conical; Wilson (1974) and Simpson (1945) considered Oreodonta
ribs prominent; metastylid situated anterior to metaconid-entoconid (Osborn, 1910) (= Merycoidodontoidea) suborder and infraorder,
juncture; preentostylid developed posterior to juncture, giving ap- respectively. Hay (1930) proposed Agriochoeriformes as a suborder,
pearance of twinned metastylid. presumably rejecting "Oreodonta" because it is ultimately based on
Merycoidodontidae: prominent Pl-3 with anterior intermediate, an indeterminate oreodontid (Oreodon). "Oreodonta" and Agrio-
median crests developed; P3-4 premolariform; P4 fully selenodont, choeriformes were preceded by Artionychia (Osborn and Wortman,
resembling anterior lobe of M 1-3; parametacone undivided; hypo- 1893), which was proposed as a suborder on the mistaken be-
cone absent; p2-3 wiith accessory crests. Molars usually more hyp- lief that its type genus (Artionyx = Agriochoerus) was an ancylo-
sodont than in agriochoerids, selenodont with prominent crests; pod perissodactyl with an artiodactyl-like tarsus. However, Romer
fossettes/ fossettids deep; enamel lining thin/absent; fossette ribs (1966) assigned oreodonts to the suborder Ruminantia, infraorder
lost. Ml-3 completely selenodont, with postprotocrista extending Tylopoda, effectively rejecting the other names. The Tylopoda is
toward mesostyle; paracone and metacone erect with prominent now considered as its own artiodactyl suborder, of equal status to
ribs; ml-3 completely selenodont, with prehypocristid extending the Ruminantia.
toward metastylid, metaconid and entoconid comparatively selen- Oreodontoidea (Gill, 1872) was replaced with Agriochoeroidea
odont, metastylid midway between metaconid and entoconid, di- by Hay (1930) (see earlier), which in turn was replaced with Mery-
rected medially to posteromedially; ml-3 stylids and metaconid/ coidodontoidea (type genus: Merycoidodon, an indeterminate ore-
entoconid ribs reduced. odontid; see Lander, 1977) by Thorpe (1937) becauseMerycoidodon
was described prior to Agriochoerus. However, Lander (1977) res-
urrected Oreodontoidea because it has priority.
POSTCRANIAL
The Merycoidodontidae perhaps should be more properly termed
The oreodont skeleton is rather piglike, being somewhat stoutly the Oreodontidae (Leidy, 1869), based on Oreodon (Leidy, 1851).
built with a comparatively long trunk and tail and a short head, This term has priority over Cotylopidae (Flower and Lydekker,
 Oreodontoidea 405

1891) (for Coty lops [Leidy, 1851 a]) and Merycoidodontidae derived from a primitive early or middle Uintan form of Protore-
(Thorpe, 1923) (for Merycoidodfm [Leidy, 1848]). The type spec- odon (P. new species or P. parvus). There appears to be no derived
imens for all three genera are oreodontids indeterminate below fa- character in these primitive forms of Protoreodon or in "P." peter-
milial level, but probably the same as Prodesmatochoerus pericu- soni that would remove them from the ancestry of the other early
lorum Cope or Eucrotaphus jacksoni Leidy (Lander, 1977). Lander merycoidodontids. However, the P4 parametacone and the Ml-3
(1977) employs the term Oreodontidae in preference to the term mesostyles of these other merycoidodontids are too primitive to be
Merycoidodontidae: Because the latter name for the family is so derived from the late Uintan P. annectens. But incipient splitting of
firmly ensconced in the recent literature, however, it is used here to the P4 parametacone to form a rudimentary paracone and metacone
avoid confusion. in certain other taxa, such as some species of Aclistomycter and at
least one specimen of Oreonetes, suggests that these lineages may
SUPRAFAMILY not have arisen from "P." petersoni. Moreover, presence of a lacrimal
fossa in these merycoidodontids suggests an origin within the Agri-
The Agriochoeridae are distinguished from the Merycoidodontidae ochoeridae separate from that of Bathygenys. Unfortunately, with-
in the retention of primitive features in the postcranium (pollex), cra- out pre-late Duchesnean merycoidodontid remains, a polyphyletic
nium (unmodified lacrimal, incomplete postorbital bar), and denti- origin for the Merycoidodontidae cannot conclusively be demon-
tion (Pl-3 lacking accessory crests, molariform teeth strated. Should separate origins from within the Agriochoeridae be
brachydont, bunoselenodont; Ml-3 protocone and ml-3 hypoconid documented for Prodesmatochoerus and any other lineage currently
incompletely selenodont) lost in merycoidodontids. Additionally, assigned to the Merycoidodontidae, the other lineage should be as-
agriochoerids have molariform P3-4; Ml-3 with very recumbent, signed to a new family.
concave, highly selenodont paracones and metacones, and enlarged
bulbous parastyles and mesostyles; and ml-3 with pre-entostylids.
INFRAFAMILY
The upper molars of later agriochoerines closesly resemble those of
some members of the Anthracotheriidae, which are only distantly Four subfamilies have been recognized previously within the Agri-
related to oredonts. Later agriochoerines also have clawed ungual ochoeridae: Oreodontinae (Gill, 1872) (for Oreodon), Agriocho-
phalanges. erinae (Gill, 1872) (for Agriochoerus), Protoreodontinae (Scott,
Three families are synonymized with Agriochoeridae (Leidy, 1890) (for Protoreodon), and Diplobunopsinae (Peterson, 1919) (for
1869) (type genus: Agriochoerus) because their type genera are agri- Diplobunops = Agriochoerus). Gill (1872) included the Agriocho-
ochoerids and Agriochoeridae has priority: Protoreodontidae (Scott, erinae and Oreodontinae, and Scott (1890) the Protoreodontinae,
1889) (for Protoreodon), Artionychidae (Osborn and Wortman, within the Merycoidodontidae. Hay (1902, 1930), however, reas-
1893) (for Artionyx = Agriochoerus), and Eomerycidae (Marsh, signed these subfamilies to the Agriochoeridae. Peterson (1919)
1894) (for Eomeryx = Agriochoerus). Earlier workers often used originally referred the Diplobunopsinae to the dichobunoid fam-
Agriochoeridae and Merycoidodontidae (or their synonyms) in- ily Anoplotheriidae, partly because the clawed ungual phalanges of
terchangeably, recognizing agriochoerids and merycoidodontids as Diplobunops resemble those of European Diplobune, but he later
subfamilies within one or the other family (see later). (1931) reassigned Diplobunops to the Agriochoeridae. Later work-
The Merycoidodontidae are distinguished from the Agriochoeri- ers have recognized the Agriochoeridae as a family distinct from the
dae in the loss of the pollex; development of a lacrimal fossa and Merycoidodontidae, but have not recognized subfamilies within it.
a complete postorbital bar in later, more derived forms; complexly Some workers (Thorpe, 193 7; Scott, 1945; Simpson, 1945), how-
crested Pl-3, more hypsodont molars with erect Ml-3 paracones ever, continued to recognize Protoreodon and the Protoreodontinae
and metacones; and completely selenodont Ml-3 protocones and as merycoidodontids. Gazin (1955) correctly assigned Protoreodon
ml-3 hypoconids. However, merycoidodontids do not develop the to the Agriochoeridae. Here only the subfamilies Protoreodontinae
specialized dental features (molariform upper premolars and an- and Agriochoerinae are recognized.
thracotherelike upper molars) or the clawed ungual phalanges that Oreodontinae (Cope, l 884a) and Cotylopinae (Flower and
characterize agriochoerids. Lydekker, 1891) were replaced with Merycoidodontinae by Hay
Gazin (1955) considered the late Uintan "Protoreodon" peter- (1902). However, all three subfamilies are invalid, being based on in-
soni (here recognized as a separate, unnamed genus) to be the determinate type specimens (see earlier). Following Lander (1977),
probable ancestor of the Merycoidodontidae because of the prim- the Merycoidodontidae includes eight subfamilies (three new) (see
itive premolariform nature of the P4, in which there is no split- later). Schultz andFalkenbach (1968) recognized eleven subfamilies
ting of the parametacone into a distinct paracone and metacone. (including Merycoidodontinae), six of which are rejected.
However, Wilson (1971) considered the merycoidodontids to be di- Figures 27.4 and 27.5 illustrate the proposed interrelationships
phyletic, deriving Bathygenys and Limnenetes (= Leptauchenia), within the genera of oreodonts recognized as valid in this chapter.
from a small species of Protoreodon ( = "P." petersoni), and de- Note in particular changes in the subfamilial positions of the gen-
riving Aclistomycter, Merycoidodon (= Prodesmatochoerus), and era Hypsiops and Brachycrus: Although their tapirlike skulls had
Prodesmatochoerus from a large species of Protoreodon ( = P. an- caused Schultz and Falkenbach (1940, 1950) to place them with
nectens). Bathygenys reveesi, like "P." petersoni, retains a vestigial the superficially similar Merycochoerus in the Merycochoerinae,
P4 hypocone and Ml-3 metaconules. Bathygenys may have arisen these skull shapes are clearly the result of parallel evolution. Details
from "P." petersoni but, along with this taxon, could also have been of the rostrum, the occiput, the basicranium (particularly the ear
 Bruce Lander 406

Figure 27 .4. Interrelationships of the

AGRlOCHOERINAE PROTOREODONTINAE
Agriochoeridae. Key to numbers
on cladogram: (I) AGRIOCHOERI-
DAE: Lacrimal fossa undeveloped;
postorbital bar incomplete; brain-
case uninflated; sagittal crest promi-
nent; supraoccipital wing projected
posteriorly; paroccipital process un-
expanded; tympanic bulla tiny, un-
inflated; dentition brachydont; fos-
settes/fossettids very shallow, lined
with thick enamel; P4 molariform;
M 1-3 paracone, metacone very re-
cumbent highly selenodont; proto-
conule retained; parastyle bulbous,
mesostyle highly inflated; Ml-3 pro-
tocone, ml-3 hypoconid incom-
pletely selenodont. (2) AGRIOCHO-
ERINAE: Larger size; muzzle con-
stricted between C-P2; Pl-2, pl-2 di-
astemata developed; ribs, protoconule
reduced/lost; parastyle bulbous; mesostyle highly inflated; m 1-3 preentostylid prominently developed; ungual phalanges clawed.
(3) Large size; tympanic bulla large, inflated; paroccipital process expanded posterolaterally; Pl-2 diastema reduced/lost; C-P2,
pl-2 diastemata elongate; 11-3, Pl reduced/lost; p4 molariform. (4) Very large size. (5) Large size; rostrum laterally expanded;
C greatly enlarged. (6) PROTOREODONTINAE: Smaller size. (7) Small size; P4 molariform; Ml-3 paracone, metacone, proto-
conule tall, conical, erect; protoconule prominent, ml-3 metaconid, entoconid tall, conical. (8) Ml-3 paracone, metacone h1g.hly
selenodont, very recumbent; ribs reduced/lost; parastyle bulbous; mesostyle highly inflated; protoconule redu.ced. (9) Smal.l size;
p4 incipiently molariform. (10) Very small size; tiny pit developed in Ml-3 transverse valley. (11) Larger size; P4 molar1form.
(12) Larger size; P4 molariform. (13) Moderately small size. (14) Medium size.

region), and the dentition have led to the reassignment of Hypsiops a guide to the synonyms of generic and species names used in this
to the Phenacocoelinae and Brachycrus to the Ticholeptinae. Simi- chapter.
larly, the less specialized Merycoides has also been transferred from Marked parallelism in dental evolution, conservative skeletal evo-
the Merycochoerinae to the Phenacocoelinae. (For a more detailed lution, and similarities between various dental and skeletal elements
discussion, see Lander, 1977.) and comparable elements in other mammalian resulted in numerous
Note also that the position of the Leptaucheniinae is problemati- misidentifications and highly artificial classifications of the agrio-
cal. As an early appearing group, it would seem probable that they choerids. In extreme cases, species were split among more than one
represent an early phylogenetic branching point. Yet leptaucheni- order or included members of more than one suborder. For example,
inines are highly derived in their own way and also possess features the members of the genus Agriochoerus were divided among five
that obviously arose in parallel with the condition in other later agriochoerid genera, oreodontids, anthracotheres, ancylopods, and
merycoidodontines (e.g., hypsodont cheek teeth, complete postor- mesonychid creodonts. The genus included members of six agrio-
bital bar, fully fan-shaped occiput, inflated tympanic bulla). Based choerid genera, merycoidodontids, and anthracotheres.
on the distribution of available characters, it is equally likely that The locality numbers listed for each genus refer to the list of uni-
they form the sister taxon to the Aclistomycteriinae or to the Min- fied localities in Appendix I. The acronyms for museum collections
iochoerinae. are listed in Appendix III.
The locality numbers may be listed in a couple of alternative ways.
Parentheses around the locality (e.g., [CPlOl]) mean the taxon in
INCLUDED GENERA IN THE question at that locality is cited as an "aff." or "cf." the taxon in
FAMILY AGRIOCHOERIDAE question. Parentheses are usually used for individual species, thus
implying the genus is firmly known from the locality, but the actual
Eight agriochoerid genera, of which six are new and unnamed, are species identification may be questionable. Question marks in front
recognized here. The Agriochoeridae are a morphologically homo- of the locality (e.g., ?CPlOl) mean the taxon is questionably known
geneous group in which genera are distinguished partly by differ- from that locality, thus implying some doubt thatthe taxon is actually
ences in average adult linear dimensions, the largest form being as present at that locality, either at the genus or the species level.
much as 3.5 times larger than its smallest contemporary. Changes
in linear dimensions through time, nearly 50 percent between strati-
PROTOREODONTINAE SCOTT, 1890
graphically superposed samples of one genus, allow recognition
of chronologically distinct and restricted subspecies within long- Characteristics: smaller than contemporary agriochoerines; ungual
ranging species. The editors' appendix (see end of chapter) provides phalanges hoofed.
 B F
A
G H
c
c:56J
MINIOCHOERINAE MERYCOCHOERINAE
LEPTAUCHENllNAE
I
t EUCROTAPHINAE
~.~
t
........ PHENACOCOELINAE
~-~~;._..;..;;..~~.;....;.;.;_~~
TICHOLEPTINAE

Figure 27.5. Interrelationships of the Merycoidodontidae. Key to skulls on cladogram (merycoidodontids redrawn from Lander, 1977 [origi-
nally taken from Schultz and Falkenbach, 1940, 1949, 1968, and Wilson, 1971], with help from Brian Regal). Scale = 5 cm. A. Agrio-
choeridae: Agriochoerus antiquus (after Wortman, 1895). B. Bathygeniinae: Bathygenys alpha. C. Aclistomycteriinae: Aclistomycter middletoni.
D. Leptaucheniinae: Sespia nitida. E. Miniochoeriinae: Prodestmatochoerus periculorum. F. Eucrotaphinae: Eucrotaphus trigonocephalus.
G. Merycochoerinae: Merycochoerus proprius. H. Phenacocoelinae: Merycoides harrisonensis. I. Ticholeptinae: Brachycrus sweetwaterensis.
Key to characters on cladogram: (1) MERYCOIDODONTIDAE: Lacrimal fossa undeveloped; postorbital bar incomplete; braincase unin-
flated; sagittal crest prominent; supraoccipital wing projected posteriorly; paroccipital process unexpanded; tympanic bulla tiny, uninflated; au-
ditory tube short; dentition brachydont; fossettes/fossettids shallow, lined with thick enamel; P4 incipiently molariform; Ml-3 paracone, meta-
cone comparatively erect; protoconule retained; Ml-3 protocone, ml-3 hypoconid incompletely selenodont; pollux lost in all but earliest
forms. (2) BATHYGENIINAE: Smaller size; lacrimal fossa developed; postorbital bar complete in later forms; braincase inflated; sagittal crest
greatly reduced;, molars completely selenodont, Ml-3 protoconule lost in later forms. (3) Lacrimal fossa developed; Ml-3 protoconule lost.
(4) ACLISTOMYCTERINAE: larger size; lacrimal fossa greatly enlarged and pocketed with well-defined margin. (5) Pl-3 with prominent ante-
rior crest; P4 premolariform. (6) LEPTAUCHENIINAE: Smaller size; facial region reduced; large facial vacuity developed in later forms; lacrimal
fossa very small, shallow; postorbital bar complete; tympanic bulla very large, highly inflated, globular; auditory tube very long; supraoccipital
wing very widely spread, incorporated into fully fan-shaped occiput; paroccipital process inflated, continuous with tympanic bulla; dentition hyp-
sodont; occlusal enamel very thin, crests, fossettes/fossettids lost with little wear; Pl-2 median crests reduced; Ml-3 labial cusps offset anteri-
orly relative to lingual cusps; ml-3 lingual cusps offset anteriorly relative to labial cusps; ml-3 ribs, stylids reduced. (7) Larger size; facial vacu-
ity larger; malar deeper below orbit; tympanic bulla larger. (8) Smaller size; paroccipital process more inflated; dentition more hypsodont; Ml-
3 styles greatly reduced; ml-3 ribs, stylids more reduced. (9) MINIOCHOERINAE: Lacrimal fossa large, deep, well defined. (10) Larger size;
postorbital bar complete; lambdoidal crest extending onto supraoccipital wing; tympanic bulla rugose; auditory tube long; Ml-3 protocone, ml-
3 hypoconid completely selenodont. ( 11) Smaller size; orbit small, posterodorsal corner flattened; supraoccipital wing widely spread, incorporated
into semi-fan-shaped occiput; tympanic bulla smooth, auditory tube short; occlusal enamel very thin; crests, fossettes/fossettids lost with little wear;
Ml-3 labial cusps offset anteriorly relative to lingual cusps; lingual cusps offset anteriorly relative to labial cusps; ml-3 ribs, stylids reduced.
(12) EUCROTAPHINAE: Paroccipital process expanded slightly anterolaterally; tympanic bulla large, inflated; auditory tube forming crest extending
onto tympanic bulla, forming anterior wall of hyoid fossa; Pl-3 transverse, robust; anterior intermediate crest reduced. (13) Pl-3 anterior intermediate crest
absent; P4 parametacone lingual surface vertically striated. (14) Larger size; facial region inflated; lacrimal fossa shallow; tympanic bulla anteroposteriorly
elongate; paroccipital process greatly expanded anterolaterally. (15) MERYCOCHOERINAE: Paroccipital process greatly expanded anteroposteriorly;
postglenoid process heavy, nearly square in outline; auditory tube cylindrical, enclosing hyoid fossa, not extending onto tympanic bulla; Pl-4 antero-
posteriorly elongate; Pl-3 anterior intermediate crest prominent, oriented anteromedially; M3 metastylid prominently projected posteriorly. (16) Smaller
size. (17) Larger size; rostral premaxillary suture extremely long; external narial opening extremely retracted; occiput fully fan shaped. (18) PHENACO-
COELINAE: Smaller size; auditory tube inflated. (19) Smaller size; rostral premaxillary suture moderately long; tympanic bulla kidney shaped and laterally
compressed. (20) Larger size; facial area higher; external narial opening more retracted, nasals shorter; tympanic bulla low, flattened. (21) Smaller size;
tympanic bulla globular. (22) Smallest size. (23) Larger size; small frontal vacuity developed. (24) Lacrimal fossa broad, shallow, fiat. (25) Tympanic
bulla conical, angular. (26) TICHOLEPTINAE: Occiput fully fan shaped; postglenoid process extremely light, transverse; tympanic bulla ventral surface
convex, auditory tube extremely inflated, saclike; P3 square; anterior intermediate crest oriented transversely; median crest reduced. (27) Smaller size;
external narial opening less retracted; tympanic bulla ventral surface convex. (28) Smaller size; rostral premaxillary suture shorter; external narial opening
less retracted; Pl-3 anterior intermediate crest forming anteroposteriorly elongated cusp. (29) Larger size; rostral premaxillary suture longer; postglenoid
process heavier; P3 anteroposteriorly elongate; intermediate crest oriented anteromedially. (30) Larger size; external narial opening more retracted; shorter
nasals. (31) Smaller size; rostral premaxi!lary suture shorter; external narial opening has retracted, nasals longer. (32) Larger size; rostral premaxillary suture
longer; external narial opening extremely retracted; deep facial fossa developed; tympanic bulla inflated, conical in later forms; M3 with small posterior
accessory lobe.
 Bruce Lander 408

Protoreodon Scott and Osborn, 1887 (synonyms: Characteristics: Smallest agriochoerid. P4 of earlier forms
Agriotherium [preoccupied], Chorotherium, Hyomeryx, less molariform than in other contemporary protoreodon-
Mesagriochoerus, Protagriochoerus) tines; tiny enamel-lined pit in transverse valley beyond
Type species: Protoreodon parvus Scott and Osborn, 1887. labial end of M 1-3 postprotocrista in latest form.
Type specimen: PU 10,398. Average length of Pl-M3: 44.0? mm.
Characteristics: Larger than other contemporary protoreo- Included species: "P." petersoni petersoni (Gazin) (known
dontines, P4 more molariform than in contemporary new from locality CP6B); "P." petersoni (Gazin), new sub-
genus A and new genus B, less molariform than in other species (small) (locality CP6B); "P." new species (small)
contemporary protoreodontines. Compared to new genus (localities CP29C, CP29D, CP39IIA, NP22).
A: Ml-3 parastyle more bulbous, mesostyle more inflated,
paracone and metacone more recumbent, ribs and proto-
New genus C
conule less prominent. Compared to other contemporary
Type species: "Agriochoerus" minimus (Douglass, 1901 ).
protoreodontines, Ml-3 parastyle and mesostyle less bul-
Type specimen: CM 709.
bous, paracone and metacone less recumbent.
Characteristics: Larger than new genus B, smaller than re-
Average length of Pl-M3: 52.0-74.0 mm.
maining contemporary protoreodontines; approximate size
Included species: P. parvus Scott and Osborn ( = P. minor
of new genus A.
Scott, Agriotherium paradoxicum Scott, Hyomeryx bre-
Average length of Pl-M3: 48.0 mm.
viceps Marsh) (known from localities CP6A, CP39IIB);
Included species: "A." minimus (Douglass) only (known from
P. annectens annectens (Scott) ( = P. medius Peterson,
localities, CP39IIA, NP22, NP24A, NP25B, NP29B).
Mesagriochoerus primus Peterson) (localities CC7C,
?CC9B SB43C, SB44A, CP6B, CP7 A, ?NPS); P. annec-
NewgenusD
tens pacijicus (Golz, new rank) (localities ?CC7B, ?CC7D,
Type species: "Agriochoerus" transmontanus (Stock, 1949).
CC9B, CC9BB, ?CP29C, ?CP29D, ?CP82); P. annectens?
Type specimen: LACM/CIT 3,558.
tardus (Scott) (localities CC7B, CC9A, CC9AA, CC9B,
Characteristics: Smaller than contemporary Protoreodon,
CP37B); P. annectens, new subspecies (large) (localities larger than remaining contemporary protoreodontines.
CPS, NP22, NP23A, NP23B, NP25A, NP25B); P. an-
Average length of Pl-M3: 57.0 mm.
nectens, new subspecies (medium) (localities CC4, CCS,
Included species: "A." transmontanus (Stock) (known from
SB43B, ?CP6A, ?CP36B, CP38B, ?CP65). locality NB2); "A." new species a (large) (localities GC23,
Protoreodon sp. is also reported from localities SB25C,
?SB25B, ?SB26B, SB43D, SB44B, CP7C); "A." new
CP65, NP9A. species b (small) (locality SB52);? "A." sp. indet. (locality
?SB43E).
New genus A
AGRIOCHOERINAE GILL, 1872
Type species: Unnamed.
(SYNONYMS: DIPLOBUNOPSINAE)
Type specimen: TMM 31,281-14.
Characteristics: Compared with new genus B: larger size (ap- Characteristics: Larger than contemporary protoreodontines. Cra-
proximate size of new genus C), P4 more molariform. nial, dental, and postcranial anatomy of earliest forms similar to
Compared with other contemporary protoreodontines: P4 that of contemporary protoreodontines. Muzzle constricted below
of earliest form less molariform; in later forms, M 1-3 para- Cl and P2 in later forms; Pl-2 and pl-2 diastemata developed;
cone and metacone taller, more conical and erect (resem- M 1-3 protoconule lost; M 1-3 preentostylid prominently developed;
bling some examples of Protoreodon parvus, but taller); ungual phalanges clawed.
parastyle much less bulbous; mesostyle much less inflated;
protoconule prominent; ml-3 metaconid and entoconid Agriochoerus Leidy, 1850b (synonyms: Agriomeryx,
taller, more conical. Artionyx, Coloreodon, Diplobunops, Eomeryx, Merycopater)
Average length of Pl-M3: 47.0 mm. Type species: Agriochoerus antiquus Leidy, l 850b.
Included species: New genus A, new species a (known from Type specimen: ANSP 10,665.
locality CP6B); new genus A, new species b, subspecies a Characteristics: Smaller than contemporary new genus E,
(medium) (localities SB42B, SB44A); new genus A, new approximate size of contemporary new genus F. Tym-
species b, new subspecies b (small) (localities SB43C, panic bulla inflated in later forms; paroccipital process ex-
SB44A); new genus A, new species b, new subspecies panded posterolaterally; Pl-P2 diastema reduced or lost;
c (medium) (locality SB43D); new genus A, new species Cl-P2 and pl-p2 diastemata expanded; Il-3 and Pl re-
indet. (small) (locality SB43B). duced/lost.
Average length of Pl-M3: 62.0-100.0 mm.
NewgenusB Included species/subspecies: A. antiquus antiquus Leidy,
Type species: "Protoreodon" petersoni (Gazin, 1955). (? = A. antiquus dakotensis Thorpe, A. latifrons Leidy,
Type specimen: PU 14,404. Mesonyx? dakotensis Scott) (known from localities
 Oreodontoidea 409

?CP40B, CP41A, CP41B, ?CP68C, CP83C, CP84A, INCLUDED GENERA IN THE

CP98C, CP99A, ?NP50B); A. antiquus guyotianus (Cope, FAMILY MERYCOIDODONTIDAE
new rank) (? = A. bullatus Thorpe, A. trifons Cope, Col-
oreodon ferox Cope, C. macrocephalus Cope, C. ryder- Riel (1964), Schultz and Falkenbach (1968), Stevens, Stevens, and
anus Cope) (localities CP84B, CP84C, PN6C); A. anti- Dawson (1969), Stevens (1970), Wilson (1971), and Ferrusquia-
quus major (Leidy, new rank) (localities ?NP27D, NP30B); Villafranca (1984) recognized a total of 48 genera (one unnamed)
A. antiquus? new subspecies (small) (localities CP49B, and 180 species (one unnamed) within the Merycoidodontidae.
CP84A, CP85D); A. antiquus? gaudryi (Osborn and Wort- However, Lander (1977) has demonstrated that the family is tax-
man) (= Agriomeryx migrans Marsh, Artionyx gaudryi Os- onomically oversplit at both levels. The taxonomy presented here
born and Wortman) (localities CP48, CP85C); A. matthewi follows that of Lander (1977) and recognizes only 19 genera and 42
matthewi (Peterson) (= Diplobunops leotensis Peterson species. The editors' appendix provides a guide to the synonyms of
[nomen nudum], D. uintensis Peterson, D. ultimus Peter- the generic and species names used in this chapter. For an alterna-
son) (locality CP6B); A. matthewi crassus (Peterson, new tive recent view of certain aspects of merycoidodont taxonomy, see
rank) (localities SB43C, CP7A); A. matthewi pearcei Stevens and Stevens (1996) and CoBabe (1996).
(Gazin, new rank) (localities CP29C, CP29D); A. pumilus
Marsh (= Eomeryx pumilus, Diplobunops vanhouteni BATHYGENIINAE (NEW)
Gazin) (locality CP6A);A. new species a (localities SB43D,
Characteristics: Smaller than other contemporary merycoidodon-
SB44B);A. ?new sp. b, new subspecies a(localities SB43E,
tids. Lacrimal fossa absent in earlier forms; postorbital bar incom-
SB44C, CP39IIC, NP29B); A. ?new sp. b, new subspecies
plete; braincase inflated in later forms; sagittal crest greatly re-
b (medium) (=A. intermedius Van Houten [nomen nudum])
duced; supraoccipital wing short, laterally projected, incorporated
(localities ?CP39B, CP39C, CP39IIA, CP66, ?NP24C,
into semi-fan-shaped occiput; tympanic bulla uninflated, low, slop-
NP24D, NP24E, NP29C); A. ?new sp. b, new subspecies
ing laterally. Dentition brachydont; vestigial P4 hypocone retained;
c (medium) (localities ?NB2, ?CP39B, CP39C, CP42B,
Ml-3 protocone incompletely selenodont in earlier forms, fossette
CP83B, CP98B).
cusp(s) developed beyond labial end of postprotocrista; protoconule
Agriochoerus sp. is also reported from localities CP83A,
retained.
NP9A, ?NP22.
Bathygenys Douglass, 1901 (synonyms: Megabathygenys,
NewgenusE Parabathygenys)
Type species: "Agriochoerus" maximus (Douglass, 1901). Type species: Bathygenys alpha Douglass, 1901.
Type specimen: CM 749. Type specimen: CM 708.
Average length of Pl-M3: 115.0 mm. Characteristics: As for Bathygeniinae.
Characteristics: Largest agriochoerid. Rostrum and tympanic Average length of Pl-M3: 33.0-36.0 mm.
bulla unknown; M 1-3 protoconule absent; paracone of ear- Included species/subspecies: B. alpha Douglass(= Megab-
liest form with prominent rib; paracone and metacone of athygenys goorisi Schultz and Falkenbach) (known from
latest form more recumbent than in other agriochoerines; localities SB43E, SB44D, CP39E, CP39IIA, ?CP42A,
ml-3 preentostylid prominent. NP24C, NP24D, NP25C); B. hedlundae (Schultz and
Included species: "A." maximus (Douglass) (known from lo- Falkenbach, new rank)(= Bathygenys alpha hedlundae)
calities SB52, CP39B, NP24D, NP24E);? "A." maximus (= Parabathygenys paralpha Schultz and Falkenbach)
(locality CP39C); "A." new species, (locality CP7C). (localities ?CP39B, CP39C, CP39G, CP42B, ?CP68A);
B. reevesi Wilson (localities SB44C, SB52).
Bathygenys sp. is also reported from localities SB43D,
NewgenusF
CP39B.
Type species: Unnamed.
Type specimen: OMSI specimen.
Characteristics: Smaller than new genus E, approximate size ACLISTOMYCTERINAE (NEW)
of contemporary Agriochoerus. Rostrum laterally expand- Characteristics: Larger than other contemporary merycoidodontids.
ed at Cl; tympanic bulla small, uninflated; Cl greatly en- Lacrimal fossa pocketed, extremely large, deep; postorbital bar in-
larged. complete; supraoccipital wing posteriorly projected; paraoccipital
Average length of Pl-M3: 88.0 mm. process U shaped; tympanic bulla rounded, slightly inflated. Den-
Included species: New genus F, new species only (C. B. Han- tition brachydont; P4 parametacone lingual surface with shallow
son, unpubl. data), known from locality PN5 only. vertical groove, vestigial hypocone retained; M 1-3 protocone, ml-
3 hypoconid incompletely selenodont.
INDETERMINATE AGRIOCHOERIDS
Aclistomycter Wilson, 1971
Unidentified agriochoerids have also been recorded from Chisos Type species: Aclistomycter middletoni Wilson, 1971.
Formation, Big Bend area, Texas (Uintan?). Type specimen: TMM 41,213-1.
 Bruce Lander 410

Characteristics: As for Aclistomycteriinae. NP38A); L. major brevifacies (Cope, new rank) (= Pithec-
Average length of Pl-M3: 66.0 mm. istes brevifacies) (? = L. densa Loomis, Brachymeryxfeli-
Included species: A. middletoni Wilson (known from locali- ceps Cope, Cyclopidius emydinus Cope, C. incisivus Scott,
ties SB43D, SB44B); A.?, sp. indet. (large) (locality C. simus Cope, Hadroleptauchenia extrema Schultz and
?CP39IIC). Falkenbach, Pithecistes decedens Cope, P. quadratus
Koerner, Pseudocyclopidius lullianus expiratus Schultz
andFalkenbach) (localities CP50, CP85D, CP86B, CP102,
LEPTAUCHEN IINAE SCHULTZ AND
NP34C, NP37); L. major loganensis (Koerner, new rank)
FALKENBACH, 1940
(= Cyclopidius loganensis) (? = L. martini Schultz and
Characteristics: Muzzle reduced; transversely paired, anterior nasal Falkenbach, Pseudocyclopidius quadratus [Koerner]) (lo-
prominences in some specimens (may merely reflect rugosity along calities CP48, ?CP85A, CP85B, CP99C, CPlOl, NP34B);
nasal/maxillary suture to support nasals); extremely large facial L. major schucherti (Thorpe, new rank) (= Cyclopidius
vacuity; lacrimal fossa very small, shallow; occiput fully fan shaped; schucherti, Pithecistes mariae Schultz and Falkenbach)
more derived forms have complete postorbital bar; paroccipital pro- (localities CP8SA, CP86A); L. major, new subspecies
cess inflated, continuous with globular, highly inflated tympanic (small) (locality CP86B); L. platyceps platyceps (Dou-
bulla; auditory tube long, meatus located high, posteriorly on skull. glass) (= Limnenetes platyceps) (localities ?SB44C,
Cheek teeth hypsodont; occlusal enamel thin; crests, fossettes/ fos- SB44D, NP23C, NP26, NP29A, NP29B); L. platyceps,
settids lost with light wear; II reduced/absent; Pl-3 anteriorly short- new subspecies (large) (localities ?CP42A, NP25B,
ened, nearly square, with central fossette; anterior intermediate crest NP25C); L. sp. indet. a (locality CP84A); L. sp. indet. b
on Pl prominent, anterior, median crests reduced; Ml-3 lingual se- (large) (localities CP52, CP103, CP104A); L. sp. indet. c.
lenes rounded, closely appressed, broadly continuous, not deeply (small) (locality CP86D); L. sp. indet. d. (locality NC2).
divided by transverse valley, forming single crest extending toward
mesostyle, offset posteriorly relative to labial selenes; ml-3 labial Sespia Stock, 1930 (synonym: Megasespia)
selenes rounded, offset anteriorly relative to lingual selenes; ribs, Type species: Sespia nitida (Leidy, 1869) (= Leptauchenia
stylids greatly reduced. nitida).
Type specimen: ANSP 10,870.
Leptauchenia Leidy, 1856 (synonyms: Brachymeryx, Characteristics: In comparison with contemporary Leptau-
Cyclopidius, C. [Chelonocephalus ], Hadroleptaucheni a, chenia: smaller size; facial vacuity smaller; malar shal-
Limnenetes, Pithecistes, Pseudocyclopidiu s, lower; paroccipital process more inflated; tympanic bulla
Pseudoleptauchen ia [lapsus calami]) smaller; dentition more hypsodont; Ml-3 styles, ml-3
Type species: Leptauchenia decora Leidy, 1856. ribs, stylids much less prominent; M3 reduced/absent.
Type specimen: ANSP 10,878. Average length of Pl-M3: 35.0-48.0 mm.
Characteristics: In comparison with contemporary Sespia: Included species/subspecies: S. nitida (Leidy) only (= S.
larger size; facial vacuity larger, malar deeper, paroccipital marianae Schultz and Falkenbach, S. ultima Schultz and
process less inflated, tympanic bulla larger; dentition less Falkenbach, Leptauchenia? [Sespia] califomica Stock, L.
hypsodont, Ml-3 styles, ml-3 ribs, stylids much more minora Schlaikjer, Cyclopidius heterodon Cope, Megas-
espia middleswarti Schultz and Falkenbach) (known from
prominent.
Average length of Pl-M3: 46.0-73.0 mm. localities ?CC9C, CC9D, CCI 1, CC12, CP48, CP50,
Included species/subspecies: L. decora Leidy (= L. harveyi CP69, CP85B, CC85C, CP86B, CP99C, CPlOl, CP102,
Schultz and Falkenbach, Hadroleptauchenia primitiva NPlOCC, NP34C, NP37).
Schultz and Falkenbach, Cyclopidius loganensis Koerner,
Pithecistes tanneri Schultz and Falkenbach, Pseudocyclo- MINIOCHOER INAE SCHULTZ AND
pidius frankforteri Schultz and Falkenbach) (known from FALKENBACH, 1956 (SYNONYMS:
localities CP42D, CP68D, CP84B, CP84C, CP99B); OREONETINA E, ?COTYLOPINA E,
L. eiseleyi eiseleyi (Schultz and Falkenbach) (= Hadrolep- ?MERYCOIDO DONTINAE, ?OREODONTIN AE)
tauchenia eiseleyi) (localities CP84A, ?CP99A); L. eise-
leyi orellaensis (Schultz and Falkenbach, new rank) Characteristics: Lacrimal fossa moderately large, deep; paraoccipital
(= Pseudocyclopidius orellaensis Schultz and Falkenbach) process with lateral surface; post-tympanic process present; tym-
(localities SBl, ?CP39H, CP40B, CP41B, CP84A); L. ma- panic bulla small, uninflated, globular; dentition brachydont.
jor major Leidy (? = L. margeryae Schultz and Falken-
bach, L. parasimus Schultz and Falkenbach, Cyclopidius Oreonetes Loomis, 1924 (synonyms: Miniochoerus,
lullianus Thorpe, Hadroleptaucheni a shanafeltae Schultz M. [Paraminiochoeru s], Platyochoerus, Stenopsochoerus,
and Falkenbach, Pithecistes altageringensis Schultz and S. {Pseudostenopsoc hoerus], Parastenopsochoe rus)
Falkenbach, P. copei Schultz and Falkenbach) (localities Type species: Oreonetes anceps (Douglass, 1901) (= Limne-
CP48, ?CP69, CP84C, ?CP85A, CP85C, CP99C, CPlOl, netes [?] anceps).
 Oreodontoidea 411

Type specimen: CM 745. Type specimen: AMNH 6397.

Characteristics: Postorbital bar incomplete in earlier forms; Characteristics: Postorbital bar complete. In comparison with
in later forms, facial region inflated. In comparison with contemporary Oreonetes: larger size; in later forms lacri-
contemporary Prodesmatochoerus: smaller size, lacrimal mal fossa larger, deeper; orbit larger, rounder; supraoc-
fossa smaller and shallower; orbit smaller, posterodorsal cipital wing longer, posteriorly projected; tympanic bulla
comer flattened; supraoccipital wing shorter, projected lat- slightly smaller; auditory tube longer; cheek teeth occlusal
erally, incorporated into semi-fan-shaped occiput; tym- enamel thicker, crests, fossettes/fossettids retained through
panic bulla smoother, slightly larger; auditory tube shorter; greater wear; Ml-3 protocone completely selenodont,
cheek tooth occlusal enamel thinner; crests, fossettes/ postprotoselene extending toward mesostyle roughly
fossettids lost with less wear; Pl-3 squarer; Ml-3 lingual parallel to, but separated from, anterior crest of meta-
selenes more rounded, closely appressed, broadly continu- conule by transverse valley; ml-3 ribs and stylids promi-
ous, less divided by transverse valley, forming single crest nent.
extending toward mesostyle, offset posteriorly relative to Average length of Pl-M3: 63.0-97.0 mm.
labial selenes; ml-3 labial selenes offset anteriorly rel- Included species/subspecies: P. periculorum periculorum
ative to lingual selenes; ribs, stylids reduced. (Dentition (Cope) (= Oreodon culbertsonii periculorum) ( = P. meek-
superficially resembling that of Leptauchenia, see earlier.) break ae Schultz and Falkenbach, ? = Merycoidodon cul-
Average length of Pl-M3: 43.0-79.0 mm. bertsonii Leidy [nomen vanum], M. culbertsonii osbomi
Included species/subspecies: 0. anceps anceps (Douglass) Schultz and Falkenbach, Otionohyus wardi Schultz and
(known from localities CP39IIA, NP23C, NP25B, NP26, Falkenbach, Paramerycoidodon georgei Schultz and Falk-
NP29B ); 0. anceps douglassi Schultz and Falkenbach (lo- enbach, ? = Cotylops speciosa Leidy [nomen vanum],
cality NP25C); 0. gracilis gracilis (Leidy) (= Oreodon ? = Oreodon priscum Leidy [nomen van um], ? = 0. robus-
gracile, Miniochoerus battlecreekenesis Schultz and tus Leidy [lapsus calami]) (known from localities CP40B,
Falkenbach, M. [Paraminiochoerus] gracilis [Leidy], CP41B, CP68C, CP84A, CP99A, NP27C, NP50B);
M.[P.] affinis [Leidy][= Oreodon affinius]), Oreodon mi- P. periculorum lewisi (Clark et al., new rank) (= Mery-
nor Cope [nomen nudum], 0. gracilis coloradoensis Cope, coidodon lewisi) (= Merycoidodon culbertsonii browni
Stenopsochoerus stembergi Schultz and Falkenbach, S. Schultz and Falkenbach, Otionohyus wardi degrooti
[Pseudostenopsochoerus] douglasensis Schultz and Falk- Schultz and Falkenbach, 0. ?vanderpooli Schultz and
enbach, Parastenopsochoerus conversensis Schultz and Falkenbach) (localities CP39B, CP3911, CP40A, CP41A,
Falkenbach (? = Platyochoerus platycephalus [Thorpe], CP68B, CP83C, CP98C, NP24C, ?NP27B, NP50A); P.
Merycoidodon platycephalus Schultz and Falkenbach), (lo- periculorum, new sub-species a (medium) (localities
calities CP40B, CP4 l B, CP68C, CP84A, CP99A); 0. gra- CP84A, CP99A); P. periculorum, new subspecies b (large)
cilis chadronensis (Schultz and Falkenbach, new rank)(= (localities CP84B, CP84C); P. macrorhinus macrorhinus
Stenopsochoerus [Pseudostenopsochoerus] chadronensis) (Douglass, new rank) (= 0. macrorhinus) (= 0. robus-
(= S.[P.] reideri Schultz and Falkenbach) (localities tumu Douglass [preoccupied]) (localities SB43E, SB44D,
CP40A, CP41A, CP68B, CP83C, ?NP29B); 0. gracilis SB44E, ?CP39B, CP39C, CP39E, ?CP39E, CP39IIA,
cheyennensis (Schultz and Falkenbach, new rank) (=Mini- ?CP42A, CP45A, CP83A, ?NPlOB, NP23C, NP24E,
ochoerus cheyennensis) (= Miniochoerus [Paraminioch- NP25B, NP26, NP29B, NP29C, NP29F, NP32A); P.
oerus] ottensi Schultz and Falkenbach, Platyochoerus hat- macrorhinus dunagani (Wilson, new rank) (= Merycoi-
creekensis Schultz and Falkenbach) (localities CP39I, dodon dunagani) (localities SB44C, SB52, NP23B,
CP42D, CP68D, CP84B, CP98C, CP99B, NP50C); NP24A, NP25A); P. macrorhinus natronensis (Schultz and
0. gracilis nicholsae (Schultz and Falkenbach, new rank) Falkenbach, new rank) (= Merycoidodonfors ythae Schultz
(= Oreonetes nicholsae) (= Stenopsochoerus berardae and Falkenbach) (localities CP39C, CP39F, CP42B,
Schultz and Falkenbach) (localities CP68C, CP84A, CP43, ?CP67, CP83B, CP98B, NP24C, NP24D,
CP99A, NP50B); 0. gracilis starkensis (Schultz and Falk- ?NP29B).
enbach, new rank)(= Miniochoerus [Paraminiochoerus]
starkensis) (= M. [P.] helprini Schultz and Falkenbach,
Stenopsochoerus joderensis Schultz and Falkenbach,
EUCROTAPHINAE (NEW)
Platyochoerus heartens is Schultz and Falkenbach) (local- Characteristics: Supraoccipital wing posteriorly projected; paraoc-
ities CP68C, CP84A, CP98B, CP99A, NP50B). cipital process expanded anterolaterally, incorporating lateral sur-
face of post-tympanic process; tympanic bulla large, inflated; au-
Prodesmatochoer us Schultz and Falkenbach, 1954 ditory tube forming anterior wall of hyoid fossa, extending onto
(synonyms: Genetochoerus, ?Merycoidodon, ?Oreodon, bulla. Dentition brachydont; Pl-3 wide transversely, massive; an-
?Cotylops, Paramerycoidodo n, Otionohyus) terior intermediate, median, and posterior crests on P3 reduced;
Type species: Prodesmatochoerus periculorum (Cope, 1884) P4 anteroposteriorly shortened; M3 metastyle weak, projected
(= Oreodon culbertsonii periculorum). labially.
 Bruce Lander 412

Eucrotaphus Leidy, 1850 (synonyms: Dayohyus, odon danai Koerner, Otionohyus [Otarohyus] hybridus
Genetochoerus [Osbornohyus 1 Merycoidodon [Leidy] [= Oreodon hybridus ], Paramerycoidodon [Gre-
[Anomerycoidodon 1 Otionohyus [Otarohyus 1 goryochoerus] wanlessi Schultz and Falkenbach) (known
Paramerycoidodon [Barbourochoerus1 from localities CP42D, CP84B, CP99B, NP34A).
Pseudogenetochoerus, Subdesmatochoerus)
Type species: Eucrotaphus jacksoni Leidy, l 850a. MERYCOCHOERINAE SCHULTZ AND
Type specimen: ANSP 10,678. FALKENBACH, 1940 (SYNONYMS:
Characteristics: Earlier forms essentially identical to later DESMATOCHOERINAE, EPOREODONTINAE,
forms of Prodesmatochoerus (see earlier), except for
PROMERYCOCHOERINAE)
inflated tympanic bulla. In comparison with contemporary
Blickohyus: smaller size; facial region uninflated; lacrimal Characteristics: Paroccipital process expanded anterolaterally, lat-
fossa deeper, more defined; zygomatic arch lighter; sagittal eral edge turned anteriorly; postglenoid process heavy; post-tympa-
crest less prominent; postglenoid process heavier; paraoc- nic process absent; tympanic bulla conical; auditory tube cylindrical,
cipital process narrower; post-tympanic process more ap- enclosing hyoid fossa or tapering venteromedially to form a promi-
parent; tympanic bulla more conical and cylindrical; P4 nence lateral to fossa; Pl-3 rectangular, anteroposteriorly elon-
parametacone lingual surface with vertical striations in gated; median crests prominent; anterior intermediate crests very
later forms. prominent, extending anteromedially; P4 anterior intermediate crest
Average length of Pl-M3: 82.0-89.0 mm. developed; M3 metastyle prominent, posteriorly projected.
Included species: E. jacksoni jacksoni Leidy (= Geneto-
choerus [Osbomohyus] norbeckensis Schultz and Falken- Eporeodon Marsh, 1875 (synonyms: Eporeodon
bach, G.[O.] dickinsonensis [Douglass] [= Eucrotaphus [Paraeporeodon 1 Desmatochoerus [Paradesmatochoerus 1
dickinsonensis], E. socialis Marsh, Merycoidodon Hypselochoerus, Mesoreodon, Promesoreodon,
[Anomerycoidodon] dani Schultz and Falkenbach, Otiono- Promerycochoerus [Desmatochoerus 1
hyus [Otarohyus] bullatus [Leidy][= Oreodon bullatus], Pseudodesmatochoerus)
0.[0.] cedrensis [Matthew][= Eporeodon majorcedren- Type species: Eporeodon occidentalis (Marsh, 1873) ( = Ore-
sis], 0.[0.] hybridus helenae [Douglass][= Eucrotaphus odon occidentalis).
helenae], Paramerycoidodon [Barbourochoerus] bacai Type specimen: YPM 10142.
Schultz and Falkenbach, P. [B.] harrisi Schultz and Falken- Characteristics: In comparison with contemporary Meryco-
bach [lapsus calami], Subdesmatochoerus socialis dakot- choerus: smaller size; facial region smaller; rostral pre-
ensis Schultz and Falkenbach) (known from localities maxillary suture shorter; malar shallower; sagittal crest
CP42C, CP68C, CP84A, CP84B, CP99A, CP99B, NP27D, less prominent; zygomatic arch lighter, apex lower; denti-
NP29D, NP30B, NP50B); E. jacksoni galushai (Schultz tion brachydont.
and Falkenbach, new rank) ( = Genetochoerus [Osbomo- Average length of Pl-M3: 84.0-135.0 mm.
hyus] geygani) (localities CP68D, CP84B, NP50C); E. tri- Included species: E. occidentalis occidentalis (Marsh)
gonocephalus Cope (= Dayohyus trigonocephalus, (=Oreodon occidentalis) (= E. davisi Schultz and Falken-
D. wortmani Schultz and Falkenbach, Pseudogenetocho- bach, E. perbullatus Thorpe, E. [Pareporeodon] paci.fi-
erus covensis Schultz and Falkenbach, P. condoni [Thorpe] cus [Cope] [= E. jacksoni paci.ficus], E.[P.] leptacanthus
[= Eporeodon condoni]) (locality PN6C). [Cope] [= Eucrotaphus jacksoni leptacanthus], E.[P.]
longifrons [Cope] [= E. major longifrons], Otinohyus
Blickohyus (Schultz and Falkenbach, 1968, new rank) [Otarohyus] alexi Schultz and Falkenbach, Promesoreo-
(synonyms: Paramerycoidodon [Blickohyus ]; don scanloni Schultz and Falkenbach, Subdesmatochoerus
Paramerycoidodon [Gregoryochoerus]) shannonensis Schultz and Falkenbach) (known from lo-
Type species: Blickohyus auritus (Leidy, 1852) (= Eucro- calities CP42D, CP84B, CP99B, PN6C); E. occidentalis
taphus auritus). major (Leidy, new rank) (? = Paramerycoidodon [Grego-
Type specimen: ANSP 10,672. rychoerus] meagherensis [Koerner][= E. meagherensis],
Characteristics: Larger than contemporary Eucrotaphus. Desmatochoerus hatcheri geringensis Schultz and Falken-
Muzzle inflated; lacrimal fossa shallow, poorly defined; zy- bach, D.[Paradesmatochoerus] grangeri Schultz and
gomatic arch heavy; sagittal crest prominent; paraoccipital Falkenbach, D. [P] wyomingensis Schultz and Falkenbach,
process laterally expanded, lateral edge turned anteriorly; Subdesmatochoerus montanus [Douglas] [= Eucrotaphus
post-tympanic process absent; tympanic bulla anteropos- montanus], Promerycochoerus [Desmatochoerus] curvi-
teriorly elongated. dens Thorpe, Mesoreodon chelonyx Scott, M. megalodon
Average length of Pl-M3: 89.0-97.0 mm. sweeti Schultz and Falkenbach, M. wheeleri, Koerner,
Included species: B. auritus (Leidy) only(= Merycoidodon Pseudodesmatochoerus milleri Schultz and Falkenbach, ?
[Anomerycoidodon] iambi Schultz and Falkenbach, = M. intermedius Scott [nomen vanum]) (localities CP85A,
M. [Blickohyus] lynchi Schultz and Falkenbach, Mesore- CP86A, CP99B, NP28, NP34A, ?PN6F); E. occidentalis
 Oreodontoidea 413

minor (Douglass, new rank) ( = E. minor) ( = Desmato- M. montanus Cope, M. temporalis Bettany, Megoreodon
choerus macrosynaphus Riel, ? = Pseudodesmatochoerus fricki Schultz and Falkenbach, M. grandis loomisi [Schlaik-
wascoensis Schultz and Falkenbach) (localities GC7, jer] [= Promerycochoerus loomisi], Promerycochoerus
CC12, CP48, CP50, CP84C, CP85B, CP85C, NP34B, grandis Douglass, P. hollandi Douglass, P. infiatus Thorpe,
NP36A, NP37, NP38A, PNlA, PNIB, PN6D, PN6E, P. latidens Thorpe, P. marshi Thorpe, P. macrostegus
PN16, PNl 7); E. occidentalis latidens (Douglass, new furlongi Schultz and Falkenbach, Superdesmatochoerus
rank) (= Mesoreodon latidens) (? = M. cheeki, Schlaik- microcephalus [Thorpe] [= P. microcephalus], S. lulli,
jer, ?M. hesperus [Stock] [= P. hesperus], M. megalodon [Thorpe] [= P. lulli, ? = P. hatcheri Douglass]) (local-
Peterson, M. scotti Schlaikjer, Desmatochoerus hatcheri ity CP85C); M. chelydra chelydra (? = Promerycochoerus
niobrarensis Schultz and Falkenbach, D. newchicagoensis [Paramerycochoerus] barbouri) (Schultz and Falke-
Schultz and Falkenbach, Hypsiops erythroceps [Stock] [= nbach) (locality PNIB); M. superbus pinensis (Schultz and
Promerycochoerus erythroceps], Merycoides giganteus Falkenbach) ( = Promerycochoerus montanus pinesis) (lo-
Schultz and Falkenbach, P. gregoryi Loomis, P. grinnelli calities CP45B, CP49B, CP50, CP85D, CP86B, CP87 A,
Koerner, P. minor Douglass, P. thorpei Koerner, Pseudo- CP102, NP33B, NP34C, NP35A, NP35B, ?NP36A,
desmatochoerus hoffmani Schultz and Falkenbach) NP36B, NP36D, NP37, PN6E, PN16); M. superbus Leidy,
(localities ?CC9C, CC9D, CC13, CP48, CP49B, CP50, new subspecies (large) (localities CP48, ?CP85C).
CP85D, CP86B, CP99C, CPIOl, CP102, NP34C, NP36B,
NP36D, NP37, PN16); ?E. major new subspecies (locali-
ties CP52, CP103, CP104A, ?CP104B); E. major pygmyus
PHENACOCOELINAE SCHULTZ AND
FALKENBACH, 1950
(Loomis)(= P. pygmyus) (locality CP86B).
Eporeodon sp. ( = "Promerycochoerus," could be Mery- Characteristics: Lacrimal fossa moderately large and deep; post-
cochoerus), is also reported from locality NPlOCC. glenoid process heavy; occiput semi fan shaped; auditory tube cylin-
drical, inflated; dentition brachydont.
Merycochoerus Leidy, 1858 (synonyms: Megoreodon,
Paracotylops, Paramerycochoerus [lapsus calami1 Merycoides Douglass, 1907 (synonym: Paramerychyus)
Promerycochoerus, P. [Parapromerycochoerus1 P. Type species: Merycoides pariogonus (Cope, 1884b)
[Pseudopromerycochoerus 1 Superdesmatochoerus) (= Merychyus pariogonus).
Type species: Merycochoerus proprius Leidy, 1858. Type specimen: AMNH 8113.
Type specimen: ANSP 10,867, 10,868, USNM 156. Characteristics: Smaller than contemporary Hypsiops and
Characteristics: In comparison with contemporary Epore- Phenacocoelus. Compared with other contemporary
odon: •larger size; malar deeper; sagittal crest more promi- phenacocoelines: larger size; rostral premaxillary suture
nent; zygomatic arch more massive, apex higher. Later longer, tympanic bulla more conical, laterally compressed.
forms with extremely long rostral premaxillary suture, re- In later forms: external narial opening slightly retract-
tracted narial opening, very short nasal; broad, shallow fa- ed; frontal slightly inflated; malar moderately deep; post-
cial fossa (distinct lacrimal fossa unrecognizable); highly glenoid process very heavy; tympanic bulla kidney-
arched, inflated braincase; broad sagittal area; fully fan- shaped.
shaped occiput; light, transversely wide postglenoid pro- Average length of Pl-M3: 73. 0-106. 0 mm.
cess; lateral surface of zygomatic arch flattened; hypsodont Included species/subspecies: M. pariogonus pariogonus
molars. (Cope) (= Pseudodesmatochoerus pariogonus [Cope])
Average length of Pl-M3: 106.0-172.0 mm. (? = M. cursor Douglass) (known from localities CC9D,
Included species/subspecies: M. proprius Leidy (known from ?CP47, CP49B, CP50, CP87A, CP102, NP34C, NP37,
localities CA 7, CP105); M. chelydra chelydra Cope PN6E, ?PN6F, PN16); M. pariogonus relictus (Loomis,
(? = Promerycochoerus [Parapromerycochoerus] barbo- new rank)(= Eporeodon relictus, Paramerychyus relictus)
uri Schultz and Falkenbach) (localities ?CP52A, ?CP86C, (=M. nebraskensis Schultz and Falkenbach, M.
PN6F); M. chelydra carrikeri (Peterson, new rank) (= nebraskensis blairi Schultz and Falkenbach, Desmatocho-
Promerycochoerus carrikeri) ( = Promerycochoerus thom- erus [Paradesmatochoerus] sanfordi Schultz and Falken-
soni) (localities ?SB31A, CP52, CP86C, ?CP103); M. chel- bach, Eporeodon thurstoni Stock) (localities CC9D, NB15,
ydra vantasselensis (Peterson, new rank) (= Promeryco- CP48, ?CP85B, CP85C, CP101, NP38B, PNlA), ?M. pari-
choerus vantasselensis) (localities CP51A, CP52, CP103, ogonus, new subspecies (small) (locality ?CP85C); M. har-
CP104A, ?PN6F); M. chelydra, new subspecies (small) risonensis harrisonensis (Peterson) ( = M. harrisonensis)
(localities GC8A, CP51B); M. magnus Loomis (localities (? = Hypsiops luskensis Schultz and Falkenbach, = H.
CP45D, CP53, CP71, CP105); M. .matthewi Loomis (lo- leptoscelos Stevens, ? = Merychyus siouxensis Loomis,
calities CP74A, CP86D, ?CP104B, PN6G, PN6H); M. su- = Phenacocoelus kayi Schultz and Falkenbach) (locali-
perbus superbus (Leidy)(= Promerycochoerus superbus ties GC8B, SB46, SB47, CP51A, CP52, CP54A, CP103,
[Leidy]) (? = M. leidyi Bettany, M. inacrostegus Cope, CP104A, CP105, NP39); M. harrisonensis Peterson, new
 Bruce Lander 414

subspecies (small) (localities CP52, CP86C, CP103); Characteristics: Approximate size of contemporary Oreodon-
M. longiceps (Douglass) (= Mesoreodon longiceps, toides, smaller than other remaining phenacocoelines;
Pseudodesmatochoerus longiceps) (= Phenacocoelus lacrimal fossa broad, shallow, flat bottomed, as in primitive
stouti Schultz and Falkenbach) (localities ?SB46, CP53, Ticholeptinae (see later); tympanic bulla globular as in pri-
CP86D, NP28B, ?PN6G). mitive M erychyus (Ticholeptinae ), to very conical, angular.
Average length of Pl-M3: 64.0-69.0 mm.
Hypsiops Schultz and Falkenbach, 1950 (synonyms: Included species: P. parvus (Thorpe) only(= Oreodontoides
Pseudomesoreodon, Submerycochoerus) [Paroreodon] parvus, Epigenetochoerus parvus) (= P. ma-
Type species: Hypsiops breviceps (Douglass, 1907) (= Ticho- rs hi Thorpe, P. stocki Schultz and Falkenbach), known
leptus breviceps). from locality PN6F only.
Type specimen: CM 1191.
Characteristics: Larger than other contemporary phenaco-
TICHOLEPTINAE SCHULTZ AND FALKENBACH,
coelines; skull comparatively high; external narial opening
1941 (SYNONYMS: MERYCHYINAE)
moderately retracted; tympanic bulla very low, flat, with
crest along anteromedial margin. Characteristics: rostral premaxillary suture long; external narial
Average length of Pl-M3: 97.0-119.0 mm. opening greatly retracted; nasals very short; lacrimal fossa, when
Included species: H. breviceps breviceps (Douglass)(= H. retained, broad, shallow, flat bottomed; occiput fully fan shaped;
johndayensis Schultz and Falkenbach, H. brachymelis pe- lambdoidal crest continuous with, not extending onto, supraoccip-
tersoni [Loomis] [= Ticholeptus petersoni], Pseudome- ital wing (unlike Merycochoerus, Merycochoerinae); postglenoid
soreodon rooneyi Schultz and Falkenbach, Ticholeptus process extremely light, transverse; tympanic bulla sloping dorso-
brachymylis Douglass) (known from localities CPSIA, medially, ventral surface convex; external auditory meatus at end
CP52, CP103, CP104A, NP35C, NP38B, NP39, PN6F); of short inflated tube, extending from extremely inflated, saclike
H. breviceps, new subspecies? (small) (locality CP104B); structure filling entire space between the postglenoid process and
H. bannackensis (Douglass) (= Submerycochoerus ban- the paroccipital process. Dentition hypsodont; P3 anteriorly short-
nackensis, Ticholeptus bannackensis) ( = Pseudomesor- ened, nearly square, with central fossette; anterior intermediate crest
eodon rolli Schultz and Falkenbach, P.? boulderensis on P3 oriented transversely, parallel to anterior cingulum; median
Schultz and Falkenbach) (localities CP54A, CP74A, crest reduced.
NP28B, NP35C).
Merychyus Leidy, 1858 (synonyms: Merychyus
Oreodontoides Thorpe, 1921 [Metoreodon 1 Ustatochoerus)
Type species: Oreodontoides oregonensis Thorpe, 1921. Type species: Merychyus elegans Leidy, 1858.
Type specimen: YPM 12329. Type specimen: ANSP 11,289, USNM 121, 12la, 438.
Characteristics: Smaller than other contemporary phenaco- Characteristics: In comparison with other contemporary
coelines; tympanic bulla globular. ticholeptines: smaller size; rostral premaxillary suture
Average length of Pl-M3: 61.0-73.0 mm. shorter; external narial opening less retracted; nasals longer.
Included species: 0. oregonensis Thorpe only (= 0. curtus Tympanic bulla globular in earliest forms, dorsomedially
[Loomis], [= Merychyus curtus] ? = ?Desmatochoerus very low, convex, sloping steeply ventrolaterally, posteri-
? [Paradesmatochoerus ?] anthonyi Schultz and Falken- orly to form a prominent apex in later forms. Pl-3
bach) (known from localities GC7, CC9E, CPS IA, CP85C, anterior intermediate crest forms an anteroposteriorly elon-
?CP86B, ?CP103, PN6D, PN6E). gated cusp separate from parametacone in later forms;
p2-4 complexly crested.
Phenacocoelus Peterson, 1906 Average length of Pl-M3: 59.0-152.0.
Type species: Phenacocoelus typus Peterson, 1906. Included species/subspecies: M. elegans elegans Leidy (=
Type specimen: CM 1263. M. elegans bluei Schultz and Falkenbach, M. verrucoma-
Characteristics: Smaller than Hypsiops, larger than other con- lus Stevens) (known from localities GC8D, NB4, ?NBS,
temporary phenacocoelines; similar to Oreodontoides ex- CP53, CP88, CP105); M. elegans arenarum (Cope)
cept for development of small frontal vacuity. (= Merychyus arenarum) (= M. arenarum leptorhynchus
Average length of Pl-M3: 78.0-79.0 mm. Cope, M. euryops Cope in Matthew [nomen nudum],
Included species: P. typus Peterson only (= P. monroensis M. delicatus Loomis) (localities ?CC16, CCl7A, ?SBS,
Peterson) (known from localities CP52, CP86C). CP45C, ?CP45D, CP72A, ?CP74A, CP86D, CP104B,
?CP105, ?NP28B, ?PN6G, PN6H); M. elegans? smithi
Paroreodon Thorpe, 1921 (synonym: Epigenetochoerus) (Douglass, new rank) (= M. calimontanus (Dougherty,
Type species: Paroreodon parvus (Thorpe, 1921) (= Eporeo- new rank) (= Ticholeptus calimontanus, T. tooheyi Schultz
don trigonocephalus parvus). and Falkenbach, ? = M. arena rum idahohensis Schultz and
Type specimen: YPM 12425. Falkenbach, M. [Metoreodon] relictus fletcheri Schultz
 Oreodontoidea 415

and Falkenbach) (localities CAI, CCI 7C, CCl9, CC21A, T. obliquidens [Cope] [= Merycochoerus obliquidens,
CC22A, NB6A, NB6C, NB17, NB22A, SB55, CP107, Promerycochoerus obliquidens], Poatrephes paludicola
NP34D, NP38C, NP42, PN18, PNl9A); M. elegansmin- Douglass, ? = Ustatochoerus? schrammi Schultz and
imus Peterson(= M. paniensis Loomis, ? = M. [Meryc- Falkenbach, Ticholeptus rileyi Schultz and Falkenbach)
hyus] jahnsi Whistler, M. calaminthus Jahns) (localities (known from localities GC4E, GC9C, CC17B, NB!6,
?GC3B, ?CC15, NB3C, CP7 l); M. crabilli crabilli Schultz ?NB17, NBl8, NB20A, ?CPlO, CP89, CP106, CPllO,
and Falkenbach, (= M. crabilli ziaensis Schultz and Falken- CPlll, [NPIOE], NP! 1, NP34E, NP41B, NP42, PN7,
bach) (localities ?NB3A, ?SB28, ?SB46, CPSIB, PN8A, PN8B, PN9A, PNl9B).
CP104B); M. crabilli Schultz and Falkenbach, new sub- Ticholeptus sp. is also questionably reported from local-
species (large) (localities CPS IA, CP52, CPI 03, CP104A); ity CP114A.
M. major (Leidy) (= Ticholeptus major, Ustatochoerus
major), (= Merycochoerus [Pronomotherium] califomi- Mediochoerus Schultz and Falkenbach, 1941
cus Merriam, Ustatochoerus califomicus raki Schultz and Type species: Mediochoerus blicki Schultz and Falkenbach,
Falkenbach, U. skinneri santacruzensis Schultz and Falk- 1941.
enbach, U. major texanus Schultz and Falkenbach) (lo- Type specimen: FAM 43172.
calities CC26B, CC30B, CC31, CC32B, NB7B, NB7C, Characteristics: Larger size than contemporary Merychyus
NB7D, NB7E, NB29, SB32D, SB32F, SB32G, SPIA, and Ticholeptus: rostral premaxillary suture longer; nasals
?[CP87B], CP90A, CP90B, CP115A, CP115B, CP116B, shorter. In comparison with Brachycrus: smaller size; ex-
CP123B, PNlO); M. medius medius Leidy (?GC6A ternal narial opening less retracted; nasals longer. Tym-
GC6B, SP2A, CP56, CP74C, CP75C, CP76, CP90A, panic bulla of earlier forms as in later forms of Merychyus.
CP114A, CP114B, CP114C, CPl23A); M. medius novo- Average length of Pl-M3: 89.0-114.0 mm.
mexicanus Frick(= Ustatochoerus medius novomexicanus Included species: M. mohavensis blicki (Schultz and Falken-
[Frick], Ustatochoerus profectus espanolensis Schultz and bach, new rank) (= M. blicki) (known from locality
Falkenbach, U. profectus nevadaensis Schultz and Falken- CPllO); M. mohavensis mohavensis Schultz and Falken-
bach, ? = U. profectus studeri Schultz and Falkenbach, bach (locality NB6E); M. johnsoni Schultz and Falken-
U. skinneri Schultz and Falkenbach, M etoreodon profectus bach (locality CP105); M. new species? (large) (locality
Matthew and Cook [nomen vanum], Metoreodon compres- NB3B).
sidens [Douglass], Merycochoerus compressidens [nomen
vanum], Merycochoerus coenopus Scott [nomen vanum]) Brachycrus Matthew, 1901 (synonym: Pronomotherium)
(localities CCI 70, CC27, CC30A, CC34, NB7 A, NB26B, Type species: Brachycrus rusticus (Leidy, 1870) (= Meryc-
NB27:<\, SB32D, SB32E, SB32F, SB32G, SB33B, SB34A, ochoerus rusticus).
SPIA, CP76, ?CP87B, CP90A, ?CP114C, CP114D, Type specimen: USNM 145.
CP115B, CP116A, CP116B, CP123B, NP41C,PN9B);M. Characteristics: In comparison with other contemporary tic-
relictus Matthew and Cook(= M. [Metoreodon] relictus holeptines: larger size; rostral premaxillary suture longer;
taylori Schultz and Falkenbach, ? = Merychyus smithii nasals shorter; very large deep facial fossa, distinct lacrimal
Douglass) (localities CC22B, NB6C, SB29D, SB32B, fossa unrecognizable; tympanic bulla of earlier forms as in
?CP106, CP108A, CP108B, CP109A, CPl 10, CPll l); later forms of Merychyus, conical and erect in later forms;
M. species indet. a (locality CPI 15C); M. species indet. M3 with small posterior accessory lobe.
b (locality CP115D). Average length of Pl-M3: 103.0-150.0 mm.
Included species/subspecies: B. rusticus (Leidy) (known
Ticholeptus Cope, 1878 (synonym: Poatrephes) from localities CP54B, NP41A); B. laticeps laticeps
Type species: Ticholeptus zygomaticus Cope, 1878. Douglass (= Merycochoerus laticeps, B. laticeps mooki
Type specimen: AMNH 8112. Schultz and Falkenbach, Merycochoerus altiramus Dou-
Characteristics: In comparison with contemporary Mery- glass) (localities CC21 B, NB6C, NB 19 A, NB20A, SB32B,
chyus: larger size; rostral premaxillary suture longer; ex- CP108B, NP34E, NP40A, NP42); B. laticeps buwaldi
ternal narial opening more retracted; nasals shorter; post- (Merriam, new rank) (= Merycochoerus buwaldi (= B.
glenoid process heavier; Pl-3 anterior intermediate crest buwaldi barstowensis Schultz and Falkenbach, B. wilsoni
not forming a cusp. In comparison with other remain- Schultz and Falkenbach, B. wilsoni longensis Schultz and
ing contemporary ticholeptines: smaller size; premaxil- Falkenbach, ? = Merycochoerus madisonius Douglass
lary suture shorter; narial opening less retracted; nasals [nomen vanum], M. elrodi Douglass [nomen vanum]) (lo-
longer; postglenoid process heavier; P 1-3 more rectangu- calities CCl7C, CC22A, NB6C, NBl9A, CP72B,
lar, longer anteroposteriorly; p2-4 complexly crested. CP108A, NP41A, NPSIB, PN19B); B. laticeps buwaldi
Average length of Pl-M3: 87.0-100.0 mm. (Merriam)? (locality NB21); B. laticeps siouense (Sin-
Included species: T. zygomaticus Cope only(= T. hypsodus clair, new rank)(= Pronomotherium siouense) (= B. rus-
Loomis, T. hypsodus leadorensis Schultz and Falkenbach, ticus riograndensis Schultz and Falkenbach, B. vaughani
 Bruce Lander 416

rioosoensis Schultz and Falkenbach) (localities CCI 7D, group has no living descendants or morphologic analogues. Their
CC21B, NB6C, NB6E, ?NB16, SB32B, SB32C, CP72B, brachydont bunoselenodont dentition suggests agriochoerids were
CP74B, CP75B, CPllO, CPl 11); B. sweetwaterensis browsers. Postcranial anatomy (particularly clawed ungual pha-
Schultz and Falkenbach (localities CP54B, CP109A); langes) of the Agriochoerinae suggests they were tree climbers/
B. vaughani Schultz and Falkenbach (localities CP54B, branch walkers (Matthew, 1911; Coombs, 1983; Wilhelm, 1993).
CP107). Agriochoerus has a large manus relative to the forearm length, a
morphological feature resembling rather closely the condition in
present-day tree kangaroos (Wilhelm, 1993). Geologic and fauna!
INDETERMINATE MERYCOIDODONTIDS
data for enclosing strata suggest Agriochoerus preferentially inhab-
Unidentified merycoidodontids have also been recorded from local- ited gallery forests, but also occupied savanna woodlands (Clark,
ities CCl6, NB19C, ?NB16 (UCMP V-6158, Hunt's canyon), SB41, Beerbower, and Kietzke, 1967; Wilson, 1975; Webb, 1977; Bjork
CP72, NP32B, PN6B, as well as from the following areas: Melrose and Macdonald, 1981; Retallack, 1983).
local fauna, Renova Formation, Montana (Chadronian?); Arika- Within the Merycoidodontidae, the Leptaucheniinae are a distinc-
ree? Formation, Yampa Valley, Colorado (Arikareean?); Suwannee tive, early-appearing lineage, differing from other merycoidodontids
River drainage, Florida (early Arikareean); and Shark Tooth Ravine, by their acquisition of highly hypsodont cheek teeth and develop-
Florida (Hemingfordian?). ment of a large facial vacuity and an inflated paroccipital process.
Their postcranial morphology shows similarities to that of present-
day rock-climbing mammals, such as hyraxes (Wilhelm, 1993), and
BIOLOGY AND EVOLUTIONARY PATTERNS the anatomy of their middle ear region is suggestive of adaptations
to open habitats (Joeckel, 1992), which would correlate with a rock-
The middle Eocene homacodontid dichobunoid Pentacemylus (late climbing mode of life. Present-day rock-climbing mammals are
Uintan to early Duchesnean) has a bunoselenodont dentition very usually highly social animals; perhaps the little nasal bosses that
similar to that of oromerycids and the earliest, most primitive ore- characterize the members of this subfamily were employed in some
odonts (well-developed P4 hypocone, Ml-3 mesostyle, and mode of social interaction, assuming that these bosses supported
enlarged metaconule; Ml-3 hypocone reduced/lost), but is more nasal horns, as suggested by Janis (1982). The fact that some, but
derived than the latter group in development of a distinct P3 proto- not all, leptaucheniinine skulls (irrespective of crushing or mode of
cone. Oreodonts and oromerycids probably arose from a Bridgerian preservation) have these nasal bosses is suggestive of possible pat-
Pentacemylus-like homacodontid. (See also Stucky, this volume, tern of sexual dimorphism. Alternatively, these bosses, which occur
Chapter 23). There is no evidence for an Asian oreodont ances- as rugosities along the nasal-maxillary suture, may have served to
try within a dichobunoid stock similar to European Choeropota- strengthen the suture and provide additional support for the nasal
mus (Choeropotamidae) or Cebochoerus (Cebochoeridae), which (for most of its length, the nasal is bordered by a large facial vacu-
are characterized by development of an enlarged caniniform pl ity). Merycoidodontids in general may have been social animals, as
(contra Thorpe, 1937; Webb, 1977). The early Uintan Protore- they have skull modifications that are suggestive of the ability for
odon (new species), currently the earliest, most primitive oreodont wide-gape threat display typical of present-day social ungulates with
known, is most likely the basal stock for the rest of the group, with enlarged canines (or caniniform incisors) such as hippos, peccaries,
merycoidodontids appearing in the late Duchesnean. and hyraxes (Janis, 1983) (see Figure 27. lB).
Agriochoerids underwent a modest radiation from the mid-Uintan Whereas many of the leptaucheniinines were of small body size,
to earliest Chadronian and achieved their greatest diversity during the merycochoerines were more prone to develop large body sizes
the late Duchesnean to earliest Chadronian (early late Eocene) when and to have postcranial morphologies similar to present-day medi-
six genera were extant; only Agriochoerus survived past the early portal ungulates, such as tapirs (Wilhelm, 1993). Several oreodont
Chadronian. However, compared to merycoidodontids, agriocho- lineages show a tendency for increasing the relative leg length over
erids represent a highly homogeneous group with respect to dental time, most predominantly the ticholeptines (Wilhelm, 1993). Tic-
and cranial anatomy. With the exception of pronounced size dif- holeptines were the only oreodonts to survive into the middle and
ferences, contemporary species of these genera exhibit little mor- late Miocene, when the prominent habitats in North America were
phologic diversity. All genera are characterized by retention of changing from woodland to savanna, and many ungulates acquired
primitive dental, cranial, and postcranial characters, most of which longer limbs at this time in correlation with increased cursoriality
are also found in homacodontids and the earliest, most primitive in more open habitats (Webb, 1977). Ticholeptines also were more
merycoidodontids, and by marked parallelism in development of de- hypsodont than most other merycoidodontines, again possibly a re-
rived dental and postcranial characters typical of Agriochoerus, but flection of a more open habitat lifestyle and a more fibrous diet
not seen in merycoidodontids or homacodontids. Molariform pre- (but note that no ticholeptine was as highly hypsodont as the earlier
molars, Agriochoerus-like molars (also observed in anthracotheriid leptaucheniinines, particularly Sespia). Although even the later tic-
suiforms ), and clawed ungual phalanges are derived characters with- holeptines did not have limb proportions resembling those of most
in the Agriochoeridae, individually and independently developed present-day open habitat cursorial ungulates, a favorable compar-
only rarely in other artiodactyl groups. ison can be made with the postcranial morphology of present-day
Agriochoerid adaptations are difficult to interpret because the suids in a variety of habitat types (Wilhelm, 1993 ). Ticholeptines are
 Oreodontoidea 417

EOCENE OLIGOCENE MIOCENE PLIOCENE

r m r m r m r
m m r m

"
?
i-
w w
i-
I~
0

0
I I I I I I I I I I I I I I I I I I I I I I I
-"ti - p ro.oreo
' d on
o 1-1--+1--+1-""1"1--1....-.1 I I I I I I I I I I I I I I I I I
ca8' 1 1
I•New, genus,
A I
, I I I I I I I I I I I I I I I I
:J>
c;') &. I 1 Ne~ gen~s ~ I I I I I I I I I I I I I I I I I
::c 0
5 3. I 1New gen~s ~ I I I I I I I I I I I I I I I I I
0 :;-
:::t: ~ I ~i:_w~s~ I I I I I I I I I I I I I I I I I
0 ·-> ·->
m :J> - I I I I I I
I I I
::c ca [ Agriochoerus
I I I I I I I I I I I I I
6

(~!-)· N~w_g_~_nu_;!i
:J> I I I I I I I I I I I I I I I I
m : :

•
I I I I I I I I I I I I I I I I
New genus F
._ :::J I I I t-i I I I I I I I I I I I I I I I I I I
~ I I I I I I I I I I I I I I I I I I
- I I Bathygenys I I I I I I I I I I I I I I
1
I I I I 1·4 I I I I I I I I I I I I I I
Aclistomycter I
I 1--t - 1-1- I I I I I I I I I I I I I I
·->
I I I I ILeptauchenia I I I I I I I I I I I I I
I I I sespia______
i----.:.... ..._~
I I I I I I I I I I I I I
I I I I I I I I I I I I I I I I I I
Oreonetes
I I I I II I I I I I I I I I I I I I I I I
Prodesmatochoerus
I I I I I I I I I I I I I I I I I I I
I I Eucrotap!!_u!j I I I I I I I I I I I I I
s::
m
·->
I I Blickohyus I
::c I I I I I I I I I I I I I
-< I I I I I I I I I I I I I I I I I I
0 Eporeodon
0 I I 1Merycochoerus I I I I I I I I I I I I I
6
0 I I I I I I I I I I I I I I I I I I
c Merycoides
0 I I I I I
I I I I I I I I I
z 1

1 Hyp~iops
-I
6 I I I I I
I I I I I I I I I
Oreodontoides
:J>
m I I ,----,--------...i....1
,-1 I I I I I I I I
I I Phenacocoelus
1-----. I I I I I I
I I Paro'!!!'don I I I I I I I
-I -
I I I I I I I
c;· I I l l l I . I I I I I I I I
:::r'
Tlcho/eptus
I
0
(i) I I 11 I I I I·~ 11 I I I I
"C Mediochoeus
e. I I I I I I I I I I I I I I I
:::J
Ill Brachycrus
... (I)
... I I I I Il--+--i1...-:-,1
I I I I I I I I
Figure 27.6. Temporal ranges of oreodont genera.
 Bruce Lander 418

rather similar to warthogs, suids that inhabit open habitat savannas periods of geological time. Different size groupings in other ungu-
today, in their postcranial proportions. lates simply might have been assigned to different genera, as was
Another evolutionary trend in merycoidodontids, also occurring originally the case with oreodonts; however, stratigraphic ally super-
in parallel in a number of different lineages, is for the elongation imposed samples might not have been analyzed for size changes, as
of the external nares and the retraction of the nasals. This occurred was also the case with oreodonts.
to a lesser extent within the genera Merychyus (most extreme in Another possibility is that oreodonts were smaller than many
M. major), Mediochoerus (most extreme in M. mohavensis), and other ungulates, and so would have had a relatively shorter life span
Hypsiops (most extreme in H. bannackensis); and to a greater extent and generation time, resulting in a higher reproductive turnover,
within the genera Merycochoerus (most extreme in M. proprius) which, in tum, might have allowed for more rapid changes in adult
and Brachycrus (most extreme in B. laticeps) (see Figure 27.51). body size through successive generations (Lander, 1988). This ef-
The skulls of these latter tax a greatly resemble those of present-day fect would be increased if oreodonts had retained the primitive ar-
tapirs, particularly with regard to the development of a deep facial tiodactyl pattern of giving birth to multiple young per litter, as seen
fossa in Brachycrus, suggesting the presence of a similar type of today in suids, rather than having the pattern typical of more derived
proboscis for selecting vegetation. modem ruminants (tragulids, for example), which have only one or
A final trend to be discussed here is the tendency for parallel size two young per litter.
fluctuations within contemporaneous oreodont lineages. At least The temporal ranges of oreodont genera (Figure 27.6) also show
eight episodes of diminishing size can be observed in oreodont lin- some interesting patterns of radiations and extinctions. The extinc-
eages through the Tertiary (Lander, 1977, 1984; Prothero, 1985). tion of two merycoidodontid and six agriochoerid genera during the
Within each period, a decrease in average adult body size of taxa is Chadronian may reflect the wave of extinctions in archaic groups of
observed, usually with a return to the original size at the end of the North American land mammals that Prothero (1985) related to the
episode. These "dwarfing episodes" can be chronicled as follows: a Terminal Eocene Event. While the agriochoerids predominated in
late middle Chadronian episode (late Eocene) involving the genera the middle Eocene, merycoidodontids were prominent in the begin-
Agriochoerus, Bathygenys, Oreonetes, and Prodesmatochoerus; a ning of the later Eocene (late Duchesnean and Chadronian), with the
late Chadronian and early Orellan (late Eocene to early Oligocene) appearance of the subfamilies Bathygeniinae, Aclistomycteriinae,
episode involving Oreonetes, Prodesmatochoerus, and Leptauche- Leptaucheniinae, and Miniochoerinae. The late Eocene through ear-
nia; an early Arikareean (late Oligocene) episode involving Agri- liest Miocene appears to be the time of greatest oreodont diversity,
ochoerus, Merycochoerus, and Leptauchenia; a latest Arikareean with Agriochoerus surviving until the end of the Oligocene and
to early Hemingfordian (early early Miocene) episode involving all the other merycoidodontid subfamilies appearing during this
Merycochoerus and Merychyus; a late early Hemingfordian (late time (the ticholeptines were last to appear at the beginning of the
early Miocene) episode involving Merycochoerus; an early Barsto- Miocene).
vian (early middle Miocene) episode involving Brachycrus; and a By the close of the early Miocene (late Hemingfordian), the ti-
late Clarendonian to early Hemphillian (early late Miocene) episode choleptines were the only remaining oreodonts. Although they re-
involving Merychyus. mained fairly abundant into the middle Miocene, both taxonomically
Episodes of diminishing body size in oreodonts appear to be cor- and in terms of numbers of individuals, by the late Clarendonian
related with the appearance of caliches (nodular zones) in the fossil (early late Miocene), Merychyus was the only surviving genus. Al-
record of the western United States, the return to a larger size cor- though records of Merychyus persist until the end of the Miocene,
related with the disappearance or reduced prominence of caliches. past the early middle Miocene, oreodonts are known only from rare,
Although caliches are not typical of any particular dimatic zone, fragmentary specimens at a limited number of localities, suggesting
it appears they may be indicative of a seasonal climatic regime of a real decrease in the number of individual animals. However, this
summer droughts (see further discussion in Lander, 1977, 1984 ). Di- apparent decline should not be considered an adaptive failure on the
minishing body size in oreodont lineages may have been in response part of oreodonts, as suggested by Thorpe (1937), who claimed that
to these types of changes in environmental conditions. Indeed, the the group lacked "resourcefulness in periods of stress." Clearly, the
"dwarfing" may reflect a phenotypic response (stunting of growth) final extinction of oreodonts merely echoes that of many other mam-
rather than true genetic dwarfing, arising out of diminished food sup- malian groups that were poorly adapted to the more open habitats
ply during growth of immature individuals because of arid summer which existed at the end of the Miocene (Webb, 1977). However, the
conditions following spring births. Dwarfing or stunting could allow later Miocene fossil record of oreodonts reflects favorably against
a more constant population to be maintained with limited resources that of forest-dwelling tapirs (see Colbert and Schoch, this volume,
(see Kurten, 1965; Lander, 1988). Chapter 39), which of course did survive to the present day. Per-
Why should oreodonts have developed this adaptive response haps it was no more than a simple twist of fate that leads us today
to periods of environmental harshness although this was not as to compare the skull and skeleton of Brachycrus to that of modem
apparent in other ungulates? One explanation may simply relate Tapirus to determine the probable oreodont lifestyle, rather than
to sampling effects: Oreodonts were very common as individual the other way around. Clearly, the oreodonts should be considered
fossils during the Oligocene and early Miocene, and thus more ma- a highly successful group because of their long-term diversity arid
terial is available to illustrate such changes over relatively short abundance during a large proportion of the Tertiary record.
 Oreodontoidea 419

ACKNOWLEDGMENTS (1983). Muscles of the masticatory apparatus in two genera of

hyraces (Procavia and Heterohyrax.) Journal of Morphology, 176,
61-87.
The senior editor would like to thank Donald Prothero and Patri-
Joeckel, R. M. (1992). Comparative anatomy and function of the leptaucheni-
cia Brady Wilhelm for advice about the editing of this chapter. The ine oreodont middle ear. Journal of Vertebrate Paleontology, 12,
author acknowledges the enormous contribution of Christine Janis, 119-30.
especially in the construction of the section on Biology and Evolu- Kurten, B. (1965). The Camivora of the Palestine caves. Acta Zoologica
tionary Patterns. Fennica, 107, 1-24.
Lander, E. B. (1977). A Review ofthe Oreodonta (Mammalia, Artiodactyla),
Parts I, II, and III. Unpublished Ph.D. dissertation. Berkeley: Uni-
versity of California.
REFERENCES (1984). Biological implications of climatically induced size changes
in mammalian adult body size through time. Abstracts with pro-
gram: Society of Economic Paleontologists and Mineralogists An-
Bjork, P. J., & Macdonald, J. R. (1981). Geology and paleontology of the nual Midyear Meeting, p. 47.
Badlands and Pine Ridge area, South Dakota. In Geology of the (1988). Fauna! events in the North American continental Tertiary mam-
Black Hills, South Dakota and tfyoming, ed. F. J. Rich, pp. 211-21. malian herbivore record and their implications regarding the courses
Rapid City, S. Dak.: Geological Society of America Rocky Mountain of extinction and diminishing average adult body size in late Quater-
Section Annual Meeting Fieldtrip Guidebook. nary mammalian herbivores. In Quaternary History of the Manix
Black, C. C. (1978). Paleontology and geology of the Badwater Creek area, Fault, Lake Manix Basin, Mohave Desert, California, eds. S. F.
central Wyoming, Part 14: The artiodactyls. Annals of the Carnegie McGill, B. C. Murray, K. A. Maher, J. H. Lieske, Jr., L. R. Rowan,
Museum, 47, 223-59. & F. Budinger. San Bernadina County Museum Quarterly, 35,
Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedimen- 36-8.
tation, stratigraphy, paleoecology and paleoclimatology in the Big Leidy, J. (1848). On a new fossil genus and species of ruminatoid, Pachy-
Badlands of South Dakota. Fieldiana: Geology Memoirs, 5, 1-158. derrnata. Merycoidodon culbertsoni. Proceedings of the Academy of
CoBabe, E. A. (1996). Leptaucheniinae. In The Terrestrial Eocene- Natural Sciences, Philadelphia, 4, 47-50.
Oligocene Transistion in North America, eds. D.R. Prothero & R. J. (1850a). Observations on two new genera of mammals, Eucrotaphus
Emry, pp. 574-80. Cambridge: Cambridge University Press. jacksoni and Archaeotherium mortoni. Proceedings of the Academy
Coombs, M. (1983). Large mammalian clawed herbivores: a compara- of Natural Sciences, Philadelphia, 5, 90-3.
tive study. Transactions of the American Philosophical Society, 20, (1850b). Descriptions of Rhinoceros nebrascensis, Agriochoerus an-
237-79. tiquus, Palaeotherium proutii, and P. bairdii. Proceedings of the
Cope, E. D. (1878). New artiodactyls of the upper Tertiary. American Nat- Academy of Natural Sciences, Philadelphia, 5, 122.
uralist, 12, 58. (1851). Untitled [remarks on Oreodon priscus and Rhinoceros occiden-
(1884a). The history of the Oreodontidae. American Naturalist, 18, talis ]. Proceedings of the Academy of Natural Sciences, Philadel-
280-2. phia, 5, 276.
(1884b). Synopsis of the species of Oreodontidae. Proceedings of the (1852). Description of the remains of extinct Mammalia and Chelonia
American'Philosophical Society, 21, 503-72. from Nebraska Territory, collected during the geological survey un-
Douglass, E. ( 1901 ). Fossil Mammalia of the White River Beds of Montana. der the direction of Dr. D. D. Owen. In Report of a geological survey
Transactions of the American Philosophical Society, 20, 237-79. of Wisconsin, Iowa, Minnesota, and incidentally a portion of Ne-
(1907). New merycoidodonts from the Miocene of Montana. Bulletin of braska Territory, D. D. Owen, pp. 534-72.
the American Museum ofNatural History, 23, 809-22. (1856). Notices ofremains in extinct Mammalia, discovered by Dr. F. V.
Ferrusquia-Villafranca, I. (1984). A review of the early and middle Tertiary Hayden, in Nebraska Territory. Proceedings of the Academy of
faunas of Mexico. Journal of Vertebrate Paleontology, 4, 187-98. Natural Sciences, Philadelphia, 8, 88-90.
Flower, W. H., & Lydekker, R. (1891). An Introduction to the Study of (1858). Notice of remains of extinct Vertebrata, from the valley of the
Mammals Living and Extinct. London, 763 pp. Niobrara River, collected during the expedition of 1857, under the
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North command of Lieut. G. K. Warren. Proceedings of the Academy of
America. Smithsonian Miscellaneous Collections, 128, 1-96. Natural Sciences, Philadelphia, 10, 20-9.
Gentry, A. W., & Hooker, J. J. (1988). The phylogeny of the Artiodactyla. ( 1869). The extinct mammalian fauna of Dakota and Nebraska, including
In The Phylogeny and Classification ofTetrapods, ed. M. J. Benton, an account of some allied forms from other localities, together with
Systematics Association Special Volume, 35B, pp. 235-72. Oxford: a synopsis of the mammalian remains of North America. Journal of
Clarendon. the Academy of Natural Sciences, Philadelphia, 2, 1-472.
Gill, T. (1872). Arrangement of the families and mammals and synoptical (1870). Untitled (remarks on a collection of fossils from Dalles City,
tables of characters of the subdivisions of mammals. Smithsonian Oregon). Proceedings of the Academy ofNatural Sciences, Philadel-
Miscellaneous Collections, 11, 1-98. phia, 22, 111-13.
Golz, J. D. (1976). Eocene Artiodactyla of southern California. Natu- Loomis, F. G. (1924). The oreodonts of the lower Oligocene. Annals of the
ral History Museum of Los Angeles County Science Bulletin, 26, Carnegie Museum, 15, 369-78.
1-85. Marsh, 0. C. (1873). Notice of new Tertiary mammals. American Journal
Hay, 0. P. (1902). Bibilography and catalogue of the fossil Vertebrata of Science, 3, 407-10, 485-8.
of North America. United States Geological Survey Bulletin, 179, (1875). Notice of new Tertiary mammals, IV. American Journal of Sci-
1-868. ence, 3, 239-50.
( 1930). Second bibliography and catalogue of the Vertebrata of North (1894). Description of Tertiary artiodactyles. American Journal of Sci-
America, Vol. II. Carnegie Institute ofWashington Publications, 390, ence, 3, 259-74.
1-1074. Matthew, W. D. (1901). Fossil mammals of the Tertiary of northwestern
Janis, C. M. (1982). Evolution of horns in ungulates: ecology and paleoe- Colorado. Memoirs of the American Museum of Natural History, 1,
cology. Biological Reviews, 57, 261-318. 353-447.
 Bruce Lander 420

(1910). On the skull of Apternodus and the skeleton of a new artiodactyl. ( 1945). The Mammalia of the Duchesne River Oligocene. Transactions
Bulletin of the American Museum of Natural History, 28, 33--42. of the American Philosophical Society, 34, 209-53.
(1911). A tree climbing ruminant. American Museum Journal, 11, Scott, W. B., & Osborn, H. F. (1887). Preliminary note on the vertebrate
162-3. fossils of the Uinta Formation, collected by the Princeton expedition
Osborn, H. F. (1910). The Age of Mammals in Europe, Asia and North of 1886. Proceedings of the American Philosophical Society, 24,
America. New York: Macmillan. 255-64.
Osborn, H.F., & Wortman, J. L. (1893). Artionyx, a new genus of Ancylopod. Simpson, G. G. (1945). The principles of classification and a classification
Bulletin of the American Museum of Natural History, 5, 1-18. of mammals. Bulletin of the American Museum of Natural History,
Peterson, 0. A. (1906). The Miocene beds of western Nebraska and east- 85, 1-350.
ern Wyoming and their vertebrate faunas. Annals of the Carnegie Stevens, M. S. (1970). Merychyus verrucomalus, a new species of oreodont
Museum, 4, 21-72. (Mammalia, Artiodactyla) from the middle Miocene Runningwater
(1919). Report upon the material discovered in the upper Eocene of the Formation. American Museum Novitates, 2425, 1-11.
Uinta Basin by Earl Douglass in the years 1908-1909, and by 0. A. Stevens, M. S., & Stevens, J. B. (1996). Reevaluation of the taxonomy
Peterson in 1912. Annals of the Carnegie Museum, 12, 40-168. and phylogeny of some oreodonts (Artiodactyla, Merycoidodonti-
(1931). Two new species of agriochoerids. Annals of the Carnegie dae ): Merycoidodontinae and Miniochoeriniae. In The Terrestrial
Museum, 20, 341-54. Eocene-Oligocene Transistion in North America, eds. D.R. Prothero
Prothero, D. R. (1985). North American mammalian diversity and Eocene- & R. J. Emry, pp. 498-573. Cambridge: Cambridge University
Oligocene extinctions. Paleobiology, 11, 389--405. Press.
Retallack, G. J. (1983). Late Eocene and Oligocene paleosols from Badlands Stevens, M. S., Stevens, J.B., & Dawson, M. R. (1969). New early Miocene
National Prak, South Dakota. Geological Society ofAmerica Special Formation and vertebrate local fauna, Big Bend National Park, Brew-
Paper, 193, 1-82. ster County, Texas. Pearce-Sellards Series, Texas Memorial Museum,
Riel, S. J. (1964). A new oreodont from the Cabbage Patch Local Fauna, 15, 1-53.
western Montana. Postilla, 85, 1-10. Stock, C. S. (1930). Oreodonts from the Sespe deposits of South Moun-
Romer, A. S. (1966). Vertebrate Paleontology. Chicago: University of tain, Ventura County, California. Carnegie Institute of Washington
Chicago Press. Publication, 404, 27--41.
Schultz, C. B., & Falkenbach, C. H. (1940). Merycochoerinae, a new sub- (1949). Mammalian fauna from the Titus Canyon Formation, California.
family of oreodonts. Bulletin of the American Museum of Natural Carnegie Institute of Washington Publication, 584, 229--44.
History, 77, 213-306. Thorpe, M. R. (1921). John Day eporeodons, with description ofnew genera
(1941). Ticholeptinae, a new subfamily of oreodonts. Bulletin of the and species. American Journal of Science, ser. 5, 2, 93-119.
American Museum of Natural History, 79, 1-105. ( 1923). The primitive and carnivorelike characters of the Merycoidodon-
(1949). Phenacocoelinae, a new subfamily of oreodonts. Bulletin of the tidae. American Journal of Science, ser. 5, 239--46.
American Museum of Natural History, 95, 1-15. (1937). The Merycoidodontidae, an extinct group of ruminant mammals.
(1950). Phenacocoelinae, a new subfamily of oreodonts. Bulletin of the Memoirs of the Peabody Museum of Natural History, 3, 1--428.
American Museum of Natural History, 95, 87-150. Webb, S. D. (1977). A history of savanna vertebrates in the New World,
(1954) Desmatochoerinae, a new subfamily of oreodonts. Bulletin of the Part I: North America. Annual Review of Ecology and Systematics,
American Museum of Natural History, 105, 143-256. 8, 355-80.
(1956). Miniochoerinae and Oreonetinae, two new subfamilies of ore- Wilhelm, P. B. (1993). Morphometric Analysis of the Limb Skeleton of Gen-
odonts. Bulletin of the American Museum of Natural History, 109, eralized Mammals in Relation to Locomotor Behavior, with Appli-
373--482. cations to Fossil Mammals. Unpublished Ph.D. thesis. Providence:
(1968). The phylogeny of the oreodonts, parts 1 and 2. Bulletin of the Brown University.
American Museum of Natural History, 139, 1--498. Wilson, J. A. (1971). Early Tertiary vertebrate faunas, Vieja Group, Trans-
Scott, W. B. (1889). The Mammalia of the Uinta Formation, Part II: The Pecos, Texas: Agriochoeridae and Merycoidodontidae. Bulletin of
Creodonta, Rodentia, and Artiodactyla. Transactions ofthe American the Texas Memorial Museum, 18, 1-83.
Philosophical Society, 16, 479-504. (1974). Early Tertiary vertebrate faunas, Vieja Group and Buck Hill
(1890). Beitrage zur kenntniss Oreodontidae. Morphologisches Jahr- Group, Trans-Pecos, Texas: Protoceratidae, Camelidae, Hypertrag-
buch, 16, 319-95. ulidae. Bulletin of the Texas Memorial Museum, 23, 1-34.
(1899). The selenodont artiodactyls of the Uinta Eocene. Transac- Wilson, R. W. (1975). The National Geographic Society-South Dakota
tions of the Wagner Free Institute of Science at Philadelphia, 6, School of Mines and Technology Expedition into the Big Badlands of
15-21. South Dakota. National Geographic Research Reports, 1890-1954
(1940). Artiodactyla. In The Mammalian Fauna of the White River Projects, 79-85.
Oligocene, eds. W. B. Scott & G. L. Jepsen. Transactions of the Wortman, J. L. (1895). On the osteology of Agriochoerus. Bulletin of the
American Philosophical Society, 28, 645-733. American Museum of Natural History, 7, 145-78.
 Editors' Appendix: Guide to oreodont synonyms

COLORODON* AGRIOCHOERUS
OLDTAXON NEWTAXON C.ferox A. antiquus? guyotianus
(Previous workers) (This chapter) C. macrocephalus
C. ryderanus
AGRIOCHOERUS
A. antiquus dakotensis A. antiquus antiquus COTYLOPS* INDET (?PRODESMATO-
A. latifrons
CHOERUS ?EUCROTAPHUS)
A. bullatus A. antiquus guyotianus C. speciosa ?P. periculorum periculorum
A. trifons
A. minimus New genus C minimus CYCLOPIDIUS** LEPTAUCHENIA
A. transmontanus New genus D transmontanus C. lullianus L. major major
A. maximus New genus E maximus C. incisivus L. major brevifacies
A. intermedius A. new sp. b, new subsp. b. C. emydinus
C. simus
AGRIOMERYX* AGRIOCHOERUS C. loganensis L. major loganensis
A. migrans A. antiquus? gaudryi C. schucherti L. major schucherti
C. heterodon Sespia nitida
AGRIOTHER!UM* PROTOREODON
A. paradoxicum Protoreodon parvus DAYOHYUS** EUCROTAPHUS
D. trigonocephalus E. trigonocephalus
ART/ONYX* AGRIOCHOERUS D. wortmani
A. gaudryi A. antiquus? gaudryi
DESMATOCHOERUS** EPOREODON
BATHYGENYS D. macrosynaphus E. occidentalis minor
B. alpha hedlundae B. hedlundae D. hatcheri geringensis E. occidentalis major
D. curvidens
BRACHYCRUS D. hatcheri niobrarensis E. occidentalis latidens
B. buwaldi barstowensis B. laticeps buwaldi D. newchicagoensis
B. wilsoni
B. wilsoni longensis D. (PARADESMATO- EPOREODON (primarily)
B. laticeps mooki B. laticeps laticeps CHOERUS)
B. rusticus riograndis B. laticeps siouense D. (P.) grangeri E. occidentalis major
B. siouense D. (P.) wyomingensis
B. vaughani rioosoensis D. (P.) sandfordi Merycoides pariogonus relictus
D. ( P?) anthonyi Oreodontoides oregonensis
BRACHYMERYX* LEPTAUCHENIA
B.feliceps L. major brevif<Jcies DIPLOBUNOPS** AGRIOCHOERUS
D. vanhouteni A. pumilus
CHO ROTHER/UM* PROTOREODON D. leotensis A. matthewi matthewi
C. paradoxicum P. parvus D. uintensis'

421
 Editors' Appendix: Guide to oreodont synonyms 422

HYPSIOPS
OLDTAXON NEWTAXON H. brachymelis H. breviceps breviceps
(Previous workers) (This chapter) H. brachymelis petersoni
H. johndayensis
D. ultimus H. leptoscelos Merycoides harrisonensis
D. crassus A. matthewi crassus harrisonensis
H. luskensis Eporeodon occidentalis latidens
EOMERYX* AGRIOCHOERUS H. erythroceps
E. pumilus A. pumilus
HYPSELOCHOERUS* EPOREODON
EPIGENETOCHOERUS** PAROREODON H. gregoryi E. major latidens
E. parvus P. parvus
LEPTAUCHENIA
EPOREODON L. harveyi L. decora
E. davisi E. occidentalis occidentalis L. martini L. major loganensis
E. paci.ficus L. margeryae L. major major
E. meagherensis E. occidentalis major L. parasimus
E. major cedrensis Eucrotaphus jacksoni jacksoni L. densa L. major brevifacies
E. socialis L.? (Sespia) califomica Sespia nitida
E. condoni E. trigonocephalus L. minora
E. relictus Merycoides pariogonus relictus L. nitida
E. thurstoni
E. trigonocephalus parvus Paroreodon parvus LIMNENETES** LEPTAUCHENIA
L. platyceps L. platyceps
EPOREODON ( PARAEPO- EPOREODON L. anceps Oreonetes anceps anceps
REODON)**
E. (P.) leptacanthus E. occidentalis occidentalis
MEDIOCHOERUS
E. (P.) longifrons
M. blicki M. mohavensis blicki
E. (P) longifrons perbullatus
MEGABATHYGENYS** BATHYGENYS
EUCROTAPHUS
M. goorisi B. alpha
E. auritus Blickohyus auritus
E. dickinsonensis Eucrotaphus jacksoni jacksoni
MEGASESPIA** SESPIA
E. helenae
Eporeodon occidentalis M. middleswarti S. nitida
E. jacksoni leptacanthus
occidentalis
E. jacksoni paci.ficus MEGOREODON** MERYCOCHOERUS
E. major longifrons M.fricki M. superbus superbus
E. montanus Eporeodon occidentalis major M. grandis
M. grandis loomisi
GENETOCHOERUS** PRODESMATOCHOERUS
G. periculorum P. periculorum MERYCHYUS
M. crabilli ziaensis M. crabilli crabilli
GENETOCHOERUS (OSBOR- EUCROTAPHUS M. paniensis M. elegans minimus
NOHYUS)** M. calaminthus
G. (0.) dickinsonensis E.jacksonijacksoni M.jahnsi
G. (0). norbeckensis M. verrucomalus M. elegans elegans
G. (0.) geygani E. jacksoni galushai M. elegans bluei
M. smithi M. elegans? smithi
HADROLEPTAUCHENIA ** LEPTAUCHENIA M. arenarum leptorhynchus M. elegans arenarum
H. primitiva L. decora M. euryops
H. densa L. major brevifacies M. delicatus
H. shanafeltae L. major major M. arenarum idahoensis M. elegans? smithi
H. extrema L. major brevifacies M. calimontanus
H. eiseleyi L. eiseleyi eisleyi M. curtus Oreodontoides oregonensis
M. pariogonus Merycoides pariogonus pariogonus
HYOMERYX* PROTOREODON M. siouxensis M erycoides harrisonensis
H. breviceps P. parvus harrisonensis
M. harrisonensis
HYOPOTAMUS* AGRIOCHOERUS
H. guyotianus A. antiquus guyotianus (Cont.)
 Editors' Appendix: Guide to oreodont synonyms 423

(Cont.)
MESAGRIOCHOERUS* PROTOREODON
OLDTAXON NEWTAXON M.primus P. annectens annectens
(Previous workers) (This chapter)
MES ONYX* (PREOCCUPIED)
MERYCHYUS MERYCHYUS M. dakotensis Agriochoerus antiquus antiquus
(METOREODON)**
M. (M.) relictus fletcheri M. relictus MESOREODON** EPOREODON
M. (M.) relictus taylori M. intermedius E. occidentalis major
M. chelonyx
MERYCOCHOERUS M. wheeleri
M. leidyi M. superbus superbus M. megalodon sweeti
M. macrostegus M. cheeki E. occidentalis latidens
M. montanus ?M. hesperus
M. temporalis M. latidens
M. coenopus Merychyus medius novomexi- M. megalodon
can us M. scotti
M. compressidens M. danai Blickohyus auritus
M. obliquidens Ticholeptus zygomaticus M. longiceps Merycoides longiceps
M.? buwaldi Brachycrus laticeps buwaldi
M. elrodi METOREODON** MERYCHYUS
M. madisonensis M. profectus M. medius novomexicanus
M. altiramus B. laticeps laticeps
M. laticeps
MINIOCHOERUS** OREO NETES
M. rusticus M. battlecreekensis 0. gracilis gracilis
B. rusticus
M. starkensis
MERYCOCHOREUS MERYCHYUS M. cheyennensis 0. gracilis cheyennensis
(PRONOMOTHERIUM)**
M. (P.) californicus M. major MINIOCHOERUS OREONETES
(PARAMINIOCHOERUS)**
MERYCOIDES M. (P.) affinis 0. gracilis gracilis
M. nebraskensis M. pariogonus relictus M. (P.) gracilis
M. nebraskensis blairi M. (P.) nicholsae 0. gracilis nicholsae
M. cursor M. pariogonus pariogonus M. (P.) ottens 0. gracilis cheyannensisi
M. siouxensis M. (P.) helprini 0. gracilis starkensis
M. giganteus Eporeodon occidentalis latidens
OREO DON* INDET. (?PRODESMATOCH-
MERYCOIDODON** INDET (?PRODESMATOCHO- OERUS or EUCROTAPHUS)
ERUS or ?EUCROTAPHUS) 0. culbertsonii ?P. periculorum periculorum
M. macrorhinus P. macrorhinus macrorhinus 0. robustum
M. dunagani P. macrorhinus dunagani 0. culbertsonii periculorum P. periculorum periculorum
M. forsythae P. macrorhinus natronensis 0. priscum
M. culbertsonii ?P. periculorum periculorum 0. macrorhinus P. macrorhinus macrorhinus
M. culbertsonii osborni 0. bullatus Eucrotaphus jacksoni
M. culbertsonii browni P. periculorum lewisi jacksoni
M. lewisi 0. hybridus Blickohyus auritus
M. platycephalus Oreonetes gracilis gracilis 0. major Eporeodon occidentalis major
0. occidentalis E. occidentalis occidentalis
MERYCOIDODON EUCROTAPHUS 0. affinis Oreonetes gracilis gracilis
(ANOMERYCOIDODON)** 0. gracile
0. gracilis coloradoensis
MERYCOIDODON BLICKOHYUS 0. minor
(BLICKOHYUS)**
M. (A.) dani E.jacksonijacksoni OREODONTOIDES
M. (B.) galushai E. jacksoni galushai 0. curtus 0. oregonensis
M. (A.) iambi B. auritus
M. (B.) lynchi OREODONTOIDES PAROREODON
(PAROREODON)**
MERYCOPATER* AGRIOCHOERUS
0. (P.) marshi P. parvus
M. guyotianus A. antiquus guyotianus
0. (P.) stocki
 Editors' Appendix: Guide to oreodont synonyms 424

PITHECISTES** LEPTAUCHENIA
OLDTAXON NEWTAXON P. tanneri L. decora
(Previous workers) (This chapter) P. mariae L. major schucherti
P. altageringensis L. major major
OTIONOHYUS** PRODESMATOCHOERUS P. copei
0. wardi P. periculorum periculorum P. brevifacies L. major brevifacies
0. wardi degrooti P. periculorum lewisi P. decedens
0? vanderpooli P. quadratus

OTIONOHYUS EUCROTAPHUS PLATYOCHOERUS** OREONETES

( OTAROHYUS)** P. hatcreekensis 0. gracilis cheyennensis
0. (0.) bullatus E.jacksonijacksoni P. heartensis 0. gracilis starkensis
0. (0.) cedrensis P. platycephalus 0. gracilis gracilis
0. (0.) hybridus helenae
0. (0.) alexi Eporeodon occidentalis POATREPHES* TICHOLEPTUS
occidentalis P. paludicola T. zygomaticus

PARABATHYGENYS** BATHYGENYS PRODESMATOCHOERUS

P. paralpha B. hedlundae P. meekae P. periculorum periculorum

PARACOTYLOPS* MERYCOCHOERUS PROMERYCOCHOERUS** MERYCOCHOERUS

P. erythroceps Eporeodon occidentalis latidens
PARAMERYCOCHOERUS** MERYCOCHOERUS P. gregoryi
P. grinnelli
PARAMERYCOIDODON** PRODESMATOCHOERUS P. ?hesperus
P. georgei P. periculorum periculorum P. minor E. occidentalis minor
P. pygmyus E. major pygmyus
PARAMERYCOIDODON EUCROTAPHUS P. thorpei E. occidentalis latidens
(BARBOUROCHOERUS)** P. grandis M. superbus superbus
P. (B.) bacai E.jacksonijacksoni P. hollandi
P. (B.) harrisi P. inflatus
P. (B.) major Eporeodon occidentalis major P. latidens
P. leidyi
PARAMERYCOIDODON BLICKOHYUS P. loomisi
(GREGORYOCHOERUS)** P. lulli
P. (G.) wanlessi B. auritus P. marshi
P. (G.) meagherensis E. occidentalis major P. microcephalus
P. montanus
PARAMERYCHYUS** MERYCOIDES P. montanus pinensis M. superbus pinensis
P. harrisonensis M. harrisonensis harrisonensis P. hatcheri M. superbus superbus
P. relictus M. pariogonus relictus P. carrikeri M. chelydra carrikeri
P. thomsoni M. chelydra thomsoni
PARASTENOPSO- OREONETES P. vantasselensis M. chelydra vantasselensis
CHOERUS**
P. conversensis 0. gracilis gracilis PROMERYCOCHOERUS EPOREODON
(DESMATOCHOERUS)**
PAROREODON P. ( D.) curvidens E. occidentalis major
P. marshi P. parvus
P. stocki PROMERYCOCHOERUS MERYCOCHOERUS
( PARAPROMERYCO-
PHENACOCOELUS CHOERUS)**
P. monroensis P. typus P. ( P.) macrostegus M. superbus superbus
P. kayi Merycoides harrisonensis P. ( P.) macrostegus furlongi
P. grandis M. superbus superbus P. (P.) barbouri M. chelydra chelydra
harrisonensis
P. stouti M. longiceps (Cont.)
 Editors' Appendix: Guide to oreodont synonyms 425

(Cont.)
SESPIA
OLDTAXON NEWTAXON S. heterodon S. nitida
(Previous workers) (This chapter) S. marianae
S. ultima
PROMERYCOCHOERUS MERYCOCHOERUS
( PSEUDOPROMERYCO- STENOPSOCHOERUS** OREONETES
CHOERUS)** S. berardae 0. gracilis nicholsae
P. (P.) montanus M. superbussuperbus S. joderensis 0. gracilis starkensis
S. stembergi 0. gracilis gracilis
PROMESOREODON** EPOREODON
P. scanloni E. occidentalis occidentalis STENOPSOCHOERUS OREONETES
( PSEUDOSTENOPSO-
PRONOMOTHERIUM* BRACHYCRUS CHOERUS)**
P. laticeps B. laticeps laticeps S. ( P.) reideri 0. gracilis chadronensis
P. siouense B. laticeps siousense S. (P.) chadronenis
S. (P.) douglasensis 0. gracilis gracilis
PROTAGRIOCHOERUS* PROTOREODON
P. annectens P. annectens annectens SUBDESMATOCHOERUS** EPOREODON (primarily)
S. shannonensis E. occidentalis occidentalis
PROTOREODON S. montanus E. occidentalis major
P. petersoni New genus B petersoni S. socialis dakotensis Eucrotaphus jacksoni jacksoni
P. minor P. parvus
P. medius P. annectens annectens SUBMERYCOCHOERUS** HYPSIOPS
P. paci.ficus P. annectens paci.ficus S. bannackensis H. bannackensis
P. tardus P. annectens tardus
P. pumilus Agriochoerus pumilus SUPERDESMATOCHOERUS** MERYCOCHOERUS
P. pearcei A. matthewi pearcei S. microcephalus M. superbussuperbus
S. lulli
PSEUDOCYCLOP IDIUS** LEPTAUCHENIA
P. frankforteri L. decora TICHOLEPTUS
P. major L. major major T. hypsodus T. zygomaticus
P. quadratus , L. major brevifacies T. hypsodus leadorensis
P. lullianus expiratus T. obliquidens
P. orellanensis L. eiseleyi eiseleyi T. rileyi
T. tooheyi Merychyus elegans? smithi
PSEUDODESMATO- EPOREODON T. brachymelis Hypsiops breviceps breviceps
CHOERUS** T. breviceps
P. milleri E. occidentalis major T. petersoni
P. wascoensis E. occidentalis minor T. bannackensis H. bannackensis
P. hoffmani E. occidentalis latidens
P. longiceps Merycoides longiceps USTATOCHOERUS** MERYCHYUS
P. pariogonus M. pariogonus parigonus U. profectus espanolensis M. medius novomexicanus
U. medius novomexicanus M. medius novomexicanus
PSEUDOGENETOCHOERUS** EUCROTAPHUS U. profectus nevadensis
P. covensis E. trigonocephalus U. profectus studeri
P. condoni U. skinneri
U. califomicus raki M. major
PSEUDOLEPTAUCHENIA** (LAPSUS CALAMI) U. major
LEPTAUCHENIA U. major texanus
U. skinneri santacruzensis
PSEUDOMERYCOCHOERUS** (LAPSUS CALAMI) U. medius M. medius medius
MERYCOCHOERUS U. medius mohavensis Mediochoerus mohavensis
mohavensis
PSEUDOMESOREODON** HYPSIOPS U. ? schrammi Ticholeptus zygomaticus
P. rolli H. bannackensis
P.? boulderensis H. bannackensis Key:* = genus or subgenus previously discontinued.
P. rooneyi H. breviceps breviceps **= genus or subgenus discontinued in this chapter.
28 Oromerycidae
DONALD R. PROTHERO

INTRODUCTION

The Oromerycidae were a middle to late Eocene group oftylopods

close! y related to camels and protoceratids. They originated in North
America in the early Uintan, during the great middle Eocene radia-
tion of artiodactyls, and became extinct during the terminal Eocene
extinction crisis at the end of the Chadronian. Of the six valid gen-
era, four are known from the Uintan, three from the Duchesnean,
and only two genera occur in the Chadronian. Most were small deer-
like animals, the size of musk deer or dik-diks, but the late Eocene
Eotylopus and Montanatylopus reached the size of small antelopes
(Figure 28.1). They had no cranial appendages or locomotory spe-
cializations. Indeed, they are so primitive in skull, dentition, and Figure 28.1. Restoration of Protylopus, by Brian Regal.
skeleton that they have long been confused with primitive camels
and protoceratids. Although Oromeryx was described by Marsh in
the last two premolars are bladelike with a lingual cusp. The mo-
1894, the family was not formally named until 1955 (Gazin, 1955).
lars are bunoselenodont with strong labial cingula and mesostyles.
Only two features help distinguish them from primitive tylopods. In
the upper molars, the protocones are bifurcate posteriorly (Wilson,
DEFINING FEATURES OF THE
1974; Prothero, 1986), which is found elsewhere only in the most
FAMILY OROMERYCIDAE
primitive camels. All oromerycids retain this primitive feature. The
only uniquely derived feature in the dentition is the deep lingual
CRANIAL cleft between the entoconid and hypoconulid on m3.
As already noted, the oromerycids have remarkably few derived
characters that distinguish them from primitive camels and proto-
POST CRANIAL
ceratids. The skull (Figure 28.2A) is like that of most primitive
tylopods, with a short, broad rostrum, incomplete postorbital bar, Postcranially, oromerycids are known primarily from the
and no facial or lacrimal vacuities. Like other tylopods, the bulla is nearly complete skeleton of Eotylopus (Matthew, 1910) and a partial
inflated and filled with cancellous bone. skeleton of Montanatylopus (Prothero, 1986). The vertebrae have
the primitive condition of the vertebral artery passing through trans-
verse processes, rather than through the pedicels of the neural arch,
DENTAL
as in camels and protoceratids. The limb elements are elongate, like
The dentition (Figure 28.2B) is relatively low crowned and bun- those of most cursorial artiodactyls, but not nearly as slender as
odont for a selenodont tylopod. All of the anterior teeth are present, those of camels. The only noticeable specialization is the fusion
and there are no diastemata. Eotylopus has slightly enlarged canines, of the radius and ulna, which happened independently in a num-
but all other oromerycids have equal-sized canines and incisors (if ber of advanced ungulates. The manus has four fully-developed
they are known). The first two premolars are simple blades, and metacarpals, although the central two are more robust. The fibula is

426
 Oromerycidae 427

1. OROMERYCIDAE
B

Figure 28.3. Interrelationships within the Oromerycidae. Characters at the

nodes are as follows: (I) OROMERYCIDAE: Deep lingual cleft separating
metaconid from entoconid on m3. (2) Lose hypocone; weak upper molar
cingula; plicate enamel. (3) Stronger selenes; higher crowned, more rectan-
gular upper molars. (4) Upper molars narrow posteriorly. (5) Much larger
size; lose enamel plications.

Figure 28.2. A. Skull and jaws of Eotylopus reedi (UW 216), a repre-
sentative and relatively completely known oromerycid (scale bar = 5 cm). p3, which does not place it with any group. The affinities of this
B. Crown view of dentition of Eotylopus reedi (scale bar= 5 cm). (Modified taxon are indeterminate.
from Prothero, 1986.)

reduced to a nodule. The lateral metatarsals have been so reduced INCLUDED GENERA IN THE
that the foot is essentially didactyl. FAMILY OROMERYCIDAE

The locality numbers listed for each genus refer to the list of unified
SYSTEMATICS localities in Appendix I. The acronyms for museum collections are
listed in Appendix III.
SUPRAFAMILY The locality numbers may be listed in a couple of alternative ways.
Parentheses around the locality (e.g., [CP101]) mean the taxon in
Although the teeth of Oromeryx were described by Marsh in 1894,
question at that locality is cited as an "aff." or "cf." the taxon in
and the complete skeleton of Eotylopus was described by Matthew
question. Parentheses are usually used for individual species, thus
in 1910, oromerycids remained poorly understood. Some taxa (e.g.,
implying the genus is firmly known from the locality, but the actual
Oromeryx) were considered primitive ruminants; others (e.g., Proty-
species identification may be questionable. Question marks in front
lopus, Eotylopus) were considered ancestral to the camels (Scott,
of the locality (e.g., ?CPlOl) mean the taxon is questionably known
1899, 1940, 1945; Simpson, 1945). The oromerycids were recog-
from that locality, thus implying some doubt that the taxon is actually
nized as distinct from camelids by Matthew in 1910, but were not
present at that locality, either at the genus or the species level.
formally named until Gazin did so in 1955. Subsequent authors
(Wilson, 1974; Golz, 1976; Black, 1978; Prothero, 1986) have up-
Merycobunodon Golz, 1976
held the validity of the group, although Wilson (1974) placed them
as a subfamily of the Camelidae. Type species: Merycobunodon littoralis Golz, 1976.
Type specimen: LACM 27351.
Characteristics: Most primitive of the oromerycids, with teeth
INFRAFAMILY
intermediate between the bunodont and selenodont condi-
Six valid genera of oromerycids are recognized (Fig. 28.3). Their tions. Differs from all other oromerycids in having lower-
temporal and geographic distribution is shown in Figures 28.4 and crowned upper molars with weaker crests, stronger cin-
22.5. Camelodon arapahovius (Granger, .1910) from the Beaver gula, and more squared occlusal outline.
Divide locality in central Wyoming (CP39IIB) is based on a left The m2 is unknown. M2 length= 8.1 mm.
mandible with such a worn dentition that there is no recognizable Included species: M. littoralis only, known from locality CC4
crown pattern. Its only unique feature is a diastema between p2 and only.
 Donald R. Prothero 428

Malaquiferus Gazin, 1955 Average length ofm2: 9.8-11 mm.

Included species: E. reedi (known from localities SB26B,
Type species: Malaquiferus tourteloti Gazin, 1955. SB43D, SB44C, SB44E, CP39B, CP39C, CP39F, CP40A,
Type specimen: USNM 20588. CP41A, CP68B, CP98C); E. n. sp. A (described but not
Characteristics: Malaquiferus also has very low crowned named by Golz, 1976) (locality CC9D); ?E. n. sp. B
teeth, with squared upper molars and small protoconules on
(Tabrum and Fields, 1980) (locality NP24C).
Ml-M3. It is distinguished from most other oromerycids
in having highly wrinkled enamel and lacking lingual up-
Montanatylopus Prothero, 1986
per molar cingula.
Average length of m2: 6.8-7 .8 mm. Type species: Montanatylopus matthewi Prothero, 1986.
Included species: M. tourteloti only (known from localities Type specimen: CM 9918.
SB43A, SB43B, CP29C). Characteristics: M ontanatylopus is readily distinguished from
all other oromerycids in its much larger size and its very
Protylopus Wortman, 1898 hypsodont teeth. Indeed, it is so hypsodont that it was long
misidentified as a camel. However, it clearly shows the bi-
Type species: Protylopus petersoni Wortman, 1898.
furcate protocones and the entoconid-hypoconulid groove
Type specimen: AMNH 2076.
that are diagnostic for oromerycids.
Characteristics: Protylopus is slightly more advanced than
Average length of m2: 16.5 mm.
Merycobunodon andMalaquiferus in having higher crown-
Included species: M. matthewi only (known from localities
ed, more rectangular molars with stronger selenes. Unlike
SB27B, CP83A, NP25C).
more advanced oromerycids, the molars do not narrow pos-
teriorly. As used by Golz (1976, Figure 15), Protylopus is
Indeterminate oromerycids
essentially a wastebasket genus for all oromerycids that
have reached a certain grade of evolution. Not all of the Fragmentary remains ascribed to oromerycids have been reported
species are referable to this genus on the basis of shared from localities CP6B, NP8, NP9A.
derived characters.
Average length of m2: 7.0--10.6 mm.
Included species: P. petersoni ( = P. minor, P. parvus) (known BIOLOGY AND EVOLUTIONARY PATTERNS
from localities CC7C, [CC9AA], CP5A, CP6A, CP6B,
CP38C); P. pearsonensis (localities CC9B, CC9BB); P. Oromerycids were very unspecialized in their skeletons and den-
stocki (localities CC7C, [CC8], CC9B); ?P. robustus (lo- titions, so it is likely they were among the many small, primitive
calities CC7C, CC9AA); ?P. annectens (CP6B). browsing artiodactyls that prowled the forests of the middle and late
Protylopus sp. is also known from localities CC7B, Eocene. In this regard, they must have shared this niche with primi-
?SB43B, ?CP38D. tive protoceratids, oreodonts, agriochoeres, leptomerycids, andhoma-
codonts. By the Oligocene, they had become considerably larger
Oromeryx Marsh, 1894 and more cursorial. Eotylopus and Montanatylopus show signs of
greater cursoriality in having a fused radius-ulna, reduced fibula, and
Type species: Oromeryx plicatus Marsh, 1894.
apes reduced to two digits. However, no oromerycid is as gracile as
Type specimen: YPM 14571
Oligocene camels, protoceratids, leptomerycids, or hypertragulids,
Characteristics: Oromeryx is slightly larger than most Proty-
which were truly cursorial. Montanatylopus did converge consid-
lopus and has posteriorly tapered upper molars. It is con-
erably on camels in its much larger body size and high-crowned
siderably smaller and less hypsodont than Eotylopus or
selenodont teeth.
Montanatylopus.
Oromerycids were always rare, so their distribution is largely an
Average length of m2: 7.9 mm.
artifact of the small sample size. However, there are some distinct
Included species: 0. plicatus only (known from localities
geographic trends. Protylopus was relatively abundant in the middle
SB44B, CP6A, CP6B).
Eocene of California, but rare elsewhere; Malaquiferus was more
common in Wyoming and Texas at the same time; Oromeryx was
Eotylopus Matthew, 1910
always rare. Eotylopus first appeared in the Duchesnean and lasts
Type species: Eotylopus reedi Matthew, 1910. until the very end of the Chadronian. It is by far the most long lived
Type specimen: UW 216. and widespread of oromerycids, ranging from New Mexico and
Characteristics: Eotylopus has slightly higher crowned teeth Texas to Montana and Wyoming. Even so, there are some surpris-
than other oromerycids except Montanatylopus and lacks ing anomalies in its distribution. It is well known from the Chadro-
the enamel plications seen in most other oromerycids. It is nian of Wyoming, Colorado, and Texas, but has so far not been
also larger than any other oromerycid except Montanaty- reported from the large Chadronian samples of the South Dakota
lopus. It differs from Montanatylopus in being smaller and Badlands. Montanatylopus has an unusual distribution: McCarty's
much less hypsodont. Mountain in Montana (Prothero, 1986), the Rubio Peak Formation
 Oromerycidae 429

w
z L L.81.
1.8
w E.81
2.4
()
0 E Lst. Hp.
4.5

:J L.Hp.
5.2
6.0

...
CL. L.E.Hp.
7.0
E.E.Hp.
L L.CI.
9.5

E.CI.

w 11.0 -
z L.L.Ba.
w 12.5
() E.L.Ba.
M
0 14.0

~ E.Ba.

15.8
L.Hm.

17.5
E.Hm.
18.8
E LL.Ar.
19.2

E.L.At.
23.0

w LE.Ar.
z
w L
()
27.7
0
(!J E.E.At.
:J 29.4
0
Wt.
E 31.9
o,.
33.4
L.Ch.
34.5
L M.Ch.

E.Ch.
35.5 -
c:
Ou
37.1
.g
0
39.5 §
gl_
..Q
L.Un.

w 41.3
~
z
w ~
E.Un.
M
() 45.9
0
w L.Br.
46.7

M.Br.
49.0
E. Br.
50.4

L.Wa,

53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0 -
55.5
Ck.3 55,6
w Ck.2 55.8
z Ck.1
Tl.6
56.0
56.2
w L Tl.5
Tl.4
57.0
() 57.3
Tl.3
0 Tl.2
58.5
w
...J
Tl.1
60.9
c( To.3
To.2
CL. 62.5
To.1
E Pu.3
Pu.2 64.5
Pu.1
Pu.a
65.0
65.1
-
Bl.=Blancan, Hp.= Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar.= Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 28.4. Temporal ranges of oromerycid genera.

 Donald R. Prothero 430

in New Mexico (Lucas, 1986), and possibly the Ahearn Member Wyoming, with descriptions of new Eocene mammals. Bulletin of
of the Chadron Formation in South Dakota (Clark, Beerbower, and the American Museum of Natural History, 28, 235-51.
Lucas, S. G. (1986). The first Oligocene mammal from New Mexico. Journal
Kietzke, 1967), but nowhere else. Clearly, oromerycids occupied a
of Paleontology, 60, 1274-6.
habitat that was not always sampled by Chadronian localities, even Marsh, 0. C. (1894). Description of Tertiary artiodactyls. American Journal
when the overall sample size is large. of Science, 48, 259-74.
The diversity trends can be related to the overall picture of fau- Matthew, W. D. (1910). On the skull of Apternodus and a skeleton of a new
nal and climatic change in the Eocene and Oligocene. Primitive artiodactyl. Bulletin of the American Museum ofNatural History, 28,
33--42.
oromerycids, like other primitive selenodont artiodactyls, were fair-
Prothero, D. R. (1986). A new oromerycid (Mammalia, Artiodactyla)
ly abundant in the forested environment of the Uintan and Duch- from the early Oligocene of Montana. Journal of Paleontology, 60,
esnean. All of these groups changed or declined during the major 458-65.
climatic change in the Chadronian (Webb, 1977; Prothero, 1986). (1994). The Eocene-Oligocene Transition: Paradise Lost. New York:
With the beginning of more mixed forest-grassland habitat in the Columbia University Press.
Retallack, G. J. (1983). Late Eocene and Oligocene paleosols from Badlands
Chadronian, Eotylopus was apparently confined to forest remnants.
National Park, South Dakota. Geological Society ofAmerica Special
This may explain its peculiar sampling and distribution. Eotylo- Paper, 193, 1-82.
pus, the last of its family, was one of the more important vic- Scott, W. B. (1899). The selenodont artiodactyls from the Uinta Eocene.
tims of the Chadronian-Orellan extinction event (Prothero, 1994; Transactions of the Wagner Free Institute of Science, Philadelphia,
Berggren and Prothero, 1992), which wiped out a number of archaic, 6, 15-121.
(1940). The mammalian fauna of the White River Oligocene, Part 4:
mostly brachydont forest dwellers. The Chadronian-Orellan extinc-
Artiodactyla. Transactions of the American Philosophical Society,
tion event was undoubtedly a result of major vegetational changes new series, 28, 363-746.
(Wolfe, 1978, 1992; Retallack, 1983) which in turn was triggered by (1945). The Mammalia of the Duchesne River Oligocene. Transactions
a worldwide cooling event caused by a pulse of Antarctic glaciation of the American Philosophical Society, new series, 34, 209-53.
(Berggren & Prothero, 1992; Prothero, 1994). Simpson, G. G. (1945). The principles of classification and a classification of
the mammals. Bulletin of the American Museum of Natural History,
85, 1-350.
Tabrum, A. R., & Fields, R. W. (1980). Revised mammalian fauna! list for
REFERENCES the Pipestone Springs local fauna (Chadronian, early Oligocene),
Jefferson County, Montana. Northwest Geology, 9, 45-51.
Berggren, W. A., & D.R. Prothero. (1992). Eocene-Oligocene climatic and Webb, S. D. ( 1977). A history of the savanna vertebrates of the New World,
biotic evolution: an overview. In Eocene-Oligocene Climatic and Part I: North America. Annual Review of Ecology and Systematics,
Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 1-28. 8, 355-80.
Princeton: Princeton University Press. Wilson, J. A. (1974). Early Tertiary vertebrate faunas, Vieja Group, Trans-
Black, C. C. (1978). Paleontology and geology of the Badwater Creek area, Pecos Texas: Protoceratidae, Camelidae, Hypertragulidae. Bulletin
central Wyoming, Part 14: The artiodactyls. Annals of the Carnegie of the Texas Memorial Museum, 18, 1-83.
Museum, 47, 223--59. Wolfe, J. A. (1978). A paleobotanical interpretation of Tertiary climates in
Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedimen- the Northern Hemisphere. American Scientist, 66, 694-703.
tation, stratigraphy, paleoecology, and paleoclimatology in the Big (1992). Climatic, fioristic, and vegetational changes near the Eocene/
Badlands of South Dakota. Fieldiana Geology Memoir, 5. Oligocene boundary in North America. In Eocene-Oligocene Cli-
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North matic and Biotic Evolution, eds. D.R. Prothero & W. A. Berggren,
America. Smithsonian Miscellaneous Collections, 128, 1-35. pp. 421-36. Princeton: Princeton University Press.
Golz, D. J. (1976). Eocene Artiodactyla of southern California. Natural Wortman, J. L. (1898). The extinct Camelidae of North America and some
History Museum of Los Angeles County, Science Bulletin, 26, 1-85. associated forms. Bulletin of the American Museum of Natural His-
Granger, W. (1910). Tertiary fauna! horizons in the Wind River Basin, tory, 10, 93-142.
29 Protoceratidae

DONALD R. PROTHERO

INTRODUCTION

The Protoceratidae were a middle Eocene to early Pliocene group of

tylopods closely related to camels. They originated in North Amer-
ica in the early Uintan, during the great middle Eocene radiation
of North American artiodactyls, and became extinct in the very lat-
est Hemphillian. Thirteen valid genera are recognized, but typically
only two or three genera were present at any given time. The Proto-
ceratidae were a relatively long-ranging, but rare group throughout
their history in most parts of North America.
Protoceratids began as small, primitive, hornless selenodont ar-
tiodactyls that are difficult to distinguish from contemporary Uintan
oromerycids, oreodonts, and ruminants. Indeed, the oldest protocer-
atids were not removed from the leptomerycids and placed in the
Protoceratidae until J. A. Wilson did so in 1974. Late Oligocene
Protoceras sported enlarged canines and horns or hornlike pro-
tuberances on the maxilla, orbit, and parietals of male individ-
uals. Later protoceratids all had some combination of horns, in-
cluding the bizarre forked occipital "propeller" of Paratoceras and Figure 29.1. Restoration of Synthetoceras, by Brian Regal.
the rostral "slingshots" and orbital horns of the synthetoceratines
(Figure 29.1). From the Whitneyan (mid-Oligocene) onward, pro-
toceratids were very like some deer or antelopes in their skeletons DEFINING FEATURES OF THE
and their range of body sizes (from 20-350 kg, according to Janis, FAMILY PROTOCERATIDAE
1982). However, they had unfused metapodials and relatively short
legs compared to more gracile cursorial artiodactyls. According to
CRANIAL
Janis (1982), most Miocene protoceratids were mooselike folivo-
rous browsers restricted to the more subtropical Gulf Coastal Plain. Although the horns and other cranial protuberances characterize the
The only Miocene High Plains protoceratid, Lambdoceras, was a more advanced members of the family, they do not occur in the most
more deerlike tree browser. Protoceratids disappeared from North primitive members. Aside from the horns, there are relatively few
America in the latest Hemphillian, along with many other character- other features that are typical of protoceratids, and many of these are
istically North American groups. These included dromomerycids, primitive. In the skull, there is a small facial vacuity, and the orbit
mylagaulid and eomyid rodents, most horses, rhinos, and antilo- is completely closed. The nasals are greatly retracted between the
caprids, and many others; altogether, sixty-two genera became ex- slender maxillae (to the level of P3 in Heteromeryx, and to the first
tinct (Webb, 1984). This extinction appart:;ntly corresponds to the molar in Protoceras and higher taxa). The upper incisors are lost in
cooling and loss of subtropical wooded habitat triggered by the the more derived taxa. The diastema between the incisors and the
Messinian crisis in worldwide climate. second premolar becomes increasingly large, and the canine and first

431
 Donald R. Prothero 432

processes of the anterior cervical vertebrae. Through the history of

the group, the limbs attain more cursorial proportions, but never
to the degree seen in camels or pecoran ruminants. Originally, the
radius and ulna were unfused, but they were completely fused in
some synthetoceratines. The lateral metacarpals become somewhat
reduced, but protoceratids always retain a fully four-toed manus.
The astragalus develops a distinct distal keel, and the proximal side
of the sustentacular facet is concave, a feature unique to the group.
The cuboid and navicular remain separate, but the ecto- and meso-
cuneiforms are fused, as in other tylopods. The lateral metatarsals
are more reduced than the lateral metacarpals, but vestiges are still
evident. The metapodials never fuse into a cannon bone in this group.
The keel at the distal end of the metatarsals is confined to the plan-
tar surface and never extends anteriorly onto the volar surface, as in
cursorial pecorans.

B
SYSTEMATICS

SUPRAFAMILY
The debate over the affinities of protoceratids has been one of the
longest and most confused in the mammalian literature (see review
in Patton and Taylor, 1973, pp. 401-2). Protoceras was the first
member of the family to be described (Marsh, 1891 ), and Marsh sug-
Figure 29.2. Skull and dentition of Protoceras ce/er from the Whitneyan gested it might be related to giraffes. Osborn and Wortman (1892),
of South Dakota. A. Left lateral view of male skull and jaws (scale bar = Scott (1895, 1899), Wortman (1898), Matthew (1905),
5 cm). B. Crown views of upper and lower dentition (scale bar = 1 cm). Colbert (1941), Stirton (1944), Simpson (1945), and most authors
(Modified from Patton and Taylor, 1973.)
in the first half of this century tended to place protoceratids with ru-
minants, particularly hypertragulids, or within the Pecora. Loomis
premolar barely fill this gap; they are often reduced ·or lost. The jaw (1925, 1928) placed Protoceras with the camels, and Scott (1940)
flexes downward at the symphysis. Although the jaw has the typical and Stirton (1967) began a modem revival of the idea that pro-
tylopod articular condyle, the coronoid process is shorter than in just toceratids were tylopods. Patton and Taylor (1971, 1973), in their
about any other ungulate, and the angular process is rounded without extensive monographic review of the post-Eocene members of the
the dorsal "hook" found in most tylopods. Like camels, protoceratids family, further supported this idea. They did not include the Eocene
have distinctive winglike basioccipital processes just anterior to the members of the group within the family because Gazin (1955) had
occipital condyles that partially cover the hypoglossal foramina. placed Leptotragulus, Leptoreodon, Poabromylus, and Heteromeryx
within the Leptomerycidae, even though he considered them ances-
tral to protoceratids (Gazin, 1955, p. 14). Ironically, "leptotragu-
DENTAL
lines" had long ago been considered tylopods by Wortman (1898),
Most of the derived features that define the Protoceratidae are found Matthew (1905), and Scott (1899). Wilson (1974), Golz (1976),
in the dentition (Figure 29.2). These include an enlarged upper ca- and Black (1978) formalized the inclusion of the "leptotragulines"
nine in males and caniniform Pl/pl isolated by diastemata. The within the protoceratids, resulting in our modem conception of the
primitively bunoselenodont upper molars are laterally wide and an- group. Webb and Taylor (1980) produced some of the strongest
teroposteriorly short. The paracone and metacone slope strongly evidence for the tylopod affinities of protoceratids.
lingually, with prominent parastyles and mesostyles and very strong
lingual cingulum. In the lower molars, the anterior crest of the hypo-
INFRAFAMILY
conid and the posterior crest of the protoconid are directed lingually.
Protoceratids are among the few Eocene artiodactyls with simple, Three subfamilial and two tribal names proposed for members of
completely selenodont teeth with no bifurcations of the lophs. the Protoceratidae are still in use. The Synthetoceratinae was pro-
posed by Frick (1937) for the taxa above node 9 in Figure 29.3, and
their monophyly has never been in doubt. Webb (1981) divided this
POST CRANIAL
group into the tribes Kyptoceratini (node 10) and Synthetoceratini
In the postcranial skeleton, protoceratid and camel vertebrae share (node 11), and both of these clades are monophyletic. Patton and
a canal that passes through the neural arch rather than the transverse Taylor (1973) used the old name "Protoceratinae" as a taxonomic
 Protoceratidae 433

Figure 29.3. Interrelationships of the

e......"' e
-
Protoceratidae. Characters at the
"':::t
:::t
"'
..::! t. se e ... "'
......e s
~ "'
e...
... nodes are as follows: (1) PROTO-
to()

~
E't.
e ... e ......
E ~
"'
~
...
'5;:: s ...
CERATIDAE: strong upper molar
lingual cingula; short coronoid

ss
E s l
~
..c:.
e ~
e i' ~ '5;:: process on mandible; proximal side
~ ~ ~ l:l., ~ ~ ~ of sustentacular facet of astragalus
concave. (2) p4 talonid closed
posteriorly. (3) Anteriorly projecting
p4 metaconid. (4) Weak columns
in lingual valley of lower molars.
(5) Brachydont teeth with thick
enamel. (6) PROTOCERATINAE
(emended): nasals retracted to level
of molars. (7) Maxillary and orbital
9. SYNTHETOCERATINAE horns in males. (8) Maxillary horn
a triangular laterally compressed
flange. (9) SYNTHETOCERATI-
NAE: Y-shaped rostral horn formed
by fused maxillary flanges in males;
longer muzzle and nasal bones;
long orbital horns in males; small
frontal protuberance including
lacrimal bone in males; reduced
premolars, and more hypsodont
molars. (10) KYPTOCERATINI:
rostral horn fused only at base,
with no stalk; horns circular in
cross section; orbital horns upright
or anteriorly directed. (II) SYN-
THETOCERATINI: rostral horn shaft
fully developed by fusion of maxil-
laries; longer diastemata and reduced anterior premolars. (12) Rostral horn with longer shaft, more anteriorly directed; frontal horns sweep outward
and backward; median pillar at base of crown between protoconid and hypoconid on ml-2.

wastebasket fc;ir the Oligocene taxa and Paratoceras (essentially the question. Parentheses are usually used for individual species, thus
nonsynthetoceratine protoceratids, as defined by Patton and Taylor). implying the genus is firmly known from the locality, but the actual
If this group includes Heteromeryx and Pseudoprotoceras, it does species identification may be questionable. Question marks in front
not appear to be monophyletic. If the Protoceratinae is restricted to of the locality (e.g., ?CP101) mean the taxon is questionably known
Protoceras and Paratoceras, then it is a valid taxon at node 8. Al- from that locality, thus implying some doubt that the taxon is actually
ternatively, the Protoceratinae could include the Synthetoceratinae present at that locality, either at the genus or the species level.
and Paratoceras plus Protoceras at some rank between family and
subfamily (node 7).
The "Leptotragulinae" (Zittel, 1893) has been widely used as a BASAL PROTOCERATIDS
wastebasket name for the Eocene taxa, which have long been placed Leptotragulus Scott and Osborn, 1887
in the Leptomerycidae. There is clearly a monophyletic clade for Type species: Leptotragulus proavus Scott and Osborn, 1887.
most of these taxa at node 3, but unfortunate! y it does not include the Type specimen: PU 11501.
genus Leptotragulus. If the hypothesis in Figure 29.3 is supported, Characteristics: Leptotragulus and Leptoreodon are small,
then another name must be sought for this clade. primitive protoceratids that differ only in features of the
lower premolars. In Leptotragulus, the p4 metaconid is
much weaker, and the anterior crest is more sharply flexed,
INCLUDED GENERA IN THE with a strong parastylid. The same is true to a lesser degree
FAMILY PROTOCERATIDAE in p3. The molars are very similar, and in most features,
both of these genera are very similar to many other Uintan
The locality numbers listed for each genus refer to the list of unified selenodont artiodactyls.
localities in Appendix I. The acronyms for museum collections are Average length ofm2: 6.5-6.8 mm.
listed in Appendix III. Included species: L. proavus (known from localities CP6A,
The locality numbers may be listed in a couple of alternative ways. CP6B); L. medius (localities CP6A, CP6B, CP29C,
Parentheses around the locality (e.g., [CPlOl]) mean the taxon in ?CP29D, [NP25A], [NP25B]); L. clarki (localities CP6A,
question at that locality is cited as an "aff." or "cf." the taxon in CP6B, CP391IB); "L." profectus (probably referable to a
 Donald R. Prothero 434

different genus, such as Trigenicus Douglass, 1903, ac- Characteristics: Toromeryx is larger than other Uintan pro-
cording to Storer, pers. comm.) (localities SB44B, CP39B, toceratids and has very prominent labial cingula. In gen-
CP83C, CP98C, NPlOB, NP24C, NP24D, NP24E, eral, the teeth are low and bulbous, with thick enamel.
NP27C). The ento- and metaconids are bunodont, but the proto- and
Leptotragulus sp. is also known from localities NP9A, hypoconids are selenodont. There are prominent recurved
NP13, CC7, NP22, CP82. lingual sty lids forming closed pockets, and the metaconid
ridge is present on p4.
Average length of m2: 9.0 mm.
Leptoreodon Wortman, 1898 (synonyms: Camelomeryx,
Included species: T. marginensis only (known from localities
Merycodesmus)
SB42B, SB44A).
Type species: Leptoreodon marshi Wortman, 1898.
Type specimen: AMNH 2064. Heteromeryx Matthew, 1905
Characteristics: The differences between Leptoreodon and 'fype species: Heteromeryx dispar Matthew, 1905.
Leptotra-gulus were discussed earlier. Leptoreodon has a Type specimen: AMNH 12326.
large, bulbous metaconid and a broadly flexed anterior Characteristics: Most primitive protoceratids are known only
crest on p4. The talonid basin on p4 also tends to be closed from teeth, jaws, and partial skulls, but Heteromeryx is
posteriorly. Both of these features indicate that Leptore- known from a fairly complete skull with associated limbs
odon is a more derived protoceratid than Leptotragulus, (described by Matthew, 1905, and Scott, 1940). There are
and place it with higher protoceratids to the exclusion of no horns or protuberances, but the narial notch is retracted
Leptotragulus. The genera Camelomeryx and Merycode- to the level of P3, leaving long, delicate nasal bones. The
smus (Scott, 1898) were synonymized with Leptoreodon upper canine is enlarged, and there are large diastemata
by Gazin (1955). anterior and posterior to the reduced Pl. Heteromeryx is
Average length of m2: 6.6--9.5 mm. much larger than earlier Eocene protoceratids and differs
Included species: L. marshi (known from localities CC4, from its contemporary Pseudoprotoceras in having larger,
CC5, SB43A, SB44A, CP6A, NP8); L. major (localities more complex P2-3, a strong protocone on P2, and less
CC4, [CC5], SB43A, SB43B); L. pusillus (localities CC7C, recessed nasals. It is considerably smaller than Protoceras
[CC8], CC9B, [CC9C], SB42B, SB43A, SB43B); L. ed- or Paratoceras and also hornless (unless the type skull was
wardsi (localities CC9A, SB43B); L. stocki (localities from a female individual).
CC9A, CC9AA, CC9B, CC9BB); L. leptolophus (local- Average length of m2: 10.5 mm.
ities CC7C, [CC9A], SB42B, SB44A). Included species: H. dispar only (known from localities
Leptoreodon sp. is also known from localities CC7B, SB44A, SB44B, CP83A, CP98A, CP98C).
CC9A, CC9B, CP6B, CP29C, NP9B, (NP22).
''PROTOCERATINAE''
Poabromylus Peterson, 1931
Pseudoprotoceras Cook, 1934
Type species: Poabromylus kayi Peterson, 1931.
Type species: Pseudoprotoceras longinaris Cook, 1934.
Type specimen: CMNH 11753. Type specimen: AMNH 81000.
Characteristics: Poabromylus is very similar to Pseudopro-
Characteristics: Originally, Pseudoprotoceras was known
toceras in some features, leading Wilson (1974) to syn-
only from a badly crushed partial skull from Chadronia
onymize the two. Emry and Storer (1981) pointed out
Pocket, Dawes County, Nebraska. Wilson (1974) synony-
some important differences that distinguish the two genera.
mized it with Poabromylus, but in a complete review of
Poabromylus is more primitive than Pseudoprotoceras in
the genus, Emry and Storer (1981) revived the name and
having more slender cheek teeth, with p4 nearly as broad
showed its distinctiveness (discussed earlier). They also
as ml and weaker lingual stylids on the lower molars. The
referred two other species to this genus. Pseudoprotoceras
upper molar cingula are much more reduced than they are
has no horns or protuberances on the skull, although the
in Leptoreodon.
only known skulls may be from female individuals. The
Average length ofm2: 8.1-11.5 mm.
nasals are retracted to the level of the first molar, and the up-
Included species: P. kayi (known from localities CP7C,
per canine is greatly enlarged. The cheek teeth are more
CP42A, CP83A); P. robustus (locality NB2); P. minor (lo-
hypsodont than in contemporary genera, without the thick
calities SB25B, SB44B, CP42A, CP83C); P. golzi (locali-
enamel of Heteromeryx. The upper and lower molars are
ties CP29D, [NP25A]).
relatively large and broad compared to the premolars. The
Poabromylus sp. is also known from locality CP29C.
P2 lacks a protocone. The upper molars have no labial
cingulum, but have a lingual column attached to the meta-
Toromeryx Wilson, 1974 conule. The P4 has a strong metaconid directly anterolin-
Type species: Toromeryx marginensis Wilson, 1974. gually, and the lower molars have weak lingual stylids but
Type specimen: TMM 31281-7. no labial cingulids.
 Protoceratidae 435

Average length of m2: 11.2-14.7 mm. SYNTHETOCERATINAE

Included species: P. longinaris (known from localities CP39C,
KYPTOCERATINI
CP39F, CP42A, CP98A, CP98B, CP98C); P. semicinctus
(locality NPlOB); P. taylori (locality CP39F). Syndyoceras Barbour, 1905
Pseudoprotoceras sp. is also known from locality NP9 A. Type species: Syndyoceras cooki Barbour, 1905.
Type specimen: UNSM 1153.
Protoceras Marsh, 1891 Characteristics: Syndyoceras is most easily recognized by
Type species: Protoceras celer Marsh, 1891. its rostral horn, which is formed by a symphysis of the
Type specimen: YPM 11078. maxillaries. Unlike the rostral horn of the Synthetoceratini,
Characteristics: Males of Protoceras can be recognized by it does not fuse to form a long stem or trunk, but branches
their combination of triangular bony protuberances ris- immediately above the base. There are long frontal horns
ing dorsally from the maxillary and supraorbital ridge and that extend over the orbit and the curve dorsomedially but
knoblike protuberances on the parietals (see Figure 29.2). do not sweep backward as in the Synthetoceratini. The
Protoceras has a relatively posteriorly placed orbit in a dentition is relatively low crowned, and the premolars are
skull with strong sagittal and parietal crests. The muzzle relatively large compared to other synthetoceratines. The
is very long, with greatly retracted nasals, and with Pl limbs are relatively robust, and there is little reduction of
midway between the canine and P2. There are three lower the lateral metacarpals.
incisors and an incisiform lower canine, with pl nearly Average length ofm2: 17.5-18.0 mm.
caniniform. The molars are elongate and slender and rela- Included species: S. cooki only (known from localities CP52,
tively low crowned. The robust ulna is fused with the radius CP103, CP104B).
only at the distal end. Most of these features easily distin-
guish both males and females of Protoceras from Parato- Kyptoceras Webb, 1981
ceras, and from Pseudoprotoceras and Heteromeryx. Type species: Kyptoceras amatorum Webb, 1981.
Average length of m2: 11.0-14.2 mm. Type specimen: UF 25711.
Included species: P. celer (known from locality CP84B); P. Characteristics: Kyptoceras, the last of the protoceratids from
skinneri (localities CP84C, CP85C, CP101); P. neatodel- the latest Hemphillian, is actually closely related to Hem-
pha (locality CP103). ingfordian Syndyoceras and not to the Synthetoceratini of
Protoceras sp. is also known from locality NPlOCC. the later Miocene. This is shown by the rostral horn that
Comments: Protoceras was the first genus of the family to be diverges immediately above the base without a trunk, tall
recognized, although the type specimen was from a female upright frontal horns, and the subcircular rather than later-
individual, so it did not have the horns of a male. This ally flattened cross section of the horns. Kyptoceras differs
led to considerable taxonomic confusion and oversplitting, from all other protoceratids in that both the orbital and
reviewed by Patton and Taylor (1973), who recognized rostral horns tilt forward and the cheek teeth are extremely
only one genus and species from the Whitneyan. hypsodont. In addition, there is no exposure of the narial
passage dorsally behind the rostral horn, as in all other
Paratoceras Frick, 1937 protoceratids.
Type species: Paratoceras macadamsi Frick, 1937. Length of m2: unknown. Length of M2: 28.6 mm.
Type specimen: F:AM 32457. Included species: K. amatorum only (known from locality
Characteristics: Paratoceras has many unique features, but GC13B, and possibly locality GC28 [could be Syntheto-
the most bizarre is the forked occipital horn in the male that ceras]).
resembles the propeller on a beanie. It also has small tri-
angular maxillary protuberances and posteriorly recurved
supraorbital horns. The orbit is shifted far forward, but the SYNTHETOCERATINI
sagittal crest is weaker than in Protoceras. There are no Lambdoceras Stirton, 1967
parietal protuberances, and the parietal crests are weak. Type species: Lambdoceras hessei Stirton, 1967.
The muzzle is short with the Pl near the upper canine, and Type specimen: UCMP 32372.
a longer Pl-2 diastema. Only three lower incisiform teeth Characteristics: Like all other Synthetoceratini, Lambdoceras
are present, so either i3 or the lower canine is lost. The has a rostral horn formed by fusion of the maxillaries into
molars are less elongate and more robust than in other pro- a distinct shaft. Compared to others within the group, the
toceratids and are also higher crowned. The ulna is gracile, shaft is relatively long and the fork is shorter. The frontal
and fused most of its length to the radius. horns extend more directly out over the orbit, and the basal
Average length of m2: 11.5-18.2 mm. part is less triangular in cross section than in Prosyntheto-
Included species: P. macadamsi (known from locality SP2A); ceras. The lower premolars are relatively heavier, poste-
P. wardi (locality GC4E). riorly wider and more wedge shaped than in Prosyntheto-
Paratoceras sp. is also known from locality CA 7. ceras. The molars are primitively lower crowned, and the
 Donald R. Prothero 436

lateral metatarsals are not as reduced as they are in Prosyn- than those of Prosynthetoceras. The lateral metapodials are
thetoceras. even further reduced than they are in Prosynthetoceras.
Average length of m2: 19.5-35.0 mm. Average length of m2: 23.0-31.5 mm.
Included species: L. hessei (known from localities CP108B, Included species: S. tricomatus only (known from localities
CP88); L. siouxensis (localities CPllO, CPll 1); L. trini- GC6A, GC6B, GCllB, GC27, SP2A).
tiensis (locality GC4E).
Lambdoceras sp. is also questionably known from lo- INDETERMINATE PROTOCERATIDS
cality CP114B.
Fragmentary remains ascribed to protoceratids also have been re-
Comments: Lambdoceras was the only synthetoceratine to ported from localities CA2, GC7, GC9C, CP114D, NPIOC.
persist in the High Plains after the Hemingfordian.
Patton and Taylor (1971) placed Lambdoceras as a sub-
BIOLOGY AND EVOLUTIONARY PATTERNS
genus of Prosynthetoceras, but their Figure 37 clearly
shows that this makes Prosynthetoceras a horizontal waste-
Protoceratids appeared in the Uintan (middle Eocene) as small, horn-
basket genus for all Barstovian and Clarendonian Syn-
less forms that were very similar to a number of contemporary
thetoceratini. For this reason, subsequent authors (e.g.,
selenodont artiodactyls, such as the protoreodonts, agriochoeres,
Webb, 1981) have treated Lambdoceras as a valid genus,
oromerycids, and hypertragulids. Although their skeletal anatomy
rather than as a subgenus of Prosynthetoceras.
is poorly known, presumably they were browsing herbivores that
lived in the forested environments of the Uintan. When the cli-
Prosynthetoceras Frick, 1937
mate changed in the late Eocene to a drier mix of forest and grass-
Type species: Prosynthetoceras francisi Frick, 1937.
land (Berggren and Prothero, 1992; Prothero, 1994), protoceratids
'fype specimen: TAMU unnumbered.
responded by becoming somewhat more hypsodont. They also be-
Characteristics: Prosynthetoceras has a longer and more an-
gan to show evidence of nasal retraction, implying that they must
teriorly directed shaft on the rostral horn than in more
have had some sort of prehensile lip for browsing.
primitive synthetoceratines. The frontal horns arise poste-
More importantly, they became restricted in distribution in the
rior to the orbit and sweep outward and backward, with the
late Eocene (see Figure 22.5, 29.4). They are among the more com-
basal part of the horn triangular in cross section. The lower
mon taxa in Uintan deposits of California, Utah, Saskatchewan,
premolars are laterally compressed compared to Lambdo-
and Texas, and Poabromylus is one of the few taxa found in most
ceras. In the lower molars, the stylids are reduced, and
Duchesnean localities. But Pseudoprotoceras and Heteromeryx are
there is a small median pillar at the base of the crown
known from only a few specimens in a few Chadronian localities,
between the protoconid and hypoconid on ml-m2. The lat-
and there are still no known Orellan protoceratids. During the later
eral metatarsals are completely reduced to proximal
Oligocene, Protoceras actually occurs in the late Whitney an "Proto-
splints.
ceras channels" in only two places in the Big Badlands (Patton and
Average length of m2: 16.1-23.8 mm.
Taylor, 1973, p. 355). The so-called "Protoceras channels" of Sheep
Included species: P. francisi (known from locality GC4E);
Mountain Table and many other places in the Big Badlands yield
P. texanus (= P. rileyi, P. australis) (1 ocalities GC2, GC3 A,
no Protoceras, and there are none known from any other late Whit-
GC3B, GC4A, GC4D, GC5, GC8D, GC9A, GC9B, NClA).
neyan channel deposit outside South Dakota. Arikareean protocer-
Prosynthetoceras sp. is also known from locality WM13.
atids are similarly scarce, despite the great abundance of Arikareean
Comments: Patton and Taylor (1971) considered Lambdo- deposits. By the Hemingfordian (late early Miocene), only Lambdo-
ceras to be a subgenus of this genus and assigned most ceras persisted in the High Plains; all other Miocene protoceratids
of the former members of the genus to Prosynthetocera.s are known from the subtropical coastal plains of Texas, Alabama,
(Prosynthetoceras). For reasons given earlier, Lambdo- Florida, New Jersey, and Panama. Clearly, from the Oligocene on-
ceras is removed from subgeneric status, obviating the ward, protoceratids were either very scarce and/or occupied habitats
need for subgenera in Prosynthetoceras. that have not been extensively sampled in the fossil record (outside
the Gulf Coastal plain).
Synthetoceras Stirton, 1932 Some clues as to their habitat are given by their anatomy. Ac-
Type species: Synthetoceras tricomatus Stirton, 1932. cording to Janis (1982, p. 286), they had purely folivorous molar
Type specimen: UCMP 31520. wear and a broad, mooselike snout (suggested by their retracted
Characteristics: Synthetoceras has the longest and most ante- nasals), which led Janis to propose that they fed on semiaquatic
riorly inclined rostral horn of any protoceratid. The frontal vegetation with little cellulose. Their relatively short limbs, with
horns are like those of Prosynthetoceras, except that they a persistently four-toed manus and unfused metapodials, suggest
are conspicuously knobbed. The pl-p4 diastema is ex- they were not open plains runners, but adapted for more brushy
tremely long, with pl-p2 lost. The p3 and p4 are propor- terrain. Only Lambdoceras persisted in the High Plains, and it
tionally smaller relative to the molars and have a high me- had a narrower snout, suggesting a tree-browsing habitat (Janis,
dian protoconid. The molars are larger and higher crowned 1982).
 Protoceratidae 437

w
zw
(.)
0
::;
L

E
L.BI.

E.BI

Lst. Hp.
L.Hp.
1.8
2.4

4.5
5.2 r----- ------ ------ ------ -----8 r.. ______-_e _
-=- i.;; (I)

:~:
6.0
Q. L.E.Hp.

iu-
7.0
e.e.Hp.
L L.CI.
8.8
9.5
U) - !::
e -~
E.CI.
w 11.0

zw L.L.Ba. 8
(.)
M
E.L.Ba.
12,5
.s
Q)
0 14.0
:S
Sil E.Ba.
§.
15.8 (I)
L.Hm.

17.5
e
Q;
E.Hm.
18.8
E L.l.Ar.
19.2

E.L.Ar.
23.0

w
zw L.E.Ar.

L
(.)
27.7
0
(!J E.E.Ar.
::;
29.4
0
Wl
(I)
E 31.9
e
~u ..----~-~--..-~-~-~~~-~,
~
Or.
33.4 sr-~~-~-~~~~~~~~~~-~~-1

eg.
L.Ch.
34,5
--l!!
L M.Ch.

E.Ch.
35.5
-e Q,,
37.1
-= -g
~
(I) Q)
~
Du
~ t:
'5ts) :::s .g -'S. ~ ~
e -~[ e
39.5 -ts)- Q)
g g 0

.s -.s- .se _e111 -Ee .:!!e

L.Un.

w 41.3
.Q
z
w
(.)
M E.Un.

45,9
.....2- _..,._
2- .....2- - ~ ~ ~
0
w L.Br.
46.7
Kyptoceratin i Synthetocer atini
M.Br.
49.0 "PROTOCERATINAE" SVNTHETOCERATINAE
e. Br.
50.4

L.Wa.

53,5
E M.Wa,
54.2
e.wa.
e.e.wa. 55.0 -
w
Ck.3
Ck.2
55.5
55.6
55.8
-
zw Ck.1
Tl.6
55.0
56.2
L Tl.5
Tl.4
57.0
(.) 57.3
Tl.3
0 TI.2
58.5
w Tl.1 59.2
...I
c( To.3
Q. To.2
62.5
E To.1
Pu.3
Pu.2 64.5
Pu.1 65.0
Pu.O 55.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
=
Wt. =Whitneyan, Or.= Orellan, Ch. Chadronlan, Du.= Duchesnean, Un. = Uintan, ar. = Bridgerian, Wa. = Wasatchian,
= =
Ck. = Clarkforkian, Ti. Tiffanian, To.= Torrejonian, Pu. Puercan.
Figure 29.4. Temporal ranges of protoceratid genera.
 Donald R. Prothero 438

Because horns were highly sexually dimorphic, there was prob- (1905). Notice of two new genera of mammals from the Oligocene of
ably extensive sexual selection for them, as is well documented in South Dakota. Bulletin of the American Museum of Natural History,
21, 21-26.
living homed ruminants. Apparently, their horns served the func-
Osborn, H.F., & Wortman, J. L. (1892). Characters of Protoceras (Marsh) the
tions of both visual display and intraspecific combat (Webb, 1981). new artiodactyl from the Lower Miocene. Bulletin of the American
The horns of Protoceras are suited only for lateral display, but those Museum of Natural History, 4, 351-72.
of the Synthetoceratini are arranged for frontal display. Ruminants Patton, T. H., & Taylor, B. E. (1971). The Synthetoceratinae (Mammalia, Ty-
show a similar progression from lateral to frontal display. Most lopoda, Protoceratidae). Bulletin ofthe American Museum ofNatural
History, 145, 119-218.
camelids and ruminants use a mixed strategy of biting, butting, pok-
(1973). The Protoceratinae (Mammalia, Tylopoda, Protoceratidae). Bul-
ing, and neck wrestling in intraspecific combat between males, and letin of the American Museum of Natural History, 150, 347--414.
the horns of protoceratids seem well designed for these purposes. Peterson, 0. A. (1931). New species from the Oligocene of the Uinta.Annals
Webb (1981, p. 364) has described how the horns of Kyptoceras of the Carnegie Museum, 21, 61-78.
were well suited for display and neck wrestling and pushing be- Prothero, D. R. (1994). The Eocene-Oligocene Transition: Paradise Lost.
New York: Columbia University Press.
cause their angle of attack would easily force males in combat to
Scott, W. B. (1895). The osteology and relations of Protoceras. Journal of
interlock without seriously gouging or killing their opponent. This Morphology, II, 303-74.
kind of behavior is well documented in modem ruminants. (1898). Preliminary note on the selenodont artiodactyls of the Uinta
formation. Proceedings of the American Philosophical Society, 37,
73-81.
(1899). The selenodont artiodactyls of the Uinta Eocene. Transactions
REFERENCES of the Wagner Free Institute of Science, Philadelphia, 6, 1-121.
(1940). The mammalian fauna of the White River Oligocene, Part 4:
Barbour, E. H. (1905). A new Miocene artiodactyl. Science, 22, 797-8. Artiodactyla. Transactions of the American Philosophical Society,
Berggren, W. A., & D.R. Prothero. (1992). Eocene-Oligocene climatic and 28, 363-746.
biotic evolution: an overview. In Eocene-Oligocene Climatic and Scott, W. B., & Osborn, H. F. ( 1887). Preliminary report on the vertebrate
Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 1-28. fossils of the Uinta formation, collected by the Princeton Expedition
Princeton: Princeton University Press. of 1886. Proceedings of the American Philosophical Society, 24,
Black, C. C. (1978). Paleontology and geology of the Badwater Creek area, 255-64.
central Wyoming, Part 14: The artiodactyls. Annals of the Carnegie Simpson, G. G. (1945). The principles of classification and a classification
Museum, 47, 223-59. of mammals. Bulletin of the American Museum of Natural History,
Colbert, E. H. (1941). The osteology and relationships of Archaeomeryx, an 85, 1-350.
ancestral ruminant. American Museum Novitates, 1135, 1-24. Stirton, R. A. (1932). A new genus of Artiodactyla from the Clarendon
Cook, H.J. (1934). New artiodactyls from the Oligocene and lower Miocene lower Pliocene of Texas. University of California Publications in
of Nebraska. American Midland Naturalist, 15, 148-65. Geological Sciences, 21, 147-68.
Douglass, E. (1903). New vertebrates from the Montana Tertiary. Annals of (1944). Comments on the relationships of the cervoid family
the Carnegie Museum, 2, 145-99. Palaeomerycidae. American Journal of Science, 242, 633-55.
Emry, R. J., & Storer, J. E. (1981). The hornless protoceratid Pseu- (1967). Relationships of the protoceratid artiodactyls, and a description
doprotoceras (Tylopoda: Artiodactyla) in the Early Oligocene of of a new genus. University of California Publications in Geological
Saskatchewan and Wyoming. Journal of Vertebrate Paleontology, Sciences, 72, 30--43.
1, 101-10. Webb, S. D. (1981). Kyptoceras amatorum, new genus and species from
Frick, C. (1937). Horned ruminants of North America. Bulletin of the Amer- the Pliocene of Florida, the last protoceratid artiodactyl. Journal of
ican Museum of Natural History, 59, 1-669. Vertebrate Paleontology, I, 357-65.
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North (1984). Ten million years of mammalian extinctions in North Amer-
America. Smithsonian Miscellaneous Collections, 128, 1-96. ica. In Quaternary Extinctions, a Prehistoric Revolution, eds. P. S.
Golz, D. J. (1976). Eocene Artiodactyla of southern California. Natu- Martin & R. G. Klein, pp. 189-210. Tucson: University of Arizona
ral History Museum of Los Angeles County, Science Bulletin, 26, Press.
1-85. Webb, S. D., & Taylor, B. E. (1980). The phylogeny of hornless ruminants
Janis, C. (1982). Evolution of horns in ungulates: ecology and paleoecology. and a description of the cranium of Archaeomeryx. Bulletin of the
Biological Reviews, 57, 261-318. American Museum of Natural History, 167, 121-57.
Loomis, F. B. (1925). Dentition of artiodactyls. Geological Society ofAmer- Wilson, J. A. (1974). Early Tertiary vertebrate faunas, Vieja Group, Trans-
ica Bulletin, 36, 583-604. Pecos Texas: Protoceratidae, Camelidae, Hypertragulidae. Bulletin
(1928). Phylogeny of the deer. American Journal of Science, 16, 531--42. of the Texas Memorial Museum, 18, 1-83.
Marsh, 0. C. (1891). A horned artiodactyl (Protoceras celer) from the Wortman, J. L. ( 1898). The extinct Camelidae of North America and some
Miocene. American Journal of Science, 41, 81-2. associated forms. Bulletin of the American Museum of Natural His-
Matthew, W. D. (1902). The skull of Hypisodus, the smallest artiodactyl, with tory, 10, 93-142.
a revision of the Hypertragulidae. Bulletin of the American Museum Zittel, K. A. (1893). Handbuch der Palaeontologie. I Abtheilung. Palaeozo-
of Natural History, 16, 311-16. ologie. IV Band. Vertebrata (Mammalia). Munich.
30 Camelidae

J. G. HONEY, J. A. HARRISON, D. R. PROTHERO, and M. S. STEVENS

INTRODUCTION

The family Camelidae, which includes the living camels, llamas, vi-
cunas, alpacas, and guanacos, first appeared in North America in the
middle Eocene (Uintan). They remained an endemic North Amer-
ican group until their late Miocene (late Turolian, = MN13 fauna!
zone) dispersal to Eurasia and probably Africa (Morales, Soria, and
Aguirre, 1980; Moya-Sola and Agusti, 1989; Pickford, Morales, and
Soria, 1993, 1995). In the early Pleistocene (Uquian) they spread
to South America (Webb, 1974; Marshall et al., 1982). Camelids
became extinct in North America during the late Pleistocene when
most of that continent's megafauna vanished. Popularly, camels are
associated with the deserts of Africa and Asia, but the first 36 mil-
lion years or so of their fossil record is confined to North America.
Restorations two fossil North American camelids are presented in
Figure 30.1.
Compared to other herbivores, camelids were relatively rare in the
Eocene. In late Chadronian and Orellan rocks, camelid fossils are
most abundant in the White River deposits of Colorado and south-
eastern Wyoming, but are extremely rare in the Dakotas not many
miles north, possibly suggesting some latitudinal control of their dis-
tribution. During the Miocene they lived over much of North Amer-
ica and are sometimes the most common large herbivore in a fossil
fauna. The main radiation of the camelids occurred in the Miocene,
starting in the Arikareean Land Mammal age. Their generic diversity
was greatest during the Hemingfordian and Barstovian, with a min-
imum of thirteen genera present; during the late Barstovian, at least
twenty species were present. Generic diversity decreased during the
late Miocene and Pliocene, although they were still relatively com-
mon animals. The last North American camelids were Camelops
hesternus, Hemiauchenia macrocephala, and Palaeolama miri.fica,
which disappeared about 11,000 years ago (Kurten and Anderson, Figure 30.1. Restorations of fossil camelids. A. Oxydactylus (by Janet
1980). Brown and Brian Regal). B. Floridatragulus (by Margaret Stevens).

439
 J. G. Honey et al. 440

DEFINING FEATURES OF THE

FAMILY CAMELIDAE

CRANIAL
Cranial armament (horns, antlers, ossicones) is lacking (Fig. 30.2).
A sagittal crest is present. The rostrum is short to extremely elon-
gate. A lacrimal vacuity is present. A maxillary fossa is usually
present and is deeply pocketed in some genera, secondarily re-
duced in later forms. Primitively, the postorbital bar is incomplete
but, except for floridatragulines, is probably complete by the early
Miocene. The mastoid portion of the periotic is exposed on the
posterodorsal surface of the skull between the exoccipital and squa-
mosal. The tympanic bulla is filled with cancellous bone. The bulla
consists of ventrally extended medial and lateral vertical plates that c
project well below the level of the basioccipital; the deep tympa-
nohyal recess is located anteriorly between the junction of the two
plates. A small to large postglenoid process is present. A distinct Figure 30.2. Protolabis heterodontus, USGS D697, from the Troublesome
angular process ("hook") is present on the mandible and may be Formation. A. Palatal view of upper dentition. B. Side view. C. Occlusal
strongly inflected mesially. The mandibular symphysis is solidly view of c, pl-4, ml-3. Scale bar= 5 cm. (Illustration by Marge C. Leggitt.)
fused.
co-ossified. The fibula is reduced to a small spine that is fused to the
proximal part of the tibia, and the distal malleolus articulates with
DENTAL
the tibia, astragalus, and calcaneum. In the carpus the trapezoid and
Il-2 reduced or absent, except in stenomylines and Miolabis, where magnum are unfused and the trapezium is usually present, although
they are large; I3 is present and usually caniniform. The lower sometimes absent in Recent forms. In the tarsus the ectocuneiform is
incisors are spatulate. The upper and lower canines are always united with the mesocuneiform, but the navicular and cuboid remain
present and usually caniniform; exceptions include some stenomy- unfused; a distal keel is present on the astragalus. Only metapodials
lines where the lower canine is incisiform and Michenia where the III and IV are functional; II and V are extremely reduced or absent
upper canine is incisiform. The Pl/pl are retained or lost; when re- (except possibly in floridatragulids where an elongate metatarsal V
tained they become caniniform in derived genera. The P2/p2 and p3 splint has been identified in Aguascalientia). Metapodials III and IV
are variably lost in derived genera. The p2--4 are relatively narrow range from completely unfused (primitive) to solidly fused except
transversely and usually simple, with an anteromedially inflected at the divergent distal ends (derived), and from extremely short to
paraconid and a posterolingual ridge off the protoconid; usually on very elongate. Metatarsals III and IV have flattened dorsal surfaces.
the p4 and sometimes on p3 the posterolingual ridge and hypoconid The metapodial keels are confined to the posterior surfaces of the
enclose a fossettid. Cheek teeth range from brachydont to hyp- distal trochlea. The fibular facet on the calcaneum has a proximal
sodont. The upper molars are transversely compressed with straight convexity and a dorsal concavity. Primitively, the stance was un-
ectolophs and with fossettes that are closed anteriorly and posteri- guligrade; later genera became digitigrade. Proportions range from
orly after moderate wear. The upper molars are four cusped, without sheep or gazelle to giraffe height.
paraconule and hypocone; the protocone is primitively bifurcated in
some Poebrotherium. The mesostyle is usually present on the upper
molars but may be secondarily lost. Labial ribs on the upper mo- SYSTEMATICS
lars are strong to subdued. The lower molars are without a lingual
notch between the metaconid and entoconid. Diastemata progres-
SUPRAFAMILY
sively develop in front of and behind Pl/p 1 in most groups. The
skull and dentition of a common Miocene North American came lid, Leidy (1847) described Poebrotherium wilsoni, the first Tertiary
Protolabis, is illustrated in Figure 30.2. camelid discovered in North America. He considered Poebroth-
erium a genus ofRuminantia (which at that time included the Camel-
idae) intermediate between Dorcatherium and the "pachyderm"
POST CRANIAL
Anoplotherium. By 1869, however, Leidy had recognized the true
The neck is elongate to giraffelike (Aepycamelus). Neural spines affinities of Poebrotherium and included it in the Camelidae; in
in the anterior cervical vertebrae are low and long. In cervicals 2 the intervening years Camelops and Procamelus were described
through 6, the vertebral artery passes within the pedicle of the neu- and recognized as camelids by Leidy (1854, 1858). Floridatragulus
ral arch in the anterior half of each vertebra, becoming confluent and Nothokemas were originally considered hypertragulids (White,
with the neural canal in the posterior half. The radius and ulna are 1940, 1947).
 Camelidae 441

The Old World camels and the South American lamines are the species later placed in Miolabis, and Camelidae to include members
only living representatives of the suborder Tylopoda, which is in- of the current Camelinae. Hay, on the other hand, included within
cluded with its sister group Ruminantia in the Neoselenodontia a broader family Camelidae not only unquestioned true camels,
(Webb and Taylor, 1980). In addition to numerous extinct camelids, but also Leptotragulus, Hypertragalus, and the oromerycid Proty-
we also include the oromerycids, protoceratids, and xiphodonts in lopus; later (1930), however, Hay included the first two genera in
the Tylopoda. Oromerycids were early thought to be related to the Hypertragulidae. Matthew (1904) subdivided the camels into
camelids, based on similarities seen with Poebrotherium in the skull three "series," groups of related animals that included only currently
(particularly the auditory bullae ), the dentition, vertebrae, and limbs recognized true camels. With some modifications including the ad-
(with unfused trapezoid and magnum, navicular and cuboid, and the dition of subsequently described genera, Matthew's series were in-
palmar restriction of the metapodial keels). Wortman (1898) con- corporated into Simpson's 1945 classification of the Camelidae,
sidered Protylopus directly ancestral to Poebrotherium. Matthew which contained five subfamilies: Poebrotheriinae (which included
(1910) thought Eotylopus and Protylopus were hypertragulids bro- some oromerycids), Camelinae, Pseudolabidinae, Alticamelinae,
adly ancestral to the Tylopoda, although not directly ancestral to and Stenomylinae.
Poebrotherium. In his 1934 phylogenetic chart of the Artiodactyla, Gazin ( 1955) established the family Oromerycidae for Oromeryx,
he showed them as basal camelids. Gazin (1955) and Golz (1976), and included Protylopus, Camelodon, and Eotylopus, genera for-
however, viewed oromerycids and camelids as originating sepa- merly included within the Camelidae; Wilson (1974), however, re-
rately from dichobunid artiodactyls, and they considered some sim- garded them as oromerycine camelids because of similarities be-
ilarities between the two families as convergent. Most authors have tween the molars of Chadronian Poebrotherium and Protylopus. His
included some or all oromerycids within the Tylopoda, either as concept of Oromerycinae was accepted by Honey and Taylor ( 1978)
members of the family Camelidae (Scott, 1940; 1945; Simpson, and by Webb and Taylor (1980), who suggested that the enlarged
1945; Wilson, 1974; Webb and Taylor, 1980) or as forming a dis- I3 may be a shared-derived feature with camels. Wilson (1974) also
tinct family of their own, Oromerycidae (Stirton, 1967; Patton and established the subfamily Poebrodoninae for what he considered to
Taylor, 1973; Black, 1978; Prothero, 1986). be a divergent line of early camelids not ancestral to Poebrotherium.
The Protoceratidae were long considered to be related to the Hy- The formal subfamilial divisions of the Camelidae (Fig. 30.3)
pertragulidae (Matthew, 1934; Scott, 1940; Simpson, 1945). Stirton adopted herein include Stenomylinae, Floridatragulinae, Protolabi-
(1967) thought they were closely related to the Camelidae, perhaps nae, Miolabinae, and Camelinae; the Poebrotheriinae and Alticame-
with a common ancestor in the Uintan Leptotragulus; this hypoth- linae are abandoned as paraphyletic. A cladogram showing our
esized relationship was based on the retention of an unfused navic- view of the interrelationships of camelid genera is presented in
ular and cuboid, a primitive feature. Patton and Taylor (1973) also Figure 30.3.
listed a number of shared primitive limb features between proto-
ceratids and camelids in support of placement of the Protoceratidae
within the Tylopoda and exclusion from the Ruminantia. Webb and
Taylor (1980) listed the passage of the vertebral artery through the INCLUDED NORTH AMERICAN GENERA IN THE
neural arch pedicles of the cervical vertebrae as a derived feature FAMILY CAMELIDAE
linking the Protoceratidae and Camelidae; camelids differ, how-
ever, in that the vertebral arterial canal becomes confluent with the The locality numbers listed for each genus refer to the list of unified
neural canal in the posterior half of cervicals 3 through 6. Black localities in Appendix I. Question marks in front of the locality (e.g.,
(1978) presented a contrary view, based almost entirely on analysis ?CP101) mean the taxon is questionably known from that locality.
of dental characters, and allied the protoceratids with leptomerycids The acronyms for museum collections are listed in Appendix III.
as the sister taxon to the combined agriochoerid-merycoidodontid
group.
Xiphodonts have been classified in the Tylopoda by some authors PRIMITIVE CAMELIDS
(Matthew, 1929; Simpson, 1945; Webb and Taylor, 1980). Similar-
Poebrodon Gazin, 1955
ities include elongate, slender, dorsally flattened metapodials III
and IV, with metapodials II and V extremely reduced; incomplete Type species: Poebrodon kayi Gazin, 1955.
fibula; shape of mandibular condyle and basioccipital process; long, Type specimen: USNM 20393.
laterally compressed premolars; and inflated auditory bulla. Characteristics: Presently known only from a few teeth. The
teeth are characteristically camelid, having straight ec-
tolophs and fossettes closed anteriorly and posteriorly; in
INFRAFAMILY
addition, the teeth are more hypsodont than those of con-
Two early approaches to the classification of North American temporary selenodont artiodactyls.
camelids were presented by Cope (1886) and Hay (1902). Cope, on Average length of m2: 7.0 mm.
the one hand, divided genera now included within the Camelidae into Included species: P. kayi (known from locality CP6B); P.
three distinct families: Poebrotheriidae to include Poebrotherium californicus (locality CC7C).
and "Gomphotherium," Protolabididae to include Protolabis and a Poebrodon sp. is also known from locality CP38C.
J. G. Honey et al. 442

~
~;i i;i
j
;::

ll I"' o
~
E;
~\
I
I

Ii \
ii! \
ij !1l

~
~-~r--~~""v-~~~~~-c~~~~~~~-=",,-~~~~~~~~~~~~~~~-====~~
i
li J~s
~
t§ ~
~ ~-~'-..p.~~~~-~-+I
M
 Camelidae 443

Figure 30.3. Interrelationships of camelid genera (metatarsal bones are depicted above the skulls). Key to characters at nodes: (1) Divergent distal
metapodials; deep tympanohyal groove; transversely narrow, high crowned, selenodont molars; straight ectoloph on Ml-3; metacarpals II and V reduced
to nodules; vertebral arterial canal confluent with neural arch in posterior half of cervicals 3-6. (2) Increased hypsodonty; larger size; M2 wider posteriorly;
lingual hypoconulid lobe of m3 reduced; lingual metaconid and entoconid flattened. (3) Reduced inner lobe ofbulla; reduced premolars. (4) Stronger sagittal
crest; smaller lacrimal vacuity; posterior extension of internal portion of bulla. (5) STENOMYLINAE: Premaxillary extended posteriorly; longer rostrum;
deep, elongate maxillary fossa; elongated upper molars, which are laterally compressed and more hypsodont; weak mesostyles; partially fused metapodials;
closed orbit. (6) Highly reduced P2-3. (7) Incisiform lower canine; il-3, c, and pl form closed row; p2-3 diastema; molars extremely hypsodont. (8)
Larger incisors; P2/p2 reduced; p3 reduced; P4 reduced; molars more hypsodont; metapodials completely fused. (9) M3/m3 extremely elongated. (10)
Deeper anteromaxillary fossa; Pl-2 lost, p2 lost; p4 very small; metapodials very short. (11) Lower molars with metastylid present, formed by overlap
of metaconid on entoconid; heavier premaxilla; large postglenoid foramen. (12) FLORIDATRAGULINAE: Il-3 lateral on premaxillary and bladelike; l3
size similar to I2; p4 equals anteroposterior diameter of p3. (13) Rostrum very elongate; p4 shorter anteroposteriorly than p3; m3 with incipient lingual
selene. (14) Posterior cleft on m3 hypoconulid; larger size. (15) p2-3 diastema; m3 hypoconulid bilophed, with medial cleft; largest size. (16) Lower
incisors more spatulate; Pl shortened; P3 internal cingulum stronger; orbits closed(?). (17) "NOTIIOKEMATINAE": Mandibular angle enlarged; Pl roots
closely appressed. (18) C larger than 13; loss of pl; p3 posterolingual stylid strong. (19) Medial plate of auditory bulla transversely compressed; rostrum
lengthened; P2-3 variably reduced; molars more hypsodont; upper molars with weaker ribs. (20) MIOLABINAE: P2-3 parastyle reduced; mandibular
angle enlarged; coronoid process short; cervical vertebrae short; metapodials very short. (21) Pl/pl absent; P2/p2 highly reduced or absent; lower molars
with anterior buttresses; elongate, slender ramus; post-cl diastema long and with sharp dorsal crest; rostrum inflated along nasal passage; auditory bulla
with anteromedial process. (22) Maxillary fossa deepened; lateral plate of auditory bulla enlarged. (23) 12 with incipient posterior cusp; Pl reduced. (24)
Skull elongate; dorsoventrally narrow rostrum. (25) Il-2 cupped; pl absent. (26) Prominent anterior buttresses on molars; metatarsals weakly fused. (27)
Lateral plate of auditory bulla flattened; bulla medially excavated. (28) p2 absent; p3-4 three lobed. (29) PROTOLABINAE: Rostrum narrow; laterally
expanded anterior nares. (30) Absence of elongate basioccipital tuberosities; P2 without strong, continuous lingual cingulum; auditory bulla less inflated
with medial plates more compressed; buccinator fossa moderate to strong. (31) Molars hypsodont; M3/m3 anteroposteriorly elongate; metastylids on lower
molars very weak to absent; ventrally produced mandibular angle with weak to strong lateral flare; metapodials fused. (32) Braincase short; 13-C weak, c
small; symphysis shallow; inflection of mandibular angle suppressed. (33) Cervical vertebrae elongate; metapodials elongate, slender, and greater than basal
length of skull; metatarsus and metacarpus approximately equal in length. (34) Mandibular ramus relatively narrow, with convex lower border. (35) Lower
premolars and molars laterally compressed. (36) CAMELINAE: Pl/pl caniniform; metapodials fused; metacarpus longer than metatarsus. (37) Maxillary
fossa pocketed; increased hypsodonty. (38) Upper canines small, peglike; pl-2 shortened, high crowned. (39) LAMINI: Arched nasals. (40) Extremely
elongate metapodials, neck. (41) Il-2 lost; P2/p2 lost; anteroexternal style present on lower molars. (42) Small Pl/pl; p3 small or absent; reduced lacrimal
vacuity; shortened rostrum. (43) Extremely elongate metapodials, neck. (44) Pl/pl lost; reduced maxillary fossa; moderate to strong anteroextemal style
on lower molars; metapodials secondarily shortened. ( 45) p3 absent; metacarpal length subequal to metatarsal length; strong anteroexternal style on lower
molars; greatly reduced lacrimal vacuity; extremely retracted nasals; greatly reduced p4. (46) Callosities on the inner foreleg. (47) Ever-growing lower
incisors. (48) Moderately to very hypsodont molars; cheek teeth narrow in relation to length; metapodials secondarily shortened. (49) Cheek teeth very
hypsodont; Pl/pl absent; p3 absent; dorsal surface of mandibular condyle transversely concave; suspensory ligament scar extends to center of proximal
phalanx and has a raised center. (50) CAMELINI: II lost; Pl/pl single rooted; prominent sagittal and lambdoidal crests. (51) Angular process on mandible
greatly enlarged; long postglenoid process on skull with correspondingly large facet on mandibular condyle; ventrally flattened auditory bulla; reduced
maxillary fossa; thickened, very heavy premaxilla; diastemal crest on mandible low and rounded; Cl/cl enlarged and rounded in cross section, especially
in males; strong labial styles and ribs on Ml-3. (52) Reduced pl; reduced p3. (53) Reduced P3; cheek teeth higher crowned than in Titanotylopus. (54) pl
absent; p3 more reduced than in Megatylopus; larger body size than Megatylopus. (55) Metapodials shorter in relation to basal length of the skull; cheek teeth
more hypsodont than in Megatylopus or Titanotylopus. (56) Spatulate lower incisors; splayed lower canines. (57) I3 enlarged and caniniform. (58) Short,
blunt chin with a shortened ramal symphysis; larger body size than Megacamelus; lower incisors arrayed almost transversely; I3 absent or vestigial. (59)
Reduced paroccipital process; metapodials subequal in length and shorter than the basal length of the skull; maxillary fossa reduced or absent; zygomatic
arch straight in lateral view; retracted nasals; center of suspensory ligament scar raised. (Illustration by Marge C. Leggitt.)

Poebrotherium Leidy, 1847 Poebrotherium sp. is also known from localities CP39B,
CP39C, CP39F, CP39G, CP391IC, CP43, CP45A, CP84A,
Type species: Poebrotherium wilsoni Leidy, 1847.
NP24C, NP24D, NP24E, NP34B.
Type specimen: ANSP 11012.
Characteristics: Complete dental formula. Poebrotherium, the
Paratylopus Matthew, 1904
common Oligocene camelid, is smaller, less hypsodont,
and more primitive than any other except Poebrodon; there Type species: Paratylopus primaevus (Matthew, 1904)
are no unique derived features. Differs from Poebrodon in (= Miolabis [Paratylopus] primaevus).
its larger size, greater hypsodonty, posteriorly wider M2, Type specimen: AMNH 6520.
reduced lingual hypoconulid lobe of m3, and flattened lin- Characteristics: Complete dental formula. Small to medium-
gual surface of metaconid and entoconid. sized camelids with relatively brachydont teeth, reduced
Average length of m2: 11.0-13.5 mm. premolars, and large skulls relative to their tooth size. Dis-
Included species: P. wilsoni (known from localities CP40A, tinguished from Poebrotherium by larger size and more
CP40B, CP41A, CP41B, CP68C, CP84A, CP84B, CP99A, reduced premolars; from Paralabis by smaller size and
CP99B, NP50B); P. chadronense (locality SB44D); P. ex- less reduced premolars; and from all other Whitneyan and
imium (localities CP40A, CP41A, CP68B, CP83C, CP84A, Arikareean camelids by its brachydont teeth, which are
CP98B, CP98C, NP29C). relatively small for its skull size.
 J. G. Honey et al. 444

Average length of m2: 12.5-16.4 mm. Miotylopus Schlaikjer, 1935 (synonyms: Dyseotylopus,
Included species: P. primaevus (known from localities Gentilicamelus, in part)
CP41C, CP84B, CP99B); P. labiatus (localities CP40B,
Type species: Miotylopus gibbi (Loomis, 1911 ).
CP41B, CP68C, CP84A, CP84B, CP98C, CP99A).
Type specimen: YPM 10328.
Characteristics: Miotylopus continues the trends seen in Pseu-
Comments: "Paratylopus" cameloides from the Haystack
dolabis, but in addition has a more elongate rostrum and
Valley Member of the John Day Formation (locality PN6G)
highly reduced premolars. Some species of Miotylopus are
is more advanced than P. primaevus, and possibly is a prim-
nearly identical to primitive Stenomylus in all features ex-
itive protolabine.
cept extreme hypsodonty.
Average length of m2: 18.0-25.0 mm.
Paralabis Lull, 1921 Included species: M. gibbi (known from localities CC9C,
Type speci~s: Paralabis cedrensis (Matthew, 1901) (=Pro- CC12, CP48, CP50, CP52, CP87, CPlOl, CP102); M.
tomeryx cedrensis). leonardi (localities CP47, CP48, CP50, CP52, CPlOl); M.
Type specimen: AMNH 8969. taylori (localities CP48, CP50).
Characteristics: Medium-sized camelid with moderately hyp- Comments: Miotylopus is a primitive but contemporary sister
sodont teeth and highly reduced premolars (ratio oflengths taxon to the Stenomylini. Contrary to Peterson (1908) and
of P2-4/ml-3 = 0.5). Distinguished from Poebrotherium Frick and Taylor (1968), it is not necessary to seek a steno-
and Paratylopus by its larger size, greater hypsodonty, and myline ancestor in the Eocene. Prothero (1996) showed
more reduced premolars; distinguished from Pseudolabis that Dyseotylopus Stock, 1935, and part of the material
or Miotylopus by lesser hypsodonty and presence of a referred to Gentilicamelus by Loomis (1936) are junior
mesostyle; otherwise very similar. synonyms of Miotylopus.
Average length ofm2: 13.3-14.7 mm.
Included species: P. cedrensis only (known from localities
CP68D, CP84B, CP99B). STENOMYLINI

Stenomylus Peterson, 1906

STENOMYLINAE Type species: Stenomylus gracilis Peterson, 1906.

Type specimen: CM 1610.
Characteristics: Because Pseudolabis and Miotylopus share derived
Characteristics: Stenomylus and the other two members of the
features with more advanced taxa such as Stenomylus, they are in-
stenomylini are extremely hypsodont and distinctive. As
cluded in an expanded Stenomylinae (Matthew, 1910), which has
shown by Frick and Taylor (1968), the high-crowned teeth
priority as a subfamily name over Pseudolabidinae (Simpson, 1945).
typically fill the entire maxilla and mandible. Also, the
The expanded Stenomylinae is characterized by a long rostrum and a
third molars are extremely elongated anteroposteriorly; the
deep, anteroposteriorly elongate maxillary fossa. The teeth are more
first and second molars are less so. The incisors, canine, and
elongate and transversely narrower than those of Poebrotherium,
first premolar form a row of closely spaced, spatulate teeth,
Paratylopus, Paralabis, or Oxydactylus. All stenomylines lack a
with a long diastema between the first and third premolars
mesostyle and have a posteriorly extended premaxilla (posterior to
(often with an alveolus for a reduced or vestigial p2). The
the level of Pl). The subfamily includes Pseudolabis, Miotylopus
skull is generally short and gracile, and the animal has
(including Dyseotylopus), and the tribe Stenomylini.
very gazellelike proportions. Advanced stenomylines are
generally smaller than contemporary Miocene camelids.
Pseudolabis Matthew, 1904 Average length of m2: 19.5-27.0 mm.
Included species: S. (Stenomylus) gracilis (known from local-
Type species: Pseudolabis dakotensis Matthew, 1904.
ities SB29A, CP51A, CP104B); S. (Stenomylus) hitchcocki
Type specimen: AMNH 9807.
(localities NB3A, CP49B, CP103, NP38B); S. (Pegomy-
Characteristics: Pseudolabis, slightly larger than Poebroth-
lus) keelinensis (locality CP52); S. tubutamensis (locality
erium, shows all the characteristic features of the Steno-
mylinae: deeply depressed maxillary fossa, posteriorly ex- SB54).
Stenomylus sp. is also known from localities CC15,
tended premaxilla, and relatively hypsodont teeth with
CC21A, SB3, SB46, SB47, CP50, CP103, CP104B, CP105:
weak mesostyles. It is also the oldest tax on of the clade to
have a fully closed orbit. Unlike more derived members, it
Blickomylus Frick and Taylor, 1968
has a relatively short rostrum and unreduced premolars.
Average length of m2: 15.0-17.0 mm. Type species: Blickomylus galushai Frick and Taylor, 1968.
Included species: P. dakotensis only (known from locali- Type specimen: F:AM 50840.
ties CP42D, CP48, CP50, CP52, CP84B, CP85C, CP99B, Characteristics: Blickomylus carried the hypsodonty seen in
CP99C, CP103). Stenomylus to greater extremes (see Figure 30.4). The in-
 Camelidae 445

c
A

D~
E

Figure 30.4. Miocene camelids. A-D. Blickomylus galushai. A, B, F:AM 50840 (type), palatal and side views; C, F:AM 50866, showing il-3, c, pl-2,
p4, ml-3; D, F:AM 50869, showing il-3, c, pl-4, ml (displaced), m2-3. E. Miolabis longiceps, AMNH 9108 (type), palatal view of rostrum with Il-3,
C, Pl. Scale bar for A-E = 5 cm. (Illustration by Brian Regal.)

cisors are enlarged, but P2/p2 and P3/p3 are greatly re- for Floridatragulus. Complete dental formula in all known taxa.
duced; in some individuals one or more of these premolars Il-3 distinctly lateral on premaxillaries. Upper teeth anterior to P2
is missing. Molarparastyles and parastylids are very weak spaced by diastemata. Lower canine recurved in Aguascalientia and
or absent. Floridatragulus. The pl small, single rooted. The p4 equal to (?P.
Average length ofm2: 16.0-27.0 mm. franki) or shorter than p3 (Aquascalientia, Floridatragulus). Lower
Included species: B.' galushai only (known from localities canine to pl and pl-p2 diastemata elongate in Aguascalientia and
SB29B, SB29C, SB29D, CP54B, CP73B). Floridatragulus; p2-p3 diastema present in Floridatragulus. Lower
molars selenodont, brachydont, with somewhat bulbous cusps, and
Rakomylus Frick, 1937 in Aquascalientia and Floridatragulus, with labial intercolumnar
Type species: Rakomylus raki Frick, 1937. tubercles. In Aguascalientia and Floridatragulus, m3 hypoconulid
Type specimen: F:AM 30990. with labial, and progressively developed lingual, selene. Upper mo-
Characteristics: The biostratigraphically youngest species of lars quadrate and brachydont, with prominent styles, weak to promi-
stenomyline is Rakomylus. It differs from other stenomy- nent ribs, and lingual intercolumnar tubercles.
lines in having a longer muzzle, a deeply recessed an- An incomplete tarsus tentatively referred to Aquascalientia has
teromaxillary fossa, and relatively short metapodials. Like unfused metatarsals III and IV and an elongated metatarsal V splint
Blickomylus, its molars are extremely hypsodont, and its (Stevens, 1977). Metapodials referred to Floridatragulus show un-
premolars are highly reduced or lost. It has completely fused metacarpals III and IV, and metatarsals III and IV with some
lost Pl and P2/p2. However, Rakomylus is not as derived proximal fusion; metatarsal V is separate from metatarsal IV
as Blickomylus in other respects because the third molars (Maglio, 1966).
are not as anteroposteriorly elongated. The subfamily Floridatragulinae includes ?Poebrotheriumfranki,
Average length ofm2: 18.4 mm. Aguascalientia, and Floridatragulus.
Included species: R. raki only, known from locality SB32B
only. ?Poebrotheriumfranki (Wilson, 1974)
Rakomylus sp. is also known from locality SB30A.
Type species: Poebrotheriumfranki Wilson, 1974.
Type specimen: TMM 40404-149.
FLORIDATRAGULINAE
Characteristics: Rostrum narrow, moderately elongated.
Characteristics: Rostrum moderately (?Poebrotheriumfranki) to ex- Teeth ahead of P2 spaced by diastemata. Il-3 small, leaf-
tremely (Floridatragulus) elongated, narrow. Skull known for ?P. like; I3 same size as I2. Upper canine very small. Pl dou-
franki and Floridatragulus. The skull of Floridatragulus shows a ble rooted. P2-4 relatively small. Upper molars quadrate,
very low profile, nasal bones nearly parallel with the palate, an open brachydont, with bulbous although selenodont cusps, and
orbit posteriorly, a relatively steep occiput with prominent sagit- with prominent styles, ribs, cingula, and intercolumnar tu-
tal and lambdoidal crests, and a well-inflated auditory bulla with bercles. Auditory bullae well inflated. Mandible very slen-
spongy bone and divided by a tympanohyal i:>it and groove (Maglio, der, presumably elongated. The p2-4 relatively small. The
1966). Lower jaw of Aquascalientia and Floridatragulus with a very p4 as long anteroposteriorly as p3. The ml brachydont
elongate rostrum and a very long, narrow symphysis (latter also with bulbous cusps. Symphysis ending beneath p2. Body
present in ?P. franki). Camelid mandibular "hook" demonstrated size small.
 J. G. Honey et al. 446

Average length of m2: unknown. Average length of ml: 9.7 Relatively small size and brachydont dentition with bul-
mm. bous although selenodont cusps with distinct stylids, ribs,
Included species: ?P. franki only (known from localities and intercolumnar tubercles led Stevens (1977) to suggest
SB44D, SB44E). an origin of floridatragulines from oromerycids. However,
the presumed oromerycid features seen in floridatragulines
Comments: ?Poebrotheriumfranki from the medial Chadro- such asAguascalientia are primitive features characteristic
nian Capote Mountain Tuff, Vieja Group, Trans-Pecos of the earliest tylopods in general.
Texas, is an odd camelid with its small, bladelike upper
incisors that are located distinctly lateral on the narrow Floridatragulus White, 1940 (synonym: Hypermekops)
rostrum and do not increase in size toward the posterior,
Type species: Floridatragulus dolichanthereus White, 1940.
p4 small and the same length as p3 instead of longer as
Type specimen: MCZ 3635.
in typical poebrotheres, long mandibular symphysis end-
Characteristics: Skull low, with notably attenuated rostrum.
ing bene.ath p2, brachydont, quadrate upper molars with
11-3 large, subequal in size, caniniform, well spaced, and
prominent styles, ribs, and intercolumnar tubercles, and
placed on the lateral margin of the rostrum. Canine size
rounded, bulbous although selenodont cusps. ?P. franki
is associated with a typical poebrothere, Poebrotherium variable, either smaller or larger than 13, and separated
chadronense (Prothero, 1996), which shows that Cameli-
from adjacent teeth by diastema. Pl (occasionally super-
dae were already well differentiated at this early date. Al- numerary possibly because of rostral attenuation) single
rooted and separated from P2 by a long diastema. P2--4
though no suitably intermediate fossils are known and con-
siderable time separates the occurrences of ?P. franki and small, with relatively simple crowns and tend to decrease
in anteroposterior length toward the posterior, unusual for
Aguascalientia sp., certain morphological features present
came lids. Upper molars brachydont, quadrate, with promi-
in the incisors and premolars of?P. franki suggests to one of
us (MSS) that ?P.franki is possibly related toAguascalien- nent para-, meso-, and metastyles, and distinct lingual in-
tia; accordingly, ?P.franki is placed tentatively within the
tercolumnar tubercles. Auditory bulla well inflated, filled
with cancellous bone, and divided by tympanohyal pit
Floridatragulinae.
and groove. Occiput steep with prominent sagittal and
lambdoidal crests. Postorbital bar incomplete. Lower ca-
Aguascalientia Stevens, 1977
nine relatively small, separated from pl by diastema of
Type species: Aquascalientia wilsoni (Dalquest and Mooser, ml-2 length or greater. The pl single rooted, separated
1974) (= Miotylopus wilsoni). from p2 by diastema of ml-2 length or greater. The p2
Type specimen: TMM 41536-26. longer anteroposteriorly than p4, separated from p3 by di-
Characteristics: Lower canine medium large, oval, canini- astema variably shorter than m3. The p3-p4 small, with
form. The pl small, subcaniniform, separated from the p4 smaller than p3. Lower molars brachydont with low,
canine by a diastema the length of m2. The p 1 separated elongated intercolumnar tubercles. The m3 hypoconulid
from p2 by a diastema the combined length of ml -m2. The divided by lingual and labial selenes. Body size medium.
p2-p3 small, brachydont, bulbous, with prominent, medi- Average length of m2: 15.0-19.2 mm.
ally inflected parastylid. The p4 slightly shorter anteropos- Included species: F. dolichanthereus (including F. barbouri
teriorly than p3 but taller crowned. Lower molars brachy- White, 1947) (known from localities GC4B, GC4C,
dont, rather broad transversely with bulbous cusps, and GC8D); F. nanus (locality GC3B); F. texanus (localities
have an isolated metaconid crest until middle or late wear. GC4B, GC4C); F. hesperus (locality GC4E).
Intercolumnar tubercles present. The m3 hypoconulid di- Floridatragulus sp. is also known from localities GC5,
vided by two unequal selenes; labial selene well developed, GC9A, CP114B.
lingual selene a small elongated cuspid. Body size medium
to small. Comments: White (1940) thought Floridatragalus dolichan-
Average length of m2: 13.3 mm. thereus, the first reported floridatraguline, to be a large,
Included species: A. wilsoni only, known from locality SB55 brachydont hypertragulid because of the spaced p2-p3, and
only. Simpson ( 1945) placed it in Hypertragulidae incertae sedis.
Aquascalientia sp. is also known from locality SB46. Another Thomas Farm taxon, Hypermekops olseni White
(1942), originally placed within the Hypertragulidae, has
Comments: Aguascalientia has been reported only from since been transferred to Floridatragalus by McKenna
Mexico and Trans-Pecos Texas. A. wilsoni, originally (1966), Maglio (1966), and Patton (1966). Based in part
placed within Miotylopus by Dalquest and Mooser (1974), on unpublished work by Bryan Patterson, Ray (1957) rec-
is perhaps slightly younger than the oldestFloridatragalus, ognized Floridatragalus questionably as a camelid, and
F. nanus. Aguascalientia sp. is 50 percent smaller than Olsen (1962) concluded that Floridatragulus was a true
F. dolichanthereus and 25 percent smaller than F. nanus and camel. Maglio (1966) and Patton (1966) believed Flori-
A. wilsoni, and may be ancestral to the last two species. datragulus was phylogenetically well separated from the
 Camelidae 447

other camelids on the basis of the greatly elongated snout, !um. The p3--4 with strong anterolingual fold. The p3 with
extremely long and narrow mandibular symphysis, reduced strong posterolingual cusp; posterolingual cusp present
premolars (especially P4/p4 ), and divided m3 hypoconulid. on p2 of some species. Intercolumnar tubercles variably
Maglio (1966) noted that the jaw has the typical camelid present on molars, and where known, are V shaped on up-
"hook." Floridatragulus was derived from a camel more per molars. Upper molars with very strong ribs and styles
primitive than Oxydactylus, as indicated by the open orbit as in Gentilicamelus and primitive Oxydactylus. Poste-
(McKenna, 1966). rior edge of m3 entoconid slightly overlaps hypoconulid.
Inv~stigators have disagreed on the identifications of the Mandible relatively deep and transversely thin as in Oxy-
teeth or alveoli anterior to P2. This disagreement arises dactylus. Angle of mandible somewhat expanded.
in part because of the failure to find the premaxillary- Average length of m2: 10.6-20.5 mm.
maxillary suture on a referred skull (MCZ 3711), and the Included species: N.floridanus (known from localities GC3B,
presence of supernumerary premolars on another referred GC8D); N. hidalgensis (locality GC3B); N. waldropi
specimen, F:AM 31864; in this second specimen, the ros- (GC8A).
trum is broken anterior to the upper canine. In this chapter, Nothokemas sp. is also known from locality GC9A.
we follow White's (1942) interpretation that Floridatrag-
ulus had a complete dental formula. A restoration of the
MIOLABINAE
head of Floridatragulus is shown in Figure 30.lB.
Characteristics: 13 caniniform, equal to or larger than upper ca-
nine. P2 reduced through shortening of the anterior and poste-
CAMELIDAE, INCERTAE SEDIS:
rior crests. Molars more hypsodont than in Gentilicamelus and
"NOTHOKEMATINAE"
Nothokemas, and usually more hypsodont than in Oxydactylus. Up-
Gentilicamelus Loomis, 1936 per molar ribs reduced compared with those of Gentilicamelus. ln-
tercolumnar tubercles variably present on molars. Orbits closed.
Type species: Gentilicamelus sternbergi (Cope, 1879)
Maximum width of rostrum occurs across the incisors. Buccina-
(= Poebrotherium sternbergi).
tor fossa usually weak; premaxillaries and anterior maxillaries not
Type specimen: AMNH 7910.
flared strongly immediately below nasals, unlike protolabines. Max-
Characteristics: Dental formula: 13(?)/3, Cl/I, P4/4, M3/3.
illary fossa very shallow to deep. Mandibular angle expanded. Coro-
Pl double rooted; pl uncertain. P2--4 less reduced relative
noid process short. Short cervical vertebrae. Metapodials short and
to Ml-3 than in 0.xydactylus. P3 internal crescent weaker
unfused or only weakly fused in metatarsus. Metatarsus longer than
than in 0.xydactylus. Cheek teeth very brachydont (more
metacarpus.
so than in 0.xydactylus) with large ribs and styles on upper
Comments: We use Hay's (1902) subfamily Miolabinae for the
molar~. Intercolumnar tubercles present on Ml-3, and at
generaMiolabis, Paramiolabis, Nothotylopus, and Cuyamacamelus.
least on m3. Maxillary fossa present, but apparently shal-
As used by Hay (1902), Miolabinae included not only Miolabis
low. Palatine notch extends as far forward as middle of
sensu stricto, but also Procamelus (which included species once
M2. Auditory bulla well inflated. Metapodials slender, un-
placed in Procamelus, Protolabis, Miolabis, and Alticamelus
fused, shorter than skull length, and with metatarsus longer
( = Aepycamelus)), Came lops, and Pliauchenia (including Giganto-
than metacarpus. Proximal phalanges with distal articular
camelus); as pointed out by Simpson (1945), Hay's Miolabinae was
surfaces unexpanded dorsally.
essentially synonymous with Zittel's (1893) Protolabinae. Simp-
Length of m2: approximately 14.7 (broken). Length of M2:
son (1945) included Miolabis doubtfully within the Alticamelinae,
12.0mm.
following, with modifications, Matthew's (1904) association of Mi-
Included species: G. sternbergi only, known from within the
olabis with Paratylopus and Oxydactylus. Matthew's arrangement
John Day Formation, ?locality PN6D.
was based on the shared-primitive characters oflow-crowned molars
with strong ribs and styles and simple P4s.
Nothokemas White, 1947
Species of Miolabis have long been known, although as shown
Type species: Nothokemas floridanus (Simpson, 1932) by Frick and Taylor (1971), most were incorrectly placed in Pro-
(= Oxydactylus floridanus). tolabis. Nothotylopus is based on a lower jaw. Patton (1969) found
Type specimen: FSGS V-5247. precise systematic placement of Nothotylopus difficult because of
Characteristics: Dental formula 13(?)/3, Cl/l, P4/3, M3/3. 13, what he termed a "strange mixture of advanced and conservative
Cl/cl caniniform. Cl larger than 13; cl large, recurved. Pl features" (p. 162), but concluded that it was probably derived from
with roots closely appressed or fused; p 1 absent in contrast an early Protolabis or from an oxydactyline-Protolabis interme-
to Gentilicamelus and Oxydactylus. Long c l-p2 diastema. diate. Comparison of the lower jaws of Nothotylopus with sam-
Cheek teeth primitively very brachydont as in Gentili- ples of an undescribed but more completely preserved miolabine
camelus. P2--4/p2--4 unreduced in primitive species; slight (miolabine sp.; see later) establishes that Nothotylopus and Mio-
reduction in more advanced species. P2 internal cingulum labis are related, an unpublished conclusion previously reached by
weak to strong; P3 with more prominent internal cingu- Beryl Taylor.
 J. G. Honey et al. 448

Paramiolabis Kelly, 1992 (synonyms: Pliauchenia, in part, Il-2 large, cupped, tending to form closed arc at front of
Miolabis, in part, "Merychenia" [Frick ms. name]) premaxillaries (Figure 30.4E). I3 caniniform, larger than
Cl. Pl usually single rooted or with two fused roots; some
Type species: Paramiolabis singularis (Matthew, 1918)
specimens have a double-rooted Pl with flaring roots (Fig.
(= Pliauchenia singularis).
30.4E). The pl is absent. Premolars stouter, less laterally
Type specimen: AMNH 17329.
compressed than in Protolabis. P2/p2 shortened and sim-
Characteristics: Dental formula I3/3, Cl/1, P2-3/2-3, M3/3.
plified through reduction of anterior and posterior crests.
Il-2 conical, nonspatulate; I3 caniniform. All premolars
P3 broadened relative to Oxydactylus, Gentilicamelus, and
reduced relative to molars, with Pl/l absent or unerupted
Paratylopus, with a reduced parastyle and a complete in-
(rare specimens show an erupted, vestigial Pl), and P2/2
ternal crescent; p3 often shortened and with a high, central
highly reduced and sometimes absent. The rn2-3 with
protoconid. The p4 more basally swollen than in Oxydacty-
protostylids and parastylids forming anterior buttresses.
lus, and with an enlarged hypoconid. Upper molars with
Diastema between cl and p2 or p3 elongate, with dor-
strong parastyle and mesostyle; ribs less prominent than in
sal border of dentary having a sharp crest and commonly
Oxydactylus and Gentilicamelus. Molars lower crowned,
slightly arched upward. Ramus elongate and slender. An-
with stronger metastylids, and M3/m3 less anteroposteri-
gle of mandible commonly expanded, flattened along the
orly expanded than in Protolabis. Intercolumnar tubercles
posterior edge, with a variable lateral rugosity or flaring in
on lower molars most common on ml. Maxillary fossa
the area of the masseteric insertion; coronoid process short.
usually deeply depressed, subcircular, and deepest pos-
Palate in region of Cl to P2 or P3 diastema slightly arched
teriorly and dorsally. Buccinator fossa weak to absent.
upward; rostrum moderately elongate and constricted, and
Palatine notch extends anteriorly only to the posterior bor-
above the Cl to P2 or P3 diastema is laterally inflated.
der of M3. Lateral plate of bulla usually projects signifi-
Maxillary fossa shallow to absent. Supraorbital region
cantly below paroccipital process and is level with or below
slightly convex. Auditory bulla inflated, with anterome-
the occlusal surface of the upper molars. Mandibular an-
dial process present. Cervical vertebrae short. Metapodi-
gle expanded with a lateral rugosity or ridge along the
als shorter than basal length of skull, with the metacarpus
posteroinferior border; narrow internal shelf sometimes
unfused and the metatarsus sometimes fused proximally.
present on posteroventral margin of angle. Mandibular
Average length of m2: 15.0-28.0 mm.
"hook" weak. Cervical vertebrae short. Metapodials un-
Included species: P. singularis (known from locality CPl 10);
fused and much shorter than basal length of skull. Proximal
P. tenuis (locality CP108B); P. taylori (locality CCI 7E).
phalanx with distal articular surface only weakly expanded
Paramiolabis sp. is also known from localities GC4B,
dorsally.
GC4C, NB3D, NB6A, NB6B, NB6C, NB6D, SB32A,
Average length ofm2: 19.0-25.0 mm.
SB32B, SB32D, SB32E, CP107.
Included species: M. transmontanus (known from locality
Comments: Kelly (1992) considered the type species of PN7); M. fissidens (localities NB6E, CP49C, CP75C); M.
Paramiolabis to be P. tenuis from the Sheep Creek Forma- montanus (locality NP41B); M. longiceps (locality CP76);
tion of Nebraska, originally described by Matthew (1924) M. princetonianus (localities CP107, CPl 10); M.fricki (lo-
as Miolabis tenuis. However, in his unpublished disser- calities CC 17C, CC 17D, CCI 7E). Questionably included:
tation, Barghoorn (1985) described the same genus un- M. califomicus (locality CC15).
der the Frick and Taylor manuscript name of Merychenia Miolabis sp. is also known from localities CC21A,
and, in agreement with their earlier conclusion, recognized CC21B, CC21C, NB6A, NB6C, NB6D, SB32A, SB32B,
the type species as Pliauchenia singularis Matthew (1918) SB32D, CP75B, CP108B, CP114A, ?CP114D, CP116B,
from the Olcott Formation of Nebraska. Barghoorn's anal- ?NPlOE.
ysis of "Merychenia" was based on study of the AMNH Comments: Frick and Taylor (1971) have shown that most
collection of this camelid. The diagnosis of Paramiolabis species of Miolabis were incorrectly placed in Protolabis,
given here relies heavily on Barghoorn's extensive diag- and clarified the distinctions between the two genera. Mi-
nosis and description of "Merychenia," supplemented by olabis califomicus is herein questionably retained in Mi-
Kelly's description of Paramiolabis and our own exami- olabis, with the realization that future revision may place
nation of the AMNH material. it in another genus. M. califomicus differs from typical
Miolabis in having nonspatulate upper incisors.

Miolabis Hay, 1899 (in Matthew, 1899) (Synonym:

Protolabis, in part) Cuyamacamelus Kelly, 1992
Type species: Miolabis transmontanus (Cope, 1879) (=Pro- Type species: Cuyamacamelusjamesi Kelly, 1992.
tolabis transmontanus). Type specimen: UCMP 65339.
Type specimen: AMNH 8196. Characteristics: Dental formula I3/?, Cl/?, P4/?, M3/?. I2
Characteristics: Dental formula I3/3, Cl/1, P4/3, M3/3. nonspatulate, smaller than I3. I3 caniniform, larger than
 Camelidae 449

C 1. P 1 small, single rooted. P2-3 less laterally compressed present in localities from early Hemingfordian through late
than in protolabines, and with bulbous central cusp. P2 Barstovian age. Probably comprising several species, these
shortened relative to P3 and P4. P3 with complete internal miolabines are closely related to Nothotylopus and, be-
crescent. Molars brachydont and with strong labial ribs and ing represented by well-preserved cranial and postcranial
styles. Skull long, with elongate, dorsoventrally narrow remains, establish the relationship between Nothotylopus
rostrum that is not extremely constricted laterally as it is in and Miolabis. They differ from Nothotylopus mainly in the
derived protolabines. Maxillary fossa deepest posteriorly presence of the p2; whether they are species of Nothotylo-
but apparently unpocketed. Lateral plate of bull a damaged, pus or belong in a new genus awaits publication of analyses
but apparently projected below the level of the damaged of this group. The camelid described by Voorhies ( l 990a, p.
paroccipital process. A219) as gen. et sp. indet. may represent a species of these
Average length of m2: unknown. Average length of M2: poorly known miolabines or their close relative, Nothoty-
30.15 mm. lopus. Characters of these miolabines include Il-2 peg-
Included species: C. jamesi only, known from locality CCI 7D like, lingually curved, and spaced; I3 conical, and equal to
only. or larger than Cl; Pl/pl double rooted, with roots often
Comments: Kelly (1992) compared Cuyamacamelus mainly widely flared; p2-4 three lobed in labial view, with deep
with Nothokemas, Flo ridatragulus, and Miolabis, and con- p4 posterolabial sulcus; lower molars with variably de-
cluded that the subfamily position of Cuyamacamelus was veloped anterior buttresses and with relatively fiat lingual
indeterminate. Although the lower jaw and postcrania of walls (through reduction or elimination of metastylids);
Cuyamacamelus jamesi are unknown, certain features pre- maxillary fossa shallow to deep, sometimes pocketed just
sent on the skull suggest a relationship with miolabines. beneath nasals; lateral plate of bulla roughly triangular
Most important are indications that the auditory bulla ex- and strongly compressed transversely, and extremely elon-
tended below the paroccipital process, as in Miolabis and gated to extend well below level of paroccipital process;
miolabine sp. (and probably also Nothotylopus). Also, in medial plate of bulla strongly compressed transversely,
miolabines the P2 is reduced, with a reduced parastyle, and smooth and deeply excavated on the medial side; mandibu-
in Miolabis and miolabine sp. the P2 and P3 are commonly lar angle greatly expanded with a posterolateral rugosity
broad transversely with a bulbous paracone on the P3, as and with the ventral margin lingually inflected; coronoid
in Cuyamacamelus. In miolabines the I3 is equal in size process reduced; metapodials shorter than basal length
or larger than the Cl (a primitive feature), and the Pl is of skull, with metacarpals unfused and metatarsals partly
often reduced. The maxillary fossa ranges from shallow to fused.
deep in miolabines, the orbit is closed, and in Miolabis and Average length ofm2: 25.0-31.0 mm.
miolabine sp. the palatine notch is posterior to the middle Miolabine sp. is known from localities GC4B, GC4C,
of M3. Other features mentioned by Kelly (1992), such SB29C, CP75B, CP75C, CP107, CP108A, CP108B,
as low, fiat cranium, laterally situated I2, and somewhat CPllO, CP114C, CP114D.
reduced premolars, are also found in some specimens of
Miolabis and miolabine sp. Nothotylopus Patton, 1969
On Cuyamacamelus, the brachydont cheek teeth and
Type species: Nothotylopus camptognathus Patton, 1969.
(apparently) lengthened lateral plate of the auditory bulla
Type specimen: TMM 31081-26, formerly in the UTBEG
indicate that this genus is a primitive member of the
collection.
Miolabis-miolabine sp.-Nothotylopus clade. Although the
Characteristics: Dental formula I?/3, C?/1, P?/3, M?/3. The cl
I2 of Cuyamacamelus is not cupped, Kelly (1992) men-
caniniform, robust. Double-rooted p 1; p2 absent. Long pl-
tioned the presence on the I2 of a "small posterior incipi-
3 diastema. The p3-4 reduced relative to molars. Three-
ent cusp," which may approximate the primitive condition
lobed p3-4, with buccal sulci anterior and posterior to the
for the cupped incisors of Miolabis; however, C. jamesi
protoconid; posterior sulcus on p4 deep. The p4 parastylid
itself is too late in time to be ancestral to Miolabis. As in
generally well developed and strongly inflected lingually.
Miolabis, the Pl of Cuyamacamelus is single rooted, and
Lower molars without metastylids and with pronounced
the P3 has a complete internal crescent (also present in
protostylids and variably developed parastylids on m2-
some specimens of miolabine sp.). The type specimen of
3, which form anterior buttresses. On lower jaw, masse-
C. jamesi is too poorly preserved to allow evaluation of its
teric insertion greatly expanded and laterally flared, angu-
morphology relative to nodes defining the Nothotylopus-
lar process ("hook") prominent but not strongly inflected,
miolabine sp. clade; until better material is described, we
and coronoid process short, weak. Referred metapodials
place Cuyamacamelus near Miolabis based on the dental
short with metacarpals unfused and metatarsals partially
features described here.
fused. Differs from Miolabis in pl present, p2 absent,
more expanded mandibular angle, flatter walled and more
Miolabine sp.
strongly buttressed lower molars, and premolars smaller
Rarely described but relatively common miolabines are relative to molars. Differs from miolabine sp. in
 J. G. Honey et al. 450

loss of p2. Differs from Oxydactylus in loss of p2, but- Characteristics: Complete dental formula. Il-2 weak, later-
tressed lower molars, expanded masseteric insertion, and ally compressed, concave lingually, and often more lat-
weak coronoid process. erally situated than in Oxydactylus. Cl larger than 13.
Average length ofm2: 27.0-31.0 mm. The pl roots closely appressed or fused. P3 lingual cin-
Included species: N. camptognathus only, known from local- gulum unbroken. Molars lower crowned and shorter than
ity GC6B only. in Protolabis. Rostrum generally wider than in Protolabis
Nothotylopus sp. is also known from localities SB32A, and Michenia, and buccinator fossa is only weakly de-
SB32B, SB32D, SP2A, CP114B, CP116B. veloped. Maxillary fossa well developed. Basisphenoid
with two elongate, strong, ventral tuberosities. Mandibular
angle unexpanded ventrally and with weak mesial inflec-
PROTOLABINAE tion. Limbs not elongated, with metapodial length less than
basal length of skull. Metacarpals unfused and metatarsals
Characteristics: Dental formula: 13-1/3, Cl/l, P4/4-3, M3/3. Il-2
weakly fused.
losJ in derived sp~cies. Distance between upper canines equal to Average length ofm2: 17.5-lS.5 mm.
or less than distance between l3s. Pl/p 1 double rooted (roots may
Included species: T. brachyodontus only (known from local-
be closely appressed), with low, blunt crown. The pl may be sup-
ities CC16, CP51A, CP104B, CP105).
pressed or lost in advanced species. Shorter cranial (postorbital)
length relative to facial length than in Camelini and Lamini. Ten-
dency toward marked rostral constriction between C 1 and P2, asso- Protolabis Cope, 1876
ciated with progressive development of buccinator fossa. Anterior
Type species: Protolabis heterodontus (Cope, 1S73) (=Pro-
nares laterally expanded. Maxillary fossa never pocketed. Paroccip-
camelus heterodontus).
ital process rarely extends below level of auditory bulla. Cervical
Type specimen: AMNH S296.
vertebrae not elongated. Metapodials unfused or weakly fused in
Characteristics: Dental formula 13-1/3, Cl/l, P4/4-3, M3/3.
primitive species and fused in advanced species. Metapodials shorter
Il-2 lost in derived species. 13, Cl/cl conical and recurved,
and stockier in some advanced species. Length of metacarpus equal
rarely transversely compressed, and without sharp ante-
to or less than length of metatarsus.
rior and posterior edges. Pl smaller than 13. The pl may
Comments: Protolabis (sometimes including species later allo-
be suppressed or absent in derived species. The p2-4 show
cated to Miolabis) and the much later described Michenia have been
progressive size reduction in advanced species, with p2 oc-
considered ancestral to some or all later came lids (Cope, l SS6; Wort-
casionally lost. The p3-4 with middle cusps usually equal
man, 1S9S; Scott, 1937; Webb, 1965, 1972), and in this century
to or wider than posterior cusps. Molar styles and ribs
have been classified within the Camelinae. Webb (1965, p. 44) es-
less prominent in advanced species. Palate narrow behind
tablished the tribe Protolabidini of the Camelinae as a "horizontal
canine. Rostral constriction extreme in advanced species,
ancestral group" for later camelines. Honey and Taylor (197S) dis-
and associated with pronounced development of buccina-
puted this view, citing a number of morphological features that they
tor fossa. Maxillary fossa shallow to deep, but not pock-
felt removed the protolabidines from the ancestry of later came lids.
eted. Angular process of dentary ("hook") more subdued
They retained the Protolabidini within the Camelinae, however,
in advanced species due to posterior and dorsal expansion
considering that subfamily monophyletic because of the features
of insertion area for masseter, and with only slight mesial
of weak buccinator fossa and elongate rostrum. They contrasted
inflection. Weak (primitive) to strong (derived) lateral flare
the Camelinae with the Aepycamelinae (Oxydactylus and Aepy-
and weak (primitive) to strong (derived) mesial tuberosity
camelus), the latter subfamily considered monophyletic by virtue
on posteroventral border of mandible. Coronoid process
of elongate limbs and cervical vertebrae, and metapodials longer
low. Metapodials fused, equal to, or less than basal length
than the basal length of the skull.
of skull, becoming shorter in some derived species. Cranial
Restudy of Aepycamelus and Procamelus, however, has shown a
material of Protolabis is illustrated in Figure 30.2.
number of features indicative of close relationship (see later), and
Protolabis differs from Tanymykter in stronger bucci-
that they are more closely related to each other than either is to Pro-
nator fossa and generally narrower rostrum; basisphenoid
tolabis. Retention of Tanymykter, Protolabis, and Michenia within
without well-developed elongate tuberosities; P2 with
the Camelinae and exclusion of Aepycamelus from that subfamily,
weaker metastyle and parastyle and discontinuous lingual
as done by Honey and Taylor, results in a polyphyletic Camelinae.
cingulum; P3 lingual cingulum weaker and generally dis-
We therefore remove the protolabines from the Camelinae and return
continuous; upper and lower molars taller crowned and an-
to the usage of Zittel's (1S93) Protolabinae for this very distinctive
teroposteriorly longer, with weaker buccal ribs above and
group consisting of Tanymykter, Protolabis, and Michenia.
weaker metastylids and lingual ribs below; mandibular an-
gle more ventrally expanded; metapodials fused; proximal
Tanymykter Honey and Taylor, 1978
phalanx less convex dorsally and distal articular surface
Type species: Tanymykter brachyodontus (Peterson, 1904) less grooved and more anteriorly expanded.
(= Oxydactylus brachyodontus). Differs from Michenia in larger size; longer braincase;
Type specimen: CM 664. stronger Il-3 in primitive species; stronger and more
 Camelidae 451

caniniform Cl/cl; more anteroposteriorly expanded M3/ Characteristics: Complete dental formula. 13, Cl/cl, canini-
m3; deeper, more ventrally produced symphysis; deeper form. l3 often as large as, or larger than, Cl; distance
horizontal ramus; relatively stouter metapodials. between l3s equal to or greater than distance between
Average length ofm2: 19.1-30.3. Cls. Pl/pl two rooted, low crowned. P2-4/p2-4 more
Included species: P. heterodontus (known from localities reduced than in Gentilicamelus and primitive Nothoke-
CA9, SB32A, SP2A, CP54B, CP74B, CP75B, CP90A, mas. P3 relatively elongate and narrow, with a prominent
CP114B, CP114D, CP116A); P. barstowensis (localities parastyle as in Gentilicamelus, and with a strong inter-
NB6C, NB6D, NB6E); P. coartatus (incl. P. notiochorinos) nal cingulum, which is usually continuous. The p3-4 with
(localities GC6A, NB23C, SB6); P. inaequidens (= well-developed, strongly lingually inflected paraconids;
Procamelus inaequidens) (localities NB6E, CP114B); P. those of p3 are usually more strongly inflected than in
gracilis (P. cf. gracilis) (= Procamelus gracilis) (locali- Miolabis. Cheek teeth brachydont, although more hyp-
ties CP114A, CP114B). sodont than in Gentilicamelus. Intercolumnar tubercles
Protolabis sp. is also known from localities CC l 7D, variably present on upper molars; rare on lower molars.
CC17G, CC17H, CC19, CC21C, NB3B, NB6B, ?NB7C, Upper molars with strong external ribs and styles; ribs (es-
NBS, ?NB2S, SB29C, SB32B, SB32C, SB32D, SB32E, pecially on M3 metacone) weaker on later Oxydactylus.
SB32F, SB34A, CP73C, ?CP73D, CP75C, CP105, CPlOSA, Paroccipital process at same level or slightly below bot-
CPlOSB, CPl 10, CP114A, CP114C, CP116B. tom of lateral plate of auditory bulla; lateral plate at same
level or above occlusal surfaces of upper molars. Auditory
bulla relatively less inflated for size of skull than in Poe-
Michenia Frick and Taylor, 1971 brotherium, Paratylopus primaevus, and Gentilicamelus
Type species: Michenia agatensis Frick and Taylor, 1971. stembergi; bullar compression less extreme than in mi-
Type specimen: AMNH 14255. olabines. Maxillary fossa shallow to deep and well de-
Characteristics: Dental formula 13-1/3, Cl/l, P4/4-3, M3/3. fined but not pocketed; maxillaries depressed immediately
Il-3 uncrowded, weak, and incisiform. Il-2 lost in beneath nasals. Palatine notch V shaped and extends as
derived species. Cl small, incisiform; cl small, laterally far forward as the anterior edge of the M3. Lower bor-
compressed, and closely associated with incisors. The pl der of mandible relatively fiat in contrast to Aepycamelus;
suppressed in advanced species. P2-4/p2-4 progressively slightly concave just in front of ascending ramus. Cervical
reduced, with p2 sometimes absent in advanced species. vertebrae elongated; where known, elongation is greatest
Molars taller crowned with weaker ribs, styles, and sty lids, in third cervical. Limbs elongate, slender. Metapodials un-
and M3/m3 more elongate, in advanced species. Slen- fused, and longer than basal length of skull. Metacarpal
der horizontal ramus and shallow mandibular symphysis. length equal to or slightly shorter than metatarsal. Distal
Rostrum elongate and extremely narrow. Braincase short, articular surface of proximal phalanx usually only slightly
rounded. Orbits large. Metapodials unfused (primitive) to expanded dorsally, and with asymmetrical upper surface.
fused (derived). Metapodials slender, becoming shorter Ungual phalanges high, narrow, and pointed. Smaller than
and stockier in advanced species. Aepycamelus, with more brachydont, less massive denti-
Differs from Tanymykter in more constricted rostrum and tion.
deeper buccinator fossa; relatively shorter braincase; Average length of m2: 19.0-25.0 mm.
weaker l3 and C 1/cl; shallower symphysis; slender meta- Included species: 0. longipes (known from localities CC52,
podials. CP51A, CP104B, NP2SB ); 0. campestris (locality CPl 03 );
Average length ofm2: 16.0-24.S mm. 0. longirostris (localities CP105, CP107); 0. lulli (local-
ity CP51A); 0. benedentatus (localities ?CA2, GC3B);
Included species: M. agatensis (known from localities NB3B,
CP104B); M. exilis (including M. australis) (localities 0. wyomingensis (locality CP52); 0. lacota (localities
SB46, CPS6D); M. yavapaiensis (localities SB6, SB33II); CPS6D, NP3SB).
M. deschutensis (locality PN6H). Oxydactylus sp. is also known from localities ?GCSC,
Michenia sp. is also known from localities NB6B, NB6C, SB4, SB2S, ?CP45D, CP71, ?CPSS.
NB6D, NBS, ?NB2S, SB32A, SB32B, SB32C, SB32D,
Comments: Oxydactylus (Figure 30. lA) is a poorly under-
?SB32E, SB32F, ?SB34A, CP51A, CP73D, ?CP74B,
stood taxon defined by elongate neck and limbs in combi-
CP75C, ?CP104B, CP105, CP107, CPlOSA, CPlOSB,
nation with a primitive dentition; as apparent from past
CP116A, NPlOD, NPlOE.
usage, Oxydactylus signifies a grade of camelid evolu-
tion and not necessarily a phylogenetically closely related
CAMELIDAE, INCERTAE SEDIS group of species. Oxydactylus has at various times included
species now referred to Miotylopus, Nothokemas, Miche-
Oxydactylus Peterson, 1904 nia, and Tanymykter, and it is questionable whether all
Type species: Oxydactylus longipes Peterson, 1904. of the species currently referred to Oxydactylus actually
Type specimen: CM 91S. belong there.
 J. G. Honey et al. 452

Priscocamelus Stevens, 1969 (in Stevens, Stevens, Metapodials fused, with metacarpus longer than metatarsus, (sec-
and Dawson, 1969) ondarily subequal in extant genera). On proximal phalanx, distal
articular surface dorsally expanded.
Type species: Priscocamelus wilsoni Stevens, 1969. Comments: In Simpson's (1945) classification, the Camelinae is
Type specimen: TMM 40849-1, formerly in UTBEG collec- a varied lot without tribal subdivision, similar to Matthew's (1904)
tion. "series B" with modifications. Webb (1965) brought some order
Characteristics: Complete dental formula. 13, Cl/cl canini- into the subfamily when he divided it into three advanced tribes,
form. l3 larger than C 1. Pl/p 1 double rooted, low crowned.
Camelini, Camelopini, and Lamini, and an ancestral tribe, the Proto-
Cheek teeth brachydont. P2-3/p2-4 long and narrow. P2
labidini. Harrison (1979) simplified this classification by discarding
internal cingulum weak but continuous; P3 internal cin- Camelopini, in effect showing that it is a polyphyletic tribe com-
gulum stronger but sometimes notched anteriorly. The p3 posed of members of the Lamini and Camelini. Procamelus was
without posterolingual cusp seen in Nothokemas. Upper
not assigned to the Lamini or Camelini because Harrison could find
molars with very strong buccal ribs and styles; intercolum-
no derived features linking it to either tribe. However, both Har-
nar tuber~les sometimes present. Lower molars less elon-
rison (1979) and Webb (1965, 1972) suggested a probable closer
gate, relatively wider, more brachydont, and with stronger
relationship of Procamelus with the camelines than the !amines.
ribs and stylids than in Australocamelus. Mandible shal-
Several derived features suggest that Hesperocamelus and Aepy-
lower than in Oxydactylus, with downwardly convex lower
camelus sensu Jato should be included within the Camelinae (see
border, and with less expanded masseteric insertion.
Figure 30.3). In common with Procamelus, these two genera have
Rostrum, C/c-Pl/pl, and Pl/pl-P2/p2 diastema shorter
a relatively narrow ramus with a convex lower border from front to
than in Oxydactylus. Palatine notch V shaped and ex-
back, a caniniform Pl/p 1, and fused metapodials. The metacarpus is
tends as far forward as middle of M2. Cervical vertebrae
longer than the metatarsus in Aepycamelus and Procamelus; the oc-
shorter than in Oxydactylus. Metapodials unfused, longer
currence of this character in Hesperocamelus is uncertain. The ear-
than basal length of skull, and with metatarsus longer than
liest supposed species of Aepycamelus, "A." priscus from the Sheep
metacarpus. Size smaller than in Australocamelus.
Creek Formation (locality CP108, early Miocene), has an unpock-
Average length ofm2: 16.8-17.8 mm. eted maxillary fossa, the primitive camelid condition that is also
Included species: P. wilsoni only (known from localities
present in protolabines. A pocketed fossa on the maxillary, which in
SB46, SB47). noncamelines is found in the otherwise very different stenomylines
(anteromaxillary fossa) and some miolabines, appears for the first
Australocamelus Patton, 1969 time in Hesperocamelus, someAepycamelus from the Olcott Forma-
Type species: Australocamelus orarius Patton, 1969. tion, and early Procamelus from the Miocene Olcott and Valentine
Type specimen: TAMU 2466A, currently curated in TMM formations (localities CPl 10, CPl 14); the latter two genera also
collections. show a distinct lateral inflation of the maxillary above the pocketed
Characteristics: Dental formula I?/?, C?/?, P?/4, M?/3. The fossa and, usually, a rather broad rostrum in palatal view (note that
pl two rooted; p2-4 unreduced, elongate, and narrow. The not all later Aepycamelus have a pocketed maxillary fossa). Hes-
pl-2 diastema short. Molars subhypsodont, narrow. perocamelus, some later Aepycamelus, and Procamelus also show
Mandible slender, becoming vertically constricted at increased hypsodonty.
pl-2 diastema. Differs from Oxydactylus in more shal- These similarities indicate to us that Aepycamelus and Hespero-
low mandible and in laterally compressed premolars and camelus are more closely related to Procamelus than are the proto-
molars (modified from Patton, 1969). labines. This idea is not entirely novel: Matthew (1932) postulated
Length of m2: 21.7 mm (one specimen). that Procamelus was a derivative of Alticamelus ( = Aepycamelus),
Included species: A. orarius only, known from locality GC3B and Frailey (1978) suggested a possible relationship between the
only. two genera based on hypsodonty. We therefore remove the proto-
Australocamelus sp. is also known from localities labines from the Camelinae and add Aepycamelus sensu Jato and
SB32B, SB32D. Hesperocamelus to that subfamily. The subfamily Aepycamelinae
Webb (1965) (= Alticamelinae Simpson [1945]), which includes
Aepycamelus, Oxydactylus, and various other genera depending on
CAMELINAE
the author, is abandoned as paraphyletic.
Characteristics: Dental formula 13-0/3, C 1/1, P4-2/4- l, M3/3. Il-2 Harrison (1979) raised the possibility that Aepycamelus and the
present in Hesperocamelus and some Aepycamelus. When present, Lamini are closely related. Aepycamelus shares with the dentally
Pl/pl caniniform and relatively high crowned, except where sec- primitive !amine, Hemiauchenia, extremely elongate limbs and cer-
ondarily reduced. Cheek teeth relatively high crowned compared to vical vertebrae; some later Aepycamelus show the !amine features
Oxydactylus, with a trend of increasing hypsodonty and premolar of strongly arched nasals and lower molars with anteroexternal
reduction and loss. Maxillary fossa ranges from deeply pocketed to sty lids ("llama buttresses"). Moreover,Aepycamelus shares with the
absent. Primitively, mandibular ramus convex downward along its Lamini and Camelini a metacarpus that is longer than the metatar-
entire lower border, rather than concave beneath ascending ramus. sus. For these reasons we include Aepycamelus within the Larnini.
 Camelidae 453

"Aepycamelus" priscus Matthew, 1924 rocamelus is uncertain, but we predict that future findings
will support its position as sister taxon of the Lamini.
Type specimen: AMNH 14189
Characteristics: Dental formula I?3/3, Cl/l, P4/4, M3/3. Cl
laterally compressed. Pl compressed and caniniform with LAMINI
one root in type specimen. Maxillary fossa broad and sub-
circular but unpocketed. Cervical vertebrae not greatly Characteristics: Dental formula 13-1/3, Cl/I, P4-3/4-3,
elongated. Metapodials fused. M3/3. Cl/cl recurved and laterally compressed, not extremely en-
Average length of m2: unknown. Length of P2: 14.1 mm. larged. Pl/pl present in Aepycamelus, Alforjas, and Hemiauchenia,
Included species: "A." priscus only (known from localities lost in Came lops, Palaeolama, and Lama. Them 1-3 with anteroex-
?CP107, CP108B). ternal stylids ("llama buttresses") varying from weak in Alforjas
to strong in Lama. Except in Aepycamelus, rostrum slender com-
pared to that of Camelini. Hemiauchenia, Palaeolama, and Lama
Hesperocamelus Macdonald, 1949
with facial shortening. Nasals arched in cross section. Mandibu-
Type species: Hesperocamelus stylodon Macdonald, 1949. lar diastemal crest sharp. Mandibular symphyseal region typically
Type specimen: UCMP 35382. more procumbent than in Camelini. Metapodials fused; metacar-
Characteristics: Complete dental formula. Il-3, Cl, and Pl pus longer than metatarsus (metacarpus and metatarsus subequal in
small, lingually curved, peglike, spaced, and forming more Lama).
or less parallel rows on either side of rostrum, in contrast
to Aepycamelus. Il-3 tending to wear blunt at tip. Pl/pl Aepycamelus Macdonald, 1956 (synonym: Alticamelus)
single or double rooted, and higher crowned than in pro-
tolabines. The cl and pl relatively small and flattened lin- Type species: Aepycamelus giraffinus (Matthew, 1909, in
gually; pl high crowned, not recurved posteriorly, and leaf Matthew and Cook, 1909) (= Alticamelus giraffinus).
shaped. Upper cheek teeth relatively large, with premolars Type specimen: AMNH 9101.
more massive, laterally expanded, and less reduced rel- Characteristics: Dental formula 13-1/3, Cl/1, P4-3/4-3,
ative to molars, than in protolabines. Rostrum long and M3/3. When present, Il-2 small, lingually curved, and
slender. Palate narrow, with weak buccinator fossa above peglike. Distance between l3s equal to or less than distance
Pl in contrast to most Aepycamelus. Maxillary fossa between Cls. 13, Cl/cl, and Pl/pl caniniform, round,
large, deeply pocketed posteriorly. Dentary with convex or laterally compressed, often heavy. Pl/pl single or
downward ventral border; angular "hook" prominent. double rooted. P2/p2 lost in derived species. The p3 with
Metapodials elongate, slender, fused, and longer than basal two roots, in contrast to most Megatylopus. Cheek teeth
length 'of skull. higher crowned than in Oxydactylus, lower crowned than
Average length of m2: 27.0-32.0 mm. in Megatylopus. The ml-3 with anteroexternal styles in
Included species: H. stylodon only (known from localities some species. Premaxillaries form more V-shaped arc at
NB19C, NB21). front of palate than in Oxydactylus, and extend anterior to
Hesperocamelus sp. is also known from localities CC21B, I1. Maxillary fossa often pocketed with the maxillaries in-
?NB4, SB32B, ?CPl 10. flated above the pocket. Palate typically broad, and usually
with little orno constriction between C 1 and P 1, in contrast
Comments: Macdonald (1949) distinguished Hesperocam- to protolabines. Nasals strongly arched in derived species.
elus from Aepycamelus (= Alticamelus) primarily on the Angular process of dentary usually more prominent than
basis ofless robustness and smaller size, and included Alti- in Miolabis or Protolabis. Dentary with convex ventral
camelus alexandrae Davidson (1923) in Hesperocamelus. border. Neck long, with greatest elongation occurring in
However, "H." alexandrae differs from Hesperocamelus anterior cervical vertebrae. Limb elements slender, elon-
stylodon in having robust, nonreduced 13, Cl/cl, and gated; where known, femur and tibia elongated more than
Pl/pl, a wide rostrum, and anteroposteriorly unexpanded metapodials. Metapodials fused, longer than basal length
molars; as pointed out by Macdonald ( 1949), it differs from of skull; metacarpus longer than metatarsus. Distal artic-
H. stylodon also in slightly larger size and loss of Il-2. ular surfaces of proximal phalanges dorsally expanded,
With its retention of Il-2 (primitive) and anteroposterior more symmetrical than in Oxydactylus. Skeleton very large
expansion of the molars (derived), H. stylodon seems to and giraffelike in some species.
represent an evolutionary line distinct from "H." alexan- Average length of m2: 30.0-43.0 mm.
drae. The latter species shows similarities withAepycame- Included species: A. giraffinus (known from locality CP76);
lus elrodi in the length of the cervical vertebrae, the un- A. proceras(localities CPl 10, NP38C);A. robustus (locali-
expanded molars, the wide rostrum, and the robust 13, ties NB23C, SP2A, CP114A, CP114B, CPl 140, NP41B);
Cl/cl, and Pl/pl; accordingly, we have included alexan- A. stocki (localities NB23B, NP38E); A. bradyi (locality
drae withinAepycamelus. Because of conflicting published NB29); A. elrodi (locality NP41B); A. major (= Megatylo-
data, metacarpal length versus metatarsal length in Hespe- pus major) (localities GCllA, GCl lB, CPl 16A);
 J. G. Honey et al. 454

A. alexandrae (= Hesperocamelus alexandrae) (localities longurio"), SB18A, SB34A, SB35, SB37, SB48, SB49,
CC17E, CC22B, NB6E). SB58B, SB60, SB65, SPlB, SPIC, SPlD, SPlF, SP3A,
Aepycamelus sp. is also known from localities GC4B, SP4B, CP90A, CP94, CP97, CP114D, CPI 16B, ?CP116C,
GC4C, GC4E, GC6B, GCllC, GC12II, CC17B, CC17C, CP117A, CP118, CP121, PN3C, PN12, ?PN14, PN23A,
CC17D, CC18, NB3D, NB6B, NB6C, NB21, NB28, PN23C.
SB32A, SB32B, SB32D, SB32F, ?SPlA, SPlB, CP45E,
CP49C, CP56, CP74B, CP75B, CP87B, CP114C, CP116B, Comments: As discussed by Webb (1974), Tanupolama is a
NP38C, ?NC3B. junior synonym of Hemiauchenia. Webb, MacFadden, and
Baskin (1981) suggested that a new generic name may be
Blancocamelus Dalquest, 1975
appropriate for "Hemiauchenia" minima from the Mixon
Bone Bed (locality GCl lB) and Love (locality GCl lA)
Type species: Blancocamelus meadei Dalquest, 1975. local faunas.
Type specimen: TMM 31179-20.
Characteristics: Known only from postcrania. Limbs, espe- Palaeolama P. Gervais, 1867
cially distal elements, extremely elongated and slender as
in Hemiauchenia and Aepycamelus giraffinus, but more Type species: Palaeolama weddelli (P. Gervais, 1855) (=Au-
than one and a half times as long as in Hemiauchenia. chenia weddelli).
Included species: B. meadei only, known from locality SP5 Type specimen: Metacarpus without published number in
only. Museum National d'Histoire naturelle, Paris.
Blancocamelus sp. is also questionably known from Characteristics: Dental formula Il/3, Cl/I, P3/2 or 3, M3/3.
locality SB50. Cheek teeth lowest crowned among the Larnini. I3 canini-
form. P3 with variable lingual crescent stronger than in
Hemiauchenia H. Gervais and Ameghino, 1880 Hemiauchenia; p3 reduced. P4 with bilobate ectoloph; p4
(synonym: Tanupolama) with complex infolding. Ml-3 with strong external styles
and anterior ribs; ml-3 with anteroexternal style. Limbs
Type species: Hemiauchenia paradoxa H. Gervais and stocky. Metapodials shortened and robust.
Ameghino, 1880. Average length of m2: 19.0-23.0 mm.
Type specimen: Museo de la Plata 36. Included species: P. mirifica (Pleistocene); P. guanajuatensis
Characteristics: Dental formula Il/3, Cl/1, P3/2-3, M3/3. (known from locality SB58B). P. weddelli and P. aequato-
Less hypsodont cheek teeth than in any other !amine except rialis are restricted to South America.
Palaeolama. Cl/cl recurved and laterally compressed, cl
reduced. Diastemal crest on mandible sharp. Post-pl di-
Alforjas Harrison, 1979
astema short in H. vera, progressively longer in H. blan-
coensis and H. macrocephala. The p3 present in H. vera, Type species: Alforjas taylori Harrison, 1979.
variably present with either one or two roots in later species. Type specimen: F:AM 40821.
Lower molars with weak anteroexternal styles. Rostrum Characteristics: Dental formula Il/3, Cl/1, P3/3, M3/3. Cheek
narrow and swollen above deep maxillary fossa with ridged teeth more hypsodont than any !amine but Camelops. I3
anterodorsal border. Nasals arched and retracted, but not and Cl/cl recurved and laterally compressed. The
so far as in Lama. Palatine notch sharply V shaped, ex- il-3 more broadly spatulate than in Lama or Hemiauche-
tending to posterior part of M2. Lacrimal vacuity not so nia. Pl/pl very small and single rooted. Ml-3 with ex-
reduced as in Lama. Limbs very long and slender, resem- ternal styles not so strong as in Palaeolama, and with
bling Blancocamelus in proportion, but of smaller size. much weaker ribs. Pl-P3 diastemal crest heavier and more
Metapodials fused and longer than basal length of skull. curved as in Camelops. Lower molars with anteroexter-
Proximal phalanx with W-shaped suspensory ligament scar nal styles weak or absent. Rostrum long and slender as
not extending onto shaft. in Camelops with little facial shortening. Maxillary fossa
Average length of m2: 23.0-35.0 mm. deeper than in Lama or Hemiauchenia. Nasals strongly
Included species: H. vera (known from localities GC13B, arched transversely. Lacrimal vacuity large. Palatine notch
NBlO, SB9, SB34C, SB58A, SP3B, CP116D, CP116E, extends only to posterior border of M3. Mandible deeper
CP116F, CP123C, CP123D, CP126); H. blancoensis (lo- than in Hemiauchenia or Palaeolama with long posteri-
calities GC14, GC17, SB14F, SB31C, SB50, SPlH, SP5, orly hooked coronoid process and strongly inflected angu-
CP80, CP117B, CP128C, CP131, PN4); H. macrocephala lar process. Limbs not as elongated as in Hemiauchenia
(locality GC 15); "H." minima (localities GCl lA, GC 1 lB, or as heavy as in Camelops or Palaeolama. Metapodials
GCllC). fused and shorter than basal length of skull. Proximal pha-
Hemiauchenia sp. is also known from localities GC 1 lB, lanx with uneven W-shaped suspensory ligament scar as in
?GC13A, ?CC40 ("Procamelus edensis"), ?CC41, ?CC53, Lama and Hemiauchenia, but extending farther down the
NB9, NB12B, NB13B, NB13C, NB31, NB32B, NB34, shaft.
SBlO, SB13, SB14B, SB15, SB16, ?SB17 ("Procamelus Average length of m2: 28.0-39.0 mm.
 Camelidae 455

Included species: A. taylori only (known from localities to Pliauchenia on the basis of lack of p2, most notably "P."
NB32A, SP3B, CP123D). magnifontis from Big Spring Canyon (locality CP90A)
Alforjas sp. is also known from localities SB9, SBlO, (Gregory, 1942). Based on a partial skeleton, "P." mag-
SB33II, ?SB58A, SPlD, CP78, ?CP116C, CP116D. nifontis supposedly showed for the first time associated
upper and lower dentitions of Pliauchenia. Pliauchenia
Camelops Leidy, 1854 has also generally been thought to be in the direct line of
Type species: Came/ops kansanus Leidy, 1854. ancestry of some later camelids, most notably the !amines.
Type specimen: USNM 154. There are problems, however, with the concept of Pli-
auchenia. The first concerns the generic diagnosis and the
Characteristics: Dental formula I 1/3, C 1/1, P2/1, M3/3. Cheek
teeth very hypsodont. Il-2 lost. I3 recurved and canini- fragmentary nature of the type specimen, and authors have
form, larger than Cl. Cl/cl very reduced and laterally disagreed on whether or not an alveolus for a small p2 is
compressed. The cl-p4 diastemal crest very sharp. Pl in fact present on the type (Matthew, 1901; Matthew and
and P2/p2 lost; pl rarely present. P3 and p4 reduced. Cook, 1909; Gregory, 1942). In fact, the jaw is brokenin the
P4 quadrate. Molars slender in relation to length. Ml-3 p2 position, making it uncertain whether or not the tooth
with reduced external styles. Lower molars with anteroex- was present. Even if the p2 was proven to be absent, phylo-
ternal stylids stronger than in Alforjas, weaker than in genetic loss of p2 by itself is not a good generic criterion,
Lama, Palaeolama, or Hemiauchenia. Premaxilla robust
as this loss occurs in several different camelid lineages
(stenomylines, miolabines, protolabines, and camelines).
as in Alforjas, but in contrast to Lama and Hemiauche-
nia. Rostrum long and slender with little facial shortening. Secondly, comparisons are usually made with the better
Maxillary fossa deep, but not pocketed. Lacrimal vacu- preserved "Pliauchenia" magnifontis instead of with the
ity very large. Nasals very strongly arched transversely. type. "P." magnifontis, however, differs from P. humphre-
siana in having a single-rooted, trenchant p I not unlike
Mandible deeper than in other !amines due to increased
hypsodonty. Angular process small. Body size largest of the pl of Procamelus. Although the pl is missing in P.
humphresiana, its alveolus indicates a double-rooted tooth.
!amines with possible exception of Blancocamelus. Limbs
long, but stocky. Metapodials robust and somewhat short- At least in some camelids (Protolabis and Procamelus), a
ened. Proximal phalanx with raised suspensory ligament single- versus a double-rooted p 1 is of generic significance
scar extending almost to center of shaft. (Honey and Taylor, 1978).
Average length of m2: 31.0-44.0 mm. It is not certain that any of the later described species of
Pliauchenia belong to the same genus as the type specimen.
Included species: C. kansanus (Pleistocene); C. traviswhitei
Galusha and Blick (1971) considered P. humphresiana to
(known from localities SB 18A, SB18B, SP5); C. sulcatus
(Pleistocene); C. huerfanensis (Pleistocene); C. minidokae be from the Skull Ridge Member of the Miocene Tesuque
(Pleistocene), C. hesternus (locality SB37 [possibly Irv- Formation (locality SB32B); of the camels we list from
ingtonian]). that member, P. humphresiana seems most closely to re-
semble Protolabis. However, detailed comparisons need
Camelops sp. is also known from localities ?CC53,
NB13B, NB13C, NB36, SB13, SB14A, SB14B, SB14C, to be made with all Skull Ridge camelids to prove any
SB14D, SB14E, SB14F, SB15, SB16, SB18C, SB31C, synonymies, if that is even possible considering the poor
preservation of the type specimen of Pliauchenia.
SB37, SB50, SB65, SPlH, CP95 (C. "vacondae"),
CP117B, CPl 18, CP121, CP130A, CP132B,PN4, PN23A.

CAMELIN!
Pliauchenia Cope, 1875
Characteristics: Dental formula 12-0/3, Cl/1, P4-2/4-2,
Type species: Pliauchenia humphresiana Cope, 1875. M3/3. Medium to giant camels more closely related to
Type specimen: USNM 2630. Asian Camelus than to New World Lama. The pl lost in
Included species: P. humphresiana (known from locality Titanotylopus. Except in Procamelus, molars with more
SB32B); "P." magnifontis (localities CP90A, CP90B, strongly developed styles and ribs and more square oc-
CP116B). clusal outline than in !amines. Nasals flattened in cross
section. Diastemal crest on mandible reduced and rounded.
Comments: Pliauchenia was named for a fragmentary ramus
Angular process moderate in Procamelus, becoming large
containing the cl and pl alveoli, the p3-4 roots, a worn,
and strongly inflected in remaining Camelini. Maxillary
broken ml, and the m2 roots, obtained from the "Santa
fossa pocketed primitively (Procamelus), secondarily re-
Fe marls." Cope diagnosed Pliauchenia as a camel hav-
duced in remaining genera. Lambdoidal and sagittal crests
ing lost the p2, thereby being interµiediate between Pro-
far more prominent than in other Camelinae.
camelus with four lower premolars and the extant Camelus
and Lama ("Auchenia") with two and one lower pre-
Procamelus Leidy, 1858
molars, respectively (Cope, 1875, 1877). Since Cape's
description, camelids have at various times been referred Type species: Procamelus occidentalis Leidy, 1858.
J. G. Honey et al. 456

Type specimen: USNM 797. Plate XIV, Figure 15) could have come from the same
Characteristics: Dental formula I2-l/3, Cl/I, P4/4, M3/3; species. These premolars are smaller than in typical pro-
I2 present in primitive species. I3 and Cl/cl laterally com- camelines, lower crowned, and morphologically similar to
pressed and bladelike in females; large, conical, and re- the premolars of protolabines. In addition, the p 1 referred
curved in males. Pl/pl single rooted, caniniform, often as by Leidy to P. gracilis is double rooted, a condition un-
large as I3; pl bladelike and recurved. Cheek teeth longer known in Procamelus but present in many protolabines.
and more hypsodont than in protolabines. The p2 little re- In fauna! lists reporting USNM material from Nebraska
duced. The p3 and p4 with posterior cusps usually wider Barstovian localities, including sites along the Niobrara
than middle cusps. P3 internal crescent incomplete. Lower River Valley from which the type of P. gracilis was col-
molars usually without anteroextemal stylids. Mandibular lected, Voorhies (1990b) reassigned P. gracilis to Pro-
symphysis long and forms a shallow angle with horizontal tolabis; herein we accept Voorhies's reassignment. Pro-
body of ramus. As in Aepycamelus but in contrast to Oxy- camelus leptognathus is a small slender-jawed form from
dactylus, the ventral border of the ramus is smoothly con- the Clarendon fauna and has never been illustrated or thor-
vex, without a slight concavity beneath the ascending ra- oughly described. Webb ( 1969) suggested a possible rela-
mus. Angular process of dentary is well developed (strong tion to "Procamelus" gracilis and "Procamelus" coartatus
"hook"), knoblike, and mesially inflected. Coronoid pro- (now Protolabis coartatus); it also may be a small !amine.
cess long. Palate wide as in most Aepycamelus, and with The premolar region of the lower jaw is not preserved. This
greatest constriction beginning posteriorly to species is almost certainly not Procamelus as that genus
Pl. Maxillary fossa usually deeply depressed and pocket- is now understood, and it is best considered indeterminate
like, with maxillaries inflated above fossa. Nasals until more adequately described.
transversely flattened. Postglenoid foramen large. Paroc-
cipital process usually projects below auditory bulla. Pro- Megatylopus Matthew and Cook, 1909
minent lambdoidal and sagittal crests. Metapodials fused Type species: Megatylopus gigas Matthew and Cook, 1909.
with metacarpals longer than metatarsals. Distal articular Type specimen: AMNH 14071.
surfaces of proximal phalanges dorsally expanded and tend Characteristics: Dental formula Il/3, Cl/l, P3/4-3, M3/3.
to be symmetrical. Cheek teeth more hypsodont than in Titanotylopus, but
Average length of m2: 24.0-35.0 mm. much less so than in Megacamelus, Gigantocamelus, and
Included species: P. occidentalis (known from localities Camelus. I3 and Cl/cl large and rounded, but smaller
GC6B, CP114A, CPI 14B, CPI 15A, NP41B); P. lepto- than in other Camelini except Procamelus. Pl/pl present.
colon (localities CP76, CPI 10); P. grandis (localities CA9, P3/p3 reduced. Nasals flattened in cross section. Limbs
CAlO, GC6B, GCI IA, SP2A, CP56, CP90A, CP90B, long without distal shortening typical of Gigantocamelus
CP114C, CP114D, CP116A, CP116B, PNlO). and Camelus. Metapodials longer than basal length of
Procamelus sp. is also known from localities GC6B, skull.
?GCllC, GC12II, CC17G, CC17H, ?NB6E, ?NB7C, Average length of m2: 39.0-50.0 mm.
NB28, ?SB7, ?SB32B, SB32D, SB33II, SPlA, SPlB, Included species: M. gigas (known from localities CP78,
SP3A, CP87B, CPI 14C, CPI 15C, CP123A, CP126, NPl 1, CP115D, CPI 16C, CPI 16E, CP123D, CP126, NP45); M.
NP42, PNl 1, ?PN13. matthewi (localities NB 10, SB9, SB34C, SB58B, SPlD,
SP3B); M.(?) cochrani (localities CPll 7 A, CP128B,
Comments: Since the naming of the genus by Leidy in 1858, PN3C); M.? primaevus (localities GC6B, CPI 16B,
numerous species have been referred to Procamelus. The NP38G).
genus was briefly reviewed by Gregory (1942) and Webb Megatylopus sp. is also known from localities GCl lB,
(1969), and studies by Patton (1967, 1969), Webb (1969), ?CC17H, ?CC41, ?NB7C,NB9, ?NB23C,NB27B, ?NB31,
Honey and Taylor (1978), and Voorhies (1986) have re- NB36, SB 10, ?SB 17 ("Procamelus coconinensis"),
duced the number of nominal species of Procamelus. Ad- SB34A, SB34B, ?SB48, ?SB57, ?SB60, SPlA, SPlB,
ditional species should be removed from Procamelus. Pro- SPIC, SPIH, SP3A, SP4B, ?CP90A, ?CP114C,
camelus inaequidens Matthew (in Cope and Matthew, CPI 14D, CPI 15B, CPI 15C, CPI 16A, CPI 16D, CP123D,
1915, Plate CLIV), figured but never described, is here ?NP5B, ?PN3B, PNlO, PNl 1, PN12, PNl3, PN15, PN23A.
referred to Protolabis based on the following features:
small braincase, constricted rostrum, double-rooted Pl/p 1, Comments: The taxonomic status of Megatylopus primae-
more upright mandibular symphysis, subdued angular pro- vus is uncertain. Described by Patton (1969) as the most
cess, and low coronoid process. Procamelus gracilis Leidy primitive species of Megatylopus, due to retaining a small,
(1858) includes "several isolated teeth and small fragments possibly nonfunctional p2 and having a larger and more
of jaws with single teeth" (Leidy, 1869, p. 155). It is not complex p3 than other species of the genus, M. primae-
certain whether all specimens came from the same species vus was considered "similar and probably closely related"
or individual, but at least the illustrated p2-4 (Leidy, 1869, (p. 167) to the slightly more advanced Megatylopus major.
 Camelidae 457

With the reassignment of M. majortoAepycamelus (Webb, Included species: M. merriami only (known from localities
MacFadden, and Baskin, 1981; Harrison, 1985), it might CC40, SBll, CPl16F).
be logical also to reassign M. primaevus to Aepycamelus,
thus returning to a diagnosis of Megatylopus, which in- Gigantocamelus Barbour and Schultz, 1939
cludes the loss of p2. However, in reporting the occurrence Type species: Gigantocamelus spatula (Cope, 1893) (= Pli-
of Megatylopus cf. primaevus in Nebraska, Leite (1990) auchenia spatula).
noted that Megatylopus has shorter limbs and more hyp- Type specimen: not numbered by Cope or Meade.
sodont teeth than Aepycamelus. Considering that the sys- Characteristics: Dental formula Il-0/3, Cl/l, P3/3, M3/3.
tematics of both Megatylopus (Voorhies and Comer, 1986) Cheek teeth more hypsodont than in Megacamelus or Meg-
and Aepycamelus are not well understood, we prefer to atylopus. I3 vestigial or lost. Cl/cl rounded and extremely
questionably retain the species primaevus in Megatylopus large. The il-3 spatulate and transversely arrayed. Mandi-
until more definitive studies are made. bular symphysis barely extends beyond widely splayed
The relationship of Megatylopus to Paracamelus, the cl. Chin blunt in contrast to Titanotylopus. The i3-cl di-
common Old World Pliocene camelid, is controversial, astema very short. Pl/pl present, but variable in female
and Megatylopus has sometimes been considered a junior rami. Ml-3 with very strong external styles. Premaxilla
synonym of Paracamelus (Macdonald, 1959). Although heavy and robust. Nasals transversely flattened. Lacrimal
this synonymy has generally been rejected in recent years vacuity reduced or absent. Maxillary fossa reduced. Dis-
(Webb, 1965), Voorhies and Comer (1986) suggested that tal limb elements shortened as in Camelus. Metapodials
further investigation of the problem is warranted. shorter than basal length of skull. Proximal phalanx very
short and stocky.
Titanotylopus Barbour and Schultz, 1934 Average length of m2: 44.0-60.0 mm.
Included species: G. spatula only (known from localities SP5,
Type species: Titanotylopus nebraskensis Barbour and CP97, CP117A, CPl17B, CPl18, CP121, CP128B).
Schultz, 1934.
Gigantocamelus sp. is also known from localities SB38,
Type specimen: USNM 2606. SB49, CP80, CP94, CP131, PN23C.
Characteristics: Dental formula I?/3, C?/l, P?/2, M?/3. Skull
and upper dentition unknown. Cheek teeth most brachy-
dont of tribe. The i3-c 1 diastema much longer than in Gi- INDETERMINATE CAMELIDS
gantocamelus. The cl large and rounded, but not splayed. Fragmentary remains ascribed to camelids are also known from lo-
The pl absent; p3 double rooted and more reduced than in calities WM21A, WM21B, GC9C, GC27, GC28, CC9B2, CC23,
M egatylopus. CC25A, CC25B, CC26B, CC32A, CC35A, CC36, CC38, CC39B,
Average length of m2: broken in type. Estimated length of NB7D, NBll, NB18, NB27A, SB12, SB19, SB59, SB63, SB64,
m3 in type: 65.0 mm. CP123E,CP130B,NP5A,NP25C,N P32B,NP34B,NP34D,NP40A,
Included species: T. nebraskensis only (known from localities PN6F, PN8B, PN9B. Undoubtedly, indeterminate camelid fossils
NB13B, NB13C, SBl8A, ?CP96, CP121). occur at many more localities than mentioned herein.

Megacamelus Frick, 1929

BIOLOGY AND EVOLUTIONARY PATTERNS
Type species: Megacamelus merriami (Frick, 1921) (= Pli-
auchenia merriami). Camelids were a highly successful group in North America in the
Type specimen: UCMP 23483. Cenozoic; in this chapter we record 95 species distributed among 36
Characteristics: Dental formula Il/3, Cl/1, P3/3, M3/3. Che- genera. Although future study will likely demonstrate synonymies,
ek teeth less hypsodont than in Gigantocamelus. I3 large we expect that the total number of genera and species of camelids
and caniniform in contrast to Gigantocamelus. Cl/cl large will increase through the description of new forms.
and rounded, but only slightly splayed. The il-3 spatulate In the broad view, there were four camelid radiations in North
but not transversely arrayed. The i3-cl diastema longer America (Figures 30.5 and 30.6). The first was the Chadronian thro-
than in Gigantocamelus, shorter than in Titanotylopus. ugh earliest Arikareean (late Eocene through mid-Oligocene) evo-
Pl/pl large and caniniform. Ml-3 with strong external lution of the archaic camelids Poebrotherium and Paratylopus. The
styles. Premaxilla very robust, but not so massive as in second was the Whitneyan through Arikareean (late Oligocene to
Gigantocamelus. Nasals transversely flattened. Lacrimal early Miocene) radiation of the stenomylines; during the early Arika-
vacuity variable. Mandibular symphysis extends beyond reean stenomylines were dominant, and nonstenomylines (Gen-
cl but not so far as in Titanotylopus. Distal limb ele- tilicamelus, Poebrotherium) were rare. Although Blickomylus
ments more shortened than in Megatylopus, less so than in (Fig. 30.4) and Rakomylus evolved later, the stenomylines never
Gigantocamelus. again achieved the success of the early Arikareean, and they went
Average length of m2: 48.0-55.0 mm. extinct in the early Barstovian (middle Miocene).
 J. G. Honey et al. 458

PLEIS.

! to
) Euras.

Oxydactylu•

EVOWTION OF THE CAMELS

Figure 30.5. Evolution of the camels

(Illustration by Clifford R. Prothero.)

Probably the most striking event in camelid history is the ap- have been the ancestral taxon for both the Lamini and Camelini;
pearance of"higher camelids" (those with metastylids on the lower early Procamelus shows a striking resemblance to Aepycamelus,
molars) in the late Oligocene and early Miocene (early and late and later Aepycamelus acquire certain lamine characteristics. By
Arikareean), marking the third major radiation. In the late Arika- the latest Barstovian at the youngest, and possibly earlier in the
reean the number of camelid genera more than doubled. The Arika- Barstovian, the camel-llama split had occurred. With the diversifi-
reean appearance of higher camelids, however, was not a single cation of the Camelinae in the Barstovian and later, the more archaic
event. Gentilicamelus, from the John Day Formation, was probably groups such as the protolabines and miolabines decline to their late
found in the Arikareean Turtle Cove Member (locality PN6D). In Miocene extinction in the later Hemphillian (based on the occur-
Nebraska the earliest higher camelid, Oxydactylus (Fig. 30.lA), is rence of an undescribed protolabine genus in the Wickieup local
found in the late Arikareean Harrison Formation (locality CP103, fauna [locality SB9] [MacFadden, Johnson, and Opdyke, 1979]),
Miocene); but Tanymykter and Michenia do not occur until the over- although they were apparently last abundant only in the Claren-
lying Marsland Formation (locality CP104B, Miocene); cf. Proto- donian (late Miocene). Total known generic diversity in the late
labis also does not occur until late late Arikareean. Gentilicamelus Blancan (late Pliocene) was 5 genera (questionably 6), down from
stembergi would appear to be a fairly good structural ancestor for highs of 11 to 13 genera in the Hemingfordian and Barstovian. Only
Priscocamelus, Oxydactylus, and Tanymykter. Gentilicamelus was 3 genera survived into the Rancholabrean (late Pleistocene).
not derived from the stenomylines with their reduced mesostyles, The appearance of camelids in the late Eocene to early Oligocene
but may have come from Paratylopus. coincided with a drying trend that produced the first areas of true
A fourth major radiation of camelids began in the late Heming- savanna in North America (Webb, 1977). The limb morphology of
fordian and early Barstovian (late early to early middle Miocene) Poebrotherium suggests that came lids were adapted to a cursorial ex-
with the evolution of the Camelinae. Aepycamelus sensu la to may istence in more open habitat by the early Oligocene; Poebrotherium
 Camelidae 459

EOCENE OLIGOCENE MIOCENE PLIOCENE

m ,... m ,... m ,...
I
m
...m
!!' ~
cm
"'• :-II
II m I IPoebrodon I I I I I I I I I I I I II I I I I
ci»
ag 1 , 1 Poebrotherium
I I I I I I I I I I
~~ ,? Paratylopus "Paratylopus" cameloides
m:i:
:" 'P I I I I I I ..- I I I I I
II 11 Paralabis
m :i:
:t (I)
c8' .g
!!1 ::r
~
m
.. I I
I I
r---i I I I I I I I I
1---+-P_s_e_,u,..do_l_a_b1_·s-+--i1 I I I
1 1 I I I
I
I
;·:: z Miotylopus
::s ;· 0
1 I
I I

~~ r[ I I I
I I I
-"'Q I I Stenomylus
·,, I I I I IB!ikomylus I I
0
;;;
z 3
§: I 1Rakomylus

[.. I I
; I I I
"'0
Cl.

"';;;· ~~ -· 1"Poebrotherium" franki I I I I I I I I

-"'m I I Aguascalientia I I I
..·,, G>O
c ::0
ca Floridatragulus
I
.
m Z)>
)> '
1 1
Gentilicamelus 1 '"'_,.1_,.., --+l-+-1--1 I
~
<
m NT -:-_, - N;t°hokemas I I I I I
;;;·
-"'J: I I I I I I I I I
Paramiolabis
3
ii
:s: I IMiolabis 1 I I I
J: 0r
CD
3
:;·
)> I lcuyama~lus I I
m
ca0 z I IMiolabine sp. I I I
a.
;;;·
)>
m 1Nothotylopus I I
I I
-"'

'~
~ I I ITanymykter I I I
II
~
["" I I 1Protolabis I I I I
..,..
:!.

~ Michenia ,
ii. m I ' 'Priscocamelus
1
I
I
'
I I I
"' z I i--__,.l....,IOxydactylus I I I I I I
~
)>
II
m - - ..,
::;: I I IAustra/ocame°Jusl I I I I
=.
.
:;
CD
'< - I I I I I I I I
"Aepycamelus "priscus
I II

-"' I I Hes;;;;;,came/us I .j
I I
~
- - --- - -
ii
0 I I I'" I I IAepy~a'metus I
~ I I I I I
I I
I I
;;; I I I IBlancocamelus
-"'0 I -
::r I I I I I I IHemiauchenia I
II r
0 Ill
I I I I I I 1Palaeolama 1
.
0
::r
Cl.
)>
:s:
3
~:
I
11 IA~! I
a m
c I I I I I
"';;;· z I I Ic~"!e~~ps
-"'0 )> I I I I I P/iauche!;;;t-
c: m , __,,......,, I I I I I
, .......
...--.-........
i1 :: I I I I I Procamelus
0
c:
" I I I I I iMegatytop~s I
::r
CD
"'"' I ITitanotylopus I'" -
?
.
CD I I I I I--'--
I 1 Megacam~s I
I I I I
Gigantocametus
- .. I I I I I I I I I I I I I I I I 111111--r-
Figure 30.6. Temporal ranges of North American camelid genera.
 J. G. Honey et al. 460

has very elongate axial metapodials, reduced lateral metapodials, Gulf Coastal Plain. Floridatraguline habitats have been interpreted
and sharp ungual phalanges. Unlike the living camelids, which are as savanna or open woodland, where the fioridatragulines were prob-
digitigrade and have padded feet, Poebrotherium was unguligrade ably selective browsers as evidenced by their brachydont, selenodont
and its ungual phalanges probably bore deerlike hooves (Webb, dentitions and greatly elongated, slender rostra. The association of
1972). The padded feet and digitigrade stance of living camelids Heloderma, the gila monster, with fioridatragulines (Stevens, 1977;
is a secondary condition probably associated with the development but also see Estes, 1963) certainly suggests that their environment
of a pacing mode of locomotion; correlative changes include fused was subtropical, but such environments then also existed elsewhere
metapodials and fore- and hindlimbs more equal in length (Webb, (Dorf, 1959; Webb, 1977). In a fauna! list, Voorhies (1990b) re-
1972). The phalanx morphology of the earliest camelines, Pro- ported Floridatragulus sp. from the late Barstovian Valentine Rail-
camelus and Aepycamelus, and that of the protolabines Protolabis way Quarries of Nebraska; this is the only reported occurrence of
and Michenia, indicate that these camelids were digitigrade; a tran- a fioridatraguline north of Texas. Why fioridatragulines rarely ven-
sitional morphology is shown by Oxydactylus and Tanymykter. It is tured northward into other subtropical savanna parklands of North
possible that this transition from unguligrade to digitigrade stance America is unresolved. The distribution of fioridatragulines may
occurred more than once within the Camelidae. have been determined by a subtle environmental factor difficult to
Miocene and later camelids evolved very different adaptations. detect from the fossil record, such as the distribution of one or more
Early and middle Miocene stenomylines were the most hypsodont sensitive plant food species.
of all camelids. Late Arikareean Stenomylus, which is found nearly
everywhere except Florida, had gazellelike proportions and an un-
guligrade stance, suggested by its slender, pointed ungual phalanges ACKNOWLEDGMENTS
(Scott, 1937) and its proximal phalanges showing only incipient
proximodorsal expansion of the distal articular surface; Stenomy- We thank Richard H. Tedford, American Museum of Natural His-
lus probably occupied open habitats. The later stenomylines were tory, for permission to study specimens and report unpublished oc-
more restricted in distribution (Wyoming, Colorado, New Mex- currences of some camelids in the American Museum collections.
ico, Mexico), and in the United States may have been adapted This chapter was reviewed by Walter Dalquest and Dave Webb.
to more arid habitats (Webb, 1977; Munthe, 1979). The "giraffe-
camel" Aepycamelus, as evidenced by its extremely long limbs and
cervical vertebrae, and only moderately high crowned teeth, was pre- REFERENCES
sumably a high-level savanna browser. Aepycamelus probably was
the New World ecomorph of the Old World giraffe. Aepycamelus Barbour, E. H., & Schultz, C. B. (1939). A new giant camel, Gigantocamelus
disappeared in the Hemphillian (late late Miocene), probably as a fricki, gen. et sp. nov. Bulletin of the University of Nebraska State
result of the restriction of woodland savanna and spread of steppe Museum, 2, 17-26.
(1934). A new giant camel, Titanoetylopus nebraskensis, gen. et sp. nov.
(Webb, 1977). Its likely predecessor, Oxydactylus (Fig. 30.lA),
Bulletin of the University of Nebraska State Museum, 1, 291-94.
shows the beginnings of these limb and vertebral specializations; Barghoom, S. F. (1985). Magnetic Polarity Stratigraphy of the Tesuque For-
a gerenuklike habit of standing on its hind legs to reach the high- mation, Santa Fe Group in the Espanola Valley, New Mexico, with a
est browse has been suggested for Oxydactylus (Janis, 1982). A Taxonomic Review of the Fossil Camels. Unpublished Ph.D Thesis.
study that included specimens of the camelids Hemiauchenia, Al- New York: Columbia University.
forjas, Palaeolama, Camelops, Procamelus, and Megatylopus indi- Black, C. C. (1978). Paleontology and geology of the Badwater Creek area,
central Wyoming, Part 14: The artiodactyls. Annals of Carnegie Mu-
cated that they also were browsers or at most intermediate feeders;
seum, 47, 223-59.
none were grazers (Dompierre and Churcher, 1996). Cope, E. D. (1873). [On Menotherium lemurinum, Hypisodus minimus, Hy-
At the opposite extreme from the "giraffe-camels" are the short- pertragulus calcaratus, H. tricostatus, Protohippus, and Procamelus
legged miolabines and derived protolabines. The short legs (seen heterodontus]. Proceedings of the Academy of Natural Sciences of
particularly in the metapodials) is a secondary, derived condition Philadelphia, 25, 419-20.
(1875). On some new fossil Ungulata. Proceedings of the Academy of
of later protolabines and probably also of the miolabines; in proto- Natural Sciences of Philadelphia, 1875, 258-62.
labines, for example, the metapodials became progressively shorter (1876). On a new genus of Camelidae. Proceedings of the Academy of
and relatively more stocky overtime. Janis (1982) suggested that the Natural Sciences of Philadelphia, 28, 144-7.
protolabines were ground-level feeders. The extremely constricted (1877). Report upon the extinct Vertebrata obtained in New Mexico by
parties of the expedition of 1874. In Report upon United States Geo-
rostrum of derived Protolabis was thought to indicate highly de-
graphical Surveys West of the One Hundredth Meridian, in charge of
veloped labial and buccal musculature (Honey and Taylor, 1978). First Lieut. Geo. M. Wheeler, Paleontology, 4, 1-370. Washington,
In conjunction with the relatively short metapodials, the constricted D.C.: U.S. Government Printing Office.
rostrum may also indicate a feeding behavior pattern of selective (1879). Observations on the faunae of the Miocene Tertiaries of Oregon.
low-level browsing and grazing. Bulletin of the United States Geological and Geographical Survey
Floridatragulines (Fig. 30. lB) are reported primarily from rela- of the Territories. F.V. Hayden, U.S. Geologist in Charge, 5, 55-69.
(1886). The phylogeny of the Camelidae. The American Naturalist, 20,
tively low latitudes. Their most primitive and/or earliest occurrences
611-25.
appear to have been in and adjacent to Mexico (Trans-Pecos Texas), (1893). A preliminary report on the vertebrate paleontology of the Llano
but during the Hemingfordian the group spread eastward along the Estacada. Geological Survey of Texas, 4th Annual Report, 137 pp.
 Camelidae 461

Cope, E. D., & Matthew, W. D. (1915). Hitherto unpublished plates of Ter- North America. Carnegie Institute of Washington Publication, 390,
tiary Mammalia and Permian Vertebrata. American Museum of Nat- 1-1074.
ural History Monograph Series Number 2, plates I-CLIV. Honey, J. G., & Taylor, B. E. (1978). A generic revision of the Protolabi-
Dalquest, W.W. (1975). Vertebrate fossils from the Blanco local fauna of dini (Mammalia, Camelidae), with a description of two new proto-
Texas. Occasional Papers, The Museum, Texas Tech University, 30, labidines. Bulletin of the American Museum ofNatural History, 161,
1-52. 371-425.
Dalquest, W.W., & Mooser, 0. (1974). Miocene vertebrates from Aguas- Janis, C. (1982). Evolution of horns in ungulates: ecology and paleoecology.
calientes, Mexico. The Pearce-Sellards Series, 21, 1-10. Biological Reviews, 57, 261-318.
Davidson, P. ( 1923). Alticamelus alexandrae, a new camel from the Barstow Kelly, T. S. (1992). New middle Miocene camels from the Caliente Forma-
upper Miocene of the Mohave Desert. University of California Pub- tion, Cuyama Valley badlands, California. PaleoBios, 13, 1-22.
lications, Bulletin of the Department of Geological Sciences, 14, Kurten, B., & Anderson, E. ( 1980). Pleistocene Mammals ofNorth America.
397-408. New York: Columbia University Press.
Dompierre, H., & Churcher, C. S. (1996). Premaxillary shape as an indicator Leidy, J. ( 1847). On a new genus and species of fossil Ruminantia. Proceed-
of the diet of seven extinct late Cenozoic New world camels. Journal ings of the Academy of Natural Sciences of Philadelphia, 3, 322-26.
of Vertebrate Paleontology, 16, 141-8. (1854). Description of a fossil apparently indicating an extinct species
Dorf, E. (1959). Climatic changes of the past and present. Contributions from of the camel tribe. Proceedings of the Academy of Natural Sciences
the Museum of Paleontology, University of Michigan, 13, 181-210. of Philadelphia, 7, 172-73.
Estes, R. ( 1963 ). Early Miocene salamanders and lizards from Florida. Quar- (1858). Notice of remains of extinct Vertebrata, from the valley of the
terly Journal, Florida Academy of Science, 26, 234-56. Niobrara River, collected during the exploring expedition of 1857,
Frailey, D. (1978). An early Miocene (Arikareean) fauna from northcentral in Nebraska, under the command of Lieut. G. K. Warren, U.S. Top.
Florida (the SB-IA local fauna). Occasional Papers of the Museum Eng., by Dr. F. V. Hayden, geologist to the expedition. Proceedings
of Natural History, the University of Kansas, Lawrence, Kansas, 75, of the Academy of Natural Sciences of Philadelphia, IO, 20-9.
1-20. (1869). The extinct mammalian fauna of Dakota and Nebraska including
Frick, C. (1921). Extinct vertebrate faunas of the badlands of Bautista Creek an account of some allied forms from other localities, together with
and San Timoteo Canon, southern California. University of Cal- a synopsis of the mammalian remains of North America. Journal of
ifornia Publications, Bulletin of the Department of Geology, 12, the Academy ofNatural Sciences of Philadephia, 7, 1-472.
277-424. Leite, M. B. (1990). Stratigraphy and mammalian paleontology of the Ash
(1929). History of the earth. Childs Frick Tertiary-Quaternary explo- Hollow Formation (upper Miocene) on the north shore of Lake Mc-
rations. Natural History, 29, 106-18. Conaughy, Keith County, Nebraska. Contributions to Geology, Uni-
( 1937). Homed ruminants of North America. Bulletin of the American versity of Wyoming, 28, 1-29.
Museum of Natural History, 69, 1-669. Loomis, F. B. (1911). The camels of the Harrison beds, with three new
Frick, C., & Taylor, B. E. (1968). A generic review of the stenomyline camels. species. American Journal of Science, series 4, 65-70.
American Museum Novitates, 2353, 1-51. (1936). The skeleton of a new fossil camel from Wyoming. University
( 1971 ). Michenia, a new protolabine (Mammalia, Camelidae) and a brief of Wyoming Publications, 2, 59-64.
review of the early taxonomic history of the genus Protolabis. Amer- Lull, R. S. (1921). New camels in the Marsh collection. American Journal
ican Museum Novitates, 2444, 1-24. of Science, series 5, 392-404.
Galusha, T., & Blick, J.C. (1971). Stratigraphy of the Santa Fe Group, New Macdonald, J. R. (1949). A new Clarendonian fauna from northeastern
Mexico. Bulletin of the American Museum of Natural History, 144, Nevada. University of California Publications, Bulletin of the De-
3-127. partment of Geological Sciences, 28, 173-94.
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North (1956). A new Clarendonian mammalian fauna from the Truckee For-
America. Smithsonian Miscellaneous Collections, 128, 1-96. mation of western Nevada. Journal of Paleontology, 30, 186-202.
Gervais, P. (1855). Recherches sur !es mammireres de l'Amerique (1959). The middle Pliocene mammalian fauna from Smiths Valley,
meridionale. In Zoologie de !'expedition de Castelnau, vol. I, Nevada. Journal of Paleontology, 33, 872-87.
pp. 1-63. Paris: Arthus Bertrand. MacFadden, B. J., Johnson, N. M., & Opdyke, N. D. (1979). Magnetic po-
(1867). Nouvelles Rescherches sur /es Animaux Vertebres Vivants et Fos- larity stratigraphy of the Mio-Pliocene mammal-bearing Big Sandy
siles. 7.oologie et Paleontologie Genera/es. l'e serie. Paris: Arthus Formation of western Arizona. Earth and Planetary Science Letters,
Bertrand. 44, 349-64.
Gervais, H., & Ameghino, F. (1880). Les mammiferesfossiles de l'Amerique Maglio, V. J. (1966). A revision of the fossil selendont artiodactyls from the
du Sud. Paris: Library F. Sary. middle Miocene Thomas Farm, Gilchrist County, Florida. Breviora,
Golz, D. J. (1976). Eocene Artiodactyla of southern California. Natural 255, 1-27.
History Museum of Los Angeles County, Science Bulletin, 26, 1-85. Marshall, L. G., Webb, S. D., Sepkoski, J. J., Jr., & Raup, D. M. (1982).
Gregory, J. T. (1942). Pliocene vertebrates from Big Spring Canyon, South Mammalian evolution and the great American interchange. Science,
Dakota. University of California Publications in Geological Sci- 215, 1351-7.
ences, 26, 307-446. Matthew, W. D. (1899). A provisional classification of the fresh-water Ter-
Harrison, J. A. (1979). Revision of the Camelinae (Artiodactyla, Tylopoda) tiary of the west. Bulletin of the American Museum of Natural His-
and description of the new genus Alforjas. The University of Kansas tory, 12, 19-75.
Paleontological Contributions, 95, 1-20. (1901). Fossil mammals from the Tertiary of northeastern Col-
( 1985). Giant camels from the Cenozoic of North America. Smithsonian orado. Memoirs of the American Museum of Natural History, I,
Contributions to Paleobiology, 57, 1-29. 355-447.
Hay, 0. P. (1899). On the names of certain North American fossil vertebrates. (1904). Notice of two new Oligocene camels. Bulletin of the American
Science, series 2, 9, 593-4. Museum of Natural History, 20, 211-5.
(1902). Bibliography and catalogue of the fossil Vertebrata of North (1910). On the skull ofApternodus and the skeleton of a new artiodactyl.
America. Bulletin of the United States Geological Survey, 179, Bulletin of the American Museum of Natural History, 28, 33-42.
5-868. (1918). Contributions to the Snake Creek fauna (with notes upon the
(1930). Second bibliography and catalogue of the fossil Vertebrata of Pleistocene of western Nebraska. American Museum expedition of
 J. G. Honey et al. 462

1916). Bulletin of the American Museum of Natural History, 38, (1945). The Mammalia of the Duchesne River Oligocene. Transactions
183-229. of the American Philosophical Society, 34, 209-53.
(1924 ). Third contribution to the Snake Creek Fauna. Bulletin of the Simpson, G. G. (1932). Miocene land mammals from Florida. Florida State
American Museum of Natural History, 50, 59-210. Geological Survey Bulletin, 10, 7-41.
( 1929). Reclassification of the artiodactyl families. Bulletin of the Geo- (1945). The principles of classification and a classification of mammals.
logical Society of America, 40, 403-8. Bulletin of the American Museum of Natural History, 85, 1-350.
(1932). New fossil mammals from the Snake Creek quarries. American Stevens, M. S. (1977). Further study ofCastolon local fauna (early Miocene),
Museum Novitates, 540, 1-8. Big Bend National Park, Texas. The Pearce-Sellards Series, Texas
(1934). A phylogenetic chart of the Artiodactyla. Journal ofMammalogy, Memorial Museum, 28, 1-69.
15, 207-9. Stevens, M. S., Stevens, J.B., & Dawson, M. R. (1969). New early Miocene
Matthew, W. D., & Cook, H. J. (1909). A Pliocene fauna from western formation and vertebrate local fauna, Big Bend National Park, Brew-
Nebraska. Bulletin of the American Museum of Natural History, 26, ster County, Texas. The Pearce-Sellards Series, 15, 3-52.
361-414. Stirton, R. A (1967). Relationships of the protoceratid artiodactyls, and
McKenna, M. C. (1966). Synopsis of Whitneyan and Arikareean camelid description of a new genus. University of California Publications in
phylogeny. Al!lerican Museum Novitates, 2253, 1-11. Geological Sciences, 72, 1-35.
Morales, J., Soria, D., & Aguirre, E. (1980). Camelido finimioceno en Venta Stock, C. (1935). Artiodactyla from the Sespe of the Las Posas Hills, Cal-
Del Moro, primera cita para Europa occidental. Estudios Geo/6 gicos, ifornia. Publications of the Carnegie Institute of Washington, 453,
36, 139-42. 119-25.
Moya-Sola, S., & Agusti, J. (1989). Bioevents and mammal successions in Voorhies, M. R. (1986). Aepycamelus robustus (Leidy) new combination,
the Spanish Miocene. In European Neogene Mammal Chronology, a giraffe camel from the medial and late Barstovian of Nebraska.
eds. E. H. Lindsay, V. Fahlbusch, & P. Mein, pp. 357-73. New York: In Proceedings of the Nebraska Academy of Sciences, Ninety-Sixth
Plenum. Annual Meeting, April 11-12, 1986, p. 53.
Munthe, J. (1979). Summary of Miocene vertebrate fossils of the Granite (l 990a). Vertebrate paleontology of the proposed Norden Reservoir area,
Mountains basin, central Wyoming. Contributions to Geology, Uni- Brown, Cherry, and Keya Paha counties, Nebraska. Technical Report
versity of Wyoming, 18, 33-46. 82-09, Division of Archeological Research, University of Nebraska,
Olsen, S. J. (1962). The Thomas Farm fossil quarry. Quarterly Journal, Lincoln, l-A593.
Florida Academy of Science, 25, 142-6. (1990b). Vertebrate biostratigraphy of the Ogallala Group in Nebraska.
Patton, T. H. (1966[67]). Revision of the selenodont artiodactyls from In Geologic Framework and Regional Hydrology: Upper Cenozoic
Thomas Farm. Quarterly Journal, Florida Academy of Science, 29, Blackwater Draw and Ogallala Formations, Great Plains, ed. T.
179-90. C. Gustavson, pp. 115-51. Austin: Bureau of Economic Geology,
(1967). Reevaluation of Hay's artiodactyl types from the Miocene of the University of Texas at Austin.
Texas Gulf Coastal Plain. The Texas Journal of Science, 10, 35-40. Voorhies, M. R., & Corner, R. G. (1986). Megatylopus? cochrani (Mam-
(1969). Miocene and Pliocene artiodactyls, Texas Gulf Coastal Plain. malia: Camelidae): a re-evaluation. Journal of Vertebrate Paleontol-
Bulletin of the Florida State Museum, 14, 115-227. ogy, 6, 65-75.
Patton, T. H., & Taylor, B. E. (1973). The Protoceratinae (Mammalia, Ty- Webb, S. D. (1965). The osteology of Came/ops. Bulletin of the Los Angeles
lopoda, Protoceratidae) and the systematics of the Protoceratidae. County Museum. Science, 1, 1-54.
Bulletin of the American Museum of Natural History, 150, 351-413. (1969). The Burge and Minnechaduza Clarendonian mammalian faunas
Peterson, 0. A (1904). Osteology of Oxydactylus. Annals of the Carnegie of north-central Nebraska. University of California Publications in
Museum, 2, 434-76. Geological Sciences, 78, 1-191.
(1906). The Miocene beds of western Nebraska and eastern Wyoming (1972). Locomotor evolution in camels.forma etfunctio, 5, 99-112.
and their vertebrate faunae. Annals of the Carnegie Museum, 4, (1974). Pleistocene llamas of Florida, with a brief review of the Lamini.
21-72. In Pleistocene Mammals of Florida, ed. S. D. Webb, pp. 170-213.
(1908). Description of the type specimen of Stenomylus gracilis Gainesville: University of Florida Press.
Peterson. Annals of the Carnegie Museum, 4, 286-300. (1977). A history of savanna vertebrates in the New World, Part I: North
Pickford, M., Morales, J., & Soria, D. (1993). First fossil camels from Eu- America. Annual Review of Ecology and Systematics, 8, 355-80.
rope. Nature, 365, 701. Webb, S. D., MacFadden, B. J., & Baskin, J. A. (1981). Geology and pa-
(1995). Fossil camels from the Upper Miocene of Europe: implications leontology of the Love bone bed from the late Miocene of Florida.
for biogeography and fauna! change. Geobios, 28, 641-50. American Journal of Science, 281, 513-44.
Prothero, D. R. (1986). A new oromerycid (Mammalia, Artiodactyla) Webb, S. D., & Taylor, B. E. (1980). The phylogeny of hornless ruminants
from the early Oligocene of Montana. Journal of Paleontology, 60, and a description of the cranium of Archaeomeryx. Bulletin of the
458-65. American Museum of Natural History, 167, 121-57.
(1996). Camelidae. In The Terrestrial Eocene-Oligocene Transition in White, T. E. (1940). New Miocene vertebrates from Florida. Proceedings of
North America, eds. D. R. Prothero & R. J. Emry, pp. 609-51. the New England Zoological Club, 18, 31-8.
Cambridge: Cambridge University Press. ( 1942). The lower Miocene mammal fauna of northern Florida. Bulletin
Ray, C. E. (1957). A list, bibliography, and index of the fossil vertebrates of of the Museum of Comparative Zoology, 92, 1-49.
Florida. Florida Geological Survey, Special Publication, 3, 1-175. (1947). Additions to the Miocene of North Florida. Bulletin of the Mu-
Schlaikjer, E. M. ( 1935). Contributions to the stratigraphy and paleontology seum of Comparative Zoology, 99, 479-515.
of the Goshen Hole area, Wyoming, IV: New vertebrates and the Wilson, J. A. (1974). Early Tertiary vertebrate faunas, Vieja Group and
stratigraphy of the Oligocene and early Miocene. Bulletin of the Buck Hill Group, Trans-Pecos Texas: Protoceratidae, Camelidae,
Museum of Comparative Zoology, 76, 97-189. Hypertragulidae. Bulletin of the Texas Memorial Museum, 23, 1-34.
Scott, W. D. (1937). A History ofLand Mammals in the Western Hemisphere. Wortman, J. L. (1898). The extinct Camelidae of North America and some
New York: Macmillan. associated forms. Bulletin of the American Museum of Natural His-
(1940). The mammalian fauna of the White River Oligocene, Part 4: tory, 10, 93-142.
Artiodactyla. Transactions of the American Philosophical Society, Zittel, K. A. (1893). Handbuch der Palaeontologie. Abtheilung I. Palaeozo-
28, 363-746. ologie. Band IV, Vertebrata (Mammalia). Munich: R. Oldenbourg.
31 Hornless ruminants
S. DAVID WEBB

INTRODUCTION

The four families of hornless ruminants treated in this chapter are

only loosely related to one another. The absence of horns (or antlers)
is not a shared-derived character (synapomorphy) that unites these
families phylogenetically, but merely a primitive feature (symple-
siomorphy). It is convenient to group them in one chapter, but in
reality they represent four separate immigrations into North Amer-
ica from the more continuous center(s) of ruminant evolution in the
Old World. The North American members of these families are very
well represented morphologically, and therefore provide some very
important insights into early ruminant evolution and adaptations.
These families help fill in the early portions of ruminant phylogeny
spanning the two great adaptive radiations of the Artiodactyla (see
Chapter 22), namely the middle to late Eocene radiation of early se-
lenodont groups and the late Oligocene to early Miocene radiation
of higher (mostly homed and antlered) ruminants.
The two earliest ruminant families, Hypertragulidae and Lep-
tomerycidae, were part of the selenodont radiation that began in the
middle Eocene, as were camelids and other possible sister groups.
The term selenodont refers to development of crescentic cusps on
the molars and molarized premolars. Some of the earliest selenodont
artiodactyls have five crescents in the upper molars, but typically
there are two pairs with their concave faces labially directed. They
interlock with two crescents in the lower molars facing in the op-
posite direction. The apices of the long row of lower crescents fall
between the apices of the double row of upper crescents. When
the lower jaw closes upward and inward in a rotary fashion, these
crescents occlude to provide a marvelously effective mechanism
for cutting and shredding leafy vegetation. Presumably selenodont
mastication was a key adaptation that helped trigger the radiation of
families, including Hypertragulidae and Leptomerycidae, that burst
on the scene during the later Eocene (Gazin, 1955).
The second two families treated here, tbe Gelocidae and the
Moschidae, appear in the late Oligocene and early Miocene and Figure 31.1. Restorations of hornless ruminants, by Brian Regal. A. Hyp-
represent key links to the higher ruminants of the second great isodus. B. Blastomeryx.

463
 S. David Webb 464

artiodactyl radiation. There is ongoing debate as to whether the the A

Moschidae originated prior to the second radiation or as part of
it. This chapter considers the four families of hornless ruminants,
saving the four homed and antlered families that reached North
America during the Miocene for subsequent chapters.
The tiny H ypertragulidae, including ti ve genera, were rare through
the late Eocene and Oligocene, but became locally abundant in
the early Miocene John Day deposits. Leptomerycidae were abun-
dant and moderately diverse during the Oligocene. They played
a major role as selective browsers and frugivores in forest un-
derstory and woodland settings. By early Miocene time, two of
1cm
the four North American leptomerycid genera had become extinct
and the few surviving taxa were evidently rare. Leptomerycids
were the longest lived family of ruminants in North America, rang- B
ing more than 30 million years from middle Eocene into middle
Miocene.
The Pseudoceratinae are the only New World representatives of
the family Gelocidae and were very rare throughout their middle
and late Miocene range. The North American Moschidae (subfamily
Blastomerycinae) were more diverse with six genera, but they too
are rare as fossils. Both families are virtually unknown west of
the Rocky Mountains. Their small sizes and brachydont dentitions 1cm
indicate that they were confined to forest edges and brushy habitats
as are their living relatives, the Chinese musk deer.

c
DEFINING FEATURES OF
HORNLESS RUMINANTS

The three defining features that can be recognized in fossil ruminants

are loss of all upper incisors, procumbent (incisiform) lower canine 4om

in the mandible, and fusion of the cuboid and navicular (proxi-

mal) elements in the tarsus. All but the Hypertragulidae share a Figure 31.2. Cranial features of hornless ruminants. A. Hypertragulus (after
deep parallel-sided astragalus (in contrast with the primitive bent Scott, 1899). B. Leptomeryx (after Scott, 1899). C. Parablastomeryx.
form of the astragalus in suiform artiodactyls). This progressive
astragalus is associated in ruminants with a fused cannon bone in laterally and posteriorly, and presence of a jugular foramen separate
the hindlimb. Additionally, evolution of the rumen, along with the from the posterior lacerate foramen.
specialized digestive physiology and morphology that support this
adaptation, are shared by living tragulids and moschids and thus
Dental
are attributable to the related fossil ruminant groups as well (with
Upper incisors absent; cl incisiform; conical Pl, P2, pl, and p2;
the possible exception of the Hypertragulidae ). The defining fea-
P3 and P4 triangular with conical protocones; p3 with tall apex
tures of each of the four hornless ruminant families are indicated
and small heel; p4 with two small lingual crests; upper molars with
next.
strong labial ribs, but lacking mesostyles.

HYPERTRAGULIDAE
Postcranial
Cranial Skeletons of Hypertragulus and its relatives are notable for their
The Hypertragulidae lie at the base of the Ruminantia, and thus small size and proportionally short forelimbs. They exhibit many
stand as the sister group to all other ruminants (Webb and Taylor, uniquely primitive features including short peglike odontoid pro-
1980). As expected of such a basal stock, most of the salient features cess, separate magnum and trapezoid, pentadactyl manus, complete
are plesiomorphic. Cranial features include narrow rostrum, orbits fibula fused distally with tibia, and "bent" astragalus in which prox-
centrally placed, postorbital bar absent, bulla with long bony meatus, imal and distal segments make an oblique angle. These plesiomor-
tympanohyal fossa small and laterally placed, petrosal-bearing sulci phies distinguish Hypertragulidae from all other ruminant families
for promontory and stapedial arteries, large mastoid exposure both (Webb and Taylor, 1980).
 Hornless ruminants 465

LEPTOMERYCIDAE
Cranial A
A number of derived features distinguish Leptomerycidae from Hy-
pertragulidae, including a broader rostrum, complete postorbital bar, 1cm
exposure of mastoid bone reduced and mainly on posterior surface,
jugular foramen confluent with posterior lacerate foramen, and tym-
panohyal depression near center of moderate! y enlarged bull a. These
features are shared with most other (more progressive) ruminant B
families. Among the primitive features shared with Hypertraguli-
dae are the centrally placed orbits; large, laterally placed mastoid 1cm
foramen; presence of two arterial sulci on the promontorium of the
petrosal; and deep subarcuate fossa on endocranial side of the pet-
rosal. A reliable autapomorphy is the long slitlike foramen ovale
seen in the Asiatic genus Archaeomeryx, as well as in Leptomeryx
(Webb and Taylor, 1980). c
Dental 1cm

The most distinctive feature, uniquely derived in Leptomerycidae, is

the enlarged, procumbent I 1. This is evident even in the very ancient
Archaeomeryx from Mongolia, and also served as the basis for the
name of the later genus, Pronodens. The upper incisors are vestigial
or absent, C vestigial, Pl absent; P2 and P3 with protocone; P4 with
strong lingual crescent; upper molars with mesostyle; pl caniniform 1cm
to reduced; other lower premolars form continuous series of three-
cusped teeth; p3 with metaconid; p4 with strong metaconid.
E
Postcranial
Leptomerycids are more progressive than hypertragulids in all but
one of their postcranial features. These include a moderately spout- 1cm
like odontoid process of the axis, fused magnum and trapezoid,
Figure 31.3. Dental features of hornless ruminants. A. Hypertragulus upper
tetradactyl manus, astragalus with parallel sides (not bent), fused
left cheek series (after Scott, 1899). B. Leptomeryx upper left cheek series
metatarsals III and IV, and fibula reduced to separate proximal rudi- (after Scott, 1899). C. Machaeromeryx upper left cheek series (after Matthew,
ment and distal malleolar bone. The forelimbs are relatively short 1926). D. Pseudoparablastomeryx scotti, lower left cheek series (after Taylor
(tragulidlike) proportionally, and are thus less progressive in lep- & Webb, 1976). E. Machaeromeryx lower left cheek series (after Matthew,
tomerycids even than in hypertragulids. 1926).

GELOCIDAE (PSEUDOCERATINAE) incorporate a narrow lenticular fossa; lower molars with strong an-
terolingual cingulum.
Cranial
Mastoid foramen small, posterodorsal; shallow subarcuate fossa
on endocranial side of petrosal; large fossa for stapedial muscle Postcranial
deeply pocketed in lateral wall of petrosal; bulla with moderately Shallow spoutlike odontoid process; fore- and hindlimbs essentially
deep tympanohyal depression near center; and large ovate fora- of equal lengths.
men ovale. Broadened basioccipital with strong fiexion stops on
condyles. These are progressive features shared with most higher
ruminant families (Webb and Taylor, 1980).
MOSCHIDAE (BLASTOMERYCINAE)
Dental Cranial
P3 with posteriorly placed and posteriorly directed protocone. Lower The following basicranial synapomorphies diagnose Blastomeryci-
canine varies from procumbent to vertical; pl premolariform, re- nae and support their alliance with Old World Moschidae: enlarged
duced or absent; lower premolars compressed transversely; p4 with suprameatal fissure; no sulci on promontorium; and laterally en-
metaconid ranging from strong to weak; p3 and p4 with four lin- closed subcentral tympanohyal depression on ventral surface of
gual crests, including two posterior crests from protoconid that may moderately large bulla (Webb and Taylor, 1980).
 S. David Webb 466

c
A B

,~ ·'

'----------'
1cm 1cm 1cm 1cm

Figure 31.4. Postcranial features of hornless ruminants. A. Forelimb of Hypertragulus (after Scott, 1899). B. Forelimb of Leptomeryx (after Scott, 1899).
C. Hindlimb of Hypertragulus (after Scott, 1899). D. Hindlimb of Leptomeryx (after Scott, 1899).

Dental dae and Protoceratidae, as subfamilies), Merycoidodontidae, and

Cl large saberlike, strongly dimorphic. Lower premolars with four Agriochoeridae.
lingual crests; metaconid crest anteriorly directed and meeting para- The critical feature on which Scott (1940) and Romer (1966)
conid crest. Lower molars with metastylids. based this broad alliance of Tylopoda was the supposed special-
ization of the first lower premolar as a caniniform tusk that oc-
Postcranial cluded posterior to the upper canine. This use of a single dental
Deep spoutlike odontoid process; closed (bridged) anterior metatarsal feature is beset by serious difficulties. By underlining the word be-
gulley. Complete distal keels on metapodials. Distal phalanges on hind in his description of this relationship between first lower pre-
distal rudiments of lateral metatarsals. molar and upper canine, Scott seems to imply something unique;
but obviously this occlusal relationship is the normal (primitive)
state in virtually all mammals with a complete, closed dental se-
SYSTEMATICS ries. A second objection is that the detailed morphology of the first
lower premolar is not really comparable between the groups that
Scott allied. In some forms, including Tragulidae, and the early
SUPRAFAMILY
camelid, Poebrotherium, pl has a low, leaftike profile that is clearly
Scott (1940) placed the North American Oligocene ruminants, along not caniniform. Webb and Taylor (1980, p. 138) argued that pos-
with several other groups, in the Tylopoda, a subordinal category session of such a leaftike, unicuspid p 1 was the primitive condition
based ultimately on the family Camelidae. Because he had a very among selenodonts, and that "the caniniform condition of the first
detailed comprehension of these animals, his views were very influ- lower premolar ... evolved independently in several selenodont ar-
ential and shaped the classifications of selenodont artiodactyls (such tiodactyl taxa." A third objection to Scott's key character is that the
as Romer [1966]) for many decades. Scott's suborder Tylopoda occlusal relationship of the lower premolar to the upper canine varies
included the following six North American Oligocene families: substantially from group to group. When Scott claimed that the
Camelidae, Hypertragulidae (in which he then includedLeptomeryci- condition was highly specialized, he alluded to the self-sharpening
 Hornless ruminants 467

(thegotic) occlusion in merycoidodontoids (indeed a very striking ruminants." Instead, these authors placed Hypertragulidae as "the
synapomorphy). This is really quite different from the arrangement next closest group, having a variety of parallelisms with the Lep-
in most Camelidae in which as many as five interlocking caninifonn tomerycidae and/or Tragulidae." The logic of their cladogram forced
teeth on each side of the rostrum (13, Cl, Pl, cl, and pl) are used Gentry and Hooker (1988) to assert that the fused cubonavicular,
for fighting and biting. In Hypertragulidae the caniniform teeth are the loss of upper incisors, and the procumbent lower canine evolved
very long and acuminate and do not directly occlude; they are more independently in Hypertragulidae, on the one hand, and in all (other)
nearly comparable with caniniform teeth of many Camelidae than ruminants, on the other.
with those ofMerycoidodontidae. Leptomerycidae, in contrast, have All three White River ruminant genera (Leptomeryx, Hypertragu-
the upper canine and lower first premolar wholly disengaged and lus, and Hypisodus) were placed by Scott ( 1940) in the family Hyper-
the latter is lost in many forms, thus resembling pecoran ruminants. tragulidae. In lumping them thus under Cape's family name, Scott
The problem of analyzing the bushy phylogenetic tree of selen- abandoned his own family name, Leptomerycidae (Scott, 1899),
odont artiodactyls is complicated by the fact that they diversified in which he had previously included the same taxa. Gazin (1955,
not only in North America but also in Europe, Asia, and Africa. p. 12) finally broke this logjam, noting that "striking dissimilarities
In his summary of North American Oligocene relationships, Scott in details of dentition" separate Hypertragulidae and Leptomeryci-
(1940, p. 560) claimed that "another character that unites these fam- dae. Webb and Taylor (1980) added other cranial and postcranial
ilies is their exclusively North American distribution ... until the features that support this separation. Leptomeryx itself is more pro-
Pliocene." This geographic argument is patently incorrect. For ex- gressive than any other Oligocene ruminant and has been compared
ample, Archaeomeryx from the late Eocene Shara Murun formation with Cervidae by such able students of the group as Leidy (1853)
of Mongolia is clearly a leptomerycid (Webb and Taylor, 1980). and Matthew (1908).
Scott also discounted the interesting suggestion of Earle ( 1901) re- Blastomerycinae were remarkably rare little ruminants scattered
garding the possible alliance of Merycoidodontoidea to European through the Miocene record until Frick (1937) substantially aug-
Mixtotherium. By ignoring the European and Asian selenodonts, mented the paleontological database. He referred to them as "diminu-
Scott made his geographic argument a self-fulfilling chauvinism. tive hornless deerlets" and contrasted them with the merycodont
His concept of Tylopoda was overly broad in including such ru- Antilocapridae and Moschus, the living musk deer of Asia (Frick,
minant stocks as Hypertragulidae and Leptomerycidae, but unduly 1937, p. 215). In some earlier studies (e.g., Lull, 1920), Blastomeryx
narrow in presuming only North American membership. was so poorly understood that its name was misapplied to excellent
Ruminantia can be based on a simpler, narrower hypothesis that material of Aletomeryx, a homed dromomerycid. Webb and Taylor
includes only families which share fused cubonavicular bones, no (1980) showed that Blastomerycinae are readily subsumed within
upper incisors, and procumbent (incisiform) lower canines. The liv- the family Moschidae, based on cranial, dental, and postcranial fea-
ing sister group of these ruminants is Camelidae, and the extinct tures. Most evidence of North American blastomerycines is cata-
content of Tylopoda is best treated not as a wastebasket but as en- logued in Frick (1937).
compassing only those early selenodonts that clearly shared a com- The Pseudoceratinae is a monotypic subfamily based on the enig-
mon ancestry with Camelidae. Tylopoda should certainly exclude matic little Pseudoceras of the late Miocene. Frick (1937, p. 649)
Merycoidodontoidea. These concepts were represented in Webb named the group and ascribed them to Camelidae. But he (Frick,
and Taylor's (1980) classification as TYLOPODA + RUMINAN- 1937, p. 648) expressed doubts about this assignment by calling
TIA = NEOSELENODONTIA. Camelidae and all of its collateral Pseudoceras "a small pseudocamelinelike form of uncertain rela-
branches forms the sister group of Hypertragulidae and all of its tionship." Tedford et al. (1987) recognized Pseudoceras as a New
collateral branches. The concept ofNeoselenodontia is accepted by World representative of the family Gelocidae. This assignment stems
Scott and Janis (1987). It is partly questioned by Gentry and Hooker from the complete osteological sample collected on the bottom of the
(1988, pp. 235, 259) but also "largely supported" by the same au- Withlacoochee River (site 4A) in Florida (locality GC1211), which
thors (p. 253). shows overwhelming evidence of ruminant affinities. The time of
The presence of a multichambered stomach, especially the rumen, appearance of this hornless ruminant in North America is remark-
is a defining feature of living camelids, on the one hand, and living ably late and decidedly out of order for, in the Old World, Ge-
tragulids, moschids, and all higher ruminants, on the other. The locidae precede Moschidae both in time and in their phylogenetic
adaptive significance of this possible neoselenodont synapomorphy placement. Not only did Pseudoceras appear late in the Miocene
is that it provides efficient symbiotic fermentation of plant cellulose (late Barstovian), but also it persisted through Clarendonian and
and also helps recycle nitrogenous wastes. It has not been possible into medial Hemphillian. Thus it became a remarkably archaic New
to decipher this diagnostic feature in fossils, and thus it cannot be World relict of the Old World Gelocidae.
invoked to sort out extinct branches of early selenodonts. The branching sequence of hornless ruminant families advocated
A family-level separation between Hypertragulidae and Lepto- by Webb and Taylor (1980) proceeds in the following phylogenetic
merycidae was advocated by Gazin (1955) using dental arguments order: Hypertragulidae; Tragulidae; Leptomerycidae; Gelocidae;
and by Taylor and Webb (1976) using crani~l and postcranial dif- and Moschidae. Only two of these families have living represen-
ferentia. Webb and Taylor (1980) proposed that Hypertragulidae lie tatives: These are Tragulidae, with living relicts in equatorial rain
at the base of the Ruminantia, and this view was accepted by Janis forests in Southeast Asia and west-central Africa, and Moschidae,
and Scott (1988). Gentry and Hooker (1988, p. 260), on the other with musk deer rather widely distributed in Asia. Whereas Moschi-
hand, advocated removing Hypertragulidae from the "undoubted dae reached North America during the Miocene, Tragulidae were
 S. David Webb 468

Figure 31.5. Proposed interrela-

tionships within families of horn-
less ruminants: A. HYPERTRAG-
ULIDAE: Key to characters on
cladogram: (1) Loss of mesostyles;
pl caniniform. (2) Loss of para-
conule. (3) Narrower premolars.
(4) Increased hypsodonty; loss of
paraconule; diastema behind p2 (5)
Diastema behind P2. (6) Extreme
root hypsodonty; loss of upper
teeth anterior to P3; flexed cranium.
c B. LEPTOMERYCIDAE: Key to
characters on cladogram: (I) Upper
incisors vestigial; lower incisors
procumbent. (2) Weak upper molar
styles; lower molar stylids reduced or
absent. (3) p3 more nearly resembles
p4. (4) Enlarged incisors; reduced
premolars. (5) Brachycephalic;
greatly reduced premolars. C. BLAS-
TOMERYCINAE: Key to characters
on cladogram: (1) Enlarged upper
canines. (2) Smaller size. (3) Broader
cranium; shorter diastema; larger pre-
molars. (4) Larger size. (5) Greater
hypsodonty; reduced premolars.
(6) Greatly reduced premolars;
elongate diastema.

confined to the Old World. Their geographic range was confined to ogram (Figure 3 l .5A) represents a preliminary attempt. As Emry
equatorial and low temperate latitudes and perhaps that is why they (1978, p. 1012) noted, "a revision of the family is badly needed."
never reached North America. This chapter thus treats four of the Hypertragulus of late Eocene and Oligocene ages serves as the
five families of hornless ruminants. central member of the group. The earlier (Duchesnean, middle
Eocene) Simimeryx shares many diagnostic features with Hyper-
tragulus, including the recurved caniniform pl and the absence of
INFRAFAMILY
mesostyles on the upper molars, but differs in other dental features,
Hypertragulidae especially the presence of a fifth crest (protoconule) in the upper
The five known genera of Hypertragulidae have never been placed in molars. Emry (1978) was so impressed by many dental differences
any phylogenetic or cladistic arrangement. The accompanying clad- that he "hesitated to include it in the family," whereas Kelly and
 Hornless ruminants 469

Whistler (1994) found more progressive samples of Simimeryx that also transferred the monotypic genus Pronodens, thought to have
helped to bridge the gaps. been an enigmatic early Miocene cervid, to the Leptomerycidae.
Hypisodus is related to Hypertragulus and ranges with it through Pronodens and Pseudoparablastomeryx differ from Leptomeryx in
the Chadronian and Orellan. This highly specialized little traguloid larger size, loss of pl, shorter postcanine diastema, and posterior
shows such marked differences from Hypertragulus as an extremely extension of symphysis beneath p2. These last two features may
narrow and pointed rostrum and a greatly reduced anterior dentition, be correlated with strengthening support for the tusklike lower in-
including loss of all teeth in front of P3. The molars of Hypisodusex- cisor. The main difference between Pronodens and Pseudopara-
hibit root hypsodonty more precociously than any other late Eocene blastomeryx is the smaller size and shorter symphysis of the latter.
or early Oligocene ungulate. Scott (1940) suggested that they might The cranium of Pseudoparablastomeryx is distinctly more brachy-
be evergrowing, but then cites Troxell's rare observation of root cephalic than that of Leptomeryx; its premolars are smaller relative
closure. The postorbital portion of the cranium is remarkably short, to the molars; the molars are more transversely compressed and
and the facial region is strongly flexed down relative to the brain- slightly higher crowned; m3 has a reduced entostylid; and its feet
case. The enlarged bullae of Hypisodus meet in the ventral midline. have fused cannon bones with complete keels on the distal ends.
Its hindfeet are unusually elongate, with tightly appressed third and
fourth metatarsals and greatly reduced side splints. "Gelocidae" (Pseudoceratinae)
The Chadronian genus Parvitragulus is particularly interesting The enigmatic position of the genus Pseudoceras has not yet been
for its relatively complex premolars, which more nearly resemble fully resolved. Although Tedford et al. (1987) indicated that its
those of other early ruminants than any other hypertragulids. This sister taxon must be some branch of Old World Gelocidae, that
is the basis for placing Parvitragulus as the sister group to all other is, a paraphyletic family at best and a waste-basket group at worst
hypertragulids in the tentative cladogram (Figure 3 l .5A). The pos- (Janis, 1987). Thus the only step forward to date has been to remove
sible relationship of Parvitragulus to the early Miocene genus Nan- Pseudoceras (and the subfamily Pseudoceratinae) from Camelidae
otragulus, tentatively proposed by Emry (1978), seems less likely to some (perhaps unknown) family of hornless ruminants. A revision
than derivation directly from Hypertragulus with some minor con- of the subfamily is still in progress.
vergences. The essential narrow structure of the premolars in Nan-
otragulus is more similar to that in Hypertragulus. The long time Moschidae (Blastomerycinae)
gap between Nanotragulus and Parvitragulus suggests that such re- Blastomeryx gemmifer Cope is typified by a single last lower molar
semblances as greater crown height, reduced intermediate conules, and therefore offers a poor basis for phylogenetic determination. But
and fewer cingula are convergent. The closed P2-P3 diastema that the subfamily represented by Blastomeryx plays an important role
characterizes Parvitragulus in the Chadronian is definitely conver- in the progression of ruminants and is now known by a substantial
gent within Nanotragulus, for this feature reappears only in the most North American sample of crania and skeletons spanning most of
progressive species such as N. ordinatus (Stevens, 1977). the Miocene. Blastomeryx was perceived by Matthew (1908) to be
Some small, incompletely known camelids from the early a small deer, somehow helping to bridge the gap in North Amer-
Miocene of Texas and Florida, Nothokemas and Delahomeryx ica between Oligocene and small modern forms such as Mazama.
(Stevens, Stevens, and Dawson, 1969), have been referred (faute de The genus was sometimes used for other small ruminants, even if
mieux) to "hypertraguloids," but this is very doubtful and has been they had horns. For example, Lull (1920) attributed the name Blas-
discounted by discovery of better material that is clearly camelid tomeryx to specimens later placed under their own rubric of Ale-
(e.g., Frailey, 1978; see also discussion later of Delahomeryx). tomeryx. When Matthew (1926) described a second blastomerycine
genus, Machaeromeryx, he noted its strong resemblance to Moschus,
Leptomerycidae but did not formalize any close affiliation with that family. Assign-
The earliest of the four North American leptomerycid genera is ment to Moschidae was formally proposed by Webb and Taylor
Hendryomeryx, characterized by lower crowned teeth with weaker (1980).
labial ribs and with less complete crescents on the upper molars In his monograph on North American ruminants, Frick (1937)
than in Leptomeryx. It is only subtly different from Leptomeryx and greatly elaborated the Blastomerycinae, adding five additional gen-
has not been compared directly with essentially contemporaneous era or subgenera, a large number of species, and dividing the group
Archaeomeryx from Mongolia. The dominant form throughout the into two subfamilies. Unfortunately, Frick did not fully designate
Oligocene and into the early Miocene is Leptomeryx. The premolars to which genera particular subgenera were attached, nor did he
are diagnostic and quite progressive with strong lingual protocones provide adequate diagnoses of most taxa. Frick (1937) restricted
on the uppers and strong paraconids, metaconids, and hypoconids use of Blastome1yx to the later Miocene. He also named two other
on p3 and p4 (Webb and Taylor, 1980, pp. 138-140). late Miocene genera: the very distinct Longirostromeryx and the
Taylor and Webb (1976) extended the Leptomerycidae from the probably distinct Parablastomeryx, basing a new subfamily (quite
Arikareean through the Hemingfordian and into the Barstovian (mid- inadequately defended) on each of these two genera. In the ear-
dle Miocene), not by new descriptions but by transfers from other lier Miocene, according to Frick's arrangement, one finds only the
ruminant families. They showed that two species of supposed blas- tiny Machaeromeryx and Frick's two new subgenera Pseudoblas-
tomerycines in the rare genus Pseudoparablastomeryx were late tomeryx and Problastomeryx. As noted earlier, another Frick (1937)
surviving (Hemingfordian and Barstovian) Leptomerycidae. They subgenus, Pseudoparablastomeryx, was reassigned (as a genus) by
 S. David Webb 470

Taylor and Webb (1976) to Leptomerycidae. Most of the diagnoses Included species: P. priscus only (known from localities
among the early blastomerycine genera appear to be nominal or SB43E, SB44D, CP39A, CP39B, CP39C, CP42B).
based on size and age. In contrast, Longirostromeryx is very clearly Parvitragulus sp. is also known from localities SB27B,
definable. ?CP98B.

Hypertragulus Cope, 1873

INCLUDED NORTH AMERICAN GENERA OF Type species: Hypertragulus calcaratus Cope, 1873.
HORNLESS RUMINANTS Type specimen: AMNH 7918.
Characteristics: Larger than any other hypertragulid except
The locality numbers listed for each genus refer to the list of unified Nanotragulus. Crown height lower than any other hyper-
localities in Appendix I. The acronyms for museum collections are tragulid except Simimeryx. Prominent anterior cingula and
listed in Appendix III. accessory conules on upper and lower molars. Pl and P2 as
The locality numbers may be listed in a couple of alternative ways. well as pl and p2 caniniform and separated from adjacent
Parentheses around the locality (e.g., [CPlOl]) mean the taxon in teeth by diastemata. P3 with very weak protocone; P4 with
question at that locality is cited as an "aff." or "cf." the taxon in conical protocone. Lower p3 secant (no lingual crests); p4
question. Parentheses are usually used for individual species, thus with posteriorly directed metaconid enclosing basin with
implying the genus is firmly known from the locality, but the actual hypoconid.
species identification may be questionable. Question marks in front Average length of m2: 6.0 mm (Scott, 1940, p. 522).
of the locality (e.g., ?CPlOl) mean the taxon is questionably known Included species: H. calcaratus (known from localities
from that locality, thus implying some doubt that the taxon is actually CC9C, CP68C, CP84A, [CP84B], [CP84C], CP98C,
present at that locality, either at the genus or the species level. CP99A, NP34B); H. heikeni (localities [SB43C], SB52);
H. hesperius (localities CC9D, CC13, PN6C, PN6D); H.
minutus (localities PN6E, PN6F).
HYPERTRAGULIDAE
Hypertragulus sp. is also known from localities CP40B,
Simimeryx Stock, 1934 CP41A, CP48, CP68B, CP98A, CP99B, CPlOl, NP34C,
Type species: Simimeryx hudsoni Stock, 1934. PN6G.
Type specimen: LACM (CIT) 1764.
Characteristics: Known from cheek teeth only. Smaller than Hypisodus Cope, 1873
Hypertragulus and Nanotragulus. Molars lower crowned Type species: Hypisodus minimus Cope, 1873.
than in Hypertragulus. Protocone and other upper cusps Type specimen: AMNH 9352.
less selenodont, with proportionally broader lingual cin- Characteristics: Much smaller even than Hypertragulus with
gula. Upper molars retain paraconule, giving a hint of a narrow abbreviated rostrum and anteriorly compressed
primitive five-crested pattern. Mesostyle absent as in all postorbital region. Postorbital bar complete. Very large
hypertragulids. Premolars simple and acuminate as in Hy- bullae meet midventrally. All upper teeth anterior to P3
pertragulus. No diastema behind p2. See Golz (1979) for lost. Molars precociously root hypsodont; uppers with flat
detailed discussion. labial walls, increasing in size posteriorly. Lower p2 lost
Average length of m2: 5.0 mm. except in very young individuals, with diastemata before
Included species: S. hudsoni (known from localities [CC7D], P3 and p3. Nearly fused posterior metatarsals three and
[CC7E], CC9B2, CC9C, CC9D); S. minutus (localities four and reduced lateral metatarsals.
CP6B, CP7C, ?CP9). Average length ofm2: 4.0 mm (Scott, 1940, p. 536).
Simimeryx sp. is also known from locality CC9A, CC9D. Included species: H. minimus only (known from localities
[SB44D], [SB44E], [CP40A], [CP41A], [CP46], CP68C,
Parvitragulus Emry, 1978 CP84A, CP98B, CP98C, CP99A).
Type species: Parvitragulus priscus Emry, 1978. Hypisodus sp. is also known from localities CP39A,
Type specimen: USNM 243971. CP39B, CP39C, CP39F, CP41A, CP41B, CP41C, CP48,
Characteristics: Smaller than all genera exceptHypisodus. No CP68B, CP68C, CP68D, CP84B, CP84C, CP85A,
diastemata separating P2 and p2 from other cheek teeth. ?CP86C, CP99B, CPlOl.
Upper and lower premolars more complex than in any other
hypertragulids. Upper premolars with protocones. Lower Nanotragulus Lull, 1922 (synonym: Allomeryx)
p3 with paraconid; p4 with paraconid, distinct metaconid, Type species: Nanotragulus loomisi Lull, 1922.
and small but distinct hypoconid. Crown height of mo- Type specimen: YPM 10330.
lars slightly more than in Hypertragulus. Cingula and ac- Characteristics: Larger and more hypsodont than Hypertrag-
cessory conules much weaker than in Hypertragulus and ulus; generally resembling that genus, but with reduced fre-
weaker than in Nanotragulus. quency and strength of cingula and intermediate conules.
Average length of m2: 4.6 mm (Emry, 1978, p. 1011). Some advanced species close diastema behind p2.
 Hornless ruminants 471

Average length of m2: 7 .0 mm. NPlOB, NP24C, NP24D); L. obliquidens (includes L.

Included species: N. loomisi (= N. intermedius) (known from agatensis) (locality [CP84C]); L. speciosus (localities
localities GC8A, ?CP42D, CP85C, CP86A, CPlOl, CP39C, CP39F, CP98C, NPIOB, NP24C, NP24D,
CP104B); N. fontanus (localities CC9C, CC12); N. or- NP27C); L. yoderi (localities CP39A, CP39B, CP42A,
dinatus (probable synonyms include N. matthewi, N. lulli [NPlOA], [NPlOB], [NP25C]).
and N. albanensis [Frick, 1937]) (localities SB46, [SB47], Leptomeryx sp. is also known from localities SB26B,
CP86C); N. planiceps ( = Allomeryx planiceps) (localities (SB52), CP39IIA, CP41A, CP43, CP45A, ?CP83B,
CC13, PN6C, PN6C). CP84B, CP85A, CP85C, CP86D, CP98B, CP98C, CP99A,
Nanotragulus sp. is also known from localities GC7, CP99B, CP102, CP103, CP105, ?NP9D, NPlOCC, NP22,
CP48, CP50, CP52, CP86D, CP99C, CP103, NP34C, NP24E, NP32B, NP34C, NP36D, NP50B, NP50C, NP51A.
NP38B, NP51A.
Pronodens Koerner, 1940
Indeterminate hypertragulids Type species: Pronodens silberlingi Koerner, 1940.
Remains assigned to the family Hypertragulidae have also been Type specimen: YPM 13952.
reported from localities CC7A, CP83A, ?CP83B, NP22. Characteristics: Larger than Leptomeryx. Known only from
mandible. Lower incisors 1 and 2 greatly enlarged; pl
LEPTOMERYCIDAE absent; c-p2 diastema much shorter than in Leptomeryx;
symphysis extends posteriorly to p2; length of premolars
Hendryomeryx Black, 1978 reduced relative to molars; premolar paraconids reduced.
Type species: Hendryomeryx wilsoni Black, 1978. Average length of m2: 8.0 mm.
Type specimen: CM 229102. Included species P. silberlingi only (known from localities
Characteristics: Known only from cheek teeth. Smaller than [CP86A], NP34B, NP34C).
Leptomeryx, with lower crowned molars and less fully de- Pronodens sp. is also known from localities NP36A,
veloped crescents. Upper molars with weaker styles than in NP36B.
Leptomeryx and with metacone rib absent. Lower p4 with
strong, inflected paraconid; metaconid joins hypoconid to Pseudoparablastomeryx Frick, 1937
enclose large basin, as in Leptomeryx. Lower molars with Type species: Pseudoparablastomeryx scotti Frick, 1937.
sty lids reduced or absent; anterior crest ofhypoconid meets Type specimen: F:AM 33763.
protoconid at very low level. Characteristics: Size smaller than Pronodens. Cranium
Average length of m2: 5.2 mm (Black, 1978, p. 257). brachycephalic. Upper premolars with stronger lingual cin-
Included species: H. wilsoni (known from localities SB43B, gula than in Leptomeryx. Lower pl absent. Premolars re-
CP29D); H. defordi (locality SB44D); H. esulcatus (prob- duced in length. Lower molars transversely compressed;
ably includes Leptomeryx transmontanus [Storer, 1981]) m3 entostylid greatly reduced.
(localities CP39E, CP68B, CP68C, CP83C, CP84C, Average length of m2: 7.0 mm (Taylor and Webb, 1976,
CP98B, NPlOB, NP29C). p. 13).
Hendryomeryx sp. is also known from localities Included species P. scotti (known from localities GC4C,
?SB43C, NP19B, NP25A, NP25B, NP32B. [CC22B], CP108B, CPllO, CPlll); P. francescita (=
Blastomeryx francescita Frick, 1937, p. 242) (localities
Leptomeryx Leidy, 1853 SB32D, CPI 14B).
Type species: Leptomeryx evansi Leidy, 1853. Pseudoparablastomeryx sp. is also known from locali-
Type specimen: USNM 157. ties CP86D, CP114A, ?NPlOD, (NC2).
Characteristics: Cranium long with central orbits. Upper in-
cisors reduced or absent; reduced upper canine; Pl absent; Comments: Storer and Bryant (1993) reported ?Pseudopara-
upper premolars triangular, with weak lingual cingulae; blastomeryx from their Locality 72Fl0-0043 along with
upper molars quadrate with strong styles and ribs. lncisi- Archaeohippus stenolophus in the Cypress Hills Forma-
form lower canine; small pl; lower premolars with three tion of Saskatchewan (locality NP! OD), thus extending the
cuspids each; camel-like pattern on p3 and p4; lower mo- range northward and confirming the late Arikareean age.
lars procumbent, ml enlarged.
Average length ofm2: 7.0 mm (Scott, 1940, p. 553).
GELOCIDAE (PSEUDOCERATINAE)
Included species: L. evansi (known from localities [CP9],
[CP46], CP68B, CP68C, CP68D, CP84A, CP84B, CP98B, Pseudoceras Frick, 1937
CP98C, CP99A, NNPlOB, NP27D, NP29C, NP29D, Type species: Pseudoceras skinneri Frick, 1937.
NP32B); L. blacki (localities NB2, [NP9C], [NPlOA], Type specimen: F:AM 33723.
[NPlOB], [NP25C]); L. mammifer (includes L. esulca- Characteristics: Small hornless ruminants. Skull with orbits
tus and L. semicinctus) (localities CP39C, CP39E, CP39F, at midlength; inflated, hollow bullae; narrow triangular
 S. David Webb 472

occiput (Frick, 1937, p. 651). First incisor large and pro- of Lusk and from Little Muddy Creek. It is likely that the
cumbent; others smaller and somewhat erect; lower canine type specimen comes from CP52 in the Harrison Forma-
large and recurved. Cheek teeth transversely compressed; tion (early late Arikareean) where there is a specimen of
p2, 3 with posteriorly directed metaconid; p4 with poste- Pseudoblastomeryx in the Frick collection.
riorly directed metaconid and hypoconid enclosing long
narrow fossettid. Parablastomeryx Frick, 1937
Average length of m2: 5.0 mm. Type species: Parablastomeryx gregorii Frick, 1937.
Included species: P. skinneri (known from localities CA9, Type specimen: F:AM 31360.
CPI 16B); P. potteri (locality CPI 16B); P. wilsoni (locality Average length ofm2: 10.8 mm.
CPl16B). Characteristics: Late Miocene, larger size group of Frick
Pseudoceras sp. is also known from localities GC6B, (1937). Brachydont molars; relatively large premolars,
GCI IA, GCl lB, GCI IC, GC12II, SB34A, SPIC, SP3A, short diastemata; premolar series length equals about 67
CPI 14D, CPI 16A, CPI 16C, CP116D. percent of molar series length (Frick, 1937).
Included species: P. gregorii (known from locality CP116B);
P.fioridanus White, 1941 (localityGC8D);P. galushiFrick,
MOSCHIDAE (BLASTOMERYCINAE) 1937 (localities CP88, CP89, CP105).
Parablastomeryx sp. is also known from localities
Problastomeryx Frick, 1937
CP90A, CP116A, NPlOE.
Type species: Problastomeryx primus (Matthew, 1908) (=
Blastomeryx primus).
Type specimen: AMNH 13822. Blastomeryx Cope, 1877
Characteristics: Early Miocene, larger size group of Frick Type species: Blastomeryxgemmifer(Cope, 1874 =Meryco-
(1937); lower premolar series length equals about 70 per- dus gemmifer).
cent of molar series length (Frick, 1937). Type specimen: AMNH 8301.
Average length of m2: 10.2 mm. Average length of m2: 8.5 mm (based on Matthew [1904]
Included species: P. primus (known from locality CP86D); referred mandible from same locality and horizon as type;
P. olcotti (Matthew, 1908 = Blastomeryx olcotti) (locality length unknown in type, which is m3 only).
CP104B). Characteristics: Late Miocene, smaller size group of Frick
Problastomeryx sp. is also known from localities (1937). Rounded skull profile; large upper canines; brachy-
?GC3A, CP52, CP71, CP88, CP105, CP109A, PN18. dont molars; large premolars; short diastemata.
Comments: Storer and Bryant (1993) referred an upper molar Included species: B. gemmifer (including B. gemmifervalenti-
to "Blastomeryx sp." from late Arikareean deposits along nus Frick, 1937) (known from localities CP74B, CP114A,
with Archaeohippus stenolophus in the Cypress Hills For- CP114B, CP114D, CP116A); B. elegans Matthew and
mation of Saskatchewan (locality NP! OD). It probably per- Cook, 1909 (localities GC6A, ?GC6B, [CP71], CPllO,
tains to this tax on and is an important early record of a large [CPI 11], [NPll], [NP41B]; B.francesca Frick, 1937 (lo-
moschid in Canada. calities [SB32A], [SB32B], SB32D, SB32F, SB34A); B.
marshi Lull, 1920 (localities [GC9B], CP106); B. medius
Pseudoblastomeryx Frick, 1937 Matthew, 1924 (including B. gemmife r medius Frick, 193 7)
Type species: Pseudoblastomeryxfalkenbachi Frick, 1937. (localities [CP71], CP108A, CP108B); B. mefferdi Frick,
Type specimen: F:AM 31530. 1937 (locality CP!14C); B. mollis Merriam, 1911 (?=
Average length of m2: 10.6 mm (calculated from Frick, 1937, Parablastomeryx, Morea, 1981) (localities NB 18,
Figure 24 and measurement table). NB20A); ?B. vigoratus Hay, 1924 (= ?Longirostromeryx
Characteristics: Early Miocene, smaller size group of Frick vigoratus Frick, 1937) (locality GC3B).
(1937); lower premolar series length equals about 70 per- Blastomeryx sp. is also known from localities GC3A,
cent of molar series length (Frick, 1937). ?GC4C, GC4D, GC4E, GC8C, GC9A, CC23A, CC23B,
Included species: P. f alkenbachi (known from Iocalities CC 16 ?SP2A, ?CP75B, CP86D, CP105, CP109A, CP109B,
[see comments later under Parablastomeryx], ?CP52); NPlOD, NP!OE NP38C, NP40A, NP40B.
P. advena (Matthew, 1907 = Blastomeryx advena) (lo-
Comments: Frick (1937) queried his reference of B. vigora-
calities CP86D, [CP88], CP104B); P. schultzi Frick, 1937
tus Hay (1924) to Longirostromeryx. The early Heming-
(locality CP106).
fordian age of the type and referred specimens predates
Comments: Frick (1937) states that the type specimen of other definite specimens of that genus. The Garvin Gully
Pseudoblastomeryx falkenbachi comes from "about material does not present the diagnostic short premolars
18 miles south of Lusk," Wyoming, and referred to it ma- and long diastema of Longirostromeryx. Pending better
terials from other areas between 10 and 20 miles southeast material, this species is referred to Blastomeryx.
 Hornless ruminants 473

Machaeromeryx Matthew, 1926 and m3, posterior edge of m3 with entoconid overlapping
Type species: Machaeromeryx tragulus Matthew, 1926. hypoconulid. Intercolumnar tubercles originate from pro-
Type specimen: AMNH 20548. tocone, large on ml-m2, reduced on m3.
Average length of m2: 8.9 mm. Length ofm2: 14.4 mm.
Characteristics: Early Miocene, smallest size group of Frick Included species: D. browni only, known from locality SB46
(1937); reduced premolar series, length about 60 percent only.
of molar series length; cementum on molars; smallest size
group; very large upper canine. Comments: This taxon is known only from two partial
Included species: M. tragulus (known from locality CP104B); mandibles from the type locality, Castolon Local Fauna in
M. gilchristensis White, 1941 (locality GC5). the Delaho Formation, Texas. Stevens (in Stevens, Stevens,
Machaeromeryx sp. is also known from localities GC8D, and Dawson, 1969) originally allied Delahomeryx with
GC9A, CP71. Nothokemas in the family "?Nothokematidae," which she
then considered to be a hypertraguloid or cameloid partly
following White (1941). MeanwhileNothokemas has come
Longirostromeryx Frick, 1937
to be regarded without serious doubt as a camelid (Patton,
Type species: Longirostromeryx wellsi (Matthew, 1904) (=
1966; Honey et al., this volume, Chapter 30). Two decades
L. merriami and L. serpentinus Frick, 1937).
later, Stevens and Stevens (1989) referred to Delahome-
Type specimen: AMNH 9823.
ryx as a "small cervid," but the age of the Castolon Local
Average length of m2: 9.6 mm (average of 8.4 mm in
Fauna is early late Arikareean, far earlier than any estab-
Matthew's type and 10.8 mm in UCMP 33506 [Gregory,
lished record of Cervidae in North America (see Webb,
1942]).
this volume). Other possible assignments for Delahome-
Characteristics: Late Miocene, medium-sized group of
ryx, worthy of further study, lie among the early Miocene
Frick (1937). Molars moderately hypsodont; very long di-
Hypertragulidae and Leptomerycidae.
astemata; premolars greatly reduced, length of series about
40 percent of molar series length; p2 lost or greatly re-
duced.
Included species: L. wellsi (= Blastomeryx wellsi Matthew,
SELENODONT ARTIODACTYL INCERTAE SEDIS
1904, also= L. merriami and L. serpentinus Frick, 1937)
(known from localities CP90A, CP90B, CP114D, CP115A, Hidrosotherium Wilson, 1974
CP115B, CP116A, [CP116B]); ?L. blicki Frick, 1937 (lo- Type species: Hidrosotherium transpecoensis Wilson, 1974.
calities GC4E, SB32D, SB32F, SB34A, CP114B); L. Type specimen: FMNH PM 454.
clare'ndoniensis Frick, 1937 (locality SP2A); L. novomex- Characteristics (taken from Wilson, 1974): Molars brachy-
icanus Frick, 1937 (locality SB32D). dont with strong styles and ribs, lacking internal cingulae
Longirostromeryx sp. is also known from locality on median pillars. Upper incisors probably reduced; Pl
CP114C. probably lost; long diastema between Cl and P2; pl elon-
gate, single rooted, with posterior cusp; p4 lacks meta-
Comments: From latest Barstovian through Clarendonian conid. Preorbital fossa present; small, noncancellous au-
Longirostrome ryx undergoes increased hypsodonty and de- ditory bullae; postorbital bar incomplete.
creased length of the premolar series. Longirostromeryx Length ofm2: 7.5 mm.
shares these tendencies with the much earlier Machaero- Included species: H. transpecoensis only, known from local-
meryx, suggesting a possible sister group (tribal) relation- ity SB44B only.
ship between these two genera.
Comments: This taxon is known only from an articulated
Indeterminate blastomerycids skull and jaws, and a posterior portion of the skull, at
Remains assigned to the subfamily Blastomerycinae have also been the type locality of Porvenir Local Fauna in Trans-Pecos,
reported from localities PN6G, PN7. Texas. Wilson (1974) originally considered Hidroso-
therium as a camelid grouped with Poebrodon in a sub-
family "Poebrodontidae." Although he thought the skull
RUMINANTIA INCERTAE SEDIS
resembled that of Leptomeryx, he found the premolar and
Delahomeryx Stevens, 1969 molar crown patterns more reminiscent of early camelids.
Type species: Delahomeryx browni Stevens, 1969. However, Honey et al. (this volume, Chapter 30) declined
Type specimen: TMM 40620-7. to include this taxon with the Camelidae, and Prothero
Characteristics (from Stevens, Stevens, and Dawson, 1969): (pers. comm.) considers it to belong possibly with the Lep-
Teeth brachydont; ml with united anterolingual and pos- tomerycidae. In my opinion Hidrosotherium is probably a
terolabial crescents, crescents closely appressed on m2 camelid, as proposed by Wilson.
 S. David Webb 474

w
z L L.81.
1.8
w E.BI
2.4
0 4.5
0 E Let. Hp.
II)

~
:i L.Hp.
5.2

a. L.E.Hp.
6.0
7.0
E.E.Hp.
L L.CI.
8.8 ~
9.5 t T

~
E.CI.
.ge ~ ?1
~
11.0
w II)
z L.L.Ba.
w .!!!
e ...~ -
12.5 Q)
~-
0 E.L.Ba. .Q [ ~ t
0 M 14.0 e t .g Q)
e ...~ § ....ee Q) -
:IE E.Ba. [
.g
e .!!!
II)
.g
.g .Q II)... - :g
- Q)

.ge 111 ·!::

15.8

.!!! .g .Q
II) .!!! Q)
:::s
...~
L.Hm.
17.5
Q)

~
.Q :::s e §
e -II)

11 ....
E.Hm. Q) Q)
_.!!! §
~
111
~ ~ 11 r:a
18.8
E L.L.Ar. a.
19.2 ..c:
.!!!
E.L.Ar.
23.0
~
w L.E.Ar.
II)
z 5i
w L 'ti
0
0
27.7 0--------------
c:
(!J
:i
E.E.Ar.
e
a.
29.4
0
Wt.
E 31.9
Or.
ll)_!g
33.4

-!g

·-~.
L.Ch.

~0 ll:
34.5 :::s
L M.Ch.
35.5 ~- Q) g> ~ Q)

~ t § t~
Q)

.§f
E.Ch.
37.1 .!!! 0
Du .§ :a ~ ~ c: §
39,5
Cl) a. :c: :c: ~ -~ ....
L.Un. 5i
:c: ....2-
w
. ._________ ._______. . . L--1------------
41.3

z
w E.Un.

0 45.9 HYPERTRAGULIDAE LEPTOMERYCIDAE + MOSCHIDAE

0
w M
L.Br.
46.7
GELOCIDAE
+
ARTIODACTYLA
M.Br. INCERTAE SEDIS
49.0
E.Br.
50.4

L.Wa.

53,5
E M.Wa.
54.2
E.Wa.
55.0
E.E.Wa.
55.5
Ck.3 55.6
w Ck.2
z
w L
Ck.1
Tl.6
Tl.5
....
55.8
56.0

57.0
0 Tl.4
TI.3
57,3
0 Tl.2
58.5
w Tl.1
...J
ct To.3
a. To.2
To.1
E Pu.3
Pu.2
Pu.1
Pu.o

Bl.=Blancan, Hp. = Hemphilllan, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To.= Torrejonian, Pu. = Puercan.
Figure 31.6. Temporal ranges of genera of hornless ruminants.
 Hornless ruminants 475

BIOLOGY AND EVOLUTIONARY PATTERNS The Blastomerycinae, as suggested earlier, may be closely com-
pared with the living musk deer. They were far more progressive
Figure 31.6 illustrates the North American temporal ranges of the than the Oligocene and earliest Miocene Hypertragulidae and Lep-
four families of hornless ruminants and their included genera. Two tomerycidae. Their large upper canines, which in a male skull in-
distinct windows open into the paleobiology of the extinct horn- spired Matthew's name Machaeromeryx, imply a deerlike stage of
less ruminants of North America. One window consists of the fossil social evolution with strong sexual selection. Moschid dentitions,
material itself. Paleobiologists bring to bear not only traditional both fossil and recent, have moderately hypsodont molars, with a
comparative functional anatomy, but also independent clues from coat of cement on their crowns and a premolar row with broad, sub-
taphonomy, soils, and isotopes. The second window on the paleo- molariform posterior premolars, and these features bespeak a more
biology of these fossil taxa consists of a handful of living hornless advanced stage of mixed feeding than in traguloids. Moschid skele-
ruminant species. Although these living fossils do not live in North tons are larger than those of tragulids, with more rigid backbones,
America, each is tied cladistically into the ruminant phylogenetic and equal-length fore- and hindlimbs with enlarged cannon bones
tree. Because they form such an important starting point for recon- and reduced side toes. These features imply deerlike running and
structing physiology, behavior, and ecology of their extinct relatives, leaping habits, as opposed to rabbitlike bounding. Modern musk
they may be considered first. deer are solitary, inhabiting forest edges and brushy habitats in the
Two genera of modern Tragulidae come from rain forest settings interior of China and Tibet, and similar settings may be invoked for
in equatorial Africa (Hyemoschus) and in southeast Asia (Tragulus ). their extinct cousins. The latest surviving genus, Longirostromeryx,
Together they constitute the nearest living relatives of the Hyper- had relatively rapidly attained a long diastema, short premolars,
tragulidae and Leptomerycidae, and at least a first approximation of and proportionally taller crowned molars, indicating a move into
the lifestyles of these earliest fossil ruminants. Similarly, the modern more open-country living and somewhat coarser mixed feeding
Chinese musk deer (Moschus) provides a close approximation to the habits.
physiology, behavior, and ecology of the extinct Blastomerycinae Pseudoceras has no confamilial living relatives, and its lifestyle
insofar as one can judge by their osteological resemblances. remains enigmatic. The skeletal proportions are closer to those of
A critical biological feature of modem ruminants is of course the Moschidae than to any of the traguloids. Its cheek teeth are
their ability to digest a wide variety of vegetation, even very coarse, roughly comparable to those of Moschus in crown height and cement
cellulose-rich grasses with the assistance of symbiotic microftora development. Presumably it can be considered a rough analogue to
and an elaborate four-chambered stomach. The capacity to do this that genus. Butthe only site in which Pseudoceras occurs abundantly
ranges widely and is most highly developed in large herding bovids (Withlacoochee 4A in Florida, locality GC 12II) evidently represents
such as Bison. The occurrence of a small but functional ruminant a small pond in a forested setting. This paleoecological clue suggests
stomach in living Tragulidae provides a phylogenetic argument that a possible analogy between the lifestyles of Pseudoceras and the
probably Hypertragulidae and certainly Leptomerycidae also ru- African water chevrotain, Hyemoschus, with an affinity for aquatic
minated. Members of these earliest families also probably availed settings and solitary, nocturnal behavior.
themselves of similar forested habitats, darting like rabbits in thick-
ets or shady undergrowth. The tiny four-toed feet and relatively
short forelimbs of Hypertragulus confirm this view. Likewise, the REFERENCES
low-crowned teeth, tiny incisors, and long narrow skulls of the Hy-
pertragulidae and Leptomerycidae confirm osteologically their ad- Black, C. C. (1978). Paleontology and geology of the Bad water Creek area,
central Wyoming, Part 14: The artiodactyls. Annals of the Carnegie
herence to the relatively dainty diets of their living counterparts that
Museum, 47, 223-59.
rely on fruits, tender shoots, and occasional insects. Cope, E. D. (1873). On Menotherium lemurinum, Hypisodus minimus, Hy-
These generalizations, however, were surely subject to a range pertragulus calcaratus, Hypertragulus tricostatus, Protohippus, and
of particularities and exceptions. In the White River Badlands, Lep- Procamelus occidentalis. Proceedings of the Philadelphia Academy
tomerycidae are decidedly more abundant than Hypertragulidae and of Natural Sciences, 25, 410-20.
(1874). Notes on the Santa Fe marls and some of the contained verte-
may have had a somewhat more gregarious social system. "At many
bratae fossils. Proceedings of the Philadelphia Academy of Natural
localities [Leptomeryx is] one of the most common taxa" (Emry, Sciences, 1984, 147-152.
Russell, and Bjork, 1987, p. 133). Both families appeared in North (1877). Report upon the extinct Vertebrata obtained in New Mexico by
America in the late Eocene when open woodland (or protosavannna) parties of the expedition of 1874. Wheeler Survey, 4, 1-370.
habitats were opening up (Webb, 1977). Hypisodus may well have Earle, C. (1901). Notes on the fossil Mammalia of Europe. American Natu-
ralist, 32, 115-17.
been precociously adapted to coarser (albeit carefully selected) fod-
Emry, R. J. (1978). A new hypertragulid (Mammalia, Ruminantia) from the
der. The progressive dental and cranial features noted here suggest early Chadronian of Wyoming and Texas. Journal of Paleontology,
that Hypisodus was an evolutionary overachiever, which, in the ab- 52, 1004-14.
sence of more progressive ruminant competition, developed a be- Emry, R. J., Russell, L. S., & Bjork, P.R. (1987). The Chadronian, Orellan,
havioral repertoire more like that of a dik-dik (Bovidae) than of and Whitneyan North American Land Mammal Ages. In Cenozoic
Mammals of North America, ed. M. 0. Woodburne, pp. 118-152.
other Tragulina. Like the present-day dik-dik (Madoqua), it may
Berkeley: University of California Press.
have lived in forest margins and fed in scrubby openings. This hy- Frailey, C. D. (1978). An early Miocene (Arikareean) fauna from North
pothesis needs to be tested by taphonomic and isotopic evidence. Central Florida (the SB- IA local fauna). Occasional Papers of the
 S. David Webb 476

Museum of Natural History, University of Kansas (Lawrence), 75, D.C.: National Zoology Symposium Volume, Smithsonian Institu-
1-20. tion Press.
Frick, C. (1937). Horned ruminants of North America. Bulletin of the Amer- Scott, W. B. (1899). The selenodont artiodactyls of the Uinta Eocene. Trans-
ican Museum of Natural History, 69, 1-669. actions of the Wagner Free Institute, Philadelphia, 6, 1-121.
Gazin, C. L. (1955). A review of the upper Eocene Artiodactyla of North (1940). The mammalian fauna of the White River Oligocene, Part 4:
America. Smithsonian Miscellaneous Collections, 128, 1-96. Artiodactyla. Transactions of the American Philosophical Society,
Gentry, A. W., & Hooker, J. J. (1988). The phylogeny of the Artiodactyla. In n. ser., 34, 209-52.
The Phylogeny and Classification of the Tetrapods, vol. 2: Mammals, Stevens, M. S. (1977). Further study ofCastolonlocal fauna (early Miocene),
ed. M. J. Benton, pp. 235-72. Oxford: Clarendon. Big Bend National Park, Texas. Pearce-Sellards Series, Texas Memo-
Golz, D. J. (1979). Eocene Artiodactyla of Southern California. Natural rial Museum, 28, 1-69.
History Museum of Los Angeles County Science Bulletin, 26, 1-85. Stevens, M. S., & Stevens, J.B. (1989). Neogene-Quaternary deposits and
Gregory, J. T. (1942). Pliocene vertebrates from Big Spring Canyon, South vertebrate faunas, Trans-Pecos Texas. In Vertebrate Paleontology,
Dakota. University of California Publications, Bulletin of the De- Biostratigraphy, and Depositional Environments, Latest Cretaceous
partment of Geological Sciences, 26, 307-446. and Tertiary, Big Bend Area, Texas, eds. A. B. Busby II & T. M.
Janis, C. M. (1987). Grades and clades: The reality of the Gelocidae and Lehman, pp. 67-90. Austin, Texas: Guidebook of the 49th Annual
the systematic position of Lophiomeryx and Bachitherium. Journal Meeting of the Society of Vertebrate Paleontology.
of Vertebrate Paleontology, 7, 200-16. Stevens, M. S., Stevens, J.B., & Dawson, M. R. (1969). New early Miocene
Janis, C. M., & Scott, K. M. (1988). The phylogeny of the Ruminantia (Ar- Formation and vertebrate Local Fauna, Big Bend National Park,
tiodactyla, Mammalia). In The Phylogeny and Classification of the Brewster County, Texas. Pearce-Sellards Series, Texas Memorial
Tetrapods, vol. 2: Mammals, ed. M. J. Benton, pp. 273-82. Oxford: Museum, 15, 1-53.
Clarendon. Stock, C. (1934). A hypertragulid from the Sespe uppermost Eocene, Cali-
Kelly, T. S., & Whistler, D. P. (1994). Additional Uintan and Duchesnean fornia. Proceedings of the National Academy of Sciences, 20, 625-9.
(middle and late Eocene) Mammals from the Sespe Formation, Simi Storer, J. E. ( 1981 ). Leptomerycid artiodactyls of the Calf Creek Local Fauna
Valley, California. Contributions in Science, Natural History Mu- (Cypress Hills Formation, Oligocene, Chadronian) Saskatchewan.
seum of Los Angeles County, 439, 1-29. Natural History Contributions, Museum of Natural History, Regina,
Koerner, H. E. (1940). The geology and vertebrate paleontology of the Fort 3, 1-32.
Logan and Deep River formations of Montana. American Journal of Storer, J.E., & Bryant, H. N. (1993). Biostratigraphy of the Cypress Hills
Science, 238, 837-62. Formation (Eocene to Miocene), Saskatchewan: Equid types (Mam-
Leidy, J. (1853). Remarks on a collection of fossil Mammalia from Nebraska. malia, Perissodactyla) and associated fauna! assemblages. Journal ·I
Proceedings of the Philadelphia Academy of Natural Sciences, 6, of Paleontology, 67, 660--9.
392-4. Taylor, B. E., & Webb, S. D. (1976). Miocene Leptomerycidae (Artiodactyla,
Lull, R. S. (1920). New Tertiary artiodactyls. American Journal of Science, Ruminantia) and their relationships. American Museum Novitates,
L, 81-130. 2596, 1-22.
(1922). Primitive Pecora in the Yale Museum. American Journal of Sci- Tedford, R.H., Skinner, M. F., Fields, R. S., Rensberger, J.M., Whistler, D.
ence, 5, 111-19. P., Galusha, R., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987).
Matthew, W. D. (1904). A complete skeleton of Merycodus. Bulletin of the Faunal Succession and biochronology of the Arikareean through
American Museum of Natural History, 20, 101-30. Hemphillian (Late Oligocene through earliest Pliocene epochs) in
(1907). A lower Miocene fauna from South Dakota. Bulletin of the Amer- North America. In Cenozoic Mammals of North America, ed. M. 0.
ican Museum of Natural History, 23, 169-220. Woodburne, pp. 153-210. Berkeley: University of California Press.
(1908). Osteology of Blastomeryx and phylogeny of the American Cervi- Webb, S. D. (1977). A history of savanna vertebrates in the New World,
dae. Bulletin of the American Museum ofNatural History, 3, 535-62. Part I: North America. Annual Review of Ecology and Systematics,
(1926). On a new primitive deer and two traguloid genera from the lower 8, 355-80.
Miocene of Nebraska. American Museum Novitates, 215, 1-8. Webb, S. D., & Taylor, B. E. (1980). The phylogeny of hornless ruminants
Patton, T. H. (1966). Revision of the selenodont artiodactyls from Thomas and a description of the cranium of Archaeomeryx. Bulletin of the
Farm. Florida Academy of Sciences, 29, 179-90. American Museum of Natural History, 167, 121-57.
Romer, A. S. (1966). Vertebrate Paleontology (3rd ed.). Chicago: University White, T. E. ( 1941 ). Additions to the Miocene fauna of Florida. Proceedings
of Chicago Press. of the New England Zoology Club, 18, 91-8.
Scott, K. M., & Janis, C. M. (1987). The phylogenetic position of the Cervi- Wilson, J. A. (1974). Early Tertiary vertebrate faunas, Vieja Group and
dae, and a case for the superfamily Cervoidea. In The Biology and Buck Hill Group, Trans-Pecos, Texas: Protoceratidae, Camelidae,
Management ofthe Cervidae, ed. C. Wemmer, pp. 3-20. Washington, Hypertragulidae. Bulletin of the Texas Memorial Museum, 23, 1-34.
 32 Dromomerycidae

CHRISTINE M. JANIS and EARL MANNING

INTRODUCTION

The Dromomerycidae were Miocene and earliest Pliocene cervoid

ruminants unique to North America, but most probably had Eurasian
I origins. They left no known descendants. They were small to medium-
·I sized animals, of a comparable size range to modem deer, pos-
sessing unbranched, paired, supraorbital cranial appendages in the
I (presumed) males only in most genera. Members of the tribe Cranio-
ceratini developed a third median occipital cranial appendage in ad-
dition to the paired supraorbital cranial appendages of the other gen-
era. They were most likely browsers living in subtropical/temperate
habitats ranging from moderately open bushland/grassland ecotone
to dense woodland and thicket. Their decline and disappearance at
the end of the epoch was probably correlated with increasing aridity
and the reduction and eventual disappearance of suitable subtropical
woodland habitat. Dromomerycids were a moderately diverse and
abundant group during the Miocene, with a peak diversity during the
early Barstovian (middle Miocene). The family is interpreted here
as containing two subfamilies (with one subfamily further divided
into two tribes) and fifteen genera or subgenera, known from a wide
variety of localities over the central, western, and southeastern parts
of the United States.

Figure 32.1. Restoration of Sinclairomeryx, by Brian Regal. The animal (a

DEFINING FEATURES OF THE male) is depicted marking twigs with a deerlike antorbital gland, in correla-
FAMILY DROMOMERYCIDAE tion with the presence of a deerlike antorbital fossa in this genus.

CRANIAL
atively obtuse. The paraoccipital processes are moderate and the
Supraorbital, unbranched, nondeciduous cranial appendages devel- auditory bullae are small. There is a double lacrimal foramen on the
oped on the frontal bones, variably round or oval in cross section: frontal bone at the anterodorsal orbital rim. The muzzle is usually
The tips of these appendages are generally expanded into a termi- fairly long and narrow; the orbit shifts posteriorly in derived forms;
nal bulb. Contrary to figures in Frick (1937, frontispiece and Fig- the premaxillae are long and have a broad contact with the nasals;
ure 1) there is no evidence for deciduous tips to these appendages. the lacrimal bone is usually rectangular in shape; a cervidlike an-
The posterior part of the occiput is generally elongated, at least in torbital vacuity is present, and a lacrimal fossa is present in some
the subfamily Dromomerycinae, tending to project into a median genera. The glenoid is broad, and the mandible is fairly shallow with
prominence at the back of the head. The basicranial angle is rel- a prominent angle.

477
 Christine M. Janis and Earl Manning 478

DENTAL
A
In the more brachydont members of the family, the premolars are
relatively large and unmolarized (although the permanent first pre-
molar is absent), and the diastema is short. The lower molars are
characterized by a Palaeomeryx fold (a short enamel ridge run-
ning posteromedially from the protoconid on the labial face of the
lower molars), and the upper molars may possess a bifurcation
of the protocone. Most genera (with the exception of Aletomeryx
and Rakomeryx) possess the suite of dental features typical of the
Eurasian palaeomerycids: a reduced metastyle in the upper molars,
an attenuated metacone on P4, and a double posterior lobe on m3 that
is closed posteriorly (Janis and Scott, 1987). Primitively, both upper
and lower mola~s possess prominent anterior and posterior cingu-
lae, and there is an entostyle in the upper molars and an ectostylid
in the lower molars. The later, more hypsodont genera tend to re-
duce the length of the premolar row, to increase the length of the
C/P2 diastema, and to simplify the form of the cheek teeth, with
the reduction or loss of the Palaeomeryx fold and reduction of the B
m3 hypoconulid. These changes occurred independently in the three
subgroups, paralleling the changes seen in the Eurasian cervoid lin-
eages. In derived forms the p4 fossette (trigonid) is closed lingually
by an enlargement of the metaconid. The incisors and the lower ca-
nines are simple and spatulate. Primitively, a small saberlike upper
canine is retained (especially in the males), as is a dpl.

POST CRANIAL c
The neck is fairly short (as in cervids ). The limbs are generally rela-
tively short and heavily proportioned, as in present-day woodland or
forest dwelling ruminants, although metapodials three and four are
fully fused in all genera, and the metapodial/phalangeal keels are
complete. A few forms have relatively long and slender limbs. Lat-
eral digits are sometimes retained in the forelimb. The metatarsal
gully is closed, and a tuberosity is present on the lateral side of
the posterior proximal metatarsus in all genera except Aletomeryx.
There is a tubercle for the fused proximal end of metatarsal II on the
medioproximal end of the metatarsus, but metatarsal V is unfused, Figure 32.2. A. Skull of Dromomeryx (Dromomeryx) borealis (Modified
and the cubonavicular facet is somewhat raised, as in the typical from Frick, 1937; illustration of CM 827) (bar= 10 cm). B. Dentition
cervid condition (see Janis and Scott, 1987). of Dromomeryx: upper dentition of D. (D.) whitfordi (modified from Frick,
1937, ofF:AM 31245); lower dentition of D. (D.) pawniensis (modified from
Frick, 1937, of F:AM 31306) (bars = 1 cm). C. Skeleton of Dromomeryx
(D.) borealis (modified from Douglass, 1909 (bar= 50 cm) (all drawings
SYSTEMATICS by Brian Regal).

SUPRAFAMILY
more brachydont molars with a Palaeomeryx fold, and the presence
The first dromomerycid was described by Cope in 1878 from a of cranial appendages. The term dromomerycid was later expanded
partial skull with a supraorbital cranial appendage as "Blastomeryx to include other species of North American Tertiary ruminants that
borealis" from the Deep River beds of Montana. Further descrip- possessed similar types of unbranched, supraorbital, apparently
tions by Scott (1895) commented on the similarity of this animal skin-covered cranial appendages, with brachydont to mesodont
to the European genus Palaeomeryx. More extensive collections cheek teeth and moderately heavily proportioned limbs.
were made by Douglass during the 1890s (summarized in Douglass, Prior to Frick's major study of 1937, only five valid dromo-
1899), and in 1909 he named the animal "Dromomeryx" (meaning merycid genera had been described. Two of these (Drepanomeryx
"running ruminant"). He distinguished this genus from Blastomeryx Sinclair, 1915 and Cranioceras Matthew, 1918) were thought to be
(as described by Matthew in 1908) on the basis of the larger size, antelope; one (Aletomeryx Lull, 1920) was considered a pronghorn
 Dromomerycidae 479

(Frick, 1937, p. 46); and one (Pediomeryx) an unidentified rumi- of palaeomerycids appear to be giraffidlike ossicones. In their view,
nant. Frick (1937) almost single-handedly established the family the apparent similarity of the cranial appendages is due to conver-
by revising the known taxa and adding all of the presently known gence.
genera. Dromomerycids are presumed to be Eurasian immigrants, not
Dromomerycids were considered to be cervids (deer), or cer- only because of a possible relationship with the Eurasian palaeomery-
voids (related to cervids ), in the last century and the earlier part of cids, but also because there are no suitable North American pecoran
this one (e.g., Cope, 1878; Douglass, 1909; Matthew, 1924; Frick, sources in the earlier Miocene. A possible Asian dromomerycid,
1937, Simpson, 1945). Frick classified them as the divisions Dromo- Asiagenes ceratinus, has been described by Vislobokova (1983)
merycini and Aletomerycini within the family Cervidae. However, from the latest Oligocene of Mongolia. Vislobokova (1983) com-
they were reclassified by Stirton (1944) as a subfamily within the pares this taxon, known from a homcore fragment and two upper
family Palaeomerycidae (the other subfamily consisting solely of molars, with Dromomeryx. The problem with this tax on is, although
the European genus Palaeomeryx), which he allied with the Giraf- the homcore fragment does resemble that of Dromomeryx (a rather
fidae in the superfamily Giraffoidea on the basis of the apparent derived genus of a derived group; see Figure 32.3), with an ovoid
similarity of the cranial appendages. More recently, most authors cross section and a lateral basal flange, the animal appears to be too
have followed this classification by allying dromomerycids with derived in terms of both the morphology of the cranial appendages
giraffids (e.g., Crusafont, 1961; Viret, 1961; Hamilton, 1978a), al- and its size (the molars are of a similar length to Dromomeryx, which
though Romer (1966) and Leinders (1983) classified them with cer- is a fairly large dromomerycid) to be a suitable Asian sister taxon
voids. More recent work (Janis and Scott, 1987) has shown that for the North American dromomerycids. Indeed, Vislobokova places
dromomerycids are clearly cervoids (see Figure 22.1), despite some Asiagenes within Frick's ( 193 7) division Dromomerycini, which re-
superficial resemblance to the form of the skull and the cranial ap- quires a complicated series of immigrations to North America ifthe
pendages seen in giraffids, on the basis of the possession of a double cladogram depicted in Figure 32.3 is correct (see later discussion).
lacrimal foramen on the orbital rim, a closed metatarsal gully, the Moreover, the molars are not especially dromomerycidlike because
presence of a saberlike upper canine in early tax a such asAletomeryx they retain a large metastyle. Although this taxon is certainly in-
and Barbouromeryx, and the presence of a Palaeomeryx fold in the teresting, and could possibly represent an Asian dromomerycid, in
lower molars. However, dromomerycid cranial appendages cannot our opinion it remains too problematical to be firmly assigned as
be simply homologized with the deciduous, branched antlers of most an Asian sister taxon to the North American taxa. A more plausible
true cervids (note that the available evidence suggests that antlers explanation would be that Asiagenes represents a reimmigration to
evolved within the Cervidae; see Janis and Scott, 1987). Asia, with the possibility that the Mongolian deposits are actually
Although dromomerycids have been referred to the family Dro- of medial Miocene age rather than latest Oligocene.
momerycidae in publications of the past couple of decades or so The poorly known European genus Triceromeryx has also been
(e.g., Hamilton, 1978a; Janis, 1982; Leinders, 1983; Webb, 1983; suggested to be a dromomerycid because it possesses a similar type
Stucky and McKenna, 1993), the family has never been formally de- of median occipital cranial appendage to the dromomerycid genus
fined as such. Janis and Scott (1987) suggested the reestablishment Cranioceras. Leinders (1983) has suggested that it may be an im-
of the family Palaeomerycidae, which was disbanded by Hamil- migrant dromomerycid from a North American stock, but the taxon
ton (1978b), to include dromomerycids with the Eurasian genera lacks the defining dental features of the dromomerycids/palaeomery-
Palaeomeryx and Amphitragulus, and possibly also the African cids (although it appears to be a cervoid of some sort) (see Janis and
genus Prolibytherium, on the basis of a derived suite of dental char- Scott, 1987). Duranthon et al. (1995) consider this taxon to have
acters, as described in the previous section. They further suggested palaeomerycidlike cranial appendages.
demoting the family Dromomerycidae to the subfamily status of
Dromomerycinae. However, given the common usage of the familial
INFRAFAMILY
taxonomic rank of the Dromomerycidae (e.g., Stucky and McKenna,
1993), we retain the family ranking in this chapter. However, it seems In his 1937 monograph on North American ruminants, Frick sepa-
likely that Palaeomeryx is the sister taxon to the Dromomerycidae, rated the dromomerycids into two divisions (taxa between the family
and the supraorbital (at least) cranial appendages may be homol- and subfamily level, within the family Cervidae).
ogous in these taxa. The recently described skull of Palaeomeryx Division A: the Dromomerycini. Characterized primitively by
tricornis (Qiu et al., 1985) bears a pair of unbranched, probably the retention of brachydont molars; more primitive features of the
skin-covered, supraorbital cranial appendages and a small median limbs and dentition in general; large premolars; lower molars with a
occipital appendage, and shows a strong resemblance to the early Palaeomeryx fold; moderate length of diastema; and short, heavily
Miocene dromomerycid Barbouromeryx trigonocorneus ..This may proportioned limbs. Dromerycini also all have a considerable poste-
suggest that Palaeomeryx could be the sister group of the Craniocer- rior projection of the occiput, with the genera in Frick's subfamilies
atini (see Figure 32.3, node 11), rather than of the family as a whole. Barbouromerycinae and Cranioceratinae possessing a median oc-
In contrast, Duranthon et al. (1995) present evidence suggesting cipital cranial appendage. The division includes most of the dromo-
that the mode of formation of cranial appendages is not homolo- merycid taxa, including the genera or subgenera Subdromomeryx,
gous between dromomerycids and palaeomerycids: the appendages Dromomeryx, and Rakomeryx (Dromomerycinae), Drepanomeryx
of dromomerycids appear to be frontal outgrowths, whereas those and Matthomeryx (Drepanomerycinae), Barbouromeryx and
 Christine M. Janis and Earl Manning 480

Figure 32.3. Proposed interrelationships within the Dromomerycidae. Key to heads (modified from illustrations in Frick, 1937, and Webb, 1983, by
Brian Regal: to approximate scale only): A. Aletomeryx. B. Sinclairomeryx. C. Drepanomeryx. D. Rakomeryx. E. Dromomeryx. F. Barbouromeryx.
G. Bouromeryx. H. Procranioceras. I. Cranioceras. J. Pediomeryx. Key to characters at nodes on cladogram: (I) DROMOMERYCIDAE Frick, 1937: tips
of supraorbital cranial appendages expanded. (2) ALETOMERYCINAE Frick, 1937 (emended): molars mesodont (cingulae and Palaeomeryx fold reduced
or lost), upper canine reduced, upper premolar protocone area reduced, orbit moved posteriorly (anterior rim over anterior M3), older females with small
supraorbital cranial appendages (known with certainty only in Aletomeryx), ramus deepened, premolars reduced, antorbital vacuity present, metapodials
lengthened. (3) Sinclairomeryx Frick, 1937: supraorbital cranial appendages lengthened and tilted forward, paired anterior nasal bosses present, deep
maxillary pits present, lacrimal fossa present, overall size increased. (4) DROMOMERYCINAE Frick, 1937 (emended): protrusion present on posterior
portion of sagittal crest, slitlike antorbital vacuity, paralleling the anterior wing of the frontal. (5) DROMOMERYCINI Frick, 1937 (expanded): flange present
on posterolateral side of the base of the supraorbital cranial appendages, tips of supraorbital cranial appendages recurved medially and lengthened, upper
canines lost, ascending ramus tilted posteriorly on the body of the dentary, c/p2 diastema lengthened (skull becomes more dolichocephalic), wide and shallow
Iacrimal depression, overall size increased. (6) Drepanomeryx Sinclair, 1915: supraorbital cranial appendages tilted strongly posteriorly, tips of supraorbital
cranial appendages flattened anteroposteriorly. (7) Supraorbital cranial appendages tilted anteriorly, orbit moved posteriorly (anterior rim over anterior M3).
(8) Rakomeryx Frick, 1937: bases of supraorbital cranial appendages strongly flattened anteroposteriorly. (9) Orbit moved further posteriorly (anterior rim
over posterior M3), p4 metaconid expanded anteriorly, supraorbital cranial appendages straightened (not recurved as strongly medially). (! 0) Dromomeryx
(Dromomeryx) Douglass, 1909: p4 trigonid closed lingually (by anterior extension of metaconid), supraorbital cranial appendages thickened and massive,
overall size increased. (11) CRANIOCERATINI Frick, 1937 (expanded): posterior sagittal protrusion and median occipital crest enlarged (to form an
occipital cranial appendage), supraorbital cranial appendages moved anteriorly (nearly vertical), lower rim of orbit expanded laterally. (12) All three cranial
appendages elongate, upper canines lost, c/p2 diastema lengthened (skull becomes more dolichocephalic), short, heavily proportioned limbs, overall size
increased. (13) All cranial appendages further elongated, dpl lost, premolars reduced, overall size increased. (14) All cranial appendages further elongated,
premolars further reduced, molars more hypsodont, c/p2 diastema lengthened, occipital horn tilts posteriorly, overall size increased. (15) p4 metaconid
expanded anteriorly (trigonid closed lingually), premolars further reduced, m3 hypoconulid reduced. (16) Pediomeryx (Yumaceras) Frick, 1937 (restricted):
overall size greatly increased. (17) Pediomeryx (Pediomeryx) Stirton, 1936: molars more hypsodont, premolars further reduced (especially P2/p2), overall
size decreased.

Bouromeryx (Barbouromerycinae), Procranioceras, and Cranio- and Yumaceras (Yumaceratinae). The relative hypsodonty of these
ceras (Cranioceratinae). two otherwise very different taxa, which was used by Frick (1937,
Division B: the Aletomerycini. Characterized by simple supraor- p. 50) to unite them, is interpreted here as having been acquired in-
bital cranial appendages without a basal flange, and a shorter occipi- dependently (Figure 32.3, nodes 2 and 15), making Frick's "division
tal region with no trace of an occipital appendage. They also possess Aletomerycini" polyphyletic.
more derived features of the limbs and dentition: mesodont molars; Stirton (1936) recognized the late Hemphillian genus Pediomeryx
reduced premolars; loss of the Palaeomeryxfold in most instances; a from the late Clarendonian and early Hemphillian. Webb (1983)
longer diastema; and longer, more slender metapodials. The division made Yumaceras (Frick, 1937) a synonym of Pediomeryx and treated
includes the genera Aletomeryx, Sinclairomeryx (Aletomerycinae), them as subgenera, as an earlier Pediomeryx (Yumaceras) and a later
 Dromomerycidae 481

Pediomeryx (Pediomeryx). Webb also synonymized Frick's (1937) hers of the Dromomerycini and Aletomerycinae in late Arikareean
supposedly first true North American cervid, Procoileus edensis, sediments.
with Pediomeryx hemphillensis. He also noted, on the basis of Even given the presently known North American dromomerycids,
associated cranial material that included a long median occi- the cladogram presented here strongly argues that the Aletomeryci-
pital appendage, that Pediomeryx should be placed in the nae separated early from the Dromomerycinae and quickly became
Cranioceratinae rather than in the Aletomerycini, despite the more very derived. It would be a fundamental error to see Aletomeryx as
mesodont cheek teeth that resembled those of aletomerycines. Webb a primitive dromomerycid simply because it is an early one: The
expanded Frick's subfamily Cranioceratinae to include Pediomeryx same could be said of Barbouromeryx.
and the members of Frick's subfamily Barbouromerycinae.
Based largely on an unpublished 1980 cladogram of the junior
author, we propose here a modification of Frick's classification INCLUDED GENERA IN THE
of dromomerycids (see Figure 32.3). There is a division between FAMILY DROMOMERYCIDAE
forms evolved from an ancestor with a lump on the posterior part
of the sagittal crest, the subfamily Dromomerycinae, and the sur- Most of the descriptions of taxa in this listing was taken from
prisingly derived early subfamily Aletomerycinae. Thus, except for Frick (1937), which remains the sole comprehensive taxonomic
the change in the position of Yumaceras, Frick's two primary "Di- work on this family. The family is badly in need of revision at
visions" are maintained. Within the Dromomerycinae we recog- the subgeneric and specific levels. Many of these genera are greatly
nize two tribes: the Dromomerycini (which groups Frick's Dromo- oversplit (most notably Aletomeryx, Rakomeryx, Cranioceras, and
merycinae and Drepanomerycinae), united by the basal flange on particularly Bouromeryx), and some species, especially in the case
the posterolateral side of the supraorbital cranial appendages; and of the Cranioceratinae, may be assigned to the wrong genus. The
the Cranioceratinae (as in Webb's [1983] usage, grouping Frick's genera were established mainly on the characters of the cranial ap-
Barbouromerycinae, Cranioceratinae, and Yumaceratinae, the latter pendages, and many of the described species are known from dental
transferred from Frick's division Aletomerycini), united by the ex- remains only.Note that, because of the cumbersome nature of Frick's
pansion of the occipital protrusion into a more definitive cranial divisions into subgenera within genera, outside of this section of the
appendage. Although Frick (1937, p. 46) treated his Barbouro- chapter the subgenera are generally referred to by their subgeneric
merycinae as only vaguely related to his Cranioceratinae, there are names only.
indications that he saw a relationship at least of Bouromeryx to his The locality numbers listed for each genus refer to the list of uni-
Cranioceratinae (Frick, 1937, p. 127). The primary distinction of fied localities in Appendix I. The locality numbers may be listed in
the cladogram presented here lies in the formal reorganization of a couple of alternative ways. The acronyms for museum collections
the taxa within Frick's "division Dromomerycini" (here treated as are listed in Appendix III.
subfamily Drnmomerycinae). Parentheses around the locality (e.g., [CPlOl]) mean the taxon
If the cranial appendages of Palaeomeryx tricornis are indeed ho- in question at that locality is cited as an "aff." or "cf." the taxon in
mologous with those of cranioceratinine dromomerycids, as previ- question. Parentheses are usually used for individual species, thus
ously discussed, then the family must be derived from more than one implying the genus is firmly known from the locality, but the actual
immigrant to North America. This scenario actually coincides with species identification may be questionable. Question marks in front
the fossil record evidence because Barbouromeryx is the only dro- of the locality (e.g., ?CPlOl) mean the taxon is questionably known
momerycid known from the late Arikareean (Tedford et al., 1987). from that locality, thus implying some doubt that the tax on is actually
The Aletomerycinae and the Dromomerycini could still represent present at that locality, either at the genus or the species level. An
monophyletic groups, the Aletomerycinae united by their more de- asterix (*) indicates the type locality.
rived dental and postcranial features, and the Dromomerycini united
by the basal flange on the supraorbital cranial appendages. This sce-
ALETOMERYCINAE FRICK, 1937
nario would entail two separate immigrations of dromomerycids to
North America in the later Miocene. If the late Oligocene Asia- Characteristics: males with supraorbital cranial appendages that are
genes ceratinus (Vislobokova, 1983) is indeed a dromomerycinine, slender, initially posteriorly directed, and secondarily recurved an-
this would suggest that all three lineages had a very early Asian ori- teriorly; smaller cranial appendages sometimes present in females
gin long before reaching North America because dromomerycinines (known with certainty only in Aletomeryx). Cranial appendages
do not appear until the late Hemingfordian (Tedford et al., 1987). somewhat triangular in cross section at the base, oval in cross sec-
Perhaps a more simple explanation, and one that would be more tion in the midshaft, and bulbous at the tips. The posterior sagittal
biogeographically parsimonious, is that given that parallelism in crest of the occiput is only moderately extended and not thickened
development of cranial appendages is so common among ungulates or projected dorsally to any great extent. An antorbital vacuity may
(see Janis, 1982), the Old World taxa represent cases of parallel horn be present. Orbit moved posteriorly (anterior rim over anterior M3,
development to that in New World taxa (a_s also suggested by Du- moved back from anterior M2). Diastema fairly long. Small saber-
ranthon et al., 1995), and thus there was only a single immigration like upper canines retained in the males (definitely for Aletomeryx,
of dromomerycids to North America. Future work may reveal mem- suggestion of canine alveolus only in Sinclairomeryx). Premolars
 Christine M. Janis and Earl Manning 482

reduced; dpl lost; premolar protocone area reduced; anterior fos- Characteristics: Supraorbital cranial appendages long (in
sette on P4 not closed. Molars mesodont; Palaeomeryx fold in lower males only), projecting anteriorly and laterally. Antorbital
molars reduced or lost. Metapodials fairly long and slender; digits vacuity and lacrimal fossa present. Nasals elongate, with
II and V retained in the manus, digit V retained in the pes. paired nasal bosses in males, and deep maxillary pits.
Unworn molars may retain a weak Palaeomeryx fold. In
Aletomeryx Lull, 1920 comparison with Aletomeryx, molars more brachydont and
Type species: Aletomeryx gracilis Lull, 1920. metapodials more heavily proportioned. Size small to
Type specimen: YPM 10732, early Hemingfordian of medium.
Nebraska. Average length ofm2: 14.5-16.5 mm.
Characteristics: Cranial appendages short (longest appendage Included species: S. riparius (known from localities
no more than one third the length of the skull) and fairly CP108A*, CP108B); S. tedi Frick, 1937 (localities
erect; posteromedial comer of cranial appendage has [CP107], CP109A *).
narrow yane, separating it from posteroinferior border of Sinclairomeryx sp. is also known from localities CP107,
the orbit. Smaller cranial appendages present in females. CP109A, CPl 10, ?NPlOE, (NPl l).
Fairly short nasals. Longest and most slender metapodials
of all the dromomerycids. Small body size. Comments: Certain publications (Storer, 1975; Skwara, 1988)
Average length ofm2: 10.0--14.6 mm. have extended the known distribution of Sinclairomeryx
Included species (an asterix denotes the type locality): A. gra- into the northern Great Plains area of Saskatchewan, with
cilis (known from localities CP105, CP106*); A. marshi the possible persistence of the genus into the early late
Frick, 1937 (= Blastomeryx marshi Lull, 1920, Blasto- Barstovian.
meryx [Dyseomeryx] marshi Matthew, 1924) (locality
CP105*); A. scotti Frick, 1937 (= Blastomeryx [Dyseo- DROMOMERYCINAE FRICK, 1937
meryx] scotti Matthew, 1924) (locality CP105*); A. lugni
Frick, 1937 (locality CP106*); A marslandensis Frick, DROMOMERYCINI FRICK, 1937
1937 (locality CP104C*); A. occidentalis Whistler, 1984 Characteristics: Males with supraorbital cranial appendages that
(locality NB4*). are broad and tall, bowed and with medially recurved tips, with a
Aletomeryx sp. is also known from localities CP71, distinct posterolateral basal flange. Posterior occipital region elon-
CP104C, CP105, ?CP109A. gated, with a pronounced median dorsal projection of the occiput,
Comments: Tedford et al. (1987) cite Aletomeryx as limited but with no occipital cranial appendage. Skull more dolicephalic.
to the early Hemingfordian. However, a single description A wide and shallow lacrimal depression may be present, occupy-
of Aletomeryx from the late Hemingfordian comes from a ing both lacrimal and nasal areas. Slitlike antorbital vacuity present,
record in the Frick collection labeled as from the "temporal paralleling the anterior wing of the frontal. Upper canines lost. Di-
equivalent of Sheep Creek, Hay Springs area, Nebraska" astema lengthened, ascending ramus of dentary tilted posteriorly
(locality CP109A). As a number of other dromomerycids relative to the body of ramus. Premolars large; posterior premo-
are also known at this locality, including the aletomerycine lars submolariform, with anterior fossette (trigonid) on p4 usually
Sinclairomeryx, this could possibly represent a case of mis- closed; dpl retained. Molars brachydont; Palaeomeryx fold usually
taken identity, and so has been recorded here as a queried present on lower molars. Limbs short to moderately elongated; ves-
occurrence. tiges of lateral digits may be retained in forelimb. Size medium to
Further research will probably greatly reduce the num- large.
ber of Aletomeryx species. Some will likely be found to
be intergradational with others and will be synonymized Drepanomeryx Sinclair, 1915
or reduced to subspecies. Others may be found to be more Drepanomeryx (Drepanomeryx) Frick, 1937
closely related to Sinclairomeryx than to A. gracilis and Type species: Drepanomeryx (Drepanomeryx) falciformis
will be added to Sinclairomeryx. Frick preferred not to (Sinclair, 1915) (= Drepanomeryxfalciformis).
synonymize species and often named any form from a new Type specimen: PU 12072, early Barstovian of Nebraska.
area as a new species. Characteristics: Supraorbital cranial appendages elongated
and posteriorly directed, with small basal flanges, shafts
''
Sinclairomeryx Frick, 1937 widely bowed inwardly, moderately twisted with bulbous
Type species: Sinclairomeryx riparius (Matthew, 1924) tips, flattened anteroposteriorly. Dentition similar to Dro-
(= Blastomeryx [Dyseomeryx] riparius) (senior synonym momeryx, but open anterior fossette in p4. Some evidence
of the original type species Sinclairomeryx sinclairi Frick exists that small supraorbital cranial appendages may be
193 7; see discussion in Skinner, Skinner, and Gooris, 1977, present in Drepanomeryx females as well as in males (see
p. 344). Frick, 1937). Limbs moderately heavily proportioned.
Type specimen: AMNH 18956, late Hemingfordian of Medium size.
Nebraska. Average length of m2: 18.4 mm.
 Dromomerycidae 483

Included species: D. (D.) falciformis only, known from lo- following a comparison of Tedford et al., 1987 with Frick,
cality CPI 10* only. 1937) from the Spring Creek area in the Deep River For-
mation of Montana (locality NP34D): cf. Rakomeryx sp. in
Drepanomeryx (Matthomeryx) Frick, 1937 the Vedder Local Fauna in the Branch Canyon Sandstone
Type species: Drepanomeryx (Matthomeryx) matthewi Frick, of California (locality CCI 9) (in the University of Cal-
1937. ifornia Museum of Paleontology collections, Berkeley);
Type specimen: F:AM 33740, early Barstovian of Nebraska. ?Rakomeryxsp. from the Box Butte Formation in Nebraska
Characteristics: Similar size to (D.) Drepanomeryx; differs (locality CP107) (Galusha, 1975); and Rakomeryx sp. in
in possession of cranial appendages that are narrowed dis- the temporal equivalent of the Sheep Creek Formation,
tally, rather than expanded. Premolars more reduced and Hay Springs area, Nebraska (locality CP109) (in the Frick
diastema longer than in (D.) Drepanomeryx. collection).
Average length of m2: 21.3 mm. Early late Barstovian records include Rakomeryx gazini
Included species: D. (M.) matthewi only, known from locality from Skull Springs Fauna, Oregon (locality PN8B)
CPll 1*, only. (Shotwell, 1968) and Rakomeryx kinseyi in the Madison
Drepanomeryx (Matthomeryx) sp. is also known from Valley Fauna of Montana (locality NP4!B) (Frick, 1937).
locality GC4C.
Dromomeryx Douglass, 1909
Rakomeryx Frick, 1937 Dromomeryx (Subdromomeryx) Frick, 1937
Type species: Rakomeryx raki Frick, 1937. Type species: Dromomeryx (Subdromomeryx) scotti Frick,
Type specimen: F:AM 31320, early Barstovian of California. 1937.
Characteristics: Cranial appendages with less pronounced Type specimen: PU 10401, late Hemingfordian of Nebraska.
basal flange, anteriorly directed, with bases strongly flat- Characteristics: Similar to Dromomeryx, but smaller size,
tened anteroposteriorly, attenuated at tips; shafts strongly molars more mesodont, premolar row shorter, and Palaeo-
bowed laterally, forming concave half-moon shape when meryx fold usually lost. Supraorbital cranial appendages
viewed from front. Orbit moved posteriorly (anterior rim relatively smaller, with basal flange less pronounced, and
over anterior M3 ). Premolars more reduced than in Dromo- limbs relatively longer and more slender than in subgenus
meryx, but p4 may lack closure of anterior fossette. Has Dromomeryx (probably relates to smaller body size; see
metastylids in upper molars and large metacone on P4, Janis and Wilhelm, 1993).
unlike other dromomerycids (with the exception of Ale- Average length ofm2: 13.5-14.6 mm.
tomeryx). Limbs of similar proportions to Drepanomeryx. Included species: D. (S.) scotti Frick (includes D. [S.] wilsoni;
Medium size. see Skinner, Skinner, and Gooris, 1977, pp. 344-345)
Average length of m2: 16.0-20.0 mm. (known from localities [NB17], CP108A*, CP108B); D.
Included species: R. raki Frick, 1937 (= Dromomeryx or (S.) antilopinus (Scott, 1893) (= Blastomeryx antilopinus,
Cervus? of Merriam, 1911) (known from localities CC l 7D, Dromomeryx antilopina) (locality NP34D*).
NB6C*); R. jorakianus Frick, 1937 (locality NB6C*); R. Dromomeryx (Subdromomeryx) sp. is also known from
yermonensis Frick, 1937 (locality NB6C*); R. gazini Frick, localities GC4C, ?PN6F.
1937 (? = R. americanus Shotwell, 1968) (localities
Comments: Unless definite unique derived characters of Sub-
[GC9B], PN7, PN8B*); R. sinclairi Skinner, Skinner, and
dromomeryx are located, the decision to use Subdromo-
Gooris, 1977 (= Cervavus sinclairi Matthew, 1924, Blas-
meryx as a separate genus or as a subgenus of Dromomeryx
tomeryx [Dyseomeryx] sinclairi Matthew, 1924, Cranio-
sensu stricto is largely a matter of taste. As sister taxa, their
ceras unicornis Frick, 1937) (locality CPI 10*); R. kinseyi
relationship remains the same.
(Frick, 1937) (= Cranioceras kinseyi) (localities NP34D,
NP38C, NP41B*, PN18).
Dromomeryx (Dromomeryx) Frick, 1937
Rakomeryx sp. is also known from localities GC4C,
Type species: Dromomeryx (Dromomeryx) borealis (Cope,
GC9C, (CC19), NBl8, NB19A, SB32B, ?CP107, CP109,
PN7,PN19B. 1878) (= Blastomeryx borealis) (= Palaeomeryx borealis
Douglass, 1899).
Comments: Tedford et al. ( 1987) state that Rakomeryx is lim- Type specimen: AMNH 8132, early Barstovian of Montana.
ited to the early Barstovian. However, records exists from Characteristics: Cranial appendages large and broad, for-
both the late Hemingfordian and the early late Barsto- wardly directed (forming an angle of approximately 60°
vian. Late Hemingfordian records include Rakomeryx cf. with the nasals) and somewhat laterally bowed; prominent
R. americanus (? = R. gazini following reports of Frick, lateral basal flange, and deeply grooved on the anterolateral
1937) from the Midway Fauna of the Dogstown Memberof face. Orbit moved posteriorly (anterior rim over posterior
the Torreya Formation (locality GC9B) (MacFadden and M3). Shallow lacrimal fossa present. Lower p4 with an-
Webb, 1982); Rakomeryx kinseyi (= Cranioceras kinseyi terior fossette closed by anterior extension of metaconid.
 Christine M. Janis and Earl Manning 4S4

Limbs short and heavily proportioned. Large size. sisting of a short, median, bulbous projection. Lacrimal
Average length ofm2: 15.6-21.0 mm. fossa present. Short diastema. Moderate-sized saberlike
Included species: D. borealis (known from localities [NB l S], upper canine present in males. Large premolars with lit-
CPl 11, [CP114A], NP34E*, NP3SC, NP42, NP43, PN7, tle molarization; anterior fossette on p4 open. Brachydont
PNSA, PNSB, [PN9A], [PN9B]); D. pawniensis Frick, molars; Palaeomeryx fold present in lower molars. Limbs
1937 (localities CP75B, CP75C*, ?CP76); D. whitfordi show moderate degree of elongation, proportionally more
Sinclair, 1915 (= Palaeomeryx sp. Matthew and Cook, so than in other cranioceratinines (probably correlated with
1909, Dromomeryx borealis Matthew, 1924) (localities small size).
[CPlOSB], CPllO*, CPl 11). Average length of m2: 11.5 mm.
Dromomeryx (Dromomeryx) sp. is also known from lo- Included species: B. (B.) trigonocorneus (known from lo-
calities GC4C, CC21A, CC21C, CC23, NB6E, ?NBS, calities CP105, CP106*); B. (B.) cursor (Cook, 1922) (=
NBlS, NB20A, NB23B, SB32B, CP49D, ?CP90A, Dromomeryx cursor) (locality CP104A *).
?CP107, CP109A, NP40A, PN7. Barbouromeryx (Barbouromeryx) sp. is also known from
localities (CP54B), CP71, ?CPSS, ?CP104B (= "Aletomeryx
Comments: Tedford et al. ( l 9S7) cite the latest appearance of pristinus" in F:AM collections), CP106, CP107.
Dromomeryx as late Barstovian. However, some specimens
in the Frick collection appear to represent early Clarendo- Barbouromeryx (Probarbouromeryx) Frick, 1937
nian occurrences of this genus. These include the Avawatz Type species: Barbouromeryx (Probarbouromeryx) sweeti
Mountain Fauna in California (locality NBS), and a ques- Frick, 1937.
tionable occurrence in the Ash Hollow Formation of South Type specimen: NSM 53-25-6-34, early Hemingfordian of
Dakota (locality CP90A). These specimens should be reex- Nebraska.
amined to determine if they truly represent Dromomeryx or Characteristics: Similar to Barbouromeryx (Barbouromeryx)
if they are instead referable to Cranioceras. A late Claren- but of smaller size.
donian specimen labeled "cf. Dromomeryx" in the Uni- Average length of m2: 10.0 mm.
versity of California Berkeley collections in the Truckee Included species: B. (P.) sweeti only (known from localities
Formation of Nevada certainly appears to be too young; it CP105, CP106*).
has been cited here as a "dromomerycid indet."
Barbouromeryx (Protobarbouromeryx) Frick, 1937
Type species: Barbouromeryx (Protobarbouromeryx) mars-
CRANIOCERATINI FRICK, 1937 landensis Frick, 1937.
Characteristics: Males with supraorbital cranial appendages that are Type specimen: NSM 3-24-7-24, late Hemingfordian of
moved anteriorly (nearly vertical), short to tall, triangular or rounded Nebraska.
in cross section, and lacking a basal flange. Posterior occipital re- Characteristics: Similar to Barbouromeryx (Probarbouro-
gion elongated, with pronounced posterior dorsal projection of the meryx), but with reduced length of premolar row.
occiput, producing a median occipital cranial appendage of varying Average length of m2: 11.0 mm.
length. Lower rim of orbit expanded laterally. Absence of lacrimal Included species: B. (P.) marslandensis Frick only, known
vacuity, but presence of narrow, slitlike antorbital vacuity parallel- from locality CP107* only.
ing anterior wing of the frontal. Upper canines present in earliest Comments: Despite Frick's claims that B. (Probarbouro-
genus only. Diastema moderate to long. Premolars variably reduced; meryx) and B. (Protobarbouromeryx) are smaller than B.
posterior premolars unmolarized to submolariform; anterior fossette (Barbouromeryx), the teeth of all the three subgenera ap-
(trigonid) on p4 may be open or closed; dpl lost in more derived pear to be about the same size. The two former subgenera
forms. Molars brachydont to mesodont; Palaeomeryx fold present in are known only from a single species each, from a single
lower molars in early genera, absent in later ones. Limbs generally locality or two, and from a limited number of specimens
shorter and more heavily proportioned than in the Dromomerycini; (77 in the case of B. [Probarbouromeryx] but only two in
lateral digits may be present. Size small, medium, or large. the case of B. [Protobarbouromeryx]). It seems probable
that the three subgenera could all be dispensed with, and
Barbouromeryx Frick, 1937 treated simply as species of Barbouromeryx. With larger
Barbouromeryx (Barbouromeryx) Frick, 1937 Barbouromeryx samples now available, the distinctness of
Type species: Barbouromeryx (Barbouromeryx) trigono- the species may also be in question.
corneus (Barbour and Schultz, 1934) (= Dromomeryx
trigonocorneus). Bouromeryx (Frick, 1937)
Type specimen: NSM 3-27-11-35, early Hemingfordian of Type species: Bouromeryx milleri (Frick, 1937) (= Barbouro-
Nebraska. meryx [Bouromeryx] milleri).
Characteristics: Short, upright supraorbital cranial append- Type specimen: AMNH 21533, late Hemingfordian of
ages with bulbous tips. Occipital cranial appendage con- Nebraska.
Dromomerycidae 4S5

Characteristics: Medium-length upright supraorbital cranial Procranioceras (Frick, 1937)

appendages, short bulbous median occipital appendage. Type species: Procranioceras skinneri (Frick, 1937) (=Cran-
Upper canine lost. Short diastema. Large premolars with ioceras [Procranioceras] skinneri).
little molarization; anterior fossette on p4 open. Brachy- Type specimen: F:AM 31250, late late Barstovian of
dont molars; Palaeomeryx fold present in lower molars. Nebraska.
Short, heavily proportioned limbs. Small to medium Characteristics: Differs from Cranioceras in the shorter me-
size. dian occipital cranial appendage, which is transversely flat-
Average length of m2: 13.5-15.6. tened and anteriorly curved. The premolars are larger than
Included species: B. milleri (known from localities WM13, in Cranioceras, but the anterior fossette on p4 tends to be
CC22B, NBl 7, CP107, CPlOSB*); B. submilleri Frick, open. The molars tend to be more brachydont, with the
1937 (localities GC4E, [CC20], NB17, [CP107], usual presence of a Palaeomeryx fold. Limbs of similar
CPlOSA*, PN6H); B. americanus (Douglass, 1S99) proportions to both Bouromeryx and Cranioceras. Medium
(= Palaeomeryx americanus, Dromomeryx? americanus size.
of Douglass, 1909) (= B. parvus [Cook, 1922] [= Dromo- Average length of m2.: l S.S mm.
meryx parvus; see Voorhies, 1990b, p. A224] [= B. nebras- Included species: P. skinneri only (known from localities
censis Frick, 1937; see Skinner, Skinner, and Gooris, 1977, CP114B, CP114C*, NPll).
p. 349]) (localities [GC9C], CPllO, [CP114A], NP41B*,
NP43); B. supernebrascensis Frick, 1937 (locality Comments: Although Procranioceras is more primitive than
CPll l *); B. pseudonebrascensis Frick, 1937 (locality Cranioceras, its first known appearance in the early late
CPl 11 *);B. madisonius(Douglass, 1909) (= Palaeomeryx Barstovian may be later than the first appearance of Cranio-
madisonius, Dromomeryx? madisonius of Douglass, 1909) ceras (known possibly from at least the early Barstovian).
(localities NP41B, NP43); B. pawniensis Frick, 1937 (lo- Because Cranioceras sensu stricto is more closely related
calities GC4C, CP75C*, CP75B). to Yumaceras and Pediomeryx than it is to Procranioceras
Bouromeryx sp. is also known from localities (GClOA), (see Figure 32.3), Frick's construction of Procranioceras
CP54B, ?CP105, ?CP106, ?NPlOE, NP43, ?PN6G. as a subgenus within Cranioceras is paraphyletic and can-
not be maintained.
Comments: Frick (1937, p. 130) originally named Bouro-
meryx as a subgenus of Barbouromeryx. However, it is
clear that Bouromeryx is a more derived animal than Bar- Cranioceras Matthew, 1918
bouromeryx: considerably larger, with longer supraorbital Type species: Cranioceras unicornis Matthew, 191S.
and o~cipital cranial appendages, and having lost the upper Type specimen: AMNH 17343, late Clarendonian of
canine. All of these features indicate that Bouromeryx is Nebraska.
more closely related to the more derived cranioceratinines Characteristics: Long, upright supraorbital cranial append-
than it is to Barbouromeryx. The taxon is usually referred ages, tapering distally and curved slightly inward; round
to as its own genus (e.g., Webb, 19S3; Tedford et al., 19S7), in midshaft cross section; long median occipital cranial ap-
as we have done here. pendage, round in cross section and directed posteriorly.
Bouromeryx supernebrascensis and B. pseudonebras- Upper canine lost. Short to moderate length diastema. Pre-
censis are known only from Observation Quarry in Ne- molar row shortened; anterior fossette on p4 open. Brachy-
braska (locality CPl 11): They may represent the same dont to mesodont molars; Palaeomeryx fold on lower mo-
species, perhaps equivalent to one of the other named lars may be lost. Short, heavily proportioned limbs.
species. As noted earlier, a detailed revision of Bouromeryx Medium size.
would likely reduce the number of Included species. Be- Average length ofm2: 17.4-20.0 mm.
sides synonymy, it is possible that some species are more Included species: C. unicornis (= C. granti Frick, 1937, C.
closely related to Procranioceras than they are to B. milleri mejferdi Frick, 1937; see Webb, 1969) (known from lo-
and should be removed from Bouromeryx. calities [CC32B], ?CPl 11, CP114C, CP114D, CP115B*,
Tedford et al. (19S7) cite the first appearance of Bouro- ?CP115C, CPl 16A, CP116B); C. dakotensis Frick, 1937
meryx as late Hemingfordian. However, there are records (locality CP90A*); C. pawniensis Frick, 1937 (locality
of early Hemingfordian age, including specimens labeled CP75C*); C. teres (Cope, 1S74) (= Cosoryx teres) (=
Bouromeryx sp. in the Frick collection from the Running- Dicrocerus teres Cope 1S75, Palaeomeryx teres, Gidley,
water Formation and its equivalent in Nebraska (locali- 1907) (localities SB32D, SB32FF, SB32G, SB34A*);
ties CP105 and CP106), and a possible Bouromeryx in the C. clarendonensis Frick, 1937 (localities GC6A, GC6B,
Haystack Valley Member of the John Day Formation, Ore- [CC32A], SP2A *); C. new sp. Kelly and Lander (localities
gon (locality PN6G) in the University of California Berke- ?CC17C, ?CC17D, CC17E*).
ley collections. We have given these occurrences question- Cranioceras sp. is also known from localities (GC4E),
able status. GCSE, CCI 7G, NBS, SB32F, SPlA, CP90A, CP115A.
 Christine M. Janis and Earl Manning 486

Comments: "Dromomeryx" texanus Hay, 1924, placed in this locality, although neither Cranioceras nor P. (Yumac-
"?Cranioceras" by Frick (1937, p. 97) was subsequently eras) are mentioned by Schultz (1977), who lists only a
placed in the protoceratid genus Prosynthetoceras (see re- "palaeomerycid." The supposed P. (Yumaceras) from this
view in Patton and Taylor, 1971, p. 140). It is possible locality may represent a large individual of Cranioceras,
that several Cranioceras species are intergradational and and so has been recorded here as of questionable status.
should be synonymized. Some may also be closer to Yu- As Yumaceras and Pediomeryx are sister taxa, the deci-
maceras than to C. unicomis. sion to treat them as subgenera of the same status, or as
Tedford et al. (1987) state that the first appearance of separate genera, is largely a matter of taste. If, however,
Cranioceras is in the late Barstovian. However, some pos- it could be shown that Pediomeryx sensu stricto did not
sible earlier records exist. Kelly and Lander (1988) recorded derive from a larger form and the large size of Yumaceras
a new (unnamed) species of Cranioceras in both the late is derived for that taxon only, Yumaceras should probably
Hemingfordian and the early Barstovian of the Caliente be a separate genus.
Formation in California (localities CC17C and CC17D),
and a questioned occurrence of C. mefferdi is recorded by Pediomeryx (Pediomeryx) (Stirton, 1936); sensu Webb 1983
Frick (1937, p. 90) at the early Barstovian Survey Quarry in Type species: Pediomeryx (Pediomeryx) hemphillensis
Nebraska (locality CPll 1). Note, however, that Frick's (Stirton, 1936) (= Pediomeryx hemphillensis) (= Procoilus
C. unicornis from the early Barstovian Olcott Formation edensis Frick, 1937, Yumaceras falkenbachi Frick, 1937).
in Nebraska (locality CPl 10) was referred to Rakomeryx Type specimen: UCMP 30703, late Hemphillian of Texas.
sinclairi by Skinner, Skinner, and Gooris (1977, p. 349), Characteristics: Similar to Pediomeryx (Yumaceras) but of
and C. unicornis was said not to occur in the formation. smaller size and with more hypsodont cheek teeth.
All these pre-late Barstovian records of Cranioceras have Average length of m2: 20.4 mm.
been accorded a questionable status. Included species: P. (P.) hemphillensis only (known from lo-
calities GCl IC, [CC25B], CC40, SPID, SP3B*, ?SP4B,
[CP115D], CP123D).
Pediomeryx Stirton, 1936
Pediomeryx (Pediomeryx) sp. is also known from local-
Pediomeryx (Yumaceras) (Frick, 1937); sensu Webb, 1983
Type species: Pediomeryx (Yumaceras)figginsi (Frick, 1937) ities SP4A, ?SP4B, CP116F, (PN14).
( = Yumaceras figginsi). Comments: Dromomerycids are commonly thought to be ex-
Type specimen: Colorado Museum 214, late early Hemphil- tinct by the end of the Miocene, but a couple of records exist
lian of Colorado. for the earliest Pliocene (latest Hemphillian). The best one
Characteristics (modified from Webb, 1983): Long supraor- is of Pediomeryx (Pediomeryx) sp. from the Santee and
bital cranial appendages; nearly vertical, with retrocurved Devil's Nest Airstrip Local Faunas within the Ash Hol-
and anterioposteriorly flattened tips, round midshaft sec- low Formation in Nebraska (locality CPI 16F) (Voorhies,
tion. Occipital appendage longer, rising more nearly ver- 1990a). The Frick collection also lists an occurrence of P.
tically than in Cranioceras, with more anteroposteriorly (P.) hemphillensis at "Christian Pit 2" in Texas, which may
flattened tip. Upper canine absent. Diastema moderate; refer to the Christian Ranch locality in the upper part of
mandible with nearly straight ventral border. Premolar row the Goodnight Beds in Texas (locality SP4B).
short (p2 greatly reduced); p4 closed lingually by long
metaconid. Mesodont molars; Palaeomeryx fold absent.
Short, heavily proportioned limbs. Large size. INDETERMINATE DROMOMERYCIDS
Average length ofm2: 23.1-24.7 mm.
Fragmentary remains ascribed to dromomerycids have been re-
Included species: P. (Y). figginsi (known from localities
ported from localities GC27, GC28, NB21, NB29, ?NP38B, PNlO,
[SPIC], [SP3A], CP78*, CP116D); P. (Y). hamiltoniWebb,
?NC3B (could be the supposed bovidNeotragoceras, but see Webb,
1983 (localities GCllA*, GCllB); P. (Y). ruminalis
this volume, Chapter 34).
Stirton, 1939 (locality CC35B*).
Pediomeryx (Yumaceras) sp. is also known from local-
ities CC38, ?SPIA, SPIB, CC115C, CC116C, CP123C,
BIOLOGY AND EVOLUTIONARY PATTERNS
(PNl l).

Comments: Tedford et al. (1987) state that P. (Yumaceras) is The Miocene of the North America High Plains was warmer and less
limited to the early Hemphillian. However, Webb (1983) arid than it is today. The predominant habitat apparently changed
described P. (Y.) hamiltoni from the late Clarendonian Love from subtropical woodland and parkland in the early
Bone Bed in the Alachua Formation in Florida (locality Miocene to more open savanna in the late Miocene, resembling that
GCl lA). There is also a record in the Frick collection of present-day East Africa (Webb, 1977; Janis, 1982). Unpublished
of P. (Y.) sp. at Cole Highway Pit in the late Clarendo- work by the senior author and Kathleen Scott shows that the cranial
nian Ogallala Formation of the Texas/Oklahoma Panhan- and postcranial proportions of dromomerycids bear a strong resem-
dle (locality SPIA). Cranioceras is also recorded from blance to present-day browsing tropical and subtropical antelope in
 Dromomerycidae 487

E.81
u ~-r
~~ Ill -~
l.Hp. ::~ r------------------------------------ lii -- Q)
L.E.Hp.
7
.0 - ? 1- u!J -e0
Ill

- ~~·
e ~ -=~=
E.E.Hp.
L 8.8
L.CI.
9.5
-
w
E.CI.
11.0
-~ - - -o8 - - -
zw L.L.Ba.
12.5
- -~~
~
-Q) -
~
~
·c:
-111
e
et
~,? c3
E.L.Ba.
0
0
M
14.0 ~ _Cb - -o - e(
Cb

SE E.Ba.

5.8
~ _....E
e1
Cb
-~I_§ I
0 ? ~
-iJ - cs_~
L.Hm.
....... u
E.Hm.
17.5

18.8
'o:( ~
....c: -
E LL.Ar.
19.2

E.L.Ar.
23.0 Dromomerycini Cranioceratini
w
zW L
L.E.Ar.
ALETOMERYCINAE DROMOMERYCINAE
0 27.7
0
CJ E.E.Ar.

:i 29,4
0
Wt.
E 31.9
Or.

L.Ch.
••.•r----------------------------------------.....J
34,5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5
1
L.Un.

w 41.3

zw E.Un.

0 45.9
0 L.Br.
W M
46.7 -

M.Br.
49.0
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa. SS.O -
Ck.3 ~~:~ t:::-:=--=---------=-----------------------------------1
w . Ck.2 55.8
zw L
Ck.1
Tl.6
Tl.5
56,0
56,2
57.0
0 Tl.4
Tl.3
57.3
0 Tl.2
w Tl.1
..J
To.3
<C
c.. To.2

E i:~
Pu.2
Pu.1
Pu.O

Bl.=Blancan, Hp. = HemlJhillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un.= Uintan, Br. = Bridgerian, Wa. = Wasatchian,
Ck.= Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 32.4. Temporal ranges of dromomerycid genera and subgenera.

 Christine M. Janis and Earl Manning 488

the bovid tribe Tragelaphini (bushbuck, kudu, and bongo), and they Cranioceratini. The occipital cranial appendage in the Craniocer-
probably filled the roles of subtropical browsers in a manner similar atini may also have been used for lateral display, behavior that is
to the present-day subtropical South American cervids. The tribes common among present-day forest-dwelling ruminants. In contrast,
Dromomerycini and Cranioceratini were the most tragelaphinelike the flared base of the supraorbital cranial appendages in the Dro-
and were probably browsers in habitats ranging from dense thicket momerycini seems more appropriate for the type of frontal display
to open bushland. Subdromomeryx appears to have occupied most typical of present bushland ruminants.
open habitat, and Cranioceras, the most closed habitat. In contrast, The evolutionary radiation of the dromomerycids reflects the
members of the Aletomerycinae had longer limbs, with limb pro- changes in habitat and vegetation during the Miocene. Limb propor-
portions more similar to those seen in ecotonal, near-water-dwelling tions of the early Miocene gazelle-sized genera (20-50 kg; all body
antelope such as reedbucks (tribe Reduncini). Their skull and dental masses taken from Janis, Gordon, and Illius, 1994) Aletomeryx,
morphology suggests a more fibrous diet, foraging at ground level. Barbouromeryx, and Subdromomeryx suggest habitat types rang-
The immigration of dromomerycids to North America in the early ing from open/ecotonal (Aletomeryx) to shrubland/thicket (Bar-
Miocene apparently had a profound effect on the radiation of brows- bouromeryx and Subdromomeryx). By the end of the Hemingfordian
ing horses, forcing a "size wedge" in the range of the medium-sized (late early Miocene), the aletomerycine genera had virtually disap-
browsing ungulates. Throughout the rest of the epoch the brows- peared (although Sinclairomeryx may have persisted into the mid-
ing horses were either of smaller or of larger body size (usually dle Miocene in Saskatchewan [Storer, 1975]), possibly displaced
larger) than the contemporaneous dromomerycids (Janis, Gordon, from the open-habitat niches by the radiation of the more special-
and Illius, 1994). ized antilocaprids. In the late Hemingfordian and early Barstovian
With the exception of Aletomeryx (and possibly also Drepano- (middle Miocene) the larger, more heavily proportioned, presum-
meryx), dromomerycids were sexually dimorphic in the possession ably more dense woodland-dwelling genera first appeared. Esti-
of cranial appendages by the males only. By analogy with the inter- mated body masses for early Barstovian species range from 40 kg
relationship of sexual dimorphism in living ruminants (see Jarman, (Bouromeryx parvus) to 180 kg (Drepanomeryx falciformis and
1974; Janis, 1982), the sexes probably lived separately for most of Dromomeryx whitfordi). Some smaller taxa such as Bouromeryx
the year in most genera, with the females feeding together in small americanus (45 kg), and medium-sized (around 100 kg) taxa such
groups and the males defending feeding and reproductive territo- as Procranioceras skinneri and Rakomeryx kinseyi persisted into
ries against other males. In contrast, the presence of small cranial the late Barstovian, but by the early Clarendonian (late Miocene),
appendages in the females of Aletomeryx, combined with the in- the remaining dromomerycids were all fairly large, around 180 kg.
ference of a more open-habitat lifestyle as evidenced by the cranio- Despite the abundance of dromomerycids west of the Rockies in the
dental and limb proportions, suggest a social system more typical of earlier part of the Miocene, few were found in this region after the
open-habitat antelope (such as many gazelles), where the sexes live medial Barstovian, perhaps reflecting a change in the western habi-
together in mixed herds for most of the year, but territorial defense tats with the increasing aridity developed during the latter part of
is still seen, at least seasonally, in the males. By analogy to living the Miocene. In contrast, antilocaprids remained common in these
ruminants, the young of Aletomeryx may have been more preco- regions, probably reflecting adaptation to more open habitats.
cial than the young of the other dromomerycids, perhaps "follower" By the late Clarendonian, the heavily proportioned Cranioceratini
rather than "hider" fawns. The more extreme sexual dimorphism were the only surviving dromomerycid tribe, found mainly in the
of Sinclairomeryx (males with additional small nasal appendages central and southern regions of the United States. In the late Claren-
and possibly also with canines) combined with the presence of a donian and early Hemphillian (late Miocene), the body size of the
deep lacrimal vacuity (which houses a scent gland in deer; see re- last surviving genus Pediomeryx increased markedly, with an esti-
construction of deerlike scent marking in Figure 32.1) suggests this mated body mass of 385 kg for Pediomeryx (Yumaceras) hamiltoni
genus was highly territorial, perhaps with the males forming lek that was not even the largest species of this subgenus (Webb, 1983).
systems like some present-day reduncine bovids, for example, the The large size of P. (Yumaceras) followed a widespread trend in
kob (Kobus kob). early to middle Hemphillian mammals, with unusually large forms
Evidence that cranial appendages in dromomerycids were indeed also seen in carnivores, camels, horses, rhinos, and others. During
used for combat can be seen in the genus Sinclairomeryx, where in this time dromomerycids were known primarily from the south-
the several preserved male skulls in the Frick collection (e.g., F:AM ern part of the United States from California to Florida, this distri-
33791) one of the cranial appendages has been broken and healed bution possibly reflecting the last remaining stands of subtropical
over during the animal's lifetime. This also provides evidence that woodland. The limbs remained short, but the dental morphology
the cranial appendages were skin covered, at least in this genus. paralleled that of the aletomerycine genera in the more mesodont
The presence of a median occipital cranial appendage in the Cran- cheek teeth, the shortening of the premolar row, and the loss of the
ioceratini suggests a giraffid type of neck wrestling and butting in Palaeomeryx fold. All these features, and the more complex poste-
male/male combat, as opposed to the more typical ritualized fight- rior premolars, are suggestive of a more fibrous diet.
ing seen in most ruminants, with the head lowered and cranial ap- The number of individuals preserved in the fossil record also de-
pendages locked (see Estes, 1991). Certain living giraffe subspecies creased in the Hemphillian, and Webb (1983, p. 273) suggests that
also develop a median cranial appendage, but one situated anteri- this reflected a real decrease in the number of dromomerycid indi-
orly, on the frontal bone, rather than on the occipital bone as in the viduals during the late Miocene. Pediomeryx (Yumaceras) reached
 Dromomerycidae 489

the peak of its size increase in the late early Hemphillian, and the Douglass, E. (1899). The Miocene lake beds of Western Montana. Ph.D.
succeeding species of Pediomeryx (Pediomeryx) became increas- Thesis. Helena: University of Montana.
ingly smaller, tending back again to the size of Cranioceras (Webb, (1909). Dromomeryx, a new genus of American ruminants. Annals ofthe
Carnegie Museum, 5, 457-79.
1983, p. 273) with the late Hemphillian P. (P.) hemphilliensis hav-
Duranthon, F., Moya Sola, S., Astiba, H., & Kohler, M. (1995). Ampelo-
ing an estimated mass of around 150 kg. The trend toward smaller meryx ginsburgi nov. gen., nov. sp. (Artiodactyla, Cervoidea) et la
body size at the end of the Hemphillian is also present in several famille des Palaeomerycidae. Compte Rendu Hebdomadaire Seances
other mammalian lineages, notably horses and the rhino Teleoceras. de l'Academie des Sciences Paris, 321, 339-346.
Note that estimates of paleotemperature from North Atlantic oxy- Estes, R. D. (1991). The Behavior Guide to African Mammals. Berkeley:
University of California Press.
gen isotope values ofbenthic forams suggest a warming trend in the Frick, C. (1937). Horned ruminants of North America. Bulletin of the Amer-
late early Hemphillian, followed by a brief period of cooling in the ican Museum of Natural History, 69, 1-669.
late Hemphillian (see Opdyke, 1990), possibly correlating with this Galusha, T. (1975). Stratigraphy of the Box Butte Formation, Nebraska.
observed changes in mammalian body sizes. Bulletin of the American Museum of Natural History, 156, 1-68.
Webb (1983) has suggested that this sudden reversal in trends in Hamilton, W. R. (1978a). Fossil giraffes from the Miocene of Africa, and a
revision of the phylogeny of the Giraffoidea. Philosophical Trans-
body size may have represented a "last-ditch" evolutionary attempt actions of the Royal Society (London), Series B, 282, 165-229.
by Pediomeryx to survive in the face of a decreasing availability (l 978b ). Cervidae and Palaeomerycidae. In Evolution of African Mam-
of suitable food .. Unfortunately, this attempt did not succeed, and mals, eds. V. J. Maglio & H.B. S. Cooke, pp. 496-508. Cambridge:
dromomerycids were largely extinct by the end of the Miocene. Harvard University Press.
As previously discussed, a couple of latest Hemphillian (earliest Janis, C. (1982). Evolution of horns in ungulates: ecology and paleoecology.
Biological Reviews, 57, 261-318.
Pliocene) specimens of Pediomeryx are known, and at the Devil's
Janis, C. M., Gordon, I. J., & Illius, A. W. (1994). Modelling equid/ruminant
Nest Airstrip Local Fauna within the upper Ash Hollow Formation competition in the fossil record. Historical Biology, 8, 15-29.
in Nebraska (locality CPl 16F), Pediomeryx is found together with Janis, C. M., & Scott, K. M. ( 1987). The interrelationships of higher ruminant
the first true cervid (genus and species undetermined) (Voorhies, families with special emphasis on the members of the Cervoidea.
1990a). However, despite the fact that dromomerycids and cervids American Museum Novitates, 2893, 1-85.
Janis, C. M., & Wilhelm, P. B. (1993). Were there pursuit predators in the Ter-
may have briefly overlapped in time, it seems unlikely that deer
tiary? Dances with wolf avatars. Journal of Mammalian Evolution,
replaced the dromomerycids by direct competition, but more likely 1, 103-26.
that cervids were better adapted to survive in the type of temperate Jarman, P. J. (1974). The social organization of antelope in relation to their
woodland habitat, previously more typical of northern Eurasia, that ecology. Behaviour, 48, 213-67.
became dominant in North America toward the end of the Tertiary. If Kelly, T. S., & Lander, E. B. (1988). Biostratigraphy and correlation of
Hemingfordian and Barstovian Land Mammal assemblages, Caliente
the Isthmus of Panama had formed a couple of million years earlier,
Formation, Cuyama area, California. In Tertiary Tectonics and Sed-
the Plio-Pleistocene cervoid radiation in South America might have imentation in the Cuyama Basin, San Luis Opisbo, Santa Barbara,
been formed by,the more tropically-adapted dromomerycids, rather and Ventura Counties, California. Pacific Section S.E.P.M., 59, 1-19.
than by the more temperately-adapted cervids. Leinders, J. J.M. (1983). Hoplitomerycidae fam. nov. (Ruminantia, Mam-
malia), from Neogene fissure fillings in Gargano (Italy). Scripta Ge-
ologica, 40, 1-51.
Lull, R. S. (1920). New Tertiary Artiodactyla. American Journal of Science
ACKNOWLEDGMENTS (4th series), 50, 82-130.
MacFadden, B. J., & Webb, S. D. (1982). The succession of Miocene (Arika-
We would like to acknowledge Michael Voorhies for a critical but reean through Hemphillian) terrestrial mammal localities in Florida.
encouraging review of an early version of this chapter, and Kath- In Miocene of the Southwestern United States, eds. T M. Scott &
S. B. Upchurch, pp. 186-99. Tallahassee: Florida Department of
leen Scott for unpublished information on dromomerycid limb pro-
Natural Resources Bureau of Geology Special Publication No. 25.
portions. Much of the work done on the Frick collection of dro- Matthew, W. D. (1908). Osteology of Blastomeryx and the phylogeny of the
momerycids after the publication of Frick (1937), including many American Cervidae. Bulletin of the American Museum of Natural
identifications, was done by Beryl Taylor of the Frick Laboratory. History, 24, 535-62.
(1918). Contributions to the Snake Creek fauna. Bulletin ofthe American
Museum of Natural History, 38, 183-229.
(1924). Third contribution to the Snake Creek fauna. Bulletin of the
REFERENCES
American Museum of Natural History, 50, 59-210.
Matthew, W. D., & Cook H. J. (1909). A Pliocene fauna from western
Barbour, E. H., & Schultz, C. B. (1934). A new antilocaprid and a new cervid Nebraska_ Bulletin of the American Museum of Natural History, 26,
from the late Tertiary of Nebraska. American Museum Novitates, 734, 361-414.
1-4. Merriam, J.C. (1911). Tertiary mammal beds of Virgin Valley and Thou-
Cook, H.J. (1922). A Pliocene Fauna from Yuma County, Colorado, with sand Creek in northwestern Nevada. University of California Publi-
notes on the closely related Snake Creek beds from Nebraska. Pro- cations, Department of Geological Sciences, 6, 199-304.
ceedings of the Colorado Museum of Natural History, 4, 1-29. Opdyke, N. D. (1990). Magnetic stratigraphy of Cenozoic terrestrial sedi-
Cope, E. D. (1878). Description of new extinct vertebrates from the Up- ments and mammalian dispersal. Journal of Geology, 98, 621-37.
per Tertiary and Dakota formations. Bulletin of the United States Patton, T. H., & Taylor, B. E. (1971). The Synthetoceratinae (Mammalia, Ty-
Geological and Geographical Survey, 4, 379-96. lopoda, Protoceratidae). Bulletin ofthe American Museum ofNatural
Crusafont, M. (1961). Giraffoidea. In Traite de Palaeontologie, Tome 6, ed. History, 145, 119-218.
J. Piveteau, pp. 1022-37. Paris: Masson_ Qiu, Z., Yan, D., Hang, J., & Sun, B. (1985). Preliminary observations on the
 Christine M. Janis and Earl Manning 490

newly found skeleton of a Palaeomeryx from Shanwag, Shandong. Storer, J. E. (1975). Tertiary mammals of Saskatchewan, Part III: The
Vertebrata Palasiatica, 23, 173-95. Miocene fauna. Life Sciences Contributions, Royal Ontario Museum,
Romer, A. S. (1966). Vertebrate Paleontology (3rd ed.). Chicago: University 103, 1-134.
of Chicago Press. Stucky, R. K., & McKenna, M. C. (1993). Mammalia. In The Fossil Record
Schultz, G. E. (ed.). (1977). Guidebook: Field Conference on Late Cenozaic 2, ed. M. J. Benton, pp. 739-77. London: Chapman and Hall.
Biostratigraphy of the Texas Panhandle and Adjacent Oklahoma. Tedford, R.H., Skinner, M. S., Fields, R. S., Rensberger, J.M., Whistler, D.
Special Publication of the Killgore Research Center, Department of P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987).
Geology and Anthropology, pp. 1-160. Canyon: West Texas State Fauna! succession and biochronology of the Arikareean through
University Press. Hemphillian (late Oligocene through earliest Pliocene epochs) in
Scott, W. B. (1893). General notes: geology and paleontology. American North America. In Cenozaic Mammals of North America, ed. M. 0.
Naturalist, 27, 656-63. Woodbume, pp. 153-210. Berkeley: University of California Press.
(1895). The Mammalia of the Deep River Beds. Transactions of the Viret, J. (1961). Artiodactyla. In Traite de Palaeontologie, Tome 6, ed.
American Philosophical Society, 18, 55-183. J. Piveteau, pp. 887-1021. Paris: Masson.
Shotwell, J. A. (1968). Miocene mammals of southeast Oregon. Bulletin of Vislobokova, I. A. (1983). The fossil deer of Mongolia. Trudy, Sovmest-
the Museum of Natural History, University of Oregon, Eugene, 17, naya Sovetsko-Mongol'skaya Paleontologicheskaya Ekspeditisiya,
1-103. Moscow, 23, 1-74.
Simpson, G. G. (1945). The principles of classification, and a classification Voorhies, M. R. (l 990a). Vertebrate biostratigraphy of the Ogallala Group
of mammals. Bulletin of the American Museum of Natural History, in Nebraska. In Geological Framework and Regional Hydrology:
85, 1-350. Upper Cenozaic Blackwater Draw and Ogallala Formations, Great
Sinclair, W. J. (1915). Additions to the fauna of the lower Pliocene Snake Plains, ed. T. C. Gustavson, pp. 115-51. Austin: Bureau of Economic
Creek beds (results of the Princeton University 1914 expedition Geology, University of Texas.
to Nebraska). Proceedings of the American Philosophical Society, ( l 990b). Vertebrate paleontology of the proposed Norden Reservoir area,
54 (217). Brown, Cherry, and Keya Paha counties, Nebraska. Technical Report
Skinner, M. F., Skinner, S. M., & Gooris, R. J. (1977). Stratigraphy and 82-09, Division of Archeological Research, University of Nebraska,
biostratigraphy of late Cenozoic deposits in central Sioux County, Lincoln. Denver: U.S. Bureau of Land Reclamation. 593 pp.
western Nebraska. Bulletin of the American Museum ofNatural His- Webb, S. D. (1969). The Burge and Minnechaduza Clarendonian mam-
tory, 158, 263-370. malian faunas of north-central Nebraska. University of California
Skwara, T. (1988). Mammals of the Topham Local Fauna: early Miocene Publications in Geological Sciences, 78, 1-191.
(Hemingfordian) Cypress Hills Formation, Saskatchewan. Natu- (1977). A history of savanna vertebrates in the New World, Part 1:
ral History Contributions, Museum of Natural History, Regina, 9, North America. Annual Review of Ecology and Systematics, 8,
1-168. 355-80.
Stirton, R. A. (1936). A new ruminant from the Hemphill middle Pliocene (1983). A new species of Pediomeryx from the late Miocene of
of Texas. Journal of Paleontology, 10, 644-7. Florida, and its relationships within the subfamily Craniocerati-
(1939). Cenozoic mammal remains from the San Francisco Bay Region. nae (Ruminantia; Dromomerycidae). Journal of Mammalogy, 64,
University of California Publications, Bulletin of the Department of 261-72.
Geological Sciences, 24, 339-410. Whistler, D. D. (1984). An early Hemingfordian (early Miocene) fossil verte-
(1944). Comments on the relationships of the cervoid family brate fauna from Western Mohave Desert. Contributions in Science,
Palaeomerycidae. American Journal of Science, 242, 633-55. Natural History, Museum of Los Angeles County, 355, 1-36.
 33 Antilocapridae

CHRISTINE M. JANIS and EARL MANNING

INTRODUCTION

The Antilocapridae (Gray, 1886) are ruminants endemic to North

America. Although the family was quite diverse in the Tertiary,
the sole surviving member is the pronghorn "antelope," Antilo-
capra americana. Tertiary antilocaprids consist of the smaller late
early to early late Miocene "merycodontines" (merycodonts; a pa-
raphyletic assemblage) and the larger predominantly later Miocene
and Pliocene antilocaprines.
Antilocaprids range in age from the early Hemingfordian (late
early Miocene, although a few possible earlier records exist) to Re-
cent, and ranged geographically from the southeastern United States
through the Grnat Plains to the Pacific coast and northern Mexico.
Relatively few fossil antilocaprids are known from the northwestern
United States and Canada: However, the historical range of Antilo-
capra stretches from central Canada to northern Mexico. Recent
pronghorn range from 36-70 kg in mass (Nowak, 1991), and many
of the late Pliocene and Pleistocene antilocaprines were of similar Figure 33.1. Restoration of Merriamoceros coronatus, by Brian Regal.
body size. Merycodonts were smaller, probably of comparable size
to present-day neotragine antelope (between 10-20 kg). Recent Antilocapra is sexually dimorphic in horn size; males
Antilocapra is unique among living ruminants in the form of have large horns with a forked horn sheath, and females possess
its cranial appendages, coupling the possession of permanent un- diminutive horns with an unforked horn sheath. Merycodont (pre-
branched bony cranial appendages (horncores) with annually de- sumed) females lack cranial appendages entirely; some Plio-
ciduous forked keratinous horn sheaths. Extinct antilocaprines had Pleistocene antilocaprines have little horn sexual dimorphism.
variously branched horns that may also have been covered with a
keratinous sheath (Webb, 1973). Merycodonts have peculiar cranial
appendages: branched horns, frequently (if probably inaccurately) DEFINING FEATURES OF THE
termed "nondeciduous antlers." The covering (if any) of these cra- FAMILY ANTILOCAPRIDAE
nial appendages is unknown (see later discussion).
Antilocapra americana has a highly hypsodont dentition, but
it both browses and grazes on a variety of forbs, grasses, shrubs, CRANIAL
and cacti; grass generally comprises no more than around 12 per- Permanent bony supraorbital cranial appendages, usually
cent of the diet on a year-round basis (Hansen and Clark, 1977). forked, resulting in two or more tines. These cranial appendages
Plio-Pleistocene antilocaprines had a similar hypsodont dentition; are referred to as "horns" elsewhere in this chapter, but by analogy
most earlier taxa had a somewhat lower crowned dentition, sug- with the bony structure beneath the keratinous sheath ofbovid horns,
gesting a less fibrous diet, or perhaps a preference for less open they should more properly be referred to as "horncores". However,
habitats. this is not to imply that the horns of all antilocaprids were covered by

491
 Christine M. Janis and Earl Manning 492

Figure 33.3. Skeleton of Cosoryx furcatus (modified from Scott and

Osborn, 1890). (Note that the reconstruction of the side toes in the feet
is largely conjectural.) Bar= 10 cm. (Drawing by Brian Regal.)

Large orbits: anterior edge above M2 primitively, moving poste-

riorly with increasing hypsodonty. Postorbital bar complete, com-
posed mainly of frontal. Antorbital vacuity present, lacrimal fossa
c absent. Auditory bulla hollow, somewhat inflated. Basicranial axis
moderately to steeply inclined.

DENTAL
Upper incisors lost in all taxa. A reduced upper canine is primitively
present in an early form (small alveolus observed in one specimen
Figure 33.2. Craniodental anatomy of antilocaprids. A. Skull of Meryceros of Paracosoryx, AMNH 3119; see Janis and Scott, 1987, p. 25), and
warreni (? = Merycodus necatus) (based on USNM 123734) (modified from lost in most taxa. Lower canine incisiform, positioned with lower in-
Frick, 1937). Note the relatively low-crowned cheek teeth and unsheathed cisors and separated from cheek teeth by a long postcanine diastema,
horn with burr. Bar= 5 cm. B. Skull of Antilocapra americana (modified
as in most other ruminants. Cheek teeth relatively hypsodont even
from Vaughan, 1985). Note the posterior position of the orbit compared with
Meryceros and the extent of horncore inside of keratinous horn sheath. Bar in earliest forms, generally lacking accessory features (although a
= 5 cm. C. Dentition of Capromeryxfurcifer (based on F:AM 32983 [P2- small entostyle, ectostylid, and metastylid are primitively present in
M3] and F:AM 31749 [p2-m3] (modified from Frick, 1937). Bar= 1 cm. the less hypsodont taxon, Paracosoryx).
(Drawing by Brian Regal.)

POST CRANIAL
keratin or even by skin. Antilocaprines probably had keratinous horn
sheaths, as in the modem Antilocapra, but it is less clear whether Antilocaprids are small (merycodonts and some antilocaprines) to
these sheaths were deciduous (see later discussion). The homcore medium-sized (some antilocaprines) ruminants, with long legs and
appears to be a frontal outgrowth, as in bovids (Solounias, 1988), a gracile appearance. As in other pecoran ruminants, the distal ulna
although homology of this structure in antilocaprids with bovid is reduced, the fibula is reduced to a malleolar bone, the astragalus is
homcores is unlikely (Janis & Scott, 1987). The homcore shaft parallel sided, the metapodials are fused, and the distal metapodial
is variable in height, with circular to somewhat oval or triangular keels are complete. The metatarsal gully is closed in antilocaprids,
cross section. The tines taper distally and are often transversely and variably open or closed in merycodonts, although the closed
compressed in Antilocaprinae. condition appears to be the primitive one (Janis and Scott, 1987).
 Antilocapridae 493

The metatarsals lack the posterior tuberosity seen in dromomerycids Janis and Scott, 1987; Scott and Janis 1987), and most merycodont
and cervids, but they resemble those taxa in the possession of a specimens in fact have an open metatarsal gully, although a closed
posterior cubonavicular facet on the proximal articulatory surface gully can be variably seen in all the genera except Meryceros,
that is raised and pointed (unlike the flatter condition in bovids and and presumably represents the primitive condition for antilocaprids
giraffids). (Janis and Scott, 1987). Unfortunately the presence or absence of
The lateral digits or side toes (digits 2 and 5) and their accompa- typical cervid features in the dentition of antilocaprids, such as the
nying splints (metapodial remnants) are absent in Antilocapra, and Palaeome ryx fold, cannot be determined because of the high degree
are assumed to be absent in the Antilocaprinae in general (although of hypsodonty of even the most primitive antilocaprids; such diag-
the condition in many antilocaprine genera is actually unknown). nostic features are lost with increased hypsodonty (Janis and Scott,
Merycodonts are generally assumed to have retained reduced side 1987).
toes, but the evidence is equivocal. One complete skeleton of Para- Gentry and Hooker (1988) united antilocaprids with bovids, but
cosoryx wilsoni shows the retention of both splints and phalanges their cladogram topology is, in part, a result of the mistaken as-
in the manus, but the condition in the pes is unknown (Frick, 1937). sumption that merycodonts have a single lacrimal orifice and an
Rudiments oflateral phalanges in the man us are also known in Coso- open metatarsal gully (see also discussion in Scott and Janis, 1993).
ryx furcatus (Frick, 1937). However, the restoration of Cosoryx by Ahearn (1994) regarded antilocaprids as the sister taxon to dromo-
Scott and Osborn (1890; see Figure 33.3), showing the retention of merycids, with giraffids as the sister taxon to that grouping. Note,
splints and phalanges in the manus and phalanges in the pes, ap- however, that molecular evidence, in contrast to all morpholog-
pears to be conjectural, based on the assumption of the limbs being ical evidence, tends weakly to group Antilocapra outside of the
similar to those of the moschid Blastomeryx. other pecorans, or with the giraffids (Kraus and Miyamoto, 1991;
Jermann et al., 1995).

SYSTEMATICS
INFRAFAMILY
The Antilocapridae is sorely in need of taxonomic revision, at both
SUPRAFAMILY
generic and species levels. Here we briefly review all the Tertiary
Traditionally Antilocapridae have been allied with the Bovidae genera named in the literature (with the exception of some of Frick's
among the Ruminantia (e.g., Matthew, 1904; Pilgrim, 1941a; Simp- [1937] subgenera, which are included within their respective gen-
son, 1945). Present-day pronghorns bear a strong resemblance to era), and note where other genera have been synonymized. The aim
Old World gazelles, both in general appearance and in their behav- of this chapter is not to present a revision of the Antilocapridae (al-
ioral ecology (see discussion in Biology and Evolutionary Patterns though the cladogram in Figure 33.4 does provide a new hypothesis
section), as exemplified by their common name pronghorn "ante- of the generic-level phylogeny of the family): Rather, we present a
lope." In particular, the cranial appendage of Antilocapra, a perma- summary of the information currently available in the literature, in
nent bony homcore with a keratinous sheath, has been considered to the hope that a revision will be attempted at a later date.
be homologous to the bovid condition (O'Gara and Matson, 1975). The cladogram presented in Figure 33.4 is a recent update of one
In addition, like bovids, living pronghorns have long limbs with the originally made by one ofus (Earl Manning) in 1974. The cladogram
loss of the side toes, highly hypsodont cheek teeth, and retain the is based largely on horn characters, partly because many antilocaprid
gallbladder. lower jaw, tooth, and limb features appear to be unreliable, due to
However, none of these features serve as definitive synapomor- intergradation of character states; in contrast, male skulls, in part
phies to link antilocaprids with bovids. The deciduous, forked horn because of their rarity, provide both the gaps that create synapomor-
sheath of Antilocapra is clearly not homologous with the permanent, phies and more evidence of morphological stages in antilocaprid
unbranched horn sheath of bovids (see later discussion) because evolution.
primitive antilocaprids lacked keratinous horn sheaths. Long limbs The cladogram presented in Figure 33.4 has both similarities with,
with side toe loss and hypsodont cheek teeth have arisen conver- and differences from, other published hypotheses of antilocaprid
gently many times among ungulates in correlation with the occupa- phylogeny. The most significant difference is the abandonment of the
tion of more open habitats (see Scott and Janis, 1993) and are more paraphyletic group "Merycodontinae" (Matthew, 1909; Simpson,
likely homoplasies than homologies in this case. Additionally, the 1945). Ahearn (1988) also supports the notion of a paraphyletic
retention of the gallbladder is a symplesiomorphic ruminant feature. "Merycodontinae." Frick ( 1937) supported the sister relationship of
Leinders (1979; Leinders and Heintz, 1980) noted some morpho- Ramoceros and Merriamoceros (his subfamily "Ramocerotinae"),
logical features of Antilocapra that appear to unite it with cervids: but grouped together the other merycodonts in his "subfamily Coso-
the possession of two lacrimal orifices situated on the orbital rim and rycinae," in contrast to the present hypothesis, in which Paracosoryx
the presence of a closed metatarsal gully. Although both of these fea- is treated as the most primitive antilocaprid (as also expressed by
tures are derived for ruminants, there are problems with using them Ahearn, 1988) and Cosoryx as the most derived. The understanding
as synapomorphies to link antilocaprids with cervids in the Cer- of the transformation of a Cosoryx type of morphology to an antilo-
voidea (as shown in Figure 22.1). The number of lacrimal orifices caprine morphology would be greatly aided by the study of material
is individually variable in both antilocaprines and merycodonts (see of undescribed primitive antilocaprines in the AMNH.
 Christine M. Janis and Earl Manning 494

"MERYCODONTINAE" ANTILOCAPRINAE

Figure 33.4. Phylogenetic relationships within the Antilocapridae (version by Earl Manning). Note that the following taxa have not been included on the
cladograrn: Meryceros, Submeryceros (uncertain taxonomic status), Ottoceros (included within the group at node 12), and Ceratomeryx (included within
the group at node 24). Key to antilocaprid heads (modified from Frick, 1937 by Brian Regal): A.= Paracosoryx. B. = Merriamoceros. C. = Ramoceros. D.
= Merycodus. E. = Proantilocapra. F. = Osbornoceros. G. = Plioceros. H. = Ilingoceros. I. = Stockoceros. J. = Capromeryx. (Note: Frick's restoration
of the horn sheath of Proantilocapra was based on the supposition that this taxon was closely related to Antilocapra). Key to numbers on cladogram:
(1) ANTILOCAPRIDAE: males with posteriorly tilting frontal horns, with burr and small anterior tine; cheek teeth moderately hypsodont; canines reduced.
(2) Upper canine lost; orbit moved posteriorly, so that anterior edge of orbit is over anterior M3; cheek teeth more highly hypsodont with complete loss of
accessory features such as entostyles, ectostylids, and metastylids; remnant of fifth metatarsal fused proximally. (3) RAMOCEROTINI: medial tine added.
(4) Merriamoceros: horn beyond the shaft becomes palmate. (5) Ramoceros: one horn shaft and its anterior tine greatly lengthened, anterior tine becomes
forwardly directed, resembling a cervid "brow tine." (6) Homs rotated anteriorly, becoming vertical on skull. (7) Overall size increased, anterior horn tine
rotated medially; premolar row shortened, molars more highly hypsodont. (8) ANTILOCAPRINAE: orbit moved further posteriorly, so the anterior edge
of the orbit is posterior to M3; horn base flattened, rudimentary horn sheath probably present, burr lost, anterior tine reduced; females with small horns;
auditory bullae expanded; metatarsal gully invariably closed; metapodials proportionally lengthened, side toes (digits 2 and 5) lost. (9) Anterior tine lost
in horn (possibly still present in the sheath); horn rotated posteriorly, mimicking the primitive antilocaprid condition; anterior edge of horn becomes a
narrow ridge. (10) Osbornoceros: horn shaft and posterior tine greatly lengthened and twisted medially (forming horn resembling that of modern impala
antelope). (11) Hom base broadened and thickened (becoming oval in cross section); overall size increased. (12) ILINGOCEROTINI: horn tines begin
to twist on axis of shaft, horn base further thickened; a lump is added to the posterior edge of the horn shaft. ( 13) Hom tines further twisted; a flange is
added to the anterior edge of the horn shaft. ( 14) Ilingoceros: horn shaft and tines develop extreme twisting around the axis of the shaft. (15) Posterior tine
lengthened. (16) Anterior tine reduced. (17) Antilocapra: anterior tine lost in homcore (but is still present in sheath); cheek teeth with extreme hypsodonty.
(18) STOCKOCEROTINI: intertine horn saddle low (greatly shortening the shaft and placing more emphasis on the tines). (19) Anterior horn tine lengthened.
(20) Hayoceros: posterior tine reduced. (21) Posterior horn tine enlarged and lengthened. (22) Third horn tine added. (23) Hexobelomeryx: common horn
base broadened; two of three homcore tines share a single horn sheath. (24) Anterior horn tine becomes cylindrical. (25) Capromeryx: overall size greatly
reduced; posterior horn tine rotated anteriorly (becoming vertical on the skull). (26) Tetrameryx: overall size greatly increased; posterior horn tine greatly
lengthened.
 Antilocapridae 495

The search for an ancestor (and later, for a sister group) of Antilo- Paracosoryx Frick, 1937
capra has long been an obsession in antilocaprid systematics. Hesse Type species: Paracosoryx wilsoni Frick, 1937.
(1935) considered Capromeryx (including Texoceros) the ancestor Type specimen: F:AM 32470.
of Antilocapra. Frick (1937) placed Proantilocapra with Antilo- Characteristics: Homs with elongate shaft and anterior tines
capra in his restricted "Antilocaprinae," in contrast to Figure 33.4, either smaller or nearly equal in length to the posterior
where Proantilocapra is considered a primitive antilocaprine related tine. Horn burr(s) positioned high on the shaft. Hom shaft
to Osbornoceros. Webb ( 1973) saw the ancestry of Antilocapra lying projects posterodorsally. Diastema primitively short and
in Sphenophalos (including Plioceros ), forms thought here to be re- premolars not reduced. Small upper canine may be primi-
lated to llingoceros. In contrast, Ahearn ( 1988) made Hayoceros the tively retained. Cheek teeth primitively less hypsodont than
sister taxon of Antilocapra, with Stockoceros and Hexobelomeryx in other antilocaprids; retain small entostyle, ectostylid,
(including Hexameryx) as other members of this clade. Here Antilo- and metastylid. Lateral splints (metapodials, as well as
capra is considered derived from a Texoceros-like antilocaprine. side toes 2 and 5) retained, at least in the manus; metatarsal
Frick (1937) allied Sphenophalos and Ilingoceros (a hypothe- rudiment V primitively unfused proximally.
sis followed here) in his subfamily Ilingocerotinae, but also in- Average length of m2: 4.6 mm.
cluded Osbornoceros with this grouping. Frick's (1937) subfam- Included species: P. wilsoni Frick, 1937 (known from lo-
ily Stockocerotinae included Stockoceros, Hayoceros, Capromeryx, calities [NB! 7], [CPI05], CPl08A*, CP108B, CPllO); P.
Ceratomeryx, and Tetrameryx, a grouping comparable to that above alticornis Frick, 1937 (localities NB6C, NB6D*, NB6E,
node 18 in Figure 33.4 (with the exclusion of Hexameryx and Hexo- [NB23C], SB32G, CP108B, CPI 14D, [NP! l]); P. dawe-
belomeryx, genera which were not known to Frick). Webb (1973) sensis Frick, 1937 (localities CP107, CPI09A*); P. fur-
considered there to be a close relationship between Stockoceros and longi Frick, 1937 (localities [CC32B], NB7A*, NB19A);
Tetrameryx (to the exclusion of Hexameryx and Hexobelomeryx), P. loxoceros (Furlong, 1935; = Merycodus) (localities
based on horn similarities, in contrast to the views expressed in NB23B*, [NB23C], [PN8A]); ?P. nevadensis (Merriam,
Figure 33.4. 1911; = Merycodus) (localities [PN9B], NB18, NB20A*).
Paracosoryx sp. is also known from localities CP75C,
CP104C, CP114C, NP38C, PN8B, PN9A.
INCLUDED GENERA IN THE
FAMILY ANTILOCAPRIDAE Comments: Paracosoryx was originally described as a sub-
species of Cosoryx (Frick, 1937). Because Paracosoryx
The locality numbers listed for each genus refer to the list of unified lacks several synapomorphies of other antilocaprids
localities in Appendix I. The acronyms for museum collections are (node 2, Figure 33.4), it is clearly distinct. As treated in
listed in AppeNdix III. the cladogram here, the genus Paracosoryx is represented
The locality numbers may be listed in a couple of alternative ways. only by its type species, P. wilsoni. As the sister group of
Parentheses around the locality (e.g., [CPIOI]) mean the taxon in the entire family, it is likely that the other assigned species
question at that locality is cited as an "aff." or "cf." the taxon in of the genus may, in fact, be more closely related to other
question. Parentheses are usually used for individual species, thus antilocaprids: This can only be determined by a detailed
implying the genus is firmly known from the locality, but the actual revision of the genus.
species identification may be questionable. Question marks in front
of the locality (e.g., ?CP!Ol) mean the taxon is questionably known Ramoceros Frick, 1937 (including Paramoceros Frick, 1937)
from that locality, thus implying some doubt that the taxon is actually Type species: Ramoceros osborni (Matthew, 1904) (=
present at that locality, either at the genus or the species level. An Merycodus).
asterix (*)indicates the type locality. Type specimen: AM 9476.
Characteristics: Homs postorbital; shaft elongate, deeply in-
curved anteriorly; two or three widely flaring tines, one
"MERYCODONTINAE" MATTHEW, 1909
prong resembling a cervid "brow tine." One horn (may be
Merycodonts, as a paraphyletic assemblage, of course have no dis- the left one or the right one, depending on the individ-
tinguishing characters other than those of the family Antilocapridae. ual) approximately twice the length of the other one (not
However, in contrast to the more derived Antilocaprinae, they can illustrated in Figure 33.4).
be identified as follows. Small body size. Hom shaft round in cross Average length of m2: unavailable. Average length of M2:
section, with twinned or multiple tines, and one or more burrs usu- 12.3 mm.
ally present at the base of the shaft; horns present in (presumed) Included species: R. osborni (= R. hitchcockensis Frick, 1937,
males only, probably initially skin covered, and later naked bone to R. howardae Frick, 1937, R. kansanus Frick, 1937 [see
the level of the burr. Cheek teeth moderately to highly hypsodont. Voorhies, 1990]) (known fromlocalitiesCP76*, [CPI 14A],
Moderately elongated metapodials, remnants of side toes apparently CP114B, CP114C, CP114D, CP123A); R. ramosus (Cope,
retained. Time range of merycodont taxa: early Hemingfordian (late 1874; = Cosoryx) (including R. ramosus quadratus Frick,
early Miocene) to late Clarendonian (early late Miocene). 1937) (localities GC6B, SB32B, SB32D*, SB32F, SB32G,
 Christine M. Janis and Earl Manning 496

SB32H); R. (Paramoceros) brevicornis Frick, I937 (local- Comments: The genus Merycodus, the type of the "subfamily
ity NB6E*); R. (P.) marthae Frick, I937 (localities Merycodontinae" (Matthew, 1909), has had a problemat-
SB32D*, ?SB34A); R. (P.) palmatus Frick, 1937 (local- ical history. Leidy's original type, from the Bijou Hills
ity SB34A*). of South Dakota (locality CP89), was never figured and
Ramoceros sp. is also known from localities CC17F, has not been located since I918 (when it was mentioned
CP56, CPl 10. by Matthew [I9I8] as being present in the American Mu-
seum of Natural History). Frick (1937) did not discuss the
Merriamoceros Frick, 1937 genus except for the species "M." sabulonis (= M. necatus
Type species: Merriamoceros coronatus (Merriam, 19I3) sabulonis, probably not even the same genus as M. neca-
(= Merycodus). tus), and treated Merycodus as synonymous with Coso-
Type specimen: UCMP 20042. ryx, a view endorsed by others, including Gregory (1942)
Characteristics: Horns small; shaft short, tips palmate with and Voorhies (1969). Merycodus has also frequently been
multiple tubercles or tine points on the dorsal or dorso- synonymized with Meryceros (see discussion later). The
lateral surface (giving the appearance of miniature moose neotype of Merycodus necatus, a specimen from the type
antlers; see Figure 33. I). Mandible primitively slender, di- locality, was described by Skinner and Taylor (I 967), who
astema short, and p2 proportionally large. also distinguished the species M. necatus and "M." sabu-
Average length of m2: 10.6 mm. lonis. The neotype selected is a horn much more primitive
Included species: M. coronatus only (known from localities than that of Cosoryx, but very close to Meryceros.
NB6C*, [NB20A]).
Merriamoceros sp. is also known from localities
NB19A, CP108B, CPllO, NP38C. Meryceros Frick, 1937
Type species: Meryceros warreni (Leidy, I858) (= Cervus).
Comments: Frick, I937, originally named Merriamoceros as Type specimen: USNM I49.
a subspecies of Ramoceros. Gregory (1942) considered it Average length of m2: 8.3 mm.
as a separate genus. Distinctive as each looks, Ramoceros Characteristics: Horns forked, supraorbital and laterally com-
andMerriamoceros appear to have been derived from a hy- pressed. Horns generally larger and more laterally com-
pothetical common ancestor with a posteriorly tilting horn pressed than in Merycodus (Skinner and Taylor, I967).
shaft and a third tine added to the posterior tine. From Metatarsal gully open.
this horn morphology, Ramoceros lengthened the entire Included species: M. warreni (including M. warreni john-
appendage, resulting in a form similar in appearance to soni Frick, I937) (known from localities CP73D, CP75C,
a deer antler; Merriamoceros retained the short horn and ?CP76, CPI 14B*, CPI I4C); M. crucensis Frick, I937
added many new short tines, creating a mooselike palmate (localities SB32B, SB32F, SB32D*, SB32G, SB34A);
horn. Frick's (1937) tribe Ramocerotini would be the ap- M. hookwayi (Furlong, 1935 =Merycodus) (locality
propriate taxon for this lineage. NB23B*); M. joraki Frick, I937 (localities [NB6A],
[NB6B], [NB6C], NB6D, NB6E*); M. major Frick, I937
(localities SB32B*, SB32D, SB32G); M. nenzelensis Frick,
Merycodus Leidy, 1854
I937 (localities CP55, CPI 14B*, CPl I4C).
Type species: Merycodus necatus Leidy, 1854.
Type specimen: Neotype F:AM 51790. Comments: Merycodus and Meryceros are very similar taxa,
Characteristics: Horns with relatively short shaft and tines of and there has been much confusion between the two. Pend-
near equal size. Tines may be as long as or longer than ing full revision we regard the genera as synonymous,
length of shaft. following Gregory (I942), Forsten (1970), and Voorhies
Average length of m2: 9.4 mm (M. sabulonis). (1990). Skinner and Taylor (I967), however, argued that
Included species: M. necatus (known from localities SB32B, the horn morphology of the two genera is distinct, a con-
CP75C, CP89*, CPl 14A, NPll, [NP41B]); "M." sabulo- clusion followed by Honey and Alzett (I988) and Tedford
nis (Matthew and Cook, I909) (= M. nectatus sabulonis et al. (1987). Even if the types are distinct, the large hy-
Matthew, I924, Cosoryx [Paracosoryx] sabulonis Frick, podigm shows intergradation. Cervus warreni Leidy I 858
1937, Cosoryx [Subparacosoryx] savaronis Frick, 1937) is most likely a junior synonym of Merycodus necatus
(localities SB32H, CP54A, CP54B, [CP75C], CP108A*, Leidy, I 854.
CP108B, CPl 10, CPI I4C, NPI I).
M erycodus sp. is also reported from the following local-
ities (note: these reports may represent any merycodont): Submeryceros Frick, 1937
CCI7B, CC17C, CC17D, CC2IA, CC21B, CC2IC, Type species: Submeryceros crucianus (Frick, I937)
CC25A, CC25B, NB3D, NB23C, SB7, SB29B, SB29C, (= Meryceros [Submeryceros]).
SB37, CP45E, [NP40A], NP43, PN9B, PNlO. Type specimen: F:AM 31495.
 Antilocapridae 497

Characteristics: Horns resembling Meryceros, but much ANTILOCAPRINAE BROOKE, 1876

smaller; burrs developed separately on each tine instead
of basally on the central shaft (Frick, I937). Characteristics: Horn shaft flattened in cross section, burr absent;
horns formed from permanent bony horncores, probably encased in
Average length of m2: unavailable: smaller size than Meryc-
eros. keratinous sheaths (see later discussion), and present in both sexes
(smaller in presumed females). Cheek teeth highly hypsodont. Audi-
Included species: S. crucianus Frick, I937 (known from lo-
tory bullae enlarged. Metapodials highly elongated, closed metatar-
calities SB32A, SB32D*); ?S. minimus (locality SB32D*,
sal gully, side toes apparently lost. Time range: early late Barstovian
SB34A); ?S. minor (including S. minor pawniensis Frick,
(late middle Miocene) to Recent. Pleistocene antilocaprine taxa not
1937 and S. minor serpentinus Frick, I937) (localities
found in the Tertiary include Stockoceros, Hayoceros, and Antilo-
SB32D*, CP75C, CPI08B, CPI09A, CPllO, CP114A,
capra sensu stricto.
CP114B, NPl 1).

Comments: Voorhies (1990) recognized Submeryceros as a Proantilocapra Barbour and Schultz, 1934
distinct genus (as did Storer, 1975), based on the size and Type species: Proantilocapra platycornea Barbour and
the distinctive burr morphology of the horns. However, Schultz, I934.
here we consider Submeryceros a likely junior synonym Type specimen: UNSM 2-5-8-30.
of Merycodus. Characteristics: Horns broad, short laterally compressed (cre-
ating a sharp anterior ridge) and with the anterior tine
greatly reduced. Distal tip of horns rounded and tuber-
Cosoryx Leidy, 1869 (including Subcosoryx Frick, 1937)
ous. Cheek teeth of similar size to Merycodus, but higher
Type species: Cosoryxfurcatus Leidy, 1869.
crowned and with a smaller p3. Angle of mandible more
Type specimen: USNM 148.
prominent than in Merycodus.
Characteristics: Horns tall, slender shafted, and bifurcate,
Average length of m2: 9.0 mm.
with elongate tines; burr generally present, shaft tilted
Included species: P. platycornea only, known from locality
slightly anteriorly and laterally, and rotated so the ante-
CP1 I6A only.
rior tines are directed medially (Frick, 1937). Teeth heav-
ier, muzzle proportionally shorter, and nasals proportion- Comments: Proantilocapra, as its name implies, was de-
ally more abbreviated and slender than Meryceros (Frick, scribed as a genus ancestral to the modern Antilocapra.
1937). Cheek teeth more hypsodont and premolars more The reduced anterior tine common to both, however, ap-
reduced than in Merycodus and Meryceros (Skinner and pears convergently in several antilocaprine lineages. The
Taylor, I967). posteriorly tilting horn and its sharp anterior edge suggest
Averag~ length of m2: unavailable. Average length of M2: a relationship to Osbornoceros.
Il.2 mm.
Included species: C. furcatus (including C. furcatus stern-
Osbornoceros Frick, 1937
bergi and C. furcatus mooki Frick, 1937 = C. burgen-
Type species: Osbornoceros osborni Frick, I937.
sis Frick, I937 [see Webb, I969]) (known from local-
Type specimen: F:AM 32980.
ities NB28, [CP56], CP75C, CP90A, CP90B, ?CPllO,
Characteristics (from Frick, I937): Horns slender, directed
CP114C, CP114D*, CPII6A, NP34E); ?C. agilis
posteriorly, twisting outward with one half of a full spiral
Douglass, I899 (locality NP41B*); C. ilfonensis Frick,
expressed by the entire length of the horn. Horns with
I937 (localities SB32B, SB32G, SB34A*); C. trilater-
dorsoventrally compressed, laterally extended bases, and
alis (locality SB34A*); C. (Subcosoryx) cerrosensis Frick,
posteriorly sweeping extremities. Cross section of horn
I937 (localities [CCI7G], [CC17H], [CC32A], [NBS],
base is sharp anteriorly, deeply concave externally, and
[NB23C], SB34A*, [SPlA]).
rounded posteriorly. Compared with other antilocaprines,
Cosoryx sp. is also known from localities NB7C, CP76,
mandibular diastema is relatively short, and m3 third lobe
CP115B.
is primitively small.
Comments: Many authors have, incorrectly in our opinion, Average length of m2: unavailable. Average length of M2:
treated Cosoryx as a junior synonym of Merycodus (e.g., Il.3 mm.
Forsten, I970; Voorhies, I990). As constructed in Figure Included species: 0. osborni only (known from localities
33.4, a cladistic analysis of primitive antilocaprids places [SB33II], SB34C*).
Cosoryx closer to the Antilocaprinae than Merycodus. Osbornoceros sp. is also known from locality SPIC.

Indeterminate Merycodonts Plioceros Frick, 1937

Fragmentary remains ascribed to merycodonts are reported from Type species: Plioceros blicki Frick, I937.
localities GC6B, NP34D. Type specimen: F:AM 3I682.
 Christine M. Janis and Earl Manning 498

Characteristics: Homs laterally compressed, with two tines; Flange present on the anterior edge of the horn shaft. Cheek
shaft tall and transversely compressed; anterior tine some- teeth almost as hypsodont as in Antilocapra.
what shorter and narrower than posterior tine; cross section Average length ofm2: 13.9 mm.
of horn at midshaft dumbbell shaped. Premolars with p3 Included species: S. nevadanus (known from localities
and p4 metaconid and entoconid not connected; long di- NB3 l *, SB9, PNl I); S. middleswarti (localities CPI 16D*).
astema; moderate-sized m3 talonid. Sphenophalos sp. is also known from localities CC25B,
Average length of m2: 11.2 mm. (CC37), PN12, PNl3.
Included species: P. blicki (known from localities SB32D*,
SB32F, SB34A, [SB34C], [NP45]); P. dehlini Frick, 1937 Comments: The lateral twisting of the horns suggests a rela-
(locality CP116B*); P. fioblairi Frick, 1937 (localities tionship to Ilingoceros.
CP90A, CP114C*, CP116B).
Plioceros sp. is also known from localities NB7C, NB7D, Ilingoceros Merriam, 1909
NB7E, SB33II, CPI 16B, (CPI 16C), ?CP123C. Type species: Jlingoceros alexandrae Merriam, 1909.
Type specimen: UCMP 11880.
Comments: Plioceros has been included within the genus
Characteristics: Homs approximately circular in cross section
Sphenophalos (Furlong, 1932; Stirton, 1938; Webb, 1969,
with two well-developed spiral ridges. The spiral ridges ex-
1973). In contrast, Frick (1937) included Sphenophalos
tend the length of the horn, which may end with a terminal
within Plioceros. The lump on the posterior edge of the
fork.
horn shaft clearly shows a relationship to Sphenophalos.
Averagelengthofm2: 15.5 mm (estimated from Frick, 1937).
It is likely that some "Plioceros" species are closer to
Included species: /. alexandrae (known from localities
Sphenophalos than they are to P. blicki and should be trans-
[NB I 9D], NB3 l *, [NB35], SB34C, SB60); I. schizoceras
ferred to Sphenophalos.
(Merriam, 1911) (locality NB31*).
Ilingoceros sp. is also known from localities SB9, NB7E,
Ottoceros Miller and Downs, 1974 NBIO, PNl2.
Type species: Ottoceros peacevalleyensis Miller and Downs,
1974. Comments: Merriam (1909) originally considered Ilingo-
Type specimen: LACM 1374. ceros to be a strepsicerine (= tragelaphine) antelope, based
Characteristics: Homs transversely flattened, with tall shaft on its possession of twisted horns, resembling the twisted
and two short tines; presence of lateral flange, just above or coiled horns of modern antelope such as the bongo and
the orbit, on horn base. thekudu (genus Tragelaphus). However, he later(Merriam,
Average length of m2: unavailable. Length of M2: 12.7 mm 1911) recognized it as an antilocaprid, based in part on its
(estimated from Miller and Downs, 1974). high degree of hypsodonty. Frick's (1937) tribe Ilingo-
Included species: 0. peacevalleyensis only, known from lo- ceratini, rather than Plioceratini, would be the appropriate
cality CC41 only. taxon for the Plioceros/Ottoceros!Sphenophalos/
Ilingoceros lineage.
Comments: Ottoceros is known only from elements of five
individuals excavated in slabs of limestone from the Kin-
sey Ranch Local Fauna. The specimens were transversely Texoceros Frick, 1937 (synonym: Dorcameryx; Capromeryx,
flattened in diagenesis thereby obscuring many cranial fea- in part)
tures. Dentally, Ottoceros lacks the derived characters of Type species: Texoceros altidens (Matthew, 1924) (=Mery-
Antilocapra and closely related taxa, such as paraconid- codus) (= Capromeryx altidens [Hesse, 1935], C. texanus
metaconid closure in p4 and development of a fourth lobe [Hesse, 1935], Dorcameryx optimae Matthew, 1932; see
on m3. The flange on the horn shaft suggests a relation- Frick, 1937, Dalquest, 1983).
ship to the Plioceros!Sphenophalos!Ilingoceros group, al- Type specimen: AMNH 18981.
though we have not placed it with confidence in the clado- Characteristics: Horns with two tines, posterior tine length-
gram in Figure 33.4. ened, anterior tine broadened and shortened. Anterior tine
rotated outwardly and posterior tine rotated inwardly. Post-
Sphenophalos Merriam, 1909 canine diastema twice the length of premolar row, jaw not
Type species: Sphenophalos nevadanus Merriam, 1909. shortened as in Capromeryx; m3 lobe expanded.
Type specimen: UCMP 11887. Average length of m2: unavailable. Average length of M2:
Characteristics: Homs positioned above posterior portion of 13.6mm.
orbit, tilting somewhat posterolaterally from the base and Included species: T. altidens (known from localities [CCl 7I],
the anterior tine twisting laterally. Hom transversely com- SPIC, SPlD, [SP3B], ?CP115C, CP115D*); T. edensis
pressed with anterior edge more compressed than the pos- Frick, 1937 (locality CC40*); T. guymonensis Frick, 1937
terior edge. Cross section of horn base wedge shaped. (localities SPlD*, [CP116D], [CPl23D]); ?T. minorei
 Antilocapridae 499

Frick, 1937 (localities ?GC6C, SB9, SB 11, SPIC, SPlD*, Comments: The very short horn shaft unites Frick's (1937)
?CPl23C); ?T. vaughani Frick, 1937 (locality CP78*). tribe Stockoceratini, comprising Hayoceros, Stockoceros,
Texoceros sp. is also known from localities NB33, SB I 0, Hexameryx, Hexobelomeryx, Capromeryx, Ceratomeryx,
?SB14A, CPl23D. and Tetrameryx. The first four of that group are united
by the lengthened anterior tine (see Figure 33.4): Note
Comments: Morgan and Morgan ( 1995) provide a discussion
that Hayoceros and Stockoceros are solely Pleistocene in
of the referral to Texoceros of species previously assigned
occurrence, and so are not discussed here.
to Capromeryx. The combination of relatively large size,
Webb (1973) noted that the sample of horns in the Bone
derived diastema and m3, and, especially, the relatively
Valley district of Florida (locality GCl3B) can be divided
long horn shaft and reduced anterior tine suggests a rela-
into two samples based on size (with some additional
tionship between T. guymonensis and Antilocapra.
differences in morphology), although no dimorphism is
apparent in the cheek teeth. H. simpsoni is representa-
?Antilocapra (Subantilocapra) Webb, 1973
tive of the smaller group, whereas H. elmorei (White,
Type species: Antilocapra (Subantilocapra) garciae Webb,
1942) is representative of the larger group. Webb consid-
1973.
ered these two morphs to represent the sexes of a single
Type specimen: UF 17354.
species (H. simpsoni), rather than two species, with the
Characteristics (modified from Webb, 1973): Homs with
larger (H. "elmorei") horns presumably belonging to the
elongate shaft, slight bifurcation at the tip, and little di-
males.
vergence between the two unequal tines. The anterior tine
Hexameryx resembles Hexobelomeryx in possessing a
is smallerthan the posterior. Shaft of horn resembles that of
total of six tines (three on each horn). However, in Hex-
Antilocapra, transversely compressed with narrower ante-
ameryx each of the three tines diverges widely from the
rior edge, medial side convex, lateral side flatter, shaft curv-
base above a short common shaft near the orbit, each tine
ing slightly outward along its long axis. Upper molars with
having its own separate sheath. In contrast, in Hexobelom-
strong styles but weak ribs. Labial inflection on p4 does not
eryx two of the three tines lie close to each other, proba-
close off to form a fossettid (as in Tetrameryx and Antilo-
bly sharing a single horn sheath, and the common shaft is
capra), but remains primitively open (as in Sphenophalos).
longer than in Hexameryx (Webb, 1973). Thus, despite the
Lower molars with nearly fiat lingual walls. Cheek teeth
dental similarities between Hexameryx and Hexobelom-
less hypsodont than in Antilocapra.
eryx, Webb (1973) considered the two forms to represent
Average length of m2: I 0.1 mm.
separate genera. However, Hexameryx and Hexobelom-
Included species: A. (S.) garciae only, known from locality
eryx were considered congeneric by Simpson (1945) and
GCl3,B only.
Ahearn (1988).
Comments: Antilocapra, sensu stricto, is recorded only from
the Pleistocene and Recent, and so is not included here.
Hexobelomeryx Furlong, 1941
Richards and McCrossin (1991) consider A. (S.) garciae
Type species: Hexobelomeryx fricki Furlong, 1941.
to belong to the genus Sphenophalos, as do we.
Type specimen: CIT 2780.
Characteristics: Three-tined horns (resulting in a "six-
Hexameryx White, 1941
homed" appearance) with unequal, tapering tines. Ante-
Type species: Hexameryx simpsoni White, 1941.
rior tine shortest with other two more nearly equal. Angle
Type specimen: MCZ 3689.
of divergence between tines either nearly equal or middle
Characteristics (modified from Webb, 1973): Supraorbital
and posterior tine less divergent than anterior and middle.
horns with three tines (resulting in a "six-homed" ap-
Upper fourth premolar with strong labial ribs and without
pearance), the saddle between the tines lowered, greatly
middle rib (unlike Antilocapra). The m3 with three major
shortening the horn shaft; tines subcircular in cross sec-
lobes and a small but variable fourth lobe.
tion, each widely divergent and separately sheathed; ante-
Average length ofm2: 13.3 mm.
rior and middle tines tilting dorsolaterally, posterior tine in
Included species: H.fricki only (known from localities SB60*,
vertical plane; anterior tine shortest, posterior tine longest. SB61).
Upper fourth premolar with strong labial ribs and without
Hexobelomeryx sp. is also known from localities CA4,
middle rib (unlike Antilocapra). Upper molars less hyp-
?NB34, SB58C, SB59, (SP4B).
sodont and with weaker styles than in Antilocapra, resem-
bling the teeth of Stockoceros. Third lower molar with three Comments: Webb (1973) noted a distinctive sexual dimor-
major lobes and small but variable fourth lobe. phism in the horns of Hexobelomeryx: In the presumed
Average length of m2: 13.3 mm. females the two anterior tines are closely paired and share
Included species: H. simpsoni only (= H. elmorei White, a common sheath; in the presumed males it is the two pos-
1942) (known from localities GCl2II, GCl3B*). terior tines that are twinned in this fashion.
 Christine M. Janis and Earl Manning 500

Capromeryx Matthew, 1902 (synonym: Breameryx anterior in position with respect to dentition. Lambdoid
Furlong, 1946) crest well developed; broad, prominent median ridge on
Type species: Capromeryxfurcifer Matthew, 1902. occiput. Basioccipital narrow, auditory bullae inflated pos-
Type specimen: AMNH 2771. terolaterally.
Characteristics: Small-sized antilocaprines with two-tined Average length ofm2: unknown. Skull "appreciably smaller"
horns (resulting in a "four-homed" appearance). Paral- than in Antilocapra (Gazin, 1935).
lel, near-vertical tines situated supraorbitally, with anterior Included species: C. prenticei only, known from locality
tine cylindrical in cross section and usually shorter than PN23Aonly.
posterior. Posterior tine flattened, rotated anteriorly, and
Comments: C. prenticei is known from two partial skulls and
deeply grooved posteriorly. A prominent temporal ridge
associated postcranials from the Hagerman Local Fauna
(parasagittal crest) forms the dorsal border of the temporal
ofldaho. Gazin (1935) considered this genus to be closely
fossa; auditory bullae greatly enlarged. Cheek teeth as hyp-
related to Tetrameryx, although it has not been exactly
sodont as in Antilocapra; diastema shorter than in Antilo-
placed in the cladogram in Figure 33.4.
capra; short premolar row; m3 lacks the fourth lobe seen
in Antilocapra; and there is a trend toward p4 metaconid-
Tetrameryx Lull, 1921
entoconid closure.
Type species: Tetrameryx shuleri Lull, 1921.
Average length ofm2: 9.7 mm (C.furcifer).
Type specimen: SMU 1.50.
Included species: C. arizonensis Skinner, 1942 (including
Characteristics: Two-tined horns (resulting m a "four-
C. arizonensis schultzi = Breameryx arizonensis) (known
homed" appearance), with deep division between ante-
from localities GC14, SB18A, SB18B, SB18C, CP117A,
rior and posterior tines; horn shaft greatly reduced; poste-
CP1l7B, and from the early Irvingtonian); C. gidleyi (lo-
rior tines much longer than anterior ones. Tines separately
cality SB14C, and from the Pleistocene); C. tauntonensis
sheathed, as in Stockoceros, as indicated by the basal rings
Morgan and Morgan, 1995 (localities [CPl 18], PN4*); and
of nutrient foramina; little sexual dimorphism in horn size
Pleistocene species C.furcifer (= C. minimus, Breameryx
(Webb, 1973). Second and third molars larger than in An-
minimus; late Irvingtonian to early Rancholabrean), C. mi-
tilocapra, premolars smaller; mesostyle more pronounced
nor (Taylor, 1911; = Breameryx minor; Rancholabrean),
on M3, butM3 lacks distinct metaconule; external styles on
and C. mexicanus (Furlong, 1925; = Breameryx mexi-
premolars less conspicuous. Degree of hypsodonty similar
canus; Rancholabrean).
to Antilocapra.
Capromeryx sp. is also known from Pliocene localities
Average length of m2: unknown. Length of M2: 17.5 mm.
NB13C, (SB12), (SB35), ?SB50, SPlH, ?SP5, CP121,
Included species: T. shuleri only (known from the Pleis-
CP131.
tocene).
Comments: The status of the genus Capromeryx was exten- Tetrameryx sp. is known from Pliocene localities
sively reviewed by Morgan and Morgan (1995). Capro- CC25B, NB13C, (SB15), SPlH.
meryx is an unusually small antilocaprine. The earliest
Comments: Tetrameryx was synonymized with Stockoceros
species, C. tauntonensis, was relatively large, about 85
by Colbert and Chaffee (1939), but the anterior and pos-
percent of the size of present-day Antilocapra: The genus
terior horn tines of Stockoceros are approximately equal
then underwent progressive dwarfing, reaching its small-
in size, unlike the condition in Tetrameryx. Although the
est form in the Rancholabrean (late Pleistocene) (Morgan
cylindrical shape of the anterior tine shows a close rela-
and Morgan, 1995). Changes also occurred in the form
tionship between Capromeryx and Tetrameryx, it is clear
of the horns, with progressive posterior movement of the
that the two genera evolved in separate directions. Like
horns (from supraorbital to more postorbital), a relative
many Irvingtonian (early Pleistocene) mammals, T. shu-
shortening of the tines, and a reduction in horn robust-
leri is unusually large for its family.
ness, especially in the anteroposterior direction (Morgan
and Morgan, 1995). Because of this regular decrease in
Indeterminate Tertiary antilocaprines
size through time, Capromeryx is one of the most useful Fragmentary remains assigned to indeterminate antilocaprines are
time indicators of the family. known from localities WM26, GC28, CC36, CC38, CC43, CC53,
NB27B, SB9, SB16, SB17, SB31C, SB58C, SB63, SB65, CP116E,
Ceratomeryx Gazin, 1935 CP116F, CP128C, CP132B, PN14, PN23C.
Type species: Ceratomeryx prenticei Gazin, 1935.
Type specimen: USNM 13760.
INDETERMINATE TERTIARY ANTILOCAPRIDS
Characteristics: Two-tined horns with tines well separated
and transversely flattened (resulting in a "four-homed" ap- Fragmentary remains assigned to indeterminate antilocaprids are
pearance). Anterior tines much larger than posterior; an- known from localities GCllA, GC12, CC26B, NB19C, NB27A,
terior tine positioned directly above orbit. Orbit relatively SB6.
 Antilocapridae 501

BIOLOGY AND EVOLUTIONARY PATTTERNS of a periodic "savanna corridor" during the Pleistocene (e.g., Webb,
1991), although antilocaprids were unknown south of Chihuahua,
northern Mexico.
ANTILOCAPRID ORIGINS AND BIOGEOGRAPHY
Antilocaprids are known only from North America, but their origins
must lie in the Old World because there are no pecoran ruminants
EVOLUTIONARY TRENDS IN
in North America prior to the early Miocene. Small hypsodont taxa
THE ANTILOCAPRIDAE
are known from the Oligocene and Miocene of Mongolia, such as As presented in the cladogram (Figure 33.4), certain trends can be
Palaeohypsodontus from the middle Oligocene Hsanda Go! Forma- perceived in the evolutionary history of antilocaprids.
tion. Such taxa have been proposed as being ancestral bovids or an- Horns: Primitively, antilocaprids had cylindrical, posteriorly tilt-
cestral antilocaprids: Unfortunately, the teeth have no distinguishing ing, supraorbital frontal horns composed of (probably) exposed
features that would ally them with any particular ruminant family bone, with one or more burrs, present in males only. The hypothet-
(see discussion in Janis and Scott, 1987, p. 74). Palaeohypsodontus ical ancestor probably had a single-pointed, burrless horn, but the
would be unusually early for a bovid, but not for an antilocaprid. most primitive known forms have a small anterior tine (in addition
Antilocaprines are primarily Hemphillian and Plio-Pleistocene to the primary posterior tine) and a burr. In more derived early an-
in distribution; merycodonts are primarily late early and middle tilocaprids, the anterior tine became equal in length to the posterior
Miocene in distribution, with a few taxa overlapping with the ear- one, and the shaft rotated anteriorly, making the shaft nearly vertical
liest antilocaprines in the Clarendonian (see Figure 33.5). Ungulate over the orbit (see Figure 33.2A). One early group (Ramoceros and
communities suggestive of present-day types of savanna habitats Merriamoceros) added extra tines and formed complex horns resem-
were first present in North America in the Clarendonian, and by bling deer and moose antlers, respectively. In the more derived an-
the Hemphillian savanna-woodland habitats were apparently largely tilocaprines, the horn was changed from a short, bifurcate, horn with
lost, with open grassland becoming the primary habitat (see Janis, a cylindrical shaft and a burr, into one with a transversely flattened
1993). Thus the early radiation of antilocaprids (with a maximum shaft and without a burr, probably encased in a keratin sheath. This
number of merycodont genera [seven] in the early Barstovian) ap- sheath would act both to strengthen the horn in intraspecific combat
pears to be one associated with a more woodland environment. The and to make it more visible in lateral view. This change in horn
radiation of antilocaprines comprises three phases: an early phase morphology may reflect a change in fighting style from stabbing
in the late middle to early late Miocene (early late Barstovian to late at the opponent to grappling with interlocking horns. Later antilo-
Clarendonian), when the genera Proantilocapra and Plioceros coex- caprine groups evolved horns that were impalalike (Proantilocapra
ist with the merycodonts; a middle phase in the late Miocene to ear- and Osbornoceros), twisted (Plioceros, Sphenophalos, and !lingo-
liest Pliocene (Hemphillian) with the radiation oflarger genera with ceros), with reduced anterior tines (Texoceros andAntilocapra), with
more complex 'horn types, such as Ilingoceros and Hexobelomeryx, lengthened anterior tines and triple tined (Hayoceros, Stockoceros,
achieving a maximum generic diversity of antilocaprines (seven) in Hexameryx, and Hexobelomeryx), and with a cylindrical anterior
the late early Hemphillian; and a third, less diverse phase of taxa tine (Tetrameryx and Capromeryx).
with two-tined horns in the Pliocene, continuing into the Pleistocene Teeth: The primitive ruminant condition is for large, caniniform
(Blancan to late Rancholabrean), with the larger genus Tetrameryx upper canines, which were lost early in antilocaprid history. The
and the smaller genus Capromeryx (see Figures 22.8 and 33.5). anterior premolars generally were progressively reduced in size,
The distribution of antilocaprids most likely reflects aspects of increasing the length of the already long diastema (probably reflect-
their ecology. Like most mammalian taxa, they are diverse in the ing an increasingly fibrous diet; see Janis, 1995). The loss of pl and
Great Plains, probably because most of our fossil data come from the molarization of P4/p4 probably occurred several times conver-
this area (as a result of abundant, long-term sedimentary deposi- gently within the family, and an additional posterior lobe was added
tion), and thus it represents the best sampling. However, especially to M3/m3 in derived forms (see Figure 33.2C). The cheek teeth were
in comparison with the contemporaneous dromomerycids (see Fig- primitively moderately hypsodont and became progressively more
ure 22.8), antilocaprids were common in the northern and southern highly crowned. Accompanying the increasing hypsodonty, to allow
Great Basin areas, but rare in Gulf Coast faunas until the latest room for the lengthened developing molar crowns, the mandibular
Miocene. This distribution probably reflects a preference for more ramus was deepened and the orbits were moved posteriorly.
arid habitats, as seen in pronghorns today; Antilocapra inhabits open Skull: In addition to the posterior shifting of the orbits, which
grasslands and deserts in the western North America, originally oc- also resulted in the modification of the zygomatic arch and the pos-
curring from eastern Washington and southern Manitoba to Baja torbital bar, antilocaprid skulls changed from a very gracile type
California and northeastern Mexico (Nowak, 1991). primitively to a heavy thick-boned type in more derived forms,
Unlike other native ungulates, such as horses, camels, and tapirs, presumably in association with the need to support heavier horns.
antilocaprids did not migrate into the Old World or into South The muzzle became deepened and lengthened. The sagittal crest
America. Perhaps they were more restricted in their ecology and became parasagittal, and the lambdoidal crest broadened with the
habitats than these other ungulates: Cold weather may have pre- relative enlargement of the braincase in the antilocaprines. The au-
vented them from migrating to Eurasia via Beringia, but their ab- ditory bullae became progressively enlarged and flattened trans-
sence from South America causes some problems with the notion versely.
 Christine M. Janis and Earl Manning 502

w
Z L L.BI.
w E.81
0
0 E
:i
c..
L L.CI.

E.CI.
w
z L.L.Ba.
~
w
0 M E.LBa.
12.5
...
0
0 14.0

:E E.Sa.

5.8
8 ~
17.5
~
E.Hm.
18.8
E L.L.Ar.
19.2

E.L.Ar.
"MERYCODONTINAE" ANTILOCAPRINAE
23.0

w LE.Ar.
z
W L
0 27.7
0
CJ E.E.Ar.
:i 29.4
0
Wt.
E 31.9
o •.
33.4
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5

L.Un.
41.3
w
z E.Un.
w
0 45.9
0 L.Br.
W M 48.7

M.Br.
49.D

E. Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0 -
55.5
Ck.3 55.6
w Ck,2 55.8
z Ck.1
Tl.6
56.0
56.2
W L Tl.5 57.0
0 Tl.4
Tl.3
57.3
0 Tl.2
w Tl.1
..J
To.3
<C
c.. To.2
To.1
E Pu.3
Pu.2
Pu.1
Pu.O

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordlan, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Brldgerian, Wa. = Wasatchian,
=
Ck. = Clarkforkian, Ti. Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 33.5. Temporal ranges of antilocaprid genera.

 Antilocapridae 503

Limbs: The feet lengthened, the metatarsal gully became invari- Antilocapra itself is the most highly hypsodont of the antilocaprids,
ably closed in the antilocaprines, and the side toes (digits 2 and 5) yet its diet includes only around 12 percent grass (Hansen and Clark,
were progressively reduced, and lost entirely in the antilocaprines. 1977). A more open-habitat lifestyle in the antilocaprines also ac-
These modifications for a greater degree of cursoriality, together cords with postcranial morphological evidence and paleobiological
with the increasing hypsodonty suggesting an increasingly abrasive evidence. In contrast to merycodonts, antilocaprines have lost the
diet, most likely related to a change in habitat from more closed side toes, and their metapodials are proportionally longer than those
woodlands to a more open savanna or plains environment. of merycodonts (Kathleen Scott, pers. comm.). Much of the mor-
Body size: The trend in antilocaprids was generally toward in- phology of Antilocapra reflects adaptation for speed; in addition to
creasing body size, although Capromeryx became progressively re- the elongated legs, it has a highly flexible spine and enlarged heart
duced in size throughout the Pleistocene. and lungs. It can run at speeds of up to 60 mph, making it the fastest
land animal in North America.
The sexually dimorphic horns of modem pronghorns, that of
ANTILOCAPRID ETHOLOGY, ECOLOGY,
AND PALEOECOLOGY smaller horns being present in the females (as is also true of gazelles,
some of which have hornless females), is typical of that of antelope
Present-day pronghorns show many parallels with gazelles in their in Jarman's (1974) Category C: part-time territoriality in open habi-
behavior as typically open-habitat ruminants. Like most gazelles, tats, with males defending resource-based territories. The presence
pronghorns are intermediate feeders (grass is taken only during the of small horns in the females of other, extinct, antilocaprines indi-
periods of spring and summer growth), and they live in open habi- cates a similar type of behavioral ecology. Several antilocaprine gen-
tats (grasslands, brushlands, and deserts). Pronghorns also resemble era (notably Texoceros, Stockoceros, and Capromeryx) are known
gazelles in their courtship behavior and in the stereotyped approach from large fossil populations, suggesting large herds.
of the male to the female (Bromley and Kitchen, 1974). Pronghorns The absence of horns in female merycodonts, in combination
give birth to a single fawn or to twins: The fawns are of the "hider" with the postcranial evidence of a preference for less open habi-
type and lie in seclusion for up to 90 percent of daytime until three tats, is suggestive of a more solitary lifestyle, perhaps with the
weeks or so following birth (Fitcher, 1974). males holding year-round territories like the antelope in Jarman's
Pronghorn males are territorial during the spring and summer Category B. However, many merycodonts (especially Meryceros,
months, when they defend territories large enough to provide for- Ramoceros, and Cosoryx) are often found in great numbers at fos-
age for themselves and the females (2-12) they attract for the season sil sites, perhaps reflecting herd associations. Certainly the more
(Kitchen and Bromley, 1974). They also have a similar type of terri- primitive morphology of Paracosoryx (which is not found in large
torial behavior to gazelles, following a ritualized urination and defe- numbers) is probably indicative that, out of the merycodonts, it had
cation sequence in territory marking, and also marking vegetation the least fibrous diet (lesser hypsodonty), the most preference for
with the cheek patches, which seem to have an analogous function closed habitats (relatively more primitive limb morphology), and
to the preorbital glands of certain bovids and cervids (Kitchen and so was probably the least social genus. However, most antilocaprid
Bromley, 1974). The black cheek patches (seen in the males only) genera appear to have been open-habitat, gregarious animals.
are also used in agonistic display (Kitchen and Bromley, 1974). In summary, although antilocaprines appear to be analogous in
Pronghorns fight horn to horn, with the fight consisting of a series their morphology and behavior to the open-habitat, group-forming
of thrusts and counterthrusts, apparently trying to gore the opponent. gazelles, the most primitive forms appear to be more similar to the
When the horns are locked, each male attempts to push the other off neotragine antelope, which are more solitary (or pair forming) in
balance, and each attempts to gain the uphill advantage (Kitchen and social behavior and more ecotonal in their habitat choice.
Bromley, 1974). Male/male fights are fairly frequent, although of
short duration, and injuries are fairly common. The severity, style,
EVOLUTION OF ANTILOCAPRID
and frequency of fighting bouts in pronghorn match the predictions
CRANIAL APPENDAGES
of Geist ( 1966) from horn morphology. That is, if, as in pronghorns,
the horns are relatively short and daggerlike, combat will be rare The unique nature of the cranial appendages of the modem prong-
but serious and the skin should be thickened in areas where injuries horn has already been mentioned. Although the deciduous kerati-
are most likely to occur. The skin is indeed thickened over the head, nous sheath of Antilocapra is often considered homologous with the
neck, and rump patch of male pronghorns (Kitchen and Bromley, permanent sheath of bovids (e.g., O'Gara and Matson, 1975), this
1974). is impossible if primitive antilocaprids lacked sheaths (see later).
To what extent might the behavior of present-day pronghorns Moreover, growth of the keratinous sheath is unlike that of bovids.
be applicable to extinct antilocaprids? All antilocaprids combine a The sheath in bovids grows by laying down incremental cones of
narrow muzzle with a high degree of hypsodonty, suggestive of an keratin directly over the bony core: Although the surface layers of
intermediate feeding style (see Janis, 1995). The greater degree of older keratin may be exfoliated, the entire structure is never shed,
hypsodonty, and the reduction of the premolar row in favor of the and exfoliation is not associated with seasonal reproductive pat-
molars, of the antilocaprines in comparison with the merycodonts terns, as in Antilocapra. The keratinous sheath may be considerably
is probably more reflective of open habitats (due to increased grit longer than the bony homcore in both groups, but although bovid
and dirt on the food; see Janis, 1988) than ofa more fibrous diet: horns can be curved or spiraled, they are never branched or forked:
 Christine M. Janis and Earl Manning 504

The bovid mechanism of growth of the sheath would not allow it to (e.g., Hexameryx, Hexobelomeryx, and Tetrameryx) may have had
spread outward over a bony fork (see discussion in Janis and Scott, individual keratin sheaths for each tine: If this was the case, then it
1987, p. 13). would be possible for their sheaths to be deciduous.
In Antilocapra, the (branched) keratinous sheath is laid down In contrast with the horns of antilocaprines, which have a pitted
over an (unbranched) skin-covered horncore; new sheath initially texture indicative of a keratinous sheath, the horns of merycodonts
forms distal to the horncore tips, keratinization of the rest of the are smooth, suggesting an initial covering of skin, as first noted by
skin proceeds downward toward the horn base, and then growth Cope ( 1877) and later by Scott and Osborn ( 1890). The skin covering
from the base forces the pronged tip of the sheath outward from the suggests the velvet of deer antlers. The variable presence of "burrs"
tip of the horncore itself (see Figure 33.2B) (O'Gara, 1990). This around the base of the merycodont horns appears similar to the burrs
basal growth, pushing the older keratinous part of the sheath distally, at the base of deer antlers. However, merycodont horns could not
eventually results in the shedding of the sheath; after shedding, have been deciduous like antlers: Neither crania with cast horns nor
the underlying horncore remains covered with skin and has usually cast horns themselves are known (Furlong, 1927). Moreover, there
already begun a new process of keratinization at the tip (O'Gara, is no evidence of merycodont crania forming an age series with
1990). horns having an increasing number of tines, as is true for deer, and
The growth and shedding of the horn sheath in Antilocapra is unbranched horns are found only in the most immature specimens
related to seasonal reproduction and under hormonal control, as (Voorhies, 1969). Histologically, the bone of merycodont horn is
is the case with cervid antlers. Interestingly enough, the horns of the same above and below the burr (Voorhies, 1969), unlike deer
female Antilocapra are not shed, although shedding of the antlers antlers, which differ greatly from the structure of the pedicle bone
does occur in female reindeer (Rangifer tarandus, the only cervid below the burr.
with antlered females) (O'Gara, 1990). Merycodont burrs (or "pseudoburrs," as they are sometimes term-
Extant male pronghorns occasionally possess additional horns: ed) are quite unlike the burrs of deer antlers, which mark the point of
Sometimes the accessory horn is present in the keratinous sheath separation between the base of the deciduous antler shaft and the top
alone; at other times there is a small horn posterior to the regular of the permanent pedicle and are cast along with the antler (Goss,
one, resembling the condition in "four-horned" antilocaprines such 1983). Ifmerycodonthorns were lost annually, in the same fashion as
as Stockoceros (O'Gara, 1990). Some bucks have horns that are deer antlers, they would have to separate from the pedicle above the
tipped forward rather than vertical: The form of the horns is also burr, with a new burr forming each year distal to the previous ones,
variable in the females, with both hornless females and females with and there is ample evidence that this is not the case (Goss, 1983). It
larger than average horns known (O'Gara, 1990). is also unlikely that the merycodont burr developed in association
It is usually assumed that fossil antilocaprines had a similar ker- with an annual shedding and regrowth of keratinous horn sheaths
atinous sheath over the bony horn: Webb (1973) noted that the (Pilgrim, 1941b; Matthew, 1924), for the reasons discussed earlier.
rings of nutritive foramina around the base of the horns in Stock- Merycodont burrs are also unlike cervid burrs in other respects:
oceros, Hexameryx, and Tetrameryx indicate the junction between Rather than representing outgrowths from the underlying basal
horny sheath and hairy skin, and the texture of the bone of an- antler bone, as in deer, they represent periodic deposition of bone on
tilocaprine horncores in general is like the slightly roughened and the horn surface, which can be easily broken off from the adjacent
pitted horncore of Antilocapra, and unlike the smooth surface found bone (Furlong, 1927), and it appears that each burr is associated with
in merycodonts. However, it would be impossible for antilocaprines a separate outerlayerof bone (Voorhies, 1969). Voorhies ( 1969) also
with forked horncores to possess a deciduous keratinous sheath. For conclusively demonstrated that the horns of adult merycodonts had
the same reasons as already discussed for bovid horns, a regenerat- no sheath or skin covering, by showing that the tips of the tines
ing horn growing from the base would not be able to spread over the sometimes show wear, at the place where a sheath would have been
divergent tines: Horn sheaths grown in this fashion cannot cover thickest.
branched tips unless each branch constitutes a separate horncore The consensus on merycodont cranial appendages is that they
with its own sheath (Goss, 1983). Note that the horn sheath itself represent branched horns, or "nondeciduous antlers" (Goss, 1983,
can be forked at the point of initial growth (as in Antilocapra), be- following Whitworth, 1958) or "pseudoantlers" (Bubenik, 1990). It
cause the initial keratinzation of the skin is at the presumptive tips, seems likely that the original horns grew to full dimension skin cov-
but branches or tines could not be added during extension of the ered: The skin was then shed to leave a permanent exposed dead bony
horn sheath shaft. horn, with the burr representing the terminal point of skin covering
If, as seems likely, a keratinous cover was present in antilo- in full-grown adults (Furlong, 1927; Voorhies, 1969; Goss, 1983).
caprines with forked or branched horncores (e.g., Ilingoceros, Plio- Fossil crania lacking burrs are those of young individuals that died
ceros, and Texoceros), it must have been laid down over an initial before the initial skin retraction (what would, in deer, be called "in
skin cover once the horn was fully grown, as appears to be the case velvet"). Multiple burrs represent successive years, with initial in-
in Antilocapra (where the dermal covering of the horncore before crease in pedicle diameter added by circumferential bone deposition
keratinization has been termed "keratovelvet"; Bubenik, 1990). But, in the area covered by skin, followed by a further retraction of the
unlikeAntilocapra, it would be impossible for this sheath to be shed, skin and the exposure of more bone (with the effect of lengthening
as it would be impossible to regrow such a structure from the base. the horn), with the formation of the new burr (Voorhies, 1969; Goss,
Some forked-horn antilocaprines with a very low intertine saddle 1983).
 Antilocapridae 505

It seems plausible that merycodont-type horns could be trans- to Management, eds. V. Geist & F. Walther, pp. 356-64. Morges,
formed into antilocaprine-type horns by the retention of the juvenile Switzerland: IUCN Publications new series, no. 24.
skin layer and its later keratinization, given the mode of horn sheath Brooke, V. (1876). (Family Bovidae). In The Geographical Distribution of
Animals, A. E. Russel, 2, 222-5. London: Macmillan.
development in Antilocapra, as discussed earlier. The addition of
Bubenik, A. B. (1990). Epigenetical, morphological, physiological, and
the keratinous sheath in antilocaprines was most likely a gradual behavioral aspects of evolution of horns, pronghorns, and antlers.
event, from stiffened outer hairs to a thin sheath, and finally to a In Horns, Pronghorns, and Antlers, eds. G. A. Bubenik & A. B.
thicker sheath whose tines extended beyond those of the horncore. Bubenik, pp. 3-113. New York: Springer-Verlag.
With a sheath and a permanent layer of skin over the horncore, the Colbert, E. H., & Chaffee, R. G. (1939). A study of Tetrameryx and associ-
ated fossils from Papago Spring Cave, Sonoita, Arizona. American
merycodont burrs would be lost. By both thickening and shorten-
Museum Novitates, 1034, 1-21.
ing the horn shaft as well as adding a sheath, antilocaprines would Cope, E. D. (1877). Report upon the extinct Vertebrata obtained in New
have greatly increased the strength of the horn. This may reflect the Mexico by parties of the expedition of 1874. Geological Surveys
adaptation to a locked horn type of male/male combat, involving West of the JOOth Meridian, Vol. 4, 1-370.
a great deal of stress on the shaft. Merycodonts most likely had a Dalquest, W. W. (1983). Mammals of the Coffee Ranch local fauna,
Hemphillian of Texas. Texas Memorial Museum, Pearce-Sellards Se-
somewhat different style of combat, where burrs were necessary to
ries, 38, 1-41.
stop the sharp tine tips from gouging the opponent's cranium. If Fitcher, E. (1974). On the bedding behavior of pronghorn fawns. In The Be-
the change in horn form in antilocaprines did indeed represent a haviour of Ungulates and Its Relation to Management, eds. V. Geist
change in fighting style, this in turn may be reflective of a change & F. Walther, pp. 352-64. Morges, Switzerland: IUCN Publications
in social behavior resulting from a change in habitat preference: As new series, no. 24.
Forsten, A. (1970). The late Miocene Trail Creek mammalian fauna. Con-
previously noted, the morphology of antilocaprines is suggestive of
tributions to Geology, University of Wyoming, 9, 39-51.
a preference for more open habitats than in merycodonts. Frick, C. (1937). Homed ruminants of North America. Bulletin of the Amer-
ican Museum of Natural History, 69, 1-669.
Furlong, E. L. ( 1927). The occurrence and phylogenetic status of Merycodus
from the Mohave Desert Tertiary. University of California Publica-
ACKNOWLEDGMENTS tions, Department of Geological Sciences, 17, 145-86.
(1932). Distribution and description of the Pliocene antelope Sphenopha-
We thank John Alroy for help with antilocaprid synonyms and lo- los from the northern Great Basin province. Publications of the
calities, and Dave Webb for comments on the manuscript. Carnegie Institute of Washington, 418, 27-36.
(1941). A new Pliocene antelope from Mexico with remarks on some
Earl Manning would like to acknowledge his debt to three mem-
known antilocaprids. Publication of the Carnegie Institute of Wash-
bers of the Frick organization (whose collection is now housed at ington, 530, 25-33.
the American Museum of Natural History in New York), on whose Gazin, C. L. (1935). A new antilocaprid from the Upper Pliocene of Idaho.
work his own antilocaprid research was built in the early 1970s. Journal of Paleontology, 9, 390-3.
Without Child~ Frick, the important F:AM collection of fossil an- Geist, V. (1966). The evolution of horn-like organs. Behaviour, 27,
175-214.
tilocaprids would never have been assembled. Frick's exhaustive
Gentry, A. W., & Hooker, J. (1988). The phylogeny of the Artiodactyla. In
1937 analysis of the group remains the major published work on The Phylogeny and Classification of the Tetrapods, Vol. II: Mam-
the family. Two members of the Frick staff also did important work mals, ed. M. J. Benton, pp. 235-72. Systematics Association Special
on material added to the Frick collection after 1937. Beryl Taylor Volume no. 35B. Oxford: Clarendon.
identified much of this material and organized the Frick collection, Goss, R. J. (1983). Deer Antlers: Regeneration, Function, and Evolution.
New York: Academic.
and EM remembers his kindness and helpfulness when he was curat-
Gray, J.E. (1886). Notes on the pronghorn buck (Antilocapra) and its position
ing the combined F:AM/AMNH antilocaprid collection. Floyd Blair in the system. Annals and Magazine of Natural History, 3, 323-6,
(for whom Plioceros fioblairi was named by Frick) also deserves 468-9.
mention. He registered and organized newly collected material into Gregory, J. T. (1942). Pliocene vertebrates from Big Springs Canyon, South
the Frick collection for many years, and it was through his efforts Dakota. University of California Publications, Department of Geo-
logical Sciences, 26, 307-466.
that most of the antilocaprid postcrania were curated, a very big job.
Hansen, R. M., & Clark, R. C. (1977). Foods of elk and other ungulates at
low elevations in northwestern Colorado. Journal of Wildlife Man-
agement, 41, 76-80.
Hesse, C. J. (1935). New evidence on the ancestry ofAntilocapra americana.
REFERENCES
Journal ofMammalogy, 16, 307-315.
Honey, J. G., & Alzett, G. (1988). Paleontology, taphonomy, and stratigra-
Ahearn, M. E. (1988). Systematics and evolution of the Antilocapridae. phy of the Browns Park Formation (Oligocene and Miocene) near
Journal of Vertebrate Paleontology, 8 (suppl. to no. 3), p. SA. Maybell, Moffat Co., Colorado. United States Geological Survey
(1994). Cranial morphology and evolutionary relationships among an- Professional Paper, 1358, 1-52.
tilocaprids and other ruminants. Journal of Morphology, 220, sup- Janis, C. M. (1988). An estimation of tooth volume and hypsodonty in-
plement to ICVM-4 abstracts. dices in ungulate mammals, and the correlation of these factors
Barbour, E. H., & Schultz, C. B. (1934). A new anti.locaprid and anew cervid with dietary preferences. In Teeth Revisited: Proceedings of the
from the late Tertiary ofNebraska.AmericanMuseumNovitates, 734, VIIth. International Symposium on Dental Morphology, Paris, 1986,
1-4. eds. D. E. Russell, J.-P. Santoro, and D. Sigogneau-Russell.
Bromley, P. T., & Kitchen, D. W. (1974). Courtship in the pronghorn (Antilo- Memoires de Museum d'Histoire naturelle, Paris (serie C), 53,
capra americana). In The Behaviour of Ungulates and Its Relation 367-87.
 Christine M. Janis and Earl Manning 506

(1993). Tertiary mammal evolution in the context of changing climates, ( 1913). A peculiar horn or antler from the Mohave Miocene of California.
vegetation, and tectonic events. Annual Review of Ecology and Sys- University of California Publications, Bulletin of the Department of
tematics, 24, 467-500. Geology, 7, 335-9.
(1995). Correlation between craniodental morphology and feeding be- Miller, W. E., & Downs, T. (1974). A Hemphillian local fauna containing
havior in ungulates: reciprocal illumination between living and fos- a new genus of antilocaprid from southern California. Los Angeles
sil taxa. In Functional Morphology in Vertebrate Paleontology, County Museum Contributions to Science, 258, 1-36.
ed. J. J. Thomason, pp. 76-98. Cambridge: Cambridge University Morgan, J. K., and Morgan, N. H. (1995). A new species of Capromeryx
Press. (Mammalia: Artiodactyla) from the Taunton Local Fauna of Wash-
Janis, C. M., & Scott, K. M. (1987). The interrelationships of higher ruminant ington, and the correlation with other Blancan faunas of Washington
families with special emphasis on the members of the Cervoidea. and Idaho. Journal of Vertebrate Paleontology, 15, 160--70.
American Museum Novitates, 2893, 1-85. Nowak, R. M. (1991). Walker's Mammals of the World, (5th ed.). Baltimore:
Jarman, P. J. (1974). The social organization of antelope in relation to their Johns Hopkins University Press.
ecology. Behaviour, 48, 213-67. O'Gara, B. W. (1990). The pronghorn (Antilocapra americana). In Horns,
Jermann, T. M., Opitz, J. G., Stackhouse, J., & Benner, S. A. (1995). Re- Pronghorns, and Antlers, eds. G. A. Bubenik & A. B. Bubenik,
constructing the evolutionary history of the artiodactyl ribonuclease pp. 231--64. New York: Springer-Verlag.
superfamily. Nature, 374, 57-9. O'Gara, B. W., & Matson, G. (1975). Growth and casting of horns by
Kitchen, D. W., & Bromley, P. T. (1974). Agonistic behavior of territorial pronghorns and exfoliation of horns by bovids. Journal of Mam-
pronghorn bucks. In The Behaviour of Ungulates and Its Relation malogy, 56, 829-46.
to Management, eds. V. Geist & F. Walther, pp. 365-81. Morges, Pilgrim, G. E. (194la). The dispersal of the Artiodactyla. Biological Reviews,
Switzerland: IUCN Publications new series, no. 24. 16, 155-8.
Kraus, F., & Miyamoto, M. M. (1991). Rapid cladogenesis among the pec- (194lb). The relationship of certain variant fossil types of "horn" to
oran ruminants: evidence from mitochondrial DNA sequences. Sys- those of the living Pecora. Annual Magazine of Natural History, 7,
tematic Zoology, 40, 117-30. 172-84.
Leidy, J. (1854). [On Bison latifrons, Arctodus pristinus, Hippodon specio- Richards, G.D., & McCrossin, M. L. (1991). A new species of Antilocapra
sus, andMerycodus necatus.] Proceedings of the Academy ofNatural from the Late Quaternary of California. Geobios - Paleontologie,
Sciences of Philadelphia, 7, 89-90. Stratigraphie, Paleoecologie, 24, 623-35.
(1858). Notice of remains of extinct Vertebrata, from the valley of the Scott, K. M., & Janis, C. M. (1987). The phylogenetic position of the Cervi-
Niobrara River, collected during the exploring expeditions of 1857, dae, and the case for a superfamily Cervoidea. In The Biology and
in Nebraska, under the command of Lieut. G. K. Warren, U.S. Top. Management of the Cervidae, ed. C. Wemmer, pp. 3-30. Washing-
Eng. by Dr. F. V. Hayden, Geologist to the expedition. Proceedings ton, D.C.: Smithsonian Institution Press, National Zoological Society
of the Academy of Natural Sciences of Philadelphia, 10, p. 23. Symposium Volume.
(1869). The extinct mammalian fauna of Dakota and Nebraska, including (1993). Relationships of the Ruminantia (Artiodactyla) and an analysis of
an account of some allied forms from other localities, together with the characters used in ruminant taxonomy. In Mammal Phylogeny:
a synopsis of the mammalian remains of North America. Journal of Placentals, eds. F. S. Szalay, M. J. Novacek, & M. C. McKenna,
the Academy ofPhiladelphia, 7, 1-472. (Naming of Cosoryxfurcatus pp. 282-302. New York: Springer-Verlag.
on p. 173.) Scott, W. B., and Osborn, H. F. (1890). Preliminary account of the
Leinders, J. J. M. (1979). On the osteology and function of the digits fossil mammals from the White River and Loup Fork Formations,
in some ruminants and their bearing on taxonomy. Zeitschrift fiir contained in the Museum of Comparative Zoology, Part II: Carnivora
Siiugertierkunde, 4, 305-18. and Artiodactyla. Bulletin of the Museum of Comparative Zoology,
Leinders, J. J. M., & Heintz, E. (1980). The configuration of the lacrimal 20, 1-87.
orifice in pecorans and tragulids (Artiodactyla; Mammalia) and its Simpson, G. G. (1945). The principles of classification and a classification
significance forthe distinction between Bovidae and Cervidae. Beau- of mammals. Bulletin of the American Museum of Natural History,
fortia, 30, 155-60. 85, 1-350.
Lull, R. S. (1921). Fauna of the Dallas Sand Pits. American Journal of Skinner, M. F. (1942). The fauna of Papago Springs Cave, Arizona, and a
Science, 5th series, 11, 159-76. study of Stockoceros, with three new antilocaprines from Nebraska
Matthew, W. D. (1902). List of the Pleistocene fauna from Hay Springs, and Arizona. Bulletin of the American Museum of Natural History,
Nebraska. Bulletin of the American Museum of Natural History, 16, 80, 143-220.
317-22. Skinner, M. F., & Taylor, B. E. ( 1967). A revision of the geology and pale-
(1904). A complete skeleton of Merycodus. Bulletin of the American ontology of the Bijou Hills, South Dakota. American Museum Novi-
Museum of Natural History, 20, 101-29. tates, 2300, 34-53.
(1909). Fauna! lists of the Tertiary Mammalia of the west. In Cenozoic Solounias, N. (1988). Evidence from horn morphology on the phylogenetic
mammal horizons of western North America, H. F. Osborn. Bulletin relationships of the pronghorn (Antilocapra americana). Journal of
of the United States Geological Survey, 361, 1-138. Mammalogy, 69, 140--3.
(1918). Contributions to the Snake Creek Fauna, with notes upon the Stirton, R. A. (1938). Notes on some late Tertiary and Pleistocene antilo-
Pleistocene of western Nebraska, American Museum Expedition caprids. Journal of Mammalogy, 19, 366--70.
of 1916. Bulletin of the American Museum of Natural History, Storer, J. E. (1975). Tertiary Mammals of Saskatchewan, Part III: The
38, 183-229. Miocene fauna. Life Sciences Contributions, Royal Ontario Museum,
(1924). Third contribution to the Snake Creek Fauna. Bulletin of the 103, 1-134.
American Museum of Natural History, 50, 59-210. Tedford, R.H., Skinner, M. S., Fields, R. S., Rensberger, J.M., Whistler, D.
Merriam, J. C. (1909). The occurrence of strepsicerine antelopes in the Ter- P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987).
tiary of Northwestern Nevada. University of California Publications, Fauna! succession and biochronology of the Arikareean through
Bulletin of the Department of Geology, 5, 319-30. Hemphillian (late Oligocene through earliest Pliocene epochs) in
(1911). Tertiary mammal beds of Virgin Valley and Thousand Creek in North America. In Cenozoic Mammals of North America, ed. M. 0.
Northwestern Nevada. University of California Publications, Bul- Woodburne, pp. 153-210. Berkeley: University of California Press.
letin of the Department of Geology, 6, 199-304. Vaughan, T. A. (1985). Mammalogy (3rd ed.). Philadelphia: Saunders.
 Antilocapridae 507

Voorhies, M. R. (1969). Taphonomy and population dynamics of an early (1973). Pliocene pronghorns of Florida. Journal of Mammalogy, 54,
Pliocene vertebrate fauna, Knox County, Nebraska. Contributions to 203-21.
Geology, the University of Wyoming, Special Paper no. 1, 1-69. (1991). Ecogeography and the Great American Interchange. Paleobiol-
(1990). Vertebrate biostratigraphy of the Ogallala group in Nebraska. ogy, 17, 266-80.
In Geologic Framework and Regional Hydrology: Upper Cenozoic White, T. E. (1941). Additions to the fauna of the Florida Pliocene. Proceed-
Blackwater Draw and Ogallala Formations, Great Plains, ed. T. C. ings of the New England Zoological Club, 18, 67-70.
Gustavson, pp. 115-51. Austin: Bureau of Economic Geology, Uni- (1942). Additions to the fauna of the Florida Pliocene. Proceedings of
versity of Texas. the New England Zoological Club, 21, 87-91.
Webb, S. D. (1969). The Burge and Minnichaduza Clarendonian mammalian Whitworth, T. (1958). Miocene ruminants of East Africa. Fossil Mammals
faunas of north-central Nebraska. University of California Publica- of Africa, no. 15. London: British Museum of Natural History Pub-
tions, Bulletin of the Department of Geological Sciences, 78, 1-191. lications.
34 Cervidae and Bovidae

S. DAVID WEBB

INTRODUCTION A B

The last two families of ruminants to enter North America from

Asia during the Tertiary were ultimately the most successful in the
sense that they still survive. After the Quaternary, as many other
large mammal groups declined or went wholly extinct (equids and
camelids to name but two), the Bovidae and Cervidae continued
to thrive. The first North American bovid genus, Neotragocerus,
arrived at about the end of the Miocene; the first cervid, as yet un-
described, appeared in the late Hemphillian (Tedford et al., 1987).
Two other cervid genera, Bretzia and Odocoileus, appeared in the
Blancan just before the end of the Tertiary. These four ruminant gen-
era are quite rare in Tertiary fossil collections. Weighed against the
vast numbers and diversity of other late Tertiary land mammals, the
true cervids and bovids seem little more than exotic curiosities. That
balance changed dramatically during the Quaternary as bovids and
cervids appeared in greater abundance and diversity, culminating in 10 cm 2cm

the present situation in which Odocoileus is the most abundant and

Figure 34.1. A. Antler of Bretzia (after Gustafson, 1985). B. Hom of Neo-
familiar of all ungulates in North America.
tragocerus (after Matthew and Cook, 1909).

SYSTEMATICS question. Parentheses are usually used for individual species, thus
implying the genus is firmly known from the locality, but the actual
The interrelationships of cervids and bovids to other ruminant ar- species identification may be questionable. Question marks in front
tiodactyls are discussed in Chapter 22, the summary chapter on of the locality (e.g., ?CPIOI) mean the taxon is questionably known
artiodactyls in this volume. from that locality, thus implying some doubt that the taxon is actually
present at that locality, either at the genus or the species level.

INCLUDED NORTH AMERICAN GENERA OF

CERVIDAE
CERVIDS AND BOVIDS
I omit discussion of Cervidae characteristics because the diagnostic
The locality numbers listed for each genus refer to the list of unified features of this existing family are well known. The two described
localities in Appendix I. The acronyms for museum collections are Tertiary genera are members of the subfamily Odocoileinae, includ-
listed in Appendix III. ing early representatives of the extant genus Odocoileus itself.
The locality numbers may be listed in a couple of alternative ways. Two distinctive cervid genera, Bretzia and Odocoileus, appeared
Parentheses around the locality (e.g., [CP101]) mean the taxon in in the Blancan of North America. Gustafson (I985) showed that
question at that locality is cited as an "aff." or "cf." the taxon in Bretzia, originally considered nonodocoileine (possibly cervine)

508
 Cervidae and bovidae 509

due to its palmate antlers, is closely related to Odocoileus. Pro- Included species: N. improvisus (known from localities
coileus (Frick, 1937), once supposed to be the late Hemphillian CPI 15C, CPI !SD); N. lingreni (locality PN23A).
predecessor of Odocoileus, was shown by Webb (1983) to be a syn- Neotragocerus sp. is also known from localities ?CP78,
onym of the dromomerycid genus Pediomeryx. NP5B.

Comments: These additional localities (especially Courtney

Bretzia Fry and Gustafson, 1974
Pit in southeastern Alberta recorded by Bums and Young,
Type species: Bretzia pseudalces Fry and Gustafson, 197 4.
1989 [locality NPSB]) are of great interest for two rea-
Type specimen: University of Washington 42971, left antler
sons. First, they confirm that the age, inferred for the geno-
(see Figure 34.1).
typic occurrence in the upper Snake Creek Fauna (locality
Characteristics: Antlers with short beam, simple brow tine,
CPI ISD), is late Hemphillian. Second, they help outline its
large, rectangular denticulate palmation; nearly horizontal
distribution, which evidently was mainly the High Plains
disposition of antler on cranium.
and west into Idaho.
Included species: B. pseudalces only, known from locality
A supposed additional record from the Calvert Forma-
PN3C only.
tion in Virginia (locality NC3B), cited by Tedford and
Hunter (I984) as "dromomerycid," is probably better as-
Odocoileus Rafinesque, 1832 signed to the Protoceratidae than to Neotragocerus. Based
Type species: Odocoileus virginianus Rafinesque, 1832. on an isolated molar, the Calvert record exhibits the subtle
Type specimen: Not designated (the extant white-tailed deer). curvature and considerable crown height of a progressive
Characteristics: Medium-sized deer with large dichotomously protoceratid molar.
branching antlers; large anterior tine branches from steeply In a later contribution on the Snake Creek Fauna,
rising beam. Long, relatively narrow cranium and slender Matthew (I9I8, p. 222) appears to doubt his own earlier
limbs. reference of Neotragocerus to the Bovidae, for that text
Included Tertiary species: 0. brachydontus only (known from has a question mark and the comment that these Bovidae
localities [SP5], CPl28A, CP128C). might equally well be Giraffidae. Further reading suggests
Odocoileus sp. is also known from localities GCl3, that this remark was intended only for Cranioceras and
GCl4, GCl5, (GCl6), CC43, (NB13C), SBI4A, SBI4C, that Matthew's brief discussion of Neotragocerus under
?SBI4F, SBI5, SB37, SB49, ?SB50, SB58C, SB63, the heading "?Bovidae" was inserted later. (See further
?SB65, SPlH, CP93, CPI I 7A, CPI28C, PN4. comments on Neotragocerus in the summary chapter on
artiodactyls, this volume, Chapter 22.)
Indeterminat~ cervids
Fragmentary remains assigned to cervids are reported from locali-
ties WM26, CA6B, GC28, CC47, NB36, CP97, CP116F, PN23A,
BIOLOGY AND EVOLUTIONARY PATTERNS
PN23C.

The appearance of bovids and cervids in North America during the

BOVIDAE late Hemphillian had limited impact due to their low diversity and
evident rarity. Yet these first records adumbrate the potential sig-
I defer discussion of the family-level characteristics, simply because nificance of later dispersals representing these two most successful
the general features of Bovidae are well known and have little bear- ruminant families.
ing on the particular problems associated with the following late In the Pliocene one can already detect an important difference in
Tertiary immigrant taxon. Neotragocerus is the only known genus the manner in which the Cervidae and Bovidae would conquer North
of the family securely recognized in the Tertiary of North America. America during subsequent Quaternary episodes. For Bovidae the
cooler more severe climates of the Quaternary favored repeated
Neotragocerus Matthew and Cook, 1909 immigrations by a variety of high-latitude taxa from Asia, including
Type species: Neotragocerus improvisus Matthew and Cook, bison, musk oxen, sheep, goats, and saiga antelopes. Neotragocerus
I909. was simply the first of this long and varied string of allochthonous
Type specimen: AMNH I414I, horn core (see Figure 34.1). bovid genera. These immigrant bovids did not extend far into equa-
Characteristics: Large straight homcore, conical, penetrated torial latitudes and none entered South America.
by frontal sinus. Cheek teeth, if correctly referred to this Most New World Cervidae, in contrast, descended from the im-
taxon, indicate a relatively brachydont group of antelopes migration of a single Odocoileine genus and extended widely in the
such as Hippotraginae, not Caprinae or Bovinae. M3 with neotropical region. Bretzia and Odocoileus are the earliest known
distinct metastyle; m3 with singleclobed heel. members of this New World radiation of odocoileines. Somewhat
Average length of m2: estimated at 24.0 mm, based on oc- later members are the extinct genera Navahoceras and Sangamona.
clusal equivalent in upper second and third molars (AMNH At least eight other living and extinct genera attributable to the tribe
14I36). Odocoileini (Groves and Grubb, 1987) appear autochthonously in
 S. David Webb 510

North and South America during the Quaternary. Only Cervus (the Cervidae) and the evolution of New World deer. Transactions of the
wapiti) and two Alcinae, the extinct Cervalces and the living Al- Nebraska Academy of Sciences, 13, 83-92.
Matthew, w. D. (1918). Contributions to the Snake Creek Fauna with notes
ces (the moose), are deer that, like bovids, entered temperate North
upon the Pleistocene of western Nebraska. Bulletin of the American
American latitudes during a cooler, more severe late Quaternary Museum of Natural History, 38, 183-229.
regime. Matthew, W. D., & Cook, H. J. (1909). A Pliocene fauna from Western
Nebraska. Bulletin of the American Museum of Natural History, 26,
361--414.
Rafinesque, C. S. (1832). Animals of Kentucky. Atlantic Journal, 1,
REFERENCES 109-10.
Tedford, R. H., & Hunter, M. E. (1984). Miocene marine-non-marine
Bums, J. A., & Young, R. R. (1989). Stratigraphy and palaeontology of correlations, Atlantic and Gulf Coastal Plains, North America.
the Hand Hills Region. Occasional Papers of the Tyrell Museum of Palaeogeography, Palaeoclimatology, Palaeoecology, 47, 129-51.
Paleontology, 9, 1-13. Tedford, R.H., Skinner, M. F., Fields, R. S., Rensberger, J.M., Whistler, D.
Frick, C. (1937). Homed ruminants of North America. Bulletin of the Amer- P., Galusha, R., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987).
ican Museum of Natural History, 69, 1-669. Fauna! succession and biochronology of the Arikareean through
Fry, W. E., & Gustafson, E. P. (1974). Cervids from the Pliocene and Hemphillian (late Oligocene through earliest Pliocene epochs) in
Pleistocene of Central Washington. Journal of Paleontology, 48, North America. In Cenozoic Mammals of North America, ed. M. 0.
375-86. Woodbume, pp. 153-210. Berkeley: University of California Press.
Groves, C. P., & Grubb, P. (1987). Relationships of living deer. In Biol- Webb, S. D. (1983). A new species of Pediomeryx from the Late Miocene
ogy and Management of the Cervidae, ed. C. Wemmer, pp. 21-59. of Florida, and its relationships within the subfamily Craniocerati-
Washington, D.C.: Smithsonian Institute Press. nae (Ruminantia: Dromomerycidae). Journal of Mammalogy, 64,
Gustafson, E. P. (1985). Antlers of Bretzia and Odocoileus (Mammalia, 261-76.
 Part V: Perissodactyla and Proboscidea

35 Perissodactyla and Proboscidea

CHRISTINE M. JANIS, MATTHEW W. COLBERT, MARGERY C. COOMBS, W. DAVID LAMBERT,

BRUCE J. MACFADDEN, BRYN J. MADER, DONALD R. PROTHERO, ROBERT M. SCHOCH,
JEHESKEL SHOSHANI, and WILLIAM P. WALL

INTRODUCTION be so simple (see summary chapter on Artiodactyla, this volume,

Chapter 22, for discussion and review).
Perrisodactyls and proboscideans are grouped together in this chap- Perrisodactyls and proboscideans also differ from artiodactyls in
ter, following the grouping by Prothero, Manning, and Fischer (1988) their specializations for locomotion. Proboscideans are graviportal,
of the "higher ungulates" or "paenungulates" at the level of the and living ceratomorphs (tapirs and rhinos) are mediportal at smaller
Perissodactyla and above. Perissodactyls and proboscideans are the sizes and graviportal at larger sizes. Equids are the only living curso-
two orders of higher ungulates with representative terrestrial mem- rial perissodactyls, although some early Tertiary ceratomorphs (e.g.,
bers in the Tertiary of North America. The relationship of these hyracodonts and certain tapiroids) were also moderately cursorial.
orders to other higher ungulates, including the debate of the rela- In contrast, cursorial adaptations of the postcranial skeleton may be
tionship ofhyracoids to the perissodactyls, is discussed in the chap- the primitive mode for artiodactyls (Rose, 1985). Only hippos could
ter on archaic ungulates and ungulatelike mammals (this volume, be described as graviportal among both living and fossil forms.
Chapter 15). Present-day representatives of these groups consist only of the
Present-day 'perissodactyls and proboscideans are all of large family Elephantidae in the order Proboscidea and the families
(greater than 200 kg) to very large body size (greater than 1,000 kg), Equidae, Tapiridae, and Rhinocerotidae in the order Perissodactyla.
and all are folivorous herbivores with a hindgut system of cellulose Of these, only tapirs have a Recent representation in North Amer-
fermentation. In both respects this contrasts with the artiodactyls, ica, although all living perissodactyl families were known during
which represent the other extensive present-day ungulate radiation. the Tertiary, as were nonelephantid proboscideans (with elephan-
Many artiodactyls (e.g., most suids) are omnivorous rather than fo- tids present in the Pleistocene). Today, feral equids, reintroduced
livorous, and folivorous artiodactyls have a primarily fore gut system from Europe in the sixteenth century, thrive well-enough in the
of fermentation (although they may also retain some hindgut fer- North American west to be considered pests (Berger, 1986). Ex-
mentation, which is the plesiomorphic condition for mammals). tinct perissodactyls found in North America in the Tertiary include
Artiodactyls also include smaller forms (such as the Asian mouse brontotheres, chalicotheres, and a variety of "tapiroids" (including
deer, Tragulus, with a body mass of as little as 2 kg), but their main isectolophids and various ceratomorphs; see Colbert and Schoch,
diversity is in the 10-300 kg range. Only hippos, giraffes, and bison this volume, Chapter 39).
may exceed 1,000 kg in body mass today. Nor was there a much The Tertiary history of perissodactyls and proboscideans parallels
greater diversity of megaherbivores in artiodactyl history, although that of the artiodactyls (see Chapter 22, this volume): a few North
extinct forms of similar mass to large bison include late Cenozoic American endemics with their own unique radiation, other North
North American camelids and Old World sivatherine giraffids. In American appearances of taxa with a primarily Old World distri-
contrast, perissodactyls and proboscideans include the largest land bution, and the radiation of a number of unique Old World forms
mammals ever known, with Oligocene indricothere rhinocerotoids (especially during the early Tertiary).
and Pleistocene mammoths attaining body masses of up to 20 tons Equids may well be of Old World origin because European
(Fortelius and Kappelman, 1993). equoids predate the earliest known North American ones (see
The difference in the siting of the fermentation chamber, foregut Hooker, 1994). Nevertheless, equids were a primarily North Amer-
versus hindgut, has often been used to explain the relative evolution- ican radiation and did not comprise a significant component of the
ary success (at least as perceived from the Recent) of artiodactyls post-Eocene Old World faunas until the emigration ofhipparionine
versus other ungulates (Janis, 1976), but the true picture may not horses in the late Miocene (see MacFadden, 1992, for summary and

511
 Christine M. Janis et al. 512

TITANOTHERIOMORPHA

~ HIPPOMORPHA MOROPOMORPHA

EQUOIDEA ANCYLOPODA CERATOMORPHA

Figure 35.1. Interrelationships among perissodactyls. * indicates taxon not found in North America (but note that Paleomoropus has been considered to
be a lophiodontid [Schoch, 1989b] and that some of the earliest Eocene species of"Hyracotherium" may actually be palaeotheres [Hooker, 1994]).

review). Brontotheres also had their greatest diversification in North Ungulata as a whole are discussed in the chapter on archaic un-
America: Unlike equids, they may have originated in this continent gulates and ungulatelike mammals (this volume, Chapter 15), and
in the early Eocene, with later (middle Eocene) dispersal to Eurasia relationships within the order Proboscidea are discussed in Lambert
(see Mader, this volume, Chapter 36; Prothero and Schoch, 1989). and Shoshani (this volume, Chapter 43). Relationships within the
Tapirids apparently had their first appearance in North Amer- Equoidea and the Hippomorpha are also ignored because equids
ica (Protapirus and an undescribed tapiroid in the late Eocene) are the only North American taxon in these groupings (see Hooker,
and, subsequent to the Oligocene occurrence of the European Pro- 1989, 1994 for a review; also note brief discussion in MacFadden,
tapirus, represent an exclusively North American radiation until the this volume, Chapter 37).
late Miocene (Schoch, 1989b). Other moropomorphs, such as isec- This chapter focuses on the interrelationships among the major
tolophids, and basal ceratomorphs Selenaletes andDesmatotherium, groups of perissodactyls: brontotheres, chalicotheres, equids, and
appear to be uniquely North American (Schoch, l 989b). In contrast, ceratomorphs (rhinos and tapirs). The interrelationships and affini-
the majority of North American moropomorphs, such as chalicothe- ties of the various North American moropomorph taxa are exten-
rioids and rhinocerotoids, are of Eurasian origin with multiple im- sively discussed in Colbert and Schoch (this volume, Chapter 39)
migrations into North America (Prothero and Schoch, 1989). and not considered further here. A review of the history of ideas of
Finally, there was a diversity of Eurasian equoids (equids and perissodactyl interrelationships was presented by Schoch (1989a),
palaeotheres), equoid-related pachynolophids, and basal cerato- and this section is largely a summary of that review.
morph lineages (e.g., lophialetids, deperetellids, and rhodopagids) The present-day concept of the Perissodactyla as comprising un-
that were entirely of early Tertiary Old World occurrence (see gulates with a mesaxonic foot posture has its origin in Blainville
Hooker, 1989, 1994; Schoch, 1989b; Prothero and Schoch, 1989). (1816). Owen (1848) named the association of horses, tapirs, and
They are not considered further in this chapter. rhinos the Perissodactyla, although note that hyracoids and various
extinct other ungulates were originally included as perissodactyls.
Previous to this, the living perissodactyls had been grouped with
SYSTEMATICS various other mammalian orders: Most commonly, tapirs and rhi-
nos were grouped with other large-bodied, short-legged ungulates
We consider only the interrelationships within the order Perisso- as "pachyderms"; horses were accorded their own status, probably
dactyla here. Figure 35.l summarizes our current hypothesis of because of their economic and social importance (see references in
the interrelationships of perissodactyls. Relationships within the Schoch, 1989a).
 Perissodactyla and Proboscidea 513

iii' iii' ii:' ii:' ii:'

~
6
~ 6 ~ ec e ~z ~
!:.
6
~
_(.) c
"' "' c ·="' u'ii c
.........
~ Ui('Cl ('Cl~
c'iii E ·;;; c.: Ui('Cl
... ... ... - .ct:: 'Iii :!:u.ct:::=
fti'iii ... ('Cl

c "'
... ('Cl

-g ~e:
... ('Cl GI ftl
...... GI -
-I!!·;:::ri
-" l! Ill GI - 0
0
(.)
('Cl .c Ill .c D. -c D.
...
D. (.)
GI 0
Q) .., GI c.>u t:: 'Iii :I ('Cl :I ('Cl GI ftl
ui
MA :; I!! (.)I!! 0
z ... GI ('Cl

~~~ (.)Iii ci:E

0 GI 0 GI
(j o!I z ... 0
en ... D. 0
z

x~
en" z
1.8
c:
, " " " "
l1l I
(.) N~ Co, Na, i;g / J;g, Na, Na, Na,
Q) c: £g I
£g / !:kl £g &g, £g

x x IX
-<a-
l1l
co I
w 2.4
...J /

x
z c: I
w ;.,
l1l
(.)
!;g,
?.!::!Q, Na, Ne, Na, Na,
0 I i;g .Eg
0 ~~ i i;g lQj Qi, i;g Eg_ Eg_
WCO

x
:::::i 4.5
Q. ' c

iX
1i) a. Co, Na,
Na, Ne, Na, Ne, Na, Ne, /
2~~ I Ne, Ps,
j J: '.2 fil!.~ ..Qi fil!,.Qi fil! ..Qi /
Ps,.Qi ,, Figure 35.2. Biogeographic ranges

x
5.2
I
of equids. A "box" (for a particular

x
I
c. c: I Na, Ne, Qi, Na, Ne, Co, Na, Na, time period in a particular biogeo-
/ Na, Ne, Na, Ne, Ne,
2~~ I fill. fill, Di. I
Ne,
graphic region) with a cross through
j J: '.2 .Qi,£! /.El On As, Qi, .El As, .Qi
I .Qi
.El it means no fossil localities are known
I/
6.0

x
~. for that time period from that area;

7.0
... c. c:
tfl ~~
_j J: :c x
Co, Hp, Na,
Ne, Ps,
~
Na, Ne,
Di ?Pl
Ne,
&El
Ne,
As Di
Co, Hp,
Ne, fill,.
~
Co, Hp, Na, Ne,
~
Pl Pr
Ne,

.El J( a single dashed line through the box

means only scant fossil information
is available (usually only a single,

x
/ /
~. ?Dia/ ' Co,/ ?Co, I
l1l c. c: Co, Na, Ne, Ps, E? / Co, Hp, Na, Ne, N,. small, locality). Key: "Hyracotheri-
w E.!1! .Pl Hp, Hp, Ne,
/ / .Pl inae" are in italics: Ep = Epihip-
w :2~
_Qg !&. .Pr
/ I/
/Qi
/
Qi!, .El Ca, ?Qj, .P1
I Qj pus, Ha = Haplohippus, Hy= Hyra-
8.8

x
/
C: c: Mg, Hh, Ne or Co, Hp, Hh, Co, I E? / cotherium, Or = Orohippus, Xe =
Cl>~-~
?Co Co, Hp, Na, Co, Hp,
Ne, Ps, Co, Hp, Co, /D. Ne, Ps, Hp, Mx, Na, Ne, Xenicohippus. "Anchitheriinae" are in
w
-<a- c:
l1l 0 I
Ne, / _l, I /
...J () "O Qi!, .Pr Ne, .Pl
?.Pr • .El or .El Qi!, .El Ps, Qi!, .P), .Pr boldface: Ar= Archaeohippus, De=
ffi 9.5
/ I v Desmatippus, Hh = Hypohippus, Ka
Hh,Mg, Hh, Co,

x
0 C: c: Hh, Co, Mg, Hh,Mg, Hh, Mg, Co, Hp, Hp,
E? I I = Kalobatippus, Me= Mesohippus,
0 ~~.~ Hp, Ne, Ps, Co, Mh, Na, Ne, Mh, Na, Ne, Ps,
... l1l c: Co, Hp, Co, Hp, I I
~ l1l- 0 Ne, ?Ps, Ps, Qi!. I Mg= Megahippus, Mi= Miohippus,
w () "O Ca, .Qi, ..Pl •.Pr .El ?Qi, .El Qi!, Qi, .El •.Pr
11.0 /
QJ,.E! Qi, .Pl •.Pr 1Mx /
/
v Pa= Parahippus. Equinae are in ro-
6 , Hh,Mg, Mg ,

x
Ar, Hh, Ka, Mg,

x
Q)
Hp,/ '/ Hh, Hp,/ ?Mp./ Co/ man type. Merychippus: Mh = hip-
§~ c: Co, Hp, Mh,
c~ Mg, Mh,
?Pa, Co, ?Hp,
I parionine Merychippus species, Mp=
Mp, Na, Ps,
_j &l ·~ /Mg
.El. Er
Mh, Ne, Ps,
V / Mx
/
primitive Merychippus species (sensu
12.5
~ • .E! • .Pr I/ I/ .El Ca, Pl, .Pr
,/
/ Mx v stricto), Mq = equinine Merychip-

x
Hh,Mg, Ar, Hh,

x
Q) '
Hh, Ka, Mh, Ps, De, Ar,/
(ii .8 ?Co, Mp, Mh, Co,Mh,
/
Mg, Hp, pus specie"S:-Hipparionini are in plain
_J ~ c: Ar, Ka, Mg, / Hh,
I
w&l ·~ Qi!, Er Mh,Mg Mg,f) Ne, ?.Qi,
El. Pr ,, /
£!
Mh,
Mg, Pl
Mh ?Mx
roman type: Co = Cormohipparion,
14.0 Hp= Hipparion, Na = Nannippus,
Mx Ar, De, Hh,
6 I Ne= Neohipparion, Ps = Pseudhip-

x
De, Hh, Ka, Ar, De, Ar, De, Hh, Ar, De, Ar, De,
Hh, Pa, Ka, Mg,
~~ c: I Hp, Mh, Mp, Hh, Ka, Ka, Pa, Mg, Hh, Mh, ?Hh,Pa parion. Equini are in roman type un-
l1l l1l l1l
w co·:;:
Mh, ?Mp, Mh, ?Mg I
Qi!, .Pr Mh,Mg Mg, .Pr Mh, Ca,Mg Mp, Mg Mp,Mh derlined: As = Astrohippus, Ca =
15.8 Calippus, Di = Dinohippus, Eq =
bi c:

x
Ar, Ka, / Ar, Ar, Ar, De, Hh, De,Hh, I

17.5
-
j
ai
.£ .!'!!
J:
E
Q)
"O
.g
...
x /
/
/
Mp
Hh, Pa,
Mh, Mg
Ka, Pa,
Mh, Mg
Mh,
.Pr
Ka, Pa,
Mp,Mh
Pa,
Mh, Mg
/
/
/

I
I
Equus, Hd = Hippidion, On =Ono-
hippidium, TI= Pliohippus, £!:=Pro-
tohippus. E? = equid indet.

Wood ( 1934) was the modem originator of the notion of a division lopoda. However, Schoch (1989a) notes that Radinsky's perceived
of living perissodactyls into the suborders Hippomorpha (equids) distinctness of chalicotheres was based in part on his assignation
and Ceratomorpha (rhinos and tapirs) (see also Wood, 1937). In this of the genus Paleomoropus to the Chalicotherioidea, an assignation
scheme, also followed by Simpson (1945), brontotheres and chali- that has since been questioned (see also discussion in Colbert and
cotheres were classified with equoids in the Hippomorpha. Borissiak Schoch, this volume, Chapter 39).
(1945) maintained that brontotheres and chalicotheres were sister In the early 1980s, both Schoch (1983, 1984, 1985) and Hooker
taxa among the hippomorphs. In contrast, Scott (1941) perceived a (1984) independently came to the conclusion that chalicotheres
fundamental difference between clawed and hoofed perissodactyls, shared derived characters with certain ceratomorphs. Schoch ( 1985)
creating the suborders Ancylopoda (chalicotheres) and Chelopoda erected the suborder Moropomorpha (including Ancylopoda and
(ceratomorphs and the remaining hippomorphs). Ceratomorpha as infraorders), leaving brontotheres and equoids as
Radinsky (1964) modified Scott's (1941) scheme to raise Hip- superfamilies within the suborder Hippomorpha. Hooker (1984)
pomorpha and Ceratomorpha to equal subordinal rank with Ancy- presented a rather more complex scheme, in which chalicotheres
 Christine M. Janis et al. 514

MA

?Me,
Ka,Mi
Mi

Me,
Mi
Mi

Me,
Me
Mi

Ep Me

Ep Ep

Figure 35.2. (Cont.)

were recognized as the sister taxon to the lophiodontid "tapiroids" "tapiroids," and ceratomorphs) are granted equal status as
and brontotheres were considered as the sister taxon to all other infraorders. Within the Moropomorpha, the parvorders Ancylopoda
perissodactyls. (chalicotherioids and lophiodontids) and Ceratomorpha (rhinocero-
Figure 35.1 is derived largely from Prothero and Schoch (1989) toids, tap irids, H eptodon, and several Asian lineages) are recognized
and presents an amalgam of the ideas of Schoch (1985, 1989a) as sister taxa. Basal moropomorph taxa include the North American
and Hooker (1984, 1989). Prothero and Schoch (1989) consider family Isectolophidae and the Eurasian genus Kalakotia (placed as
hyracoids to be the sister taxon to other perissodactyls, dividing more derived than the isectolophids; not included in Figure 35.1).
the Perissodactyla into the suborders Hyracoidea and Mesaxonia. Perhaps if hyracoids are excluded from the Perissodactyla (as sug-
Within the Mesaxonia (equal to the Perissodactyla of common us- gested by, e.g., Novacek, Wyss, and McKenna, 1988; Shoshani,
age), Titanotheriomorpha (brontotheres), Hippomorpha (equoids 1993), these infraordinal ranks would be raised from infraorder to
and pachynolophoids), and Moropomorpha (chalicotherioids, suborder, and from parvorder to infraorder, respectively.
 Perissodactyla and Proboscidea 515

Figure 35.2. (Cont.)

EVOLUTIONARY AND todon and Hyrachyus. In the latest Wasatchian Xenicohippus was
BIOGEOGRAPHICAL PATTERNS replaced by Orohippus, and there was the appearance of two new
taxa: the brontothere Eotitanops and the indeterminate perissodactyl
Lambdotherium (these taxa were confined to the late Wasatchian,
EOCENE
or nearly so) (see Figures 35.2, 35.4).
Perissodactyls first appear in North America at the start of the A more distinct fauna! shift occurred at the start of the mid-
Eocene, although they may be known from slightly earlier sedi- dle Eocene (Bridgerian), with the appearance of the larger, more
ments in Europe (Hooker, 1994). The earliest Wasatchian contains derived brontothere Palaeosyops, the replacement of more primi-
the equid Hyracotherium and the isectolophid Cardiolophus: These tive isectolophids by lsectolophus, and the appearance of the first
taxa were joined in the later Wasatchian by the equid Xenicohippus, true tapiroid Helaletes (see Figure 35.4). Hyracotherium was rare
the isectolophid Homogalax, and basal ceratomorphs such as Hep- in the early Bridgerian and did not survive this stage. By the late
 Christine M. Janis et al. 516

iii' iii' ii:" ii:" ii:"

6 ~ !!!. !!!. ~ ~ z 6~
Ci' Q. 6
_(.) c c Ill Ill
c
Ill e:. u;
u; ...... c .5

-g ~e 111 ...5i ...E

~
l1l
..,~

... Ill
c "iii
._ OI
c ·-
m=
c.5
._ OI
111- f... :19 ... Ill
Ill-
.g J OI

- ...... 0
"lU.~ 0 c"' .2!t:: Ill .c Ill .c D. D. .c D. (.)

MA
... Ill
(.)

:;
Ill 0
(.)(.) 0
...
OI :I"'
0 Ill
'50 m
Q)
c ...
~!
t:: ...
0
z 12
OI
() .c
l1l t:: ui
zf en ... en ... D. 0
z
~:1!~ (.) c( CJ (j oil CJ CJ CJ CJ CJ z
w 1.8
z

IX
c
w l1l
(.) I / '
0 Q) c I
/
jffi

xx x
0

x
I I
::J I/

x
D.. 2.4 c I
l1l
;., (.)
?Te
I
~~

x
WC!l I

x
4.5
t) 6.. c: /
.$ ~:lll
I Ap, Te ?Ap, Te Ap, Te
/
j J: '.E I 1,
5.2 ,

~ x
I
a. c Te
Te / /
.$~Jg I / r?Ap, Te Ap, Te Ap, Te / Te

x
j J: :E I / /

x
6.0 Figure 35.3. Biogeographic ranges
~. I
.. c. c ?Ap, of rhinocerotoids. A "box" (for a
Te Ap, Te Ap,Te Ap, Te Ap, Te
il'.l ~~ I
Ap, Te
Te
particular time period in a particu-
..J J: :c: I
lar biogeographic region) with a cross

x
7.0 ;.,
~ 6. c / ?Ap/ / Te / through it means no fossil locali-
w E.!!! Te Ap, Te Ap, Te Ap, Te Te
w / / / / ties are known for that time period
z u.i ~ :E I/
8.8
/ I/ / from that area; a single dashed line

x
w
0 cc ?Te/ Ap,?Pe, R? / / I through the box means only scant
0 Q) ~ .!!! Ap, Te ?Ap, Te Pe, Te Ap, Te
-<a- c
ct! 0 I / / I fossil information is available (usu-
Te
:E ,

xx
...J () "C
I/ / II ally only a single, small, locality).
9.5 Key: Amynodontidae are in boldface:
cc Ap,?Pe, Ap, Pe, Ap, Te Ap, Pe, / / I
~~.~
.. l1l c Ap, ?Te I Ay = Amynodon, At = Amynodon-
Te Te / /
w ()l1l- 0
"C Te topsis, Mt = Metamynodon, Mg =
/ I/ I
11.0 Megalamynodon, "Pr"= "Procadur-
I

12.5
.2? 0
ca u; c
-1 (ij ca
..J co ·s:
Ap, Pe,
Te
/
/
/

/
/
/

I/
/
/

x Ap, Pe, Te
/
?Ap/
/
Ap /

II
/
II
I
codon." Hyracodontidae are in ital-
ics: Ep = Epitriplopus, Fa = Forster-
cooperia, Hy = Hyracodon, Td =

~
Ap, Ap1
m.9~
x Triplopides, Tr = Trip/opus, H? =
Q) '
?Ap, / Pe,
c Ap, Pe, ?Ap Pe Ap,Pe /
...J
Te
Pe, I hyracodontid indet. Rhinocerotidae
u.i 8l ·~ Te
/
/ Te Te I/ are in roman type: Am = Amphicaeno-
14.0
6 I pus, An= new acerathere genus, Ap =

x
;.,- I Ap, Pe, Ap, Di,
~ c
-.:::: Ap, Pe, R? Ap,Pe Ap, Pe, Te Aphelops, Br = Brachypotherium, Di
l1l l1l l1l I Te Te Te I
W CD"> Te = Diceratherium, Fl = Floridaceras,
/ It
15.8 I Me = Menoceras, Pe= Peraceras, Pt
bi

IX
c

~
c ct! Ap / /
An, Ap, = Penetrigonias, Sh= Subhyracodon,
.$ -~ 'E / Ap, ?Di Ap,Pe Di / I
Br, Pe Te= Teleoceras, Tg = Trigonias, Tt=
.'.:J:i::.l2 I/ II II
17.5 Teleteceras. R? = rhinocerotid indet.

Bridgerian, brontothere diversity increased, including the appear- episode of perissodactyl diversification, the abundance of equids
ance of the first homed brontotheriine Telmatherium. Basal was low. Although represented by two genera in the early Uintan,
ceratomorphs such as Heptodon and Selenaletes were replaced by and by Epihippus alone in the late Uintan (see Figure 35.2), the abso-
taxa such as Desmatotherium and Dilophodon, and the first amyn- lute numbers of fossil equids known from this time interval are very
odont rhinocerotoid was apparent (see Figures 35.3, 35.4). sparse, in contrast to the great abundance of individuals belonging
The Uintan marked the zenith of perissodactyl diversity at the to species of Hyracotherium in the early Eocene (see discussion in
family level. "Eomoropid" chalicotheres, the more derived tapiroid MacFadden, 1992; Janis, 1993).
Colodon, and the first true tapirid appeared in the early Uintan; There is little in the way of biogeographic variation during this
brontothere diversity increased to include larger, more specialized earlier part of the Eocene, although perissodactyls in general appear
homed forms. By the late Uintan almost all brontotheres were to be sparsely represented in the California coast and northern Great
large, homed forms, and hyracodontid rhinocerotoids made their Plains faunas. Note that rhinocerotoids and tapiroids are found in
first appearance (see Figures 35.3, 35.4). Ironically, during this California in the late middle Eocene; equids are absent.
 Perissodactyla and Proboscidea 517

iii' iii' ii:' ii:' ii:' 0

0 ~ !!!. !!!. ~ ~ z
.......
_o ~
c
c ·;;;
.. OI
~·=
c
c .5
Ill
.. OI
111- !:!
Ill
c c.5
.. OI
Ill-
Ill e:.
....
() :l
~
u;
OI
0
-C Ill
01 _!m .cm .c D.
-c D.
.c D. =~ 0
Ill 0
~1ii ... () .c
Oo
(3 oil
t: 10
z0 ..
Ill 8i
Ill ..
0 Ill
Ill ..
Ill 01
Of
t: -
0"'
Zf "' t:
D. 0
ui
Cl Cl Cl Cl Cl z :Z
/ / Di,
Br, Fl, / Di,
Me, Pe / Fl

/
Di, Me
/
Di
Di /
Di, Me
/

Di Di
Di

Hy,
Di Di

Mt, Hy, Di
Am, Di,
Sh
?Hy,

Sh, Tg

Hy,
Sh,Tg

Hy,
Sh,Tg

Hy, Td,
Pt, Tg

Hy,

Figure 35.3. (Cont.)

WHITE RIVER CHRONOFAUNA (unique to this time interval) (see Figure 35.4). The only remaining
basal ceratomorph taxon was Toxotherium, first appearing in the
For the other groups discussed in this volume, a distinctive change is Duchesnean, although amynodontid and hyracodontid rhinocero-
apparent at the start of the Chadronian (late Eocene), heralding the toids maintained their late Uintan diversity. Distinctive additions
appearance of the "White River chronofauna" (see discussions in include definitive rhinocerotid genera (Penetrigonias and Teletac-
summary Chapters 4 and 22, this volume, relating to Tertiary North eras ), and the "rebound" of the equids with the appearance of the
American chronofaunas). In contrast, a more distinctive change in hyracotheriine Haplohippus (known only from the Duchesnean) and
perissodactyl diversity is seen earlier, in the late middle Eocene the anchitheriine Mesohippus (see Figures 35.2, 35.3).
(Duchesnean), with a great reduction in generic diversity. By this Further, although less dramatic, change is apparent at the start
time, "eomoropid" chalicotheres and isectolophids were extinct, of the Chadronian. Brontotheres rediversified with the appearance
and brontotheres were reduced to the single genus Duchesneodus of the extremely large, specialized eubrontotheres Brontops,
 Christine M. Janis et al. 518

Figure 35.4. Biogeographic ranges

u m m ii:'
e
ii:'
~
ii:'
~ z 0 of tapiroids, chalicotherioids, bron-

--
u ~ ec ~

-I! ·c:
<
~ - .. -
~
I ll
_o
Ill_.
c
c "iii
.. Ill
c·-
.. Ill E·~
Ill

~s
c
Ill
c.5
Ill
.. Ill
() :fl ~
!:.
Ill
totheriids, and other perissodactyls. A
"box" (for a particular time period in a

--
CD-

-g ~e
Ill ~ -c Ill
Ill !m
CD Ill
.cm CD-
.c D. -c_
D.
.c D. !E ;:: 0 particular biogeographic region) with
() .c
&i
()
0
MA
Q)
- CD =
CD O
Oo t: 1a
I!!
:I Ill
0 CD
CD Ill
0 I!!
0 -Ill
t:
z I!! Ill t:: w a cross through it means no fossil lo-
!)~~ occ 5i E :I
CJ (j oil 0
ZCJ
U) ..
CJ
U) ...
CJ CJ CJ
D. 0
z z calities are known for that time pe-
1.8 riod from that area; a single dashed

~
/
c:
!3c:
' /
I line through the box means only scant
I Ta Ta fossil information is available (usually
Q) Ta / I
-<O -<O
co I I/ only a single, small, locality). Key:
w 2.4 ...J
z "Tapiroids" are in roman type. Isec-

~ ~
w
c:
<O I
>- 0 Ta tolophidae are in italics underlined:
0 -.:: ~ I
0 Ca= Cardiolophus, Ho = Homo-

xx
<O -
:i 4.5 wee II

x
galax, !.§. = lsectolophus. Basal cer-
0. I
- C..c
,SJ ~ Jl1
jJ:'.2
I T?
Ta
IX /
/ atomorphs (and others) are in regu-
Jar italics: De = Desmatotherium, Di

x
5.2 I = Dilophodon, Fo = Fouchia, Hp =

x
/ / Heptodon, Hy = Hyrachyus, Pm =
6. c: I /
/ Paleomoropus, Se = Selenaletes, Sz
2 ~ Jl1 I Ta
/
Ta
/

x
j J: :2 = Schizotheriodes, Tx = Toxotherium.
I/ /
Tapiroidea are in roman type under-

x
6.0

~
~. lined: Co= Colodon, He= Helaletes,
... c. c
~~~ Tv Ta Ta Ht= Heteraletes. Tapiridae are in reg-
_j I :C ular roman type: Mi = Miotapirus,

x
7.0 / I / ,/
Ta= Tapirus, Tn = new tapiroid
~a. c / / / / genus, Tv = Tapiravus. T? = tapiroid
i1J E .!!! Ta / / / / indet. Chalicotherioids are in plain
w~~ I/ I/ I/ I/

x
8.8 boldface: Ep = Eomoropus, Gr =
/ /
C: c I T? I Grangeria, Mo = Moropus, Ty =
Cl>~-~ Ta T? T? / Ta, /
-<O -<O c I / Tylocephalonyx. C? = chalicotheriid
w 0 / /
...J (.) "C 1, C? indet. Brontotheriidae are in bold-
zw 9.5
I/
"' /
I/

~
' face italics; homed taxa (Brontotheri-
0 C: c I I
~~-~ Ta Ta Ta / inae) are underlined: Br = Bron-
0 ... <O c I
:!] <O - 0 / I tops, Dp = "Diplacodon" progressum,
w (.) v

x
"C I/ 11
11.0 Du = Duchesneodus, Dx = prim-
I I / /
I itive diplacodont (sensu Jato, genus

ix
Q) '
- 0 / / / Ta, Tv, /
..5 "F! c
/ / I undetermined), Eo = Eotitanops, !J!.
_j ~ ·~ /
I/
/ C?
I/ v = Eotitanotherium, Eu = eubron-
I/ I/
12.5 / tothere (genus undetermined), Me =

~
26 / Tv I
('Cl- Mo, Mesatirhinus, Mg= Megacerops, Mn
Ty

14.0
...J ~
u..i ~ ·~
c T? C?
IX I/
/
/
C? I = Menops, Mr = Metarhinus (=
Rhadinorhinus), Mt = Metatelmath-
6
x
I Tv/ erium, No = Notiotitanops, Pa =
>.- T?,
i::::: ~ c: Mo Ty C? Palaeosyops, Pr= Protitanotherium,
I I
i1l ~ ·~ Mo
! v Pp= Protitanops, Sp= Sphenocoelus

x
I
15.8 ,

x
Cl c I I (= Dolicorhinus), St= Sthenodectes,
c <O / Ty /
2·E~
Ty
? I Te = Telmatherium. Perissodactyla in-
/ Mo Mo /
«! ~ J2 , / I certae sedis are in boldface italics don-
17.5 ble underlined: La= Lambdotherium.

Megacerops, andMenops. Colodon was the only surviving tapiroid. of more derived genera such as Trigonias, Amphicaenopus, and the
True tapirids appear to be absent during the Duchesnean, but this diceratheriine Subhyracodon (see Figure 35.3).
may represent a sampling artifact because the tapirid Protapirus is The transition from Eocene to Oligocene was marked by the ex-
known from the late Uintan and Duchesnean, and tapirids reemerged tinction of the brontotheres and the appearance of the tapirid Pro-
in the Oligocene (see Figure 35.4). Anchitheriines were now the only tapirus, but otherwise perissodactyl diversity was little affected.
surviving equids, with the additional appearance of Miohippus in Some fauna! changeover was apparent during the later Oligocene.
the early Chadronian (see Figure 35.2). Among the rhinocerotoids, The last definitive appearance of the tapiroid Colodon, the
amynodonts and hyracodonts were represented only by a single amynodontid Metamynodon, and the rhinocerotids Penetrigonias
genus, Metamynodon and Hyracodon, respectively, but diversity in- and Amphicaenopus was in the Whitneyan, and the rhinocerotid
creased among the rhinocerotids, with the Chadronian appearance Diceratherium made its first appearance (see Figures 35.3, 35.4).
 Perissodactyla and Proboscidea 519

u m m ii:'
e ii:' ii:'
ze:. u~
< ~ u ~ ec ~ ~

~
...
Ill
_o
......
('Cl_.
c
c 'iii ...cCD ·-.,, c
Ill
.5 c
Ill
E·i () ...:fl Ui
Ill

-- ~i
... ('Cl ('Cl

-g ~e
('Cl ...('Cl
... Ill ftl CD- CD-
m
- l! m 0
-('Cl·-fl 0 c .c .c D. ... D. .cD. !E ;::
Q)
...... 0 CD 0
Oo
('Cl
t::-
0 ('Cl '50 'taCD :I Ill
c_
CD ftl t:: 'Iii () .c
('Cl t::
0
w
MA - CD :; z I!! 0 I!! 0 I!! z0 ...CD
5i E
!I:!!!~ oci:
D. 0
CJ ~oil CJ
U) ...
CJ en CJ CJ CJ z z
17.5
g> c I Pt, / / Mi, / Pt T?/
;.,·- ct!
w -.:: E '5 / ?
z ct!
w J: J2
Q) ... / / Mo, Ty
w
0 18.8
2 c:
0 '
ct! Mo/ / Mi, /
ct!
:!] ....
...J~
Q)
fl! / / /
...J <( <O
19.2 Q) c:
iii
...J,,:,, Q)
ct! / /
fl!
u..i~ ct! Mo /
23.0
£ c_
ct! •
w~ a>
m Mi,
_j <m Mo
27.7
~ c:
w a; ct!
Mi
w,,;, Q)
z fl!
w u..i~
g29.4 c:
<O

CJ :c>-
ct! Co
:i ~ Q)
c:
0
31.9
c:
.!'!! QQ, Pt
~
0
33.4
-g' .!!!c: QQ,

:5 B e
Q)
QQ

34.5
Q) ' c: QQ,
~ -g .!!! QQ I
::2 ..c
·- l5
(.) ...
?!f!!, ?Mg,
?Mn I
~35.5
w ;., "C
'
ct!
c: Tx,QQ, Tx, Co, I
~ t'O·-
0 ct! ..c c: ?!f!!, ?Mg, !f!!, Mg, I
0 WU e ?Mn Mn
w 37.1
"'
Q)
c:
..c ct!
0
::i Q)
c:
QQ,
Du
Co,!::!l,
?12M
0
39.5
c: ?Di, Sz,
Q) .!l! QQ,
c:
-ct!·-
...J :::> ?He, !::!l
41.3 Figure 35.4. (Cont.)

The earliest Arikareean marked the last appearance of the hyra- tion by the end of the Whitneyan in the central Great Plains (see
codontid Hyracodon and the rhinocerotid Subhyracodon, and the Figure 35.4). Rhinocerotids are virtually the only perissodactyls to
first definitive appearances of the equid Kalobatippus and the tapirid be found in the California coast faunas.
Miotapirus. By the late early Arikareean the first true chalicothe-
riid, Moropus, immigrated from Eurasia (see Figures 35.2, 35.3,
RUNNINGWATER CHRONOFAUNA
35.4).
There is little in the way of biogeographical diversity in this The early Miocene heralded the appearance of the "Runningwa-
Eocene/Oligocene time period, except for the relict survival of the ter chronofauna." The major change in perissodactyl diversity at
tapirid Protapirus, and possibly also of the tapiroid Colodon, in this time was in the equids, with the appearance and radiation at
the John Day faunas of the Pacific Northwest, after their extinc- this time of more derived anchitheriines such as Archaeohippus,
 Christine M. Janis et al. 520

iii' iii' £' a. £'

ec ~ 2. e z e0
0
.......
_<.>
I'll ._.
E 'iii
~
c
....ccGI ·-m
"' E·~"' ~:!
"'
c "'
c.5 ..
e:. Ui
c "'
... l'CI
gi I'll

GI O
I'll
(.) (.)
l! .!
t::-
0 l'CI
.....
:I l'CI
I'll GI -
.c a.
'S 1ii
-c a.
...
GI I'll
(.)I!!
GI -
.c a.
t:: ..
0 l'CI
lE
u
u
l'CI
:=
.c
t::
0
(.)
ui
MA z I!! en0 ...GI ~I!! z I!! a. 0
~oil CJ CJ CJ CJ CJ z z
41.31---+-+-....,i~-+-...,.--+--r---+--~~,...,
/
/

§,De, Di, /
Hy,]jg, Hy
/

/ I
/ I

/ I I
I I

Figure 35.4. (Cont.)

Desmatippus, and Parahippus. The large, specialized anchitheriine immigrants appeared in the early Hemingfordian, including the
Hypohippus first appeared in the early Hemingfordian. Mesohip- aceratheriines Floridaceras and Aphelops and the teleoceratine
pus was now extinct, although Miohippus survived as a rare fau- Brachypotherium (see Figure 35.3). Chalicotheres were relatively
na! element through the early Hemingfordian. The earliest equine abundant in this time period, with the appearance of Tylocephalonyx
equids may be present at the start of the Miocene, although the genus in the early Hemingfordian. Tapirids were the only surviving
Merychippus is not definitively known until the early Hemingfor- tapiroids, with little change in their diversity over their Oligocene
dian (see Figure 35.2). condition (see Figure 35.4).
Rhinocerotid diversity increased with the appearance of Meno- In contrast to the patterns observed in other large mammals,
ceras in the early late Arikareean, although the similarly paired- where the Gulf Coast faunas were rather distinctive, there appears
horned Diceratherium continued to survive until the late Heming- to be little biogeographical differentiation during this time period
fordian, and possibly into the early Barstovian. More rhinocerotid among perissodactyls. Note, however, that Aphelops is known only
 Perissodactyla and Proboscidea 521

from this region, and that Protapirus is known from here but is Great Plains faunas). The derived equinine Dinohippus, sister taxon
absent in the central Great Plains; Miotapirus shows the opposite to the extant Equus, first appears in the early Clarendonian (see
pattern. Figure 35.2).
Gomphotheres remained moderately diverse among the
CLARENDONIAN CHRONOFAUNA proboscideans, with the appearance of the immigrant shovel-tusked
gomphothere Platybelodon. Zygolophodon was replaced by the
The late Hemingfordian (early late Miocene) represents the start larger, more specialized mammutid Mammut (see Figure 35.5).
of the Clarendonian chronofauna. Although proboscideans are gen- Rhinocerotid generic diversity was little changed over their mid-
erally thought to be absent from North America until the middle dle Miocene condition. Only one possible record of a chalicothere
Miocene (e.g., Tedford et al., 1987), note the isolated occurrence is known from the Clarendonian, but tapirs persisted as occasional
of two specimens of the mammutid Zygolophodon in the late Hem- fauna! elements.
ingfordian (from the Massacre Lake Local Fauna in Nevada [local- There are few patterns ofbiogeographic variation in the Clarendo-
ity NB 17] and the Deep River Local Fauna in Montana [locality nian chronofauna. The apparent continued survival of chalicotheres
NP34D]). However, as also noted by Tedford et al. (1987), pro- into the central Great Plains region alone is probably a sampling
boscideans are apparently absent from most areas until the late artifact. Dwarf rhinocerotids appear to be confined to southern and
Barstovian. Even the well-sampled central Great Plains yields only western regions in the Clarendonian. The Gulf Coast faunas do not
one dubious proboscidean. Gomphotheriid proboscideans first ap- appear to have acted as a refuge for the browsing anchitheriine
peared in the early Barstovian, with a diversification in the late equids. Although sampling was poor during the late late Barstovian
Barstovian, including the appearance of the shovel-tusker Serbe- and Clarendonian in the northern Great Plains and Pacific North-
lodon in the late Barstovian/early Clarendonian (see Figure 35.5). west, perissodactyls (especially rhinocerotids) were rather conspic-
Among the perissodactyls, the inception of the Clarendonian uously underrepresented from these regions during this time period.
chronofauna appears to be marked by new appearances, rather than Mammutid proboscideans appear to have been confined to the Gulf
by extinctions of earlier taxa (in contrast to other groups of large Coast and the Pacific Northwest during the Clarendonian.
mammals). Equine equids showed a diversity of species of
"Merychippus," representing both equinine and hipparionine lin-
MIO-PLIOCENE CHRONOFAUNA
eages of this paraphyletic genus. The hipparionine Hipparion and
the equinines Protohippus and Calippus appeared early in this time By the start of the Hemphillian, specialized browsing perissodactyls,
interval. Anchitheriines remained moderately diverse, although such as anchitheriine equids and chalicotheres, had become extinct.
more primitive anchitheriines such as Miohippus were now extinct. In contrast, the browsing rhinocerotid Aphelops survived, possibly
In the early late Barstovian there was the appearance of a greater di- indicative of a different habitat for this animal than for the other
versity of more derived equids: the hipparionines Cormohipparion, browsers (perhaps resembling the present-day savanna-dwelling
Neohipparion, and Pseudhipparion, and the equinine Pliohippus browsing rhinoceros, Diceros bicomis ). Dwarf species of rhinocero-
(see Figure 35.2). toids were now extinct. Aphelops and Teleoceras survived into the
Rhinocerotid diversity was now almost entirely comprised of ac- earliest Pliocene, and there is a single record of a possible Teleo-
eratheriines and teleoceratines, although Diceratherium survived ceras from the early Blancan (see Figure 35.3). Apparently beating
into at least the late Hemingfordian. The short-legged, probably all evolutionary odds, the extant tapirid Tapirus (also a browser)
semiaquatic, grazing teleoceratine Teleoceras first appeared in the continued through the Pliocene, although it is only known from
late Hemingfordian. Dwarf rhinocerotids were common in the mid- occasional specimens.
dle Miocene, with the appearance of the aceratheriine Peraceras Grazing equids continued at moderate diversity through the late
hessei in the early Barstovian and Teleoceras meridianum in the Miocene, although all species assigned to the genus "Merychippus"
early late Barstovian (see Figure 35.3). Chalicotheres and tapirids were now extinct. The monodactyl equinine Astrohippus appeared
continued to appear as rare elements of the fauna during the middle in the late early Hemphillian. Equid diversity showed a steady de-
Miocene, with the appearance of the extant genus Tapirus in the late cline through the Mio-Pliocene chronofauna, however: Protohip-
late Barstovian (see Figure 35.4). pus and Hipparion were extinct by the late Hemphillian, and the
Perissodactyl diversity patterns showed a shift in the Clarendo- Gulf Coast faunas appeared to act as a refuge, especially during
nian (early late Miocene), paralleling a similar shift in artiodactyl the Pliocene, for genera such as Cormohipparion, Neohipparion,
diversity (see Chapter 22, this volume), with a decline in the num- Calippus, and Pseudhipparion. The cause for this drop in diver-
bers of browsing taxa and an increase in the numbers of grazers, sity seems related to a major global change related to increased
probably representing a reduction of woodland habitat. aridity and seasonality and the spread of lower productivity prairie
Anchitheriine equids persisted until the end of the Clarendonian, grasslands (Cerling, Wang, and Quade, 1993; MacFadden and Cer-
but were now represented only by the very large, highly specialized, ling, 1994). The extant Equus first appeared in the Blancan and was
genera Hypohippus and Megahippus (see also Janis, Gordon, and virtually the only surviving genus by the end of the Pliocene (see
Illius, 1994). In contrast, the diversity of equine equids increased. Figure 35.2).
The early Clarendonian was the zenith of equine generic diversity, Proboscideans were diverse in the late Miocene, with the appear-
with ten sympatric taxa (all of which could be found in central ance of the shovel-tusker Amebelodon in the early early Hemphillian
 Christine M. Janis et al. 522

~ ~
u
~ u ~
iii iii
!!!.
~
!!!. 2. ~ ze:. u~
~
2. Ui("Cl
_(.) c c
c·- c .E"' "'c c .E"'("Cl Ui("Cl
...!c ::-"'
c"iii
... ("Cl ...GI "'l1l ... ("Cl
!:i ...
GI -
~i
......
-"
GI -

-g m
-fl
......
("Cl·-
0
(.) GI O
("Cl 1! Ill .c Ill .c D. ... Q.
c ...
.c D.
t:: ... u .c
0
(.)
e: Q) .., GI (.) (.) t:: "Iii 'S -; :I ("Cl GI l1l 0 ("Cl ("Cl t:: ui
MA 5j E :; 0 !!! 0
...
GI
... 0 GI (.) !!! z !!!
!I:!!!~ (.) ca: (j oil z"
(/) (/) D. 0
z z
1.8

Q)
c
l1l
u
c I Cu, Go,
Ma, Rh/
/
/

St
Ma,
Cu, Rh,
" "
Rh, St
Ma,
Rh, St
"
x "
Ma,
St I
I

xx x xx
-<a-
l1l
w ...J co I Mt / St
z 2.4
w
0
0
:i 4.5
~~
wc:o
c
l1l
>- u
I
I
St Ma Cu, Rh,
St
M?,
St
Ma,
Rh, St
Ma
x
Q.

5.2
en a. c:
2~~
.~b: '.E
I
I
i Ma,
Rh,
Go
Rh, St Rh G?
,
/
/

x
~ x
/ /
6. c I Ma, Ma, Ma, Ma,
/ Ma, /
2 ~ ,m I Rh, Ak, Ak,?Gn, M?,
'.E /?Rh /

x
j J: Go l,1 . Rh, St, Rh Go G? Figure 35.5. Biogeographic ranges
/

x
6.0 of proboscideans. A "box" (for a par-
~. Am, To,
... c. c Ak,Am, ticular time period in a particular

t><
Ma, Ma Am M?
.11 ~ :lll IQ
Ma, ?Rh,
biogeographic region) with a cross
_j J: :E Go Go Go
through it means no fossil localities

x
7.0 I /
>- May '
I I
Am, are known for that time period from
~ 6. c: Am, / ~/ / Am, Am/
w E.!!l Rh ?Rh, that area; a single dashed line through
Go / / / / I
ui~:E v 11 I/ Go
Go
I/ V Am the box means only scant fossil infor-
8.8 / / mation is available (usually only a sin-
C: c Go; Ma,/ '
Cl>~-~ Go Go ~ / .EL~ / gle, small, locality). Key: Mammuti-
l1l c
-<a- I Go / Go /
1X
xx
w 0
dae are in boldface: Ma = Mammut,
zw 9.5
...J () -0 I
11 11 v /G?.
/
Zy = Zygolophodon. M? = mammu-
0 C: c Ma, ' I
~ Go, Eu, Go, / I tid indet. Gomphotheriidae are in ro-
0 ~~-~ Gn,Go, Go Go
:e 11.0 .... l1l c
l1l- 0
W() -o Me Go Te Te / I I man type face. "Shovel-tuskers" are in

6
/ I /
I/
/
I/
. I/
I
roman type underlined: Ak = Amebe-
lodon (Konobelodon), Am= Amebe-

x
,S? Se/ G?/
j~c M?, Go/ Zy, / /
lodon, I'! = Platybelodon, Se = Ser-
I I
_j &l ·~ Go
11
/
I/
/
I/
/ Eu, Go, Me
I/
/
I/ 11 belodon, To = Torynobelodon. "New
12.5 / World forms" are in italics: Cu =
Go;

x
Q)
iU .8
'
Zy, Zy, / Zy, M?, Cuvieronius, Rh = Rhyncotherium,

14.0
...J ~
ui &l ·~
c:
t>< P?
Go
Go, Me
I/
/
/
Go Go G?
I/
I St = Stegomastodon. Other gom-
photheres are in regular roman type:
Eu = Eubelodon, Gn = Gnathabe-

x
>--
6 Go; P? Zy P? Go
I
-;;::: ~ c lodon, Go = Gomphotherium, Me =
l1l l1l l1l I I
wc:o·s; I I Megabelodon. G? = gomphotheriid
15.8 indet. Other elephantoids are in bold-

~
/
bi c I

17.5
ai
-
j J: .l2
c
.E '5
Q)
l1l

....
I/
/
/ Zy
x Zy
/
/
/

I
I
face italics: Mt = Mammuthus, Te =
Tetralophodon. P? = proboscidean in-
det.

and of other shovel-tusked taxa such as Torynobelodon and Gnatha- and possibly Gomphotherium), there were few patterns of biogeo-
belodon, which may also have survived during the Hemphillian. graphical variation. Rhinocerotids were now known from latest
However, shovel-tuskers did not survive into the Pliocene. The gom- Miocene Pacific Northwest faunas (perhaps their earlier apparent
phothere Gomphotherium may have survived into the Pliocene only absence merely represents a sampling artifact). Despite the diver-
in the Gulf Coast faunas. The "New World" form Rhynchotherium sity of equids elsewhere on the continent, Pliohippus was almost
may have been present as early as the earliest Hemphillian, joined the only equid known from the latest Miocene Pacific Northwest.
by Stegomastodon in the late Hemphillian. These New World forms, In contrast, Dinohippus was almost the sole equid in northern Great
joined by Cuvieronius in the Blancan, formed the majority of Plains faunas. This apparent difference could well represent nomen-
Pliocene proboscidean diversity, although the mammutid Mammut clature bias because species of Dinohippus and Pliohippus are fre-
was also present during the Pliocene (see Figure 35.5). quently mistaken for each other, and both genera are in exten-
With the exception of the taxa mentioned here that survived sive need of revision (see comments in MacFadden, this volume,
only in Gulf Coast faunas into the Pliocene (various equid taxa Chapter 37).
 Perissodactyla and Proboscidea 523

SUMMARY REFERENCES
Perissodactyls were diverse in Eocene faunas, with a zenith of di-
Berger, J. (1986). Wild Horses of the Great Basin. Chicago: University of
versity at the family level in the Uintan. This late middle Eocene Chicago Press.
perissodactyl diversity comprised brontotheres, "eomoropid" chal- Blainville, de, H. M. D. (1816). Prodrome d'une nouvelle distribution
icotheres, and a diversity of tapiroids and ceratomorphs: Perhaps systematic du regne animal. Bulletin des Sciences, Societe Philo-
paradoxically, equid diversity was low at this time. mathique de Paris, Serie 3, 3, 105-124.
Borissiak, A. (1945). The chalicotheres as a biological type. American Jour-
A later Eocene reduction in perissodactyl diversity occurred
nal of Science, 243, 667-9.
slightly earlier (Duchesnean) to the fauna shifts seen in other large Cerling, T. E., Wang, Y., & Quade, J. (1993). Global ecological change
mammal groups (early Chadronian). Extinctions at this time in the late Miocene: expansion of C4 ecosystems. Nature, 361,
included chalicotheres and nontapirid "tapiroids." A "White River" 344-5.
(late Eocene-Oligocene) perissodactyl chronofauna was comprised Fortelius, M., & Kappelman, J. ( 1993). The largest land mammal ever imag-
ined. Zoological Journal of the Linnean Society, 107, 85-101.
primarily of anchitheriirie equids, tapirids, and rhinocerotine rhino-
Hooker, J. J. (1984). A primitive ceratomorph (Perissodactyla, Mammalia)
cerotoids. Brontotheres were prominent as large specialized forms from the early Tertiary of Europe. Zoological Journal of the Linnean
in the Chadronian, but did not survive the Eocene. Amynodontine Society, 82, 229--44 .
and hyracodontine rhinocerotoids were also most diverse in the late (1989). Character polarities in early perissodactyls and their significance
Eocene and did not survive the Oligocene. for Hyracotherium and infraordinal relationships. In The Evolution
of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 79-101.
A Runningwater chronofauna, commencing in the early Miocene,
Oxford: Clarendon.
was marked by the radiation of more larger and specialized anchith- ( 1994). The beginning of the equoid radiation. Zoological Journal of the
eriine equids, newly immigrant chalicotheriid chalicotheres, and Linnean Society, 112, 29-63 .
a diversity of immigrant rhinocerotids. Change at the start of the Janis, C. M. (1976). The evolutionary strategy of the Equidae, and the origins
Clarendonian fauna, in the late Hemingfordian (late early Miocene), of rumen and cecal digestion. Evolution, 30, 757-74.
(1993). Tertiary mammal evolution in the context of changing climates,
contrasts with changes in faunal diversity seen in other large mam-
vegetation, and tectonic events. Annual Review of Ecology and Sys-
mals at this time in that there were few generic extinctions. However, tematics, 24, 467-500.
this time was marked by the appearance of numerous new taxa, pri- Janis, C. M., Gordon, I., & Illius, A. (1994). Modelling equid/ruminant
marily representing the endemic diversification of equine equids competition in the fossil record. Historical Biology, 8, 15-29.
and the immigration of the mammutid proboscideans . MacFadden, B. J. (1992). Fossil Horses: Systematics, Paleobiology, and
Evolution of the Family Equidae. Cambridge: Cambridge University
Proboscideans were not generally common in North America un-
Press.
til the later middle Miocene, with gomphotheriids appearing for the MacFadden, B. J., & Ceding, T. E. (1994). Fossil horses, carbon isotopes,
first time in the early Barstovian. Both large specialized rhinocero- and global change. Trends in Ecology and Evolution, 9, 481-5.
tids (including the probable grazing hippo mimic Teleoceras) and Novacek, M. J., Wyss, A. R., & McKenna, M. C. (1988). The major groups
dwarf rhinos were common in the middle Miocene, and a diversity of eutherian mammals. In The Phylogeny and Classification of the
Tetrapods, Vol. II: Mammals, ed. M. J. Benton, pp. 31-71. System-
of both browsing and grazing equids was apparent.By the Clarendo-
atics Association Special Volume no. 35B. Oxford: Clarendon.
nian (early late Miocene), although generic diversity of the grazing Owen, R. ( 1848). Description of teeth and portions of jaws of two extinct An-
equine equids was at its zenith, diversity was greatly reduced in thracotherioid quadrupeds (Hyopotamys vectianus and Hyop. bovi-
the browsing anchitheriines. Both anchitheriine equids and chali- nus) discovered by the Marchioness of Hastings in the Eocene de-
cotheres (also browsers) were extinct by the end of the Clarendo- posits on the N.W. coast of the Isle of Wight: with an attempt to
develope Cuvier's idea of the classification of pachyderms by the
nian, but some browsers, such as the rhino Aphelops and the extant
number of their toes. Quarterly Journal of the Geological Society of
tapir Tapirus, survived. London,4, 103--41.
In the Mio-Pliocene chronofauna a diversity of equids and large Prothero, D.R., Manning, E. M., & Fischer, M. (1988). The phylogeny of the
rhinocerotoids continued during the late Miocene, but had declined ungulates. In The Phylogeny and Classification of the Tetrapods, Vol.
by the Pliocene. The last North American rhinocerotoid was known II: Mammals, ed. M. J. Benton, pp. 201-34. Systematics Association
Special Volume no. 35B. Oxford: Clarendon.
from the early Blancan, and by the late Blancan the extant Equus was
Prothero, D.R., & Schoch, R. M. (1989). Origin and evolution of the Perisso-
virtually the only surviving equid. The latest Miocene saw a flour- dactyla: summary and synthesis. In The Evolution of Perissodactyls,
ishing of shovel-tusked proboscideans, but both these and regular eds. D.R. Prothero & R. M. Schoch, pp. 504-29. Oxford: Clarendon.
gomphotheres were replaced by "New World form" gomphotheres Radinsky, L. B. (1964). Paleomoropus, a new early Eocene chalicothere
in the Pliocene. The mammutid Mammut survived from the late (Mammalia, Perissodactyla), and a revision of Eocene chalicotheres.
American Museum Novitates, 2179, 1-28.
Miocene through the Pliocene.
Rose, K. D. (1985). Comparative osteology of North American dichobunid
Tertiary perissodactyls and proboscideans do not exhibit pro- artiodactyls. Journal of Paleontology, 59, 1206-26.
found patterns ofbiogeographical diversity, although perissodactyls Schoch, R. M. ( 1983). Relationships of the earliest Perissodactyls. Geolog-
appear to have been generally sparse in their distribution in the ical Society of America, Abstracts with Programs, 15, 144.
Northwest and along the California coast. The Gulf Coast fau- (1984). Introduction. In Vertebrate Paleontology, ed. R. M. Schoch,
pp. 1-16. New York: Van Nostrand Reinhold.
nas were not particularly distinctive in their perissodactyl and pro-
(1985). Concepts of the relationships and classification of major peris-
boscidean components, except for the role that the Florida faunas sodactyl groups: notes for a workshop on fossil perissodactyls held
appear to have played as a refuge for the earlier diversity of equine at Fourth International Theriological Congress, Edmonton, Canada,
equids in the Pliocene and latest Miocene. August 1985. Privately printed and distributed.
 Christine M. Janis et al. 524

(1989a). A brief historical review ofperissodactyl classification. In The Simpson, G. G. (1945). The principles of classification and a classification
Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch, of mammals. Bulletin of the American Museum of Natural History,
pp. 13-23. Oxford: Clarendon. 85, 1-350.
(1989b). A review of the tapiroids. In The Evolution of Perissodactyls, Tedford, R.H., Skinner, M. F., Fields, R. W., Rensberger, J.M., Whistler,
eds. D. R. Prothero & R. M. Schoch, pp. 298-320. Oxford: D. P., Galusha, T., Taylor, B. E., Macdonald, B. E., & Webb, S.
Clarendon. D. (1987). Fauna! succession and biochronology of the Arikareean
Scott, W. B. (1941). Perissodactyla. The mammalian fauna of the White through Hemphillian interval (late Oligocene through earliest
River Oligocene. Transactions of the American Philosophical Pliocene epochs) in North America. In Cenozoic Mammals of North
Society, 28, 747-980. America, ed. M. 0. Woodburne, pp. 153-210. Berkeley: University
Shoshani, J. (1993). Hyracoidea-Tethytheria affinity based on myological of California Press.
data. In Mammal Phylogeny: Placentals, eds. F. S. Szalay, M. J. Wood, H. E., II (1934). Revision of the Hyrachyidae. Bulletin of the
Novacek, & M. C. McKenna, pp. 235-56. New York: Springer- American Museum of Natural History, 67, 181-295.
Verlag. (1937). Perissodactyl suborders. Journal of Mammalogy, 18, 106.
36 Brontotheriidae
BRYN J. MADER

INTRODUCTION

The Brontotheriidae is an extinct perissodactyl family whose ex-

act phylogenetic relationship to other members of the order has
not been firmly established. Brontotheres (=titanotheres) ranged
in size from small dog-size forms to giants rivaling medium-sized
proboscideans. The group is characterized by the distinct bunoselen-
odont pattern of the molars and, in most cases, by a shortening of the
face and elongation of the skull posterior to the orbit. Members of the
subfamilies Brontotheriinae ( = Telmatheriinae, sensu Mader, 1989)
and Dolichorhininae evolved frontonasal hornlike prominences that
became spectacular bifurcated structures in certain Chadronian (late
Eocene) brontotheriine genera.
Brontotheres were obligatory browsers occupying warm temper-
ate to subtropical environments with habitats ranging from forest to
relatively open woodland. The group probably originated in North
America either during the latest Wasatchian (early-middle Eocene
boundary) or earliest Bridgerian (early middle Eocene) and reached
its maximum diversity in the Uintan (early late middle Eocene).
Brontotheres became extinct in North America at the end of the Figure 36.1. Restoration of Duchesneodus, by Brian Regal.
Chadronian (latest Eocene), but may have survived in Asia as late
as the middle Oligocene. the entire skull length. Although shortening of the face is clearly
a synapomorphic character that defines the vast majority of bron-
totheres, these derived cranial proportions were probably not shared
DEFINING FEATURES OF THE by the most primitive brontothere genus, Eotitanops. Eotitanops is
FAMILY BRONTOTHERIIDAE known only from fragmentary remains, but one partial skull (AMNH
14887) suggests that the face was probably long like in other early
perissodactyls.
CRANIAL
Another derived condition shared by the majority ofbrontotheres,
Brontotheres exhibit a great diversity of cranial morphologies rang- but lacking in Eotitanops, concerns the morphology of the external
ing from extreme brachycephaly (Palaeosyops) to extreme dolicho- nares and configuration of the bones surrounding the lateral nasal
cephaly (Sphenocoelus ). In most genera, the part of the skull anterior incision. In Eotitanops the external nares are not very deep and are
to the orbit is shorter than the part of the skull behind the orbit. The bordered either entirely, or nearly entirely, by the premaxilla (based
face is short, therefore, and the rest of the skull elongated. In cer- on AMNH 14887). The maxilla is thus excluded or nearly excluded
tain members of the subfamily Brontotheriinae (diplacodonts sensu from the lateral nasal incision. In all other brontotheres, however,
Jato), this shortening of the face became quite extreme, and in these the external nares are deeply recessed so they are broadly bordered
brontotheres the face typically represents less than 40 percent, of by the maxilla.

525
 Bryn J. Mader 526

a f~
b ~~ g k

~ c
Figure 36.2. Lateral views of North
American brontothere skulls. A. Eoti-
tanops. B. Palaeosyops. C. Tel-
matherium. D. Metatelmatherium. E.
Sthenodectes. F. Mesatirhinus. G.
Sphenocoelus. H. Eotitanotherium I.
"Diplacodon" progressum. J. Proti-
tanotherium. K. Menops. L. Bron-
d tops. M. Megacerops. Scale bar: A-J
= 50 cm; K-M = 89 cm. Illustra-

~ tions A, B, F, G, K, L, and M af-

~
ter Osborn, 1929; C and D after Os-
m born, 1908; E after Douglass, 1909; H
e after Peterson, 19 l 4a; I after Peterson,
Scale
1934; and J after Hatcher, 1895.

On the ventral surface of most brontothere skulls, a posterior con- (Schoch and Lucas, 1985), however, the horns of (presumed) males
tinuation of the nasal septum extends from the internal nares to the were larger than those of females.
anterior edge of the basicranium. This posterior extension {vomer- At least two brontothere genera independently developed fron-
ine plate of Peterson, 1924) forms a stiff flange of bone that pro- toparietal domes or convexities, which, like the horns, were present
trudes ventrally between the pterygoids. A vomerine plate occurs in both sexes. In the hornless Bridgerian genus Palaeosyops, the
in Mesatirhinus, Metarhinus, Sphenocoelus, Metatelmatherium, a convexity was never very large, although in males it may have been
skull similar to that of Protitanotherium (AMNH 117163), "Dipla- larger than in females (Mader, 1989). In the primitive homed genus
codon" progressum, Notiotitanops, Duchesneodus, and in at least Duchesneodus, dome size was strongly sexually dimorphic. In fe-
one Chadronian brontothere, possibly Megacerops (MCZ 3413). males of Duchesneodus the dome was low, but prominent; in males
Although the presence of this flange may be primitive for bron- the dome was very large and similar to that of the dome-skulled
totheres, its occurrence has yet to be documented in the most primi- chalicothere Tylocephalonyx.
tive brontothere genera, Eotitanops and Palaeosyops. It is possible, As in other early perissodactyls, primitive brontotheres had
therefore, that the flange may define all brontotheres except for the prominent sagittal crests. In the derived Brontotheriinae and in most
most primitive genera. The posterior narial flange appears to have species of Sphenocoelus, however, the cranial vertex is widened and
been reduced or lost in specimens of certain Chadronian taxa (for the sagittal crest is lost as a result. In the eubrontotheres the cranial
example, some specimens of Brontops ), but it is not known whether vertex became quite wide and took on a distinctive "saddle-shaped"
this reduction or absence has any taxonomic significance. morphology.
Perhaps the most striking feature of many brontothere skulls is
the two hornlike prominences that arise anterior to the orbit. Horn-
DENTAL
like prominences evolved in the subfamily Brontotheriinae and in-
dependently in the dolichorhinine genus Sphenocoelus. The horns Brontotheres are distinguished from all other perissodactyls by the
developed along the dorsal border of the facial concavity near the bunoselenodont pattern of the upper molars in which there are a
junction of the frontal, nasal, and maxillary bones immediately dor- well-developed W-shaped ectoloph and isolated lingual cusps. Para-
sal to the presumed origin of the nasolabialis muscles (see Gregory, conules and metaconules are variably developed, large in primitive
1929). The absence of grooves for nutrient blood vessels suggests genera and reduced or lost in more derived genera. A protoloph is
that in life the horns were probably covered with tough hide rather never present and I know of only a single specimen (F:AM 128600,
than by a keratinized sheath. Primitive! y the horns were rather small the holotype of Megacerops kuwagatarhinus) in which a metaloph
structures, but in the more derived brontotheriines they became quite is present.
large and exhibited a variety of morphologies. Homs were present On the lower molars there is a distinct M-shaped loph pattern.
in both sexes and, primitively (Telmatherium and Sphenocoelus), The paraconid is typically reduced, and on all of the molars except
horn size was probably not a sexually dimorphic character. In the for m3, the hypoconulid is rudimentary. On m3 the hypoconulid is
derived members of the Brontotheriinae known as eubrontotheres always very large, a retained primitive perissodactyl character.
 Brontotheriidae 527

b
Figure 36.4. Skeletal reconstructions of A. Brontops, a derived Chadronian
eubrontothere. B. Palaeosyops, a primitive Bridgerian genus. After Osborn,
1929.

c
demonstrated that canine size is sexually dimorphic in at least some
(and possibly all) brontothere species.

d
POSTCRANIAL

Figure 36.3. Upper cheek dentition of North American brontotheres. Most brontotheres were large, bulky creatures with short necks,
A. Menops. B. Diplacodon (genoholotype = Eotitanotherium?). C. Telmath- stocky bodies, and relatively short limbs. The primitive perisso-
erium. D. Palaeosyops. Illustrations A, C, and Dafter Osborn, 1929; B after dactyl pattern of four digits on the manus (II, III, IV, and V) and
Peterson 1914a. Note the reduction of the paraconules and metaconules and
three on the pes (II, III, and IV) was never altered. As a whole, the
the progressive molarization of the premolars from D (primitive) through A
(derived). postcranial skeleton was rather conservative, and I have not been
able to identify any synapomorphic characters that can be used to
define the family. It has also proven difficult to identify postcra-
The brontothere dentition was remarkably weak and the cheek nial synapomorphic characters that define subgroupings within the
teeth were very low crowned. It is not uncommon to find very Brontotheriidae.
old individuals in which the anterior molars have been worn com- Hatcher (1893) noted that certain Chadronian brontotheres lacked
pletely fiat. In primitive brontothere genera, such as Eotitanops and the trapezium, whereas most other brontotheres clearly possessed
Palaeosyops, wear on the paracone and metacone was fairly direct this bone. The loss of this element from the carpus could be used
(producing in many specimens of Palaeosyops circular wear facets). as a synapomorphic character defining subgroupings within the eu-
In all of the more derived genera, however, shear is more greatly brontotheres, but the exact distribution of this character has yet to
emphasized and wear is concentrated along the lingual edge of the be documented.
ectoloph. Numerous postcranial morphologic changes occurred as the
In most brontotheres, the upper and lower incisors are subcanini- mediportal forms characteristic of the middle Eocene gave way to the
form and distinctly triangular in shape with a well-developed lingual graviportal forms characteristic of the late Eocene. These changes
cingulum. This character is another synapomorphy that unites the include the development of short, stocky limb elements, the acquisi-
vast majority of brontotheres but excludes Eotitanops, in which the tion of a widely splayed pelvis similar to that of proboscideans, and
upper incisors are relatively blunt and peg-shaped. The morphology the elongation of the neural spines on the thoracic vertebrae to help
of the lower incisors in Eotitanops is unknown. support the head. Because similar adaptations could have evolved
Primitive brontotheres had a total of six upper and lower incisors, in parallel in several lines of descent with increasing body size, the
relatively large canines, and unmolarized premolars. As a whole, the value of these characters in a phylogenetic analysis is questionable,
dentition remained relatively conservative throughout the group's and I refrain from using them here.
evolutionary history, and in most cases there was little change from Osborn (1929, pp. 797-798) reconstructed the pattern of arterial
the primitive pattern. By far the most morphologic change occurred circulation around the atlas in several brontotheres and concluded
in the subfamily Brontotheriinae where, in derived genera, the in- that in certain Uintan and Chadronian genera the vertebral artery
cisors became small and globular in shape and were reduced to four did not perforate the transverse process, but instead passed ventral
or fewer in number, the canines became smaller and ofless use as of- to it. If this interpretation is correct, then this character could be
fensive weapons, the postcanine diastema was severely restricted or recognized as a synapomorphic condition and used to define sub-
eliminated entirely, and the premolars became partially molarized. groups within the Brontotheriinae. My own examination of Osborn's
In eubrontotheres the first upper and lower premolars were occasion- materials, however, suggests that at least some of his observations
ally lost, but this character probably has no taxonomic significance. may have been inaccurate, and at present, there js not enough in-
Although I have previously stated that canine size is not sexu- formation to determine at what level (if any) this character may be
ally dimorphic in brontotheres (Mader, 1989), further research has synapomorphic.
 Bryn J. Mader 528

SYSTEMATICS brontotheres as the primitive sister group to all other perissodactyls.

At present there is no consensus on the correct systematic position
of the brontotheres within the Perissodactyla (see also discussion
SUPRAFAMILY
in summary chapter on perissodactyls and proboscideans, this vol-
Traditionally, brontotheres were included in the Hippomorpha (usu- ume, Chapter 35; Colbert and Schoch, this volume, Chapter 39 and
ally assigned subordinal or infraordinal status) along with equoids Coombs, this volume, Chapter 38).
(Wood, 1937; Scott, 1941; Simpson, 1945; Radinsky, 1964; Romer, The Wasatchian perissodactyl genus Lambdotherium has usually
1966; Schoch, 1983a, 1984) and occasionally with chalicotherioids been regarded as a primitive brontothere (Earle, 1892; Osborn, 1929;
(Wood, 1937; Simpson, 1945). Hooker (1984), however, regarded Simpson, 1945; Hooker, 1989; Prothero and Schoch, 1989a, l 989b),

Menop~

Figure 36.5. Hypothesis ofrelationship for selected North American brontothere genera. Key to characters at nodes: (1) BRONTOTHERIIDAE: bunoselen-
odont molars. (2) Face relatively short and skull posterior to orbit elongated, lateral incision of external nares very deep and broadly bordered by maxilla, all
incisors subcaniniform with a posterior cingulum. (3) Shear more emphasized on ectoloph of molars, presence of facial concavity. (4) DOLICHORHININAE:
presence of suborbital protuberance, gracile canines? (5) BRONTOTHERIINAE: frontonasal hornlike swelling or prominence over facial concavity, trian-
gular projection of frontal overlapping nasal, cranial vertex relatively wide. (6) DIPLACODONTS SEN SU LATO: frontonasal swelling developed into a true
horn, facial region of skull greatly abbreviated, external nares widened vertically, cranial vertex widened with complete loss of sagittal crest, lateral cranial pro-
file distinctly concave. (7) "EUBRONTOTHERES": cranial vertex extremely broad, four upper incisors that are reduced and globular in shape, diastema lost,
medifossette present, hypocone well developed and consistently present on M3. (8) Four (or fewer) lower incisors. (9) Palaeosyops: strongly brachycephalic
skull, pronounced convexity or dome in region of frontoparietal boundary, massive zygomatic arches that are strongly curved, strongly curved nasals that are
tapered distally, extreme reduction or loss of upper diastema. (10) Sphenocoelus: hyperdolichocephalic skull. (11) Telmatherium: prominent pitlike fossa in
center of sagittal crest. (12) "Diplacodon" progressum: relatively large bulbous horns that are roughly circular in cross section. (13) Duchesneodus: prominent
convexity or dome in region of frontoparietal boundary, moderately expanded zygomatic arches that are apparently wing shaped in males. (14) Brontops:
massive zygomatic arches that are swelled posteriorly, relatively large horns that are directed anteriorly and are circular to elliptical in cross section.
(15) Megacerops: broadly expanded zygomatic arches with a distinct wing shape, long to very long horns that are directed laterally or forward and are circular
to elliptical _in cross section. (16) Menops: moderate-sized horns that are trihedral in cross section. The genera Eotitanops and Mesatirhinus cannot at present
be defined in terms of unique derived characters because they are primitive within higher brontothere groupings. Valid North American genera not included
in the cladogram: Metarhinus, Protitanotherium, Eotitanotherium (= Diplacodon?), Notiotitanops, Protitanops, Metatelmatherium, and Sthenodectes.
 Brontotheriidae 529

but unpublished research has suggested that Lambdotherium is not Because most of my metric data are restricted to skulls and upper
a member of this family (Wallace, 1980; Mader, 1991). The exact dentitions, I have provided as an indication of size the average length
phylogenetic position of Lambdotherium relative to other perisso- of the M2 rather than the average length of m2. Because the species-
dactyls remains unclear, however, although the genus bears some level taxonomy is still in the process of revision, locality information
similarity to equoids. Lambdotherium probably includes only a sin- is provided for genera only.
gle species, L. popoagicum (Bonillas, 1936), and is known from lo- The locality numbers listed for each genus refer to the list of
calities SB22B, CP20D, CP20E, CP25G, CP25H, CP25I, CP27C, unified localities in Appendix I. Question marks in front of the
CP27D, CP28B, and NP14. locality (e.g., ?CPlOl) mean the taxon is questionably known from
that locality. The acronyms for museum collections are listed in
Appendix III.
INFRAFAMILY
In the present chapter, four brontothere subfamilies are recognized as EOTITANOPINAE OSBORN, 1914
valid: the Eotitanopinae, the Palaeosyopinae, the Dolichorhininae,
and the Brontotheriinae (= Telmatheriinae, sensu Mader, 1989). Al- Characteristics: Not distinguished from the type genus, Eotitanops.
though I had previously refrained from assigning subfamily names
Eotitanops Osborn, 1907
to Eotitanops and Palaeosyops (Mader, 1989), I now recognize that
Type species: Eotitanops borealis (Cope, 1880).
it is appropriate to do so. Eotitanops is the sole member of the Eoti-
Type specimen: AMNH 4892, a right maxilla with complete
tanopinae Osborn, 1914, and Palaeosyops is the only member of
P4 to Ml and fragmentary M2 to M3.
the Palaeosyopinae Steinmann and DOderlein, 1890. Because the
Characteristics: Small-sized brontothere with a well-
subfamilies Eotitanopinae and Palaeosyopinae each consist of only
developed upper canine; long upper and lower diastema;
a single genus, the diagnosis of these two subfamilies does not differ
unmolarized premolars; relatively large paraconules; and
from that of its member genus.
hypocone or pseudohypocone variably present on M3.
The subfamily name Brontotheriinae as used here is a senior syn-
Average length ofM2: 14.0-20.5 mm.
onym ofTelmatheriinae sensu Mader, 1989. The name Telmatheri-
Included species: E. borealis (= E. brownianum [Cope,
inae was established by Osborn in 1914; the name Brontotheriinae
1881],E.princepsOsborn, 1913, ?E. majorOsborn, 1913);
was established by Marsh in 1873 through the Principle of Coordi-
E. gregoryi Osborn, 1913 (=E. minimus Osborn, 1919).
nation (Article 36, International Code of Zoological Nomenclature,
(Wallace, 1980, unpublished manuscript, has expressed the
Ride et al., 1985). Diplacodonts sensu Jato and eubrontotheres are
opinion that E. minimus is distinct from E. gregoryi and
valid subgroups of the Brontotheriinae.
belongs to a new genus distinct from Eotitanops).
The phylogenetic relationship of Sthenodectes and Metatelmath-
Eotitanops is known from localities CC50, SB22C, CP20E,
erium to other brontothere genera is uncertain, and these taxa are
CP25H, ?CP25I, CP27C, CP27E.
classified here as Brontotheriidae incertae sedis.

PALAEOSYOPINAE STEINMANN AND

INCLUDED NORTH AMERICAN GENERA IN THE DODERLEIN, 1890
FAMILY BRONTOTHERIIDAE Characteristics: Not distinguished from the type genus, Palaeosyops.

Note: Generic names are those recognized as valid by Mader, 1989 Palaeosyops Leidy, 1870b (synonyms: Limnohyus Marsh,
or 1991. Included species ofpre-Chadronian brontotheres are those 1872; Limnohyops Marsh, 1890; Eometarhinus
recognized as valid by Mader, 1991. Because the present chapter Osborn, 1919)
is the first published summary of the species-level taxonomy pre- Type species: Palaeosyops paludosus Leidy, 1870b.
sented in an otherwise unpublished dissertation (Mader, 1991), all Type specimen (lectotype): USNM 759, an isolated m2.
junior synonyms for valid species are listed. Justifications for these Characteristics: Medium-sized brontothere with six upper
synonymies can be found in the dissertation. and lower incisors; large canines; very small or no upper
With regard to the Chadronian brontotheres, species of Mega- diastema and a moderate lower diastema (mostly between
cerops are taken from Mader and Alexander, 1995, and species of p 1 and p2); unmolarized premolars; large paraconules on
Brontops andMenops are from Osborn, 1929 (Menops = Menodus, the molars; hypocone or pseudohypocone variably present
sensu Osborn). Osborn severely oversplitthese two Chadronian gen- on M3; strongly brachycephalic skull; robust zygomatic
era, but the recognition of synonymous species will require much arches that are sharply curved; sharply curved nasals that
further work. Both Brontops and Menops have junior synonyms taper distally; and a low convexity or dome in the region
(Diploclonus andAllops, respectively) that were recognized as valid of the frontoparietal suture.
genera by Osborn (1929). Because I have_ not attempted to deter- Average length M2: 27.0-41.5 mm.
mine valid species of Brontops andMenops, I have listed as included Included species: P. paludosus (=P. major Leidy, 1871, P.
species for these two genera all species that Osborn included under montanus [Marsh, 1871], P. humilis Leidy, 1872, P. laevi-
their junior synonyms. dens [Cope, 1873d], P. longirostris Earle, 1892, P. priscus
 Bryn J. Mader 530

[Osborn, 1908], P. matthewi [Osborn, 1908], P. monoconus Characteristics: Moderately large-sized brontothere with six
[Osborn, 1908]); P.fontinalis (Cope, 1873d) (=P. huerfa- upper and lower incisors; large- to medium-sized canines;
nensis [Osborn, 1919]); P. robustus (Marsh, 1872) (= P. long upper and lower diastema (lower diastemata occur be-
diaconus Cope, l 873a, P. leidyi Osborn, 1908, P. grangeri tween pl and p2 and between pl and the canine); unmolar-
Osborn, 1908, P. copei Osborn, 1908). ized premolars; small paraconules and metaconules vari-
Palaeosyops is known from localities SB22C, ?SB26A, ably present on the molars; hypocone or pseudohypocone
CP20F, ?CP25I, CP31C, CP34A, CP34B, CP34C, CP34D, often present on M3; hyperdolichocephalic skull; subor-
CP36A, CP38B. bital protuberance well-developed; nasals broad and flared
distally; small angular horn at frontonasal boundary over
orbit in most (if not all) species; and in most species, a
DOLICHORHININAE, RIGGS, 1912
wide, rounded cranial vertex with no sagittal crest.
Characteristics: Medium- to moderately large-sized brontotheres Average length ofM2: 33.5--48.0 mm.
with a suborbital.protuberance. The canine is usually somewhat Included species: S. uintensis (= S. blairi [Cook, 1926], S.
reduced and the molars tend to be more elongated than in other bridgeri (Cook, 1926), ?S. harundivorax [Cook, 1926]);
brontothere genera. S. intermedius (Osborn, 1908) (=S. heterodon [Douglass,
1909], S. fluminalis [Riggs, 1912], S. superior [Riggs,
Mesatirhinus Osborn, 1908 1912]); S. hyognathus (Osborn, 1890) (= S. cornutum
Type species: Mesatirhinus megarhinus (Earle, 1891). [Osborn, 1895], S. longiceps [Douglass, 1909]); S. angusti-
Type specimen: YPM-PU 10008, a partial skull. dens (Granger and Gregory, 1943).
Characteristics: Medium-sized brontothere with six upper in- Sphenocoelus is known from localities CP6A, CP38C,
cisors; reduced canine; upper diastema usually short but CP65, and from Asia.
occasionally long; unmolarized premolars; small paracon-
ules and metaconules variably present on the molars; no
hypocone on M3; well-developed suborbital protuberance; BRONTOTHERIINAE MARSH, 1873
and nasals that are moderately flared distally. Characteristics: Medium- to very large-sized brontotheres with a
Average length of M2: 28.5-33.5 mm. distinct frontonasal prominence or horn over the facial concavity,
Included species: M. megarhinus ( = M. petersoni Osborn,
a triangular projection of the frontal overlapping the nasal, and a
1908) only, known from localities CP34D, CP38A, CP38B.
widened cranial vertex.
(M. junius [Leidy, 1872] is regarded here as a nomen
dubium, but is a potential senior synonym of M. megarhi- Telmatherium Marsh, 1872 (synonyms: Leurocephalus
nus.) Osborn, Scott, and Speir, 1878; "Telmatotherium" Marsh,
1880; Manteoceras Hatcher, 1895)
Metarhinus Osborn, 1908 (synonyms Rhadinorhinus Riggs, Type species: Telmatherium validus Marsh, 1872.
1912; Heterotitanops Peterson, 1914b) Type specimen (lectotype): YPM 11120, a partial palate with
Type species: Metarhinusfluviatilis Osborn, 1908. most of the upper dentition and fragments of the zygomatic
Type specimen: AMNH 1500, a partial skull lacking the arches, nasals, and occipital region.
nasals. Characteristics: Medium- to moderately large-sized bron-
Characteristics: Medium-sized brontothere with six upper in- tothere with six upper and lower incisors; large canines;
cisors; reduced canine; often no upper diastema; unmo- moderate to long upper diastema; unmolarized premolars;
larized premolars; hypocone occasionally present on M3; paraconules and metaconules usually small and variably
small to large suborbital protuberance; prominent orbits; present on molars; hypocone or pseudohypocone rarely
very deep lateral nasal incision; and nasals that are either present on M3; small hornlike swelling or prominence de-
strongly tapered distally or broadly flared distally giving veloped on frontonasal boundary over facial concavity; and
them a distinct spoon shape. a deep pitlike fossa in the middle of the sagittal crest near
Average length ofM2: 30.0-38.0 mm. the back of the skull.
Included species: M. fluviatilis (= M. riparius [Riggs, 1912]); Average length ofM2: 35.0--43.5 mm.
M. diploconus(Osborn, 1895) (= M. abbotti [Riggs, 1912], Included species: T. validus (= T. cultridens [Osborn,
?M. pater Stock, 1937). Scott, and Speir, 1878], T. manteoceras [Hay, 1902],
Metarhinus is known from localities CC4, CP6A, T. washakiensis [Osborn, 1908]) only, known from locali-
?CP36B, CP38C. ties ?SB26A, CP34D, CP38B.

Sphenocoelus Osborn, 1895 (synonyms: Dolichorhinus Protitanotherium Hatcher, 1895

Hatcher, 1895; Tanyorhinus Cook, 1926; Dolichorhinoides Type species: Protitanotherium emarginatum (Hatcher,
Granger and Gregory, 1943) 1895).
Type species: Sphenocoelus Osborn, 1895. Type specimen: YPM-PU 11242, the anterior part of a skull
Type specimen: AMNH 1501, the posterior part of a skull. with lower jaws.
 Brontotheriidae 531

Characteristics: Large-sized brontothere with six upper and Type specimen: USNM 16646, basal part of a skull with
lower incisors; large canines; long upper and lower di- dentition and portions of both rami.
astema; short, broad nasals; and short but prominent fron- Characteristics: Large-sized brontothere with four upper in-
tonasal horns that are laterally directed and elliptical in cisors that are reduced and globular in shape; reduced ca-
cross section. nine; long upper diastema; single lingual cusp on Pl and
Average length M2: 52.0-56.0 mm (based on unpublished two poorly separated lingual cusps on P2 to P4; no para-
specimens that are uncertainly assigned to this genus). conules or metaconules on M2 and M3 of the type (and
Included species: P. emarginatum only. only) specimen (Ml is too worn to tell if paraconules or
Protitanotherium is known from locality CP6B (type lo- metaconules were present); and a rudimentary hypocone
cality). Specimens that might be referable to Protitanoth- onM3.
erium are known from localities SB43A, SB43B, CP6A, Average length M2: 48.5 mm (based on the type and only
CP37B. specimen).
Included species: N. mississippiensis only, known from lo-
"Diplacodon" progressum cality EM2 only.
Type species: "Diplacodon" progressum Peterson, 1934.
Type specimen: CM l 1879A, a skull and lower jaw.
Characteristics: Large-sized brontothere with six upper in- Protitanops Stock, 1936
cisors; reduced canine; long upper diastema; one lingual Type species: Protitanops curryi Stock, 1936.
cusp on Pl, one or two lingual cusps on P2, and two poorly Type specimen: LACM (CIT) 1854, a skull and lower jaws.
separated lingual cusps on P3 and P4; paraconules and Characteristics: Large-sized brontothere with four upper in-
metaconules occasionally present on molars; hypocone cisors that are reduced and globular in shape; reduced
or pseudohypocone occasionally present on M3; robust canine; long upper diastema; single lingual cusp on Pl
nasals that are rounded distally and curved downward; rel- and two lingual cusps on P2 to P4; paraconules and meta-
atively large, bulbous horns that are roughly circular in conules lacking on M2 and M3 in the type (and only) spec-
cross section; and a widened cranial vertex with no sagit- imen (M 1 is too worn to tell if paraconules or metaconules
tal crest. were present); hypocone on M3; zygomatic arches that are
Average length of M2: 50.0-56.0 mm. moderately wing shaped; robust nasals that are rounded
Included species: "D." progressum only, known from locali- distally and curve downward; relatively large horns that
ties ?CP6A, CP6B. are roughly elliptical in cross section; and a widened cra-
nial vertex with no sagittal crest.
Comme~ts: This taxon is generically distinct from Dipla- Average length of M2: 61.0 mm (based on the type and only
codon Marsh (1875), which may be a nomen dubium or specimen).
synonymous with Eotitanotherium Peterson. A new gen- Included species: P. curryi only, known from locality NB2
eric name will be proposed for it in the near future. only.

Eotitanotherium Peterson, 1914c (synonym: Diploceras Duchesneodus Lucas and Schoch, 1982
osborni Peterson, 1914a) Type species: Duchesneodus uintensis (Peterson, 1931 ).
Type species: Eotitanotherium osborni (Peterson, 1914a). Type specimen: CM 11809, a lower jaw.
Type specimen: CM 2859, the anterior part of a skull with Characteristics: Large-sized brontothere with four upper and
lower jaws, atlas, axis, scapula, pelvis, and various foot six lower incisors that are reduced and globular in shape;
bones. reduced canines; no diastema; first premolars occasionally
Characteristics: Large-sized brontothere with six upper in- lacking; single lingual cusp on Pl and two lingual cusps
cisors; reduced canine; long upper and lower diastema; on P2 to P4; hypocone consistently present on M3; zygo-
one lingual cusp on Pl, one or two lingual cusps on P2, matic arches moderately expanded posteriorly and devel-
and two well-separated lingual cusps on P3 and P4; no oping a wing shape in some individuals (probably males);
hypocone on M3 in any of the three specimens known; small frontonasal horns that are roughly circular in cross
gracile nasals that are distally rounded and downturned; section; small to large (sexually dimorphic) convexity or
and short, angular horns that are elliptical in cross section. dome in the region of the frontoparietal border; and a broad,
Average length of M2: 57.5-59.5 mm (based on a single "saddle-shaped" cranial vertex.
specimen). Average length M2: 48.0-66.0 mm.
Included species: E. osborni only, known from locality CP6B Included species: D. uintensis only, known from localities
only. SB25B, SB25C, ?SB27, ?SB43C, SB44B, ?CC9B, CP5B,
CP7C, ?CP82, NPlO.
Notiotitanops Gazin and Sullivan, 1942
Type species: Notiotitanops mississippiensis Gazin and Comments: Lucas and Schoch, 1989, recognize D. primitivus
Sullivan, 1942. (Lambe, 1908) as a second valid species.
 Bryn J. Mader 532

Brontops Marsh, 1887 (synonym: Diploclonus Marsh, 1890) BRONTOTHERIIDAE INCERTAE SEDIS
Type species: Brontops robustus Marsh, 1887.
Metatelmatherium Granger and Gregory, 1938
Type specimen: YPM 12048, a skull and skeleton.
Type species: Metatelmatherium cristatum Granger and Gre-
Characteristics: Very large-sized brontothere with four (or
gory, 1938.
fewer) upper and lower incisors that are reduced and glob-
Type specimen: AMNH 26411, a skull and lower jaw.
ular in shape; no diastema; partially molarized premolars;
Characteristics: Large-sized brontothere with six upper in-
hypocone on M3; massive zygomatic arches that are thick-
cisors; large canines; long upper and lower diastema; un-
ened or swelled posteriorly; short, anteriorly directed horns
molarized premolars; paraconules and metaconules not
that are round to elliptical in cross section; and a wide,
present in the few skulls known; pseudohypocone occa-
"saddle-shaped" cranial vertex. sionally present on M3; zygomatic arch often with a promi-
Average length ofM2: 59.0-80.0 mm. nent flange on the ventral surface of the j ugal near to where
Included species: B. robustus; B. amplus (Marsh, 1890); B.
it borders on the squamosal; and lateral incision of the ex-
bicornutus (Osborn, 1902); B. brachycephalus (Osborn,
ternal nares shifted forward so as to lie over the upper
1902); B. dispar Marsh, 1887 (= B. validus Marsh, 1891);
diastema or Pl in an uncrushed specimen.
B. selwynianus (Cope, 1889); B. tyleri (Lull, 1905).
Average length of M2: 42.5-52.0 mm.
Brontops is known from localities CP98, CP41, CP83,
Included species: M. cristatum: M. ultimum (Osborn, 1908)
NPIO.
( = M. uintensis [Douglass, 1909]).
Metatelmatherium is known from localities CP6B,
Menops Marsh, 1887 (synonyms: Allops Marsh, 1887;
CP65, and from Asia.
Menodus, sensu Osborn, 1929)
Type species: Menops varians Marsh, 1887.
Sthenodectes Gregory, 1912
Type specimen: YPM 12060, a skull. Type species: Sthenodectes incisivum (Douglass, 1909).
Characteristics: Very large-sized brontothere with four (or
Type specimen: CM 2398, a skull crushed dorsoventrally.
fewer) upper and lower incisors that are reduced and glob-
Characteristics: Large-sized brontothere with six very large
ular in shape; reduced canine; no diastema; partially mo-
spatulate upper incisors; very long pointed canine; no up-
larized premolars; hypocone on M3; massive zygomatic
per diastema; and unmolarized premolars.
arch that is not thickened or expanded laterally; horns rel-
Average length of M2: 43.5--45.5 mm. (based on two speci-
atively large, laterally directed, and strongly trihedral in
mens only).
cross section; and a wide, "saddle-shaped" cranial vertex.
Included species: S. incisivum only, known from locality
Average length ofM2: 67.5-100.0 mm.
CP6Aonly.
Included species: M. varians; M. crassicornis (Marsh, 1891 );
M. heloceras (Cope, l 873b ); M. marshi (Osborn, 1902); M. In addition to the localities noted here, unidentified brontotheres
serotinus (Marsh, 1887); M. trigonoceras (Cope, 1873c); are known from localities SB41, SB42B. Unidentified paleobron-
M. walcotti (Osborn, 1916). totheres are known from localities NP14, PN5A. Unidentified prim-
Menops is known from localities CP98, CP83, CP68B, itive diplacodonts sensu lato are known from localities SB44A,
NPIO. CP7 A,? CP37 A. Unidentified eubrontotheres are known from lo-
calities SB26B?, SB43E, CP7A, CP39IIC, CP42, PN5B.
Megacerops Leidy, 1870a (synonyms: "Megaceratops" Cope,
1873b; Brontotherium Marsh, 1873; Titanops Marsh, 1887)
Type species: Megacerops coloradensis Leidy, 1870a. BIOLOGY AND EVOLUTIONARY PATTERNS
Type specimen: ANSP 13362, fragmentary horns and nasals.
Characteristics: Large- to very large-sized brontothere with Brontotheres, with their relatively weak dentitions, were obligatory
four (or fewer) upper and lower incisors that are reduced browsers. Although most brontotheres probably browsed on twigs
and globular in shape; reduced canine; no diastema; par- and leaves, forms with heavy canine tusks and large caniniform in-
tially molarized premolars; hypocone on M3; zygomatic cisors (such as Palaeosyops and Sthenodectes) were probably well
arches that are broadly expanded and wing shaped; horns suited for plucking bulbs, tubers, and roots as well. In eubrontotheres
that are long, directed laterally or forward, and roughly (subfamily Brontotheriinae), the incisors were greatly reduced and
circular to elliptical in cross section; and a wide, "saddle- were probably quite useless in plucking vegetation. As a replace-
shaped" cranial vertex. ment for the incisors, a partially prehensile upper lip appears to have
Average length of M2: 62.5-84.5 mm. evolved (Gregory, 1929, pp. 708-9; Osborn, 1929, p. 795).
Included species: M. coloradensis; M. curtus (Marsh, 1887); As Osborn (1929, p. 269) noted, the brontothere chewing mech-
M. kuwagatarhinus Mader and Alexander, 1995; M. plat- anism was essentially a combination of two functions: crushing and
yceras (Scott and Osborn, 1887). cutting (shearing). Grinding does not appear to have been empha-
Megacerops is known from localities CP98, CP83, sized. Crushing was accomplished by a double pestle and mortar
CP68B, NPIOB, NP291I. arrangement in which the protocone and hypocone of the upper
 Brontotheriidae 533

w
zw L L.BI.
1.8
2.4
E.BI
0 4.5
0 E L•t Hp.
::i L.Hp.
5.2
6.0
Q. LE.Hp.
7.0
E.E.Hp.
8,8
L L.CI.
9.5

E.ct.
11.0
w
zw L.L.Ba.
12.5
0 E.L.Ba.
M
0 14.0

:ii E.Ba.

5.8
L.Hm.

17.5
E.Hm.
18.8
E L.L.Ar.
19.2

E.L.Ar.
23.D

w
zw L.E.Ar.

L
0 27.7
0
CJ E.E.Ar.
::i 29.4
0
Wt.
E 31.9
Or.
33.4
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5
,
L.Un.

w 41.3

z E.Un.
w
0 45.9
0
w L.Br.
M 46.7 -§-
M.Br.

E.Br.
49.0 -~-
.g
-~[
50.4
I I
L.Wa.
"Eubrontotheres"
53.5
E M.Wa.
L..J ..._ _ _ _.. L-1
54.2
E.Wa. DOLICHORHININAE BRONTOTHERllNAE
55.0
E.E.Wa.
55.5
Ck.3 55.6
w Ck.2 55.8 PALAEOSYNOPINAE BRONTOTHERllDAE INCERTAE SEDIS
zw Ck.1
Tl.6
56.0
56.2
EOTITANOPINAE PERISSODACTYLA INCERTAE SEDIS
L Tl.5
Tl.4
57.0
0 Tl.3
57.3
0 Tl.2
w Tl.1
...I
c( To.3
Q. To.2
62.5
To.1
E Pu.3
Pu.2
Pu.1
Pu.o
Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
=
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br. Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To.= Torrejonian, Pu. = Puercan.
Figure 36.6. Temporal ranges of North American brontothere genera.
 Bryn J. Mader 534

cheek teeth and the various conids of the lower cheek teeth pressed this species. In most species of Megacerops and Brontops, however,
into fossae on the opposing dentition. Shear was achieved by the the side of the body probably remained the primary focus of attack,
action of reversed double-bladed crescents, in which the W-shaped a behavior that is suggested by one skeleton of Brontops (AMNH
ectoloph of the upper molars slid past the M-shaped loph of the 518) in which one of the ribs had been broken and healed during
lower dentition. the life of the animal. Both Brontops and Megacerops have broadly
Brontotheres were evidently contentious animals, and most mem- expanded zygomatic arches, which may have furnished additional
bers of the family possessed horns or cranial domes that were prob- protection to the head while grappling with an opponent (Osborn,
ably used in intraspecific combat. The horn morphology was quite 1929, p. 792).
variable, and it is likely that different horn shapes were employed in Brontotheres such as Duchesneodus, Menops, and most of the
different ways. Although somewhat speculative, it may be possible more primitive homed genera were equipped with relatively short,
to infer from living homed mammals some of the general charac- laterally directed horns and, as Osborn (1929, p. 792) pointed out,
teristics of brontothere intraspecific aggression. blows were probably dealt with a sideward thrust of the head. It
Bovids with short horns (such as the mountain goat, Oreamnos) seems likely, therefore, that lateral attacks similar to those of Ore-
typically engage in contests in which the participants circle one amnos were the usual form of aggression in these animals. Because
another while using their horns to deliver lateral blows to the trunk the majority of homed brontothere genera fall into this morpholog-
and haunches of the opponent (Geist, 1966). According to Geist ical category, Schoch's (1983b) contention that broadside intimi-
(1966), this horn morphology and this form of combat are primitive dation displays were the dominant form of intraspecific combat in
for homed mammals in general. brontotheres is probably correct.
The brontothere genera Telmatherium and Sphenocoelus are both Given the wide variety of horn sizes and shapes, it is likely that the
roughly analogous to this grade of horn development, having short, horns also served as intraspecific displays and as a means of species
laterally directed prominences anterior to or just above the orbit. recognition. Although not sufficiently studied, it seems that many
As Geist pointed out, the horns of these brontotheres were probably Chadronian homed brontothere species were contemporaneous and
capable of delivering blows upward and to the side only, suggesting thus likely to benefit from a visual means of species identification.
that flank attacks, similar to those of Oreamnos, were the usual
method of fighting. Unlike the horns of the mountain goat, however,
the horns of these brontotheres were blunt and were not covered with
ACKNOWLEDGMENTS
a keratinized sheath. It is unlikely, therefore, that they were capable
of inflicting serious injury. I would like to thank Margery C. Coombs and Spencer Lucas for
The next grade of horn development envisioned by Geist is found their thoughtful comments on an earlier version of this chapter. The
in the bovids Bos and Bison, which have relatively large horns that systematic conclusions published here are the result of research
are widely spread. These animals fight head to head, locking their
partially funded by grants from the Theodore Roosevelt Memorial
horns with those of the opponent. In addition to their function as Fund of the American Museum of Natural History and by the Rose
offensive weapons, the horns of these bovids thus also serve to hold
M. Louer Fellowship Fund of the Field Museum of Natural History.
and control the weapons of the opponent, preventing theiruse against
The financial support of these institutions is greatly appreciated.
the otherwise unprotected body. The primary focus of the attack
remains the opponent's flank, however, and this will be struck with
the horns if the opportunity arises. According to Geist, the antlers
of cervids perform a similar function, although the wrestling aspect REFERENCES
of the contest is more greatly emphasized.
The brontotheres Brontops and Megacerops appear to have horns Bonillas, Y. (1936). The dentition of Lambdotherium. Journal of Mammal-
ogy, 17, 139-42.
that are roughly analogous to those of Bos, Bison, and the cervids.
Cook, H. J. (1926). New Eocene titanotheres from Moffat County, Col-
In Brontops, the horns are relatively short and anteriorly directed, orado. Proceedings of the Colorado Museum of Natural History, 6,
apparently adapted for use in frontal attacks. Primitive species of 8-12.
Megacerops have laterally directed horns, but, because of their great Cope, E. D. (1873a). On some Eocene mammals obtained by Hayden's
length, they could probably be employed frontally by lowering the geological survey of 1872. Paleontological Bulletin, 12, 1-6 (Inde-
pendent Publication).
head. In Megacerops platyceras and M. kuwagatarhinus, in which
(1873b). Second notice of extinct Vertebrata from the Tertiary of the
the horns are long and sweep far forward, frontal aggression seems plains. Paleontological Bulletin, 15, 1-6 (Independent Publication).
especially likely. The horns of both Brontops and Megacerops seem (1873c). Synopsis of new Vertebrata from the Tertiary of Colorado, ob-
well suited for locking with the horns of an opponent and for pre- tained during the summer of 1873, Washington, October, 1873. 1-19.
venting their use against the unprotected flank. Because the dis- (Washington, D.C.: U.S. Government Printing Office).
(1873d). On the new perissodactyls from the Bridger Eocene. Paleon-
tally bifurcated horns of M. kuwagatarhinus are roughly similar to
tological Bulletin 11, Proceedings of the American Philosophical
the branching pattern of cervid antlers, they seem particularly well Society, 13, 35-6.
adapted as a locking mechanism. It is likely, therefore, that wrestling, (1880). The badlands of the Wind River and their fauna. American Nat-
rather than striking with the horns, was emphasized in the contests of uralist, 14, 745-8.
 Brontotheriidae 535

(1881). On the Vertebrata of the Wind River Eocene beds of Wyoming. Perissodactyls, eds. D. R. Prothero & R. M. Schoch, pp. 458-84.
Bulletin United States Geological Survey of the Territories, 6, Oxford: Clarendon.
183-202. (1991). The Systematics and Phylogeny of North American Eocene
( 1889). Vertebrata of the Swift Current River, III. American Naturalist, Brontotheres (Mammalia, Perissodactyla). Unpublished Ph.D. The-
23, 628-29. sis. Amherst: University of Massachusetts.
Douglass, E. (1909). Preliminary descriptions of some new titanotheres from Mader, B. J., & Alexander, J. P. (1995). Megacerops kuwagatarhinus n.
the Uinta deposits. Annals of the Carnegie Museum, 6, 304--13. sp., an unusual brontothere (Mammalia, Perissodactyla) with distally
Earle, C. (1891 ). On a new species of Palaeosyops, Palaeosyops megarhinus forked horns. Journal of Paleontology, 69, 581-7.
sp. nov. American Naturalist, 25, 45-7. Marsh, 0. C. (1871). Notice of some new fossil mammals and birds from
(1892). A memoir upon the genus Palaeosyops Leidy and its allies. the Tertiary formation of the west. American Journal of Science, 3d
Journal of the Academy of Natural Sciences of Philadelphia, 2d ser., ser., 2, 120-7.
9, 267-388. (1872). Preliminary description of new Tertiary mammals, Part I. Amer-
Gazin, C. L., & Sullivan, J. M. ( 1942). A new titanothere from the Eocene of ican Journal of Science, 3d ser., 4, 122-30.
Mississippi, with notes on the correlation between the marine Eocene (1873). Notice of new Tertiary mammals. American Journal of Science,
of the Gulf Coastal Plain and continental Eocene of the Rocky Moun- 3d ser., 5, 407-11, 485-8.
tain region. Smithsonian Miscellaneous Collections, 101, 1-13. (1875). Notice of new Tertiary mammals, N. American Journal of Sci-
Geist, V. (1966). The evolution of horn-like organs. Behaviour, 27, 175-214. ence, 3d ser., 9, 239-50.
Granger, W., & Gregory, W. K. (1938). Addendum. A new titanothere genus (1880). List of genera established by Professor 0. C. Marsh, 1862-1879,
from the upper Eocene of Mongolia and North America. In Fossil 1-12 (privately published).
Mammals from Burma in the American Museum of Natural History, (1887). Notice of new fossil mammals. American Journal of Science, 3d
ed. E. H. Colbert. Bulletin of the American Museum of Natural His- ser., 34, 323-31.
tory, 74, 435-366. ( 1890). Notice of new Tertiary mammals. American Journal of Science,
(1943). A revision of the Mongolian titanotheres. Bulletin of the Ameri- 3d ser., 39, 523-5.
can Museum of Natural History, 80, 349-89. (1891). Notice of new vertebrate fossils. American Journal of Science,
Gregory, W. K. (1912). Note on the upper Eocene titanotheroid Telmath- 42, 265-9.
erium? incisivum Douglass, from the Uinta Basin. Science, new ser., Osborn, H. F. (1890). The Mammalia of the Uinta formation, Part 3: the
35, p. 545. Perissodactyla. Transactions of the American Philosophical Society,
( 1929). The muscular anatomy and the restoration of the titanotheres. new ser., 16, 461-572.
In The Titanotheres of Ancient Wyoming, Dakota, and Nebraska, (1895). Fossil mammals of the Uinta Basin, Expedition of 1894. Bulletin
ed. H.F. Osborn. United States Geological Survey Monographs, 55, of the American Museum of Natural History, 7, 71-105.
703-25. (1902). The four phyla of Oligocene titanotheres. Bulletin of the Ameri-
Hatcher, J. B. (1893). The Titanotherium beds. American Naturalist, 27, can Museum of Natural History, 16, 91-109.
204--21. (1907). Tertiary mammal horizons of North America. Bulletin of the
( 1895). On a new species of Diplacodon, with a discussion of the relations American Museum of Natural History, 23, 237-53.
of that genus to Telmatotherium. American Naturalist, 29, 1084--90. (1908). New or little known titanotheres from the Eocene and Oligocene.
Hay, 0. P. (1902). Bibliography and catalogue of the fossil vertebrata of Bulletin of the American Museum of Natural History, 24, 599-617.
North America. Bulletin of the United States Geological Survey, (1913). Lower Eocene titanotheres. GeneraLambdotherium, Eotitanops.
179, 1-868. Bulletin of the American Museum of Natural History, 32, 407-15.
Hooker, J. J. (1984). A primitive ceratomorph (Perissodactyla, Mammalia) (1914). Recent results in the phylogeny of the titanotheres. Bulletin of
from the early Tertiary of Europe. Zoological Journal of the Linnean the Geological Society of America, 25, 403-5.
Society, 82, 229-44. (1916). Two new Oligocene titanotheres. Bulletin of the American Mu-
(1989). Character polarities in early perissodactyls and their significance seum of Natural History, 35, 721-3.
for Hyracotherium and infraordinal relationships. In The Evolution (1919). New titanotheres from the Huerfano. Bulletin of the American
of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 79-101. Museum of Natural History, 41, 557-69.
Oxford: Clarendon. (1929). The titanotheres of ancient Wyoming, Dakota, and Nebraska (2
Lambe, L. M. (1908). The Vertebrata of the Oligocene of the Cypress Hills, vols.). United States Geological Survey Monographs, 55, 1-953.
Saskatchewan. Contributions to Canadian Paleontology, 3, 1-64. Osborn, H. F., Scott, W. B., & Speir, F. (1878). Paleontological report
Leidy, J. (1870a). Remarks onMegacerops coloradensis. Proceedings of the of the Princeton scientific expedition of 1887. Contributions from
Academy of Natural Sciences of Philadelphia, 22, 1-2. the Museum of Geology and Archaeology of Princeton College, 1,
(1870b). On fossils from Church Buttes, Wyoming Territory. Proceed- 1-106.
ings ofthe Academy ofNatural Sciences ofPhiladelphia, 22, 113-14. Peterson, 0. A. (1914a). A new titanothere from the Uinta Eocene. Annals
(1871). Remarks on fossil vertebrates from Wyoming. Proceedings of of the Carnegie Museum, 9, 29-52.
the Academy of Natural Sciences of Philadelphia, 23, 228-9. (1914b). A small titanothere from the lower Uinta beds. Annals of the
(1872). On fossils from Wyoming. Proceedings of the Academy of Nat- Carnegie Museum, 9, 53-7.
ural Sciences of Philadelphia, 24, 277. (1914c). Correction of generic name. Annals of the Carnegie Museum,
Lucas, S. G., & Schoch, R. M. (1982). Duchesneodus, a new name for some 9, 220.
titanotheres (Perissodactyla, Brontotheriidae) from the late Eocene (1924). Osteology of Dolichorhinus longiceps Douglass, with a review
of western North America. Journal of Paleontology, 56, 1018-23. of the species of Dolichorhinus in the order of their publication.
(1989). Taxonomy of Duchesneodus (Brontotheriidae) from the late Memoirs of the Carnegie Museum, 9, 405-73.
Eocene of North America. In The Evolution of Perissodactyls, eds. (1931 ). A new species of the genus Teleodus from the upper Uinta of
D.R. Prothero & R. M. Schoch, pp. 490--503. Oxford: Clarendon. northeastern Utah. Annals of the Carnegie Museum, 20, 307-12.
Lull, R. S. (1905). Megacerops tyleri, a new species of titanothere from the (1934). New titanotheres from the Uinta Eocene in Utah. Annals of the
Bad Lands of South Dakota. Journal of Geology, 13, 443-56. Carnegie Museum, 22, 351-61.
Mader, B. J. (1989). The Brontotheriidae: a systematic revision and pre- Prothero, D. R., & Schoch, R. M. (1989a). Origin and evolution of the
liminary phylogeny of North American genera. In The Evolution of Perissodactyla: summary and synthesis. In The Evolution of Peris-
 Bryn J. Mader 536

sodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 504-29. Oxford: and Eastern Europe. In Abstracts of Papers and Posters, eds. W. A.
Clarendon. Fuller, M. T. Nietfeld, & M.A. Harris, Abstract 0558. Alberta: Fourth
(l 989b). Classification of the Perissodactyla. In The Evolution of Peris- International Theriological Congress.
sodactyls, eds. D. R. Prothero & R. M. Schoch, pp. 530-37. Oxford: Scott, W. B. (1941). The mammalian fauna of the White River Oligocene,
Clarendon. Part 5: Perissodactyla. Transactions of the American Philosophical
Radinsky, L. B. (1964). Paleomoropus, a new early Eocene chalicothere Society, 28, 747-980.
(Mammalia, Perissodactyla), and a revision of Eocene chalicotheres. Scott, W. B., & Osborn, H. F. (1887). Preliminary account of the fossil
American Museum Novitates, 2179, 1-28. mammals from the White River formation contained in the Museum
Ride, W. D. L., Sabrosky, C. W., Bernardi, G., & Melville, R. V. (1985). of Comparative Zoology. Bulletin of the Museum of Comparative
International Code of Zoological Nomenclature (3rd ed.). London: Zoology, 13, 151-71.
International Trust for Zoological Nomenclature. Simpson, G. G. (1945). The principles of classification and a classification
Riggs, E. S. (1912). New or little known titanotheres from the lower Uin- of mammals. Bulletin of the American Museum of Natural History,
tah formations. Field Museum of Natural History Publication 159, 85, 1-350.
Geological Series, 4, 17-41. Steinmann, G., & Doderlein, L. (1890). Elemente der Paliiontologie. Verlag
Romer, A. S. (1966) . .Vertebrate Paleontology (3rd ed.). Chicago: University von Wilhelm Engelman, Leipzig, 337-848.
of Chicago Press. Stock, C. (1936). Titanotheres from the Titus Canyon Formation, Cal-
Schoch, R. M. (1983a). Relationships of the earliest perissodactyls (Mam- ifornia. Proceedings of the National Academy of Sciences, 229,
malia, Eutheria). Geological Society ofAmerica, Abstracts with Pro- 656--61.
grams, 15, 144. Stock, C. (1937). An Eocene titanothere from San Diego County, California,
(1983b). The function of the horns oftitanotheres (Mammalia, Perisso- with remarks on the age of the Poway conglomerate. Proceedings of
dactyla). Evolutionary Theory, 6, 274. the National Academy of Sciences, 23, 48-53.
(1984). Two unusual specimens of Helaletes in the Yale Peabody Mu- Wallace, S. M. (1980). A Revision of North American Early Eocene
seum collections, and some comments on the ancestry of the Tapiri- Brontotheriidae (Mammalia, Perissodactyla). Unpublished master's
dae (Perissodactyla, Mammalia). Postilla, 193, 1-20. thesis. Boulder: University of Colorado.
Schoch, R. M., & Lucas, S. G. (1985). The Brontotheriidae, a group of Wood, H. E. (1937). Perissodactyl suborders. Journal of Mammalogy, 18,
Eocene and Oligocene perissodactyls from North America, Asia, 106.
37 Equidae

BRUCEJ.MACFADDEN

INTRODUCTION

Fossil horses are the most commonly encountered medium to large

mammals at many Tertiary fossil sites in North America. This is
particularly true during the late Miocene, when horses reached
their maximum diversity. At that time as many as twelve appar-
ently sympatric species occur at some fossil sites. Horses also had
a widespread paleogeographic distribution that resulted in at least
three separate dispersal events into the Old World during the
Miocene. Nevertheless, the principal center of diversification of the
Equidae was in North America. Because of their long temporal du-
ration and relatively rapid evolution, fossil horses are biochronolog-
ically useful in t>ubdividing terrestrial sediments of Cenozoic age,
especially in the rich terrestrial sequence in North America.
Based on collections from the western U.S. territories, nineteenth-
century paleontologists such as 0. C. Marsh believed that the avail-
able evidence for fossil horses provided a relatively complete
sequence of evolutionary change. This supported the notion of "or-
thogenesis," or directed change to a more perfect form. We know,
however, that evolution does not occur in this manner, and the bushi-
ness of the horse family tree (see later) supports the modem concepts
of experimentation and opportunism. Since Marsh's time, the family
Equidae is frequently cited in textbooks as one of the best exam-
ples of a well-known phylogeny and long-term evolution that can
be derived from the fossil record.

DEFINING FEATURES OF THE FAMILY EQUIDAE Figure 37.1. Reconstruction of the fossil horse Parahippus leonensis from
the middle Miocene Thomas Farm locality, Florida. This illustration depicts
It is not easy to characterize the entire family Equidae for two the hypothesized social structure of a harem band with one stallion, three
mares, and two juveniles that were foaled at different times within a nonsyn-
reasons. First, the earliest member of this family, Hyracotherium,
chronized breeding season. From MacFadden (1992), illustration by Wendy
is very primitive dentally and postcranially, and therefore lacks Zomlefer, and reproduced with permission of Cambridge University Press.
any synapomorphies in these structural systems. Second, there are
so many adaptive morphologies that evolved, particularly in later
CRANIAL
horses, that few character states can be used to describe all mem-
bers of the Equidae. Nevertheless, the following characters, the most Two cranial characters have been recognized as synapomorphies for
important of which are depicted in Figures 37.2-37.4, are prominent the Equidae (MacFadden, 1976). These include the lack of a sepa-
in an understanding of the family. rate foramen ovale and the position of the optic foramen relative to

537
 BruceJ.l'vlacFadden 538

A A

P1

B
B P4

c D

protoloph

Figure 37.3. Dentitions of representative equid genera: A. Upper dentition

Figure 37.2. Skulls of representative equid genera: A. Hyracotherium. B. of Hyracotherium showing the primitive tooth morphology for Equidae.
Mesohippus. C. Merychippus. D. Equus. This greatly simplified view of Note the development of the triangular-shaped premolars and degree of
selected genera illustrates some of the principal characters and trends seen formation of the principal cusps. B. Upper dentition of Mesohippus showing
in fossil Equidae (e.g., increase in preorbital region and increase in the the quadrate, relatively molarized premolars (i.e., P2-P4) and development
depth of the jaw). The rudimentary preorbital depression in Merychippus of transverse lophs characteristic of later, nonhypsodont horses (i.e., mostly
is elaborated in numerous later hipparionines and equines and is then lost subfamily Anchitheriinae ). In general, hypsodont horses can be subdivided
in Equus, which has a smooth preorbital region. Scale bar represents 5 cm. into two morphologies of the upper molars: C. Hipparionines, in which
From MacFadden (1992) and reproduced with permission of Cambridge during middle wear stages the protocone is isolated, and D. Equines, in
University Press. which the protocone is connected to the protoloph as in equines. Scale bar
represents 1 cm.

the ethmoid and anterior lacerate foramina. l'vlembers of the subfam-

ily Hyracotheriinae retain a relatively primitive skull morphology
Homogalax. Also in Hyracotherium the premolars are triangular
with the preorbital and postorbital lengths roughly similar. A major
shape and thus retain the basic tritubercular plan. In the lower cheek
cranial reorganization occurred in the Equidae during the l'vliocene
teeth the ancestral tribosphenic trigonid has a reduced paraconid and
(Radinsky, 1984; Figure 37.2) in which the preorbital region ex-
paracristid (Hooker, 1989; l'vlacFadden, 1992). The more advanced
panded greatly. This seems to have resulted from the greatly modi-
hyracotheriine horses (Orohippus, Haplohippus, andEpihippus) are
fied and enlarged dentition associated with a changeover to feeding
characterized by progressive molarization of the premolars both in
on more abrasive foodstuffs, that is, a specialization for including
the upper and lower teeth.
varied amounts of grass in ancient horses' diets.
In Mesohippus and later horses, this morphology is fixed so that
all upper and lower cheek teeth (P2-l'v13 inclusive) are molariform
(Figure 37.3B ). In parahippines and later hypsodont forms, fossettes
DENTAL
(lakes) form in the upper cheek teeth. Thereafter, one of the principal
Horses are characterized by a wide range of dental morphologies. changes occurs during the l'vliocene when horses become hypsodont,
Hyracotherium is quite primitive relative to Equus, but all Equidae that is, where the depth of the unworn crown exceeds the length of
can be united by the shared presence of the postprotocrista on the the occlusal surface of the tooth. This advanced character state is
P3 (Hooker, 1989). Besides this character, Hyracotherium itself well exhibited in the subfamily Equinae and represented in about
lacks significant synapomorphies and, as such, is the plesion within half of the known genera from North America. Within the Equinae
the Equidae. In the general pattern seen in the upper dentition of two general morphologies are represented (in middle wear and in
primitive hyracothere horses, the principal crests are connected by most, but not all species): (1) Hipparionine, in which the protocone
the proto-, meta-, and ectolophs (Figure 37.3A), but these are less is isolated; and (2) Equine, in which the protocone is connected to
well developed than, for example, in the contemporaneous tapiroid the protoloph (Figure 37.3C, D).
 Equidae 539

A B because we happen to have more nearly complete lines back

to it from later members of the family than from other fami-
lies. There is no particular time at which the Equidae became
a family rather than a genus or species; the whole process is
gradual and we assign the categorical rank after the result is
before us. (p. 345.)

However, recent work has shown that based on basicranial char-

acters, Hyracotherium shares the derived character states of the po-
sitions of the optic foramen and foramen ovale with other horses.
Thus the Equidae from North America does seem to be a natural, or
monophyletic group. Note, however, that recent work on primitive
Hyracotherium from the Old World (Hooker, 1989, 1994) indicates
that some taxa traditionally assigned to this genus are more closely
related to palaeotheres, extinct Old World perissodactyls that are the
closest sister taxon to the Equidae. If so, then the Equidae may need
to be restricted to those taxa within Hyracotherium that are more
III III closely related only to other horses.
Although not the most primitive perissodactyl, Hyracotherium
Figure 37.4. Feet of representative equid genera. All fossil horses (and other (and the tapiroid Homogalax) is very close to the base of the adap-
perissodactyls) have metapodial III as the central weight-bearing digit. A. tive radiation of this order. A suitable primitive sister species for
Hyracotherium, with four functional digits, ll-V in the forelimb (and three
in the hindlimb, II-IV). B. Mesohippus, showing the tridactyl condition, as
the Perissodactyla seems to be close to Radinskya, a perissodactyl-
seen in Anchitheriine and hipparionine (and a few primitive equine) horses. like form from the Paleocene of China (l'vlcKenna et al., 1989).
Fully monodactyl horses occur in most, but not all, taxa within the tribe l'vlost workers support the monophyly of the Perissodactyla, and
Equini. Scale bar represents 1 cm. Modified from MacFadden (1992) and good characters support this hypothesis, for example, the shape
reproduced with permission of Cambridge University Press. of the navicular facet on the astragalus (Radinsky, 1966), mesax-
onic reduction of lateral metapodials, and particular cecal morphol-
POSTCRANIAL ogy (l'vlacFadden, 1988a, 1992). The closest outgroup that could
have given rise to this order seems to be the extinct early Tertiary
As for the order Perissodactyla, all Equidae are mesaxonic, that
phenacodontid condylarths from North America, which share the
is, they have the axis of symmetry passing through the central di-
reduction of the paralophid/paraconid complex in the lower cheek
git (i.e., l'vlP Ill'; Figure 37.4). l'vlembers of the Hyracotheriinae, and
teeth and incipient mesaxonic limb symmetry/metapodial reduction
possibly some species within Mesohippus, have four metacarpals
with perissodactyls (l'vlacFadden, 1976). Note that other hypothe-
and three metatarsals. All other anchitheriine and primitive equine
ses also have been proposed: For example, Prothero, l'vlanning, and
horses have reduced their digits so they have three metacarpals and
Fischer (1988; also see Prothero and Schoch, 1989) have proposed
three metatarsals. Advanced equine horses, that is, most members
that the Embrithopoda (including Radinskya, if indeed this is its
of the tribe Equini, are functionally or actually monodactyl.
proper phylogenetic position) are the outgroup of the Perissodactyla.
Gingerich (1989) proposed the novel hypothesis that perissodactyls
SYSTEMATICS may have arisen from some primitive mammals, as yet unknown, in
Africa.
Within modem taxa, anatomical and biochemical data suggest
SUPRAFAMILY
that rather than uniting Perissodactyla and Artiodactyla as ungulate
The phylogenetic position of Hyracotherium, the common early sister taxa, the closest extant perissodactyl outgroup may be the
Eocene perissodactyl, is central to the question of the hyracoids (Fischer, 1989). l'vlcKenna (1975) united phenacodontids,
relationships of the Equidae to other perissodactyl families, both perissodactyls, and hyracoids in the "mirorder" Phenacodonta. The
extinct and extant. It has been traditionally stated in textbooks that close relationship of the extant taxa within this clade is corroborated
Hyracotherium was the ancestral taxon from which all other peris- by several characters, including, for example, the external position
sodactyls could have been descended. As such, its placement in the of the internal carotid artery (Wible, 1986).
family Equidae (rather than elsewhere within the Perissodactyla)
was somewhat arbitrary. Underscoring this view, Simpson (1953)
notes the following: INFRAFAMILY
The family Equidae is usually divided into three subfamilies (Fig-
In effect, there was no family Equidae when eohippis [sic] ure 37.5): the Hyracotheriinae, Anchitheriinae, and Equinae (e.g.,
lived. The family and all its distinctive developed gradually Simpson, 1945; l'vlacFadden, 1992); the first two of these are para-
as time went on. Eohippus [sic] is referred to the Equidae phyletic. These paraphyletic taxa have traditionally been grouped
 Bruce J. MacFadden 540

Figure 37.5. Interrelationships of the Equidae. Description of the shared-derived characters that corroborate principal nodes 1-6 are presented in text.
Problematic regions of poorly, or unresolved, interrelationships are indicated by boxes and "?."Note that those genera depicted in this figure include all
those currently recognized in North America. However, several others are endemic to the Old World (Sinohippus, Stylohipparion, and at least two invasions
of the Old World hipparion complex) and to South America. From MacFadden (1992) and reproduced with permission of Cambridge University Press.

together because of generally similar suites of characters, or stages that are more closely related to equids than they are to palaeotheres.
of evolution. In contrast, modem phylogenetic interpretations of the Between nodes 1 and 2 (Orohippus-Epihippus), the important
subfamily Equinae indicate it is a monophyletic clade that includes character-state transformations relate principally to the progressive
the principal taxonomic diversification of some fifteen North Amer- molarization of the premolars (e.g., MacFadden, 1976, 1992). Node
ican genera within the Equidae during the last 18 myr (MacFadden 2, representing the "subfamily Anchitheriinae", is defined by all
and Hulbert, 1988; Hulbert and MacFadden, 1991). equid genera (starting withMesohippus) with fully molarized P2/p2s
In addition to the cladistic analysis (Figure 37.5), the generic through P4/p4s as well as almost all of these anchitheriine (and
interrelationships within the Equidae are presented in the phylogeny equine) horses in which the fourth digit (MC V) in the manus is
in Figure 37.6. As with any complex group, some of these genera vestigial or lost. Prothero and Shubin (1989) also list several other
are better studied than others, as will be evident later. These generic synapomorphies to corroborate node 2, including Il-3 with pitted
interrelationships are taken from many different studies; the last crowns, extended premaxilla, longer diastema, and jaw angle pos-
comprehensive overview of these was presented by Stirton (1940), teriorly rounded lacking a notch.
and although a valuable classic, it is a half century out of date. The genera North American genera united at node 3, that is,
Given the scope of this chapter and the diversity within this Kalobatippus (= New World "Anchitherium"), Hypohippus, and
family, it is impossible to describe the cladistic interrelationships Megahippus, are traditionally viewed as primitive browsing, three-
of each individual genus within the Equidae. Instead, the impor- toed horses, lacking any derived characters (e.g., Stirton, 1940).
tant character states I use to unite major groups of equid genera at However, functionally these horses were probably quite specialized
nodes 1-7 in Figure 37.5 are described later (also see MacFadden, for browsing. As represented by node 3, derived characters such as
1992). very rapid increase of both tooth crown surface area (MacFadden,
Node 1 defines the Equidae, and therefore includes all genera 1988b) and body size (MacFadden, 1987) can be used to unite these
within the family (sensu stricto, excluding the Old World Palaeothe- in a monophyletic group, herein called Anchitheriinae s. s. The other
riidae), as described in MacFadden (1992). This taxon is united by taxa included in the subfamily "Anchitheriinae," that is, Mesohip-
several characters, including the shared loss of a separate foramen pus, Miohippus, Archaeohippus, Desmatippus, and Parahippus, are
ovale, position of optic foramen (also see earlier discussion), and plesions or incertae sedis relative to node 3. Within this phylogenetic
presence of a P3 protocrista (Hooker, 1989). Note that based on complex, Archaeohippus is quite interesting because it represents
Hooker's studies of European Hyracotherium (e.g., 1989, 1994), one of the first examples of phyletic dwarfing in late Cenozoic horses
this genus is paraphyletic (see also further discussion later). As (MacFadden, 1987; work in progress).
such, Hyracotherium s. s. is used for those species within this genus As represented by node 4, the monophyletic subfamily Equinae
 Equidae 541

~ PHYLOGENY OF THE EQUIDAE

Myr W~i--~~~~ .......~~~~~~~~~~~~~~~~~~~~-..-~~~~~~
S. AMERICA NORTH AMERICA OLD WORLD
01;::-t---.;;;:::==::::::=:::t=============================:==-::-;-.;;;:;~7"1
15 Equus *
,g * *
a..
5

10

.,
~
15 -~
::;:

20

25

.,
30 .,c:
8
.!2>
0
35

40

45
2!
8 12] Grazing horses
w
50
Gill Browsing horses

Figure 37.6. Phylogeny of the

55 Equidae. From MacFadden (1992)
and reproduced with permission of
Cambridge University Press.

are united by the presence of cement-filled, high-crowned cheek hipparion, Hipparion, Nannippus, and Cormohipparion and were
teeth, degree of development of pli caballin, presence of pli ento- three-toed, principally high-crowned taxa that were very abundant
flexid, moderately deep ectoflexid on P2, and unworn Ml and M2 during the Miocene.
crown heights of greater than about 23-28 mm (Hulbert, 1989a;
MacFadden, 1992). These taxa represent the major adaptive radia-
tion and diversification of fossil horses, imparting the "bushiness" INCLUDED NORTH AMERICAN GENERA
to the phylogeny in Figure 37.6. Within the Equinae are two tribes IN THE FAMILY EQUIDAE
(Figure 37.5). As represented at node 6, the Equini, usually con-
sidered the monodactyl fom1s (although some primitive species of For a group that is as abundant and well known as North Ameri-
this clade have not entirely lost the lateral metapodials), include the can Equidae, the modem systematics for each genus is not always
extant Equus, the North American Miocene genera Protohippus, available. Stirton ( 1940) presented the last balanced, comprehensive
Calippus, Pliohippus, Astrohippus, and Dinohippus, and two gen- overview of all valid genera and included species, although his pa-
era traditionally thought to be South American endemics (Hippidion per was not an exhaustive, monographic revision. Osborn (1918) is
and Onohippidium), which are now also known to have occurred in also a required reference for the Oligocene through Pliocene forms.
North America (MacFadden and Skinner, 1979). The Hipparion- Based on more recent work, the taxonomy of some genera is better
ini, united at node 5 by numerous dental characters, including a known than others. Although the present chapter certainly is not
persistent pli caballin, MC V and IV articulation, and an isolated intended to be a revision, I have attempted to synthesize the work
protocone in the upper molars in most taxa. These include the North since Stirton (1940) using the principal references, which are cited
American genera Pseudhipparion, Merychippus s. s., in part, Neo- later for the interested reader (also see MacFadden, 1992).
 Bruce J. MacFadden 542

As is standard in this volume, molar dimensions are given as a Comments: Hyracotherium was first recognized by
general indicator of size. These include all taxa, and depending on Sir Richard Owen (1840) based on specimens from the
availability, either Ml/M2 and/or ml/m2 lengths are listed (these Spamacian (early Eocene) deposits in the London Basin
yield generally similar dimensions and are highly correlated to one (also see Hooker, 1980). Subsequently, Marsh (1876)
another). The abbreviations and conventions are consistent with the described the often cited genus Eohippus based on mate-
usage throughout this book; in addition, DPOF is dorsal preorbital rials from the western United States. Most workers subse-
fossa. quently considered the two genera to be distinct,
The locality numbers listed for each genus refer to the list of uni- principally because of their diverse geographic distribu-
fied localities in Appendix I. The acronyms for museum collections tion. However, the detailed anatomical studies of Cooper
are listed in Appendix III. (1932) and Simpson (1952) indicated that they were
The locality numbers may be listed in a couple of alternative ways. congeneric. Thus Eohippus, still the familiar vernacular
Parentheses around the locality (e.g., [CPlOI]) mean the taxon in "dawn horse," is now universally regarded as the junior
question at that locality is cited as an "aff." or "cf." the taxon in synonym.
question. Parentheses are usually used for individual species, thus More recent work by Hooker (1994) throws doubt on the
implying the genus is firmly known from the locality, but the actual identity of some of the North American species of Hyra-
species identification may be questionable. Question marks in front cotherium as truly belonging to this genus, or even be-
of the locality (e.g., ?CPlOl) mean the taxon is questionably known longing within the family Equidae. He identifies the ear-
from that locality, thus implying some doubt that the taxon is actually liest equid as actually belonging to the genus Pliolophus
present at that locality, either at the genus or the species level. (Owen, 1858), a synonym of Hyracotherium. Pliolophus
was named for a European taxon, but Hooker claims that
"HYRACOTHERIINAE" the North American H. pernix should also be considered
as properly belonging to Pliolophus. He also considers H.
Hyracotherium Owen, 1840 (synonym: Eohippus
sandrae to be a "stem equoid," related to the European
Marsh, 1876)
Cymbalophus, and H. seekinsi to be even more distantly
Type species: Hyracotherium leporinum Owen, 1840.
related to true equids, possibly related to the isectolophid
Type specimen: BMNH M 16336, from the London Clay of
moropomorph Cardiolophus.
Heme Bay, Kent, England (Kitts, 1956; Hooker, 1980).
Principal references: Kitts (1956), Robinson (1966), Gin-
Characteristics: Small size, skull primitive in that pre- and
gerich (1989, 1991).
postorbital lengths are of roughly equal size, premolars
essentially unmolarized, manus consists of four digits (MC
II to V) and pes of three digits (MT II to IV). Xenicohippus Bown and Kihm, 1981
Approximate molar length: Ml/M2 = 6.0-10.0 mm; ml/rn2 Type species: Xenicohippus grangeri.
= 6.0-11.0 mm (Kitts, 1956). Type specimen: USGS 292 from Wasatchian of Big Hom
Included species: H. seekinsi Morris, 1968 (known from lo- County, Wyoming (Bown and Kihm, 1981).
cality CC50); H. angustidens (Cope, 1875) (localities Characteristics: As stated in Bown and Kihm (1981, p. 257),
?SB21 [could be H. vasacciense], ?SB24, SB39G, this taxon differs fromHyracotherium "in having enlarged
[CP20A], CP23A, CP25B, [CP25C], [CP25D], CP26D, p2-p3, pl relatively close to p2, premolariform p3, and a
CP27 A, CP63); H. sandrae Gingerich, 1989 (locality squared m3 heel." The skull is unknown, the upper den-
CP19); H. index (Cope, 1873) (localities SB24, [CP20C], tition is poorly known, and the postcranials are unde-
CP27B); H. grangeri Kitts, 1956(localityCP19);H. vasac- scribed.
ciense (Cope, 1872) (localities ?SB21 [could be H. an- Approximate molar length: ml/m2 = 7.5-10.0 mm (taken
gustidens], SB22A, SB22B, SB22C, SB24, SB40, CP5A, from Bown and Kihm, 1981, Table 2).
CP25J, CP27B, CP27D, CP27E, [CP28B]); H. aemulor Included species: X. grangeri Bown and Kihm, 1981 (known
Gingerich, 1991 (locality CP20); H. pemix (Marsh, 1876) from localities CP20A, CP64B); X. osborni Bown and
(locality uncertain); H. cristatum Wortman, 1896; H. tapir- Kihm, 1981 (locality CP64B).
inum (Cope, 1875) (= H. craspedotum of Kitts, 1956, Xenicohippus sp. is also known from localities SB24
Robinson, 1966; see Gingerich, 1981) (combined local- and (CP27B).
ities SB22A, SB22B, SB24, CP20E, [CP25E], [CP25F],
[CP25G], CP25H, [CP25J], CP27B, CP27D, CP27E, Comments: Where it occurs in early Eocene faunas, Xenico-
[CP28A]). hippus is rare. If indeed it is distinct, the diagnostic char-
Hyracotherium sp. (includes unsynonymized species H. acters used by Bown and Kihm (1981) place this taxon
venticolum, H. index, and H. borealis) is also known from morphologically between Hyracotherium and Orohippus.
localities GC2111, SB25A, CP20A, CP20B, CP20C, CP20D, It is provisionally listed here as distinct, but further com-
CP20E,CP20F,CP24C,CP25A,CP27C,CP27D,CP28A, parisons should be made in the future to determine the
CP28B, CP28C, CP31A, CP64A, CP64B, CP64C, NP14, validity of this genus (Froehlich, in progress).
NP49. Principal reference: Bown and Kihm (1981).
 Equidae 543

Orohippus Marsh, 1872 Haplohippus McGrew, 1953

Type species: Orohippus pumulis Marsh, 1872. Type species: Haplohippus texanus McGrew, 1953.
Type specimen: YPM 11336, from the Bridgerian deposits Type specimen: FMNH PM 17, from the Chadronian Vieja
(horizon B) of the Bridger Basin, southwestern Wyoming. Formation of west (Trans-Pecos) Texas.
Characteristics: In contrast to Hyracotherium, Orohippus has Characteristics: Small size, very long diastema, p2 very prim-
P3-P4 with four cusps of roughly equal size, and these itive, p3 submolariform, p4 fully molarized. Skull, upper
teeth are fully molariform in advanced species. Other diag- dentition, and postcranials unknown. In contrast to Meso-
nostic characters include p3 and p4 with entoconids, and in hippus, McGrew (1953) notes that although the teeth are
the latter this cusp is as large as the other principal cusps. lower crowned, the mandibular ramus is deeper in Haplo-
The m3 heel is relatively shorter than in Hyracotherium hippus.
(Kitts, 1957). Unlike Hyracotherium, there is a paucity Approximate molar length: ml/m2: 9.5-10.0 mm (McGrew,
of cranial and postcranial material and therefore these are 1953).
poorly described in the literature. Included species: H. texanus only (known from localities
Approximate molar length: M l/M2 = 6.0-9 .0 mm; m l/m2 = SB44B, PN5B).
6.0-10.0 mm (Kitts, 1957; also Korth and Evander, 1982).
Included .species: Stirton (1940) lists ten species of Orohip- Comments: Haplohippus is aberrant because it is at the ap-
pus, all from the Bridger Basin of Wyoming and all of proximate same grade of premolarization of Orohippus,
Bridgerian (middle Eocene) age. Korth and Evander but occurs 10 Ma later in time, with no known morpholog-
(1982) extend the range of this genus into the late ical intermediates. Haplohippus is also interesting because
Wasatchian (late early Eocene). Based on Kitts (1957), the it has a very long postcanine diastema in contrast to other
following appear to be valid species: 0. pumulis Marsh, hyracotheres.
1971 (known from localities CP20F, [CP25J], [CP25K], Principal reference: McGrew (1953).
CP34A, [CP34B], CP34C); 0. major Marsh, 1874 (local-
ities [CP32B], CP34C); 0. progressus Granger, 1908 (lo- Epihippus Marsh, 1878 (synonym: Duchesnehippus
calities CP34C, CP34D); 0. agilis, Marsh, 1873 (locality Peterson, 1931)
CP34D); 0. sylvaticus (Leidy, 1870) (localities CP34D, Type species: Epihippus gracilis (Marsh, 1871).
CP65); possibly 0. proteros Korth and Evander, 1982 (lo- Type specimen: YPM 11271, from the late Eocene (Uintan)
cality CP27D). Uinta Basin of Utah.
Orohippus sp. is also known from localities ?SB43A Characteristics: Small size, p2/P2 is advanced, that is, they
(could be Epihippus), CP27E, (CP31B), CP31C, CP31E, are submolariform with three primary cusps of nearly equal
CP32B, CP35, CP38B, PN5A. size and a smaller fourth cusp. P3/p3 and P4/p4 are fully
Comments: In contrast to the relatively widespread Hyra- molariform. In contrast to other hyracotheres, the first
cotherium, later hyracotheres like Orohippus are relatively
lower premolar is single rooted (as it also is in all anchith-
eriine and equine horses; see MacFadden, 1976). Postcra-
rare elements of their faunas. This taxon retains small size
and is generally similar to Hyracotherium except for a nial specimens are neither abundant nor have they been
slightly more advanced dentition. Although there are cra- described in the literature, but Stirton ( 1940) states that in
contrast to primitive hyracotheres, Epihippus has a rela-
nial and skeletal remains for Orohippus, these are poorly
tively larger MP III.
known in the literature.
Approximate molar length: Ml/M2 = 7.0-11.0 mm; ml/m2
Korth and Evander (1982) describe a new species based on a
single specimen, a maxillary fragment with P4-M3 from
= 6.0-10.0 mm (MacFadden, 1980a).
Included species: E. gracilis (Marsh, 1871) (known from lo-
the late Wasatchian Wind River Formation of central Wyo-
calities [SB42B], [SB44A], CP6A, CP6B, [CP29C],
ming. They base their diagnosis on the presence of a small
[CP36A], [NP8]); E. uintensis (Marsh, 1875) (localities
hypocone on P4, which would make their specimen ad-
SB43B, CP36B, CP37 A, CP37B, CP82, [NP22]); E. inter-
vanced relative to contemporaneous Hyracotherium. How-
medius (Peterson, 1931) (localities CP7B, CP7C, CP82).
ever, without more specimens from those deposits, that
Epihippus sp. (including unsynonymized species of E.
also show this advanced character, despite the argument
parvus) is also known from localities ?SB43A (could be
made in Korth and Evander (1982), it is possible that this
Orohippus), CP6B, CP29C, PN5B.
specimen represents a morphologically advanced individ-
ual from a Hyracotherium population. If Korth and Evan- Comments: Like other advanced hyracotheres (i.e., exclud-
der (1982) are correct about the presence of Orohippus in ing Hyracotherium ), and in contrast to later Tertiary horses,
the late Wasatchian, then, as they note, it may be transi- this genus is usually not an abundant component of mam-
tional between this genus and Hyracotherium. malian faunas.
Principal references: Kitts (1957) and Korth and Evander Principal references: Stirton (1940), Forsten ( 1971 ), and
(1982). MacFadden (1980a).
 Bruce J. MacFadden 544

"ANCHITHERIINAE'' Type specimen: YPM 11275, from the late Oligocene/early

Miocene (Arikareean) of the John Day, Oregon.
Mesohippus Marsh, 1875 (synonyms: Pediohippus Characteristics: According to Prothero and Shubin (1989,
Schlaijker, 1935, in part; Miohippus, Marsh, 1874, in part) p. 154), Miohippus differs from Mesohippus in "the artic-
Type species: Mesohippus bairdii (Leidy, 1850). ulation between the cuboid and third metatarsal, the larger
Type specimen: USNM 8632, probably from Bear Creek, hypostyles, the longer face, and the deeper facial fossa."
South Dakota (Osborn, 1918). Of these, they believe that the cuboid/MT articulation is of
Characteristics: Small size, preorbital region with shallow, great importance because it relates to the expansion of the
relatively poorly defined DPOF, P2-M3/p2-m3 molari- central metatarsal.
form, MC V reduced, most species tridactyl. Approximate molar length: Ml/M2=11.0-15.0mm;ml/m2
Approximate molar length: M l/M2: 11.0-12.0 mm (from = 11.0-17 .0 mm (Prothero and Shubin, 1989).
Prothero and Shubin, 1989);ml/m2=9.0-13.0mm(Clark, Included species: M. annectens Marsh, 1874 (= M. crassi-
Beerbowe,r, and Kietzke, 1967). cuspis) (known from localities CC9C, CP84B, [NP34B],
Included species: M. bairdii (Leidy, 1850) (= M. hypostylus) [PN6D], PN6E);M. obliquidens (Osborn, 1904) (=Meso-
(known from localities CP39E, CP40A, CP41A, CP43, hippus obliquidens, M. eulophus, M. meteulophus, Mio-
CP45A, CP83A, CP83B, CP83C, [CP98B], CP98C, hippus meteulophus, Miohippus/Pediohippus antiquus [in
CP99A, [NP10B2], NPIOBB, NP25C, NP50B); M. tex- part], Mesohippus/Pediohippus brachystylus) (localities
anus McGrew, 1971 (localities SB44B, SB44C, SB44D); CP40A,CP41A,CP42C,CP68B,CP68C,CP83C,CP84A,
M. celer (Marsh, 1874) (localities [CP42A], NP24C); M. CP84B, CP84C, CP98C, NP50C, NP51A);M. assiniboien-
latidens Douglass, 1903 (locality NP23); M. propinquus sis (Lambe, 1905) (= Mesohippus planidens, M. assini-
Lambe, 1905 (locality NPIOB); M. grandis Clark, Beer- boiensis, M. grandis) (localities CP83C, NPIOB, NPIOC,
bower, and Kietzke, 1967 (locality CP83); M. westoni ?NP50A); M. intermedius (Osborn and Wortman, 1895)
(Cope, 1889) (= M. montensis, M. praecocidens) (local- ( = Mesohippus/Pediohippus intermedius) (localities
ities CP41A, CP84A, CP98C, CP99A, NPIOA, NPIOB, CP42D, [CP101], CP84B); M. gidleyi (Osborn, 1904)
NP24C, NP24D, NP27C); M. exoletus, (Cope, 1874) (= (= Mesohippus/Pediohippus gidleyi, Miohippus validus,
M. trigonostylus, Pediohippus exoletus) (localities CP41A, Mesohippus grallipes) (locality CP84B); ?M. equinanus
CP68C, CP83C, CP84A, CP99A, NP50B, NP50C); M. Osborn, 1918 (= Archaeohippus equinanus) (localities
barbouri Schlaikjer, 1931 (= Miohippus antiquus, in part) CP84B, CP85C, CP86A, CP86B, CP86C, CP87 A, PN6C);
(localities CP42C, CP84A). M. gemmarosae Osborn, 1918 (localities [CP86C],
Mesohippus sp. (including unsynonymized species M. [PN16]).
portentus,M. proteulophus, andM. equiceps) is also known Miohippus sp. (includes unsynonymized species M. acu-
from localities GC7, NB2, SB44, SB52, CP39B, CP39CC, tidens, M. anceps) is also known from localities CP46,
CP39F, CP68B, CP98B, CP98C, NP9C, NPIOB, NP24C, CP68D, CP99C, NPIOBB, NPlOC, NPIOCC, NP36B,
NP24D, NP24E, NP25B, NP25C, NP27D, NP29C, NP37, ?NP50A, PN6F, PN6G.
NP30A, NP32B, ?NP34B, PN6C, PN6D, PN6E.
Comments: Although Prothero and Shubin (1989) have sug-
Comments: Morphologically, the cheek teeth in Mesohippus
gested that several diagnostic characters define Mesohip-
are fully molarized, which is in contrast to the hyracotheri-
pus, as Stirton (1940) noted, these two genera are indeed
ines. This genus was functionally tridactyl, although a
difficult to separate. Stirton ( 1940) lists eighteen named
Chadronian (late Eocene) specimen from Bates Hole,
species of Miohippus; however, like Mesohippus, a com-
Wyoming, preserves a rudimentary MC V (Emry, pers.
plete revision of all Oligocene and Miocene valid species
comm., 1975). Despite the work of Prothero and Shubin
(including those dealt with in Prothero and Shubin, 1989)
(1989) for the Orellan and Whitneyan forms, the specific
referable to this genus is still needed.
taxonomy of all valid species of Mesohippus still needs
Principal references: Stirton (1940) and Prothero and Shubin
to be done in a synthetic manner. Mesohippus was very
(1989).
widespread; in addition to its numerous occurrences in the
western United States, it also is the first equid to occur in
the southeast (Patton, 1969), and it also has been reported Kalobatippus Osborn, 1915 (synonym: Anchitherium Meyer,
from Canada (Emry, Russell, and Bjork, 1987). 1844)
Principal references: Clark, Beerbower, and Kietzke (1967), Type species: Kalobatippus praestans Cope, 1879.
Emry, Russell, and Bjork (1987); Prothero and Shubin Type specimen: AMNH 7269, upper John Day beds, Oregon.
(1989). Characteristics: Relatively large size, large preorbital fossa
poorly developed relative to Hypohippus, primitive
Miohippus Marsh, 1874 (synonyms: Pediohippus Schlaijker, low-crowned dentition, in upper cheek teeth crochet absent
1935; Mesohippus, Marsh, 1875, in part) and ribs present or weakly developed between styles; tri-
Type species: Miohippus annectens Marsh, 1874. dactyl metapodials.
 Equidae 545

Approximate molar length: Ml/M2 = 17.0 mm; ml/m2 = calities GC3B, GC4C, GC6B, CCl 7C, CC21A, CC21B,
19.0 mm (Osborn, 1918). CC23, NB19A, NB20A, SB32A, SB32D, SB33B, CP54B,
Included species: K. praestans Cope, 1879 (known from lo- CP106, CP107, CPI 10, CPll 1, CP114C, CP115B,
cality PN6E); K. gracilis (Marsh, 1892) (locality PN6E); CPI 16B, ?CP123A, NP40A, NP42, NP43, (PN7).
K. agatensis (Osborn, 1918) (localities CP52, [CP88],
CP104B); K. avus (Schlaijker, 1935) (locality [NB6E]); Comments: Species of this genus are usually larger than those
· K. clarenci Simpson, 1932 (localities GC9A, GC9C); K. of Anchitherium, except for the recently described Hy-
australis Leidy, 1873 (locality GC3II); K. navasotae (Hay, pohippus chico from Florida (MacFadden, 1982). Some
1924) (localities GC3B, GC8D, GC9B). forms like H. "large sp." in the F:AM collection attained
Kalobatippus sp. (including unsynonymized species K. very large body size estimated to be about 600 kg (Janis,
condoni) is also known from localities WM 13, CA 7, CCl 4, Gordon, and Illius, 1994), surpassing known Megahippus
NB 17, (SB32B), CP85B, CP103, CP104A, CP105, CP107, and similar to extant large domestic Equus. Thus, like
CP108A, CP108B, CP109, CPllO, CP114A, CPI 14B, Megahippus, Hypohippus underwent very rapid body size
PN6G,NC1A. increase (MacFadden, 1987). Based on their low-crowned
dentitions, they seem to have been browsers.
Comments: The genus Kalobatippus was proposed by Principal references: McGrew (1938), Stirton (1940), Webb
Osborn (1915) to include primitive low-crowned, three- (1969), and MacFadden (1982).
toed horses from the early Miocene of North America.
Subsequent workers (e.g., Stirton, 1940; Simpson, 1945) Megahippus McGrew, 1938
synonymized this genus with its presumed senior synonym
Type species: Megahippus matthewi McGrew, 1938.
Anchitherium Meyer, 1844, the latter of which is very abun-
Type specimen: USNM 10-16-5-13, from the Devil's Gulch,
dant throughout the Old World. Although these two genera north central Nebraska.
are indeed very similar in their cranial and dental charac-
Characteristics: Large size, well-developed facial fossa, dis-
ters, they are very different in the relative robustness of tinctive cup-shaped symphysial region with procumbent
their metapodials. Thus Kalobatippus is the primitive an-
incisors, premolars usually with cingula.
chithere from the early Miocene of the New World that Approximate molar length: Ml/M2 = 25.0-27.0 mm (Ted-
has relatively elongated and slender metapodials. So far
ford and Alf, 1962); ml/m2 = 27.0 mm (McGrew, 1938).
as is known, with this revised interpretation, Anchitherium
Included species: M. matthewi (Barbour, 1914) (known from
sensu stricto is not recognized in North America, although, localities [NB23C], [SB34A], CP114D, CP116A);
like the other larger, low-crowned equids, the systematics M. mckennai Tedford and Alf, 1962 (localities NB6E,
of Kqlobatippus is poorly known. This genus dispersed [SB32D], CP114A, CP114B).
into the Old World during the early Miocene, where it un-
Megahippus sp. is also known from localities GClOB,
derwent an adaptive radiation resulting in numerous named CCI 70, CCI 7H, NB7A, SB32B, CP73D, CP74B, CP75B,
species of Anchitherium (Abusch-Siewert, 1983). CP90A, CPI 14C, NP43.
Principal references: Osborn (1918) and Stirton ( 1940).
Comments: Megahippus is a very distinctive low-crowned
Hypohippus Leidy, 1858 horse that, like Hypo hippus and Anchitherium, is a rare el-
Type species: Hypohippus affinis Leidy, 1858. ement in the faunas in which it occurs. Megahippus evolved
Type specimen: USNM 573, from the Niobrara River, near a highly distinctive cropping mechanism in its anterior den-
Ft. Niobrara, Nebraska (Osborn, 1918). tition. It also underwent rapid increase in body size. Thus,
Characteristics: Large size, well-developed, and deeply pock- although primitive in many respects, it also had several
eted preorbital fossa, teeth low crowned, crochet absent, highly specialized and derived morphologies, and with re-
M3 always smaller than M2. gard to feeding, it seems to have been a browser.
Approximate molar length: Ml/M2 = 27.5 mm; ml/m2 = Principal references: McGrew (1938), Stirton (1940), and
26.0 mm (McGrew, 1938). Tedford and Alf (1962).
Included species: H. affinis Leidy, 1858 (known from locali-
ties GC6A, [NB23B], [SB32FF], SP2A, CP56, CP90A, Archaeohippus Gidley, 1906
CP114B, CP114D, CP116A); H. equinus (Scott, 1893) Type species: Archaeohippus ultimus Cope, 1886.
(localities CPI 10, [NPIOE], ?NP34E); H. osborni Gidley, Type specimen: AMNH 8174, from the Mascall Formation
1907 (localities [NB18], [NB23B], CP75B, ?CP76, of Oregon.
CP114A, CP114B, [NPl 1], [NP41B], [PN8B], [PN9A]); Characteristics: Very small size, shallow DPOF, malar fossa
H. nevadensis Merriam, 1913 (lo~ality NB23C); H. chico relatively well defined with posterior pit, low-crowned
MacFadden, 1982 (locality GCIOB). dentition, crochet varies in development, but never as
Hypohippus sp. (includes unsynonymized species H. strongly developed as in advanced Parahippus; tridactyl
pertinax, H. tehonense, H. wardi) is also known from lo- metapodials.
 Bruce J. MacFadden 546

Approximate molar length: Ml/M2 = 10.0-13.0 mm; ml/m2 stricted sense to include the low-crowned more primitive
= 10.0-13.0 mm (Downs, 1956). "parahippines." Because the taxonomy of the parahippines
Included species: A. ultimus Cope, 1886 (known from local- sensu Jato is exceedingly complex and no modern revi-
ities [CC21C], PN7, PN8A); A. blackbergi (Hay, 1924) sion is available, the species included here as pertaining to
(localities GC3A, GC3B, GC5, GC8D, GC9A, GC9B, Desmatippus are done so in a provisional sense following
[CP88], NC2); A. mourningi (Merriam, 1913) (localities Osborn ( 1918), pending a thorough revision of all relevant
CCI 7D, CC22B, CC23, NB6D); A. penultimatus, species.
Matthew, 1924 (localities CP108A, CP108B); A. minimus Principal reference: Osborn (1918).
(Douglass, 1900) (locality NP42); A. stenolophus Storer Parahippus Leidy, 1858
and Bryant, 1993 (= Mesohippus stenolophus, M. plani- Type species: Parahippus cognatus (Leidy, 1858).
dens) (locality NPlOD). Type specimen: USNM 567, from the Niobrara River region,
Archaeohippus sp. (includes unsynonymized species A. near Ft. Niobrara.
minutus) is also known from localities CA 7, GC2, GC4A, Characteristics: Cheek region with moderately developed
?CC16 (could be Merychippus), CC19, SB32A, CP52, DPOF, malar fossa absent; orbit closed posteriorly; in con-
CP105, CP106, CP107, CPllO, CPl14A, NP!l, PN6H, trast to primitive horses, in advanced Parahippus and
PN19B, NC3B. equine horses, preorbital skull length is greater than pos-
torbital length; cheek teeth mesodont to incipiently hyp-
Comments: This is a tiny, low-crowned genus; the species-
sodont; advanced forms with cement and crochet; meta-
level systematics are poorly known and in need of revision.
conids and metastylids separated.
From a paleobiological point of view, Archaeohippus is one
Approximate molar length: Ml/M2 = 16.0 mm; ml/m2 =
of the exceptions to Cope' s rule for body size in the Equidae
15.0 mm (for P. leonensis; Hulbert and MacFadden, 1991,
(MacFadden, 1987); it is one of the clearest examples of
Table 2).
phyletic dwarfing in horses, that is, the descendant species
Included species: P. cognatus (Leidy, 1858) (known from
are smaller relative to their presumed ancestors.
localities [CP86D], [CP88], CP104A, CP105, [CPllO],
Principal references: Bode (1933), Stirton (1940), and Downs
CPI 14A, CP114B, NPlOD); P. pristinus Osborn, 1918 (lo-
(1956).
cality CP86C); P. wyomingensis Schlaijker, 1925 (locali-
Desmatippus Scott, 1893 (synonyms: Anchippus; Parahippus, ties CP103, CP104B); P. leonensis Sellards, 1916
in part) (= P. vellicans, P. barbouri) (localities GC3B, [GC5],
Type species: Desmatippus crenidens Scott, 1893. GC8D, GC9A, CP105, NC2); ?P. agrestis (Leidy, 1873)
Type specimen: PU 10437, from the early Miocene Deep (locality NP40II); P. pawniensis Gidley, 1907 (localities
River beds of Montana. NB17; CP71, CP75B, CP104B, CP105); P. coloradensis
Characteristics: Low-crowned dentition lacking crochet, lit- Gidley, 1907 (locality CP86C).
tle, if any cement, relatively short limbs. Parahippus sp. (includes unsynonymized species P. max-
Approximate molar length: Ml/M2 = 15.0 mm; ml/m2 = imus and could include Desmatippus sp.) is also known
14.5 mm (Osborn, 1918). from localities GC5, (GC8A), (GC8C), CC15, CC16,
Included species: D. crenidens Scott, 1893 (known from lo- CC17B, CC17C, NB21, SB28, CP52, CP106, ?CP108B,
cality NP34D); D. integer (Matthew, 1924) (locality CP114A,CP114B, ?CPI 16B,NP10D,NP10E,PN6H,PN7.
CPI 10); D. tyleri Loomis, 1908 (localities [SB29C],
Comments: As traditionally envisioned (e.g., Matthew, 1926;
CP45D); D. nebrascensis Peterson, 1906 (localities
Stirton, 1940), the genus Parahippus represents a very het-
CP104B, [CP107]); D. texanus (Leidy, 1868) (localities
erogeneous assemblage of forms that span a morphocline
GC2, GC3A, GC3B, GC4A, CP86D); D. avus (Marsh,
of dental and postcranial characters. As used here, Parahip-
1874) (= Parahippus brevidens) (localities CC23, NB 18,
pus is set apart from the more primitive Desmatippus in
NB20A, CP108A, CP108B, PN7, PN8B, PN9A).
being on the derived end of the morphocline, that is, trend-
Desmatippus sp. is also known from localities GC4C,
ing toward or demonstrating the presence of crochets, in-
CP75B, NP40A, NP42.
cipient hypsodonty (in some taxa with cement), and re-
Comments: The parahippines sensu lato are represented by duced lateral metapodials. With regard to phylogeny, the
horses that are more advanced than Miohippus but lack advanced parahippines, that is, Parahippus leonensis, is
high-crowned cement-covered teeth. However, within the closest sister species to Merychippus (MacFadden and
these parahippines there is a vast array of differences in Hulbert, 1988; Hulbert and MacFadden, 1991). As with
dentitions ranging from primitive low crowned to subhyp- Desmatippus, the species within Parahippus sensu stricto
sodont. The genus Desmatippus was originally described are in great need of revision, as can be seen from the nu-
by Scott (1893) for low-crowned horses within the evo- merous named species, several of which coexist during the
lutionary grade. Despite the synonymies of earlier work- middle Miocene in the mid-continent.
ers (e.g., Stirton 1940), this genus is used here in the re- Principal references: Schlaijker ( 1937), Stirton ( 1940),
 Equidae 547

Downs (1956), Forsten (1975), and Hulbert and Merychippus sp. is also known from localities CAI,
MacFadden (1991). CA2, GC4D, GC8A, GC9B, ?CC16 (could beArchaeohip-
pus), CC18, CC20, NB 17, SB6, (SB29C), CP45E, CP54B,
CP56, CP71, CP73B, CP107, CP108B, CPlll, CP116B,
EQUINAE
NP5A, NP34D, NP40A, NP40B, PN6H, PN9B,
Merychippus Leidy, 1857 ?NC3B.
Type species: Merychippus insignis Leidy, 1857.
Type specimen: ANSP 11276, from the Bijou Hills, South Comments: The genus Merychippus has traditionally been
Dakota (Skinner and Taylor, 1967). used in the literature as a form genus (i.e., grade mostly
Characteristics: Skull with DPOF usually developed, upper based on stage of dental evolution) to include a
cheek teeth moderately hypsodont and cement covered, polyphyletic assemblage of subhypsodont to hypsodont
crochet present forming fossettes (lakes), lower teeth with horses. Stirton (1941, p. 178) aptly noted, "The genus
separated metaconids and metastylids; tridactyl meta- Merychippus is extremely difficult to diagnose, since its
podials. species intergrade almost imperceptibly with those of
Approximate molar length: Ml/M2 = 16.0-21.0 mm (Skin- Parahippus on the one hand and those of Pliohippus,
ner and Taylor, 1967); ml/m2 = 17.0 mm (Osborn, 1918, Calippus, Nannippus, Hipparion, andNeohipparion on the
Figure 86:2). other." In the strict interpretation, Merychippus includes
Included basal "Merychippus" species: M. gunteri Simpson, the species listed here. But some workers have also in-
1930 (known from localities GC3B, GC4B, GC4C, GC4D, cluded the subgenus Protohippus within Merychippus
GC4E, GC9B, GC9C, [NB20A]); M. primus Osborn, 1918 (sensu Jato), although this no longer is the preferred inter-
(localities GC9, CP54B, [CP107], ?CP108A, CP108B); M. pretation (e.g., Hulbert and MacFadden, 1991; MacFad-
re/ictus (Cope, 1889) (localities NP42, PN7, PN9B). den, 1992). Recent systematic studies have attempted to
"Merychippus" species included within the tribe Hippar- unravel the traditional horizontal merychippine grade and
ionini: M. insignis Leidy, 1858 (known from localities allocate species previously placed inMerychippus to other,
?GC4B, [GC4C], GClOA, [NB6C], [CP75B], CP89, more advanced equine or hipparionine taxa. Hulbert and
CP108B, CPl 10, CPI 14A, [CP114B], [CP114D], MacFadden (1991) have shown that within Merychippus
[CP116A], [NP38E]); M. sejunctus (Cope, 1874) (local- sensu stricto, the most primitive sister taxa are M. gunteri
ities [GC4B], [GC4C], CP75B, CP75C, [NP38E]); M. ter- and then M. primus followed by the other merychippine
tius Osborn, 1918 (localities CP107, CP108B); M. isone- taxa within the monophyletic tribe Hipparionini (includ-
sus (Cope, 1889) (localities [GC9C], [NB6B], NB 18, ing Merychippus s.s. and related forms) and tribe Equini
SB32A, SB32B, CPl 10, [NPl l], NP38C, PN7, PN8A, (including Protohippus and related genera; Figure 37.5).
PN9A, PN20); M. calamarius (Cope, 1875) (localities As such, some of the original ideas used for the split be-
NB23B, SB32D); M. californicus Merriam, 1915 (local- tween Merychippus and Protohippus proposed earlier are,
ities GClOB, CC23, [NB18], NB20A, ?NB23B [could in general, corroborated.
be M. sumani], NB20A); M. seversus (Cope, 1878) (lo- Principal references: Stirton (1940), Downs (1956), Skinner
calities [CCI 7B], PN7, PN8B); M. brevidontus (locali- and Taylor (1967), and Hulbert and MacFadden (1991).
ties WM13, [GClOA], CC17D, CC21C, CC23, NB18,
NB20A, PN8B, PN9A); M. goorisi (MacFadden and Skin-
HIPPARIONINI QUINN, 1955
ner, 1981) (= Cormohipparion goorisi) (localities GC4B,
GC4C, GC4D, GC4E, GClOA); M. coloradense (Osborn, Pseudhipparion Ameghino, 1904 (synonym: Griphippus
1918) (= Neohipparion coloradense,? = Hipparion san- Quinn, 1955)
fondensis) (localities [SB30B], [SB30C], SB32D, SB32F, Type species: Pseudhipparion retrusum (Cope, 1889).
SB33A, CP56, CP75B, CP76, CPl 10, CP114B, CP114D); Type specimen: No type specimen designated, original spec-
M. republicanus Osborn, 1918 (localities [GClOA], CP76, imen of Pseudhipparion retrusum from Phillips County,
CPI 14A, CPI 14B). Kansas, from the Ogallala Group (see also Webb and Hul-
"Merychippus" species included within the tribe Equini: M. bert, 1986, p. 240).
carrizoensis Dougherty, 1940 (= M. tehachapiensis) Characteristics: Medium to small hypsodont to very hyp-
(known from localities CC17B, CC17C, ?CC17D, CC19, sodont to incipiently hypselodont hipparionine horses;
CC21A, NB3D, NB6A, NB6C, [NB 17]); M. sumani Merr- DPOF poorly developed and with indistinct boundaries,
iam, 1915 (? = M. "sty/onus") (localities CC17C, CC17D, located far anterior on nasal and maxillary bones; malar
CC17E, CC17F, [CC21C], NB6E, NB6F, ?NB23B [could fossa absent; short postcanine diastema; protocone large
beM. californicus]); M. intermontanus Merriam, 1915 (lo- and elliptical, isolated during early wear but connected
calities [CC21C], [CC22B], ?NB6D, NB6E, NB6F, to protoloph during later ontogeny; in lower cheek teeth,
[SB32B], CPI 10, [NP41B], [NP38C]); M. stylodontus metaconids and metastylids widely separated, ectofiexid
Merriam, 1919 (localities CC21B, NB6C, NB6D). shallow in premolars; tridactyl metapodials.
 BruceJ.MacFadden 548

Approximate molar length: MI/l\12: I3.0-I8.0 mm; ml/m2 Neohipparion Gidley, 1903 (synonym: Hesperohippus
= I3.0-I 9.0 mm (Webb and Hulbert, I986). Dalquest, 1981)
Included species: P. retrusum (Cope, I889) (known from lo- Type species: Neohipparion affine (Leidy, 1869).
calities CP56, CP87B, CPI I4C, CPI I4D); P. "early Type specimen: USNM 584, Niobrara River region, near Ft.
species" (Webb and Hulbert, I986) (locality GCIOB); P. Niobrara, Nebraska.
curtivallum (Quinn, I 955) (= Astrohippus curtivallum) Characteristics: Medium to large size; DPOFrelatively poorly
(localities GC6A, GC6B, GCIOC); P. hessei Webb and defined; malar fossa absent; in upper cheek teeth isolated
Hulbert, I986 (= Nannippus gratus) (localities SPIA, elongate-oval to very elongated protocone; in lower cheek
SP2A); P. gratum (Leidy, I 869) (localities ?SB34A, teeth well-developed pli caballinids and widely separated
CP90A, CP90B, CPI I5B, CPI I6A, CPI I6B); P. skinneri metaconids and metastylids, flattened protoconids and
Webb and Hulbert, I986 (localities [GCllA], [GCilB], hypoconids; slender, elongated, and tridactyl metapodials.
[GCI3A], CPI I6B); P. simpsoni Webb and Hulbert, I986 Approximate molar length: MI/l\12 = 20.0-24.0 mm
(localities ?GClOD, GCI3B, [GC28], CPI27). (MacFadden, l 984b; Hulbert, l 987a); ml/m2 = 22.0-26.0
Pseudhipparion sp. is also known from localities GC4E, mm (Osborn, 1918; Hulbert 1987a).
GC6B, ?GC6D (could be Calippus), CP75C, CPI I4B. Included species: N. affine (Leidy, 1869) (= N. whitneyi, N.
dolichops,? = Hipparion sanfondensis,? = Hippotherium
Comments: Particularly in later forms, this genus is highly speciosum) (known from localities GC6A, GC6D, SB32D,
distinctive in its small size, dental pattern, and very SP2A, CP90A, CPI I4B, CPI 15B, CPI I6A, CPI 16B,
high-crowned dentition. Pseudhipparion simpsoni is the CP123A); N. trampasense (Edwards, 1982) (= Hipparion
only known fossil horse to have a partially ever-growing trampasense) (localities GCl IA, GCl lB, GCl IC,
dentition, termed incipient hypselodonty, in which growth CC30A, CP116B, CPI23C, [CP126]); N. leptode Mer-
of the incisors and cheek teeth continues past the early on- riam, 1915 (localities NB31, SPlB, SPIC, SPlD, [SP3A],
togeny of the individual. In this regard, Pseudhipparion is CPI 16D, CPI 16E, [PNI2]); N. eurystyle (Cope, I893) (=
the most derived equid genus. N. phosphorum, N. molle, N. arellenoi, N. floresi, N. otomi,
Principal reference: Webb and Hulbert (1986). in part, N. manias, "Hesperohipparion stirtoni") (local-
Hipparion Cristol, 1832 (synonym: Hippotherium ities CA4A, GC8F, GClOD, GC12II, GC13A, GC13B,
Kaup, 1833) GC13C, GC27, [GC28], CC26B, [CC32B], CC36, CC37,
Type species: Hipparion prostylum Cristo!, 1832. [CC38], CC39B, SB58A, SB60, SPlD, [SPlE], SP3B,
Type specimen: Unnumbered specimen in Avignon, France SP4A, SP4B, CP115C, CP115D, CP116C, [CP116F],
(see MacFadden, 1980b). NP5B); N. gidleyi Merriam, 1915 (= N. otomi, in part)
Characteristics: Medium size; DPOF well developed with (localities CC41II, SB34B, SB58A, SP3A).
prominent anterior and posterior rims, sometimes pock- Neohipparion sp. is also known from localities NB I 9C,
eted posteriorly; hypsodont dentition, upper cheek teeth ?SB48 (could be Cormohipparion), SPIA, CP78, CPI 14C,
with isolated protocones and moderately to strongly pli- PN14, PNI5.
cated enamel borders; relatively deep ectoftexids in lower
Comments: Neohipparion is a highly distinctive genus, par-
molars; tridactyl metapodials.
ticularly in the more advanced forms because they evolve
Approximate molar length: of Ml: I5.6-21.0 mm (Mac-
a derived dental pattern. Dalquest (I98I) described the
Fadden, I984b).
new genus Hesperohippus for what are here considered
Included species: H. shirleyae MacFadden, I 984b (known
advanced species of Neohipparion, and, as such, the former
from localities GC4C, GC4E); H. tehonense (Merriam,
is a junior synonym. Neohipparion was very widespread
I9I6) (= Nannippus tehonense, Hipparion lenticulare)
in the New World during the late Miocene. Horses with
(localities GC6B, [GCl IA], [GCl IC], [GCI2II], CCI 7G,
Neohipparion-like morphology also occur in the Old World
CC25A, CC27, CC32A, NB7C, NB27A, SPIA, SP2A,
(Zhegallo, I 978; MacFadden, I 984b) and, although phylo-
CP90A, ?CPI 14C, CP116A, CP116B, CP123C); H. for-
genetic affinities are suggested, parallel evolution of these
cei Richey, 1948 (localities [CC25B], CC26B, CC30A,
structures cannot be ruled out without further comparisons.
CC32B, NB7C, NB7D, NB7E, SP3A, CPI I5B, CPI 16D).
Principal references: MacFadden (I984b), Hulbert (1987a),
Hipparion sp. is also known from localities GClOA,
and Hulbert (1987b).
SPlB, SPIC, CPI 14A, NP5A, NPI 1.

Comments: Hipparion is the central concept for the genus of Nannippus Matthew, 1926
tridactyl fossil horses with isolated protocones in the upper Type species: Nannippus peninsulatus (Cope, I885).
molars and hence the nomen for the tribe Hipparionini. In Type specimen: AMNH 8345, from the "Loup Fork shales"
addition to the presence of Hipparion in North America, of Tehuichila, Mexico.
this genus was also very widespread in the Old World. Characteristics: Small to medium-sized hypsodont hippari-
Principal reference: MacFadden (1980b, 1984b). onine, elongated rostrum, DPOF varies from moderately
 Equidae 549

well developed in primitive species to absent in advanced rims and posterior pocket; malar fossa absent; moderately
species, cheek teeth hypsodont to very hypsodont with hypsodont cheek teeth; upper cheek teeth with isolated pro-
moderately complex enamel plications, upper cheek teeth tocones, moderately to very plicated fossette borders and
with oval, isolated protocones; lower cheek teeth with well-developed pli caballins; lower cheek teeth with pli ca-
poorly developed pli caballinids and deep ectoftexids; ballinids, which are better developed in younger species;
anterior projection of P2/p2 (anterostyle and paraconids) tridactyl, moderately stout, tridactyl metapodials.
poorly developed; gracile, tridactyl metapodials. Approximate molar length: Ml/l\12 = 17.0-23.0 mm
Approximate molar length: MI/l\12 = 14.5-18.0 mm (MacFadden, 1984b); ml/m2 = I8.0-24.0 mm (Hulbert,
(MacFadden, 1984b); ml/m2 = 15.0-18.0 mm (Hulbert, 1988).
1987a, Table 9). Included species: C. occidentale (Leidy, I856) (= Neohip-
Included species: N. peninsulatus (Cope, 1885) (= N. phle- parion occidentale, Hipparion mohavense) (known from
gon, N. hesperides, N. montezuma) (known from localities localities [CAIO], GC6A, ?GC6B, CC26B, CC27, CC35A,
CA4B, GCI4, GCl5, GCl6, GCl7, GC18, SB12, SB13, NB7B, NB7C, NB7D, NB7E, [NB l 9C], [NB23C], NB29,
SB14A,SB14B,SBl4C,SBI4D,SBl4~SBI5,SB18A, SB30C, SB33B, [SB34A], ?SPlA, ?[SPlB], ?[SPIC],
SB 18B, SB3 l C, SB35, SB49, SB50, SB58C, SB63, SB64, SP2A, SP2B, [SP3A], SP3B, CP78, CP87B, CP90A,
SB65, SPlF, SPlH, SP5, CPl21, CP128, CPI30A, CP90B, CPll4D, CPI 15B, CPI I6A, CPI I6B, CP116C,
CP130B); N. ingenuus (Leidy, I885) (= Cormohipparion CPl23A, CP123C, CPI26); C. sphenodus (Cope, 1889)
ingenuus, Neohipparion ingenuus, Neohipparion lenticu- (localities GC6B, [GCIOA], GClOB, GClOC, CC27,
laris, Hipparion lenticularis, Nannippus lenticularis, Pro- CC31, SB32H, SP2A,CP76, CPI 14A,CPI I4B, CPI I4C);
tohippus lenticularis) (localities GC8F, GCl IA, GCl lB, C. plicatile (Leidy, 1887) (localities CA9, ?GCSE, GCI IA,
GC12II, GC13A, GCI3B, GC27, SPlD, SP2A, SP3B, GCI IC, GC12II, GCI3B); C. emsliei Hulbert, I988 (lo-
SP4A, SP4B, CPI I5C, CPI I6D, CP116E, CP116F, calities GCIOD, GCl IC, GCI3B, GCI5, GCl7, GC29).
CP123D, NP5B, NP45); N. minor (Sellards, 1916) (= N. Cormohipparion sp. is also known from localities (CA3),
aztecus) (localities [GC6D], GC8F, GCIOD, [GCl lB], NB27A, SB32D, SB32F, ?SB48 (could be Neohipparion),
GCllC, GCI2II, GCl3A, GC13B, GCI3C, GC29, NC5.
SB58A, SB60, SP3B, [CPI 18]); N. westoni (Simpson,
1930) (localities GCllA, GCI lB); N. morgani Hulbert, Comments: Cormohipparion was originally recognized by
I993 (localities GCllC, GC12II); N. beckensis Dalquest Skinner and MacFadden (1977) based on the well-
and Donovan, 1973 (localities [GC6D], SPIF). developed and highly distinctive DPOF and complicated
Nannippus sp. is also known from localities GC4E, GC6D, dental pattern; Hulbert (1987a, 1987c, I 989b) has added in
GCIOB, CC25B, SB59, SP2A, CP90A, ?CP95, CPI 15B, his revised diagnosis several distinctive dental characters,
CPI 16A, CPI I6B, CPI 16C, CP123A, CPI23D, PNI4. as listed earlier. As recognized by Woodburne, MacFad-
den, and Skinner (I98I), Cormohipparion sphenodus is
Comments: Matthew (1926) originally proposed Nannippus probably the closest sister species of the Old World radia-
as a subgenus of Hipparion to include small hipparionine tion of hipparionine horses.
species from North America. However, most subsequent Principal references: Skinner and MacFadden (1977),
workers (e.g., Stirton, 1940) have elevated Nannippus to MacFadden (1984b), and Hulbert (I987a, I987c,
full generic rank. In addition to its generally small to very 1989b).
small size, Nannippus has distinctive dental characters,
as listed earlier. Although the species of Nannippus are
small, there is not a unidirectional size decrease; N. minor, EQUINI GRAY, 1821
which is a species of intermediate age, is the smallest of Protohippus Leidy, 1858 (synonyms: Hippotherium, in part;
the genus. As noted elsewhere (MacFadden I 987; Hulbert, Merychippus, in part; Eoequus; Pliohippus, in part;
I 988, 1993), several kinds of dwarfism exist in hipparions Dinohippus, in part)
in North America. This evolutionary phenomenon also oc- Type species: Protohippus perditus (Leidy, 1858).
curs in the Old World (Bernor and Tobien, I 989). Type specimen: USNM 619, from Niobrara River region,
Principal references: MacFadden (l 984b) and Hulbert (1993). near Ft. Niobrara, Nebraska.
Characteristics: Medium to large size; DPOF large, tear-
Cormohipparion Skinner and MacFadden, 1977 dropped shaped, lacrimal bone large and included in poste-
Type species: Cormohipparion occidentale (Leidy, I856). rior portion of DPOF, malar fossa absent; hypsodont teeth
Type specimen: ANSP 11287, probably from along Little with protocones that are isolated only during relatively
White River, South Dakota (Ski.nner and MacFadden, early wear stages, simple fossette enamel borders, tridactyl
1977). metapodials.
Characteristics: Medium to large size, distinctive tear-dropped Approximate molar length: MI/l\12 = 20.0-23.0 mm; ml/m2
shaped DPOF with well-developed anterior and posterior = 21.0-23.0 (Hulbert, I988).
 Bruce J. MacFadden 550

Included species: P. perditus (Leidy, 1S5S) (= P. ["Hippo- pus], in part, Protohippus martini, in part, Astrohippus sel-
therium"] profectus, Merychippus perditus, Merychip- lardsi, in part, Equus laparensis) (localities GC6A, GC6B,
pus circulus, in part, "Eoequus" wilsoni) (known from lo- GClOC, GC13B, [SPIA], [SP2A], [CP90A], CP90B,
calities GC4E, GC6B, GClOB, CP114A, CP114B, CP123A); C. cerasinus Hulbert, l 9SS ( = Astrohippus mar-
CP114C); P. supremus Leidy, 1S69 (= Pliohippus supre- tini, in part) (localities GCllA, GC13B, CP116B);
mus, in part, Pliohippus!Protohippus simus, "Hippotigris C. hondurensis (Olson and McGrew, 1941) (= Pliohip-
sellardsi") (localities GC6A, ?GC6B, GClOC, GC90A, pus hondurensis and C. castilli) (localities CA4A, CA9,
SP2A, CP114D, CP115B, CP116A, CP116B); P. gidleyi CAlO, GCllB); C. maccartyi Hulbert, 19SS (=cf. Plio-
Hulbert, 19SS (localities GCl lA, GCl lB, GC12II, CP7S, hippus hondurensis) (localities GCI IC, GC12II).
CP116B, CP116D);P. vetusQuinn, 1955 (localities GC4B, Calippus sp. is also known from localities ?GC6D (could
GC4D). be Pseudhipparion), GC lOA, GC l 3B, SPlB, SP! C, SP3A,
Protohippus sp. is also known from localities GClOA, CP75B, CP7S, CPI 16B, CPI 16C, CP116D, CP126.
NB21, SB29D, SB32A, SB32B, SB32FF, CP126.
Comments: Calippus is another one of the lineages of prin-
Comments: Protohippus has variously been considered a sub- cipally small horses in the Tertiary of North America.
genus (first of Equus, then of Merychippus) or a distinct However, unlike its contemporary Pseudhipparion, Calip-
genus. As discussed earlier, the affinities to Merychippus pus, although initially small, exhibits size increase through
were mostly based on primitive dental characters. Hulbert time - some early Hemphillian forms from Nebraska are
( l 9SS) has shown that Protohippus is distinct and has char- relatively large. Calippus is distinctive in its highly spe-
acters that place it as the primitive sister taxon within the cialized linear and enlarged muzzle and incisor arcade.
tribe Equini. Principal reference: Hulbert ( l 9SS).
Principal references: Stirton (1940), Hulbert (19S7a), and Pliohippus Marsh, 1874
Hulbert (19SS). Type species: Pliohippus pernix Marsh, 1S74.
Calippus Matthew and Stirton, 1930 (synonyms: Protohippus, Type specimen: YPM 13007, from Niobrara River region,
in part; Astrohippus, in part) Nebraska.
Type species: Calippus placidus (Leidy, 1S5S). Characteristics: Medium to large hypsodont equine; DPOF
Type specimen: USNM 621, from Niobrara River, near Fort and malar fossa well developed, oftentimes complex and
Niobrara. deeply pocketed; upper cheek teeth strongly curved with
Characteristics: Very small to medium-sized horses; when protocones connected to protoselene and frequently shal-
present (i.e., it is poorly developed or absent in some low hypoconal groove; lower cheek teeth with poorly sepa-
species), DPOF long, shallow, and not pocketed; malar rated metaconids and metastylids; moderate to deep
fossa absent; incisor arcade linear and relatively expanded; ectoflexid; early populations tridactyl, later populations
short postcanine diastema; cheek teeth hypsodont to very monodactyl.
hypsodont; upper cheek teeth with large oval to elongated Approximate molar length: Ml/M2 = 22.0-25.0 mm; ml/m2
protocones that remain connected to the protoselene, pli ca- = 22.0-27.0 (Osborn, 191S, Figure 120; Matthew and
ballin rudimentary or absent, relatively shallow hypoconal Stirton, 1930).
groove which frequently closes to form an isolated lake Included species: P. pernix Marsh, 1S74 (known from locali-
during wear; generally simple enamel pattern. ties [SB30C], [SB34A], SP2A, CPS7B, [CP90A], CP90B,
Approximate molar length: Ml/M2: 13.0-17.0 mm; ml/m2 CPI 14D, CP115A, CP116A); P. supremus (Leidy, 1S69)
= 14.0-17.0 (Hulbert, 19SS). (localities GC6B, GClOC, [SB32D]); P. fairbanksi
Included species: C. placidus (Leidy, 1S5S) (= C. optimus, C. Merriam, 1915 (= P. tantalus,? = P. leardi,? = P. teho-
anatinus, in part, Protohippus placidus, Protohippus!Astro- nense) (localities [Cl 7G], CC25B, [CC26B], [CC32A],
hippus martini, in part) (known from localities GC6A, CC32B, NB7B, NB7C, NB7D, [NB23C], [NB31]);
GC6B, SPlA,CPI 14B,CP116A); C.proplacidus(Osbom, P. mirabilis (Leidy, 1S5S) (= Merychippus mirabilis, M.
191 S) ( = C. francisi, Merychippus francisi, Proto hippus campestris) (localities [GClOB], [NB7C], CP75C, CP76,
proplacidus, in part, Merychippus proplacidus) (localities CPI 14B, CP114C); P. nobilis Osborn, 191S (localities
GC4E, GCSE, GClOB, CP76, CP114A, CP114C); C. reg- CPI 16D, CP116E, CP123C); P. spectans (Cope, lSSO) (lo-
ulus Johnston, 1937 (= C. anatinus, in part, Protohip- calities [CCI 7H], CC36, [CC3S], [NB27B], PNl 1, PN12,
pus proplacidus, in part) (localities GC4E, GC6A, GC6B, PN13, PN15); P.fossulatus (Cope, 1S93) (localities SPIA,
SPlA, SP2A, CP76, CP114C, CP114D); C. sp. (see Hul- SP2A); P. coalingensis (Merriam, 1914) (localities CC37,
bert, 19SS) (localities GC4B, GCSE, GClOB, GClOC); CC39B); P. bakeri Azzaroli, 19SS (locality SP3B).
C. elachistus Hulbert, 19SS (localities GCllA, GCllB, Pliohippus sp. (includes currently valid, although pos-
GC13A, GC13B); C. circulus(Quinn, 1955) (=Merychip- sibly junior synonyms P. robustus, P. castilli, P. profec-
pus perditus, in part, Pliohippus circulus, in part) (locality tus, P. minor, and P.francesca) is also known from locali-
GC4E); C. martini Hesse, 1936 (= Pliohippus [Astrohip- ties ?WM21A and ?WM21B (both could be Dinohippus),
 Equidae 551

CC17H, CC35A, CC41, (NB7B), (NB7C), (NB7D), Principal references: Matthew and Stirton (1930), Stirton
(NB7E), ?NBS, (NB9), NB19C, NB27A, NB29, NB33, (1940), Lance (1950), and MacFadden (19S4a).
SB6, ?SB4S (could be Astrohippus), SB30B, SB32D,
Dinohippus Quinn, 1955 (synonym: Pliohippus, in part)
SB32F, SB32FF, SB34A, SPlB, SPIC, SP3A, (CP54C),
Type species: Dinohippus leidyanus (Osborn, l 91S).
CP76, CP90A, CPI 14D, NP43, PN14.
Type specimen: AMNH 17224, from western Nebraska.
Comments: The systematics of Pliohippus are certainly in Characteristics: Large size; malar fossa absent; DPOF poorly
need of a modem revision; many of the names listed here developed, rudimentary, or absent; moderately curved up-
are probably synonymous. It has traditionally been stated per dentition (relative to Pliohippus); protocone strongly
in the literature that Pliohippus is the fossil monodactyl connected to protoselene, well-developed hypoconal
genus that gave rise to Equus (e.g., Matthew, 1926; Stir- groove, simple enamel plications; lower dentition with rel-
ton, 1940). This idea is no longer considered valid, prin- atively deep ectofiexids; so far as is known, monodactyl
cipally because Pliohippus has very complex facial fos- metapodials.
sae, whereas in Equus these structures are rudimentary or Approximate molar length: Ml/M2 = 26.0-27.0mm; ml/m2
absent (MacFadden, 19S4a). As such, Dinohippus, with = 23.0-26.0 mm (Osborn, 191S; Quinn, 1955).
relatively poorly developed facial fossae (as well as cer- Included species: D. leidyanus (Osborn, 191S) (known from
tain derived dental characters), is now considered the sister localities CP115C, CPI 15D, [CPI 16A], CPI 16D,
taxon of Equus (Quinn, 1955; Bennett, 19SO; MacFadden, CP116E, NP5B); D. pachyops (Cope, 1S93) (locality
19S4a). Pliohippus used to be considered the dominant SP2A); D. subvenus Quinn, 1955 (locality GC6B); D. in-
monodactyl horse in the Miocene and Pliocene. With the terpolatus (Cope, l S93) (localities CA4, CCI 7I, CC25B,
splitting apart of pliohippines sensu Jato, that is, Pliohip- SB34B, SPlD, SP3B, SP4A, [CP7S], CP123D, NP45);
pus and Dinohippus, the former genus is actually only D. mexicanus (Lance, 1950) (localities GC13B, GC13C,
common during the middle Miocene, whereas the latter SB5SA, SB5SB, SB60, SB61, [SP4]); D. osborni (Frick,
is widespread from the middle Miocene to early Pliocene 1924) (locality [SB64]); D. edensis (Frick, 1924) (locali-
(Figure 37.7). ties CC40, [NB 10]).
Principal references: Stirton (1940) and Webb (1969). Dinohippus sp. is also known from localities ?WM21A
(could be Pliohippus), ?WM21B (could be Pliohippus),
Astrohippus Stirton, 1940 (synonym: Pliohippus, in part) GCllC, ?NB13, ?NB14, SBlO, SB13, SB31B, SB34A,
Type species: Astrohippus ansae (Matthew and Stirton, SB59, CP116F.
1930).
Type specimen: UCMP 30225, from Coffee Ranch, Texas Comments: Dinohippus is a common late Miocene through
panhoodle. early Pliocene monodactyl horse in North American
Characteristics: Small to medium size; well-developed DPOF faunas. This genus was first recognized by Quinn (1955)
and malar fossa that are, so far as is known, confluent; and includes several species that were previously assigned
high-crowned dentition; upper cheek teeth with elongated, to Pliohippus. Dinohippus exhibits a facial morphology
connected protocones, simple enamel pattern, weakly de- and dental characters which indicate that it is the closest
veloped hypoconal groove; lowers with widely separated sister taxon of Equus (Figure 37.5), in contrast to the pre-
metaconids and metastylids, shallow ectofiexids, and well- vious interpretations of either Pliohippus or Astrohippus
developed metaconids and metastylids; so far as is known, as the ancestors of Equus (e.g., Matthew, 1926; Stirton,
Astrohippus is monodactyl. 1940).
Approximate molar length: Ml/M2 = 16.0-21.0mm; ml/m2 Principal references: Lance (1950), Quinn (1955), Webb
= 16.0-21.0 (Matthew and Stirton, 1930; Lance, 1950). (1969), and MacFadden (19S4a).
Included species: A. ansae (Matthew and Stirton, 1930)
(known from localities SB34B, SPlD, SP3B, SP4A, Hippidion Owen, 1869
CP116D, CP123D); A. stockii (Lance, 1950) (=A. al- Type species: Hippidion principale (Lund, l S45) ( = H. "neo-
bidens) (localities GC6D, GC13B, GC13C, SB13B, gaeum," H. bonaerense).
SB5SA, SB5SB, SPlE, [SP4], SB60). Type specimen: Right M2 from Lagoa Santa, Brazil, figured
Astrohippus sp. is also known from localities CC40, by Lund (1S45; plate XLIX), probably lost (Alberdi and
NB19D, ?SB4S (could be Pliohippus), SP3A, CP90A. Prado, 1993).
Characteristics: Very large, retracted nasal notch even farther
Comments: Astrohippus was originally described as a sub- posterior than for Onohippidium, preorbital cheek region
genus of Pliohippus (Stirton, 1940), but several more re- smooth with no fossae, upper cheek teeth with connected
cent papers elevated this nomen to generic rank (Quinn, protocones and crescentic fossettes, lower cheek teeth with
1955; Webb, 1969; MacFadden, 19S4a). Based on facial characteristically very deep ectofiexids almost separating
morphology (i.e., presence of facial fossae) and dental pat- the metaconids and metastylids, very simple enamel pat-
tern, Astrohippus is closely related to Pliohippus. tern.
 BruceJ. rvlacFadden 552

Approximate molar length: ml/m2 = 25.0-28.0 mm (from North American species; Sefve (1912) for South Amer-
rvlacFadden and Skinner, 1979; upper dentitions from ican forms.
North America are unknown).
Included species (North America): Hippidion sp. only
Equus Linnaeus, 1758 (synonyms: Plesippus and many
(known from localities ?[NBl 3B], [NB13C], SPIC). others, see Simpson, 1945)
Type species: Equus caballus Linnaeus, 1758.
Comments: Like Onohippidium, which was classically Type specimen: Not designated (the living domestic horse).
thought to be endemic to the Plio-Pleistocene of South Characteristics: rvledium to large size; facial region lack-
America, Hippidion has recently been reported from North ing fossae; very hypsodont, connected protocones, lower
America (rvlacFadden and Skinner, 1979). The temporal cheek teeth usually with shallow ectoftexids and well-
and paleobiogeographic range of this genus is based developed pli caballinids; complex enamel plications;
on two specimens, which are characterized by large size monodactyl metapodials.
and the di~tinctive deep ectoftexids in the lower cheek Approximate molar length: rvll/rv12 22.0-27.0
teeth. Because of the paucity of material, however, new (rvlacFadden, 1989; Hulbert, 1995); ml/m2 = 24.0 mm
species were not designated, and it is impossible to (Hulbert, 1995).
determine if these two occurrences represent the same Included species: Note: The genus Equus was very common
species. and widespread throughout North America during the Plio-
Principal references: rvlacFadden and Skinner (1979) for Pleistocene. Only the following species are recognized
North American species, Sefve (1912) and Alberdi and in the Tertiary of North America (Winans, 1989): Equus
Prado (1993) for South American forms. (Dolichohippus) simplicidens Cope, 1892 (= E. shoshon-
ensis, E. pons, E. occidentalis, E. calobatus, E. cumminsi,
Onohippidium rvloreno, 1891 E. achates, E. tau) (known from localities GC14, GCl5,
Type species: Onohippidium muiiizi rvloreno, 1891. CC43,CC44, CC45, CC47,NB12B,NB13B, SB12, SBl3,
Type specimen: Skeleton in the rvluseum of Natural Sciences, SB14A, SB14B, SB14C, SB14D, SB15, SB16, [SB31C],
La Plata, Province of Buenos Aires, Argentina. SB33C,SB35,SB36,SB38,SB49,SB50,SB58C,SB62,
Characteristics: Very large; cheek teeth moderately curved SB63, SB64, SB66, SPIF, SPIH, SP5, CP80, CP94, CP95,
with connected protocones in upper molars, moderate to CP97, CPl17A, CP117B, CP118, CP128B, CP128C,
deep ectoftexid in lower molars, very simple enamel pat- CP130A, CPI 30B, CP13 l, CP132A, PN3C, PN4, PN23A,
tern; very well-developed DPOF, malar fossa rudimentary PN23C); E. scotti Gidley, 1900 (localities [GC 14], [GC 15],
or absent; nasal notch retracted posteriorly to a position [GCl 7], [SB18A], [SB 18B], [SB36], SB37, SB38, SB50).
that lies dorsal to P3-P4. Equus sp. (probably= E. simplicidens, includes unsyno-
Approximate molar length: rvll/rv12 = 27.0 mm; ml/m2 = nymized species E. giganteus and E. idahoensis) is also
28.0 mm (rvlacFadden and Skinner, 1979). known from localities Wrv126, CA4B, GC18, CC53,
Included species (North America): Onohippidium galushai NB13C, NB36, SB14E, SB14F, SB17, SB18C, SB3411,
only, known from locality SB9 only. SB37, CP96, CP121, CP132B.
Comments: Alberdi and Prado (1993) question the valid-
Comments: Fossil representatives of the genus Equus were
ity of the genus Onohippidium (as separate from Hippid-
exceedingly widespread in North America during the
ion) as well as its supposed presence in North America.
Pliocene until the late Pleistocene. Accordingly, some sixty
Based on its unique morphological characters, Onohip-
names have been attributed to this genus for extinct species.
pidium is still considered valid here and it indeed occurs
Winans (1985, 1989) only considers five to be valid, of
in North America. Its presence in North America was
which the two listed here are recognized from Tertiary
only recently determined because of very well-preserved
localities. Equus also dispersed into the Old World and
material from Wikieup, Arizona (locality SB9), that pre-
South America during the late Pliocene and Pleistocene
serves the characteristic facial configuration of the DPOF
(Lindsay, Opdyke, and Johnson, 1980; rvlacFadden and
and retracted nasal notch (rvlacFadden and Skinner, 1979).
Azzaroli, 1987).
Surprisingly, this genus does not occur elsewhere in the
Principal references: Skinner (1972) and Winans (1985,
southwestern United States despite that fact that there are
1989).
numerous other localities of suitable age. It therefore seems
that ungulates, particularly horses and antilocaprids, were
highly endemic during that time (rvlacFadden, Johnson,
INDETERMINATE EQUIDS
and Opdyke, 1979; rvlacFadden, 1992). Onohippidium is a
common Plio-Pleistocene horse throughout much of South With regard to the composite locality list prepared in this book, sev-
America. eral fragmentary or nondiagnostic equid occurrences are currently
Principal references: rvlacFadden and Skinner (1979) for unassignable to a particular taxon. These fossil horses have been
 Equidae 553

reported from localities NB28, SB33II, NP9A, NP9B, NP34C, PN2, of these methods. With regard to taxonomic rates of evolution,
PNIO, PNl 1. Stanley ( 1979) quantified the intrinsic rate of increase of species
diversity within a clade as

BIOLOGY AND EVOLUTIONARY PATTERNS R =(In N -In No)/fj,_t,

where N 0 is the original number of taxa (usually species or genera),

With its widespread fossil record and familiar and beloved modem
and N is the number of taxa after a duration Dt. rvlacFadden and
analogue, horses have been frequently used in the literature as prime
Hulbert ( 1988) calculated the intrinsic rate of increase for species of
examples of long-term evolutionary patterns and processes. This
North American rvliocene horses. During the early phase of the adap-
section briefly touches on the more important of these examples.
tive radiation ofhypsodont species, R increased dramatically to 0.97
between 18-16 million years ago. (This compares to, e.g., murid ro-
TRENDS IN EVOLUTION dent species [R = 0.35] and bovid species [R = 0.15] during the late
Cenozoic [Stanley, 1979].) Thereafter, during the rvliocene, back-
The linear sequence of morphology depicted from Hyracotherium
ground levels of R were relatively low at 0.06 (a value similar to
to Equus is a time-honored graphic reproduced in numerous text-
Phanerozoic bivalves), indicating that a steady-state carrying ca-
books and museum displays. This straight-line, or orthogenetic, se-
pacity of equid species had been reached. In a taxonomic sense,
quence has its foundations in nineteenth-century paleontological
therefore, hypsodont horses experienced a very rapid burst of evo-
and evolutionary thought. Although the simplistic orthogenetic pro-
lution during the rvliocene.
gression of fossil horses was the paradigm of rvlarsh, Huxley, and
With regard to morphology, Haldane (1949) proposed the
other prominent paleontologists during the last century, later dis-
"darwin" as follows:
coveries during the early twentieth century did much to modify the
horse/orthogenesis couplet among active researchers (e.g., rvlatthew,
1926; Stirton, 1940; Simpson, 1951), although this simplistic no-
tion still persists in the derivative literature. The standard story has where r is the rate of morphological change (in darwins, d), x 1 is
been that horses got larger through time, elongated their grinding the initial dimension of a character, x 2 is the final dimension of a
teeth to eat abrasive grasses, and reduced the number of limb digits character, and /j,_f is the time duration between the ages of x 1 and
and elongated their limbs to better evade predators. Although all of x 2 • Using darwins, rvlacFadden (1988b) examined rates of morpho-
these character transformations potentially represent trends, that is, logical change in four linear dimensions of first upper molars (rvll );
a gradual, or progressive directional change through time, there are that is, (1) length and (2) width of the occlusal surface, (3) proto-
many exceptions to each of these so-called rules. cone length, and (4) unworn crown height. He found that during the
One of the universally recognized examples of "Cope's rule" explosive radiation of hypsodont taxa during the rvliocene, crown
(i.e., progressive increase in body size through time) is with the height evolved very rapidly (0.2 d for Parahippus and Merychippus;
55 million-year fossil record of horses - from the tiny, dog-sized eo- 0.08 d for hipparions), whereas occlusal dimensions evolved sig-
hippus (Hyracotherium) to the large and powerful Equus. nificantly more slowly (ca. 0.04 d for occlusal lengths and widths
rvlacFadden (1987) estimated the body sizes of fossil horses and of Parahippus, Merychippus, and hipparions). These slower rates
examined the extent to which the Equidae from North America are similar to the rates of evolution (0.04-0.05 d) for the same four
follow Cope's rule. Two important conclusions resulted from that dimensions in early horses (Hyracotherium-Mesohippus). Interest-
study: ( 1) Evolution in body size is not gradual, that is, the first 35 ingly, the low-crowned anchitheres (Anchitherium, Hypohippus, and
million years of horse evolution (before about 20 million years ago) Megahippus), although generally regarded as being conservative in
is characterized by relative stasis in body size; it is only during the their evolution, experienced very rapid rates of morphological evo-
late Tertiary that there is rapid increase in body size in some clades. lution (e.g., 0.11 d for crown height increase). Several general evolu-
(2) Even if there is a unidirectional trend in some clades, for less tionary phenomena are illustrated by these results: ( 1) All organisms
than half of the 55 million-year history of the Equidae numerous are "mosaics" (de Beer, 1954) of characters evolving at independent
exceptions to Cope's rule exist during this time, for example, the rates even within the same functional complex. (2) rvlls of different
five or so dwarfing lineages that evolved during the rvliocene and horses evolved at different rates; not all were gradual or progressive.
Pliocene. Thus, depending on one's point of view, Cope's rule as Although there were numerous periods of relative stasis in evolu-
applied to fossil horses is either a gross simplification of the more tion of equid rvlls, there also were times of rapid evolution in certain
complex evolutionary pattern, or it cannot be applied to this group characters.
because of the many exceptions within the Equidae.

FUNCTIONAL MORPHOLOGY
RATES OF EVOLUTION
The evolution of the feeding and locomotory complexes in fossil
Simpson (1953) describes two methods commonly available to pa- horses has also been another classic story told innumerable times
leontologists for assessing rates of evolution, that is, taxonomic in the textbooks. Traditionally it has been stated that primitive
and morphological. Fossil horses have been used to exemplify both low-crowned horses such as Hyracotherium and Mesohippus were
 BruceJ.l\1acFadden 554

w
z L 1.1
-c:- c:
w
0
LBL

E.BI
2A
.§ g: ,0 .£?r..;;
4.S :s:! Q. -a::C\'I 6
0
:i
E Lot.Hp.
LHp.
5.2
-Cll
-ic:
II) 0
8:re:
::i :s .... II>~ -Q. ·a::
c: g: .e;
Q.
C\'I
Cl. LE.Hp.
l.O -~·
.s ::i Q. ....
Q. .e; Q.
.i:::
.i::: 0 ::i :g- .e; C\'I
.i::: .e; g .s &·-.,.-,i:: 8:
-u ,0 Q. .i:::
E.E.Hp.
7.0

:s .e: a:: .e; 0

.e:a
.i::: 0 ::z: otl ud~i-"g
L LCI.
I.I
1.5
II) - - 0 ~ - -·.: QI '.§ c:
c: e
ca
~ ~ ~
::i ·-
E.CI. II) .5 g: :::::
Q. "':(
V)
a. ::z: 8
w 11.0
g: ?I &15- 8: -a.
zw
0 M
L.LBL

E.L.Ba.
12.5 -g: g: g: .e; Q. ::i
Q. Q. .i:::
-~ .e; .e; 0QI ~
II)

Q.
II)•~!:!!. :c ::i
::ii 11)111) 0 8:
II)
-;;I :c
..:....
---II)

~~
_::i
0 14.0
Q.
Q.Q. 1;°a:i
SE E.Bo. -
&
! '§ .i:::& .i:::C\'I eC\'I .i:::
e .e; .e; Q.1; -(j .e;
-~ -.C
L.Hm.
5.1
e
~ ~ ~ "':( Q ~ QI~ ~
m .i::: .i::: -
~
17.5 Q) - -Q)
::!!: :5_
-==
E.Hm.
11.1
E L.L.Ar.
1J
11.2

E.L.Ar.
23.0

w ? I
z
w
L.E.Ar.

L
0
0
'ZT.7 - - - - - - - - - - - - - - - - - -
CJ E.E.Ar.
:i 29.4
0
Wt. g:
E 31.9 g: Q. - - - - - - - - -
Or. Q. .e;
33.4
II)
.e; .i:::
'§
~
L.Ch.
34.5 ::i-
L M.Ch. Q.
35.5 .e;- II)
E.Ch.
~
Du
37.1
-~[
~
39.S _::z:
L.Un.

w 41.3 g: g:
z Q. Q.
.e; .e;
w E.Un.

~
0
0
w L.Br.
45.9
==
.fr
M 46.7

M.Br.
49.0
E.Br.
50.4 .§ g:
LWa.
t
:S
.e:Q.

~
E
53.6
g
M.Wa.
e
t
54.2
E.Wa.
55.0 ~
E.E.Wa.
55.5
Ck.3 55,6
w Ck.2 55.8
z Ck.1
Tl.6
56.0
56.2
Equini Hipparionini
w L Tl.5
Tl.4
57.0
0 Tl.3
57.3
0 Tl.2
58,5
w
..J
Tl.1
59.2
II ANCHITHERllNAE" EQUINAE
60.9
<( To.3
Cl. Ta.2
61.3 "HYRACOTHERllNAE"
62.5
Ta.1
E Pu.3
63.8
64.0
Pu.2 64.5
Pu.1 65,0
Pu.O 65.1

Bl.=Blancan, Hp.= Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar.= Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un. = Uintan, Br. = Bridgerlan, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti.= Tiffanian, To. = Torrejonian, Pu. = Puercan

Figure 37.7. Temporal ranges of North America equid genera.

 Equidae 555

forest-dwelling browsers and that the change to more open-country state than tridactyly. Although it is very difficult to compare the
grazing adaptations did not occur until the advent of hypsodonty relative merits of functionally analogous locomotory systems in
and extensive grasslands during the middle l\1iocene about 15 mil- extinct taxa, I would argue that both monodactyly and tridactyly
lion years ago. Although in the broadest outlines this scenario is were highly adaptive limb configurations. Tridactyly existed from
not incorrect, based on recent research the functional and adaptive the early Oligocene (with Mesohippus) for some 35 million years
morphology of horses is far more complex than the tidy picture just ago until the Pliocene (late surviving Nannippus and Cormohip-
presented. parion), some 2 million years ago. It has even been suggested that
Although it is possible that some species of Hyracotherium may although many later tridactyl forms (e.g., hipparions) were func-
have been forest-dwelling browsers (fide Janis, 1982), it is equally tionally monodactyl, their lateral toes may have provided additional
plausible that other species within this genus inhabited more open- stability in muddy substrates, thus imparting an advantage in cer-
country biomes and fed on soft vegetation such as bushes, herba- tain environments (Shotwell, 1961; Renders, 1984). l\1onodactyly in
ceous dicots, and other plants forming the ground cover (Gingerich, horses is first known in late l\1iocene equines from about 10 million
1981). years ago, and, as such, is a recent arrival on the scene.
It cannot be disputed that during the l\1iocene there was a major
functional reorganization in horses. As shown by Radinsky (1984),
POPULATION BIOLOGY
the major elongation of the preorbital region, deepening of the jaw,
and increase in hypsodonty in hipparionines and equines occurred Understanding of the autecology, or population biology of fossil
during this time. Since the time of Kowalevsky (1873), it has been species, requires extraordinary fossil deposits containing large, well-
believed that these changes were related to the changeover from preserved samples with many individuals. We are fortunate that
browsing to grazing along with the spread of savanna/grassland several important fossil horse species are known from large quarry
habitats during the l\1iocene. Although there appears to be some samples which appear to represent ancient populations.
truth to this, recent work indicates that again, this scenario is far too Gingerich (1981) describes a very important sample of Hyra-
simplistic. Using techniques such as enamel microwear (i.e., small cotherium tapirinum from Castillo Pocket of Colorado. Based on
distinctive scratches diagnostic of diet made on the occlusal tooth the degree of sexual dimorphism and ratio of sexes represented at
surface during chewing of food) and the gross morphology of the that quarry (as determined from relative size), Gingerich (1981)
muzzle region, even though middle l\1iocene horses like advanced asserted that Hyracotherium lived in small harems consisting of a
Merychippus and Cormohipparion were relatively high crowned, single dominant male, a few females, and a few juveniles. He also
their diet most probably consisted of both browse and grass, that is, believed that Hyracotherium had more of a range of habitat and
they were mixed feeders (Janis, 1989; Hayek et al., 1991). Other feeding preferences than has traditionally been interpreted, that is,
work using carbon isotopes (Wang, Cerling, and l\1acFadden, 1994) from browsers to more open-country foragers (although not graz-
found no corre'spondence between the advent of hypsodonty and ers). In contrast, based on its size, patterns of gross dental wear,
spread of grasses as we know them today (i.e., predominantly C4 and presumed forest-dwelling adaptations, Janis (1982) speculated
photosynthesis). Thus, when hypsodont horses underwent their ma- that Hyracotherium was solitary, and as such, was similar to several
jor adaptive radiation during the middle l\1iocene, they exploited small, extant forest-dwelling artiodactyls, for example, chevrotains,
primitive C3 grasslands (which are relatively rare today). which also show sexual dimorphism in canine size.
With regard to locomotion, Hyracotherium and other primitive Several important population studies have been done for l\1iocene
horses were subunguligrade in their stance. A close modem ana- horses. Hulbert ( 1984) studied the population of Parahippus leonen-
logue in terms of their running ability is the tapir (Camp and Smith, sis from Thomas Farm, Florida. This population lacked distinctive
1942). It has classically been stated that Hyracotherium was a rel- sexual dimorphism seen in early horses (i.e., Hyracotherium) and
atively slow and clumsy runner and, in contrast, Equus is a highly also lacked well-developed age cohorts (based on wear classes of
perfected "running machine." Although it is almost certainly true dentitions). Based on these observations, Hulbert concluded that
that Equus can run faster than Hyracotherium could have, compar- P. leonensis probably did not have peaked seasonal breeding, also
ing the locomotory adaptations of an Eocene horse to that of a mod- suggesting that the local climate and forage was, respectively, more
em relative is like comparing apples to oranges. It has been tradi- equable and available year round. Hulbert also suggested that P.
tionally thought that the increased cursorial prowess (both stamina leonensis had a potential longevity of about nine years (in contrast to
and speed) of advanced horses was as a coevolutionary and selective about four years for Hyracotherium; l\1acFadden, 1992). Van Valen
response to evading predators as part of the so-called "evolution- (1964) analyzed a large sample (ca. 150individuals) of Merychippus
ary arms race" (Janis and Wilhelm, 1993). Although there may be primus from Thomson Quarry, Nebraska. He found discrete wear
some truth in this, other equally plausible explanations have also classes representing seasonal cohorts. He also estimated that the in-
been offered. For example, it has been suggested that the evolu- tensity of selection was greatest in the juvenile cohorts, dropping in
tion of cursoriality was an adaptation to being able to migrate long middle age, and then increasing slightly in older individuals. Van
distances to exploit locally available, patchy food resources as do Valen estimated that the potential longevity of any given individual
certain modem zebras (Janis and Wilhelm, 1993). was about eleven years. Voorhies (1969) did a classic taphonomic
A corollary to interpreting the locomotory history of horses is study of the middle l\1iocene horse Protohippus cf. perditus from
that monodactyly is a more highly evolved (i.e., "better adapted") Verdigree Quarry, Nebraska. The age/wear cohorts in this sample
 Bruce J. MacFadden 556

are quite distinct, and Voorhies believed this to be a catastrophic during the middle Miocene supported the greatest diversity of fos-
accumulation of a population about two months before a peaked sil horses known from the history of the family Equidae. This high
breeding season (the first cohort represented unborn foals). He esti- level of diversity was maintained for almost 10 million years until
mated that the potential longevity of any given individual was about the late Miocene and early Pliocene when at most sites in North
12-15 years, which represents a significant increase over the stud- America there are about a half-dozen three-toed and one-toed hyp-
ies of Merychippus reported above but still significantly shorter than sodont species (the browsing anchitheres become extinct in the late
that of Equus (> 20 years in natural populations). Miocene). Traditionally this drop in diversity has been attributed
Thus we know that Miocene hypsodont horses such as Merychip- to a global aridification and the onset of lower productivity steppe
pus and Protohippus probably bred seasonally, migrated to exploit grasslands. Based on recent studies of carbon isotopes of fossil tooth
seasonally available resources and, according to Wang, Ceding, and enamel, we now also know that during the late Miocene there was
MacFadden (1994), were mixed feeders (browsers/grazers). All of a major global change in which C3-based savannas gave way to
these parameters suggest an social structure similar to extant equids more C4-based pure grasslands (e.g., Ceding, 1992; Ceding, Wang,
like the Burchell's ..:ebra in which there are harems and bachelor and Quade, 1993; Wang, Ceding, and MacFadden, 1994). It was
bands. If this is correct (as has been suggested by Klingel, 1975), on this latter system of global terrestrial ecosystems that Equus
then the antiquity of this social structure is at least 20 million years radiated and became the most widespread of any genus of fossil
old (unless Gingerich [1981] is correct, in which case it extends horse. During the late Pliocene and Pleistocene Equus dispersed
back until the early Eocene). throughout Holarctica and into South America and Africa (Lindsay,
Relatively little is known about the population biology/age struc- Opdyke, and Johnson, 1980). Its greatly reduced range (restricted
ture of fossil Equus. There are some excellent quarry samples from to parts of Eurasia and Africa) during the Holocene resulted from
which studies of this kind could be made (e.g., the enormous sample a large extinction wave (along with many other large-bodied mam-
of E. simplicidens from Hagerman, Idaho), and this certainly would mals) apparently caused by a combination of climate change and
be an attractive project for the future. overhunting by the human species that rapidly populated the North-
ern Hemisphere during this time.

COMMUNITY ECOLOGY
Early horses such as Hyracotherium are a relatively small compo- CONCLUDING REMARKS
nent of their terrestrial mammalian community (e.g., For more than 150 years, fossil horses have fascinated paleontolo-
Gingerich, 1989), and there is a slight increase in relative abundance gists and provided some of the most robust evidence for long-term
during the Oligocene (e.g., Mesohippus). By the middle Miocene, evolutionary change. In more recent years, fossil horses have also
however, horses are the dominant component of most terrestrial been the source of many studies that have attempted to understand
mammalian communities throughout North America. For example, the paleobiology and paleoecology of the Equidae. As future fos-
at the late Miocene Clarendonian Love Bone Bed of Florida, horses sil discoveries are made and new methodologies and techniques
comprise 59 percent of all of the ungulates collected at that site are developed, further paleontological advances both specific to
(MacFadden, 1992). Within these and other Miocene localities after equids and of more general interest will no doubt be made with
the explosive diversification between 18-15 million years ago, there this widespread and common group.
can potentially be as many as a dozen genera of equids coexisting.
How can phylogenetically closely related species apparently co-
exist in the same terrestrial communities? Not surprisingly, they ACKNOWLEDGMENTS
divide up their niche spaces in several different ways. The minority
of this diversity (usually two species in any given fauna) is repre- The author extends sincere thanks to Christine Janis for her immense
sented by the anchitheres (i.e., three-toed, medium to large-bodied, help with the locality records, careful scrutiny of various versions
low-crowned, presumed browsers). The majority, both in terms of of the manuscript, and for her patience and encouragement during
diversity and usual abundance, is represented by three-toed and one- the development of this chapter. Wendy Zomlefer and Linda Chan-
toed high-crowned forms of small, medium, and large body size. dler either prepared or assisted in the preparation of the figures.
Based on different dental morphologies, cropping mechanisms, and Earlier versions of this chapter were written during tenure of NSF
body sizes, these horses were apparently partitioning their food re- grant BSR 85-15003. This is University of Florida Contribution to
sources in order to minimize competition. Although traditionally Paleobiology no. 454.
it would have been assumed that all of these high-crowned forms
were almost pure grazers, new studies of enamel microwear (Hayek
et al., 1992) and carbon isotopes (Wang, Ceding, and MacFadden,
REFERENCES
1994; MacFadden and Ceding, 1994, 1996) suggest a spectrum
of feeding adaptations ranging from browsers/mixed feeders (but
Abusch-Siewert, S. (1983). Gebissmorphologische Untersuchungen en
probably not many pure browsers), to mixed feeders, to almost pure eurasiatischen Anchitherien (Equidae, Mammalia) unter besonder
grazers. Thus the advent of relatively rich savannas (although pre- Beriicksichtigung der Fundstelle Sandelzhausen. Courir Forschun-
dominantly C3 based; see Wang, Ceding, and MacFadden, 1994) ginstitut Senkenberg, 62,1-361.
 Equidae 557

Alberdi, M. T., & Prado, J. (1993). Review of the genus Hippidion Owen, northwestern Wyoming: composition and diversity in a rarely sam-
1869 (Mammalia: Perissodactyla) from the Pleistocene of South pled high-floodplain assemblage. University of Michigan Papers in
America. Zoological Journal of the Linnean Society, 108, 1-22. Paleontology, 28, 1-97.
Ameghino, F. (1904). Recherches de morphologie phylogenetique sur Jes (1991). Systematics and evolution of early Eocene Perissodactyla (Mam-
molaires superieures des ongules: Ana/es del Museo Nacional de malia) in the Clarks Fork Basin, Wyoming. Contributions of the Mu-
Historia Natural de Buenos Aires, 3rd Ser, v. 3, 1-541. seum of Paleontology, University of Michigan, 28, 181-213.
Azzaroli, A. (1988). On the equid generaDinohippus Quinn 1955 and Plio- Gray, J. E. (1821 ). On the natural arrangement of vertebrose animals. London
hippus Marsh 1874. Bollettino della Societa Paleontologica Italiana, Medical Repository Review, 15, 296-310.
27, 61-72. Haldane, J.B. S. (1949). Suggestions as to quantitative measurement of rates
Beer, G. de (1954). Archaeopteryx and evolution. Advances in Science, 42, of evolution. Evolution, 3, 51-6.
1-11. Hayek, L.-A. C., Bernor, R. L., Solounias, N., & Steigerwald, P. (1992).
Bennett, D. K. ( 1980). Stripes do not a zebra make, Part I: A cladistic analysis Preliminary studies of hipparionine horse diet as measured by tooth
of Equus. Systematic Zoology, 29, 272-87. microwear. Annals Zoologica Fennici, 28, 187-200.
Bernor, R. L., & Tobien, H. (1989). Two small species of Cremohippar- Hooker, J. J. (1980). The succession of Hyracotherium (Perissodactyla,
ion (Equidae, Mamm.) from Samos, Greece. Mitteilungen der Bay- Mammalia) in the English early Eocene. Bulletin British Museum
ererischen Stattssammlung Paliiontologie und Historische Geologie, (Natural History), Geology, 33, 101-14.
29, 207-26. (1989). Character polarities in early perissodactyls and their significance
Bode, F. D. (1933). Anchitheriine horses from the Merychippus zone of for Hyracotherium and infraordinal relationships. In The Evolution
the North Coalinga district, California. Publications of the Carnegie of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 79-101.
Institution of Washington, 440, 43-58. Oxford: Clarendon.
Bown, T. M., & Kihm, A. J. (1981). Xenicohippus, an unusual new hyra- (1994). The beginning of the equoid radiation. Zoologica/Journal of the
cothere (Mammalia, Perissodactyla) from lower Eocene rocks of Linnean Society, 112, 26-63.
Wyoming, Colorado, and New Mexico. Journal of Paleontology, 55, Hulbert, R. C., Jr. (1984). Paleoecology and population dynamics of the
257-70. early Miocene (Hemingfordian) horse Parahippus leonensis from
Camp, C. L., & Smith, N. (1942). Phylogeny and functions of the digital lig- the Thomas Farm site, Florida. Journal of Vertebrate Paleontology,
aments of the horse. University of California Memoirs, 13, 69-124. 4, 547-58.
Cerling, T. E. (1992). Development of grasslands and savannas in East Africa (1987a). Phylogenetic Systematics, Biochronology, and Paleobiology of
during the Neogene. Palaeogeography, Palaeoclimatology, Palaeoe- late Neogene horses (Family Equidae) of the Gulf Coastal Plain and
cology (Global and Planetary Change Section), 97, 241-47. the Great Plains. Unpublished Ph.D. Thesis. Gainesville: University
Cerling, T. E., Wang, Y., & Quade, J. (1993). Expansion ofC4 ecosystems as of Florida.
an indicator of global ecological change in the late Miocene. Nature, (1987b). Late Neogene Neohipparion (Mammalia, Equidae) from the
361, 344-5. Gulf Coastal Plain of Florida and Texas. Journal of Paleontology,
Clark, J. F., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedimen- 61, 809-30.
tation, stratigraphy, paleoecology, and paleoclimatology in the Big (1987c). A new Cormohipparion (Mammalia, Equidae) from the
Badlands of South Dakota. Fieldiana: Geology Memoirs, 5, 1-158. Pliocene (latest Hemphillian and Blancan) of Florida. Journal of
Cooper, C. F. (1932). The genus Hyracotherium. A revision and description Vertebrate Paleontology, 7, 451-68.
of new specimens found in England. Philosophical Transactions of ( 1988). Calippus and Protohippus (Mammalia, Perissodacty la, Equidae)
the Royal Society, London, 221, 431-48. from the Miocene (Barstovian-early Hemphillian of the Gulf Coastal
Cristo!, J. de. (1832). [No title: description of Hipparion.] Annales Science Plain). Bulletin of the Florida State Museum, Biological Sciences,
Industrie, Midi, France, I, 180-1. 32, 221-340.
Dalquest, W.W. (1981). Hesperhippus (Mammalia: Equidae) anew genus of (l 989a). Phylogenetic interrelationships and evolution of North
horse from the Hemphillian of North America, with the description American late Neogene Equidae. In The Evolution of Perissodactyls,
of a new species. Southwestern Naturalist, 25, 505-12. eds. D.R. Prothero & R. M. Schoch, pp. 176-96. Oxford: Clarendon.
Downs, T. (1956). The Mascall Fauna from the Miocene of Oregon. Univer- (1989b). Cormohipparion and Hipparion (Mammalia, Perissodactyla,
sity of California Publications in Geological Sciences, 31, 199-354. Equidae). Bulletin of the Florida State Museum, Biological Sciences,
Emry, R. J., Russell, L. S., & Bjork, P.R. (1987). The Chadronian, Orellan, 33, 229-338.
and Whitneyan North American Land Mammal Ages. In Cenozoic (1993). Late Miocene Nannippus (Mammalia: Perissodactyla) from
Mammals of North America, ed. M. 0. Woodburne, pp. 118-52. Florida, with a description of the smallest hipparionine horse. Jour-
Berkeley: University of California Press. nal of Vertebrate Paleontology, 13, 350-66.
Fischer, M. (1989). Hyracoids, the sister-group of perissodactyls. In The (1995). Equus from Leisey Shell Pit IA and other early Pleistocene
Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch, localities from Florida. Bulletin of the Florida Museum of Natural
pp. 37-56. Oxford Clarendon. History, 37 (part II), 553-602.
Forsten, A. M. (1971). Epihippusfrom the ViejaGroup, Trans-Pecos Texas. Hulbert, R. C., Jr., & MacFadden, B. J. (1991 ). Morphological transformation
The Pearce-Sellards Series, Texas Memorial Museum, 18, 1-5. and cladogenesis at the base of the adaptive radiation of Miocene
(1975). The fossil horses of the Texas Gulf Coastal Plain: a revision. hypsodont horses. American Museum of Natural History Novitates,
Pearce-Sellards Series, Texas Memorial Museum, 22, 1-86. 3000, 1-61.
Gidley, J. W. (1903). A new three-toed horse. American Museum of Natural Janis, C. M. (1982). Evolution of horns in ungulates: ecology and paleoe-
History Bulletin, 19, 465-76. cology. Biological Reviews, 57, 261-318.
(1906). A new genus of horse from the Mascall beds, with notes on (1989). Estimation of diets in fossil ungulate mammals. Journal of Ver-
the small collection of equine teeth in the University of California. tebrate Paleontology, 9 (suppl. to no. 3), p. 27A.
American Museum of Natural History Bulletin, 22, 385-8. Janis, C. M., Gordon, I., & Illius, A. (1994). Modelling equid/ruminant
Gingerich, P. D. ( 1981). Variation, sexual dimorph.ism, and social structure in competition in the fossil record. Historical Biology, 8, 15-29.
the early Eocene horse Hyracotherium (Mammalia, Perissodactyla). Janis, C. M., & Wilhelm, P. B. (1993). Were there mammalian pursuit preda-
Paleobiology, 7, 443-55. tors in the Tertiary? Dances with wolf avatars. Journal ofMammalian
(1989). New earliest Wasatchian mammalian fauna from the Eocene of Evolution, 1, 103-25.
 Bruce J. MacFadden 558

Kitts, D. B. (1956). American Hyracotherium (Perissodactyla, Equidae). Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 128-41.
American Museum of Natural History Bulletin, 110, 1-160. Oxford: Clarendon.
(1957). A revision of the genus Orohippus (Perissodactyla, Equidae ). (1992). Fossil Horses: Systematics, Paleobiology, and Evolution of the
American Museum of Natural History Novitates, 1864, 1-40. Family Equidae. Cambridge: Cambridge University Press.
Klingel, H. (1975). Social organization and reproduction in equids. Journal MacFadden, B. J., & Azzaroli, A. ( 1987). Cranium of Equus insulatus (Mam-
of Reproduction and Fertility, 23, 7-11. malia, Equidae) from the middle Pleistocene ofTarija, Bolivia. Jour-
Korth, W.W., & Evander, R. L. (1982). A new species of Orohippus (Peris- nal of Vertebrate Paleontology, 7, 325-34.
sodactyla, Equidae) from the early Eocene of Wyoming. Journal of MacFadden, B. J., & Cerling, T. E. (1994). Fossil horses, carbon isotopes
Vertebrate Paleontology, 2, 167-71. and global change. Trends in Ecology & Evolution, 9, 481-86.
Kowalevsky, V. ( 1873). Sur Anchitherium aurelianense Cuv. et sur L'histoire (1996). Mammalian herbivore communities, ancient feeding ecology,
paleontologique des chevaux. Memoirs L'Academie lmperiale Sci- and carbon isotopes: a 10 million-year sequence from the Neogene
ence St. Petersbourg, 20(3)(7th series), 1-73. of Florida. Journal of Vertebrate Paleontology, 103-5.
Lance, J. F. (1950). Paleontologia y estratigrafia de! Plioceno de Yepomera, MacFadden, B. J., & Hulbert, R. C. (1988). Explosive speciation at the base
Estado de Chihuahua - la parte: Equidos, excepto Neohipparion. of the adaptive radiation of Miocene grazing horses. Nature, 336,
Universidad Nat;ional Autonoma de Mexico, Instituto de Geologia 466-8.
Boletin, 54, 1-81. MacFadden, B. J., Johnson, N. M., & Opdyke, N. D. (1979). Magnetic po-
Leidy, J. (1850). Descriptions of Rhinoceros nebrascensis, Agriochoerus an- larity stratigraphy of the Mio-Pliocene mammal-bearing Big Sandy
tiquus, Palaeotherium proutii, and Palaeotherium bairdii. Proceed- Formation of western Arizona. Earth and Planetary Science Letters,
ings of the Academy of Natural Sciences, Philadelphia, 5, 121-2. 7, 11-22.
(1857). Notices of extinct Vertebrata discovered by F. V. Hayden, during MacFadden, B. J., & Skinner, M. F. (1979). Diversification and biogeography
the expedition to the Sioux Country under the command of Lieut. of the one-toed horses Onohippidium and Hippidion. Yale Peabody
G. K. Warren. Proceedings of the Academy of Natural Sciences, MuseumPostilla, 175, 1-9.
Philadelphia, 8, 311-12. McGrew, P. 0. (1938). The Burge Fauna, a lower Pliocene mammalian
(1858). Notice of remains of extinct Vertebrata, from the Valley of the assemblage from Nebraska. University of California Publications,
Niobrara River, collected during the exploring expedition of 1857, Bulletin of the Department of Geological Sciences, 24, 309-28.
in Nebraska, under the command of Lieut. G. K. Warren, U.S. Top. (1953). A new and primitive early Oligocene horse from Trans-Pecos
Eng, by Dr. F. V. Hayden, Geologist to the Expedition. Proceedings Texas. Fieldiana: Geology, 10, 167-71.
of the Academy of Natural Sciences, Philadelphia, I 0, 20-9. McKenna, M. C. (1975). Toward a phylogenetic classification of the Mam-
Lindsay, E. H., Opdyke, N. D., & Johnson, N. M. (1980). Pliocene dispersal malia. In Phylogeny of Primates, eds. W. P. Luckett & F. S. Szalay,
of the horse Equus and late Cenozoic mammalian dispersal events. pp. 21-46. New York: Plenum.
Nature, 287, 135-8. McKenna, M. C., Chow, M., & Ting, S., & Zhexi, L. (1989). Radinskya
Linnaeus, C. (1758). Systema Naturae per Regna tria Naturae, secundum yupingae, a perissodactyl-like mammal from the late Paleocene of
Classes, Ordines, Genera, Species, cum Characteribus, Dif.ferentiis, China. In The Evolution of Perissodactyls, eds. D.R. Prothero & R.
Synonymis, Locis, 10th ed., Stockholm. M. Schoch, pp. 24-36. Oxford: Clarendon.
Lund, P. W. (1845). Meddelelse afdet Udbytte de i 144 undersogte Knogle- Marsh, 0. C. (1871 ). Notice of some new fossil mammals from the Tertiary
huler have av give! ti! Kundskaben om Brasiliens Dryeverden for formation. American Journal of Science, 3, 35-44.
sidste Jordomvaeltning. Ebenda, Kobenhavn 1846, 12, 57-94. (1872). Preliminary description of new Tertiary mammals, Parts II, III,
MacFadden, B. J. (1976). Cladistic analysis of primitive equids, with notes and IV. American Journal of Science, Ser. 3, 4, 202-24.
on other perissodactyls. Systematic Zoology, 24, 1-14. (1874). Notice of new equine mammals from the Tertiary formation.
(1980a). Eocene perissodactyls from the type section of the Tepee Trail American Journal of Science, Ser. 3, 7, 249-50.
Formation of northwestern Wyoming. University of WYoming Con- (1875). Notice of new Tertiary mammals. American Journal of Science,
tributions to Geology, 18, 135-43. Ser. 3, 9, 239-50.
(1980b). The Miocene horse Hipparion from North America and from (1876). Notice of new Tertiary mammals, V. American Journal of Sci-
the type locality in southern France. Palaeontology, 23, 617-35. ence, Ser. 3, 12,401-4.
(1982). New species of primitive three-toed browsing horse from the (1878). The introduction and succession of vertebrate life in America.
Miocene phosphate mining district of central Florida. Florida Sci- Popular Science Monthly, 7, 513-27; 672-97.
entist, 45, 117-24. Matthew, W. D. (1926). The evolution of the horse: a record and its inter-
(1984a). Astrohippus and Dinohippus from the Yepomera local fauna pretation. Quarterly Review of Biology, 1, 139-85.
(Hemphillian, Mexico) and implications for the phylogeny of one- Matthew, W. D., & Stirton, R. A. (1930). Equidae from the Pliocene of
toed horses. Journal of Vertebrate Paleontology, 4, 273-83. Texas. University of California, Bulletin Department of Geological
(1984b). Systematics and phylogeny of Hipparion, Neohipparion, Sciences, 19, 349-96.
Nannippus, and Connohipparion (Mammalia, Equidae) from the Moreno, F. P. (1891). Onohippidium munizi, Breve noticia sobre los restos
Miocene and Pliocene of the New World. American Museum of Nat- f6siles de un genero nuevo de la familia de los Equidae conservados
ural History Bulletin, 179, 1-196. en el Museo de La Plata. Revista Museo La Plata, 2, 65-71.
(1987). Fossil horses from "Eohippus" (Hyracotherium) to Equus: scal- Osborn, H.F. (1915). [No title; description of Kalobatippus gen. nov.]. In
ing, Cope's law, and the evolution of body size. Paleobiology, 12, Hitherto Unpublished Plates of Tertiary Mammalia and Pennian
355-69. Vertebrata, eds. E. D. Cope & W. D. Matthew, plate CVIII. American
(l 988a). Horses, the fossil record, and evolution - A current perspective. Museum of Natural History Monographs, Ser. 2.
In Evolutionary Biology (Vol. 22), eds. M. K. Hecht, B. Wallace, & (1918). Equidae of the Oligocene, Miocene, and Pliocene of North Amer-
G. T. Prance, pp. 131-58. New York: Plenum. ica. Iconographic type revision. American Museum of Natural His-
(1988b). Fossil horses from "Eohippus" (Hyracotherium) to Equus, 2: tory Memoirs, 2, 1-326.
Rates of dental evolution revisited. Biological Journal ofthe Linnean Owen, R. (1840). Description of the fossil remains of a mammal, a bird,
Society, 35, 37-48. and a serpent, from the London clay. Proceedings of the Geological
(1989). Dental character variation in paleopopulations and morphos- Society of London, 3, 162-6.
pecies of fossil horses and extant analogues. In The Evolution of (1858). Description of asmall lophiodont mammal (Pliolophus vulpiceps
 Equidae 559.

Owen), from the London Clay, near Harwich. Quarterly Journal of (1953). The Major Features of Evolution. New York: Columbia Univer-
the Geological Society of London, 14, 54-71. sity Press.
(1869). On fossil teeth of equines from Central and South America. Skinner, M. F. (1972). Order Perissodactyla. In Early Pleistocene Pre-glacial
Proceedings of the Royal Society, London, 17, 267-8. and Glacial Rocks and Faunas ofNorth-central Nebraska, eds. M. F.
Peterson, 0. A. (1931). New species from the Oligocene of the Uinta. Car- Skinner & C. Hibbard. American Museum of Natural History Bul-
regie Museum Annals, 21, 61-78. letin, 148, 1-148.
Prothero, D. R., Manning, E. M., & Fischer, M. (1988). The phylogeny Skinner, M. F., & MacFadden, B. J. (1977). Connohipparion n. gen. (Mam-
of ungulates. In The Phylogeny and Classification of the Tetrapods, malia: Equidae) from the North American Miocene (Barstovian-
Vol. 2: Mammals, ed. M. E. Benton, pp. 201-34. Oxford: Clarendon. Clarendonian). Journal of Paleontology, 51, 912-26.
Prothero, D.R., & Schoch, R. M. (1989). Origin and evolution of the Peris- Skinner, M. F., & Taylor, B. E. (1967). A revision of the geology and pale-
sodactyla. In The Evolution of Perissodactyls, eds. D.R. Prothero & ontology of the Bijou Hills, South Dakota. American Museum Novi-
R. M. Schoch, pp. 504-29. Oxford: Clarendon. tates, 2300, 1-53.
Prothero, D. R., & Shubin, N. (1989). The evolution of mid-Oligocene Stanley, S.M. (1979). Macroevolution: Pattern and Process. San Francisco:
horses. In The Evolution of Perissodactyls, eds. D. R. Prothero & Freeman.
R. M. Schoch, pp. 142-75. Oxford: Clarendon. Stirton, R. A. (1940). Phylogeny of North American Equidae. University of
Quinn, J. H. (1955). Miocene Equidae of the Texas Gulf Coastal Plain. California Publications, Bulletin of the Department of Geological
University of Texas Bureau of Economic Geology, Publication no. Sciences, 25, 165-98.
5116, 1-102. Tedford, R.H., & Alf, R. M. (1962). A new Megahippus from the Barstow
Radinsky, L. R. (1966). The adaptive radiation of the phenacodontid condy- Formation San Bernardino County, California. Bulletin of the South-
larths and the origin of the Perissodactyla. Evolution, 20, 408-17. ern California Academy of Sciences, 61, 113-22.
( 1984). Ontogeny and phylogeny in horse skull evolution. Evolution, 38, Van Valen, L. (1964). Age in two fossil horse populations. Acta Zoologica,
1-15. 45, 93-106.
Renders, E. (1984). The gait of Hipparion sp. from fossil footprints in Lae- Voorhies, M. R. (1969). Taphonomy and population dynamics of an early
toli, Tanzania. Nature, 308, 179-81. Pliocene vertebrate fauna, Knox County, Nebraska. Contributions to
Richey, K. A. (1948). Lower Pliocene horses from Black Hawk Ranch Mount Geology, University oflfyoming, Special Paper, 1, 1-69.
Diablo, California. University of California Publications, Bulletin Wang, Y., Cerling, T. E., & MacFadden, B. J. (1994). Fossil horses and
Department of Geological Sciences, 28, 1-44. carbon isotopes: New evidence for Cenozoic dietary, habitat, and
Robinson, P. (1966). Fossil Mammalia of the Huerfano Formation, Eocene, ecosystem changes in North America. Palaeogeography, Palaeocli-
of Colorado. Yale Peabody Museum Bulletin, 21, 1-95. matology, Palaeoecology, 107, 269-80.
Schlaijker, E. M. (1935). Contributions to the stratigraphy and paleon- Webb, S. D. (1969). The Burge and Minnechaduza Clarendonian mam-
tology of the Goshen Hole area, Wyoming. IV. New vertebrates malian faunas of north-central Nebraska. University of California
and the stratigraphy of the Oligocene and early Miocene. Bulletin Publications in Geological Sciences, 78, 1-191.
of the Museum of Comparative Zoology, Harvard University, 76, Webb, S. D., & Hulbert, R. C., Jr. (1986). Systematics and evolution of
97-189. Pseudhipparion (Mammalia, Equidae) from the late Neogene of the
(1937). A study of Parahippus wyomingensis and a discussion of the Gulf Coastal Plain and the Great Plains. In Vertebrates, Phylogeny,
phylogeny of the genus. Bulletin of the Museum of Comparative and Philosophy, eds. K. M. Flanagan & J. A. Lillegraven, pp. 237-72.
Zoology, Harvard University, 80, 255-83. Laramie: University of Wyoming Contributions to Geology.
Scott, W. B. (1893). The mammals of the Deep River beds. American Nat- Wible, J. R. (1986). Transformations in the extracranial course of the inter-
uralist, 27, 659-62. nal carotid artery in mammalian phylogeny. Journal of Vertebrate
Sefve, I. (1912). Die fossilen Pferde Siideramerikas. Svenska Vetenskap- Paleontology, 6, 313-25.
sakademiens Handlingar, 48, 1-185, 3 plates. Winans, M. C. (1985). Revision of North American fossil species of the
Shotwell, J. A. (1961). Late Tertiary biogeography of horses in the northern genus Equus (Mammalia: Perissodactyla: Equidae). Unpublished
Great Basin. Journal of Paleontology, 35, 203-17. Ph.D. Thesis. Austin: University of Texas.
Simpson, G. G. ( 1945). The principles of classification and a classification of (1989). A quantitative study of the North American fossil species of the
mammals. American Museum ofNatural History Bulletin, 85, 1-350. genus Equus. In The Evolution of Perissodactyls, eds. D.R. Prothero
(1951). Horses: The Story of the Horse Family in the Modern World and & R. M. Schoch, pp. 262-97. Oxford: Clarendon.
Through Sixty Million Years of History. Oxford: Oxford University Zhegallo, V. I. (1978). Hipparioni Zentralnoi Asii (the hipparions of cen-
Press. tral Asia). Trudy sovneztnoi Sovetsko-Mongolskoi paleont. ekspedizii
(1952). Notes on British hyracotheres. Journal of the Linnean Society of (Transactions Joint Soviet-Mongolian Paleontological Expedition),
London, Zoology, 2, 195-206. 7, 1-156.
38 Chalicotherioidea
MARGERY C. COOMBS

INTRODUCTION

The Chalicotherioidea were clawed perissodactyls that first appeared

in the Eocene and became extinct in the Pleistocene. The group was
never particularly numerous or diverse and is, in many cases, known
only from scattered and fragmentary fossils. Nonetheless, several
deposits, most notably those of the Upper Harrison Formation at
the Agate Spring Quarries in western Nebraska, have yielded good
quantities of complete skeletal material.
Chalicotheres were primarily a Eurasian group, as suggested by
greater taxonomic diversity in Eurasia and a continuous fossil record
from the Eocene into the Pleistocene. The first African record is
from the Miocene. The fossil record in North America is discontin-
uous. Primitive chalicotheres are known from North America in the
Eocene but evidently became extinct by the late Eocene. There is
then a hiatus in which no North American chalicotheres are known
until the latest Oligocene or early Miocene. In the Arikareean, other,
more derived chalicotheres (Figures 38.l, 38.2) referable to the
family Chalicotheriidae, subfamily Schizotheriinae, reached North
America, where they survived into the late Miocene. It is unlikely
that chalicotheres ever reached South America.

DEFINING FEATURES OF THE

Figure 38.1. A restoration of social interaction in the North American
SUPERFAMILY CHALICOTHERIOIDEA
schizotheriine chalicotheriid Tylocephalonyx, by Brian Regal.

Chalicotheres are defined dentally by upper molars having a com-

plete metaloph with no metaconule and an incomplete protoloph The most prominent postcranial character of chalicotheres is the
in which the paraconule (= protoconule) is retained (see Figure clawed ungual phalanges (Figure 38.2A, D), which are pointed and
38.2E). A W-shaped ectoloph occurs in most chalicotheres but is fissured at their tips. All chalicotheres are assumed to have had
not yet developed in the earliest representatives. Although some claws, although this point has only recently been confirmed for any
chalicotheres increased the surface area and durability of the cheek of the Eocene chalicotheres (Lucas, pers. comm., based on Grange-
teeth by elongating the molars, increasing crown height, and de- ria canina). In many chalicotheres, derived specializations of the
veloping larger premolars, chalicotheres as a group have relatively forelimb and manus allow the claws to be used as hooks during
low-crowned cheek teeth and little molarized premolars. They ap- browsing and to keep them clear of the ground during walking.
pear to have been browsers on leaves, twigs, and other nonresistant The combination of claws with a browsing dentition has made
vegetation. Anterior teeth of chalicotheres show a wide range of chalicotheres one of the more famous members of the odd hetero-
variation (Coombs, 1978a). geneous group termed clawed herbivores (see Coombs, 1983, for a
560
 Chalicotherioidea 561

functional discussion and comparison of various large clawed her-

bivores).

SYSTEMATICS

A
SUPRA-SUPERFAMILY
The superfamily Chalicotherioidea is one of the major subdivisions
within the order Perissodactyla, but the exact relationships to other
perissodactyls are still not completely clear. Some authors (for ex-
ample, Simpson, 1945) have allied Chalicotherioidea with Equoidea
and Brontotherioidea on the basis of the W-shaped ectoloph on upper
molars. This grouping, termed the Hippomorpha, is almost certainly
invalid, and the W-shaped ectoloph probably evolved independently
in the three groups. Borissiak (1946) listed a number of characters
shared by chalicotheres and brontotheres, but most or all of these
B characters are either convergent or primitive for perissodactyls. Scott
(1941) and Radinsky ( 1964) considered chalicotheres to be an in-
dependent perissodactyl lineage without special relationship to any
particular group. They placed the Chalicotherioidea in a separate
perissodactyl suborder, the Ancylopoda. Hooker (1984) suggested
that the strong, uninterrupted metaloph of chalicotheres was a de-
rived character linking them with tapiroids and rhinocerotoids. He
- ... --... joined another perissodactyl group, the Lophiodontidae, with the
Chalicotherioidea under the name Ancylopoda. Most previous au-
' thors had placed the Lophiodontidae in the Tapiroidea, although the
' group has no clear affinities with true tapirs (Tapiridae). Dental sim-
'
c ilarities between chalicotheres and lophiodontids are underscored
by controversy over the proper placement of two Eocene genera,
Paleomoropus and Lophiaspis (considered to be chalicotheres by
Radinsky, 1964, and Hooker, 1986, and lophiodontids by Fischer,
1964, 1977), and by the fact that one of the early named species of
the well-known chalicothere genus Moropus, M. oregonensis, was
originally referred to the genus Lophiodon (Leidy, 1873). Hooker's
D concept of the Ancylopoda has been generally accepted in recent
years. A grouping of the Isectolophidae, Ancylopoda, and Cerato-
morpha was termed Tapiromorpha by Hooker (1984, 1989) and
Moropomorpha by Schoch (1984, without Isectolophidae; 1989a,
1989b), Prothero and Schoch ( 1989), and Colbert and Schoch (this
volume, Chapter 39). (See also summary chapter on perissodactyls
and proboscideans, this volume, Chapter 35.)
E Almost all of the preceding comparisons have been based on den-
tal evidence. Comparisons of skull and postcranial elements of early
perissodactyls are very few and, in the case of chalicotheres, made
especially difficult by the dearth of postcranial elements known for
Eocene representatives.
Figure 38.2. Representative material of North American Miocene
schizotheriine Chalicotheriidae. A. Skeleton of Moropus elatus, as recon-
structed by Osborn, 1919 (scale bar= 1 m). B. Skull of Moropus elatus, INFRA-SUPERFAMILY
after Coombs, 1978a (scale bar = 10 cm). C. Outline of domed skull of Ty-
locephalonyx skinneri, UCMP 115867 (from locality CP54B), with dotted Romer ( 1966) listed two families of the Chalicotherioidea, the Eo-
line representing the dome outline ofNMNH 187129 (from locality WM7), moropidae and the Chalicotheriidae. This classification scheme orig-
after Munthe and Coombs, 1979 (scale bar= 10 cm). D. Side view ofungual
phalanx (claw) of digit II of the manus of M. elatus, CM 1604, from Holland
inated with Matthew ( 1929), who separated the Eocene chalicotheres
and Peterson, 1914 (scale bar= 10 cm). E. Left cheek teeth P2-M3 of M. (= Eomoropidae) from later forms(= Chalicotheriidae) in which
elatus, AMNH 11322 (scale bar= 5 cm). claws were well known.
 Margery C. Coombs 562

erium Schizotherlum Borissiakia MOROPUS TYLOCEPHALONYX Ancylotherium

Figure 38.3. A hypothesis of relationships within the Chalicotherioidea. Genera that appear in North America are capitalized. Derived characters and the
groups they define include (1) CHALICOTHERIOIDEA: Upper molars with complete metaloph and incomplete protoloph; clawed ungual phalanges(?).
(2) Mesostyle developed. (3) Grangeria: Robust mandible and canines. (4) CHALICOTHERIIDAE: Emphasis of ectoloph over transverse lophs; reduced
parastyle and protoloph and more posterior protocone; reduced m3 hypoconulid; loss of Pl; metatarsals shorter than metacarpals; reduction of astragalus
neck; modified metacarpals, metatarsals, and phalanges in association with claws. (5) CHALICOTHERIINAE: Extreme shortening of tibia and pes; Mc
and Mt IV much longer than Mc and Mt III, respectively. (6) SCHIZOTHERIINAE: Longer molars, with upper molar ectolophs forming a longer, taller
W; specialized claw retraction mechanism. (7) Strongly elongated molars; loss of m3 hypoconulid. (8) Presence of duplex bone (Fig. 38.2D) on digit II
of manus. (9) Moropus: Mt IV subequal in length to Mt III. (10) Tylocephalonyx: Domed skull; reduced hypocone on M3; reduced lower incisors. (11)
Ancylotherium: Crochet on upper molars (also in Phyllotillon, not shown in diagram), reduced hypocone on M3; reduced lower incisors; strongly modified
carpals and metacarpals for extreme wrist flexion and strong dorsal extension of the claws.

The Eomoropidae is clearly a paraphyletic group of chalicotheres. W, especially the part between the parastyle and mesostyle. In con-
Analysis of the relationships of the included genera to each other and trast, the postcranials of members of the Schizotheriinae are rela-
to the Chalicotheriidae is still incomplete, although one hypothesis tively conservative, with much less disproportion between forelimbs
is given in Figure 38.3. The term "Eocene chalicothere" replaces and hindlimbs than occurs in the Chalicotheriinae. Nonetheless,
the use of Eomoropidae in the present chapter. members of the Schizotheriinae for whom adequate skeletal material
The Chalicotheriidae, first known in the fossil record in the early is known have fairly long limbs and necks, suggesting high browsing
Oligocene of Eurasia, is a monophyletic group defined on the basis of habits that could be improved by rearing up on the hindlimbs. The
a number of derived dental and postcranial characters (Figure 38.3 ). latest surviving schizotheriine genus, Ancylotherium, from the late
Especially notable are the emphasis of the ectoloph on upper molars, Miocene to Plio-Pleistocene of the Old World, had very robust fore-
the shortening of the pes compared to the man us, and the modifica- limbs that could probably hook down large branches. Schizotheriine
tion of metacarpals, metatarsals, and phalanges in association with chalicotheres also developed a specialized claw retraction mecha-
more specialized claws. Based on the analysis of Coombs ( 1983), the nism by which the claws could be pulled dorsally free of the ground
disproportionate-sized manus and pes in the Chalicotheriidae sug- while the animal was walking. Two genera of the Schizotheriinae
gest that bipedal browsing occurred in all members of this family. occur in North America during the Miocene and are treated further
The Chalicotheriidae is subdivided into two subfamilies, the Chal- later.
icotheriinae and the Schizotheriinae. The Chalicotheriinae retain rel-
atively short, low-crowned molars but have highly modified limbs.
The tibia and pes are extremely shortened, and both Mc (metacarpal) INCLUDED NORTH AMERICAN GENERA IN THE
IV and Mt (metatarsal) IV are considerably longer than Mc III and SUPERFAMILY CHALICOTHERIOIDEA
Mt III, respectively. Body proportions of members of the Chali-
cotheriinae were rather gorillalike (see Zapfe, 1979). The Chali- The locality numbers listed for each genus refer to the list of unified
cotheriinae contains only two genera, Chalicotherium and the very localities in Appendix I. Question marks in front of the locality (e.g.,
closely related, probably synonymous genus Nestoritherium. The ?CPlOl) mean the taxon is questionably known from that locality.
subfamily is primarily Eurasian, with one early Miocene species The acronyms for museum collections are listed in Appendix III.
known from Africa. No member of the Chalicotheriinae is known
to have reached North America, and the subfamily is therefore not
EOCENE CHALICOTHERES
considered any further in this chapter.
The Schizotheriinae includes chalicotheriids with more elon- Radinsky (1964) recognized five genera of Eocene chalicotheres:
gated, higher crowned molar teeth than occur in the Chalicotheri- Paleomoropus, Lophiaspis, Eomoropus, Grangeria, and Litolophus.
inae. In particular, the upper molar ectolophs form a taller, longer Paleomoropus is the only one of the five genera known only from
 Chalicotherioidea 563

North America. Lophiaspis is known only from Europe and Litolo- lophs and weaker ectolophs, with some development of
phus only from Asia. Grangeria and Eomoropus are both broadly a mesostyle on the ectoloph (more than in Litolophus).
distributed genera with representatives in both Asia and North Amer- Paraconule and protocone slightly displaced posteriorly
ica. BothLophiaspis and Paleomoropus have controversial affinities. on the protoloph. Derived relative to all Eocene chali-
They have been placed in the Lophiodontidae by some authors (Fis- cotheres in having proportionately large canines, shorter
cher, 1964, 1977; Prothero and Schoch, 1989; Schoch, 1989b) and and deeper cranial rostrum, shorter and deeper mandibular
in the Chalicotherioidea by others (Hooker, 1986, 1989; Gingerich, symphysis and mandibular ramus, and larger, more tren-
1991). Proper placement in the Lophiodontidae or Chalicotherioidea chant p2. Premolar row longer relative to molar row, and
depends on determination of dental characters that can distinguish lower molar metalophids less oblique than in Litolophus.
early representatives of the two groups. Paleomoropus, still known For additional dental characters, see Radinsky (1964) and
only from PU 13254, three separate upper molars, is treated with Lucas and Schoch (1989). Lucas (pers. comm.) has studied
the Lophiodontidae by Colbert and Schoch in the present volume postcranials of Grangeria canina; these are quite similar
(Chapter 39). Gingerich (1991) noted problems concerning the type to those of Litolophus and do not show strong modifica-
locality of Paleomoropus and suggested a medial Wasatchian age tions of the manus or pes for claw use. Grangeria is the
for the specimen. only Eocene chalicothere for which ungual phalanges are
known, and these are clawed (Lucas and Schoch, 1989).
Eomoropus Osborn, 1913 Average length of m2: 19.0 mm (based on three specimens
Type species: Eomoropus amarorum (Cope, 1881). ranging from 16.9-21.9 mm).
Type specimen: AMNH 5096. Included species: G. canina (known only from the Eocene of
Characteristics: A small chalicothere (smaller than Grange- China); G. anarsius (known from localities CP6A, CP29D).
ria and Litolophus) with low-crowned teeth. Upper molars
Comments: Grangeria is known from much more extensive
with transverse lophs better developed than ectoloph but
material from China than from North America. The North
with definite development of a mesostyle (mesostyle much
American species, G. anarsius, was originally referred to
more clearly developed than in Litolophus). Mandible long
Eomoropus by Gazin (1956). Radinsky (1964) and Lucas
and slender with a long shallow symphysis, low ascend-
and Schoch (1989) have given good reasons for assigning
ing ramus, and unenlarged lower canines (primitive in
this species to Grangeria. G. anarsius is slightly smaller
mandibular morphology compared to Grangeria). Premo-
and has somewhat smaller upper molar mesostyles than G.
lar row longer relative to molar row, and less oblique mo-
canina.
lar metalophids than in Litolophus; pl smaller but more
molariform than that of Grangeria. For additional dental The robust canines and lower jaws of Grangeria resemble those
comparisons with Litolophus and Grangeria, see Lucas seen in the later chalicotheriid genus Chalicotherium but proba-
and Schoch (1989). Although ungual phalanges are not bly evolved autapomorphically. Postcranial characters of Grangeria
known for Eomoropus, the type specimen of E. amarorum show no evidence of relationship to Chalicotherium (Lucas, pers.
preserves much of the manus and some of the pes. Neither comm.).
shows the special modifications for claw use that occur in
chalicotheriids. The man us has four unelongated toes, and Other Eocene Chalicotheres
the astragalus has a long neck. Upper cheek teeth of a small chalicothere have been collected from
Average length of m2: 16.0 mm (based on five specimens the Tepee Trail Formation (locality CP36B). This specimen has not
with range 14.3-18.1 mm). yet been described.
Included species: E. amarorum (known from localities CP6A,
CP38C); E. quadridentatus (known from the Eocene of CHALICOTHERIIDAE
China).
Of the two subfamilies of Chalicotheriidae, only the Schizo-
Comments: The two species of Eomoropus are quite similar theriinae are known to have reached North America. The subfam-
in size and morphology despite the large geographic dis- ily Schizotheriinae includes the following genera (see Figure 38.3):
tance between them. E. amarorum includes E. annectens Schizotherium, Borissiakia, Phyllotillon, Moropus, Tylocephalonyx,
(see arguments by Radinsky, 1964), and E. quadridenta- and Ancylotherium. Of these, only Moropus and Tylocephalonyx
tus includes E. minimus (see arguments by Radinsky, 1964, reached North America. Moropus, also known from the early
and Lucas and Schoch, 1989). Miocene of Europe (Coombs, 1974), is first known to occur in North
America in the Arikareean. Tylocephalonyx, known only from North
Grangeria Zdansky, 1930 America, appears in the fossil record somewhat later (Figure 38.4),
Type species: Grangeria canina Zdansky, 1930. and it is not altogether clear whether it is more closely related to
Type specimen: PMUM 3018. Moropus or to the Old World genus Ancylotherium. If it is closer
Characteristics: A small chalicothere (but larger than Eo- to Ancylotherium, it could represent a second, slightly later, chali-
moropus). Upper molars with well-developed transverse cothere immigrant than Moropus.
 Margery C. Coombs 564

Moropus Marsh, 1877 pes to form a duplex is rare in M. elatus (only about 10
Type species: Moropus distans Marsh, 1877. percent of known cases). M. distans is one of the smallest
Lectotype specimen: YPM 12193a. and probably one of the earliest known representatives of
Characteristics: A medium- to large-sized chalicothere. The Moropus in North America. Moropus oregonensis (Leidy,
most distinctive derived character is the subequal length 1873) is also relevant to any discussion of M. distans. Ma-
of Mt III and Mt IV (in most schizotheriine chalicotheres, terial of M. oregonensis, also probably from the John Day
Mt III remains distinctly longer). This length relationship Formation, consists on! y ofupper cheek teeth, which there-
is reflected in the presence of an articulation between Mt fore cannot readily be compared with the postcranial ele-
IV and the ectocuneiform. Other characters of Moropus ments of M. distans. M. oregonensis differs from M. elatus
are discussed and compared with those of other members in smaller size and more primitive morphology, such as less
of the Schizotheriinae elsewhere (Coombs, 1978b, 1979, elongation of the molars and the closeness of the labial
1989) and are summarized in Figure 38.3. Skeletal ele- origin of the metaloph to the mesostyle. M. distans and
ments of early species of Moropus are extensively figured M. oregonensis may represent the same species, although
by Coombs (1978b). the teeth of M. oregonensis are slightly larger than those
Average length of m2: 43.0 mm (average for fifteen speci- that might be expected to go with the postcranials of M.
mens of Moropus elatus = 44 mm, with range for these distans (sexual dimorphism might explain this difference,
specimens= 40.6-48.0 mm; some, less well-known species however).
of Moropus are smaller, with known m2 length as low as Moropuselatusis by farthe best known species of Moro-
34.0 mm). pus and probably is the best known of all the chalicotheres.
Included species: M. distans (known from locality PN6); M. Numerous specimens from the Agate Spring Quarries
oregonensis (locality PN6); M. elatus (localities CP103, (Nebraska; locality CP104A), Morava Ranch Quarry
CP104A, CP104B); M. hollandi (localities CP53, (Nebraska; locality CP104B), and Harper Quarry
CP104C); M. matthewi (locality ?CP75B); M. merriami (Wyoming; locality CP104A) are very close morpholog-
(localities NB 17, NB 18, NB20A, NB22A, NB23B, CPl 10, ically to one another and to the type material described
CPl 11). by Marsh (1877; lectotype of M. elatus is YPM 13081).
M. elatus has the largest average size of any species of
Comments: Fragmentary material of Moropus from a variety Moropus; size sexual dimorphism is readily documented
oflocalities of various ages is referable to Moropus sp. This in this species (Coombs, 1975). The known geographic
material cannot readily be assigned to established species distribution of M. elatus is, however, quite limited. One
but is insufficient to serve as the basis of new species. Some problem relevant to M. elatus is the status of material
of the unresolved material comes from the Runningwater from Florida (locality GC8A) referred by Frailey (1979)
Formation and equivalents of Nebraska (localities CP105, and from Texas (locality GC5) referred by Albright (pers.
CPl 06), the Batesland Formation of South Dakota (locality comm.) and Manning (1990) to Moropus sp. These spec-
CP88), and from several localities in California (localities imens are quite similar to M. elatus in morphology but
CCl 7C, CC21A). These materials are of Hemingfordian are much smaller in size. Moropus hollandi (holotype CM
age. Another problematic set of material is that represent- 1424; locality CP104C) is morphologically quite similar
ing the latest occurrences of chalicotheres in North Amer- to M. elatus and is slightly later in age. It differs from
ica (localities CP114A, CP114C, CP115B). all other species of Moropus in lacking a trapezium in
Marsh (1877) named three original species of Moropus: the manus. Metatarsals of M. hollandi are proportionately
M. distans, M. senex, and M. elatus. M. senex is consid- shorter than those of M. elatus, although longer than those
ered a nomen dubium for reasons cited in Coombs (l 978b, of M. matthewi and M. merriami. It can be difficult to dif-
p. 47). Holland and Peterson (1914) designated M. dis- ferentiate between M. elatus and M. hollandi when only
tans, collected from Oregon, probably from the John Day dental material is available, as in the case of AMNH 13765
Formation (locality PN6), as the type species of Moropus. (locality CP51B or CP52). The preceding species of Moro-
This choice was not particularly felicitous because the lec- pus have all been discussed in detail by Coombs (1978b).
totype of this species is a fused (but broken) proximal and Specimens of Moropus from Oregon (locality PN6),
medial phalanx(= duplex) of digit II of the pes. This spec- Florida (locality GC8A), and Nebraska (locality CP103),
imen clearly belongs to a schizotheriine chalicotheriid but all considered to be oflate Arikareean (early early Miocene)
is marginally diagnostic at the generic level, much less age, are the earliest clearly confirmed occurrences of this
so than the better material of M. elatus, which Marsh de- genus in North America. However, among collections at
scribed at the same time. M. distans can be distinguished Amherst College made by Loomis from the Muddy Creek
from M. elatus on the basis of much smaller size, on certain area of eastern Wyoming (locality CP50) are chalicothere
details of the morphology of a cuboid bone collected with phalanges extremely similar in morphology and size to
the lectotype, and on the greater flatness of the lectotype those of M. elatus. The Muddy Creek fauna is generally
duplex bone. Furthermore, the fusion of phalanges of the considered to be late early Arikareean (late late Oligocene),
Chalicotherioidea 565

which, if correct, would represent the earliest known ma- tion (locality CP114C). This metacarpal differs from those
terial of Moropus in North America. However, caution is known for other species of Moropus. It might be refer-
advisable until the collections from Muddy Creek are fully able to Tylocephalonyx, and it is clearly not modified like
studied. metacarpals of the medial Miocene to early Pleistocene Old
Moropus matthewi Holland and Peterson (1914) is a World schizotheriine genus Ancylotherium. Other frag-
rather enigmatic species. Locality information for the type mentary specimens suggest that schizotheriine chalico-
material is inexact. All the material referred by Holland theriids survived into at least the Clarendonian (late
and Peterson to M. matthewi comes from northeastern Col- Miocene) of North America.
orado, but it was collected in different years (1898 and
1901 AMNH expeditions) from different localities. Gal- Tylocephalonyx Coombs, 1979
breath (1953) suggested that the specimens came from the Type species: Tylocephalonyx skinneri Coombs, 1979.
Hemingfordian (late early Miocene) Martin Canyon local Type specimen: F:AM 54915.
fauna (locality CP71), but Skinner (1968) found the evi- Characteristics: A large schizotheriine chalicotheriid, approx-
dence for this unconvincing. The specimens could as easily imately the same size as Moropus elatus. The most notable
have come from the early Barstovian (early late Miocene) character of Tylocephalonyx is the domed skull, formed
Eubanks local fauna (locality CP75B) (as might be implied by the dorsal expansion of the frontal and parietal bones
from the original collecting data) or a later local fauna. (Figures 38.1, 38.2C). Other characters of Tylocephalonyx
Material of M. matthewi includes a variety of elements (a compared with those of other members of the Schizotheri-
magnum, single upper molar, a Mt II, two Mt III speci- inae are summarized in Figure 38.3. Known metatarsal
mens, a scaphoid, phalanges, a vertebra, and fragments of proportions of Tylocephalonyx are intermediate between
an ilium, ischium, and femur). The metatarsals are inter- those of Moropus elatus and M. merriami, fairly simi-
mediate in length/width proportions between those of M. lar to those of M. matthewi. Fusion of the proximal and
elatus and M. merriami. Some aspects of the morphology medial phalanges of the manus to form a duplex bone oc-
of M. matthewi are similar to Tylocephalonyx skinneri, but curs throughout Tylocephalonyx; similar fusion for digit II
reference to Tylocephalonyx is not warranted on the basis of the pes is common but not universal.
of known material. M. matthewi is clearly not referable Average length of m2: 45.0 mm. (based on nine specimens
to Chalicotherium, despite a suggestion to this effect by from the Sheep Creek Formation of Nebraska, locality
Matthew (1929; he suggested reference to Macrotherium, CP108, with range 41.3-48.7 mm).
now subsumed in Chalicotherium). At present, Moropus Included species: T. skinneri only (known from localities
matthewi is known only from the ?early Barstovian of CP54B, CP106, CP108A, CP108B, CP109A).
northeastern Colorado (?CP75B), but it differs from early Additional material belonging to this genus is referred
Barstovian chalicotheres (probably Moropus merriami) to Tylocephalonyx sp. (localities WM7, CPlO, NP34D or
from nearby Nebraska. Earlier chalicotheres that clearly E, NP38E) or ?Tylocephalonyx sp. (locality CP72B).
come from the Martin Canyon local fauna of northeastern
Comments: Tylocephalonyx was figured, described in de-
Colorado (represented by F:AM 54913; locality CP7 l) are
tail, and compared with Moropus and other schizotheriine
probably not referable to M. matthewi but closely resemble
chalicotheriids by Coombs (1979). Possible functions of
material of M oropus sp. from the Runningwater Formation
the domed skull were explored by Munthe and Coombs
of Nebraska (locality CP105).
(1979), who concluded that the dome most likely func-
Moropus merriami is based on a variety of material from
tioned in display and low-impact butting during intraspe-
the early Barstovian of Nevada. Chalicothere material from
cific social interactions (see Figure 38.1).
the early Barstovian Lower Snake Creek fauna of Nebraska
(locality CPl 10) is quite similar and can probably also be Other Miocene chalicotheres
referred to this species. M. merriami has proportionately Several Miocene specimens that have been identified as schizo-
the shortest and widest of any metatarsals that can be re- theriine chalicotheriids cannot be allied with certainty at this time
ferred to Moropus. All known specimens of proximal and with a given genus (localities CC16, CC17B, NB19A, NB25A,
medial phalanges of digit II of the pes are fused to form PN8B, PN9A, PN18, NC2). These specimens are either fragmen-
a duplex in M. merriami. In this respect M. merriami re- tary, not yet described, or unfigured and currently not located. Most
sembles the lectotype of M. distans and specimens referred are probably referable to Moropus.
to Moropus sp. from the Runningwater and other early to
medial Hemingfordian faunas of Nebraska. M. merriami
is the latest species of Moropus for which good material is BIOLOGY AND EVOLUTIONARY PATTERNS
available.
After the early Barstovian, chalicothere specimens are The history of the Chalicotherioidea in North America is actually
extremely rare in North America. The best such specimen two separate histories, one occurring in the Eocene and one in the
is the Mc III noted by Skinner from the Valentine Forma- Miocene.
 Margery C. Coombs 566

w
z L L.BI.
1.8
w E.BI
2.4
0 4.5
0 E Lat.Hp.
5.2
:::::i L.Hp.
6.0
£L. L.E.Hp.
7.0

L
E.E.Hp.

L.CI.
...
9.5

E.CI.
11.0
w
z L.L.Ba.
w 12.5
0 M
E.L.Ba.
0 14.0

:i E.Ba.

5.8
L.Hm,
17.5
E.Hm.
18.B
E L.L.Ar.
19.2

E.L.Ar.
23.0

w L.E.Ar.
z
w L
0 27.7
0
Cl E.E.Ar.
:::::i 29.4
0
Wt.
E 31.9
Or.
33.4
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

!l.
w
L.Un.
39.5

41.3
Q.
e -qif
-th
·~

z ec:
w
0
0
w
E.Un.

L.Br.
45.9 11 -
(!)
-
M 46.7

M.Br.
PRIMITIVE CHALICOTHERES CHALICOTHERIDAE
49.0 "EOMOROPIDAE"
E.Br.
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.

E.E.Wa.
55.0
55.5
-
Ck.3 55.6
w Ck.2
Ck.1
55.B
z Tl.&
56.D
56.2
w L Tl.6
Tl.4
57.0
0 Tl.3
57.3
0 Tl.2
58.5
w TI.1
..J
<( To.3
To.2
£L.
To.1
E Pu.3
63.8
64.0
Pu.2 64.5
Pu.1
65.0 -
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphllllan, Cl.= Clarendonian, Ba.= Barstovlan, Hm.= ~emingfordian: Ar. ".' Arikareean, .
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronlan, Du.= Duchesnean, Un. = Umtan, Br. = Br1dgerian, Wa. = Wasatch1an,
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.

Figure 38.4. Temporal ranges of North American chalicothere genera.

 Chalicotherioidea 567

The known Eocene radiation in North America is limited to two Both Moropus and Tylocephalonyx were, on the basis of dentition,
species, Eomoropus amarorum and Grangeria anarsius. Both have browsers on vegetation such as leaves and twigs. As noted by Osborn
related species in Asia referred to the same genera. Eocen.e chali- (1919, for M. elatus), body proportions were somewhat similar to
cotheres were small and had low-crowned, unelongated upper mo- those of the living okapi (Okapia, Giraffidae ), with longer forelimbs
lars in which the transverse lophs were dominant over the ectoloph. than hindlimbs and a somewhat elongated neck. These attributes
Recent evidence suggests that claws were present, at least in Grange- suggest specialized browsing on high vegetation, a habit probably
ria, but the feet of Eocene chalicotheres were not particularly mod- enhanced by the ability to rear up on the hindlimbs (see discussion
ified for claw bearing and resemble those of other perissodactyls. in Coombs, 1983).
Chalicotheres are not known from North America from the Chad- The rarity of chalicothere fossils raises questions about the ecol-
ronian through Whitneyan (late Eocene-Oligocene) time interval. ogy of these animals. A low rate of fossilization can be associated
Cope (1891) did name Chalicotherium bilobatum, based on a frag- with any or all of the following: (I) low absolute numbers, (2) pref-
mentary, edentulous left lower jaw ramus, from a supposed Chadro- erence for environments where fossilization is less likely (such as
nian fauna in the Cypress Hills of Saskatchewan (locality NPlOB). forests), and (3) solitary habits. Absolute numbers of chalicotheres
This material was later placed in a new genus Oreinotherium by are difficult, if not impossible, to estimate. The existence of size
Russell ( 1934). More recently Oreinotherium was mentioned again sexual dimorphism (Coombs, 1975) and morphological structures
by Storer and Bryant (1993) in the context ofbiostratigraphic mix- such as the domed skulls of Tylocephalonyx (Munthe and Coombs,
ing of the early Cypress Hills collections, which actually include 1979) suggest agonistic behavior in some sort of social setting. Oc-
Chadronian through Hemingfordian fauna! elements. Nonetheless, casional cases of chalicothere mass concentrations, such as those
examination of the very poor material of Oreinotherium suggests at the Agate Spring Quarries (locality CP104A) and Morava Ranch
that it does not resemble any known chalicothere and probably be- Quarry (locality CP104B), suggest that some species (in this case
longs to a brontothere (see also Skinner, 1968). It seems reasonable M. elatus) frequented particular spots or grouped together at least
to conclude that chalicotheres became extinct in North America at some times of year. My own opinion is that, although absolute
before the late Eocene and that the Miocene schizotheriine chali- numbers, solitary habits, and habitats might all play a part in the rar-
cotheriids are new immigrants from Eurasia. ity of chalicothere fossils, the primary factor was lack of a suitable
There are two genera of Miocene schizotheriine chalicotheres in preserved ancient habitat. The main reason for this is the relative
North America: Moropus and Tylocephalonyx. Moropus is the ear- abundance of chalicotheres in assemblages where they do occur.
lier of these to appear in the fossil record, with definite occurrences After the early Barstovian, chalicotheres were extremely rare in
in the late Arikareean (early early Miocene) and possibly somewhat North America. It is not at present clear whether both Moropus
earlier. Tylocephalonyx first appears in the fossil record in the mid and Tylocephalonyx survived the early Barstovian, or, if not, which
to late Hemingfordian (late early Miocene). It is not clear whether lineage persisted. Despite the probable similarity of habits of chali-
the later appearance of Tylocephalonyx represents a new immigra- cotheres to those of certain ground sloths, it is unlikely that the influx
tion from Eurasia or is an artifact of the lack of representation of its of ground sloths at about 7-6 million years ago had anything to do
biotope in the hitherto sampled earlier fossil record. with chalicothere disappearance in North America. Chalicotheres
As discussed by Coombs (1979), Moropus and Tylocephalonyx, had become increasingly rare and may have been extinct in North
even though contemporaneous in North America through the late America before any ground sloths arrived from South America.
Hemingfordian and into the Barstovian, are rarely, if ever, entombed
in the same deposits. For example, in Nebraska Moropus is found
in the Harrison, Upper Harrison, and Runningwater formations (lo- REFERENCES
calities CPI03, CPI04, CPI05, CP106) but is replaced by Tylo-
cephalonyx in the Sheep Creek Formation (locality CPI08). In the Borissiak, A. A. (1946). (A new chalicothere from the Tertiary of Kaza-
Olcott Formation (locality CPI 10), Moropus (probably M. merriami) khstan.) Akademia Nauk SSSR. Trudy Paleontologicheskovo lnsti-
is present; Tylocephalonyx is known from an early Barstovian fauna tuta, 13, 1-134.
from Montana (locality NP38E). These separate distributions corre- Coombs, M. C. (1974). Ein Vertreter von Moropus aus dem europiiischen
Aquitanien und eine Zusammenfassung der europliischen pos-
late with those of certain other mammals such as the erinaceid insec- toligozlinen Schizotheriinae (Mammalia, Perissodactyla, Chali-
tivore Brachyerix (see Rich and Rich, 1971), although comparative cotheriidae ). Sitzungsberichte Osterreichische Akademie der Wis-
information about a variety of taxa is meager. Some evidence in- senschaften, Mathematisch-naturwissenschaftlichen Klasse, 182,
dicates that Tylocephalonyx was associated with mixed mesophytic 273-88.
forests dominated by deciduous hardwoods and conifers and with a (1975). Sexual dimorphism in chalicotheres (Mammalia, Perisso-
dactyla). Systematic Zoology, 24, 55-62.
temperate climate. Such forests were concentrated in the northwest- (1978a). A premaxilla of Moropus elatus Marsh, and evolution of chal-
ern United States but may have expanded or contracted at various icotherioid anterior dentition. Journal of Paleontology, 52, 118-21.
times. Moropus seems to have occupied somewhat drier environ- (1978b). Reevaluation of early Miocene North American Moropus
ments with more open country. Thus the occurrences of these genera (Perissodactyla, Chalicotheriidae, Schizotheriinae ). Bulletin of the
Carnegie Museum, 4, 1-62.
at particular times and places can be considered within the frame-
(1979). Tylocephalonyx, a new genus of North American dome-skulled
work of our developing understanding of ecological changes in the chalicotheres (Mammalia, Perissodactyla). Bulletin of the American
Miocene of North America. Museum of Natural History, 164, 1-64.
 Margery C. Coombs 568

(1983). Large mammalian clawed herbivores: a comparative study. Marsh, 0. C. (1877). Notice of some new vertebrate fossils. American Jour-
Transactions of the American Philosophical Society, 73, 1-96. nal of Science, ser. 3, 14, 249-56.
(1989). Interrelationships and diversity in the Chalicotheriidae. In The Matthew, W. D. (1929). Critical observations upon Siwalik mammals. Bul-
Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch, letin of the American Museum of Natural History, 56, 437-560.
pp. 438-57. Oxford: Clarendon. Munthe, J., & Coombs, M. C. (1979). Miocene dome-skulled chalicotheres
Cope, E. D. (1881). The systematic arrangement of the order Perisso- (Mammalia, Perissodactyla) from the western United States: a pre-
dactyla. Proceedings of the American Philosophical Society, 19, liminary discussion of a bizarre structure. Journal of Paleontology,
377-403. 53, 77-91.
(1891). On Vertebrata from the Tertiary and Cretaceous rocks of the Osborn, H.F. (1913). Eomoropus, an American Eocene chalicothere. Bul-
North West Territory, I: The species from the Oligocene or lower letin of the American Museum of Natural History, 32, 261-74.
Miocene beds of the Cypress Hills. Geological Survey of Canada, (1919). Seventeen skeletons of Moropus: probable habits of this animal.
Contributions to Canadian Paleontology, 3, 1-25. Proceedings of the National Academy of Science, 5, 250-2.
Fischer, K.-H. (1964). Die tapiroiden Perissodactylen aus der eozanen Prothero, D.R., & Schoch, R. M. (1989). Origin and evolution of the Perisso-
Braunkohle des Geisaltales. Geologie, Berlin, 45, 1-101. dactyla: Summary and synthesis/Classification of the Perissodactyla.
(1977). Neue Funde von Rhinocerolophiodon (n. gen.), Lophiodon, In The Evolution of Perissodactyls, eds. D. R. Prothero & R. M.
und Hyrachyus {Ceratomorpha, Perissodactyla, Mammalia) aus dem Schoch, pp. 504-29, 530-7. Oxford: Clarendon.
Eozan des Geiseltals bei Halle (DDR). Zeitschrift far Geologische Radinsky, L. B. (1964). Paleomoropus, a new early Eocene chalicothere,
Wissenscha.ften ( Gesellscha.ft far Geologische Wissenscha.ften der and a revision of Eocene chalicotheres. American Museum Novitates,
DDR), Berlin, 5, 909-19. 2179, 1-28.
Frailey, D. (1979). The large mammals of the Buda local fauna (Arikareean: Rich, T. H. V., & Rich, P. V. (1971). Brachyerix, a Miocene hedgehog from
Alachua County, Florida). Bulletin of the Florida State Museum, western North America, with a description of the tympanic regions of
Biological Sciences, 24, 123-73. Paraechinus and Podogymnura. American Museum Novitates, 2477,
Galbreath, E. C. (1953). A contribution to the Tertiary geology and paleontol- 1-58.
ogy of northeastern Colorado. University of Kansas Paleontological Romer, A. S. (1966). Vertebrate Paleontology (3rd ed.). Chicago: University
Contributions, 4, 1-120. of Chicago Press.
Gazin, C. L. (1956). The geology and vertebrate paleontology of the up- Russell, L. S. (1934). Revision of the lower Oligocene vertebrate fauna of the
per Eocene strata in the northeastern part of the Wind River Basin, Cypress Hills, Saskatchewan. Transactions of the Royal Canadian
Wyoming, Part 2: The mammalian fauna of the Badwater area. Smith- Institute, 20, 49-65.
sonian Miscellaneous Collections, 131, 1-35. Schoch, R. M. (1984). Two unusual specimens of Helaletes in the Yale
Gingerich, P. D. (1991 ). Systematics and evolution of early Eocene Perisso- Peabody Museum collections, and some comments on the ances-
dactyla (Mammalia) in the Clarks Fork Basin, Wyoming. Contribu- try of the Tapiridae (Perissodactyla, Mammalia). Postilla, Peabody
tions from the Museum of Paleontology, University of Michigan, 28, Museum of Natural History, Yale University, 193, 1-20.
181-213. ( l 989a). A brief historical review of perissodactyl classification. In The
Holland, W. J., & Peterson, 0. A. (1914). The osteology of the Chali- Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch,
cotheroidea with special reference to a mounted skeleton of Moropus pp. 13-23. Oxford: Clarendon.
elatus Marsh, now installed in the Carnegie Museum. Memoirs of the (1989b). A review of the tapiroids. In The Evolution of Perissodactyls,
Carnegie Museum, 3, 189-406. eds. D.R. Prothero & R. M. Schoch, pp. 298-320. Oxford: Claren-
Hooker, J. J. (1984). A primitive ceratomorph (Perissodactyla, Mammalia) don.
from the early Tertiary of Europe. Zoological Journal of the Linnean Scott, W. B. (1941). The mammalian fauna of the White River Oligocene,
Society, 82, 229-44. Part V: Perissodactyla. Transactions of the American Philosophical
( 1986). Mammals from the Bartonian (middle/late Eocene) of the Hamp- Society, new series, 28, 747-980.
shire Basin, southern England. Bulletin of the British Museum (Nat- Simpson, G. G. (1945). The principles of classification and a classification
ural History) Geology Series, 39, 191-478. of mammals. Bulletin of the American Museum of Natural History,
(1989). Character polarities in early perissodactyls and their significance 85, 1-350.
for Hyracotherium and infraordinal relationships. In The Evolution Skinner, M. F. ( 1968). A Pliocene chalicothere from Nebraska, and the distri-
of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 79-101. bution of chalicotheres in the late Tertiary of North America. Amer-
Oxford: Clarendon. ican Museum Novitates, 2346, 1-24.
Leidy, J. (1873). Contributions to the extinct vertebrate fauna of the Western Storer, J. E., & Bryant, H. N. (1993). Biostratigraphy of the Cypress Hills
Territories. Report of the U.S. Geological Survey of the Territories, Formation (Eocene to Miocene), Saskatchewan: equid types (Mam-
1, 1-358. malia: Perissodactyla) and associated fauna! assemblages. Journal
Lucas, S. G., & Schoch, R. M. (1989). Taxonomy and biochronology of of Paleontology, 67, 660-9.
Eomoropus and Grangeria, Eocene chalicotheres from the western Zapfe, H. (1979). Chalicotherium grande (Blainv.) aus der mioziinen Spal-
United States and China. In The Evolution of Perissodactyls, eds. D. tenfiillung von Neudorf an der March (Devinska Nova Yes), Tsche-
R. Prothero & R. M. Schoch, pp. 422-37. Oxford: Clarendon. choslowakei. Neue Denkschriften des Naturhistorischen Museums in
Manning, E. M. (1990). The late early Miocene Sabine River. In Trans- Wien, 2, 1-282.
actions of the Gulf Coast Association of Geological Societies, eds. Zdansky, 0. (1930). Die alttertiiiren Siiugetiere Chinas nebst stratigraphis-
E. Kinsland & T. Cagle, 40, 531-49. chen Bemerkungen. Paleontologia Sinica, (C), 6, 5-87.
39 Tapiroidea and other moropomorphs

M. W. COLBERT and R. M. SCHOCH

INTRODUCTION

The tapiroid lineage and the rhinocerotoid lineage diverged from

a common ancestor in the early Eocene (approximately 50 Ma).
Living tapirs are characterized by a stout body, short legs, and a large
tapered head that is carried close to the ground. Their thick skin and
streamlined shape are ideal for moving through dense underbrush.
Tapirs are nocturnal, having a keen sense of hearing and smell, but
relatively poor eyesight. The mesaxonic forefoot has four digits (the
fifth smaller than the rest, the first lost) and the hindfoot bears three
digits. Because they retain a large number of primitive features,
extant tapirs are often considered "living fossils" (e.g., Janis, 1984).
In contrast to their generally "primitive" morphology, extant tapirs
have a short, fleshy, prehensile proboscis, composed of muscles of
Figure 39.1. Restoration of Helaletes, by Brian Regal.
the snout and upper lip. This feature's evolutionary inception traces
back to at least the Bridgerian (middle Eocene) Helaletes, and its
modification through time is one of the keys to understanding tapir is provisionally considered the most basally diverging lineage of
evolution. the rhinocerotoids (see Radinsky, 1966b; Prothero, Manning, and
The traditionally defined Tapiroidea is a paraphyletic group be- Hanson, 1986; Emry, 1989; Schoch, l 989b, for discussion of this
cause it does not contain the descendent rhinoceroses (see Schoch, problem), it is described here. Toxotherium is also presented here,
1989a, 1989b). This traditional paraphyletic concept of the although considered a rhinocerotoid. This chapter is organized into
Tapiroidea refers to those perissodactyls having complete cross discussion of four intemested clades: the Moropomorpha, the Cer-
lophs on the upper and lower molars, but having short ectolophs and atomorpha, the Tapiroidea, and the Tapiridae, shown as succes-
brachydont teeth (Radinsky, 1963). In order to recognize a mono- sive nodes on the cladogram in Figure 39.5. Linnaean ranks follow
phyletic (sensu Hennig, 1966) Tapiroidea, we follow the suggestions Prothero and Schoch (1989b).
of Prothero and Schoch (1989a, 1989b ), Schoch ( 1989a, 1989b), and
Hooker (1989) and restrict and redefine the Tapiroidea as the sister
taxon of the Rhinocerotoidea. The name Moropomorpha (Schoch, DIAGNOSTIC FEATURES OF
1984a, 1989b; Prothero and Schoch 1989a, 1989b) is now applied THE MOROPOMORPHS
to the more inclusive group including most taxa traditionally con-
sidered tapiroids, but additionally including the Rhinocerotoidea. The Moropomorpha are diagnosed by the development of distinct
The Moropomorpha as used here also includes the chalicotheres, shearing upper molar metalophs and lower molar hypolophids,
following the analyses of Hooker ( 1984, 1989) and Schoch ( 1984a, both advancements toward bilophodonty. Additionally, the moropo-
1989b). morphs have a transversely oriented lophoid postprotocrista (met-
This chapter deals only with North American moropomorphs ex- aloph) joining the ectolophs on P3 and P4. This character is modified
clusive of the Rhinocerotoidea and the Chalicotherioidea, which in those later moropomorphs that develop molariform premolars
are investigated elsewhere in this volume. Although Hyrachyus (i.e., well-developed cross lophs on the premolars).

569
 M. W. Colbert and R. M. Schoch 570

~--
Figure 39.4. Skeleton of Helaletes nanus. Modified from Radinsky (1965,
Fig. 23), Troxell (1922c, Figure 3), and descriptions of Schoch (1984a). Bar
Figure 39.2. Skull of Protapirus validus, partially restored (modified from
=5cm.
Hatcher, 1896). Scale bar= 5 cm.

the polytomy among the Ceratomorpha, Hippomorpha, and Ancy-

A lopoda.
All three classificatory schemes just reviewed regard the tapiroids
and rhinocerotoids as being closely related, with the rhinocero-
toids considered derived from tapiroids. Additionally, none of these
B schemes regard the chalicotheres as particularly closely related to
tapiroids or rhinocerotoids.
Several workers have independently suggested that perhaps chal-
c icotheres and ceratomorphs share a close relationship. In 1964,
Radinsky described Paleomoropus and classified it as an early eo-
moropoid chalicothere, although Fischer (1964, 1977) regarded Pa-
leomoropus as a lophiodont tapiroid (see also Coombs, 1982).
D Schoch (1983) suggested that the Ceratomorpha and chalicotheres
might be sister groups, and later (Schoch, 1984a) coined the name
Moropomorpha for this grouping; in this scheme the Moropomorpha
and Hippomorpha (composed of the Brontotherioidea and Equoidea)
E
were regarded as sister groups (Schoch, 1989a, 1989b). Indepen-
dently, Hooker (1984) linked chalicotheres and lophiodonts in the
Figure 39.3. Representative upper dentitions of tapiroids and other moropo- Ancylopoda, making this the sister group of all other ceratomorphs
morphs. Key: A. Tapirus (based on MCZ 6037). B. Miotapirus (based on except the Isectolophidae. To name his concept of ceratomorphs
MCZ 2949). C. Protapirus (based on PU 10899). D. Hyrachyus (based on plus ancylopods plus isectolophids, Hooker (1984) resurrected and
AMNH 12362). E. Heptodon (based on AMNH 14884). (Drawing by Brian
redefined the old name Tapiromorpha, which is essentially identical
Regal.)
to Schoch's (1984a) concept of the Moropomorpha.
In summary, the moropomorphs are here considered to be the
SYSTEMATICS sister taxon to the Hippomorpha and include the Rhinocerotoidea,
the Ancylopoda, and various other taxa including the redefined
Tapiroidea.
INFRAORDER: THE CONCEPT OF
THE MOROPOMORPHA
RELATIONSHIPS OF GENERA WITHIN
Wood (1937; adopted by Simpson, 1945a) united the Tapiroidea
THE MOROPOMORPHA
and Rhinocerotoidea in the Ceratomorpha and aligned the Cerato-
morpha with the Hippomorpha (composed of the Equoidea, Bron- Figure 39.5 shows the relationships of genera of the North Ameri-
totherioidea, and Chalicotherioidea) within the Perissodactyla. Scott can moropomorphs discussed here (see also Schoch, l 989b; Hooker,
(1941) united the Tapiroidea and the Rhinocerotoidea in the Cer- 1989). Assessment of the polarities of character transformations as-
atomorpha, restricted the Hippomorpha to the Brontotherioidea and sumes equoids as the first outgroup and brontotheres as the
Equoidea, and recognized a distinct Ancylopoda (Cope, 1889) for second outgroup (see Maddison, Donoghue, and Maddison, 1986,
the Chalicotherioidea. Scott (1941) united his Ceratomorpha and for methodology of outgroup comparison). Characters were taken
Hippomorpha in the Chelopoda. Radinsky (1964) adopted essen- primarily from Hooker (1989). Swofford's 1989 PAUP program
tially the same classification as Scott (1941), but did not resolve was used for tree determination, and Maddison and Maddison's
 Tapiroidea and other moropomorphs 571

\ \
\ \
\ \
\ \
\ \
\ \
\
\
\
\
\
\ \
,,,,
,,
~ 4 CERATOMORPHA

'
Figure 39.5. Interrelationships of moropomorph perissodactyls. Key to characters at nodes: (I) MOROPOMORPHA: Distinct shearing upper molar met-
aloph and lower molar hypolophid; transversely oriented lophoid metaloph joining the ectoloph on P3 and P4. (2) ISECTOLOPHIDAE: postcanine diastema
reduced, M3 trapezoidal in outline (metacone reduced). (3) UNNAMED TAXON: Metacone somewhat flattened. (4) CERATOMORPHA: Paraconules lost
on P4 and upper molars, with the concomitant development of a complete shearing loph; loss of metastylids and reduction of metalophids on the lower
molars; parall~l sided to divergent hypoconid and entoconid; anterposteriorly oriented metaloph. (5) RHINOCEROTOIDEA: m3 hypoconulid lost; lingually
creased P4 protocone; enlarged metalophids and paralophids (characters from Prothero, Manning, and Hanson, 1986). (6) TAPIROIDEA: enlarged narial
incision. (7) UNNAMED TAXON: Upper canine size less than incisors. (8) UNNAMED TAXON: upper molar metacones convex; m3 hypoconulid lost.
(9) TAPIRIDAE: Groove developed for expanded nasal diverticulare posteriorly from the narial incision. (10) UNNAMED TAXON: Skull "telescoped"
rearward.

MacClade version 2.1 program was used to investigate character lsectolophus, presumably occurring in parallel) with the concomi-
evolution. Because the results of this analysis are biased by the tant development of a complete shearing loph and the reduction of
exclusion of Eurasian taxa, a conservative classification scheme is the metalophids. In addition, the ceratomorphs are characterized by
adopted here to create a more defensible phylogenetic hypothesis, the presence of a postcanine diastema (Schoch, 1989b), parallel-
with poorly supported clades "collapsed" into polytomies. sided to divergent lower molar protoconid and metaconid and par-
In agreement with Hooker (1989), the Isectolophidae (node 2) is allel sided to divergent hypoconid and entoconid (Hooker, 1989),
found to be the most basally diverging member of the Moropomor- and an anteroposteriorly oriented metaloph (Hooker, 1989). Hep-
pha; its monophyly is weakly supported by a reduction of the post- todon forms an unresolved polytomy with the Tapiroidea and the
canine diastema. This arrangement suggests that the development of Rhinocerotoidea at the base of the Ceratomorpha.
unbroken shearing cross lophs on the upper molars in lsectolophus The Rhinocerotoidea (node 5) has been diagnosed by Prothero,
is convergent with the Ceratomorpha. Manning, and Hanson (1986) by the loss of m3 hypoconulid, posses-
Paleomoropus and Schizatheriodes are the next diverging genera sion of a lingually creased P4 protocone, enlarged metalophids and
within the Moropomorpha, but do not share unambiguous synapo- paralophids, increased hypsodonty, and a cylindrical odontoid pro-
morphies uniting them into a clade. If Paleomoropus is a chali- cess on the atlas (see Emry, 1989, for a critique of these characters).
cothere, then this group would be linked with the moropomorphs at Hyrachyus is apparently the sister taxon of at least some rhinocero-
this level. We include these taxa as incertae sedis at the level of the toids (node 5; also see Emry, 1989; Hooker, 1989; Prothero, Man-
Moropomorpha. ning, and Hanson, 1986; Radinsky, 1963, 1965, 1966b) and is best
Our concept of the Ceratomorpha is seen at node 4 on the clado- referred to the Rhinocerotoidea, although discussed in this
gram, here diagnosed by the loss of the paraconules (also seen in chapter.
 M. W. Colbert and R. M. Schoch 572

Our analysis places Toxotherium within the Rhinocerotoidea, as posterior migration of the nasal bones, and an increase in vertical
have most recent studies (e.g., Skinner and Gooris, 1966; Radinsky, height of the posterior portion of the skull. These modifications are
1969; Emry, 1979). Radinsky (1969) suggests that Toxotherium is probably associated with both the further development of the pro-
an aberrant amynodont, with an unusual buccal expansion of its boscis and the realignment of jaw musculature concomitant with
parastyle. The exact taxonomic position of Toxotherium is problem- premolar molarization, and might serve as a useful diagnosis for the
atical owing to the lack of preserved cranial and postcranial features Tapirinae.
and its possession of many dental autapomorphies. Toxotherium is The following discussion provides characterizations of various
placed incertae sedis within the Rhinocerotoidea. genera of North American moropomorphs, exclusive of most rhino-
The Tapiroidea is now defined as all taxa more closely related cerotoids and ancylopods, which are treated elsewhere in this
to Tapirus than to Rhinoceros, diagnosed by their possession of an volume.
enlarged narial incision. Helaletes is hypothesized to be the most
basally diverging tapiroid, and its narial incision morphology (see
Radinsky, 1963) is suggested to be representative of the ancestral INCLUDED GENERA OF NORTH
tapiroid condition. Note that the old concept of the "Helaletidae" AMERICAN MOROPOMORPHS
is disregarded here because it is clearly paraphyletic (Emry, 1989;
Hooker, 1989; Schoch, 1989b). Recently collected cranial material The locality numbers listed for each genus refer to the list of unified
referable to Heteraletes under study by Alan Tabrum of the Carnegie localities in Appendix I. The locality numbers may be listed in a
Museum shows that its narial incision is similar to that of Helaletes, couple of alternative ways. The acronyms for museum collections
indicating its inclusion within the Tapiroidea. Colodon, also with a are listed in Appendix III.
narial incision, shows a further development in proboscis evolution Parentheses around the locality (e.g., [CPlOl]) mean the taxon
by its reduced nasal bones. in question at that locality is cited as an "aff." or "cf." the taxon in
The relationships of Desmatotherium, Selenaletes, Dilophodon, question. Parentheses are usually used for individual species, thus
and Fouchia are ambiguous, largely because these taxa are poorly implying the genus firmly known from the locality, but the actual
represented by diagnostic fossil material (see Emry, 1989; Radinsky, species identification may be questionable. Question marks in front
1963, 1966a; Schoch, 1984a). Although it is likely that most of of the locality (e.g., ?CPlOl) mean the taxon is questionably known
these forms are either members of the Tapiroidea or possibly the from that locality, thus implying some doubt that the tax on is actually
Tapiridae, we conservatively place these forms as Ceratomorpha present at that locality, either at the genus or the species level.
incertae sedis.
The loss of the hypoconulid on m3 within the Ceratomorpha is
ISECTOLOPHIDAE
a somewhat problematical character (see Emry, 1989). Helaletes
possesses a well-developed hypoconulid, but the feature is vari- Characteristics: A possibly paraphyletic grade taxon of small to
ably developed both in Colodon (Radinsky, 1963), and in Pro- medium-sized cursorial moropomorphs characterized by the lack of
tapirus from the Clarno Formation (B. Hanson, pers. comm.). The any major postcanine diastemata, the possession of a full placental
hypoconulid is reduced or lost in Heteraletes, Desmatotherium, Se- dentition, and the lack of an enlarged narial incision. The premo-
lenaletes, Dilophodon, Fouchia, the Tapiridae, and (presumably in- lars are not molarized. The molars have cross lophs combined with
dependently) in the rhinocerotoids. Caution should be exercised short ectolophs. The paracones and metacones on Ml-3 tend to be
when using this feature to diagnose major groups. subequal and convex. M3 tends to be trapezoidal in outline, with
The Tapiridae are here diagnosed based on further evolution of the metaloph as long as protoloph. The metalophids on ml-3 tend to
narial region, primarily the development of grooves extending pos- be well developed; m3 has a large hypoconulid.
teriorly from the narial incision hypothesized to house expansions
ofi;:artilaginous nasal diverticulae (see Gregory, 1920). Tapirids are Homogalax Hay, 1899 (synonym: Systemodon)
further characterized by having relatively convex and labially placed Type species: Homogalax protapirinus (Wortman, 1896)
metacones on the upper molars and by the lack of a hypoconulid (= Systemodon primaevus Wortman, 1896).
onm3. Type specimen: AMNH 144.
Protapirus and new undescribed species from Uintan aged de- Characteristics: Small, cursorial forms with completely non-
posits in southern California form an unresolved polytomy at the molariform premolars (P3 always has only one lingual
base of the Tapiridae. Discovery of cranial material showing the cusp); Ml-3 sometimes bear visible (but poorly devel-
tapirid condition of the narial incision in the new species from Cali- oped) protoconules and metaconules; the cusps on Pl, pl,
fornia is surprising because of the dental resemblance of these forms p2-3 are high and sharp, and metastylids are present on
to Dilophodon minusculus from the Western Interior. This similar- ml-3. Recent fossil finds demonstrate that Homogalax has
ity suggests that Dilophodon might be a tapirid, but discovery of distinctive, apomorphous, "doglike" incisors (described by
cranial material of D. minusculus is needed to test this hypothesis. Bakker, Cooke, and Schain, MS). Homogalax can be dis-
Miotapirus, Tapiravus, and Tapirus share a derived condition tinguished from lsectolophus by its possession of weak ·
within the Tapiridae of a backward "telescoping" of the skull that paraconules and metaconules on Ml-3. Homogalax also
involves the relative reduction in length of the frontal bones, the possesses small diastemata that occur variably between P2,
 Tapiroidea and other moropomorphs 573

Pl, Cl, and I3, although the taxonomic usefulness of these Zetes) latidens, Osborn, in Scott and Osborn, 1889; = Pari-
variable features is questionable (Radinsky, 1963). sectolophus latidens, Peterson, 1919; = Homogalax
Average length of m2: 9.2-11.8 mm. bridgerensis, Troxell, l 922a) (localities NB 14, CP34C,
Included species: H. protapirinus (= S. primaevus Wort- CP34D, CP38A, CP38B); I. radinskyi Schoch, 1983 (lo-
man, 1896; = S. protapirinum Wortman, 1896; = Ho- calities CP34C, CP34D).
mogalax primaevus, Hay, 1899) (known from localities lsectolophus sp. is also known from locality CP6B.
SB24,SB25A,CP20A,CP25A,CP2 5B,CP25C,CP27A,
CP27D, CP64A); H. aureus Gingerich, 1991 (=Homo-
MOROPOMORPHA INCERTAE SEDIS
galax cf. H. protapirinus, Radinsky, 1963) (locality NP49);
Homogalax small sp. (localities CP64A, CP64B); Homo- Paleomoropus Radinsky, 1964
galax new small species (locality CP20A [Bakker, Cooke, Type species: Paleomoropusjepseni Radinsky, 1964.
and Schain, MS]). Type specimen: PU 13254 (only known specimen).
Homogalax sp. is also known from localities (SB24), Characteristics: Large, anterolabially displaced parastyle; no
?CP25F, (CP63). mesostyle; metacone lingually displaced and slightly con-
vex labially; protoloph with a lophoid protoconule; lingual
Cardiolophus Gingerich, 1991 (synonyms: Systemodon, half of protoloph straight, protocone not deflected posteri-
in part; Homogalax, in part; Hyracotherium, in part) orly; metaloph high with no metaconule, terminating high
Type species: Cardiolophus radinskyi Gingerich, 1991. at the ectoloph; basal cingula on anterior, lingual, posterior,
Type specimen: UM 78915. and posterolabial sides; M3 larger with larger and more
Characteristics: This primitive moropomorph was recently labially displaced parastyle and a slightly shorter meta-
accorded its own genus, having previously been referred cone than Ml and M2. A large wear facet on the posterior
to Homogalax (Radinsky, 1963). lingual side of the metaloph indicates that m3 had a large
hypoconulid (Radinsky, 1964).
Cardiolophus differs from Homogalax and other isectolo- Average length ofm2: unknown. Length ofM2: 14.0 mm.
phids in having lower crowned and consequently less lopho- Included species: Pjepseni only, known from locality CPI 9II
dont upper and lower cheek teeth, differs in having narrow only (see Gingerich, 1991).
double-rooted Pl-2 like those of Hyracotherium, differs in
having narrower upper molars (making them relatively square Schizotheriodes Hough, 1955
like those of Hyracotherium), and differs in having less mo- Type species: Schizotheriodes parvus Hough, 1955.
larized p4. (Gingerich, 1991, p. 194.) Type specimen: USNM 20205 (only known specimen).
Characteristics: Extreme development of parastyles on the
Average length of m2: 8.3-8.8 mm. upper molars; paracone and metacones weakly developed;
Included species: C. radinskyi (= cf. Homogalax primae- protoloph displays a distinct paraconule; protocone is
vus, McKenna, 1960; = Homogalax sp., Radinsky, 1963; slightly deflected posteriorly; continuous cingulum devel-
= Hyracotherium sp., Bown, 1979) (known from locali- oped from the parastyle, across the anterior, lingual, and
ties CP19, CP20A); C. semihians (Cope, 1882b) (= Syste- posterior borders of the tooth; labial cingulum present, and
modon semihians Cope, 1882b; = Homogalax semihians, a small mesostyle is developed on M3. Schizotheriodes can
Hay, 1899) (locality CP20E). be distinguished from Paleomoropus by its anterolingually-
posterolabially compressed parastyles, and its slight devel-
Isectolophus Scott and Osborn, 1887 (synonyms: opment of a sharp cingular cuspule, possibly an incipient
Parisectolophus; Schizolophodon; Helaletes, in part; mesostyle (Radinsky, 1964).
Homogalax, in part) Average length of m2: unknown. Length of M2: 11.7 mm.
Type species: lsectolophus annectens Scott and Osborn, 1887. Included species: S. parvus only, known from locality NP22
Type specimen: PU 10400. only.
Characteristics: In general larger than Homogalax, and lacks
a diastema; Pl is relatively small and P3 may be submolar-
CERATOMORPHA
iform (it bears one or two lingual cusps); Ml-3 relatively
larger than in Homogalax, with protoconules and meta- Heptodon Cope, 1882c (synonyms: Lophiodon, in part;
conules never distinct; metastylids lacking on ml-3. Pachynolophus, in part)
Average length of m2: 10.8-15.3 mm. Type species: Heptodon calciculus (Cope, 1880) (= Lophio-
Included species:/. annectens (=I. scotti Peterson, 1919; = don ventorum Cope, 1880).
Schizolophodon cuspidens Peterson, 1919; = Homogalax Type specimen: AMNH 4850.
uintensis, Troxell, 1922a) (known from locality CP6A); I. Characteristics: Small- to medium-sized ceratomorphs with
latidens (Osborn, Scott, and Speir, 1878) (= Helaletes lati- a complete eutherian dentition; P2-4 nonmolariform with
dens Osborn, Scott, and Speir, 1878; 1sectolophus (Hela- only one lingual cusp, protoloph better developed than
 M. W. Colbert and R. M. Schoch 574

metaloph, and protocone steeply conical; M 1-2 metacones Length of known m2: 16.0 mm.
slightly convex to flat, and relatively unshortened and not as Included species: D. intermedius ( = Hyrachyus intermedius
lingually displaced as in otherceratomorphs; M3 metacone Osborn, Scott, and Speir, 1878 = Desmatotherium guy-
relatively unreduced; p2 is a labiolingually compressed otti Scott, 1883; = Chasmotheriodes intermedius, Wood,
blade; p3-4 trigonids relatively high; narial incision not en- 1934; Helaletes intermedius, Radinsky, 1963) only (known
larged. Heptodon can be distinguished from isectolophids from localities CP34D, ?CP38B).
by its long postcanine diastemata, by its lack of para- and
Comments: Desmatotherium is known from four fragmen-
metaconules on Ml-3, and by its relatively long and nar-
tary specimens that are only doubtfully referred to a single
row M3; m3 with short, narrow hypoconulid.
species (see Schoch, 1984a).
Average length ofm2: 10.0-13.4 mm.
Included species: H. calciculus (= Lophiodon calciculus Selenaletes Radinsky, 1966a
Cope, 1880; = L. ventorum Cope, 1880; = Pachynolophus Type species: Selenaletes scopaeus Radinsky, 1966a.
calciculus, Cope, 1882a; = P. ventorum, Cope, 1882a; = Type specimen: AMNH 8320.
Heptodon ventorum, Cope, 1882c) (known from localities Characteristics: Very small ceratomorphs with relatively non-
CP20A, CP20C, CP25E, CP25F, CP25G, CP25H, CP27D, molariform premolars; lower molars similar to those of
CP27E, [CP28A], [CP28B], CP64B, CP64C); H. posticus Heptodon, but much smaller; differs from Dilophodon in
(Cope, 1882b) (= Pachynolophus posticus Cope, 1882b; having a narrower p4 talonid with a smaller entoconid and
= Heptodon brownorum Seton, 1931) (localities SB22B, less reduced molar metalophids; m3 is smaller than m2
SB22C,CP20A,CP20C,CP20D,CP20E,CP25G,CP27C, and lacks a hypoconulid.
CP27D, CP27E, CP28B, ?CP31B, CP64C). Length of known m2: 7.5 mm.
Heptodon sp. is also known from localities CP20A, Included species: S. scopaeus (known from localities SB22B,
CP20C,CP28B,CP31C. SB22C, CP27D, CP27E); ?Selenaletes new species (local-
ity CP34A).
CERATOMORPHA INCERTAE SEDIS
Comments: Selenaletes is only known from lower jaw frag-
Fouchia Emry, 1989 ments and isolated teeth.
Type species: Fouchia elyensis Emry, 1989.
Type specimen: USNM 417339. Dilophodon Scott, 1883
Characteristics: Smallest known ceratomorph; pl/Pl present, Type species: Dilophodon minusculus Scott, 1883.
nonmolariform to submolariform premolars; upper mo- Type specimen: PU 10019.
lar metacones lingually displaced and somewhat labially Characteristics: Very small tapiroids; premolars relatively
flattened, bearing a small labial cingulum; m3 lacks a nonmolariform to submolariform; lacking hypocones on
hypoconulid. Fouchia is superficially similar to Dilopho- P3-4; Ml-2 metacones lingually displaced, slightly con-
don minusculus, from which it differs in its smaller size vex labially, and bear small labial cingula; pl absent; m3
and possession of a p 1. lacks a hypoconulid.
Length of m2: 6.0 mm. Average length of m2: 8.6-9.5 mm.
Included species: F. elyensis only, known from locality NB14 Included species: D. minusculus only (known from localities
only. CP34C, CP36B, CP38B, ?NP22).
Comments: Fouchia and Dilophodon are probably closely Dilophodon sp. is also known from CP37B.
related. Emry (1989) hypothesized that these two forms
Comments: The dental resemblance in both morphology and
are rhinocerotoids based on the derived absence of m3
size of D. minusculus to the new species from California
hypoconulid in both of these forms. Recent discovery of
(included within the Tapiridae) suggests that Dilophodon
a new species of tapirid in California with dentition re-
might be a tapirid.
sembling Dilophodon, but possessing the derived narial
incision of the Tapiridae, calls for a reassessment of the
relations of these diminutive moropomorphs. RHINOCEROTOIDEA

Desmatotherium Scott, 1883 (synonyms: Chasmotheriodes; Hyrachyus Leidy, 187la (synonyms: Lophiodon, in part;
Hyrachyus, in part; Helaletes, in part) Colonoceras; Chasmotherium; Metahyrachyus;
Type species: Desmatotherium intermedius (Osborn, Scott, Ephyrachyus)
and Speir, 1878) ( = Desmatotherium guyotti Scott, 1883). Type species: Hyrachyus modestus (Leidy, 1871 a) (= Lophio-
Type specimen: PU 10166. don agrestis; see Leidy, 187lb; Wood, 1934).
Characteristics: Desmatotherium is superficially similar to Type specimen: USNM 660.
Helaletes, but is larger, with metalophs that bypass the Characteristics: Hyrachyus is here considered the most
hypocones on P3-4 and an extremely reduced hypoconulid basally diverging rhinocerotoid lineage. It is relatively
onm3. plesiomorphic, preserving general dental and skeletal
 Tapiroidea and other moropomorphs 575

features observed in nonrhinocerotoid ceratomorphs such ented cross crests on both upper and lower dentition; large
as Heptodon (Radinsky, 1967). Hyrachyus is medium- to parastyle on M3; labial cingulum absent on Ml and M2;
large-sized relative to other nonrhinocerotoid ceratomorphs relatively massive dentary, with short symphysis and rela-
(Wood, 1927, 1934). It typically possesses a full eutherian tively wide angle between lower jaws (Emry, 1979).
dentition, spatulate incisors, and nonmolariform to submo- Lengthofm2: 14.1.
lariform premolars with no tendency toward bilophodonty. Included species: T. hunteri ( = Schizotheriodes jackwilsoni
It is distinguished by the development of crestlike upper Schiebout, 1977; = Toxotherium woodi Skinner and
molar metacones, especially posteriorly, and the loss of Gooris, 1966) only (known from localities SB44B, SB44C,
the metacone's labial cingula. The M3 metacone is lin- SB44E, CP39B, CP39C, ?CP39G, CP39IIA, NPlOB).
gually depressed, and the M3 is relatively triangular; the Toxotherium sp. is also known from localities SB43C,
hypoconulid is absent on m3; the narial incision is slight. and ?SB44C.
Hyrachyus is judged to have been moderately cursorial and
has a tetradactyl manus and a tridactyl pes. Comments: It has been suggested that because of the wedge-
Average length ofm2: 14.2-25.3 mm. shaped mandible and short symphysis that Toxotherium
Included species: H. modestus ( = Lophiodon modestus Leidy, is relatively brachycephalic (Skinner and Gooris, 1966), a
1870; = Hyrachyus agrestis Leidy, 187la; = H. agrar- possible indication of amynodont affinities. Except for the
ius Leidy, 187la; = Lophiodon hairdianus Marsh 187la; relatively large parastyles, the cusp pattern of Toxotherium
= Lophiodon affinus Marsh, 187la; = Hyrachyus hair- is basically rhinocerotoid.
dianus Leidy, 1872; = Colonoceras agrestis Marsh, 1873;
= Hyrachyus crassidens Osborn, Scott, and Speir, 1878; =
TAPIROIDEA
Chasmotherium stehlini Deperet, 1904; = Metahyrachyus
hicornutus Troxell, 1922b; = Ephyrachyus cristalophus Helaletes Marsh, 1872 (synonyms: Lophiodon, in part;
Wood, 1934; = Chasmotherium minimum [in part], Fis- Hyrachyus, in part)
cher, 1964) (known from localities NB14, SB22C, [SB41], Type species: Helaletes nanus (Marsh, 187la) (= Helaletes
CP5A, CP20F, CP25H, CP25K, CP27B, CP27D, CP31D, hoops Marsh, 1872).
CP34A, CP34B, CP34C, CP34D, CP35, CP38B, CP65); Type specimen: YPM 11080.
H. eximius Leidy, 1871 b (= H. princeps Marsh, 1872; = H. Characteristics: Small tapiroids with small canines; P2-4
imperialis Osborn, Scott, and Speir, 1878; = H. hyposty- submolariform with a tendency for the hypocone to sep-
lus Wood, 1934; = Metahyrachyus troxelli Wood, 1934) arate from the protocone, but metalophs do not bypass
(localities CP31B, [CP32B], CP34C, CP34D, ?CP36A, the hypocones as in Desmatotherium; Ml-2 metacones
CP36B). slightly convex to flat; M3 metaloph relatively shorter and
Hyrachyus sp. (including unsynonymized species H. metacone not as reduced as in Colodon; small entoconids
"grandis" and H. "imperialis") is also known from locali- present on p3-4; m3 with short hypoconulid; narial in-
ties CC2, ?SB43A, CP6A, CP6B, CP31C, CP34A, CP38C, cision greatly enlarged (see Radinsky, 1963). Relative to
?NP14, PN5A. Heptodon and Tapirus, Helaletes shows modifications of
the hindlimb (e.g., a longer ilium and a higher greater
Comments: As noted by Schoch (1984a), the genus is in
trochanter of the femur) toward a more cursorial condi-
need of a thorough revision. Whereas Wood's (1934) re-
tion (see Schoch, 1984a).
vision may have been "oversplit," Radinsky's (1967) revi-
Average length of m2: 9.9-11.8 mm.
sion probably suffers from an extreme case of taxonomic
Included species: H. nanus ( = Lophiodon nanus Marsh,
"lumping" (Radinsky replaced Wood's four genera and
1871 a; = Hyrachyus nan us Leidy, 1872; = H elaletes hoops
over a dozen species with one genus and three species).
Marsh, 1872) only (known from localities NB 14, [SB22C],
Note added in press: "Hyrachyus grandis" (known from
CP5A, ?CP31B, CP31E, CP34A, CP34C, CP34D, ?NP22).
localities CP6A, CP6B, CP38C) has been synonymized
Helaletes sp. is also known from localities ?CP6B,
with the newly described rhinocerotid Uintaceras radin-
CP20F, CP25K, CP32, ?CP36A, NP14 (termed "Hyrach-
skyi (Holbrook and Lucas, 1997) (see also Prothero, this
yus" by West and Dawson, 1978).
volume, Chapter 41).
Heteraletes Peterson, 1931 (synonym: Dilophodon, in part)
Type species: Heteraletes leotanus Peterson, 1931.
RHINOCEROTOIDEA INCERTAE SEDIS
Type specimen: CM 11849.
Toxotherium Wood, 1961 (synonym: Schizotheriodes, Characteristics: Very small tapiroids; premolars relatively
in part) nonmolariform to submolariform; metalophs bypass the
Type species: Toxotherium hunteri Wood, 1961. hypocones on P3-4; Ml-2 metacones lingually displaced,
Type specimen: NMC 8919. slightly convex labially, with small labial cingula; pl ab-
Characteristics: Reduced premolar series (lacking p 1 and per- sent; m3 lacks a hypoconulid.
manent p2); teeth relatively high crowned, obliquely ori- Average length ofm2: 7.4-8.6.
 M. W. Colbert and R. M. Schoch 576

Included species: H. leotanus (= Dilophodon leotanus, Ml-2 metacones lingually displaced, slightly convex labi-
Hough, 1955) only (known from localities [SB44A], ally, with small labial cingula; pl variably present; m3
CP6B, CP7A, CP29C, [CP39E], NP22, NP23A, NP25A). lacks a hypoconulid. Dentally, this diminutive new taxon
of tapirid is almost identical to Dilophodon minusculus.
Comments: Recently collected skull material of Heteraletes It can be diagnosed by the presence of an enlarged post-
under study by Alan Tabrum displays a degree of narial in- canine diastema and relatively delicate bones. Pl/pl are
cision development similar to that of Helaletes. Heteraletes variably present, and the canines are long (although this
is superficially similar to D. minusculus (see Hough, 1955) is potentially sexually dimorphic). The cranial morphol-
but is smaller and has more molariform premolars with dis- ogy resembles Protapirus, including the morphology of
cernible hypocones. the narial incision and nasal bones. As already mentioned,
the dental similarities of this new species to Dilophodon
Colodon Marsh, 1890 (synonyms: Desmatotherium, in part; minusculus raises the possibility that Dilophodon is also a
Lophiodon, in part; Mesohippus, in part) tapirid.
Type species: Colodon occidentalis (Marsh, 1890) (= Average length of m2: 9.3-10.3 mm.
Colodon luxatus Marsh, 1890; = "Mesohippus longipes Included species: "Unnamed genus and species /!\' (=
Osborn and Wortman, 1894," an equid generic name, is Dilophodon sp. [in part], Stock, 1936; = D. leotanus [in
also a junior synonym of Colodon occidentalis; see part], Golz and Lillegraven, 1977) (known from locali-
Prothero and Shubin, 1989). ties CC7B, [CC9A]); "Unnamed genus and species B" (=
Type specimen: YPM 11830. Dilophodon sp. [in part], Golz, 1976; = D. leotanus [in
Characteristics: Medium-sized to large tapiroids with rela- part], Golz and Lillegraven, 1977; = D. minusculus [in
tively short, wide cheek teeth; canines small or absent; Pl part], MacFadden, 1980) (localities CC4, CC5, CC6A).
with lingual loph or cusp; P2-4 essentially molariform,
protocone and hypocone more or less separated, metaloph
Protapirus Filhol, 1877 (synonyms: Tanyops;
as prominent as protoloph and extending to hypocone;
Plesiocolopirus; Colodon, in part; Tapirus, in part)
Ml-2 metacone flat to concave, more lingually displaced
Type species: Protapirus priscus (Filhol, 1874) (= Tapirus
and shortened than in Helaletes; p 1 absent; p2-4 with wide
priscus Filhol, 1874, from the Phosphorites du Quercy,
talonids and relatively large entoconids; m3 hypoconulid
France).
variably developed; greatly enlarged narial incisions with
Characteristics: Medium-sized tapirids with nonmolariform
reduced nasals (Radinsky, 1963).
to submolariform (incipient hypocone on P4, or P3 and
Average length of m2: 14.0-20.0 mm.
P4) premolars; small canines present; Ml-2 metacones
Included species: C. occidentalis (= Lophiodon occiden-
convex labially and lingually displaced; hypoconulid ab-
talis Leidy, 1868; = Colodon luxatus Marsh, 1890; Meso-
sent on m3; enlarged narial incision, nasals extend to above
hippus longipes Osborn and Wortman, 1894; = Colodon
canines (Hanson, 1996), with grooves developed on their
dakotensis Osborn and Wortman, 1895; = C. procuspi-
posterolateral margin.
datus Osborn and Wortman, 1895) (known from locali-
Average length of m2: 18.0-27 .0 mm.
ties CP83A, ?CP83B, ?CP83C, CP84A, CP84B, CP98A,
Included species: P. simplex Wortman and Earle, 1893 (known
CP99A, CP99B, NPIOA, ?NPIOB); Colodon cingulatus
from locality CP84A); P. obliquidens Wortman and Earle,
Douglass, 1901 (localities NP24D, [NP25A], [NP25B],
1893 (= Tanyops undans Marsh, 1894) (locality CP84B);
?NP25C, ?NP27D, ?NP30A); C. kayi (Hough, 1955) (=
P. validus Hatcher, 1896 (locality CP84B); P. robustus
Desmatotherium kayi Hough, 1955) (localities ?NP8,
Sinclair, 1901 (localities PN6E, PN6F, PN6G); P. han-
NP22, NP24C, NP25A); C. wocdi (Gazin, 1956) (=Des-
cocki (Radinsky, 1963) (= Colodon? hancocki Radinsky,
matotherium kayi [in part], Hough, 1955; = D. woodi
1963, Plesiocolopirus hancocki Schoch, 1989b) (localities
Gazin, 1956) (localities CP29C, [CP37 A], [CP37B],
[SB44A], PN5B).
[NP25A]); Colodon stovalli Wilson and Schiebout, 1984
Protapirus sp. is also known from (GC5), CP84A, PN6C,
(localities ?SB44B, ?SB44C).
PN6D,PN6E.
Colodon sp. is also known from localities NB2, CP29C,
CP39A, CP39C, CP39E, NP8, NP9A, NP9B, NP9C, Comments: Hatcher (1896) recognized three North Ameri-
NP24D, PN6C. can species based on differences in the premolar region,
but Scott (1941), noting the variability of premolar struc-
ture in many perissodactyl species, cautioned that these
TAPIRIDAE
differences might be attributable to individual variation.
Unnamed new genus (see Colbert, 1993) (synonym: As construed here, Protapirus includes material from the
Dilophodon, in part) Clarno Formation (locality PN5B) that Radinsky (1963)
Characteristics: Very small tapiroid; premolars relatively non- questionably referred to Colodon and that Schoch (1989b)
molariform to submolariform, lacking hypocones on P3-4; later referred to a new genus, Plesiocolopirus. Hanson
 Tapiroidea and other moropomorphs 577

(1996) justifies referring this material to Protapirus based 1960 (= Tapiravus?, White, 1942; = Tapiravus cf. polken-
on the variability observed in a larger Clarno sample than sis, Schultz, Martin, and Comer, 1975) (localities GC13A,
was available to either Radinsky or Schoch, and by its [CPI 14B]); ?T. rarus, Marsh, 1877, nomen dubium,
similarity to the type Protapirus in France. The mono- Schoch, 1984b (= Tapiravus rarus Marsh, 1877) (local-
phyly of Protapirus is dubious, however, judging from its ity uncertain - late Miocene or early Pliocene - aged strata
lack of known apomorphies. Clearly, the genus is in dire of the Rocky Mountains region; see Schoch, 1984b ).
need of revision. Tapiravus sp. is also known from locality ?CP76.

Miotapirus Schlaijker, 1937 Tapirus Briinnich, 1772 (synonyms: Hippopotamus, in part;

Type species: Miotapirus harrisonensis Schlaijker, 1937. Tapir; Hydrochoerus, in part; Hydrochaerus; Syspotamus;
Type specimen: MCZ 2949. Rhinochoerus; Elasmognathus; Tapyra; Tapiriella;
Characteristics: Medium-sized tapirids; in many respects in- Acrocodia; Megatapirus)
termediate in general morphology between Protapirus and Type species: Tapirus terrestris (Linnaeus, 1758) (=Hip-
Tapirus (and Tapiravus); i 1 larger than i2, and i3 diminu- popotamus terrestris).
tive; p2 proportionately short (anteroposteriorly); upper Type specimen: Not designated (the extant lowland tapir, see
and lower premolars relatively molariform, with Hershkovitz, 1954; Honacki, Kinman, and Keoppl, 1982:
well-developed hypolophid and hypocone; P4 larger than type area is Pernambuco, Brazil).
Ml, p4 larger than ml; superior premolar length greater Characteristics: Medium- to large-sized tapirids; premolars
than superior molar length; molar crests are higher and highly molariform; upper canine vestigial and functionally
better developed than in Protapirus, but not as well de- replaced by 13; prehensile proboscis present.
veloped as in Tapirus; Ml-2 relatively square in occlusal Average length ofm2: 21.4-30.8 mm.
view; nasals reduced and with nasal excavations; skull Included species: T. haysii Leidy, 1859 ( = T. copei Simp-
"telescoped" posteriorly; symphysis constricted but mas- son, 1945b) (localities SB37, SB49, CP80, plus Irving-
sive (Schlaikjer, 1937). tonian, and possibly Rancholabrean of Florida, Georgia,
Average length ofm2: 18.7 mm. South Carolina, North Carolina, Pennsylvania, Indiana,
Included species: M. harrisonensis (known from localities Arkansas, Oklahoma, Texas, Nebraska, ?Colorado [see
CP42D, CP51B, CP86B, CP104B); M. marslandensis Ray and Sanders, 1984]); T. californicus Merriam, 1913
Schoch and Prins, in Schoch l 984b (locality CPI 05). (= T. haysii californicus Merriam, 1913) (Rancholabrean,
Miotapirus sp. is also known from locality ?CP!Ol. California and Oregon; ?New Mexico and ?Arizona); T.
veroensis Sellards, 1918 (= T. excelsus Simpson, 1945b)
Comme,nts: Miotapirus is known from only a very few spec-
(Rancholabrean of Florida, Georgia, Kansas, Missouri,
imens (see Macdonald, 1970; Schlaikjer, 1937; Schoch,
Tennessee, Texas); T. merriami Frick, 1921 (Irvingtonian,
1984b). Albright (unpublished ms.) argues that the spec-
Bautista Creek, California); ?T. johnsoni Schultz, Martin,
imen referred by Macdonald (1970) to M. harrisonensis
and Comer, 1975 (known from localities ?SP2A, [CP49C],
and the specimens referred to M. marslandensis by Schoch
?CP114B, ?CPl 140, CP116B); ?T. simpsoni Schultz, Mar-
l 984b actually belong to a new genus that he is currently
tin, and Comer, 1975 (localities GCl lA, ?GCl lB, SP3A,
describing.
?CP116B).
Tapirus sp. is also known from localities GC 14, ?CC26B,
Tapiravus Marsh, 1877 (synonym: Lophiodon, in part)
CC30A, CC41, SB38, CP90A, CP116E, CP116F, CP121
Type species: Tapiravus validus (Marsh, 187lb) (= Lophio-
(plus numerous Pleistocene sites south of the glaciated
don validus Marsh, 187lb).
area).
Type specimen: YPM 13474.
Characteristics: Small- to medium-sized tapirids; dentition Comments: There is no current revision of all species, recent
similar to that of Tapirus, but teeth lower crowned (rela- and fossil, that have been referred to Tapirus. Although
tively brachydont); lower cheek teeth elongated anteropos- this genus is very uniform dentally, cranial differences (and
teriorly; m2-3 trigonids and talonids separated by a deep in the case of extant species, differences in pelage) have
groove or constriction; maxilla relatively deep as com- been used to warrant full generic rank for at least each of
pared to Tapirus; lachrymal process less prominent than the living species (Simpson, 1945b; Hershkovitz, 1954).
in Tapirus; sulcus on the lateral face of the ascending por- Most paleontologists have found it convenient to lump all
tion of the maxilla shallower and more constricted than in these forms into the genus Tapirus (e.g., Simpson, 1945b;
Tapirus; mandibular symphysis narrow relative to Tapirus. Ray and Sanders, 1984) because most isolated tapir teeth
Average length ofm2: 12.0-19.8 mm. from this group would be generically indeterminate if more
Included species: T. validus (= Lophiodon validus Marsh, than one genus were recognized. Hershkovitz (1954), the
187lb; = Tapiravus cf. validus, Gazin and Collins, 1950) last reviser of extant forms, being influenced by Simpson
(known from localities NCIB, NC3B); T. polkensis Olsen, (1945b), chose to recognize a distinct subgenus for each of
 M. W. Colbert and R. M. Schoch 578

the four extant species. Whether or not to recognize more Moropomorphs were relatively common in the Northern Hemi-
than one genus for Tapirus is a semantic issue, as long as sphere during the Eocene, at which time they underwent a prolific
the named supraspecific taxa named are monophyletic. radiation. In North America at this time, several lineages, described
earlier, were present: the conservative isectolophids, Paleomoro-
pus, Schizotheriodes, eomoropid chalicotheres (i.e., Eomoropus and
INDETERMINATE MOROPOMORPHS Grangeria), Heptodon, the Rhinocerotoidea, and the Tapiroidea in-
Fragmentary material assigned to "tapiroid indet." has also been cluding primitive tapirids (e.g., the Clarno Formation Protapirus).
reported from localities WM13, CC9AA, GC8B, GC28, GC26B, Tapiroid evolution is characterized by dental adaptations result-
NB29, SB51, CPI 14A, PNlO. ing in an improvement of shear along the transverse crests of the
molars, molarization of the premolars, and especially by enlarged
narial incisions associated with the development of the probos-
BIOLOGY AND EVOLUTIONARY PATTERMS cis, with its complex facial musculature and cartilaginous nasal
diverticulae.
The four extant species of tapirs are tropical in distribution: three Among North American moropomorphs (aside from the rhinos),
species (Tapirus terrestris, T. pinchaque, and T. bairdii) living in only the Colodon-tapirid clade (all tapiroids) survived past the
Central and South America and Tapirus indicus living in southeast Eocene. Additionally, the forms surviving the Eocene-Oligocene
Asia (see Hershkovitz, 1954; Eisenberg, 1981; Nowak, 1991, for transition are generally extremely rare as fossils, and it is assumed
discussions of living tapirs and references to more detailed works). that they were low in both diversity and absolute numbers.
Extant tapirs live in tropical forests, woody, and grassy habitats. Because this clade is characterized by its primitive possession, and
They range from sea level to 4,500 m. The head and body length of further development of, a prehensile proboscis, it was possibly the
extant species of Tapirus range from 180-250 cm, and their body possession of this feature, and its correlated behavioral strategies,
mass varies from approximately 225-300 kg. Lowland tapirs seem that permitted their survival through the Eocene-Oligocene tran-
to depend on permanent water supplies. Tapirs tend to be solitary, sition (see discussion later). These tapiroids are characterized by
nonterritorial, and have no distinct breeding season; all features are precocious proboscis development, and by the middle Miocene es-
believed to be primitive behavioral characteristics for ungulates in sentially modem tapirs were already present (e.g., Miotapirus ). The
general (Janis, 1984; Nowak, 1991). conservatism in many features of tapiroid evolution, coupled with
Overall, the postcranial skeleton of tapirs is also relatively ple- the retention of a large number of plesiomorphic features, especially
siomorphic, and it has been noted that the skeleton of Tapirus is as compared to other ungulates, has led many to consider the recent
similar to that of the basal ceratomorph Heptodon, with the pri- tapirs "living fossils" (Janis, 1984).
mary differences related to Tapirus's larger size (approximately 40 The evolution of the tapirids is largely a story of the refinement of
percent larger), relatively larger head, and more graviportal nature the proboscis, molarization of the premolars, and a general increase
(Radinsky, 1965; Janis, 1984). The dentition of tapirs is also rela- in size. These are all presumably correlated with the fine-tuning of
tively plesiomorphic, with the bilophodont pattern established early tapirid life history strategies. It is probable that tapirids have al-
in the group's history apparently constraining further dental evo- ways lived in humid mesothermal areas, where the quantity and
lution. In contrast, the skull (particularly in the Tapiroidea) shows diversity of foliage is great. The paucity of tapiroid remains in the
many modifications, especially noticeable in the development of a Oligocene and Miocene of North America undoubtedly reflects the
short, fleshy, prehensile proboscis formed by the snout and upper restriction of these areas in contrast to their widespread Eocene dis-
lips. Extant tapirs use this prehensile proboscis to browse selectively tribution. The development of the proboscis, and related selective
on leaves, sprouts, and small branches, including aquatic plants, foraging strategies, might have allowed the Oligocene and Miocene
and also ingest a great deal of fruit and seeds (Eisenberg, 1981; tapirids to maximize their utilization of food resources in restricted
mesothermal refugia (e.g., riparian areas, lakeshore habitats, etc.) in
Janis, 1984).
One of the earliest well-known moropomorphs (as discussed ear- an otherwise relatively dry temperate environment. The rhinocero-
lier, moropomorphs are the sister tax on to the Hippomorpha) is Ho- toids, sister tax on to the tapiroids, generally dealt with the changing
mogalax of the early Eocene. Homogalax is a relatively small and environmental conditions by developing higher crowned teeth with
generalized perissodactyl with a skull length of approximately 16 cm increased ectoloph shear (paralleled in the European tapiroid fam-
(PU 16168) and an estimated body mass of lOkg (Janis, 1984). Com- ilies Lophialetidae and Lophiodontidae), and they adopted various
pared to Hyracotherium (which in some respects approximates an strategies for dealing with more fibrous vegetation, including in-
ancestral perissodactyl morphotype), Homogalax and all later mo- creased body size.
ropomorphs are characterized by the development of higher and Simpson (1945b) demonstrated that all post-Pliocene North
unbroken crests on the molars that increased the amount of shear American tapir occurrences are south of the limit of continental
along the fronts of the protolophs and metalophs of the upper mo- glaciation in what are presumed to have been humid mesothermal
lars with a concomitant reduction in crushing function of the molars areas. Tapirs appear to have been present when humans first entered
(Radinsky, 1969). Homogalax was relatively folivorous compared the North American continent, but became extinct in these latitudes
to other contemporaneous Eocene perissodactyls (Janis, 1984). at the end of the Pleistocene when mesic woodlands disappeared
 Tapiroidea and other moropomorphs 579

4.5
Lal Hp.
5.2
L.Hp.
6.0
LE.Hp.
7.0
E.E.Hp.
L 8.8
L.CI.
9.5

E.CI.
11.0
w
z L.L.Ba.
w E.L.Ba.
12.5

0 M
0 14.0

:i E.Ba.

5.8
L.Hm.

17.S
E.Hm.
18.8
E LL.Ar.
19.2

E.L.Ar.
23.0

w L.E.Ar.
? I
z
W L
0 27.7 _J - - - - -
0
CJ E.E.Ar.
:J ~
0 29.4 _.!::
W1. ~.?
.....
E 31.9 _,OL
Or.
33.4
:i
L.Ch.
34.5
L M.Ch.
35.5
E.Ch. c:
37.1
-.g -
Du
,g
39.5
-O..r
~
'L.Un. ...... ?
w 41.3
-9 .
z E.Un.
W M
0 45.9
0
w L.Br.
46.7

M.Br.
49.0
e.er.
50.4 ~ ~ -C:
l.Wa. =
.....
Q.
..c:Q. e .gs
53.S ~ 0
..., Ql-Q.
E
.~ § ~
eg>Q
M.wa.
54.2 i:s Q) -
E.WI.
E.E.Wa. ::: 1-::r: 1 ~ ~
_____..;;'J;...L....;_;:__________________________ _.::::_=.....::=--==-=.....:
Ck.3 ss.& I
w ~
1
55.8 L. -
z T1.e 11:~
w L Tl.S 570 TAPIRIDAE
0 Tl.4 57'3
0 ~~ sa:s
w Tl.1 59.2 TAPIROIDEA
..J 60.9
<C To.3 61.3
£L. To.2
To.1
62.5
ISECTOLOPHIDAE CERATOMORPHA
E Pu.3
63.6
64.0
Pu.2 64.5
Pu.1
Pu.O
65.0
65.1
MOROPOMORPHA
Bl.=Blan~an, Hp.= Hemp.hillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemlngfordian, Ar. = Arikareean
Wt. =Whitneya~, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerlan, Wa. = w~satchian
Ck. = Clarkfork1an, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan. '
Figure 39.6. Temporal ranges of North American "tapiroid" genera.
 M. W. Colbert and R. M. Schoch 580

(Janis, 1984). Tapirs invaded South America from the north upon (1989). A tiny new Eocene ceratomorph and comments on "tapiroid"
the emergence of the Isthmus of Panama, during the Pliocene or systematics. Journal of Mammalogy, 70, 794-804.
Filhol, H. (1874). Note relative a la decouverte d'un animal appartenant
Pleistocene. The evolutionary relationships of the South and Cen-
au genre des tapirs dans Jes gisements de phosphates de chaux du
tral American species and other Pleistocene species are currently Quercy. Memoires de la Societe Philomatique et Naturelles, 1-3.
unknown, however, and it is unclear whether these Recent forms (1877). Recherches sur Jes Phosphorites du Quercy. Bibliotheque de
form a monophyletic group relative to Pleistocene North American l'ecole des hautes Etudes, Paris, 16, 1-338.
species. The presence of Tapirus in Southeast Asia today and in the Fischer, K.-H. (1964). Die tapiroiden Perissodactylen aus der eozanen
Braunhohle des Geiseltales. Geologie, Berlin, 45, 1-101.
Pleistocene obviously records a dispersal event, most likely across
(1977). Neue Funde von Rhinocerolophiodon (n. gen.), Lophiodon
the Bering land bridge, but the direction and timing of this dispersal undHyrachyus (Ceratomorpha, Perissodactyla, Mammalia) aus dem
are currently unknown. Eozan des Geiseltals bei Halle (DDR), I. Tei!: Rhinocerolophiodon.
Zeitschrift Geologische Wissenschaften, 5, 909-19.
Frick, C. (1921). Extinct vertebrate faunas of the badlands of Bautista Creek
and San Timoteo Canyon, southern California. University of Cal-
ACKNOWLEDGMENTS ifornia Publications, Bulletin of the Department of Geology, 12,
277-424.
This chapter would not have been possible without the help and Gazin, C. L. (1956). The geology and vertebrate paleontology of upper
encouragement of numerous people. In particular, it benefited from Eocene strata in the northeastern part of the Wind River Basin, Part 2:
The mammalian fauna of the Bad water area. Smithsonian Miscella-
the suggestions of C. Janis, S. Lucas, R. Emry, M. Coombs, D.
neous Collections, 131, 1-35.
Prothero, and T. Demere. M. C. particularly thanks S. Lucas for Gazin, C. L., & Collins, R. L. (1950). Remains of land mammals from the
his role in introducing him to this opportunity and C. Janis for her Miocene of the Chesapeake Bay region. Smithsonian Miscellaneous
understanding and helpful criticism. Collections, 116, 1-21.
Gingerich, P. D. (1991 ). Systematics and evolution of early Eocene Perisso-
dactyla (Mammalia) in the Clarks Fork Basin, Wyoming. Contribu-
tions from the Museum of Paleontology, The University ofMichigan,
REFERENCES 28, 181-213.
Golz, D. J. (1976). Eocene Artiodactyla of southern California. Los Angeles
Bakker, R. T., Cooke, J. C., & Schain, J. (MS). The dawn horses revisited Natural History Museum Bulletin, 26, 1-85.
and the basal bushiness of the Perissodactyla. Hunteria. Golz, D. J., & Lillegraven, J. A. (1977). Summary of known occurrences
Bown, T. M. (1979). Geology and mammalian paleontology of the Sand of terrestrial vertebrates from Eocene strata of southern California.
Creek Facies, lower Willwood Formation (lower Eocene), Washakie University of Wyoming Contributions to Geology, 15, 43-65.
County, Wyoming. Geological Survey of Wyoming Memoir, 2, 1-151. Gregory, W. K. (1920). Studies of comparative myology and osteology, V:
Briinnich, M. T. (1772). Zoologiae fundamenta praelectionibus acad. acco- On the anatomy of the preorbital fossae of Equidae and other un-
modata (Lat. et Dan). Transactions of the Linnean Society ofLondon, gulates. Bulletin of the American Museum of Natural History, 42,
7, p. 241. 265-83.
Colbert, M. W. (1993). New Species of Tapiroids from the Eocene of San Hanson, C. B. (1996). Stratigraphy and vertebrate faunas of the Bridgerian-
Diego County, California, and Their Implications to Tapiroid Phy- Duchesnean Clarno Formation, north-central Oregon. In The Ter-
logeny and Evolution. Unpublished Master's Thesis. San Diego: San restrial Eocene-Oligocene Transition in North America, eds. D. R.
Diego State University. Prothero & R. J. Emry, pp. 206-39. Cambridge: Cambridge Univer-
Coombs, M. C. (1982). Chalicotheres (Perissodactyla) as large terrestrial sity Press.
mammals. Proceedings of the Third North American Paleontological Hatcher, J.B. (1896). Recent and fossil tapirs. American Journal of Science
Convention, I, 99-103. (4th series), 1, 161-80.
Cope, E. D. (1880). The badlands of the Wind River and their fauna. Amer- Hay, 0. P. (1899). On the names of certain North American fossil vertebrates.
ican Naturalist, 14, 745-8. Science, 9, 593-4.
(1882a). On the Vertebrata of the Wind River Eocene beds of Wyoming. Hennig, W. (1966). Phylogenetic Systematics. Urbana: University of Illinois
Bulletin of the United States Geological and Geographical Survey Press.
of the Territories, 6, 183-202. Hershkovitz, P. (1954). Mammals of Northern Colombia, Preliminary Report
(1882b). Contributions to the history of the Vertebrata of the lower No. 7: Tapirs (Genus Tapirus), with a systematic review of Amer-
Eocene of Wyoming and New Mexico, made during 1881, Part. I; ican species. Proceedings of the United States National Museum,
The fauna of the Wasatch beds of the Basin of the Bighorn River. Smithsonian Institution, 103, 465-96.
Proceedings of the American Philosophical Society, 20, 139-97. Holbrook, L. T., & Lucas, S. G. (1997). A new genus ofrhinocerotidfrom the
( l 882c ). Two new genera of Mammalia from the Wasatch Eocene. Amer- Eocene of Utah and the status of North American "Forstercooperia."
ican Naturalist, 16, 1029. Journal of Vertebrate Paleontology, 17, 384-96.
(1889). The Vertebrata of the Swift Current River, II. American Natu- Honacki, J. H., Kinman, K. E., & Koeppl, J. W. (1982). Mammal Species of
ralist, 23, 151-5. the World. Lawrence, Kans.: Association of Systematics Collections.
Deperet, C. ( 1904). Sur Jes caracteres et Jes affinites du genre Chasmotherium Hooker, J. J. (1984). A primitive ceratomorph (Perissodactyla, Mammalia)
Riitimeyer. Bulletin de Societe Geologique de France, 4, 569-87. from the early Tertiary of Europe. Journal of the Linnaean Society,
Douglass, E. (1901). Fossil Mammalia of the White River Beds of Montana. 82, 229-44.
Transactions of the American Philosophical Society, 20, 237-79. (1989). Character polarities in early perissodactyls and their significance
Eisenberg, J. F. (1981). The Mammalian Radiations. Chicago: University of for Hyracotherium and infraordinal relationships. In The Evolution
Chicago Press. of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 79-lOI.
Emry, R. J. (1979). Review of Toxotherium (Perissodactyla, Rhinocero- Oxford: Clarendon.
toidea) with new material from the early Oligocene of Wyoming. Hough, J. (1955). An upper Eocene fauna from the Sage Creek area, Beaver-
Proceedings of the Biological Society of Washington, 92, 28-41. head County, Montana. Journal of Paleontology, 29, 22-36.
 Tapiroidea and other moropomorphs 581

Janis, C. M. (1984). Tapirs as living fossils. In Living Fossils, eds. N. Eldredge Prothero, D. R., & Schoch, R. M. (1989a). Origin and evolution of the
& S. Stanley, pp. 80--6. New York: Springer-Verlag. Perissodactyla: summary and synthesis. In The Evolution of Peris-
Leidy, J. (1859). Description of vertebrate fossils. In 1858-1860, Post- sodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 504-29. Oxford:
Pleistocenefossils of South Carolina, ed. F. S. Holmes, pp. 99-122. Clarendon.
Charleston: Russell and Jones. (l 989b ). Classification of the Perissodacty la. In The Evolution of Peris-
(1868). Notice of some Eocene remains of extinct pachyderms. Pro- sodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 530-7. Oxford:
ceedings of the Academy of Natural Sciences Philadelphia, 20, Clarendon.
230--3. Prothero, D.R., & Shubin, N. (1989). The evolution of Mid-Oligocene
(1870). Remarks on a collection of fossils from the western territories. horses. In The Evolution of Perissodactyls, eds. D. R. Prothero &
Proceedings of the Academy of Natural Sciences Philadelphia, 22, R. M. Schoch, pp. 142-75. Oxford: Clarendon.
109. Radinsky, L. B. (1963). Origin and early evolution of North American
(187la). Report on vertebrate fossils from the Tertiary formations of Tapiroidea. Yale University Peabody Museum Natural History Bul-
the West. Annual Report of the United States Geological Survey of letin, 17, 1-115.
Wyoming and portions of Contiguous Territories, 340--70. (1964). Paleomoropus, a new early Eocene chalicothere (Mammalia,
(1871 b ). Remarks on fossil vertebrates from Wyoming. Proceedings of Perissodactyla), and a revision of the Eocene chalicotheres. American
the Academy of Natural Sciences Philadelphia, 23, 229. Museum Novitates, 2179, 1-28.
(1872). Remarks on fossils from Wyoming. Proceedings of the Academy (1965). Evolution of the tapiroid skeleton from Heptodon to Tapirus.
of Natural Sciences Philadelphia, 24, 19-20. Bulletin of the Museum of Comparative Zoology, Harvard, 134,
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundum 69-106.
classes, ordines, genera, species, cum characteribus, differentiis, (1966a). A new genus of Eocene tapiroid (Mammalia, Perissodactyla).
synonymis, locis. Stockholm: Laurentii Salvii. Journal of Paleontology, 40, 740-2.
Macdonald, J. R. (1970). Review of the Miocene Wounded Knee faunas (l 966b ). The families of the Rhinocerotoidea (Mammalia, Perisso-
of southwestern South Dakota. Bulletin of the Los Angeles County dactyla). Journal of Mammalogy, 47, 631-9.
Museum Natural Hist01y, 8, 1-82. (1967). Hyrachyus, Chasmotherium, and the early evolution of helaletid
MacFadden, B. J. (1980). Eocene perissodactyls from the type section of the tapiroids. American Museum Novitates, 2313, 1-23.
Tepee Trail Formation of northwestern Wyoming. Contributions in ( 1969). The early evolution of the Perissodactyla. Evolution, 23, 308-28.
Geology, University of Wyoming, 18, 135-43. Ray, C. E., & Sanders, A. E. (1984). Pleistocene tapirs in the Eastern United
Maddison, W. P., Donoghue, M. J., & Maddison, D. R. (1986). Outgroup States. In Contributions in Quaternary Vertebrate Paleontology: A
analysis and parsimony. Systematic Zoology, 33, 83-103. Volume in Memorial to John E. Guilday, eds. H. Genoways & M.
Marsh, 0. C. (187la). Notice of some new fossil mammals from the Tertiary Dawson. Special Publications of the Carnegie Museum of Natural
formation. American Journal of Science (3rd series), 2, 35-44. History, 8, 283-315.
(187lb). [Untitled]. Proceedings of the Academy of Natural Sciences Schiebout, J. A. (1977). Schizotheriodes (Mammalia, Tapiroidea) from
Philadelphia, 23, 9-10. the Oligocene of Trans-Pecos Texas. Journal of Paleontology, 51,
(1872). Preliminary descriptions of new Tertiary mammals. American 455-8.
Journal of Science (3rd series), 4, 202-24. Schlaikjer, E. M. (1937). A new tapir from the Lower Miocene of Wyoming.
(1873). Notice of new Tertiary mammals. American Journal of Science Bulletin of the Museum of Comparative Zoology, Harvard, 80,
(3rd series), 5, 407-8. 231-51.
(1877). Notice of some new vertebrate fossils. American Journal of Sci- Schoch, R. M. (1983). A new species of Isectolophus (Mammalia,
ence (3rd series), 14, 249-56. Tapiroidea) from the middle Eocene of Wyoming. Postilla, Peabody
(1890). Notice of new Tertiary mammals. American Journal of Science Museum, Yale University, 188, 1-4.
(3rd series), 39, 523-35. (1984a). Two unusual specimens of Helaletes in the Yale Peabody Mu-
(1894). A new Miocene tapir. American Journal of Science (3rd series), seum collections, and some comments on the ancestry of the Tapiri-
48, 348. dae (Perissodactyla, Mammalia). Postilla, Peabody Museum, Yale
Merriam, J. C. (1913). Tapir remains from the late Cenozoic beds of the University, 193, 1-20.
Pacific coast region. University of California Publications, Bulletin (1984b). The type specimen of Tapiravus validus and ?Tapiravus rarus
of the Department of Geology, 7, 169-75. (Mammalia, Perissodactyla), with a review of the genus, and a new
Nowak, R. M. (1991). Walker's Mammals of the World (5th ed.). Baltimore: report of Miotapirus (Miotapirus marslandensis Schoch and Prins,
John Hopkins University Press. new species) from Nebraska. Postilla, Peabody Museum, Yale Uni-
Olsen, S. J. (1960). Age and fauna! relationship of Tapiravus remains from versity, 195, 1-12.
Florida. Journal of Paleontology, 34, 164-7. ( 1989a). A brief historical review of perissodactyl classification. In The
Osborn, H. F., Scott, W. B., & Speir, F. (1878). Paleontological report of Evolution of Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp.
the Princeton Scientific Expedition of 1877. Contributions of the 13-23. Oxford: Clarendon.
Museum of Geology and Archeology of Princeton College, 1, 1-107. (l 989b ). A review of the tapiroids. In The Evolution of Perissodactyls, eds.
Osborn, H. F., & Wortman, J. L. (1894). Fossil mammals of the Lower D.R. Prothero & R. M. Schoch, pp. 298-320. Oxford: Clarendon.
Miocene White River beds. Collection of 1892. Bulletin of the Amer- Schultz, C. B., Martin, L. D., & Corner, R. G. (1975). Middle and late
ican Museum of Natural History, 6, 199-228. Cenozoic tapirs from Nebraska. Bulletin of the University of
(1895). Perissodactyls of the lower Miocene White River beds. Bulletin Nebraska State Museum, 2, 1-21.
of the American Museum of Natural History, 7, 343-75. Scott, W. B. (1883). On Desmatotherium and Dilophodon, two new Eocene
Peterson, 0. A. (1919). Report upon the material discovered in the upper lophiodonts. Contributions of the E. M. Museum of Geology and
Eocene of the Uinta Basin. Annals of the Carnegie Museum, 12, Archeology, Princeton College Bulletin, 3, 46-53.
40--168. (1941). Perissodactyla. The mammalian fauna of the White River
(1931). New species from the Oligocene of the Uinta. Annals of the Oligocene. Transactions of the American Philosophical Society, 28,
Carnegie Museum, 21, 61-78. · 747-980.
Prothero, D.R., Manning, E., & Hanson, C. B. (1986). The phylogeny of the Scott, W. B., & Osborn, H.F. (1887). Preliminary report on the vertebrate
Rhinocerotoidea (Mammalia, Perissodactyla). Zoological Journal of fossils of the Uinta formation. Proceedings of the American
the Linnaean Society, 87, 341-66. Philosophical Society, 24, 255-64.
 M. W. Colbert and R. M. Schoch 582

(1889). The Mammalia of the Uinta formation. Transactions of the (1922c). Helaletes redefined. American Journal of Science (5th series),
American Philosophical Society, 16, 461-572. 3, 365-70.
Sellards, E. H. (1918). The skull of a Pleistocene tapir including description West, R. M., & Dawson, M. R. (1978). Vertebrate paleontology and the
of a new species and a note on the associated fauna and flora. Cenozoic history of the North Atlantic region. Polarforshung
Florida State Geological Survey, 10th and 11th Annual Reports, (German Society of Polar Research), 48, 103-19.
57-70. White, T. E. (1942). Additions to the fauna of the Florida phosphates.
Seton, H. (1931). A new Heptodon from the Wind River of Wyoming. Proceedings of the New England Zoological Club, 21, 87-91.
Proceedings of the New England Zoological Club, 12, 45-8. Wilson, J. A., & Schiebout, J. A. (1984). Early Tertiary vertebrate faunas,
Simpson, G. G. (1945a). The principles of classification and a classification Trans-Pecos Texas: Ceratomorpha less Amynodontidae. The
of mammals. Bulletin of the American Museum Natural History, Pearce-Sellards Series, 39, 1-47.
85, 1-350. Wood, H. E., II (1927). Some early Tertiary rhinoceroses and hyracodonts.
(1945b). Notes on Pleistocene and Recent tapirs. Bulletin of the Bulletin of American Paleontology, 13, 165-265.
American Museum Natural History, 86, 33-81. (1934). Revision of the Hyrachyidae. Bulletin of the American Museum
Sinclair, W. J. (1901). The discovery ofa new fossil tapir in Oregon. Journal of Natural History, 67, 181-295.
of Geology, 9, 702-7. (1937). Perissodactyl suborders. Journal of Mammalogy, 18, 106.
Skinner, S. M., & G~oris, R. J. (1966). A note on Toxotherium (Mammalia, (1961). Toxotherium hunteri, a peculiar new Oligocene mammal
Rhinocerotoidea) from Natrona County, Wyoming. American from Saskatchewan. Natural History Papers, National Museum of
Museum Novitates, 2261, 1-12. Canada, 13, 1-3.
Stock, C. (1936). Perissodactyla of the Sespe Eocene, California. Proceed- Wortman, J. L. (1896). Species of Hyracotherium and allied perisso-
ings of the National Academy of Science, 22, 260-5. dactyls from the Wasatch and Wind River beds of North America.
Troxell, E. L. (l 922a). The status of Homogalax with two new species. Bulletin of the American Museum of Natural History, 8, 81-111.
American Journal of Science (5th series), 2, 288-92. Wortman, J. L., & Earle, C. (1893). Ancestors of the tapir from the lower
(1922b). Homed Eocene ungulates. American Journal of Science (5th Miocene of Dakota. Bulletin of the American Museum of Natural
series), 4, 31-7. History, 5, 159-80.
40 Amynodontidae

WILLIAM P. WALL

INTRODUCTION

The Amynodontidae were Holarctic rhinocerotoids with a middle

Eocene to early Oligocene stratigraphic range in North
America. Amynodon is a characteristic Uintan (middle Eocene)
taxon; Metamynodon planifrons is a good indicator for the
"Metamynodon channel sandstones" during the Orellan (early
Oligocene). Amynodontids were medium to large-sized animals tra-
ditionally thought to have had a semiaquatic mode oflife. However,
life habits of amynodontids were more diverse than generally rec-
ognized. The two advanced tribes, Cadurcodontini and Metamyn-
odontini, exhibit different morphologic adaptations. Cadurcodon-
tines were lightly built animals possibly resembling modem tapirs in Figure 40.1. Restoration of Amynodon, by Rick Spears.
appearance. Metamynodontines are the only members of the family
with skeletal modifications conducive to a semiaquatic mode of life DENTAL
(Figure 40.1 ).
All amynodontids have greatly enlarged canines, a unique feature
within the Rhinocerotoidea. Sexual dimorphism in canine size is
also apparent. Premolar reduction is common, and Pl/pl is always
DEFINING FEATURES OF THE
absent. The upper molars tend to have a simple pi shape with
FAMILY AMYNODONTIDAE
emphasis on ectoloph shear (Figure 40.2B). M3 architecture is very
different from the triangular pattern of other rhinocerotoids, being
CRANIAL quadratic with a distinct valley between the metaloph and metastyle.
The protolophid and hypolophid on the lower molars are oriented
A preorbital fossa is the most significant cranial synapomorphy.
posterolingually rather than lingually as in other rhinocerotoids. The
Changes in the relative size and position of this fossa are correlated
external groove separating the trigonid from the talonid is reduced
with structural modification of the skull in various lineages. Prim-
or absent.
itive amynodontids (e.g., Amynodon) have a large fossa anterior to
the orbit. The fossa extends medial to the orbit in cadurcodon-
tines, producing a distinct antorbital bar. In metamynodontines, the
POST CRANIAL
preorbital fossa is reduced.
Amynodontids have a large sagittal crest (Figure 40.2A), a fea- The postcranial skeleton is unknown for some taxa, so only a few
ture also seen in other primitive ceratomorphs, such as Hyrachyus, generalities can be given here. Primitive amynodontids are sub-
and may merely represent a primitive retention. The emphasis on cursorial to mediportal. Cadurcodontines apparently retained the
vertical ectoloph shear in amynodontids, however, indicates that a locomotor traits of their primitive relatives, but the postcranial skele-
strong attachment site for the temporalis muscle was important in ton is poorly known in advanced members of this tribe. Metamyn-
this family. odontines are heavy-limbed, barrel-chested animals exhibiting

583
 William P. Wall 584

the rhinocerotoid families: Amynodontidae, Hyracodontidae, and

Rhinocerotidae. Osborn pointed out the large size of the canines and
shortened facial region as means of separating amynodontids from
other rhinocerotoids (this latter characteristic is true only for ad-
A
vanced amynodontids ). Since the tum of the century, Wood ( 1927),
Gromova (1954), Belyayeva (1971), and Wall (1982a) have refined
the family diagnosis of amynodontids.
The monophyly of the Rhinocerotoidea and separate family sta-
tus for amynodontids is accepted today by most authors (Prothero,
Manning, and Hanson, 1986; Wall and Manning, 1986); how-
ever, the relationship between amynodontids and other rhinocero-
B toids is not as clear. Radinsky (1969) argued that amynodontids
evolved from a tapiroid stock independently from hyracodontids
and rhinocerotids. Prothero (this volume, Chapter 41) places Amyn-
odontidae within the Rhinoceratoidea as the sister taxon to Hyra-
c codontidae plus Rhinocerotidae.

INFRAFAMILY
Kretzoi (1942) presented the first serious interpretation of the re-
lationships between amynodontid genera, dividing them into four
subfamilies: Amynodontinae, Metamynodontinae, Paramynodon-
tinae, and Cadurcotheriinae. Based on study of new material, Wall
D (1982a) and Wall and Manning (1986) modified Kretzoi's classifi-
cation. In this chapter only two amynodontid subfamilies are recog-
nized, Rostriamynodontinae, including only Rostriamynodon from
the middle Eocene of Asia, and Amynodontinae, including all other
amynodontids ( Caenolophus, a problematic primitive Asiatic amyn-
odontid, is not included in this classification). The Amynodontinae
Figure40.2. A. Skull of Metamynodon (from Osborn (1898)). B. Upper and is divided into three tribes based on the relative development of the
C. Lower dentition of Amynodon, by Rick Spears. D. Skeleton of Amynodon, preorbital fossa. Amynodon, the sole member of the Amynodontini,
by Rick Spears.
has a large preorbital fossa that does not extend medial to the orbit.
Amynodontopsis, the North American member of the Cadurcodon-
numerous skeletal adaptations associated with riverine habits in tini, has a large preorbital fossa extending medial to the orbit. The
modem hippos (Figure 40.2D). The manus has four digits and the Metamynodontini, including Megalamynodon and Metamynodon,
pes three. has a reduced preorbital fossa. The phylogenetic relationships of
North American amynodontids are shown in Figure 40.3 (see also
Wall, 1982a, 1982b, 1989).
SYSTEMATICS

INCLUDED NORTH AMERICAN GENERA

SUPRAFAMILY
IN FAMILY AMYNODONTIDAE
The first known amynodontid fossil is a skull from the Uintan beds
of Utah described by Marsh in 1877 that he named Amynodon The locality numbers listed for each genus refer to the list of unified
advenus (= Diceratherium advenum, Marsh, 1875). Marsh did localities in Appendix I. The acronyms for museum collections are
not separate Amynodon from other rhinocerotids above the generic listed in Appendix III.
level. Amynodontids were placed in a separate family by Scott and
Osborn (1883) to differentiate two middle Eocene genera, Amyn- Amynodon Marsh, 1877
odon and Orthocynodon, from hyracodontids and rhinocerotids. Type species: Amynodon advenus (Marsh, 1875) (=Dice-
Scott and Osborn (1887) amended the family diagnosis after the dis- ratherium advenum).
covery of a new genus, M etamynodon, in the White River Oligocene, Type specimen: YPM 11763.
citing several important amynodontid characters: loss of Pl and Characteristics: Preorbital fossa large, but not extending me-
enlargement of the sagittal crest. Osborn ( 1889) was the first to rec- dial to the orbit; preorbital portion of skull long com-
ognize the uniqueness of the preorbital fossa and quadratic M3 of pared to other North American amynodontids; nasal inci-
amynodontids, and also presented the first detailed comparison of sion ending above the canine-cheek tooth diastema; skull
 Amynodontidae 585

Amynodontopsis Metomynodon
Amynodon
Megolomynodon

Figure 40.3. Interrelationships of North American amynodontids. Key to number on cladogram: (1) Quadratic M3, enlarged canines, loss of upper and
lower Pl. (2) Preorbital region of skull reduced. (3) Preorbital fossa extends medial to the orbit. (4) Preorbital fossa reduced, zygomatic arches thickened,
skull increasingly brachycephalic. (5) Loss of P2, lower cheek teeth narrow with trigonid and talonid confluent labially.

dolichocephalic, with a relatively narrow occiput; p2 large Characteristics: Preorbital portion of skull intermediate in
and double rooted; length of upper and lower premolar se- length between Amynodon and Metamynodon; zygomatic
ries approximately one half that of molar series; distinct arch large and thickened, skull mesaticephalic; p2 absent,
labial groove separating trigonid from talonid. p3 large and double rooted; labial groove separating trigo-
Average length of m2: 29.0-42.0 mm. nid from talonid greatly reduced.
Included species: A. advenus (known from localities CC5 Average length ofm2: 44.0-54.0 mm.
or CC6B, CC7B, CC7C, CC7D, CC7E, CC9B, SB43A, Included species: M. regalis only (known from localities
SB44A, CP6A, CP6B, CP7 A, CP7B, CP29B, CP29C, CP7 A, CP7C, NP22).
CP36A, CP36B, CP38C, CP38D, NP22); A. reedi (locality
CC4). Metamynodon Scott and Osborn, 1887
Type species: Metamynodon planifrons Scott and Osborn,
Amynodontopsis Stock, 1933 1887.
Type species: Amynodontopsis bodei Stock, 1933. Type specimen: MCZ 9157.
Type specimen: LACM/CIT 150/1087. Characteristics: Preorbital fossa greatly reduced; preorbital
Characteristics: Preorbital fossa large and extending far back portion of skull reduced to an extent comparable to Amyn-
medial to the orbit; preorbital portion of the skull greatly odontopsis but orbits are positioned high on the skull; zy-
reduced, orbit positioned low on skull; nasal incision end- gomatic arch very large and massive; skull brachycephalic;
ing above Ml; skull dolichocephalic; p2 single rooted or p2 absent, p3 small or absent; trigonid and talonid conflu-
absent thereby reducing premolar series length compared ent labially.
to molar series length; labial groove separating trigonid Average length ofm2: 47.0-62.0 mm.
from talonid reduced compared to Amynodon. Included species: M. planifrons (known from localities
Average length of m2: 35.0-40.0 mm. GC24, CP84A, CP84B, NP50B); M. chadronensis (local-
Included species: A. bodei only (known from localities ities SB44, CP83C); M. mckennyi (locality SB43B).
CC9B, SB43C, CP82, NP22).

Indeterminate amynodonts
Megalamynodon Wood, 1945
Type species: Megalamynodon regalis Wood, 1945. An indeterminate amynodont (= "Procadurocodon") is re-
Type specimen: CM 11953. ported from locality PN5B. Note that "Mesamynodon" Peterson,
 William P. Wall 586

w
zw L L.BI. 1.8
2.4
E.BI
(.)
4.5
0 E Lat. Hp.
5.2
:J L.Hp.
6.0
CL. L.E.Hp.
E.E.Hp.
7.0 -
8.8
L L..CI.
9.5

E.CI.
11.0
w
z LL.Ba.
w E.L.Ba.
12.5
(.)
M
0 14.0

:i E.Ba.

15.8
L.Hm.

17.5
E.Hm,
18.8
E LL.Ar.
19.2

E.LAr.
23.0

w L.E.Ar.
z
w L
(.)
27.7
0
CJ E.E.Ar.

:J 29.4
0
WI.
E 31.9 -
Or. •
c:
L.Ch.
33.4
34.5
.!!,!
Ill
c:
.g -c:
_.g_
~c:
0
L M.Ch.
35.5 0 -.g -(.)-

E
-:g[
E.Ch.
§, _o
Ou
37.1
.g E §,
0
§,
.!!! E ~
39,5 l§, -.s -0
E
~ ~ ~
-.gc:
L.Un.
q:
41.3
w
z E.Un. 0
w §,
(.)
0
45.9
-eq:
w M
L.Br.
46.7

M.Br.

49.0
E. Br.
50.4

L.Wa.

53.5
E M.Wa.
54.2
E.Wa.

E.E.Wa.
55.0
55.5
-
Ck.3 55.6
w Ck.2 55.8
z Ck.1
Tl.6
56.0
56.2
w L Tl.5
Tl.4
57.0 -
(.) 57.3
Tl.3
0 Tl.2
58.5
w Tl.1
59,2
..J To.3
60.9
<C To.2
61.3
CL. 62.5
To.1
E Pu.3
63.B
64.0
Pu.2 64.5
Pu.1 65.0
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. =Tiffanian, To.= Torrejonian, Pu. = Puercan.

Figure 40.4. Temporal ranges of North American amynodontid genera.

 Amynodontidae 587

1931, reported from locality CP7 by Anderson and Picard (1972), nean (late middle Eocene) fauna and not much is known about its
is not an amynodont and may be synonymous with Hyracodon. postcranial skeleton. This genus does, however, exhibit similarities
to Paramynodon from the late Eocene of Asia. Paramynodon clearly
represents an early stage in the transition of metamynodontines to
BIOLOGY AND EVOLUTIONARY PATTERNS a hippo like mode of life.
Asiatic amynodontids show a similar adaptive radiation around
Migration between North America and Asia via Beringia played a the Eocene-Oligocene boundary. Comparison of Asiatic and North
significant role in amynodontid evolution. Asia appears to be the American amynodontids from this time period reveals a complex
center of origin for the family. Caenolophus, a possible ancestral interchange between these two regions. Cadurcodontines appear
form, and Rostriamynodon, the oldest well-defined amynodontid, to have evolved more rapidly in North America. Amynodontopsis
come from the early or middle Eocene Irdin Manha Beds of Asia. displays an evolutionary stage not equaled in Asia until the early
The oldest occurrence of amynodontids in North America is Amyn- Oligocene. Dispersal of Amynodontopsis into Asia may account
odon from the middle Eocene Uinta Beds. Taxonomic diversity also for the radiation of cadurcodontines there. An undescribed taxon
appears to be greater in Asia, but some of those taxa may not be from the early Oligocene of Asia is similar to Amynodontopsis and
valid (see Wall and Manning, 1986). either belongs in the same genus or to a sister taxon (see Wall,
North American amynodontids show close evolutionary relation- 1980, l 982a, 1989). Cadurcodontines were more successful in Asia
ships to Asiatic members of the family. Amynodontopsis is more than in North America, surviving well into the middle Oligocene;
closely related to the Asiatic genera Sianodon and Cadurcodon North American Amynodontopsis probably became extinct prior to
than to any North American amynodontid. Although Megalamyn- the Chadronian (late Eocene). The picture for metamynodontine
odon and Metamynodon are more closely related to each other than radiation is not as clear. Paramynodon and Megalamynodon exhibit
to other North American amynodontids, the sister taxon for each, roughly equivalent early stages in metamynodontine evolution, but
Paramynodon and Zaisanamynodon, respectively, is Asiatic. more advanced forms were also present in both continents at the
The radiation of amynodontids into North America during the same time. Zaisanamynodon in Asia and Metamynodon (Wilson
early Uintan came at a time of significant environmental change. and Schiebout, 1981) exhibit comparable evolutionary stages during
During the Bridgerian (early middle Eocene), North America was the late Eocene.
covered by lush "tropical" forests adapted to a homogeneous equable Environmental change probably played an even more important
humid climate (Savage and Russell, 1983). During the Uintan (later role in amynodontid evolution during the latest Eocene, when sea-
middle Eocene), however, mountain building and increasing aridity sonality was even more pronounced and habitats were becoming
were having an impact on the ecosystem. By the end of the Uintan, more open (Savage and Russell, 1983). Cadurcodontines probably
barriers to widespread dispersal in the form of a more heterogeneous did not survive past the Duchesnean in North America. Metamyn-
environment became important (Lillegraven, 1979). These barriers odontines, however, fared better than their tapirlike counterparts.
could be responsible for the modest radiation of amynodontids dur- Metamynodon chadronensis is a rare component of the late Eocene
ing the middle to late Eocene. fauna. M. planifrons from the Chadronian Byram Formation of
Amynodon is one of the best known genera in the family, but the Mississippi (locality GC24) is probably the oldest occurrence of
life habits of this animal are still misinterpreted. Its skeleton ex- this taxon (Manning, Dockery, and Schiebout, 1986). This species
hibits none of the traits associated with semiaquatic life found in is better known from the early Oligocene beds of the northwest-
Metamynodon (Wall, 1981), and it was most likely a subcursorial ern plains. Metamynodon fossils are typically found in channel
to mediportal terrestrial browser. Its broad geographic range, com- sandstones, a fact that strengthens the argument that this animal
bined with well-developed locomotor abilities, makes it likely that was semiaquatic. Metamynodon apparently inhabited river systems
Amynodon preferred the open forest-savanna habitat that dominated throughout much of North America, but except for the Orellan this
North America during the Uintan (Savage and Russell, 1983). animal was a rare component of the Oligocene fauna. By the end of
Climatic change continued to bring about an increase in habi- the Whitneyan (late early Oligocene), perhaps earlier, amynodon-
tat diversity through the Duchesnean, which also represents a time tids became extinct in North America. As previously discussed, the
of significant change in amynodontid evolution. Members of the family was more successful in Asia, with one taxon surviving into
two advanced tribes, Cadurcodontini and Metamynodontini, make the Miocene of India.
their first appearance at this time. These two tribes exhibit dis-
tinctly different adaptations, indicating a dichotomous radiation of
amynodontids (Wall, l 982a, 1989). Cadurcodontines evolved in a REFERENCES
direction leading to life habits similar to modern tapirs, Advanced -------------------------
members of the tribe display numerous cranial features indicating Anderson, D. W., & Picard, M. D. (1972). Stratigraphy of the Duchesne River
they had a tapirlike proboscis. Amynodontopsis displays many of Formation (Eocene-Oligocene), northern Uinta Basin, northeastern
these characteristics, including development of large snout mus- Utah. Bulletin of Utah Geological and Mineral Survey, 97, 1-29.
Belyayeva, E. l. (1971). Novye dannye po aminodon tam SSSR. (New data
cle scars; enlarged nasal incision; and reduced nasal bones (Wall, on the amynodonts of the USSR). Akademia Nauk, USSR, Trudy
1980). Metamynodontines became specialized for an amphibious Paleontological Institute, 130, 39-61.
mode of life. Megalamynodon is a rare component of the Duches- Gromova, V. (1954). Boltany noosorogi (Amynodontidae) Mongolic.
 William P. Wall 588

Akademia Nauk, USSR, Trudy Paleontological Institute, 55, River Formation contained in the Museum of Comparative Zool-
85-189. ogy. Bulletin of the Museum of Comparative Zoology, Harvard, 13,
Kretzoi, M. (1942). Auslandische saugetierfossilien der Ungarischen Mu- 151-71.
seum. Foldtony Koslony, 72, 139-48. Stock, C. (1933). An amynodont skull from the Sespe deposits, California.
Lillegraven, J. A. (1979). A biogeographical problem involving comparisons Proceedings of the National Academy of Science, 19, 762-7.
of late Eocene terrestrial vertebrate faunas of western North Amer- Wall, W. P. (1980). Cranial evidence for a proboscis in Cadurcodon and
ica. In Historical Biogeography, Plate Tectonics, and the Changing a review of snout structure in the family Amynodontidae (Perisso-
Environment, eds. J. Gray & A. J. Boucot, pp. 333-47. Corvallis: dactylas, Rhinocerotoidea). Journal of Paleontology, 54, 968-77.
Oregon State University Press. (1981). Systematics, Phylogeny, and Functional Morphology of the
Manning, E., Dockery, D. T., III, & Schiebout, J. A. (1986). Preliminary Amynodontidae (Perissodactyla: Rhinocerotoidea). Unpublished
report of a Metamynodon skull from the Byram Formation (lower Ph.D. Thesis. Amherst: University of Massachusetts.
Oligocene) in Mississippi. Mississippi Geology, 6, 1-16. (1982a). Evolution and biogeography of the Amynodontidae (Perisso-
Marsh, 0. C. (1875). Notice of new Tertiary mammals. American Journal dactyla, Rhinocerotoidea). Proceedings ofthe Third North American
of Science, 3, 239-50. Paleontological Convention, 2, 563-7.
(1877). Notice o{ some new vertebrate fossils. American Journal of Sci- (1982b). The genus Amynodon and its relationship to other members of
ence, 14, 249-56. the Amynodontidae (Perissodactyla, Rhinocerotoidea). Journal of
Osborn, H.F. (1889). The Mammalia of the Uinta Formation, Part III: The Paleontology, 56, 434-43.
Perissodactyls. Transactions of the American Philosophical Society, (1989). The phylogenetic history and adaptive radiation of the Amyn-
16, 505-30. odontidae (Perissodactyla, Rhinocerotoidea). In Evolution of the
(1898). The extinct rhinoceroses. Memoirs of the American Museum of Perissodactyla, eds. D. R. Prothero & R. M. Schoch, pp. 341-54.
Natural History, 1, 75-164. Oxford: Clarendon.
Prothero, D.R., Manning, E., & Hanson, C. B. (1986). The phylogeny of the Wall, W. P., & Manning, E. (1986). Rostriamynodon grangeri n. gen., n. sp.
Rhinocerotoidea (Mammalia, Perissodactyla). Zoological Journal of of amynodontid (Perissodactyla, Rhinocerotoidea) with comments
the Linnean Society, 87, 341---06. on the phylogenetic history of Eocene Amynodontidae. Journal of
Radinsky, L.B. (1969). The early evolution of the Perissodactyla. Evolution, Paleontology, 60, 911-19.
23, 308-23. Wilson, J. A., & Schiebout, J. A. (1981). Early Tertiary vertebrate fau-
Savage, D. E., & Russell, D. E. (1983). Mammalian Paleofaunas of the nas, Trans-Pecos Texas: Amynodontidae. Texas Memorial Museum,
World. Reading, MA: Addison-Wesley. Pearce-Sellards Series, 33, 1-62.
Scott, W. B., & Osborn, H.F. (1883). On the skull of the Eocene rhinoceros Wood, H. E. (1927). Some early Tertiary rhinoceroses from the Eocene of
Orthocynodon, and the relation of the genus to other members of the Mongolia. Bulletin of American Paleontology, 13, 165-249.
group. Contributions E. M. Museum and Archives, Princeton College (1945). Family Amynodontidae. In The Mammalia of the Duchesne
Bulletin, 3, 1-22. River Oligocene (W. B. Scott). Transactions of the American Philo-
(1887). Preliminary account of the fossil mammals from the White sophical Society, 34, 209-52.
41 Hyracodontidae
DONALD R. PROTHERO

INTRODUCTION

The hyracodonts were a group of rhinocerotoids that flourished

during the middle and late Eocene and Oligocene of Eurasia and
North America (Radinsky, 1967; Prothero, Guerin, and Manning,
1989; Prothero, 1996). Their most distinctive feature is their elon-
gate limbs. Primitive hyracodonts were about the size of sheep
(Figure 41.1) and quite cursorial. One group of hyracodonts, the
paraceratherines, grew to gigantic size. An Oligocene hyracodont
from Asia, Paraceratherium (= Indricotherium, Baluchitherium)
reached 6 m (18 feet) at the shoulder and weighed up to 20,000 kg
(Lucas and Sobus, 1989). It was the largest land mammal that ever
lived, and pr,obably browsed on the tops of trees. Smaller hyra-
codonts, however, were among the most cursorial animals of their
Figure 41.1. Restoration of Hyracodon, by Brian Regal.
time. They had relatively simple browsing dentitions, and large
herds of them probably roamed the mixed forest-grasslands of the
Oligocene, browsing on low vegetation. The hyracodont Triplo- and a deep mandible with an upward-curved symphysis (Fig-
pus was very common in the middle Eocene. Together with the ure 41.2A).
small primitive paraceratherine Forstercooperia, it migrated freely
between Eurasia and North America. In the late Eocene of North DENTAL
America, there were two genera ofhyracodonts. By the late Chadro-
nian, the group was reduced to the small cursorial genus Hyracodon, Most genera had a full complement of incisors and canines, which
whose last occurrence is in the lowermost Sharps Formation in South were small spatulate or conical teeth. A few of the more specialized
Dakota (earliest Arikareean), about 30 Ma (Tedford et al., 1985, forms lost incisors and developed other anterior teeth into tusks.
1996). There is typically a short diastema between the anterior and cheek
teeth, and in some taxa, both the rostrum and the symphysis are later-
ally constricted. The upper molars show the characteristic rhinocero-
DEFINING FEATURES OF THE toid pi pattern formed by are were not fully molarized, and there are
FAMILY HYRACODONTIDAE various combinations of crests uniting the protocone and hypocone.
The most characteristic tooth is the M3, which has a highly re-
duced, lingually deflected metacone, not the large, labially deflected
CRANIAL
ridge of amynodonts or the triangular M3 without a metastyle of
In most features of the skull and skeleton, hyracodonts differ lit- rhinocerotids. Most hyracodont molars also have reduced parasty Jes.
tle from more primitive rhinocerotoids, such as Hyrachyus. The The lower dentition shows the stereotyped rhinocerotoid pattern of
skull has a well-developed sagittal crest, slender but flaring zygo- L-shaped paralophids and metalophids connecting the cusps of the
matic arches, an unretracted nasal notch and elongate nasal bones, trigonid and talonid, and are not diagnostic of the group.

589
 Donald R. Prothero 590

of shared primitive characters, I agree with his choice of inclusion of

North American taxa in this family (Figure 41.3). The relationships
of the Rhinocerotoidea have been discussed by Prothero, Manning,
A and Hanson (1986). Hyracodontids are clearly the sister group of
the Rhinocerotidae, and the Amynodontidae are a more primitive
sister taxon. The family Hyracodontidae is defined by the shared de-
rived presence of elongate limbs and metapodials, a tridactyl manus
(i.e., metacarpal V greatly reduced), and also by a reduced, lingually
deflected metacone on M3.

INFRAFAMILY
Two monophyletic subfamilies can be recognized. The subfamily
B Indricotheriinae includes all the gigantic forms of Eurasia, plus their
smaller sister tax on Forstercooperia from the middle Eocene of both
Eurasia and North America (Lucas, Schoch, and Manning, 1981).
They are recognized by their greatly enlarged size, enlarged p 1, and
reduced M3 metacone (paralleling the rhinocerotids). The subfam-
ily Hyracodontinae includes the smaller, more cursorial forms of
the late Eocene and Oligocene of Eurasia and North America. They
are defined by their flattened premolar ectolophs and greatly re-
c duced parastyles. Of these, only Trip/opus is common in the middle
Eocene, and Hyracodon in the late Eocene and Oligocene; the other
two taxa are known only from a few localities.

Figure 41.2. Osteology of Hyracodon nebraskensis. A. Skull and mandible INCLUDED NORTH AMERICAN GENERA
(scale bar= 5 cm). B. Upper and lower dentitions in crown view (scale bar IN THE FAMILY HYRACODONTIDAE
= 5 cm). C. Articulated skeleton (scale bar= 50 cm). (After Scott, 1941.)
The locality numbers listed for each genus refer to the list of unified
POST CRANIAL localities in Appendix I. The acronyms for museum collections are
listed in Appendix III.
The skeleton is generally primitive in overall proportions and not The locality numbers may be listed in a couple of alternative ways.
very different from that of Hyrachyus, except in one area. Hyra- Parentheses around the locality (e.g., [CP101]) mean the taxon in
codont limbs and particularly the feet are much more elongate and question at that locality is cited as an "aff." or "cf." the taxon in
gracile than those found in any other group of rhinocerotoids. All question. Parentheses are usually used for individual species, thus
hyracodonts have a tridactyl pes, unlike amynodonts and primitive implying the genus is firmly known from the locality, but the actual
rhinocerotids. The metapodials are long and slender (or laterally species identification may be questionable. Question marks in front
compressed), a sure indication of a more cursorial habit. The family of the locality (e.g., ?CP 101) mean the taxon is questionably known
Hyracodontidae is primarily diagnosed by its limb proportions. Iron- from that locality, thus implying some doubt that the taxon is actually
ically, these proportions persist even when individual taxa become present at that locality, either at the genus or the species level.
gigantic. The largest indricotheres still had long limbs and toes,
even though they were larger than graviportal elephants, titanoth-
eres, rhinocerotids, and even most dinosaurs, whose metapodials INDRICOTHERIINAE
are greatly compressed dorsoventrally. The gigantic indricotheres
Forstercooperia Wood, 1939
are a clear example of how inherited patterns can persist in a group,
Type species: Forstercooperia totadentata (Wood, 1938).
even after they have moved to a completely different adaptive niche.
Type specimen: AMNH 20116.
Anatomy is not infinitely adaptive and plastic, but subject to inher-
Characteristics: As discussed by Lucas, Schoch, and Manning
ited constraints.
(1981 ), Forstercooperia is a small indricothere found in the
middle and late Eocene of both Asia and North America. It
SYSTEMATICS is the primitive, small sister taxon of all the giant Oligocene
indricotheres. The nasal incision is above the canine, and
the upper incisors are conical and pointed. Unlike most
SUPRAFAMILY
indricotheres, it has a full dentition with unmolarized P3.
The family Hyracodontidae was last fully revised by Radinsky The M3 metacone is variable, but usually absent. Its crown
(1967). Although I disagree with some of his conclusions and his use height index is about 0.6. The manus is fully tridactyl.
 Hyracodontidae 591

incisors, and a very faint metacone on M3. As the name

implies, the manus is tridactyl.
Average length of M2: 11.0-17.3 mm.
Included species: T. cubitalis (known from locality CP38C);
T. obliquidens (= T. amorarum) (localities CP6A, CP6B,
CP38D); T.? woodi (localities CC9C, CC9BB); T. impli-
catus (localities [SB44A], CP6B, CP38C); T. rhinocerinus
(localities [SB43A], [NP22]).
Trip/opus sp. is also known from localities SB43B, CP65.

Hyracodon Leidy, 1856

Type species: Hyracodon nebraskensis (Leidy, 1850).
Type specimen: USNM 336207.
Characteristics: Hyracodon is diagnosed as a medium-sized
hyracodontid, with pointed incisors and small canines. The
dPl is replaced by Pl, and pl is usually lost. Premolar and
1. HYRACODONTIDAE molar paracones are not separated posteriorly from the
ectolophs. M 1-M2 have a large antecrochet and protocone
set off with vertical grooves. M3 has a persistent small
metacone.
Figure 41.3. Interrelationships of the Hyracodontidae (after Prothero,
Manning, and Hanson, 1986). Characters at the nodes are as follows: Average length ofM2: 15.0-25.0 mm.
(1) HYRACODONTIDAE: tridactyl manus; long limbs and metapo- Included species: H. nebraskensis (known from localities
dials; laterally compressed carpus and tarsus. (2) INDRICOTHERI- CP40A,CP40B,CP41A,CP41B,CP 68B,CP68C,CP68D,
INAE: Large size; enlarged pl; M3 metacone reduced. (3) HYRA- CP84A,CP84B,CP84C,CP85C,CP 98B,CP98C,CP99A,
CODONTINAE: flattened premolar ectolophs; parastyles further reduced.
CP99B, CP99C, NPlOB, NP50C); H. primus (localities
(4) Increased hypsodonty (crown height index greater than 0.75); labial cin-
gulum on P3-4; Pl replaces dPl; p 1 rarely present; paracone folded toward SB44B, SB44C, SB44D, CP7C, CP42A); H. petersoni
parastyle on Ml-2; p2-4 labial cingula; high paralophids and metalophids; (= H. selenidens) (localities SB43E, SB44E, CP39IIA,
sharply constricted symphysis and rostrum. (5) Lower canine reduced; elon- CP42A, CP98A, NPlOA, NPlOB); H. priscidens (locali-
gate narrow M2; oblique cross lophs; Ml-2 metacones deflected labially ties [CP39B], CP39C, [CP39F], CP39G, CP42A, [CP83A],
with sharp crease; trace of M3 metacone lost.
[CP83B], CP83C, NPlOB, NP24C, NP25C); H. leidyanus
(localities CP68D, CP84B, CP84C, CP85C, CP99B).
Average length ofM2: 19.0-26.0 mm. Hyracodon sp. is also known from localities ?SB52,
Included North American species: F. grandis (known from (CP43), ?CP45A, CP83C, CP85A, CP86A, CP99C, NP9A,
localities CP6A, CP38C); F. minuta (localities SB25B, NP24E, (NP25B), NP27C, NP32B.
CP36B).
Comments: Hyracodon (see Figures 41.1, 41.2) is the most
Comments: In an paper published after this manuscript was common and best known of the hyracodontid rhinoceroses
submitted, Holbrook and Lucas ( 1997) claim that the North because many skulls and skeletons have been found in the
American specimens of Forstercooperia do not belong to White River beds of the High Plains. It is the largest of the
this taxon, but to an new taxon, Uintaceras radinskyi, that hyracodontine hyracodonts and also the most cursorial. It
forms the sister taxon of the Rhinocerotidae. If their asser- is known from the Duchesnean to the latest Whitneyan
tions are correct, Indricotheriinae are confined to the Asian (about 40-30 Ma), so its IO-million-year range is among
subcontinent. the longest in the mammals. The genus was greatly over-
split based on subtle variations in the upper premolars,
leading Radinsky to lump them all into one species. Many
HYRACODONTINAE authors (e.g., Wilson and Schiebout, 1984) considered this
lumping to be excessive, and I recognize only five easily
Trip/opus Cope, 1880
distinguished species (Prothero, 1996).
Type species: Trip/opus cubitalis Cope, 1880.
Type specimen: AMNH 5095. Epitriplopus Wood, 1927
Characteristics: Trip/opus is the commonest rhinocerotoid Type species: Epitriplopus uintensis (Peterson, 1919).
from the Uintan of North America. It was once greatly Type specimen: CMNH 3007 a.
oversplit taxonomically, but most of the invalid genera and Characteristics: Epitriplopus is a small, relatively hypsodont
species have been eliminated by Radinsky (1967). Trip/o- (crown height index = 0.78) hyracodontid from the late
pus has almost no derived characters of its own, but is Uintan and Duchesnean. It is recognized by its upper mo-
diagnosed by its size (slightly larger than Hyrachyus and lars, which have curved cross crests and paracones project-
smaller than later rhinocerotoids), its equal-sized spatulate ing labial to the parastyles. Ml-M2 are relatively long and
 Donald R. Prothero 592

M.Br.
49.0
E.Br.
INDRICOTHERllNAE HYRACODONTINAE
50.4

L.Wa.
53.5
E M.Wa.
54.2
E.Wa.

E.E.Wa. ::::
Ck.3 55.6
L----------- -----------= --==--=:--= :---:=---:=- -=:--=:-=-- :=--:::--:j
w Ck.2 55.8
56 o
z
w
Ck.1
TI.6 56:2 -
L Tl.5 57.0
(.) Tl.4 57.3
0 ~:: 58.5
w Tl.1
..J
<( To.3
To.2
11.
E i:.~ 63.8
64.0 -
PU.2 64.5
PU.1 65,0
Pu.o 65.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= ~emingfordlan? Ar. ".' Arikareean, .
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un. = Umtan, Br.= Bridgerian, Wa. = Wasatch1an,
Ck. = Clarkforkian, Ti. = Tiffanian, To.= Torrejonian, Pu. = Puercan.

Figure 41.4. Temporal ranges of North American hyracodontid genera.

 H yracodontidae 593

narrow, with a pronounced antecrochet. The M3 metacone between North America and Asia during the late Eocene because
is completely lost. several genera occur in both regions at the time. Yet within North
Average length of M2: 11.3 mm. America, they were very restricted. Trip/opus occurs in great abun-
Included species: E. uintensis only (known from localities dance in the Washakie Basin of Wyoming and in the Uinta Basin
CP6B, CP7 A, CP7C, CP29C, [CP36]). of Utah, but was completely absent from the late Eocene of Cal-
ifornia (Golz and Lillegraven, 1977) and the Clarno Formation of
Triplopides Radinsky, 1967 Oregon (which had the oldest North American rhinocerotid, and
Type species: Triplopides rieli Radinsky, 1967. an amynodont). It was one of many organisms that showed the en-
Type specimen: CMNH 9370. demism typical of the Uintan (Lillegraven, 1979). Throughout the
Characteristics: Triplopides is a small hyracodontid from the Chadronian, Orellan, and Whitneyan, Hyracodon remained a rare
early Chadronian McCarty's Mountain locality in Mon- but ubiquitous member of the fauna. When it vanished, the group
tana. It is known from the type skull, which is badly crushed also disappeared from North America. In Eurasia, however, the gi-
and missing the dorsal portion, and a mandible. The in- gantic indricotheres persisted into the early Miocene.
cisors are very spatulate, and the canine is small and slen-
der. The premolars are greatly shortened relative to the
molars, and both upper and lower first premolars are lost. REFERENCES
As in Epitriplopus, the cross crests are obliquely deflected
and the Ml-M2 are very long and narrow. The paracones
Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedi-
on these teeth project labially, and the protocones are not mentation, stratigraphy, paleoecology, and paleoclimatology in the
constricted. M3 has no trace of a metacone. Big Badlands of South Dakota. Fieldiana Geology Memoir, 5,
Average length ofM2: 25.0 mm. 1-158.
Included species: T. rieli only, known from locality NP25C Cope, E. D. (1880). A new genus of tapiroids. American Naturalist, 14,
382-3.
only.
Golz, D. J., & Lillegraven, J. A. (1977). Summary of known occurrences
of terrestrial vertebrates from Eocene strata of southern California.
University of Wyoming Contributions in Geology, 15, 43-65.
INDETERMINATE HYRACODONTIDS Holbrook, L. T., & Lucas, S. G. (1997). A new genus of rhinocerotoid from
the Eocene of Utah and the status of North American "Forstercoope-
Fragmentary remains ascribed to hyracodontids have also been re-
ria." Journal of Vertebrate Paleontology, 17, 384-96.
ported from localities CC7B, CP46, NP9, NP22, NP29C, CP86A. Leidy, J. (1850). [Untitled note]. Proceedings of the Academy of Natural
Sciences of Philadelphia, 5, 121.
(1856). Notices of several genera of extinct Mammalia, previously less
BIOLOGY'AND EVOLUTIONARY PATTERNS perfectly characterized. Proceedings of the Academy of Natural Sci-
ences of Philadelphia, 8, 91-2.
Lillegraven, J. (1979). A biogeographical problem involving comparisons of
As many authors have noted, the elongate limbs and feet of hyra- later Eocene terrestrial vertebrate faunas of western North America.
codonts clearly indicate that they were among the most cursorial In Historical Bio geography, Plate Tectonics, and the Changing Envi-
ungulates in the late Eocene and Oligocene of North America. Their ronment, eds. J. Gray & A. J. Boucot, pp. 333-46. Eugene: University
limb proportions are comparable with the horses of their time. Their of Oregon Press.
Lucas, S. G., Schoch, R. M., & Manning, E. (1981). The systematics of
teeth are among the most hypsodont of any mammal of the time, so
Forstercooperia, a middle to late Eocene hyracodontid (Perisso-
they may have been able to browse on tougher vegetation. The total dactyla: Rhinocerotoidea) from Asia and western North America.
sample size of hyracodonts has always been small, so it is difficult Journal of Paleontology, 55, 826-41.
to make many inferences about their ecology. It is possible they Lucas, S. G., & Sobus, J. (1989). The systematics of indricotheres. In The
may have roamed in herds, although no mass death assemblages Evolution of Perissodactyls, eds. D. R. Prothero & R. M. Schoch,
pp. 358-78. Oxford: Clarendon.
have been found to my knowledge. There is no obvious sexual di-
Peterson, 0. A. ( 1919). Report upon material discovered in the Upper Eocene
morphism or other indications of unusual social structure. Clark, of the Uinta Basin. Annals of the Carnegie Museum, 12, 40-168.
Beerbower, and Kietzke (1967) were not able to find any particu- Prothero, D. R. (1996). Hyracodontidae. In The Terrestrial Eocene-
lar ecological association for Hyracodon. It was rare but present in Oligocene Transition in North America, eds. D.R. Prothero & R. J.
near-stream, open plains, and swampy plains facies. Emry, pp. 652-{i3. Cambridge: Cambridge University Press.
Prothero, D. R., Guerin, C., and Manning, E. (1989). The history of the
Hyracodon split into two species in the Whitneyan, the normal-
Rhinocerotoidea. In The Evolution of Perissodactyls, eds. D. R.
sized H. nebraskensis and the much larger H. leidyanus (Prothero, Prothero & R. M. Schoch, pp. 322-40. Oxford: Clarendon.
1996). It then disappears at the end of the Whitneyan, after a Prothero, D.R., Manning, E., & Hanson, C. B. (1986). The phylogeny of the
IO-million-year history. Its disappearance is probably related to the Rhinocerotoidea (Mammalia, Perissodacty la). Zoological Journal of
the Linnean Society, 87, 341-{i6.
increase in aridity that marked the late Whitneyan and affected many
Radinsky, L.B. (1967). A review of the rhinocerotoid family Hyracodon-
other persistent White River taxa.
tidae (Perissodactyla). Bulletin of the American Museum of Natural
Hyracodontids had a rather peculiar geographic distribution History, 136, 1-45.
(Figure 35.3). They were apparently very successful in migrating Scott, W. B. (1941). The mammalian fauna of the White River Oligocene,
 Donald R. Prothero 594

Part V: Perissodactyla. Transactions of the American Philosophical in North America, eds. D.R. Prothero & R. J. Emry, pp. 312-34.
Society, 28, 747-90. Cambridge: Cambridge University Press.
Tedford, R.H., Swinehart, J.B., Hunt, R. M., Jr., & Voorhies, M. R. (1985). Wilson, J. A., & Schiebout, J. A. (1984). Early Tertiary vertebrate fau-
Uppermost White River and lowermost Arikaree rocks and faunas, nas, Trans-Pecos Texas: Ceratomorpha less Amynodontidae. Pearce-
White River Valley, northwestern Nebraska, and their correlation Sellards Series, Texas Memorial Museum, 39, 1-47.
with South Dakota. Dakoterra, Museum of Geology, South Dakota Wood, H. E., IL (1927). Some early Tertiary rhinoceroses and hyracodonts.
School of Mines and Technology, 2, 335-52. Bulletin of American Palaeontology, 13, 5-105
Tedford, R. H., Swinehart, J. B., Prothero, D. R., Swisher, C. C., King, (1938). Cooperia totadentata, a remarkable rhinoceros from the Eocene
S. A., & Tierney, T. E. (1996). The Whitneyan-Arikareean transition of Mongolia. American Museum Novitates, 1012, 1-20 (addendum
in the High Plains. In The Terrestrial Eocene-Oligocene Transition dated Feb. 23, 1939, replaces Cooperia with Forstercooperia).
42 Rhinocerotidae

DONALD R. PROTHERO

INTRODUCTION

The Rhinocerotidae was one of the most successful groups of

mammals in North America. After the extinction of titanotheres in
the late Eocene, rhinos were the largest land mammals on the conti-
nent until the appearance of mastodonts in the mid-Miocene. They IJf'
occurred in enormous herds, especially in the High Plains Miocene, 1

and some quarries yield thousands of their bones. Rhinocerotids

were very diverse ecologically. There were large hippolike graz-
ers (Teleoceras, Brachypotherium, and Peraceras superciliosum);
prehensile-lipped browsers (most of the aceratherines); four in-
dependent examples of dwarfing (Peraceras hessei, Teleoceras
meridianum, and new undescribed species of Teleoceras and Dicera-
therium); pig-sized herding rhinos (Menoceras arikarense; see Figure 42.1. Restoration of Menoceras, by Brian Regal.
Figure 42.1); and many other less specialized kinds. Rhinocerotids
occupied the large-bodied herbivorous niches in North America
from the early Oligocene to the end of the Miocene, and many
other niches besides. change in the rhinos as well. Diceratherium and Menoceras became
The family Rhinocerotidae first appears in North America, extinct as invading aceratherines (Floridaceras, Peraceras, Aph-
probably as immigrants from Asia, in the Duchesneau (late elops, and a new undescribed genus) and teleoceratines (Brachy-
middle Eocene) of Oregon, California, and Texas. A new genus, potherium) established themselves. From the late Hemingfordian
Teletaceras, has been erected by Hanson (1989) to refer to the until the late Hemphillian (latest Miocene), the rhino faunas were
most primitive North American rhinocerotids from the Clarno For- stable elements of the "Clarendonian chronofauna" (Tedford, 1970;
mation of Oregon. By the latest Eocene (Chadronian) and early Webb, 1984; Tedford et al., 1987). There was a single genus of
Oligocene, there were several rhino genera represented in the High browsing aceratherine (typically Aphelops) and grazing hippolike
Plains: Trigonias, Penetrigonias, Subhyracodon, and Amphicaeno- teleoceratine (typically Teleoceras) in nearly every North American
pus. In the Whitneyan and Arikareean (later Oligocene through ear- locality of Barstovian, Clarendonian, or Hemphillian age.
liest Miocene), however, rhino diversity was reduced to one genus, At the end of the Hemphillian, rhinoceroses were nearly extinct
Diceratherium, which persisted from 31-17 Ma, the longest un- in this continent, along with the extinction of protoceratids, dromo-
opposed reign of any rhino. In the latest Arikareean, the pig-sized merycids, mylagaulid and eomyid rodents, most horses, and antilo-
rhino Menoceras arikarense (misidentified as Diceratherium cooki caprids, totaling 62 genera (Webb, 1984). One isolated tooth scrap
in most books and museum labels) immigrated from Europe. It from the Blancan of Texas is known (Madden and Dal quest, 1990),
occurs in large numbers at the famous Agate Springs Quarries in but no other rhinos are known after the Hemphillian. This extinction
Nebraska. was probably related to the loss of subtropical wooded habitat dur-
During the Hemingfordian (late early Miocene), North American ing the cooling and drying associated with the Messinian worldwide
mammal faunas were in great flux, and there was considerable climatic event.

595
 Donald R. Prothero 596

DEFINING FEATURES OF THE Hanson, 1986, Figure 6). Primitive rhino premolars were not fully
FAMILY RHINOCEROTIDAE molarized, so there may be crests connecting the protocone and
hypocone. In the past, the subtle differences in premolars were used
to oversplit the group (e.g., Wood, 1927). Quarry samples show
CRANIAL that most of this premolar variation is normal intrapopulation vari-
Despite the popular association of rhinos with horns, most extinct ability, and thus the species lists here reflect considerable lump-
rhinos were hornless. Rhino horns are made of agglutinated hair ing. Lower teeth, on the other hand, have the stereotyped L-shaped
loosely attached to a rugose area on the skull, so they are rarely protolophid and metalophid, which changes only in becoming more
preserved. Nevertheless, from the rugosities on the skull surface, it hypsodont.
is clear that horns were present in only a few groups. Paired nasal
flanges occur in Diceratherium, and small paired horns (indepen-
POST CRANIAL
dently derived) in Menoceras. Teleoceratines have a small terminal
nasal horn, and m~le Peraceras superciliosum is the only acerather- When prospecting in almost any Oligocene or Miocene locality,
ine to bear a horn. Instead, rhinocerotids must be defined on a large rhino postcranial bones are easily recognized by their size and ro-
suite of cranial, dental, and postcranial characters. bustness compared to any other mammal of the time (except titan-
The skull is low and saddle shaped, with broad parasagittal crests otheres and mastodonts, which were even larger). Only Menoceras
and laterally flared lambdoid crests (Fig. 42.2). The premaxillary- and the dwarf species became more gracile, but even their bones
nasal contact is lost. The nasals are long and slender, and in the cannot be mistaken for any other mammal. In particular, the rhino
aceratherines, the narial notch is deeply retracted. The dentary has axis has a fused intervertebral canal and a deep alar notch in the
a broad ascending ramus with a straight posterior border and a anteromedial margin of the transverse process. There are also deep
distinct postcondylar process, a unique rhinocerotid synapomorphy. dorsal and ventral notches on the anterior articulations of the atlas.
The mandibular condyle is broad, flat surfaced, and horizontal. The medial process of the proximolateral tuberosity of the humerus
is very long and posteromedially inflected. The limbs in teleocer-
atines are very robust and proximodistally shortened; this is espe-
DENTAL
cially true of the manus and pes. Most rhinocerotids had tridactyl
As reviewed by Prothero, Manning, and Hanson ( 1986), the Rhino- feet, but the most primitive ones, like Trigonias, still retain a func-
cerotidae are defined by the presence of a chisel-shaped II occlud- tional fifth metacarpal. This bone also reappears in a few individuals
ing with a tusklike i2. On M3, the metacone is extremely reduced of the aceratherines Aphelops, Peraceras, and Floridaceras, possi-
or absent, and the posterior cingulum is short. In most rhinocerotids bly as an atavism (Prothero, Manning, and Hanson, 1986, p. 359).
except the most primitive ones, the other incisors and canines are Teleoceratines also have a posterior articulation of the scaphoid and
lost. The upper cheek teeth are the most characteristic part of the lunar.
anatomy. The molars typically have the diagnostic Jr-shaped pat- Most of the systematics in this chapter is based on Prothero,
tern formed by the junction of the protoloph, metaloph, and ec- Manning, and Hanson (1986), Prothero and Manning (1987),
toloph. In very derived forms, these teeth can become very hyp- Prothero, Guerin, and Manning (1989), and Prothero (in a com-
sodont and develop various internal ridges (crochets, antecrochets, plete monograph on the North American Rhinocerotidae, which is
and cristae) to increase surface area (see Prothero, Manning, and currently in preparation).

Figure 42.2. Osteology of the Orel-

lan rhinocerotid Subhyracodon occi-
dentalis. A Skull (after Osborn, 1898)
(scale bar= 50 cm). B. Skeleton (after
Scott, 1941) (scale bar= Im).
 Rhinocerotidae 597

SYSTEMATICS Rhinocerotidae, but only a few are monophyletic (Figure 42.3). The
subfamily Aceratheriinae is easily recognized by their retracted nar-
SUPRAFAMILY ial notch and loss ofll. Presumably, these rhinos had a prehensile lip
or snout for browsing. The tribe Teleoceratini are highly derived and
The Hyracodontidae are the closest sister taxon to the Rhinoceroti- easily recognized by their short, stout, proximodistally compressed
dae. The two groups are united by the following derived features: limbs and feet, broad brachycephalic skulls with flaring lambdoid
reduced parastyles; longer cheek tooth series relative to skull length; crests, and nasal bones with a U-shaped cross section supporting a
Ml-M2 metacone flange lengthened; M3 metacone reduced and small terminal horn. They also have an elongate calcaneal tuber and
lingually deflected. a posterior articulation of the scaphoid and lunar.
In the past, there was a paraphyletic subfamily "Caenopinae" for
all the primitive taxa; it has no legitimacy (even the name Caeno-
INFRAFAMILY
pus is invalid). Until recently, Diceratherium and Menoceras were
A number of subfamilies and tribes have been proposed within the placed together in a subfamily Diceratheriinae for the paired-homed

Figure 42.3. Interrelationships of

~
the North American Rhinocerotidae.
~ ::t c: Characters at the nodes are as fol-
.9 ~ ~ ~

c:~
c:
~
l ~
~
~
lows: (I) RHINOCEROTIDAE: II
chisel shaped; i2 tusklike; very re-

·sc:
~
-~
..c ~
~
~
~
duced M3 metacone; Ml-2 cristae
lost; premaxillary-nasal contact lost;

~
~ reduced posterior cingulum on M3;
~ ::t ..9 shorter posterior ectoloph on Ml-2.
~ ~ t'-l ~ (2) Ml-3 parastyle folds more open;
antecrochets enlarged; M3 metacones
lost; i3, lower canine lost in adults;
metacone ribs on molars lost; para-
cone and metacone ribs separate and
reduced on premolars; greater hyp-
sodonty; postcondyloid process on
ramus; broad ascending ramus on
dentary, with straight posterior bor-
der; long, posteromedially curved
process on anterolateral tuberosity
of humerus. (3) Broad parasagit-
tal crests; laterally flared lambdoid
crests; concave dorsal skull pro-
file; long nasals. (4) Extended oc-
ciput; anterodorsally inflected basi-
cranium; long, flattened postglenoid
process. (5) Third upper and lower
incisors, upper canine lost; meta-
cone ribs of P2-4 lost; P2 molar-
ized; mandibular condyle broader, flat
surfaced and nearly horizontal; dis-
tal condyle of humerus more asym-
metrical; dorsoventrally compressed
posterior articular surface on at-
las; postglenoid process faces anteri-
orly; fifth metacarpal reduced to ves-
tige. (6) DICERATHERIINAE: long,
broad supraorbital ridges; paired nasal
ridges in males. (7) Strong crochet
present; 12 lost; reduced sagittal crest;
pre maxillary further reduced; nasal in-
1. RIDNOCEROTIDAE cision over posterior P2; basicranium
shortened relative to palate; upper mo-
lar lingual cingula weak or absent;
shallow anteroventral notch on atlas. (8) Upper premolars fully bilophodont; overall size increase. (9) ACERATHERIINAE: medial flange of i2 re-
duced; long diastema posterior to i2; fifth metacarpal enlarged. (10) Premaxillary reduced; 11 lost; nasal incision over anterior P4. (11) Dorsal skull profile
flattened. (12) TELEOCERATINAE: TELEOCERATINI: metapodials shortened; carpals and tarsals compressed dorsoventrally; strong antecrochets; broad
zygomatic arches; lateral edge of nasals downturned and thinned, resulting in U-shaped cross section; calcaneal tuber elongate; brachycephalic skull; nasal
incision retracted to level of anterior P3; p2 lost in some Teleoceras.
 Donald R. Prothero 598

rhinos (Peterson, 1920). Tanner (I969) showed that the two genera Average length ofm2: 25.0-27.4 mm.
were very distinct, and Prothero, Manning, and Hanson (1986) and Included species: P. hudsoni (known from localities [SB44B],
Prothero and Manning (I 987) have shown that they are only distantly CP39C, CP39G, CP39IIC, CP42A, CP83A, CP98A); P.
related. A large suite of derived characters (Figure 42.3, node 7) sagittatus (locality NPlOB); P. dakotensis (localities
unites Menoceras with higher rhinos, and the paired-horn condition CP39C, CP68D, CP84B).
is a parallelism. The Diceratheriinae includes only Diceratherium
Comments: Small Oligocene rhinocerotids that are slightly
sensu stricto and Subhyracodon, and Menoceras is grouped with
larger than Teletaceras but smaller than any species of Sub-
European Pleuroceros in the subfamily Menoceratinae (Prothero,
hyracodon or Trigonias have been incorrectly referred to a
Manning, and Hanson, I986; Prothero, Guerin, and Manning, I989).
number of genera. The first valid name to be proposed for
this group was Penetrigonias, although the original diag-
nosis was based on invalid differences in upper premolar
INCLUDED NORTH AMERICAN GENERA IN THE
crests. Ignoring the premolars, there are distinct features
FAMILY RHINOCEROTIDAE
in the size and the reduction of the premaxilla that unite
this genus.
The locality numbers listed for each genus refer to the list of unified
In their concept of Penetrigonias, Prothero, Manning,
localities in Appendix I. The locality numbers may be listed in a
and Hanson (1986) included not only the type, but also
couple of alternative ways. The acronyms for museum collections
specimens from the Yoder Formation (locality CP42A) and
are listed in Appendix III.
Flagstaff Rim, Ledge Creek (localities CP39C, CP39G),
Parentheses around the locality (e.g., [CPlOI]) mean the taxon
and Beaver Divide areas of Wyoming (locality CP39IIC),
in question at that locality is cited as an "aff." or "cf." the taxon in
the Porvenir Local Fauna of Texas (locality SB44B), and
question. Parentheses are usually used for individual species, thus
specimens referred to "Subhyracodon" sagittatus by
implying the genus is firmly known from the locality, but the actual
Russell (1982). "Caenopus" dakotensis probably also be-
species identification may be questionable. Question marks in front
longs in this genus.
of the locality (e.g., ?CPlOl) mean the taxon is questionably known
from that locality, thus implying some doubt that the taxon is actually
present at that locality, either at the genus or the species level. Trigonias Lucas, 1900
Type species: Trigonias osborni Lucas, I900.
BASAL RHINOCEROTIDS Type specimen: USNM 3924.
Characteristics: The medium- large-sized Chadronian rhino,
Teletaceras Hanson, 1989 Trigonias, is known from a number of complete skeletons.
Type species: Teletaceras radinskyi Hanson, 1989. It retains all the anterior teeth except for i3 and the lower
Type specimen: UCMP 129000. canine. The skull profile is distinctively concave and saddle
Characteristics: Small rhinocerotid with Il/i2 chisel-tusk shaped, with low sagittal crests and broad lambdoid crests.
combination, but these are not as enlarged as they become It is one of the few nonaceratherine rhinos that retain a
in later rhinocerotids. Teletaceras has an unreduced ante- functional fifth metacarpal.
rior dental series, sharp creases between molar parastyles Average length ofm2: 37.0-50.0 mm.
and paracones, more lingually inflected molar metacone Included species: T. osborni (known from localities CP68B,
axes, and low connection of lower molar metalophids to CP83A, CP83B, CP98A, NPlOB, NP27C, CP42A,
protolophids. There is a marked postcanine diastema and NPIOB); T. wellsi (localities CP83A, CP98A).
a single-rooted p I. Trigonias sp. is also known from localities NB24C,
Average length of m2: 20.5-24.8 mm. NB27D, (CP39F), CP98B, NP24C, NP27D.
Included species: T. radinskyi (known from locality PN5B);
T. mortivallis (locality NB2). Comments: Trigonias was once grossly oversplit by Gregory
and Cook ( 1928) based on slight premolar variations within
Penetrigonias Tanner and Martin, 1976 (synonyms: a single population, but this same evidence is now grounds
Caenopus in part; Subhyracodon, in part) for regarding most premolar differences as intrapopula-
Type species: Penetrigonias hudsoni Tanner and Martin, tional. Besides the type species, only the much larger T.
I976. wellsi is valid.
Type specimen: UNSM 62049.
Characteristics: Penetrigonias is readily distinguished from Amphicaenopus Wood, 1927
Teletaceras not only by its larger size, but also by a Type species: Amphicaenopus platycephalus (Osborn and
whole suite of more advanced rhinocerotid characters Wortman, 1894).
(Figure 42.3, node 2), including a completely developed Type specimen: AMNH 542.
rhinocerotid incisor complex, the absence of an M3 meta- Characteristics: Amphicaenopus is a precociously large
cone, the loss of i3 and the lower canine, and the charac- rhinocerotid from the Chadronian and Whitneyan of the
teristic features of the mandible and humerus. South Dakota Big Badlands and from Stark County, North
 Rhinocerotidae 599

Dakota. It has a very broad, dolichocephalic skull with flar- serve, Goshen Co., Wyoming (locality CP42C), clearly
ing lambdoid crests. The nasals are relatively short, with shows that most of the "species" based on premolar vari-
no lateral notches, and the posterodorsal part of the pre- ants belong to the same population. The species list here re-
maxillary is reduced, allowing the maxilla to contact the flects a more modem concept of taxonomy and will be for-
nasal incision. The lower jaw has a strongly procumbent mally substantiated elsewhere (Prothero, in preparation).
i2 tusk and is nearly cylindrical in cross section. The first
two upper premolars are very primitive, but P4 is nearly
Diceratherium Marsh, 1875 (synonyms: Subhyracodon,
completely molarized.
in part)
Average length of m2: 46.0-50.0 mm.
Type species: Diceratherium armatum Marsh, 1875.
Included species: A. platycephalus only (known from local-
Type specimen: YPM 10003.
ities CP83A, CP84B, NP51A).
Characteristics: Typical Diceratherium armatum is a large
Comments: It is peculiar that this rhino is known from the rhino with paired flanges on the tips of the nasals on males.
Chadronian and Whitneyan, but not from the intervening This was previously confused with the rounded nasal
Orellan. Because it was probably a large, amphibious form, bosses of the much smaller rhino, Menoceras, so most
found in river channel sandstones, perhaps it was driven Menoceras was incorrectly labeled Diceratherium (as dis-
from its habitat in the Orellan by the amphibious amyn- cussed earlier). Besides the nasal ridges and the completely
odontid rhinocerotoid Metamynodon. molarized upper premolars, few derived characters distin-
guish Diceratherium.
Average length ofm2: 31.0-47.0 mm.
Included species: D. armatum (= D. matutinum) (known
DICERATHERIINAE
from localities CC9D, CC13, CP48, CP50, CP51A, CP52,
Subhyracodon Brandt, 1878 (synonyms: Caenopus, in part) CP84C, CP85C, CPI 0 I, NPl OC, NP34C, NP36A, NP36B,
Type species: Subhyracodon occidentalis (Leidy, I 850). PN6C, PN6D, PN6E); D. tridactylum (= Subhyracodon
Type specimen: USNM I I4. tridactylum) (localities CP42D, CP84B, CP99B, CP99C,
Characteristics: Subhyracodon is easily distinguished from NP51A); D. annectens (localities CC9D, CP48, CP50,
other Chadronian and Orellan rhinos by its larger size (ex- ?CP51A, CP52, CP84C, CPIOl, NP34C, PN6C, ?PN6D,
cept for Amphicaenopus, which is much larger). It is also ?PN6E, NClA); D. niobrarense (localities CCI3, CP103,
distinguished from more primitive rhinos by the loss of CP104A, CP104B, NPlOD, PN19B); D. n. sp. (locality
I3 and the upper canine, and a molarized P2 (Figure 42.3, CPlOl).
node 5). It has a mandibular condyle that is broader, flat Diceratherium sp. is also known from localities ?CAI,
surfaced, and nearly horizontal. The postglenoid process CA7, NB3D, SB2, SB4, CP86B, (NP33B), (NP34D),
faces anteriorly. The feet are completely tridactyl. (PN6F), PN6G.
Average length of m2: 29.0-39.0 mm.
Included species: S. occidentalis (known from localities Comments: Subhyracodon gives rise to Diceratherium in the
CP40B, CP4IA, CP41B, CP68C, CP84A, CP99A, Whitneyan (31 Ma) and this genus then persists into the
NP50C); S. mitis (localities CP39C, CP39G, CP39IIC, Hemingfordian and possibly the early Barstovian (about
CP40A,CP4IA,CP42A,CP68B,CP83A,CP83B,CP83C, I 7 Ma), a record for generic longevity among rhinos. As
CP98, NP29C); S. kewi (locality CC9C). with Subhyracodon, the taxonomy of Diceratherium has
Subhyracodon sp. is also known from localities CC12, long been badly misunderstood.
NP24C, NP24D, NP32B. Nevertheless, identification is usually unambiguous be-
cause it was the practically the only rhino present through
Comments: Subhyracodon is the commonest larger Chadro- the Whitneyan and most of the Arikareean. This low in
nian and Orellan rhinocerotid, but is a taxon that has been generic diversity was compensated for by several paral-
greatly misunderstood. For over a century, the invalid name lel species lineages that persisted unchanged through most
"Caenopus" (Cope, I880) has been attached to this rhino, of the Arikareean: D. armatum, the large type species,
and it still appears in recent works (e.g., Russell, I982; and D. annectens, a smaller species. There is also a new
Wilson and Schiebout, I984) because there has been no dwarf species (Prothero, in preparation) from the earli-
comprehensive systematic work on Oligocene rhinos since est Arikareean (formerly considered the Gering Formation
Scott (1941). Lucas, Schoch, and Manning (198I, p. 835) at Roundhouse Rock, Nebraska, but now correlated with
formally placed "Caenopus" in synonymy with Subhyra- the "brown siltstone" member of the Brule Formation, ac-
codon, and my work substantiates this (Prothero, in prepa- cording to Swinehart et al., 1985) and a late Arikareean
ration). to early Hemingfordian species, D. niobrarense. In this
Like Trigonias and Hyracodon, Subhyracodon was over- study, the Whitneyan species "Subhyracodon" tridacty-
split into several invalid species and genera based on slight lum is also referred to Diceratherium because some male
variations of molarization of the tipper premolars. The skulls clearly show the beginnings of the paired nasal
large Orellan quarry sample from the Harvard Fossil Re- ridges.
 Donald R. Prothero 600

MENOCERATINAE gests that it was one of several immigrant aceratherines in

the early Hemingfordian which eventually lost out to the
Menoceras (Troxell, 1921) (synonyms: Diceratherium, dominant genera, Aphelops and Peraceras.
in part; Moschoedestes)
Type species: Menoceras arikarense (Barbour, 1906). New genus (Prothero, in preparation)
Type specimen: UNSM 62008. Characteristics: Galusha (1975) briefly described a speci-
Characteristics: Menoceras was long confused with men (F:AM 95544) from the early late Hemingfordian
Diceratherium because both have paired horns at the tips Box Butte Formation, Box Butte Co., Nebraska (locality
of their nasals (Peterson, 1920). As already discussed, the CP107). It differs from all other aceratherines in its smaller
horns are not homologous in detail, and apparently they size, extreme nasal retraction (to the level of anterior Ml,
were independently derived. Menoceras shows many other a feature paralleled only in the latest, most derived Aph-
characters (Figure 42.3, node 7) that clearly show it is a elops), and its uniquely flat dorsal skull profile. It cannot be
much more derived rhino than Dice ratherium sensu stricto. referred to any existing genus, so Prothero (in preparation)
Indeed, i~ female skulls without the paired horns, Meno- will erect a new genus to contain this material. Although
ceras was often mistaken for more derived rhinos. it is one of the earliest aceratherines known, it is also one
Average length of m2: 31.0-46.0 mm. of the most derived.
Included species: M. arikarense (known from localities The m2 is unknown. Average length ofm2: 41.0 mm.
GC3A, GC3B, GC8B, ?GC8C, CP51A, CP84C, CP103,
CP104A, CP104B); M. barbouri (= M. "cooki," M. "mars- Aphelops Cope, 1874
landensis," and "Moschoedestes delahoensis") (localities Type species: Aphelops megalodus (Cope, 1873).
GC5, GC8D, SB29B, SB46, CP71, CP88, CP105, CP106, Type specimen: AMNH 8292.
NCIA, NC2). Characteristics: Aphelops has the derived features of a dor-
sally arched nasal-frontal profile and an unusually long di-
Comments: Pig-sized Menoceras arikarense (usually astema between the lower tusk and the first lower premolar.
mislabeled "Diceratherium cooki") immigrated to North Like almost all aceratherines, it lacks a chisel-shaped II.
America in the late Arikareean and is by far the common- Average length ofm2: 41.0-63.0 mm.
est mammal in the famous Agate bone bed. It evolved to a Included species: A. megalodus (known from localities GC3B,
slightly larger species, M. barbouri, in the early Heming- GC4B, GC4C, GC4D, GC4E, NB6B, [NB20A], NB25B,
fordian, which has been badly oversplit (M. ''falkenbachi," NB29, SB32A, SB32B, SB32D, SB32F, CP75B, [CP75C],
M. "marslandensis," and the Castolonrhino, "Moschoedes- CP76, CP78, CP90A, CP108B, CPI 10, CP114A, CP114B,
tes"). Some of these synonymies were demonstrated by CP114C, CP114D, CP116A, CP116B, ?NC3B); A. mala-
Prothero and Manning (1987), and others are forthcom- corhinus (localities GCl lA, GCl lB, SB32G, SPIB,
ing (Prothero, in preparation). After spreading over much SPIC, SP3A, CP76, CP78, CP116A, CP116B, CP116C,
of the continent from New Mexico to Florida in the early CP123C); A. mutilus (localities GCl lC, GC12II, GC13A,
Hemingfordian, Menoceras disappeared as the wave ofim- SB34C, SB58A, SB58B, SPID, SP3B, SP4A, CP115C,
migrant aceratherines and teleoceratines apparently drove CP116D, CP116F, CP123D).
it out in the mid-Hemingfordian. Aphelops sp. is also known from localities GC6D,
(GC8F), (GC9A), GC9B, GC9C, GCIOB, GC12I, ?CC23,
NB7D, (NB9), (NBll), NB17, NB19C, NB23C, NB27B,
ACERATHERIINAE (SB43A), SB55, CP127, (NPll), NP42, PN8B, (PN9B),
Floridaceras Wood, 1964 PN13.
Type species: Floridaceras whitei Wood, 1964.
Comments: Aphelops is the most common and best known
Type specimen: MCZ 4046.
aceratherine in North America. Three successive species
Characteristics: Floridaceras is the first aceratherine whose
are recognized here, each becoming progressively larger
functional fifth metacarpal was noted (Wood, 1964). Its
from the late Hemingfordian to the late Hemphillian. The
aceratherine affinities are somewhat equivocal. The only
primitive species, A. megalodus, persists with little change
skull material is badly crushed, so the degree of nasal
from the late Hemingfordian to the late Clarendonian. In
retraction is difficult to determine. Unlike any other ac-
the Hemphillian, the genus gets noticeably larger, with
eratherine, it had an upper II, as indicated by a thegosis
more hypsodont teeth and more extreme nasal retraction.
facet on the lower tusk.
Aphelops, a browser, is typically found with the grazing
Average length ofm2: 47.2 mm.
Teleoceras in most North American Miocene localities,
Included species: F. whitei only (known from localities GC5,
although in smaller numbers. It reached its acme in the
GC8D, CP106, PN6H).
latest Hemphillian, when it occurred in great numbers in
Comments: Floridaceras is an unusually large Hemingfor- the Panhandle of Texas (Coffee Ranch, locality SP3B, and
dian rhino known primarily from the Thomas Farm Local comparable localities) and Kansas (Edson Quarry) (local-
Fauna in Florida (locality GC5). Most of its anatomy sug- ity CP123D). Teleoceras, by contrast, was rare at this time.
 Rhinocerotidae 601

Peraceras Cope, 1880 (synonyms: Diceratherium, in part); that is level with the postglenoid process. There are strong
Aphelops, in part; Teleoceras, in part) lingual cingula on the upper premolars, and M2 is approx-
Type species: Peraceras superciliosum Cope, 1880. imately equal in length to M3. The distal limb elements,
Type specimen: AMNH 8380. especially the second metapodials, are longer and less ro-
Characteristics: Peraceras is characterized by a brachycep- bust than is typical of Teleoceras. In all other features, it
halic skull with a procumbent lambdoid crest and occiput, shows the typical teleoceratine anatomy of a robust, short-
shortened nasals, flat dorsal skull profile, an upturned sym- limbed skeleton and a brachycephalic skull with relatively
physis in females, a short lower diastema, and lingual cin- hypsodont teeth and flaring lambdoid crests.
gula on the lower cheek teeth. Average length of m2: 44.0-54.0 mm.
Average length of m2: 31.0-61.0 mm. Included North American species: B. americanum (known
Included species: P. superciliosum (= P. "crassus," P. "trox- from localities CP71, CP107, CP108A, CP108B).
elli") (known from localities NB23B, SB32D, SB32F,
SB34A, [SPIA], CP71, CP73C, CP76, CP89, CP90A, Comments: Yatkola and Tanner (1979) assigned late Hem-
CPI 14A, CP114B, CP114C, CPI 14D, ?CPI 16, NP41B); ingfordian teleoceratines from the Martin Canyon Local
P. profectum (= "Diceratheriumjamberi," "Aphelops mon- Fauna of northeast Colorado (locality CP71) to Brachy-
tanus") (localities ?NB7C, NB18, SB32A, SB32B, potherium americanum (gender corrected). It is unques-
SB32D, SB32F, CP71, CP108B, CPI 10, CPI 14B, NP42); tionably the most primitive teleoceratine from North
P. hessei (localities GC4B, GC4C, GC4D, GC4E, GClOA, America, although whether it is truly Brachypotherium is
SB32D). still debatable. For the present, all of the late Hemingfor-
Peraceras sp. is also known from localities SB34A, dian teleoceratine material is referred to Brachypotherium
PN9A. americanum; early Barstovian material is referred to Teleo-
ceras medicomutum.
Comments: Peraceras has long been one of the most mis-
Teleoceras Hatcher, 1894
understood rhinos of North America. The type species
Type species: Teleoceras major Hatcher, 1894.
is the most robust, derived species of the genus, and the
Type specimen: PU 10645.
primitive members have been erroneously referred to
Characteristics: Teleoceras is easily recognized by its char-
Aphelops, Aceratherium, and Diceratherium. Prothero and
acteristically robust and proximodistally shortened limbs
Manning (1987) cleared up some of the confusion, and
and feet, and by its brachycephalic skull with broad flaring
further discussion will be presented by Prothero (in prepa-
lambdoid crests and broad zygomatic arches. The nasal
ration).
bones are U-shaped in cross section, and typically there
Th,ree species are currently recognized. The primitive,
is a small horn rugosity at the very tip of the nasals. The
medium-sized species, P. profectum (= "Diceratherium
upper II is still present, unlike the condition in the ac-
jamberi," "Aphelops montanus"), overlaps in size and mor-
eratherines. The cheek teeth are usually very hypsodont,
phology with A. megalodus. Most specimens of the prim-
with the molars greatly enlarged at the expense of the pre-
itive Hemingfordian and Barstovian P. profectum are hard
molars. The p2 is lost in most species of Teleoceras. The
to distinguish from A. megalodus from the same deposits.
nasal incision is retracted to the level of anterior P3, much
There is a dwarf species, P. hessei, from the Texas Gulf
shallower than in derived aceratherines. In addition to the
Coastal Plain (Prothero and Sereno, 1980; Prothero and
robustness of the limbs, Teleoceras has an elongate cal-
Manning, 1987), and the large type species, P. supercil-
caneal tuber and an articulation between the scaphoid and
iosum (= P. "crassus," P. "troxelli"), which paralleled the
lunar.
teleoceratines in many features. During the Barstovian and
Average length of m2: 50.0-63.0 mm.
early Clarendonian, P. superciliosum is most common in
Included species: T. major (known from localities GC6B,
northern localities, such as South Dakota, northern Ne- SB32G, SP2A, CP90A, CP114D, CPI 16A, CP116B); T.
braska, and Montana and is not found in southern localities medicomutum (localities GC4C, GC4D, GC4E, NB7C,
except in New Mexico and California. If it competed with [CP75C], [CP76], CP87B, CPI 10, CP114A, CPI 14D,
Teleoceras for the same large-bodied aquatic grazer niche,
NP41B, PN7); T. meridianum (localities GC4B, GC4C,
then it was less successful because it was much more rare
GC4D, GC4E, NB7C, CP114A); T. fossiger (localities
than Teleoceras and died out in the Clarendonian.
[CA9], NB27A, NB31, NP45, SB31B, [SPIB], [SPIC],
SP3A, CP116B, CP116C, CP116D, CP116E, CPl23C,
CP127, [PN12]); T. proterum (localities GCllA, GCI IB);
TELEOCERATINAE: TELEOCERATINI
T. hicksi (= T. "ocotensis") (localities GC13B, GC13B,
Brachypotherium Roger, 1904 GC13C, CC37, CC40, CC41, NB33, SBl 1, SB34C,
Type species: Brachypotherium brachypus (Lartet, 1837). SB58A, SB62, SPID, CP78, CPI 16D, CPI 16E); T. n. sp.
Type specimen: Unavailable. A. (localities SB32B, SB32D, SB32F); T. n. sp. B (SPID,
Characteristics: Brachypotherium differs from Teleoceras in SP3B, CP123D).
having a slender post-tympanic process with a ventral edge Teleoceras sp. is also known from localities (CA3), CA4,
Donald R. Prothero 602

(GCIOB), (GCl2II), GCl3A, GC27, CC25B, CC26B, elops seems to be related to the climatic and floral changes
CC36, CC37, CC38, NBl9C, NB23C, NB27C, SB32G, triggered by the Messinian worldwide climatic event.
SB34A, SB60, SPlA, (SPlF), (CP56), CP76, CP87B,
CPI 14C, CPI !SD, CPI 16F, NP38E, PN9A, PNI 1, PN13, INDETERMINATE RHINOCEROTIDS
PN14, PNl5.
Fragmentary remains ascribed to rhinocerotids have been reported
Comments: Teleoceras is by far the most abundant and best from localities CA7, GCSE, CC3, CC19II, CC21C, SB47, SB48,
known Miocene rhinoceros from North America. It is SB52, CP54B, CP87A, CP116E, NP9A, NPIOBB, NP22, NP25A,
known from enormous quarry samples all over the con- NP25B, NP25C, PN6B, PN13, PN16.
tinent, including complete articulated individuals in death
poses killed during an ash fall, from Ashfall Fossil Bed,
Antelope Co., Nebraska (locality CP116A; Voorhies and BIOLOGY AND EVOLUTIONARY PATTERNS
Thomas~on, 1979; Voorhies, 1981).
The taxonomy of Teleoceras was in great confusion, North American rhinoceroses showed a considerable variety of eco-
but Prothero (in preparation) has simplified this consider- logical adaptations. Most were relatively large-bodied herbivores
ably. By my count, there are eight valid species in North and among the largest animals of their ecosystem, but the pig-sized
America, including two that are new. Overall, the Brachy- Menoceras was much smaller than contemporary entelodonts and
potherium-Teleoceras lineage shows a general size increase chalicotheres. Eocene and Oligocene rhinocerotids have relatively
from Brachypotherium americanum in the late Hemingfor- low-crowned teeth and efficient cutting incisor chisel-tusk combi-
dian to T. medicornutum of the Barstovian, T. major of the nations for browsing, and most of them are found in river chan-
C!arendonian, and T. fossiger of the early Hemphillian. nel sandstones. Clark, Beerbower, and Kietzke (1967) found that
After reaching maximum size in the early Hemphillian, Subhyracodon was more abundant in the near-stream facies of the
the remaining species are smaller: T. proterum, the Florida Big Badlands than in the swampy plain; it was absent from the
endemic, and T. hicksi, the late Hemphillian species. Two open plain. Trigonias and Amphicaenopus are known exclusively
dwarf species of Teleoceras are known: T. meridianum of from river channel deposits. As the titanotheres, amynodonts, and
the Texas Gulf Coastal Plain Barstovian, and Teleoceras n. other large-bodied browsers disappeared during the late Eocene
sp. B from the latest Hemphillian of Guymon, Oklahoma. and the Oligocene, only the rhinocerotids persisted in the brows-
There is an new endemic species of Teleoceras from the ing, near-stream niche (Figure 35.3). By the Arikareean, the con-
Barstovian and Clarendonian of the Santa Fe Group of ditions were drier, and overall mammalian diversity was at an all-
New Mexico, which differs from contemporary T. medi- time low (Prothero, 1985; Stucky, 1990). During most of this time,
cornutum in having unusually short nasals and robust pre- Diceratherium was the only large-bodied mammal in North
maxillae, apparently for some sort of prehensile lip, as is America, apparently hiding in what remained of the riparian wood-
seen in aceratherines. land. Judging from the large numbers of these rhinos in quarries such
Teleoceras was unquestionably a successful analogue of as 77 Hill, Niobrara Co., Wyoming (locality CP52), they may have
the hippopotamus, which lives in large herds in the riverbed formed herds. Diceratherium was the first North American rhino to
during the day and comes out at night to graze along the show significant sexual dimorphism in horns or tusk development,
riverbanks. This is suggested not only by their occurrence so it probably had a more complicated social structure than its more
in river channels and their hippo like build, but also by their solitary predecessors.
population structure (D. Wright, pers. comm.). Teleoceras In the late Arikareean (Upper Harrison Formation, Agate Springs
occurs in large numbers in just about every Miocene local- Quarry, locality CPI 04A), the first of the early Miocene immigration
ity in the High Plains after the Hemingfordian, usually ac- events brought the tiny European Menoceras on the scene. Judging
companied by the browser Aphelops. Contrary to popular from its great abundance in these quarries, it probably lived in large
belief, Teleoceras did not disappear before Aphelops in the herds, with subequal numbers of males and females. By contrast,
latest Hemphillian. It just happens to be scarce in the richest trueDiceratherium was very rare and apparently losing ground to the
late Hemphillian quarries, such as Coffee Ranch and Ed- invader. It is unknown from the rest of the Hemingfordian (late early
son. Actually, Teleoceras reached its maximum geographic Miocene), although it may occur as late as the Barstovian in Railroad
extent in the latest Hemphillian, even if it is scarce. It occurs Canyon, Idaho (locality PNl 9B). Menoceras, in turn, had to compete
in the Mt. Eden Local Fauna in southern California, Panaca with a mid-Hemingfordian wave of immigrant aceratherines and
Local Fauna in Nevada, Keams Canyon and Jeddito Local teleoceratines and was gone before the late Hemingfordian. By the
Fauna in Arizona, the Chamita Formation in New Mexico, late Hemingfordian, the aceratherines Aphelops and Peraceras and
the Wray localities in Colorado, the Snake Creek Forma- the teleoceratine Brachypotherium had established the dominant
tion in Nebraska, the Sawrock Local Fauna in Kansas, rhino lineages for the rest of the Miocene.
Coffee Ranch Local Fauna and the Guymon area in the From the late Hemingfordian to the late Hemphillian (latest
Texas-Oklahoma Panhandle, the Upper Bone Valley For- Miocene), North American rhinos show the browser-grazer pair
mation in Florida, and in Yepomera and Ocote in Mexico. combinations that are typical of savannas everywhere, including
As discussed earlier, the extinction of Teleoceras and Aph- the modern East African savanna. The browser typically feeds on
 Rhinocerotidae 603

w
z L L.BI. 1.8
w E.BI
2.4
(.)
0 E L•t Hp.
4.5

:J L.Hp.
5.2
6.0
11. L.E.Hp.
7.0
E.E.Hp.
L 8.8
L.CI.

E.CI.
9.5
e
.~
w 11.0 i::
Cll
z L.L.Ba.
.s0
w 12.5 -111
(.)
0 M
E.L.Ba.
14.0
-e-Ill
-111- ~ -~
e
::§ E.Ba. Cll
Cl
:I
cQ) a Ill
e '5 0
Cll

~
5.8 -111-
:tJ -~[
0
Qi ~ -;!!
L.Hm.

17.5 -IQ i::

-Q) '§. e
~
-0[
-~ LL: z
E.Hm.
18.B
'o:(
E L.L.Ar.
_o
19.2
c: - - -
E.L.Ar.
23.0
~

w L.E.Ar.
z
w L
(.)
27.7
0
CJ E.E.Ar.

:J 29.4
0
Wt.
E 31.9
Or.
33.4
l.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Du

39.5
Teleoceratini
L.un.
w 41.3
DICERATHERllNAE ACERATHERllNAE
z E.Un.
w M
(.) 45.9
0
w L.Br.
46.7

M.Br.
49.0
E.Br.
50.4

L.Wa.

53.5
E M.Wa.
54.2
E.Wa.
E.E.Wa.
55.0
56.5
-
Ck.3 55.6
w
z
Ck.2
Ck.1
Tl.&
56.8
56.0
-
56.2
w L Tl.5 57.0
(.) Tl.4 57.3
Tl.3
0 Tl.2
w Tl.1
..J
<( To.3
11. To.2
To.1
E Pu.3
Pu.2 64.5
Pu.1 65.0
Pu.a 65.1

Bl.=Blancan, Hp. = Hempt)illian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar. = Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch. = Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. = Wasatchian,
Ck. = Clarkforkian, Ti. = Tiffanian, To.= Torrejonian, Pu.= Puercan.
Figure 42.4. Temporal ranges of North American rhjnocerotid genera.
 Donald R. Prothero 604

medium- and high-level leaves and tender shoots and uses a pre- fauna, and Teleoceras proterum was also a Florida endemic. The
hensile lip and/or trunk for the purpose. Aceratherines, particularly Rio Grande rift was remarkably rich in Peraceras (all three species)
Aphelops and primitive Peraceras, performed this role in the North and had its own endemic, prehensile-lipped species of Teleoceras
American Miocene savanna. These rhinos have a greatly retracted (Teleoceras "n. sp. Pl' earlier). The details of the faunas from the
nasal incision for anchoring muscles of a prehensile lip and have re- Miocene of California, Oregon, Nevada, and Arizona have not been
placed their upper incisors with a nipping pad, as modem rhinos and fully worked out yet, but if the fauna has any rhinos at all, typically
ruminants have. The modem browsing black rhino is rather solitary they include Aphelops megalodus and less frequently Teleoceras
with a wide home range, and with few exceptions, mostAphelops are or Peraceras. Further details of this biogeographic picture are pre-
found in small numbers in any given quarry. Wright (pers. comm.) sented elsewhere.
has found that Aphelops has relatively low infant mortality at the
Love site in Florida (locality GCI IA), which is comparable to the
population structure of the browser Diceros. REFERENCES
By contrast, there is little doubt that teleoceratines were good
analogues of the modem hippo. Not only was their body very hippo- Barbour, E. H. (1906). Notice of a new fossil rhinoceros from Sioux County,
like, but they are found in great numbers in river channel deposits, Nebraska: Diceratherium arikarense. Publication of the Nebraska
indicating that they were gregarious herders like hippos. Wright Geological Survey, series 1, 2, 313-18.
Brandt, J. F. (1878). Testamen synopseos rhinocerotidum viventium et fossil-
(pers. comm.) found that the Love Bone Bed Teleoceras had a very
ium. Memoires de l'Academie Imperiale des Sciences, St. Petersburg,
high infant mortality, like that of hippos. Their teeth are very high 26, 1-66.
crowned and suitable for grazing, and the throat cavities of the Poi- Clark, J., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedi-
son Ivy Quarry (locality CPI 16A) rhinos even contain grass seeds mentation, stratigraphy, paleoecology, and paleoclimatology in the
(Voorhies and Thomasson, 1979). Although modem white rhinos BigBadlandsofSouthDakota. Fieldiana Geology Memoir, 5, 1-158.
Cope, E. D. (1873). On some new extinct Mammalia from the Tertiary of
are not as aquatic as Teleoceras or hippos, they are efficient grazers
the Plains. Palaeontological Bulletin, 14, 1-2.
that live in small herds. The browser-grazer pair combination was (1874). Report on the stratigraphy and Pliocene vertebrate paleontology
very successful, for similar pairs were found in most Miocene sa- of northern Colorado. United States Geological and Geographical
vanna localities in Eurasia and Africa, usually involving a browsing Survey of the Territories, Bulletin, 14, 9-28.
Aceratherium or Dicerorhinus and a grazing Brachypotherium or (1880). A new genus of rhinoceros. American Naturalist, 14, 540.
Galusha, T. (1975). Stratigraphy of the Box Butte Formation. Bulletin of the
Chilotherium. When the savannas disappeared from North America
American Museum of Natural History, 156, 1-68.
in the early Pliocene, so did the rhinos. Today the browser-grazer Gregory, W. K., & Cook, H.J. (1928). New material for the study of evo-
pairs are found only in the East African savanna, one of the few lution, a series of primitive rhinoceros skulls (Trigonias) from the
remaining on earth. lower Oligocene of Colorado. Proceedings of the Colorado Museum
Because of the wide geographic spread of North American of Natural History, 8, 3-39.
Hanson, C. B. (1989). Teletaceras radinskyi, a new, primitive rhinocerotid
Miocene localities, it is possible to examine rhino biogeography
from the late Eocene Clarno Formation, Oregon. In The Evolution of
as well. The most striking fact is that rhinos are far more abundant the Perissodactyls, eds. D.R. Prothero & R. M. Schoch, pp. 379-98.
in the Plains states of South Dakota, Nebraska, Kansas, Oklahoma, Oxford: Clarendon.
and Texas than they are in the intermontane localities in Califor- Hatcher, J.B. (1894). A median horned rhinoceros from the Loup Fork beds
nia, Oregon, and Nevada (see Figure 35.3). There are many good of Nebraska. American Geologist, 13, 149-50.
Lartet, E. (1837). Sur Jes debris, fossiles trouves a Sansan, et sur Jes ani-
Miocene localities from the intermontane region, with a great abun-
maux antediluviens en general. Comptes-rendus Hebdomadaires des
dance of camels, horses, and mastodonts, but rhinos are extremely Seances de l'Academie des Sciences, Paris, V, 158-9.
rare. Because this cannot be due to sampling, it is apparent that rhi- Leidy, J. (1850). Remarks on Rhinoceros occidentalis. Proceedings of the
nos preferred the open savanna of the Plains to the more wooded Academy of Natural Sciences, Philadelphia, 5, 119.
habitats of the intermontane states. There are few Miocene local- Lucas, F. A. (1900). A new rhinoceros, Trigonias osborni, from the Miocene
of South Dakota. United States National Museum Proceedings, 23,
ities east of the Mississippi, but Florida produces rhinos in great
221-4.
numbers. Lucas, S. G., Schoch, R. M., & Manning, E. (1981). The systematics of
Besides this overall trend, certain species clearly had biogeo- Forstercooperia, a middle to late Eocene hyracodontid (Perisso-
graphic preferences. The dwarf species of Peraceras and Teleoceras dactyla: Rhinocerotoidea) from Asia and western North America.
were found mainly in the Texas Gulf Coastal Plain, which had ape- Journal of Paleontology, 55, 826-41.
Madden, C. T., & Dalquest, W. W. (1990). The last rhinoceros in North
culiar endemic fauna that lived in the swampy, wooded habitat near
America. Journal of Vertebrate Paleontology, 10, 266--7.
the coast (Prothero and Sereno, 1980). These dwarfs were analo- Marsh, 0. C. (1875). Notice of new Tertiary mammals. IV. American Journal
gous to the dwarfing that occurs when savanna mammals adjust of Science, 109, 239-50.
to the more limited spaces and resources of the forest, as happens Osborn, H.F. (1898). The extinct rhinoceroses. Memoirs of the American
with pygmy hippos, forest elephants and Cape buffalo, and many Museum of Natural History, 1, 75-164.
Osborn, H. F., & Wortman, J. L. (1894). Fossil mammals of the lower
island dwarfs. Peraceras superciliosum, in contrast, is found almost
Miocene White River beds. Collection of 1892. Bulletin of the Amer-
exclusively in more northerly Barstovian and Clarendonian local- ican Museum of Natural History, 6, 199-223.
ities, especially in Montana, South Dakota, and Nebraska; it was Peterson, 0. A. (1920). The American diceratheres. Memoirs ofthe Carnegie
absent from Florida or Texas. Florida had endemism in much of its Museum of Natural History, 7, 399-476.
 Rhinocerotidae 605

Prothero, D.R. (1985). North American mammalian diversity and Eocene- Toronto: Royal Ontario Museum, Life Sciences Miscellaneous Pub-
Oligocene extinctions. Paleobiology, 11, 389-403. lication.
Prothero, D. R., Guerin, C., & Manning, E. (1989). The history of the Tedford, R.H. ( 1970). Principles and practice of mammalian geochronology
Rhinocerotoidea. In The Evolution of the Perissodactyls, eds. D. R. in North America. Proceedings of the North American Paleontolog-
Prothero & R. M. Schoch, pp. 320-40. Oxford: Clarendon. ical Convention, F, 666--703.
Prothero, D.R., & Manning, E. (1987). Miocene rhinoceroses from the Texas Tedford, R. H., Galusha, T., Skinner, M. F., Taylor, B. E., Fields, R. W.,
Gulf Coastal Plain. Journal of Paleontology, 61, 388-423. Macdonald, J. R., Rensberger, J. M., Webb, S. D., & Whistler,
Prothero, D.R., Manning, E., & Hanson, C. B. (1986). The phylogeny of the D. P. (1987). Fauna! succession and biochronology of the Arika-
Rhinocerotoidea (Mammalia, Perissodactyla). Zoological Journal of reean through Hemphillian interval (late Oligocene through earliest
the Linnean Society, 87, 341-66. Pliocene Epochs) in North America. In Cenozoic Mammals ofNorth
Prothero, D. R., & Sereno, P. C. (1980). Allometry and paleoecology of America, ed. M. 0. Woodburne, pp. 152-210. Berkeley: University
medial Miocene dwarf rhinoceroses from the Texas Gulf Coastal of California Press.
Plain. Paleobiology, 8, 16--30. Troxell, E. L. (1921). A study of Diceratherium and the diceratheres. Amer-
Roger, 0. (1904). Wirbeltierreste aus dem Obermiocan der bayerisch- ican Journal of Science, 202, 197-208.
schwabishcen Hochebene. Tei! V. Berichte des Naturwissens- Voorhies, M. R. (1981). Ancient ashfall creates Pompeii of prehistoric ani-
chaftlichen VereinsfiirSchwaben und Neuburg, Augsburg, 36, 1-21. mals. National Geographic, 159, 66--75.
Russell, L. S. (1982). Tertiary mammals of Saskatchewan, Part VI: The Voorhies, M. R., & Thomasson, J. R. (1979). Fossil grass anthoecia within
Oligocene rhinoceroses. Life Sciences Contribution, Royal Ontario Miocene rhinoceros skeletons: diet in an extinct species. Science,
Museum, 133, 1-58. 206, 331-3.
Scott, W. B. (1941). The Mammalian Fauna of the White River Oligocene, Webb, S. D. (1984). Ten million years of mammalian extinctions in North
Part V: Perissodactyla. Transactions of the American Philosophical America. In Quaternary Extinctions, a Prehistoric Revolution, eds.
Society, 28, 747-980. P. S. Martin & R. G. Klein, pp. 189-210. Tucson: University of
Stucky, R. K. (1990). Evolution of land mammal diversity in North America Arizona Press.
during the Cenozoic. Current Mammalogy, 2, 375-432. Wilson, J. A., & Schiebout, J. A. (1984). Early Tertiary vertebrate fau-
Swinehart, J.B., Souders, V. L., Degraw, H. M., & Diffendal, R. F., Jr. (1985). nas, Trans-Pecos Texas: Ceratomorpha less Amynodontidae. Pearce-
Cenozoic paleogeography of western Nebraska. In Cenozoic Paleo- Sellards Series, Texas Memorial Museum, 39, 1-47.
geography of the West Central United States, eds. R. M. Flores & Wood, H. E., II. (1927). Some early Tertiary rhinoceroses and hyracodonts.
S. Kaplan, pp. 209-29. Special Publication of the Rocky Mountain Bulletin of American Palaeontology, 13, 5-105.
Section, Society of Economic Paleontologists and Mineralogists. (1964). Rhinoceroses from the Thomas Farm Miocene of Florida. Bul-
Tanner, L. G. (1969). A new rhinoceros from the Nebraska Miocene. Bulletin letin of the Museum of Comparative Zoology, Harvard University,
of the University of Nebraska State Museum, 8, 395-412. 130, 361-86.
Tanner, L. G., & Martin, L. D. (1976). New rhinocerotoids from the Yatkola, D., & Tanner, L. G. (1979). Brachypotherium from the Tertiary
Oligocene of Nebraska. In Athlon, Essays in Paleontology in of North America. Occasional Papers of the Museum of Natural
Honor of Loris Shano Russell, ed. C. S. Churcher, pp. 210--19. History, University of Kansas, 77, 1-11.
43 Proboscidea
W. DAVID LAMBERT and JEHESKEL SHOSHANI

INTRODUCTION

The order Proboscidea currently consists of two extant species,

the Asian elephant (Elephas maximus) and the African elephant
(Loxodonta africana), both of which belong to the family Elephan-
tidae. However, modem proboscideans represent only a vestigial
remnant of a formerly much greater radiation, including Paleogene
taxa of Africa (e.g., Palaeomastodon and Phiomia ), the conservative
mammutids (family Mammutidae ), the gomphotheres (family Gom-
photheriidae sensu lato ), the stegodonts (family Stegodontidae), and
elephantid cousins of the modem elephants such as Stegotetrabe-
lodon and the mammoths. Other groups may also belong to the
Proboscidea, although there is disagreement on this subject. Taxa
placed in the Proboscidea by some authors include the moeritheres
(family Moeritheriidae), deinotheres (family Deinotheriidae), bary-
theres (family Barytheriidae), Numidotherium, and Anthracobune
(e.g., Tassy and Shoshani, 1988; Shoshani, 1996). Conflicting views
are expressed by other authors, however (e.g., Coppens et al., 1978).
The reader is referred to Tassy and Shoshani (1988) for a more de-
tailed discussion of this subject.
Proboscideans were the largest land animals in North America
during the late Cenozoic. They originated in the late Paleocene
along the shores of the Tethys Sea in northern Africa and south- Figure 43.1. Restoration of Amebelodon (by Brian Regal).
ern Asia (where they became extinct). Some spread into Eurasia
from Africa during the early Miocene and crossed Beringia to en- Gomphotheriidae), of which the gomphotheres were by far the more
ter North America in the late early Miocene. Many were present diverse and abundant. Conservatively about 20 generic names for
throughout the rest of the Tertiary and into the Quaternary and did North American Tertiary gomphotheres exist in the literature (many
not vanish from the continent until approximately 10,000 years ago of which are almost certainly invalid), versus only 4 for marnmu-
(Coppens et al., 1978; Kurten and Anderson, 1980; Shoshani, 1996). tids (of which only 2 are considered valid here). In addition, gom-
During the Tertiary they were morphologically diverse, geograph- photheres are known from far more Tertiary sites in North America
ically widespread (ranging from the Atlantic to the Pacific coasts, than the mammutids, particularly those of Miocene age.
as far north as Saskatchewan and as far south as Honduras), and This chapter is a review of the Tertiary proboscideans of North
abundant, being represented in an extraordinary number of Tertiary America. The phylogeny and classification schemes used here are
faunas (Gazin and Collins, 1950; Downs, 1952; Webb and Perrigo, presented in Figure 43.5 and Table 43.1, and are based on a cladistic
1984; Madden and Storer, 1985). The Tertiary proboscideans of analysis by Shoshani (1996). The species-level taxonomy of many
North America consisted of two major groups, the Mammutidae proboscidean groups is sorely in need of revision. Such a revision
and the gomphotheres sensu lato (mostly consisting of the family is a task beyond the scope of this chapter, and thus discussion of

606
 Proboscidea 607

Figure 43.3. Skeleton of Mammut americanum (modified from Osborn

[1936] by Brian Regal).

Figure 43.2. A. m2-m3 of Zygolophodon proavus (scale bar = 5 cm).

B. m2-m3 of Gomphotherium sp. (scale bar = 5 cm). C. Skull of Gom-
photherium sp. (scale bar= 18 cm). All modified from Osborn (1936) (by Figure 43.4. Skeleton of Gomphotherium sp. (modified from Osborn [1936]
Brian Regal). by Brian Regal).

species is minimal. The defining features provided here are charac-

ters by which given taxa can be recognized, and may or may not be the mammutid genus Zygolophodon and conservative gomphothere
strictly apomorphic. genera such as Gomphotherium. Both groups have well-developed
lower and upper tusks, enamel bands on their upper tusks, and cheek
teeth whose pretrite cones (i.e., those cones that are on the leading
DEFINING FEATURES OF THE edge of the chewing stroke, lingual on the upper dentition, labial on
ORDER PROBOSCIDEA the lower) are associated with conules (small accessory cones) that
together form the classic "trefoil" pattern (reminiscent of a three-
Proboscidean taxa such as Palaeomastodon, Phiomia, and the more leaf clover) with wear (see Figure 43.2A). (Note that the tendency
advanced forms are classically diagnosed on the basis of the fol- for the cheek teeth to form trefoils can vary considerably within
lowing characters: (1) columnar limbs, with the proximal elements Zygolophodon, with some Zygolophodon cheek teeth lacking them
longer than the distal ones; (2) limb bone shafts lacking medullary altogether [Coppens et al., 1978].) Nevertheless, all mammutids
cavities, with cancellous bone filling the interior; (3) carpals and are characterized by short, broad skulls and cheek teeth with sim-
tarsals short and broad; (4) skull pneumatized; (5) nasal bones re- ple cross crests lacking a central conule (i.e., zygodonty) (Tobien,
tracted (to accommodate a proboscis or "trunk"); (6) premaxilla 1975; Coppens et al., 1978; Shoshani, 1996) (see Figures 43.2A
expanded; (7) 12/i2 enlarged to form tusks, with 11/il and 13/i3 ab- and 43.3).
sent; (8) canines absent; and (9) cheek teeth (i.e., premolars and The family Gomphotheriidae is diagnosed by the following char-
molars) bunolophodont and composed of plates that form distinct acters: intermediate cheek teeth (i.e., dP4/dp4--M2/m2) trilopho-
lophs on wear. These characters are present in the conservative gen- dont, and M3/m3 tetralophodont (seeFigure43.2B). Note, however,
era Phiomia and Palaeomastodon, although they are better devel- that secondary tetralophodonty in the intermediate cheek teeth is
oped in more advanced taxa (Coppens et al., 1978; this diagnosis known in some gomphotheriids (e.g., the Amebelodon subgenus A.
can be roughly applied to the suborder Elephantiformes within the [Konobelodon] and Platybelodon), and the number oflophs on the
Proboscidea as proposed by Tassy and Shoshani [1988]). M3/m3 can be significantly greater than four (e.g., Amebelodon, Ste-
The family Mammutidae as a whole is difficult to distinguish from gomastodon) or occasionally less than four (i.e., Gomphotherium)
the gomphotheres sensu lato because of a strong similarity between (see later).
 W. David Lambert and Jeheskel Shoshani 608

SYSTEMATICS Table 43.1. A suggested classification of the

Proboscidea of North America (note that no subfamilial
nodes are named or ranked due to potential instability
SUPRAORDINAL at this taxonomic level).
The proboscideans sensu lato are united with extant sirenians (sea
cows) and extinct desmostylians in the Tethytheria by McKenna Order Proboscidea Illiger, 1811
(1975). Tethytherian monophyly is defined by the following char- Suborder Elephantiformes Tassy, 1988
lnfraorder Elephantoidea Gray, 1821
acteristics: bifed apex of the heart; a pair of pectoral mammae; for-
Superfamily Mammutoidea Hay, 1922
wardly positioned orbit, placed above the anterior premolars, with
Family Mammutidae Hay, 1922
the infraorbital canal placed under the orbit; amastoidy, with reduc- Zygolophodon Vacek, 1877
tion of the mastoid process; and incipiently bilophodont/bunolopho- Mammut Blumenbach, 1799
dont dentition (see Tassy and Shoshani, 1988). Traditionally, the Unnamed taxon (node C in Figure 43.5)
Hyracoidea has been considered the sister taxon to the tethytheres Family Gomphotheriidae Hay, 1922
among living mammals, with hyracoids plus tethytheres and Gomphotherium Burmeister, 1837
the extinct arsinoitheres (order Embrithopoda) comprising the Megabelodon Barbour, 1914b
Paenungulata or "subungulates" (e.g., Simpson, 1945). However, Amebelodon Barbour, 1927
more recently, other workers have sought to place hyracoids as Serbelodon Frick, 1933
the sister group to the Perissodactyla (see Prothero, Manning, and Torynobelodon Barbour, 1929a
Platybelodon Borissiak, 1928
Fischer, 1988), although this suggestion is by no means universally
Gnathabelodon Barbour & Sternberg, 1935
accepted (e.g., Tassy and Shoshani, 1988).
Eubelodon Barbour, 1914a
Rhynchotherium Falconer, 1868
INFRAORDINAL Stegomastodon Pohlig, 1912
Cuverionius Osborn, 1923
All of the proboscideans known from North America can be referred Unnamed taxon (node E in Figure 43.5)
to the infraorder Elephantoidea, which includes the families Mam- Tetralophodon Falconer, 1857
mutidae, Gomphotheriidae, Stegodontidae, and Elephantidae (node Mammuthus Brookes, 1828
A on Figure 43.5; see Tassy and Shoshani, 1988, and Shoshani,
1996, for a discussion of the Old World Stegodontidae). Of these
families, only the mammutids and gomphotheres (with one possi-
ble exception being Tetralophodon; see later) were present in North siderably throughout the group, a tetralophodont M3. Based on this
America during the Tertiary, the elephantid mammoths not reaching character variability one could justify placing the nodes Dl, 04,
North America until the early Pleistocene (Kurten and Anderson, Gnathabelodon, and Eubelodon into a polytomy joined at node C,
1980; Webb, 1992). A diagnosis for the elephantoid proboscideans although in the cladistic analysis this action always increased the
is presented later. length of the tree by a significant number of steps. Gomphotherium
A phylogeny of the North American proboscideans is presented and the "shovel-tusked" gomphotheres are united at node Dl by a
in Figure 43.5, with a brief discussion later. This phylogeny is taken single suspect character, the presence of a linear band of enamel on
from a cladistic analysis of the Proboscidea performed by the sec- the upper tusks. A similar enamel band is found on the conserva-
ond author (Shoshani). Refer to Shoshani (1996) for an in-depth tive mammutid genus Zygolophodon as well as the more conserva-
description and discussion of the methodology used. Proboscidean tive Phiomia, and thus this character may represent an elephantoid
phylogeny is difficult to analyze due to a lack of information for plesiomorphy. The "shovel-tusked" gomphotheres, united at node
some taxa and a strong tendency toward homoplasy within the group 02, are characterized by distinctly flattened lower tusks and can
(e.g., tetralophodonty in both Platybelodon and Tetralophodon, the be distinguished from rare Gomphotherium individuals with flat-
absence of lower tusks in Mammut, Cuvieronius, and the elephan- tened lower tusks by their straight (as opposed to rounded) medial
tids, and double trefoiling in a large number of groups). Thus this tusk surfaces. Platybelodon and Torynobelodon are united within
cladogram should be considered tentative, at least at the subfamilial the shovel tuskers at node 03 by the shared possession of denti-
level. nal rods with a unique morphology in their lower tusks, although
The monophyly of the elephantoids (node A) is supported by an they differ considerably in their tusk shape (Lambert, 1990, 1992).
impressive number of characters, many of which (particular! y the ba- Rhynchotherium, Stegomastodon, and Cuvieronius, united at node
sicranial characters) are unlikely homoplasies. The family Mammu- 04, constitute a strictly New World radiation of proboscideans. They
tidae (node B) is diagnosed primarily by cheek tooth characters. The differ in a number of respects, but all share a helicoidal (spiral) curva-
unranked taxon at node C encompasses the entire gomphothere ra- ture in their upper tusks. Stegomastodon and Cuvieronius are some-
diation, including the gomphotheres sensu Jato, the elephantids, and what questionably united at node 05, with the possibility existing
the Old World stegodonts; the lower tusk synapomorphies are highly that Rhynchotherium is the sister group of Cuvieronius rather than
variable within the group. The family Gomphotheriidae (node D) is Stegomastodon (e.g., May and Repenning [1982]). Stegomastodon
a weak assemblage. It is based on a single feature that varies con- and Cuvieronius both lack P2/p2-P4/p4 and are brevirostrine (i.e.,
 Proboscidea 609

ELEPHANTOIDEA

Mammutidae Gomphotheriidae

"Shovel-tuskers" "New World forms" Elephantidae

Figure 43.5. Interrelationships of the North American proboscidean genera (slightly modified after Shoshani, 1996). Note that Megabelodon is not
included in this cladogram because of its questionable taxonomic status. Key: # = Old World taxon. * = Taxon of questionable status. Key to
nodes on cladogram: A. ELEPHANTOIDEA: dp2 parastyle reduced; dp3 posterior !aphid oblique; P2 absent; M3 postentoconule enlarged; M3
with three or more lophs; fourth lophid on m3 with central conule in third interlophid; ml-m3 anterolingual cingulum on post-trite side re-
duced; horizontal displacement and vertical succession of cheek teeth associated with remodeling of adjacent bones; sagittal crest absent; premax-
illa with the posterodorsal process in the midline, at the floor of the nasal fossa; canalis temporalis (= squamosal sinus canal) absent; foramen
ovale and foraihen lacerum confluent; bulla (ectotympanic area) slightly enlarged posteriorly. B. MAMMUTIDAE: M3 reduced; cheek teeth with
a central conule that does not block the valley; cheek teeth zygodont. C. UNNAMED TAXON - HIGHER ELEPHANTOIDS: Lower tusks elon-
gate, protruding more than 10 cm; lower tusks pyriform (note that these characters can vary considerably in some groups); cranium swollen; nar-
ial openings with deep depressions laterally; narial openings extend laterally up to or beyond width of rostrum; stylohyoid with posterior ramus.
D. GOMPHOTHERIIDAE: M3 tetralophodont. DI. UNNAMED TAXON including Gomphotherium, Amebelodon, Platybelodon, Torynobelodon, and
Serbelodon: Linear band of enamel on upper tusks. D2. UNNAMED TAXON including Amebelodon, Platybelodon, Torynobelodon, and Serbelodon:
Lower tusks dorsoventrally flattened; lower tusk medial surface straight rather than rounded. D3. UNNAMED TAXON including Platybelodon and
Torynobelodon: Lower tusks filled with short, morphologically homogeneous dentinal tubules. D4. UNNAMED TAXON including Rhynchotherium,
Stegomastodon, and Cuvieronius: Upper tusks with helicoidal (spiral) curvature. D5. UNNAMED TAXON including Stegomastodon and Cuvieronius:
P2-P4 absent; M3 with five or more lophs; brevirostrine mandibular symphysis; lower tusks absent. E. UNNAMED TAXON including Tetralophodon and
Mammuthus: Intermediate cheek teeth tetralophodont.

have strongly foreshortened mandibular symphyses lacking tusks; INCLUDED NORTH AMERICAN GENERA
see discussion later); Rhynchotherium and Cuvieronius share a spi- IN THE ORDER PROBOSCIDEA
raled enamel band on their upper tusks. Stegomastodon lacks an
enamel band on its upper tusks and no likely near ancestor with The locality numbers listed for each genus refer to the list of unified
an enamel band is known, so its ancestral condition cannot be di- localities in Appendix I. The acronyms for museum collections are
rectly determined. If, however, the brevirostrine condition and lack listed in Appendix III.
of P2/p2-P4/p4 are true synapomorphies of Stegomastodon and Cu- The locality numbers may be listed in a couple of alternative
vieronius at node 05 and not convergences, then one can deduce that ways. Parentheses around the locality (e.g., [CP101]) mean the taxon
the ancestor of Stegomastodon most likely had a spiraled enamel in question at that locality is cited as an "aff." or "cf." the taxon
band as well. An unnamed group consisting of Tetralophodon and in question. Parentheses are usually used for individual species,
Mammuthus (Elephantidae) is united at node E strictly by cheek thus implying the genus is firmly known from the locality, but the
tooth characters, principally their tetralophodont intermediate cheek actual species identification may be questionable. Question marks
teeth. The intermediate cheek tooth tetralophodonties in Platybe- in front of the locality (e.g., ?CP101) mean the taxon is questionably
lodon andAmebelodon (Konobelodon) are presumably homoplasic. known from that locality, thus implying some doubt that the taxon
 W. David Lambert and Jeheskel Shoshani 610

is actually present at that locality, either at the genus or the species Included species: M. americanum (known from localities
level. GC14, PN3C, PN23A, PN23C); M. matthewi (locality
CP115D).
Mammut sp. (=M. adamsi, M. cosoensis, M.furlongi,
MAMMUTIDAE HAY, 1922 M. nevadanus, M. praecursor, M. vexillianus, and M. sel-
Zygolophodon Vacek, 1877 (junior synonym: Miomastodon lardsi) is also reported from localities GC13B, GC13C,
Osborn, 1922) CC171, CC39B, NBlO, NB12B, NB27B, NB31, SBS,
Type species: Zygolophodon turicensis. ?SBlO, SB35, SPlD, CP117A, CP121, CP128B, CP128C,
Type specimen: Unavailable. PNlO, PNll.
Characteristics: Lower tusks well developed; upper tusks
Comments: Unlike Zygolophodon, Mammut is well repre-
with a well-developed enamel band; M2/m2 trilophodont;
sented in North America by a large amount of material,
M3/m3 tetralophodont, with the fourth loph(id) weakly
including several skeletons (see Figure 43.3). Also unlike
developed; and cheek teeth characterized by short ridges
Zygolophodon it had a long chronological range, extend-
medially divided by V-shaped sulci, cones commonly ac-
ing from the Clarendonian (early late Miocene) to the ter-
companied by weakly developed conules (Coppens et al.,
minal Pleistocene. It was uncommon during the Claren-
1978; Madden and Storer, 1985) (see Figure 43.2A).
donian and Hemphillian (late Miocene-earliest Pliocene)
Average length ofm3: 171.0-179.0 mm.
where it is commonly referred to as Pliomastodon in the lit-
Included species: Z. proavus (= Z. merriami, Z. tapiroides
erature, but became much more abundant during the Blan-
nomen nudum) only (known from localities NB 18, CP56,
can (late early late Pliocene) to become a major part of the
NP! 1).
North American fauna. Mammut was widespread across
Zygolophodon sp. is also reported from localities NB6E,
the continent, even during the Clarendonian and Hemphil-
NBl7, NB20A, CP75C, (CPl14A), CPI 14B, NP34D.
lian (Osborn, 1926; Simpson, 1930; Osborn, 1936;
Comments: Zygolophodon is rare in North America and is Shotwell, 1963; Schultz, 1977; Kurten and Anderson,
known from scant material, mostly dentition. It has a 1980). The lower tusks of Mammut in nearly all cases
widespread distribution, being known from the Old World are either extremely reduced or else completely absent;
and North American sites in the West, Great Plains, and however, one isolated mandible from the Blancan Santa Fe
Gulf Coastal Plain (Frick, 1933; Osborn, 1936; Madden River site in Florida (locality GC 14) possessed lower tusks
and Storer, 1985). Zygolophodon had a relatively narrow of near-Zygolophodon proportions as indicated by the size
chronological range. It is best known from the late Barsto- of its alveoli, although it was Mammut-like in all other re-
vian (late middle Miocene), although it did reach the spects. Whether this specimen represents an aberrant indi-
Clarendonian (early late Miocene) (Savage and Russell, vidual or a member of an aberrant population remains to be
1983). Tedford et al. (1987) stated that it entered determined.
North America contemporaneously with Gomphotherium
during the Barstovian (see discussion of Gomphotherium Indeterminate Mammutidae
later). However, there is evidence that Zygolophodon en- Reported from localities GC6, GC I OA, NB 11, NB36, SB3 l C, SB50,
tered North America during the late Hemingfordian (early SPlF, PNSB, PN12, PN14.
Miocene, 18-16 million years ago), significantly earlier
than Gomphotherium (Webb, 1992).
GOMPHOTHERIIDAE HAY, 1922

Mammut Blumenbach, 1799 (junior synonym: Gomphotherium Burmeister, 1837 (junior synonyms:
Pliomastodon Osborn, 1926) Trilophodon Falconer, 1847; Serridentinus Osborn, 1923;
Type species: Mammut americanum. Trobelodon Frick, 1933; Ocalientinus Frick, 1933;
Type specimen: Unavailable. Tatabelodon Frick, 1933)
Characteristics: Upper tusks with vestigeal enamel bands Type species: Gomphotherium angustidens.
or lack of them; lower tusks vestigial or absent (but see Type specimen: Unavailable.
later); cones transversely elongated and lacking accessory Characteristics: M3 with three to four and a fraction lophs,
conules, forming extremely simple lophs on wear (note m3 with three to five and a fraction lophids; trefoils present
that some Tertiary individuals possess rudimentary acces- only on the pretrite half of the cheek teeth, that is, sin-
sory conules that form simple crests); medial sulcus be- gle trefoiling (see diagnosis for the mammutids earlier);
tween cheek teeth cones largely (Tertiary individuals) or elongated lower tusks circular, oval, pyriform, or rarely
completely absent (Pleistocene individuals) (Kurten and dorsoventrally flattened in cross-sectional profile; and an
Anderson, 1980). elongated mandibular symphysis, commonly called the
Average length of m3: 164.0-194.0 mm. longirostrine condition (Tobien, 1973; Coppens et al.,
Proboscidea 611

1978) (see Figures 43.2B, C, and 43.4). that swept the Northern Hemisphere (Webb, 1983; see
Average length ofm3: 172.0-245.0 mm. later). It is known with certainty from only one latest
Included species: G. productum (known from localities Hemphillian (earliest Pliocene) locality, the Whidden
NB28, SB32D, SB34A, SP2A); G. nov. sp.? (localities Creek site in the Bone Valley Formation of Florida (locality
GC13B, ?GC14, GC17). GC13B), where it is represented by a possible new dwarf
Gomphotherium sp. (including G. anguirivalis, G. species (material in the collection of the Florida Museum of
calvertensis, G. cimarronis, G.fricki, G. giganteum, G. gre- Natural History in Gainesville). Gomphotherium persisted
gori, G. nebrascensis, G. obscurum, G. osbomi, G. perfec- into the Blancan only in Florida, where the Bone Valley
tum, G. phippsi, G. republicanum, G. simpsoni, G. willis- species is represented in the Macasphalt Local Fauna (lo-
toni) is also reported for localities CA2, CA3, GC4E, cality GCl 7) and possibly the Santa Fe River (locality
GC6A, GC6B, GClOA, GCl lA (Amebelodon of Webb, GC14).
MacFadden, and Baskin [1981]), GCl lB, GC13B, CC17F,
CC17G, CC17H, CC26B, CC31, CC36, CC37, CC40, Eubelodon Barbour, 1914a (synonym: Gomphotherium in
NB6E, NB7D, NBS, NB23C, NB29, SB34C, SPlB, Tobien, 1973)
CP75C, CP76, CP114A, CP114B, CP114C, CPl15B, Type species: Eubelodon morrilli.
CP115C, CPl15D, CP116A, CP124, NPll, PN7, NC3B, Type specimen: UNSM 1416 (mounted skeleton).
NC4. Characteristics: Lower tusks absent, with the mandibular
symphysis elongated and highly tapered; in most other
Comments: Gomphotherium is the most morphologically respects resembles Gomphotherium.
conservative of all of the gomphotherioid proboscideans, Average length ofm3: 202.0 mm.
and probably is close to the basal stock from which this Included species: Eubelodon morrilli only (known from lo-
group radiated. Which taxa should actually be included in calities CP87B, CP114C, CPl 140, CP116A).
this genus is controversial. Several Gomphotherium-like
Comments: Eubelodon is a rare genus known from a largely
genera exist in the literature, such as Serridentinus, Tata-
intact skeleton, including a mandible and skull (Barbour,
belodon, Trobelodon, and Ocalientinus. These taxa were
1914a). It is one of a small group of Gomphotherium-like
united with Gomphotherium by Tobien (1972, 1973), who
taxa including Gnathabelodon and Megabelodon that lack
demonstrated that their key diagnostic features lay within
lower tusks and have unusually shaped mandibular sym-
continual character distributions encompassing this genus.
physes. It is controversial whether these taxa represent dis-
Tobien also included in Gomphotherium the aberrant gen-
tinct lineages of their own or whether they are merely aber-
era Eubelodon, Gnathabelodon, and Megabelodon (see
rant, perhaps pathological Gomphotherium individuals (a
laterfor discussion of these genera).
view advocated by Tobien, 1973). Nevertheless, the ap-
Gomphotherium first entered North America from Asia
parent adaptiveness of the mandibular symphyses of these
during the Barstovian (middle Miocene). There is, how-
animals (see later) suggests that they indeed represent valid
ever, a question as to exactly when during the Barsto-
taxa rather than mere Gomphotherium aberrations.
vian it arrived. Tedford et al. (1987) used Gomphotherium
to help diagnose the beginning of the early late Barsto- Megabelodon Barbour, 1914b (synonym: Gomphotherium in
vian based on its stratigraphic record in the Great Plains. Tobien, 1973)
For example, Gomphotherium is unknown from the early Type species: Megabelodon lulli.
Barstovian Lower Snake Creek deposits of Nebraska (the Type specimen: UNSM 1643 (cranium and mandible).
Olcott Formation, locality CPl 10) but abundant in the Characteristics: Lower tusks absent, as in Eubelodon; mandi-
overlying early late Barstovian Valentine Formation (lo- bular symphysis elongated but not tapered, with the distal
cality CPl14) (Voorhies, 1990). Downs (1952), however, end slightly expanded; distal end of the mandibular sym-
described Gomphotherium material from the Mascall Lo- physis with a symmetrically cleft coarse scar on its dorsal
cal Fauna of Oregon (locality PN7), dated by Tedford et surface; in most other respects resembles Gomphotherium.
al. (1987) as early Barstovian, and Ferrusquia-Villafranca Average length ofm3: 222.0 mm.
( 1984) reported Gomphotherium from the early Barstovian Included species: M. lulli (known from locality CPl 140); M.
El Orama! Local Fauna of Central America (locality CA2). minor (locality GC6B); M. joraki (locality SB32D).
After entering North America, Gomphotherium quickly
became widespread and abundant throughout the conti- Comments: Megabelodon is much better known than Eube-
nent, having been found at an extraordinary number of lodon; several specimens including two complete skeletons
late Barstovian to early Hemphillian (late Miocene) sites. having been found (Frick, 1933; Barbour, 1934; Mawby,
It became scarce in the late Hemp.hillian (latest Miocene), 1968; Tobien, 1973). Note that following Tobien (1973),
evidently having been decimated along with the rest of the Shoshani (1996) considers Megabelodon to be a junior
Clarendonian chronofauna by major ecological changes synonym of Gomphotherium.
 W. David Lambert and Jeheskel Shoshani 612

Gnathabelodon Barbour and Sternberg, 1935 (synonym: suggested that this variation is the result of ontogeny and
Gomphotherium in Tobien, 1973) accidental breakage during life.
Type species: Gnathabelodon thorpei. Specimens in both the Northern Great Plains and the Old
Type specimen: FHSM VP-18. World have been referred to Rhynchotherium, although all
Characteristics: Lower tusks absent; mandibular symphysis currently accepted Rhynchotherium specimens are from
elongated, but strongly concave dorsally, widening distally sites in southern North America (Frick, 1933; Osborn,
to form a broad, spoonlike tip; in other respects resembles 1936; Olsen, 1957; Tobien, 1973; Kurten and Anderson,
Gomphotherium. 1980; Miller, 1980; Webb and Perrigo, 1984; Miller, 1990).
Average length of m3: 166.0 mm. Rhynchotherium is classically considered to range through
Included species: G. thorpei (known from locality CP123); the late Hemphillian and Blancan (latest Miocene-Plio-
G. buckneri (locality GC6A). cene ), and indeed Tedford et al. ( 1987) used the appearance
of Rhynchotherium to help define the late Hemphillian.
Comment.s: Gnathabelodon has possibly strangest appear- Webb and Perrigo (1984), however, described Rhyncho-
ance of the gomphotheres, lacking lower tusks, its bizarre therium material from the Gracias Formation in Honduras,
symphysis inspiring the moniker "spoon-billed mastodont" which is either late Clarendonian or early Hemphillian
(Barbour and Sternberg, 1935; Osborn, 1936; Sellards, (early late Miocene) in age. Thus it is likely that Rhyn-
1940; Tobien, 1973; Tedford et al., 1987). chotherium first evolved in Central America during the
Rhynchotherium Falconer, 1868 (junior synonyms: late Clarendonian or early Hemphillian, and then dispersed
Blickotherium Frick, 1933; Aybelodon Frick, 1933) northward into the main body of North America at the be-
Type species: Rhynchotherium tlascalae. ginning of the late Hemphillian. Rhynchotherium appar-
Type specimen: Unavailable. ently did not survive into the Pleistocene.
Characteristics: Broadly resembles Gomphotherium in cheek
tooth morphology; enamel bands of upper tusks spiral to Cuvieronius Osborn, 1923 (junior synonym: Cordillerion
varying degrees (a character shared with Cuvieronius; see Osborn, 1926)
later); lower tusks mediolaterally compressed and com- Type species: Cuvieronius hyodon.
monly but not always having enamel bands; and mandibu- Type specimen: Unavailable.
lar symphysis longirostrine but significantly deflected Characteristics: Lower tusks absent; mandibular symphysis
downward (Miller, 1980; Webb and Perrigo, 1984; Miller, strongly shortened (i.e., "brevirostrine"); upper tusk with
1990). spiraled enamel bands as per Rhynchotherium; cheek teeth
Average length of m3: 168.0-180.0 mm. varying enormously in morphology, ranging from strong
Included species: R. tlascalae (known from locality SB56); double trefoiling (i.e., accessory conules/conulids that wear
R. blicki (= R. simpsoni and R. edense) (localities CA9, to form trefoils on both halves of the cheek teeth) to sin-
CAlO, CC40, ?NBlO, GC13B). gle trefoiling (primitive Gomphotherium condition), and
Rhynchotherium sp. (including R. praecursor and R. eu- ptychodont (i.e., plicated enamel folds) to aptychodont
hypodon) has also been reported from localities CC53, (i.e., lacking plicated enamel folds), in their simplest form
GC28, ?SB13, SB18A, SB36, SB58A, SB58B, SB64, being indistinguishable from Gomphotherium; and M3/m3
SPlB, SP3B, SP4B, SP5, CP125, CP130A, CP131. with four and a fraction to five and a fraction lophs
(Simpson and Paula-Couto, 1957; Miller, 1990; Kurten
Comments: Rhynchotherium has been subject to consider- and Anderson, 1980).
able confusion, in part because of its distinct resemblance Average length of m3: 188.0-225.0 mm.
to Gomphotherium. There has been controversy regard- Included species: Intraspecific variation in cheek tooth mor-
ing the morphology of the enamel on its upper and lower phology makes it extremely difficult to identify valid Cu-
tusks. The spiraled enamel band on the upper tusks led vieronius species. Pending future clarification of this situa-
May and Repenning (1982) to refer gomphothere material tion, it is safest to refer Cuvieronius material to Cuvieronius
from the Mount Eden site in California (locality CC40, late sp. (known from localities GC17, SB13, SB14A, SB14B,
Hemphillian) to Cuvieronius rather than Rhynchotherium. SB14C, SB14E, SB18A, SB18B, SB37, SB63).
However, more definate Rhynchotherium upper tusk ma-
terial with spiraled enamel bands has been found in the Comments: Cuvieronius, likeRhynchotherium, was a Neotro-
Bone Valley Formation of Florida (locality GC13B, latest pical genus. It is rare in the Tertiary, being largely known
Hemphillian). The controversy over the presence of enamel from the Pleistocene. Near the end of the Blancan (late
bands on the lower tusks (a feature found in no other gom- Pliocene) or the beginning of the Irvingtonian (latest
phothere) stems from their inconsistent presence (Tobien, Pliocene or early Pleistocene) it entered South America,
1973). Webb and Perrigo (1984) found that this feature where it may have given rise to such endemic gomphothere
varied significantly in a Rhynchotherium population from genera as Notiomastodon and Haplomastodon (Tobien,
the Gracias Formation of Honduras (locality CA9), and 1973; Kurten and Anderson, 1980).
 Proboscidea 613

Stegomastodon Pohlig, 1912 (synonyms: Rhabdobunus (Gidley, 1926; Osborn, 1936; Savage, 1955; Woodburne,
Hay, 1914; Anancus Gidley, 1926) 1961; Tobien, 1973; Kurten and Anderson, 1980; Madden,
Type species: Stegomastodon mirificus. 1986). It is commonly considered to have entered South
Type specimen: Unavailable. America (e.g., Simpson and Paula-Couto, 1957; Kurten
Characteristics: Lower tusks absent; mandibular symphysis and Anderson, 1980; Simpson, 1980). This is disputed
brevirostrine; upper tusks lacking enamel bands; cranium by Webb (1992; pers. comm.), however, who believes
high domed, reminiscent of elephantids; cheek teeth char- that the South American gomphothere referred to Stego-
acterized by secondary trefoiling, ptychodonty, choero- mastodon is in fact Haplomastodon (a genus endemic to
donty (i.e., the presence of numerous tubercles within the South America).
lophs ), and an abundance of cement; M2/m2 eithertrilopho-
dont or tetralophodont (with fourth loph poorly developed Amebelodon Barbour, 1927
when present); and M3 with from five and a fraction to Type species: Amebelodonfricki.
seven and a fraction lophs (Osborn, 1936; Savage, 1955; Type specimen: UNSM 1226 (mandible).
Tobien, 1973). Characteristics: Mandibular symphysis more elongated than
Average length ofm3: 210.0-230.0 mm. is typical of Gomphotherium; lower tusks elongated and
Included species: S. primitivus (= S. rexroadensis) (known strongly flattened (although less so than Torynobelodon
from localities CP118, CP128C); S. mirificus (localities and Platybelodon ), with a distinct dorsal and in most cases
CC43, SB 12, SB 15, SPIH, SP5, CP94, CP95, CP96, CP97, also a ventral concavity (Lambert, 1990). This diagno-
CP117B, CP130A, CP131, PN23C). sis is based strictly on mandibular characters. Although
Stegomastodon sp. has also been reported from localities mandibular material is reasonably common, inexplicably
WM26, NB13C, SB14D, SB14F, SB36, SB58A, SB58B, Amebelodon cranial material beyond upper dentition is
SB60, SB63, CP121. largely unknown (but see later).
Average length ofm3: 173.0-286.0 mm.
Comments: Stegomastodon can be divided into two distinct
Included species: A.floridanus (=A. paladentatus, A. hicksi)
forms, although whether these forms represent subgenera
(known from localities GCl lB, CP78, CP123C); A. nov.
or two species within a single subgenus is debatable. The
sp. (locality GClO); A. fricki (localities SPlC, SP3A,
more conservative of these forms (S. primitivus) is char-
CP78, CPl 160, ?NP45); A. (Konobelodon) britti (local-
acterized by purely trilophodont M2/m2s, typically five
ities GClOD, GCl lC, CP123D, SPID).
and a fraction lophs on its M3/m3, and cheek teeth with a
Amebelodon sp. has also been reported from localities
moderate amount of cement (although considerably more
SPIB, CP116C, CP125, PNl 1.
than is typical of gomphotheres) (Osborn, 1936; Wood-
burne, 1961; Tobien, 1973). The more advanced form (S. Comments: Amebelodon, by far the most common of the
mirificus) is characterized by M2/m2s commonly with a North American "shovel-tusked" gomphotheres, has been
poorly developed fourth loph (a condition termed "quasi- subject to considerable confusion in the literature primar-
tetralophodonty" by Tobien [1973]), M3/m3s with six and ily because of controversy over the proper diagnosis of
a fraction to seven and a fraction lophs, and cheek teeth two key features: the number of lophs on the interme-
with a large amount of cement, greater than that found diate cheek teeth and the internal structure of the lower
in the conservative Stegomastodon forms (Savage, 1955; tusks (e.g., Gregory, 1945; Bennett, 1977). ApparentAme-
Tobien, 1973). belodon mandibles and lower tusks from North America
In the cladogram in Figure 43.5, Stegomastodon is linked can be divided into two morphotypes: (1) trilophodont in-
with Rhynchotherium and Cuvieronius as part of an au- termediate cheek teeth and lower tusks with a distinctly
tochthonous New World radiation. This view is also ex- laminated internal structure, both gomphothere plesiomor-
pressed by Savage (1955) and Tobien (1973). Webb (1992) phies; and (2) Ml/ml or both Ml/ml and M2/m2 tetralo-
proposed that Stegomastodon is an Old World immigrant phodont (all North American specimens are tetralophodont
with Anancus affinities (a view shared with Gidley [ 1926] in their Ml/ml and M2/m2) and lower tusks filled with
and Mebrate [1987]), implying that the brevirostrine dentinal rods, both characters apparently shared with Platy-
mandibles shared by Stegomastodon and Cuvieronius are belodon and at least one (dentinal rods) shared with
convergences. Torynobelodon (its cheek teeth being completely unknown;
Stegomastodon first appeared in the late Hemphillian see later) (Lambert, 1990). Some authors (e.g., Gregory,
(latest Miocene) and extended well into the Irvingtonian 1945) downplayed the phylogenetic significance of the
(early Pleistocene), although it did not survive to the end conflicting characters; others (e.g., Bennett, 1977) con-
of the epoch. The conservative fo.rm is known from the sidered them a sufficient basis to refer the anomalous Ame-
late Hemphillian and early Blancan; the advanced form belodon specimens to the Platybelodon/Torynobelodon
extended across the entire Blancan into the Irvingtonian. group, removing them from Amebelodon. Lambert ( 1990),
Stegomastodon is characteristically a western genus however, demonstrated that the similarities between the
 W. David Lambert and Jeheskel Shoshani 614

anomalous Amebelodon individuals and Platybelodon/ ical of Gomphotherium; cheek teeth similar to those of
Torynobelodon are convergent, and placed the anomalous Gomphotherium.
specimens into a subgenus of Amebelodon, Amebelodon Averagelengthofm3: 170.0-173.0mm.
(Konobelodon ). Included species: S. barbourensis (= S. barbouri) (known
Amebelodon shows intrageneric variation in other im- from locality CP116B); S. burnhami (localities NB7B,
portant features as well, though they have not been the NB7C, NB7D, NB7E). Note added in press. "S. burnhami"
subject of controversy. Amebelodon lower tusks can dif- is now considered to be Amebelodon sp. (Guan, 1996).
fer significant! y in shape, ranging in their degree of flat-
Comments: Serbelodon, another "shovel-tusked gompho-
tening from marginal (conservative species, e.g., A. flori-
there," strongly resembles Amebelodon in the cross-sec-
danus) to extreme (advanced species, e.g.,A.fricki). Lower
tional profile of its lower tusks and was united withAmebe-
tusks with marginal flattening can be distinguished from
lodon by Tobien (1972) on this basis. Nevertheless, it does
rare flattened Gomphotherium specimens by virtue of their
squared-off medial tusk surfaces, the same surface being not strikingly resemble Amebelodon beyond this one fea-
ture, although both genera share a squared-off medial edge
rounded for Gomphotherium. Amebelodon also varies con-
on their lower tusks. The existence of rare Gomphotherium
siderably in its cheek tooth morphology, ranging from a
very simple single trefoiling indistinguishable from Gom- lower tusks that resemble Amebelodon in profile (presum-
photherium (conservative species) to elaborate double tre-
ably a convergence) suggests that the similarity between
Amebelodon and Serbelodon may also be the result of con-
foiling (advanced species).
vergence (note thatAmebelodon and Serbelodon are united
Amebelodon is widespread geographically, known from
Africa and Asia as well as North America (Tobien, 1986; at a polytomy in Figure 43.5 due mainly to their shared
Gaziry, 1987), although it is rare outside of North America. flattened tusks). Two Serbelodon lower tusk morphotypes
The distribution of Amebelodon within North America is exist, each associated with a distinct species. Serbelodon
barbourensis has a lower tusk profile that is nearly indis-
mainly restricted to the Great Plains and the Gulf Coastal
Plain (Florida) (Cook, 1922; Barbour, 1927; Barbour and tinguishable from that of Amebelodon. Serbelodon burn-
hami, in contrast, has lower tusks that are proportionately
Hibbard, 1941; Gregory, 1945; Bennett, 1977; Lambert,
broader and flatter than those of Amebelodon, with a dor-
1990, 1992).
There has been confusion about the biochronology of sal concavity present but extremely shallow and little or
no ventral concavity (Osborn, 1933).
Amebelodon in the literature. For example, Tedford et al.
Serbelodon is mainly restricted to North America, al-
(1987) use the appearance of Amebelodon to help de-
though Guan ( 1996) reports its presence in China. It is very
fine the beginning of the late Clarendonian (early late
Miocene), whereas Lambert (1990) restricts Amebelodon rare, known from only a few localities in North America
to the Hemphillian (late-latest Miocene) excluding the lat- (Frick, 1933; Osborn, 1933; Whistler and Burbank, 1992).
est Hemphillian (earliest Pliocene). This confusion stems
from the union of Serbelodon with Amebelodon by Tobien Torynobelodon Barbour, 1929a
(1972) (see later), which extended the range of Amebe- Type species: Torynobelodon loomisi.
lodon back to the Clarendonian. (The use of Amebelodon Type specimen: UNSM 2601 (distal tusk tip).
as an indicator for the late Clarendonian by Tedford et al. Characteristics: Lower tusks narrow and highly flattened,
[1987] apparently is based on the erroneous referral by with strong dorsal curvature; as in Platybelodon, inter-
Webb, MacFadden, and Baskin [1981] of the Love site nal space of lower tusks filled with short, densely packed
Gomphotherium to Serbelodon, which following Tobien dentinal rods.
[1972] they called Amebelodon.) However, nothing that Average length of m3: unknown.
can be called traditional Amebelodon (which we recognize Included species: T. loomisi only (known from localities
here) has ever been reliably found from a site older than ?GC12II, CPI 16II, CP126).
the early Hemphillian. Surprisingly, the youngest known
Amebelodon material is not from North America where Comments: Torynobelodon has been the source of consider-
it is best known but rather from the Sahabi site in North able confusion in the literature. The name Torynobelodon
Africa, which has been dated as Ruscinian (equivalent to was originally given to a single distal tusk tip with fea-
the latest Hemphillian) (Gaziry, 1987). tures as described here. For unknown reasons, however,
Barbour ( 1931) assigned this name to an isolated mandible
Serbelodon Frick, 1933 that bore little resemblance to the original tusk specimen
Type species: Serbelodon barbourensis. beyond possessing dentinal rods. This specimen was later
Type specimen: F:AM 25728 (mandible, partial cranium). reassigned by him to Platybelodon (Barbour, 1932). In
Characteristics: Lower tusks shorter than typical of Gom- addition, as noted earlier, some Amebelodon lower tusks
photherium and flattened, with a profile identical to that possess dentinal rods, and Bennett (1977) incorrectly re-
of Amebelodon; mandibular symphysis shorter than typ- ferred these specimens to Torynobelodon. Tobien (1973)
 Proboscidea 615

considered Torynobelodon to be a distinct taxon closely re- lodon affinities. Specimens from the Withlacoochee River
lated to Platybelodon, and united it with Platybelodon as 4A site (locality GC12II, early Hemphillian) in Florida
a subgenus. Tobien's view is followed here, although we may also conceivably represent Platybelodon, although a
consider the morphological differences between Toryno- referral to Torynobelodon may be somewhat more plau-
belodon and Platybelodon great enough to warrant Toryno- sible based on the apparent narrowness of the tusk frag-
belodon keeping its traditional status as a separate genus ments (see earlier). Platybelodon first appeared in the early
(note that Shoshani [1996] follows Tobien in subsuming Miocene of Africa (Maglio, 1969), well before the en-
Torynobelodon under Platybelodon). Torynobelodon was trance of proboscideans into North America, and thus is
probably an Old World immigrant like Platybelodon. most likely an immigrant to the continent as opposed to an
autochthon.
Platybelodon Borissiak, 1928 (synonym: Torynobelodon in
Barbour, 1931) Indeterminate Gomphotheriidae
Type species: Platybelodon danovi. Reported for localities GC13A (= "Rhynchotherium euhyphodon"
Type specimen: Catalog number unknown, stored in the St. of Webb and Tessman [1968]), CC24, CC32B, SB6, SB7, SB9,
Petersburg Museum, Russia (mandible). CP80, CP90A, CP116E, CP116F, CP124, CP132A, PN8B, PNlO
Characteristics: Lower tusks heavily flattened and consid- (= "Platybelodon" of Shotwell [1963]), PN14.
erably shorter and broader than those of Amebelodon or
Torynobelodon sensu stricto, lacking a dorsal or ventral
OTHER ELEPHANTOIDS
concavity; internal space of lower tusks filled with densely
packed dentinal rods like those found in Torynobelodon; Tetralophodon Falconer, 1857
upper tusks reduced (nearly absent in females), with the Type species: Tetralophodon longirostris.
enamel band vestigial or completely absent in adults; man- Type specimen: Unavailable.
dibular symphysis short and broad; cheek teeth with strong Characteristics: Tetralophodont in all of the intermediate
double trefoiling; and intermediate cheek teeth fully tetra- cheek teeth; five or greater lophs on the M3/m3; reduc-
lophodont (Osborn, 1936, pp. 463-473). tion in the mandibular symphysis; and a tendency toward
Average length of m3: 165.0 mm (North American speci- double trefoiling on the cheek teeth (Tobien, 1978).
men). Average length of m3: 195.0 mm.
Included species: P. barnumbrowni only, known from locality Included species: T. campester ( = T. fricki) only (known from
CP116B only. localities SP2A, CPI 23A).

Comments: The only described Platybelodon specimen in Comments: Of all the gomphotheres sensu Jato found in
North America is a mandible with well-preserved lower North America, only Tetralophodon does not lie within
tusks from the late Clarendonian Ash Hollow Formation the Gomphotheriidae as it is defined in the cladogram in
of Nebraska (locality CP116B), referred to the species Figure 43.5, and instead lies in the unnamed sister group
P. barnumbrowni (Barbour, 1931; Skinner and Johnson, of the gomphotheriids (node E) that contains Mammuthus
1984). This specimen differs from Asian P. grangeri in (family Elephantidae). Tetralophodon is primarily an Old
having dorsoventrally thicker, less dorsally curved lower World genus, although a few North American specimens
tusks, and shallower, less pronounced suprasymphyseal have been referred to it. A major problem with these re-
ridges and alveolar regions on the mandibular symphysis. ferrals is that the most important feature for diagnosing
No intermediate cheek teeth are present on this specimen, this taxon, tetralophodonty in all of the intermediate cheek
so tetralophodonty in this feature cannot be confirmed. teeth, cannot be confirmed in any of the specimens. Indeed,
An undescribed isolated upper tusk of Platybelodon bar- in some cases, even M2/m2 tetralophodonty cannot be con-
numbrowni in the collection of the Nebraska State Mu- firmed, the specimens in question only having a M3 or m3.
seum in Lincoln has been examined by one of the authors The situation is made even more ambiguous by the fact that
(Lambert); this specimen differs from P. grangeri upper the other features characteristic of Tetralophodon (e.g., the
tusks in being distinctly flattened rather than conical and shortening of the mandibular symphysis, double trefoil-
in having a vestigial strip of enamel rather than being with- ing) are strongly prone to parallelism within gomphotheres
out enamel. (Tobien, 1978). Only two North American specimens with
A mandible and lower tusk from the Black Butte site in tetralophodont M2s (the anterior intermediate cheek teeth
Oregon (locality PNlO, late Clarendonian) were referred being unknown) have been referred to Tetralophodon, both
to Platybelodon by Shotwell (1963), based on the extreme from the Clarendonian of the Central Great Plains (Frick,
broadness and flatness of the lower tusk. This referral is 1933; Osborn, 1936). Tobien (1978) has proposed that
erroneous as indicated by the complete absence of dentinal these specimens may represent one or two parallelisms
rods in the lower tusk, and in all likelihood this animal rep- from native North American Gomphotherium stock rather
resents a new gomphothere genus, possibly with Amebe- than immigrants. Thus the status of Tetralophodon in North
 W. David Lambert and Jeheskel Shoshani 616

America is unclear at best. Note added in press. Tassy in some Gomphotherium individuals. Lambert also found that lower
(1996) considers that no Tetralophodon is actually known tusk digging was practiced extensively within the Gomphotherium
from North America, and that specimens of "T. campester' population from the Love site (locality GCl IA), although this be-
are referable to the Amebelodontidae. havior has not been observed elsewhere beyond a single tusk tip
lacking stratigraphic data from the Bone Valley of Florida.

INDETERMINATE PROBOSCIDEA
Megabelodon
Reported for localities GC4D, GC9C, CC25A, CC25B, CC27, The unusual distal end of the mandibular symphysis of Megabelodon
CC30A, CC30B, CC31, CC33, CC35A, CC47, NB19C, NB22B, appears to have had an adaptive significance, the dorsal scar surface
NB23B, NB36II, SPlG, CP75B, CP112. quite likely representing an anchor for a tough pad (presumably
Note added in press: Contrary to the previously believed maxim useful for manipulating vegetation). The apparent adaptiveness of
that no true elephantid is known in North America prior to the Irv- these unusual mandibular symphyseal tips and their morphological
ingtonian (earliest Pleistocene) (see, e.g., Webb, 1992), a specimen uniformity within the taxon strongly argues against this condition
of Mammuthus has recently been described from the Santa Fe River being the result of a simple aberration in Gomphotherium individu-
IB locality in Florida (locality GC14) (Lambert, Cordier, and Chan- als as proposed by Tobien (1973). The presence of this hypothetical
dler, 1995). pad suggests that this animal fed in a manner different from other
gomphotheres, although the exact nature of this difference is unclear.

BIOLOGY AND EVOLUTIONARY PATTERNS Gnathabelodon

The functional significance of Gnathabelodon's broad, spoon-shaped
The Tertiary proboscideans of North America were both taxonom- symphysis is uncertain. Barbour and Sternberg ( 1935) proposed that
ically and morphologically diverse, and this diversity most likely Gnathabelodon was a "mud-grubber," wading and scooping up wa-
reflects a wide ecological range for these animals, particularly re- ter plants with its mandibular symphysis, and there is no reason to
garding feeding habits. Their abundance in Tertiary faunas, in con- believe that this scenario is implausible.
junction with evidence for their feeding habits, strongly suggests
that proboscideans had a major impact on communities through- Stegomastodon
out the continent, in some cases possibly being keystone species In many respects, Stegomastodon is convergent with the elephantids,
like their extant relatives in Africa. Both the taxonomic diversity and with its complex teeth it has been proposed as having been
and overall abundance of proboscideans varied over the course of a grazing specialist (e.g., Kurten and Anderson, 1980), although
the North American Tertiary, and these shifts may have been both a whether it was as specialized a grazer as the elephantids is debatable.
result and a partial cause of habitat changes that swept the continent. Its geographical restriction to western North America may reflect
an ecological dependence on grasslands.

AUTECOLOGY
Amebelodon
Family Mammutidae The feeding behavior of Amebelodon has been a subject of con-
Little direct information is available about the paleoecology of the siderable interest and speculation over the years since the genus
Tertiary mammutids. Pleistocene M ammut is purported to have been was first described, largely because of the strange morphology of
a browser that inhabited forest rather than grassland savanna or open its mandible and lower tusks. Authors traditionally have consid-
grassland, although its wide distribution across the continent sug- ered Amebelodon to have been an aquatic "shoveler," wading in
gests it must have been a habitat generalist to some degree. Stomach shallow ponds and lakes and scooping up water plants with its
contents of Pleistocene Mammut specimens consist of a variety of flattened, shovel-like lower tusks (see also Gnathabelodon earlier)
coarse vegetation including spruce needles, pine cones, and grass, (e.g., Barbour, 1927, 1929b; Osborn, 1936), and this view has be-
with a gourd and grape leaves known for one Florida specimen come prevalent in popular paleontological literature (e.g., Fenton
(Kurten and Anderson, 1980; Webb, 1992; Webb, pers. comm.). The and Fenton, 1958). In addition, for unknown reasons Amebelodon
overall morphological conservatism of mammutids through their is usually provided in restorations with a short, flaplike trunk rather
history suggests that the ecology of Tertiary mammutids did not than a more flexible trunk like that found on modem elephants.
differ greatly from that of their Pleistocene descendants. Lambert (1992), however, demonstrated on the basis of an exami-
nation of both upper and lower tusk wear patterns that Amebelodon
Gomphotherium was a feeding generalist, in all likelihood scooping up aquatic plants
The severe reduction of Gomphotherium in the late Hemphillian as in the traditional view but also scraping bark from tree trunks and
(latest Miocene) with the rest of the Clarendonian chronofauna sug- stripping leaves and twigs from trees in a fashion reminiscent of
gests that this animal favored savanna habitats (see later). Based in modem African elephants. (Note that these behaviors varied widely,
large part on its simple bunolophodont cheek teeth, it is commonly even within at least one population [Moss Acres Racetrack, locality
considered to have been a browser. This hypothesis was supported by GCl lC], with no trends observed at the specific level whatsoever.)
Lambert (1992), who found evidence of tree bark stripping behavior The latter two behaviors were in all likelihood aided by a long,
 Proboscidea 617

w
-~Q) -~
1.B
Z L L.BI.
w E.BI
2.4
_!g
0 4,5 :S
0 E
:::i
Lst. Hp.
L.Hp.
S.2
6.0 e
:;1----+-------,-- -----r---------:S
a !g - .s - ::s
·-.-Clla----e
-~ "§ 11 ~1 =c: ~ =-~ =~ [
=a -- ;~~;
a. L.E.Hp.
E.E.Hp.
...
7.0
e
~ t c: "g 1- -~ ~ § -~ -
§
~ 1- ~
"fi e
·~
c: -
~ )E
~
=~-f
L "lloo
L.CI.

E.CI.
9.5
sQ §
1:J Q)
Q;I- .8-
.Q
-.QQ)
Q)-g
-
-i:J
.Q
-.Q-
Q -e._
Q) - cc - -21-
Q.
--2 -.8 - ~- "'"'
w
z
w
LL.Ba.
11.0

12.5
c:
~
0
.c: -
~
Q -
Q -.Q

-eQi - lJ,
Q)
Ill
:S
Ill
- Q)
E E-9
q: -q: ~ -
c:
i..;
(/)Q)
o
-i-; -
~
-
e-
Q

....
0
0
M E.L.Ba.
14.0 i (!J Iii ::E c3 ~
:ii E.Ba. .Q
Q
5.8 - ~
L.Hm.

17.5 "'
E.Hm.
18.8
E L.L.Ar.
19.2 "Shovel-tuskers" "New World forms"
E.L.Ar. MAMMUTIDAE GOMPHOTHERllDAE
~.or-------------------------------------.JIL----1

w OTHER
L.E.Ar.
z ELEPHANTOIDS
W L
0 27.7
0
CJ E.E.Ar.

:::i 29.4
0
Wl
E 31.9
Or.
33A
L.Ch.
34.5
L M.Ch.
35.5
E.Ch.
37.1

Ou

39.5

L.Un.

w 41.3

z E.Un.
w
0 45.9
0 L.er.
W M
45.7

M.Br.
49,0
E.Br.
50.4

L.Wa.
53,5
E M.W•.
54.2
E.W•.
E.E.W•. SS.O -
Ck.3 ~:~i-----------~~~~~---~~~~~---~-~~~~~~--~~~---l
w ~~ 55.8
Z Tl.6 ~:~
W L TI.5 57.o
0 Tl.4 57.3
0 ~~ 58,5
~ Tl.1
592
To.3- 60.9
ct 81.3
a. To.2
62.5
E To.1 63.8
Pu.3 64.0
Pu.2 64.5
Pu.1 55.0
Pu.O 65.1

Bl.=Blancan, Hp. = Hemphillian, Cl.= Clarendonian, Ba.= Barstovian, Hm.= Hemingfordian, Ar.= Arikareean,
Wt. =Whitneyan, Or.= Orellan, Ch.= Chadronian, Du.= Duchesnean, Un.= Uintan, Br.= Bridgerian, Wa. Wasatchian, =
Ck. = Clarkforkian, Ti. = Tiffanian, To. = Torrejonian, Pu. = Puercan.
Figure 43.6. Temporal ranges of North American proboscidean genera.
 W. David Lambert and Jeheskel Shoshani 618

flexible trunk like that found in modem elephants, and occurred in this case, mixed feeding as opposed to strict grazing is quite plau-
savanna woodland as well as marshes and ponds (see Figure 43.1). sible). As large browsers and hence tree damagers and/or destroy-
Preliminary preparation of oneAmebelodon skull recently recovered ers (presumably those gomphotheres that stripped bark like Gom-
from the Moss Acres Racetrack site reveals enormously retracted photherium, Amebelodon, and Serbelodon killed the trees that they
and deep nasals, providing morphological evidence that the trunk fed upon), the Tertiary North American proboscideans probably eco-
of Amebelodon was immense and presumably very flexible. logically resembled both the modem African and Asian elephants to
a significant degree. Both modem animals feed extensively on trees,
Serbelodon although unlike the African elephant the Asian elephant seldom kills
Lambert (1992) found evidence oflower tusk aquatic scooping and them (Eisenberg, McKay, and Seidensticker, 1990).
tree bark scraping with the upper tusk in some Serbelodon bar- As an abundant tree predator, the African elephant has a major
bourensis specimens, and no sign of tusk feeding behavior in the impact on the landscape structure of modem African savannas. The
one S. barbourensis specimen (lower tusks present on! y) examined. African elephant can be thought of as a "mobile ecological dis-
Thus tusk feeding behavior varied in Serbelodon as in Amebelodon, turbance," consuming and destroying large numbers of trees when
although the extent of this variation is unknown. foraging in woodlands, although the woodlands themselves are ac-
tually destroyed and converted into grassland only under unusual
Torynobelodon conditions. This consistent (although not necessarily continuous)
Little can be said about the paleoecology of Torynobelodon consid- disturbance of woodland has two major ecological effects: (1) a
ering the paucity of material referable to it. Lambert (1992) found canopy is strongly discouraged, forcing the woodland to remain
evidence for lower tusk shoveling in abrasive substrates on the type open; and (2) a stable successional state that would exclude col-
specimen, a behavior also observed in Gomphotherium from the onizer species is not reached, so that plant diversity is enhanced.
Love site (see earlier). These effects, in tum, affect the associated fauna in the following
manner: (1) due to the significantly greater ground insolation result-
Platybelodon ing from the absence of a canopy, the amount of food available for
As withAmebelodon, there has been considerable speculation about relatively short, nonscansorial herbivores is greater than it would
the functional significance of the strongly flattened tusks of Platybe- be otherwise; and (2) the increased plant diversity allows for the
lodon (e.g., Barbour, 1932). Like Amebelodon it is commonly por- possibility of finer niche partitioning among herbivores than would
trayed as an aquatic shoveler wading in shallow ponds and marshes, be possible otherwise, and hence a greater supportable herbivore
and in restorations it is provided with a short, flaplike trunk (see diversity. Because of its major ecological impact on the savanna
Osborn, 1936). Lambert (1992) found no evidence for this behav- community as a whole, the African elephant can be considered a
ior in Platybelodon based on an examination of tusk wear patterns, prime example of a keystone species (Owen-Smith, 1988).
and rather proposed that these unusual lower tusks were used in For most of the Miocene, North America was physiognomically
conjunction with a relatively long, flexible trunk for cutting tough dominated by savanna, with grassland savanna having replaced
vegetation in a sawing fashion (presumably in a terrestrial habitat). woodland savanna in the early and middle Miocene (Webb, 1977; see
He also found that the nasal bones of Asian Platybelodon skulls later). The proboscideans in all likelihood played an ecological role
are as retracted as those of modem elephants, and thus showed that analogous to that of the modem African elephant in this situation,
there is no morphological basis for portraying Platybelodon with a helping to maintain a grassland savanna state. Indeed, it is conceiv-
short, flaplike trunk. able that the immigration ofproboscideans into North America en-
couraged the large-scale Miocene conversion of woodland savanna
Eubelodon, Rhynchotherium, and Cuvieronius to grassland savanna (see Janis [1982] and later discussion).
Little is known about the paleoecology of these genera. As indicated
by their conservative cheek tooth morphologies, all three taxa were
EVOLUTIONARY PATTERN AND ITS
most likely browsers like Gomphotherium. There is no evidence
ECOLOGICAL SIGNIFICANCE
that either Rhynchotherium or Cuvieronius used their upper tusks
extensively in feeding like other browsing gomphotheres, although Proboscideans were a part of the North American fauna from the
distinct wear patterns on a few upper tusk tips referable to these taxa middle Miocene to the end of the Pleistocene (late Hemingfordian-
indicate that they were used to some degree (insufficient Eubelodon Rancholabrean), a span of roughly 17 million years. However, their
material has been examined for any assertions to be made about it, abundance and taxonomic diversity were not uniform over this pe-
although there are no signs of feeding wear on the upper tusks of riod. Their abundance during the Tertiary peaked during the period
the type specimen). from the late Barstovian to the early Hemphillian (late Miocene,
roughly 14--7 million years ago), and they reached a peak in generic
diversity during the late Clarendonian and late early Hemphillian
COMMUNITY ECOLOGY
(roughly 10-7 million years ago) with the following genera known:
A major theme in the paleoecology of Tertiary North American Mammut, Gomphotherium, Megabelodon, Rhynchotherium (in Cen-
proboscideans is browsing; of all the discussed taxa, only Stego- tral America), Amebelodon, Serbelodon, Torynobelodon, and Platy-
mastodon shows adaptations consistent with grazing (and even in belodon. This diversity was greatly reduced by the late Hemphillian
 Proboscidea 619

(7-5.5 million years ago), with only Mammut, Gomphotherium, ACKNOWLEDGMENTS

Rhynchotherium, and Amebelodon surviving (of which only Mam-
mut and Rhynchotherium were even moderately common). Pro- We thank Christine Janis for providing additional proboscidean lo-
boscidean diversity rebounded somewhat in the Pliocene with the calities from the literature, and Brook Hall for supplying additional
appearance of Stegomastodon and Cuvieronius, but proboscideans data on proboscidean localities from her museum notes.
remained a relatively small component of the North American fauna
until the Pleistocene with the arrival of the mammoths.
This diversity pattern roughly mirrors the overall pattern observed REFERENCES
for North American ungulates during the late Miocene. Beginning
at roughly 17 million years ago, the generic diversity of North Barbour, E. H. (1914a). Mammalian fossils from Devil's Gulch. Nebraska
American ungulates (the Clarendonian chronofauna of Webb, Geological Survey, 4, 177-90.
(1914b). A new longirostral mastodon from Cherry County, Nebraska.
[1983]) increased dramatically until it reached a peak at 12 mil-
Nebraska Geological Survey, 4, 212-22.
lion years ago (the early Clarendonian) and then declined, crash- ( 1927). Preliminary notice of a new proboscidean Amebelodon fricki,
ing at approximately 6 million years ago (the late Hemphillian) gen. et sp. nov. Bulletin of the Nebraska State Museum I, 131-34.
(Webb, 1983). This rise in ungulate diversity is attributed to an (1929a). Torynobelodon loomisi gen. et sp. nov. Bulletin of the Nebraska
autochthonous evolutionary burst resulting from a replacement of State Museum,!, 147-53.
(1929b). The mandibular tusks of Amebelodon fricki. Bulletin of the
wood! and savanna, dominant in North America during the Oligocene
Nebraska State Museum, 1, 155-8.
and early Miocene, by a more open grassland savanna whose greater (1931). A new amebelodont, Torynobelodon barnumbrowni sp. nov.: a
habitat complexity allowed more ungulates to coexist without ex- preliminary report. Bulletin of the Nebraska State Museum, l, 191-8.
cessive niche overlap, particularly grazers. As grasslands continued (1932). The mandible of Platybelodon barnumbrowni. Bulletin of the
to spread near the end of the Miocene, however, this habitat mosaic Nebraska State Museum, !, 252-8.
(1934). A mounted skeleton of Megabelodon lulli. Bulletin of the Ne-
was disrupted, woodland loss systematically converting grassland
braska State Museum, l, 303-10.
savanna into steppe. This trend had a devastating effect on both Barbour, E. H., & Hibbard, C. (1941). A shovel-tusked mastodon, Amebe-
browsers and ultimately grazers, resulting in a collapse in ungu- lodon fricki, from Kansas. Bulletin of the Nebraska State Museum,
late diversity over the entire continent (Webb, 1977, 1983; Janis, 2, 35-46.
1993). Barbour, E. H., & Sternberg, G. (1935). Gnathabelodon thorpei, gen. et sp.
nov. A new mud-grubbing mastodont. Bulletin of the Nebraska State
Much of the proboscidean diversity in North America during the
Museum, 1, 395-404.
Miocene was the result of immigration rather than autochthonous Bennett, D. K. (1977). Paleontology, Paleoecology, and Biostratigraphy of
evolution (e.g., Gomphotherium, Torynobelodon, Platybelodon, the Rhinoceros Hill Fauna (Hemphillian, latest Miocene) and Sedi-
possibly Amebelodon and Serbelodon). Origins aside, however, the mentology of the Enclosing Ogallala Formation, Ash Hollow Mem-
overall success- of the proboscideans in the late Miocene indicates ber, Wallace County, Kansas. Unpublished M.S. Thesis. Lawrence:
University of Kansas.
that they were ecologically integrated into the North American
Blumenbach, J. F. (1799). Handbuch Der Naturgeschichte. Giittingen.
grassland savanna community (indeed, it is conceivable that Borissiak, A. A. ( 1928). On a new mastodon from the Chokrak Beds (middle
Gomphotherium entered North America later than the presumed Miocene) of the Kuban region, Platybelodon danovi, n. gen., n. sp.
forest-dwelling mammutid Zygolophodon because of a lag in sa- Annual of the Paleontological Society Russie, 7, 105-20.
vanna development on the continent). If many of these proboscidean Brookes, J. (1828). A Catalog of the Anatomical and Zoological Museum of
Joshua Brookes, Esq., Part I. London: Taylor.
taxa were ecological analogues to the modern African elephant as
Burmeister, H. (1837). Handbuch der Naturgeschichte. Berlin.
evidence suggests (see earlier), then it is highly likely that they were Cook, H.J. (1922). Two new bunomastodonts from Colorado. Proceedings
a major factor in maintaining the mosaic nature of the North Amer- of the Colorado Museum of Natural History, 2, 2-15.
ican grassland savanna, thus helping to support overall ungulate Coppens, Y., Maglio, V. J.,Madden, C. T., &Beden, M. (1978). Proboscidea.
diversity. In Evolution ofAfrican Mammals, eds. V. J. Maglio & H.B. S.Cooke,
pp. 336-67. Cambridge: Harvard University Press.
The hypothetical role of the proboscideans in the North Amer-
Downs, T. ( 1952). A new mastodont from the Miocene of Oregon. University
ican savanna communities has implications for the collapse of the of California Publications in Geological Sciences, 29, 1-20.
Clarendonian chronofauna at the end of the Miocene. The encroach- Eisenberg, J. E., McKay, G. M., & Seidensticker, J. ( 1990). Asian Elephants.
ment by steppe into the woodlands within the savanna most likely Washington, D.C.: Friends of the National Zoo, National Zoological
had a severe negative impact on the proboscidean fauna. As pro- Park, Smithsonian Institution.
Falconer, H. (1857). On the species of mastodon and elephant occurring in
boscidean (particularly gomphothere) abundance and diversity de-
the fossil state in Great Britain Part I. Mastodon. Quarterly Journal
clined, the landscape presumably became much more homogeneous of the Geological Society of London, 13, 307-60.
in the surviving woodland areas and thus of reduced quality. This Falconer, H. (1868). Paleontological Memoirs and Notes of the Late Hugh
effect, in combination with woodland loss as a whole, most likely Falconer with a Biog rap hi cal Sketch of the Author, ed. C. Murchison.
considerably reduced the capacity of the savannas to support a di- London.
Fenton, C. A., & Fenton, M. A. (1958). The Fossil Book. Garden City, NY:
verse ungulate fauna and thus contributed_ to the collapse of the
Doubleday.
late Miocene savanna fauna. Owen-Smith (1988) proposed a sim- Ferrusquia-Villafranca, I. (1984 ). A review of early and middle Miocene Ter-
ilar model incorporating megaherbivore extinction to help explain tiary faunas of North America. Journal of Vertebrate Paleontology,
the terminal Pleistocene extinction event. 4, 187-98.
 W. David Lambert and Jeheskel Shoshani 620

Frick, C. (1933). New remains of trilophodont-tetrabelodont mastodons. Miomastodon of the middle Miocene, new genus. American Museum
Bulletin of the American Museum of Natural History, 59, 505-12. Novitates, 49, J-4.
Gazin, C. L., & Collins, R. E. ( 1950). Remains of land mammals from the (1923). New subfamily, generic, and specific stages in the evolution of
Miocene of the Chesapeake Bay region. Smithsonian Miscellaneous the Proboscidea. American Museum Novitates, 99, 1-4.
Collections, 116, 1-21. (1926). Additional new genera and species of the mastodontid Pro-
Gaziry, A. W. (1987). Remains of Proboscidea from the early Pliocene of boscidea. American Museum Novitates, 238, 221-31.
Sahabi, Libya. In Neogene Paleontology and Geology of Sahabi, eds. (1933). Serbelodon burnhami, a new shovel-tusker from California
N. T. Boaz, A. El-Arnauti, A. W. Gaziry, J. de Heinzelin, & D. D. American Museum Novitates, 639, 1-5.
Boaz, pp. 183-204. New York: Alan R. Liss. (1936). Proboscidea, Vol. I. New York: American Museum Press.
Gidley, J. W. (I 926). Fossil Proboscidea and Edentata of the San Pedro Val- Owen-Smith, N. (1988). Megaherbivores. Cambridge: Cambridge Univer-
ley, Arizona. United States Geological Survey Professional Paper, sity Press.
140B, 83-95. Pohlig, H. (1912). Sur une vieille mandibule de "Tetracaulodon ohiotocum"
Gray, J.E. ( 1821 ). On the natural arrangement of vertebrose animals. London Blum, avec defense in situ. Bulletin de la Societe Beige Geologique,
Medical Repository, 15, 295-310. 26, 187-93.
Gregory, J. T. (1945). An Amebelodon jaw from the Texas Panhandle. Uni- Prothero, D.R., Manning, E. M., & Fischer, M. (1988). The phylogeny of
versity a/Texas Publications, 4401, 477-84. the ungulates. In The Phylogeny and Classification of the Tetrapods,
Guan, J. (1996). On the shovel-tusked elephantoids from China. In The vol. 2, ed. M. J. Benton. Systematics Association Special Volume,
Proboscidea: The Evolution and Palaeoecology of Elephants and 35B, pp. 201-34. Oxford: Clarendon Press.
Their Relatives, eds. J. Shoshani & P. Tassy, pp. 124-35. Oxford: Savage, D. (1955). A survey of various late Cenozoic faunas of the Panhandle
Oxford University Press. of Texas, Part II: Proboscidea. University of California Publications
Hay, 0. P. (1922). Further observations on some extinct elephants. Proceed- in Geological Sciences, 31, 51-74.
ings of the Biological Society of Washington, 35, 97-102. Savage, D. E., & Russell, D. E. (1983). Mammalian Paleofaunas of the
Janis, C. (1982). Evolution of horns in ungulates: ecology and paleoecology. World. Reading, Mass.: Addison-Wesley.
Biological Reviews, 57, 261-318. Schultz, G. E. (1977). Guidebook: field conference on late Cenozoic bios-
Janis, C. M. (1993). Tertiary mammal evolution in the context of changing tratigraphy of the Texas Panhandle and adjacent Oklahoma, August
climates, vegetation, and tectonic events. Annual Review of Ecology 4-6 1977. Kilgore Research Center Special Publication, 1, 1-160.
and Systematics, 24, 467-500. Sellards, E. H. (1940). New Pliocene mastodon. Bulletin of the Geological
Kurten, B., & Anderson, E. (1980). Pleistocene Mammals of North America. Society of America, 51, 1659-64.
New York: Columbia University Press. Shoshani, J. (1996). Para- or monophyly of the gomphotheres and their
Lambert, W. D. (1990). Rediagnosis of the genus Amebelodon (Mam- position within Proboscidea. In The Proboscidea: the Evolution and
malia, Proboscidea, Gomphotheriidae), with a new subgenus and Palaeoecology of Elephants and their Relatives, eds. J. Shoshani &
species, Amebelodon ( Konobelodon) britti. Journal of Paleontology, P. Tassy, pp. 149-77. Oxford: Oxford University Press.
64, 1032-40. Shotwell, J. A. (1963). The Juntura Basin: Studies in earth history and pale-
(1992). The feeding habits of the shovel-tusked gomphotheres: evidence oecology. Transactions of the American Philosophical Society, 53,
from tusk wear patterns. Paleobiology, 18, 132-47. 1-77.
Lambert, W. D., Cordier, D., & Chandler, R. (1995). Mammuthus from the Simpson, G. G. (1930). Tertiary land mammals of Florida. Bulletin of the
late Blancan Santa Fe River !B locality: the oldest record of Mam- American Museum of Natural History, 59, 149-211.
muthus from North America. Journal of Vertebrate Paleontology, 15 ( 1945). The principles of classification and a classification of mammals.
(suppl. to no. 3), p. 40A. Bulletin of the American Museum of Natural History, 85, 1-350.
Madden, C. T. (1986). Stegomastodon associated with Mammuthus during (1980). Splendid Isolation. New Haven: Yale University Press.
the Quaternary. Quaternary Research, 26, 266-71. Simpson, G. G., & Paula-Couto, de, C. (1957). The mastodonts of
Madden, C. T., & Storer, J.E. (1985). The Proboscidea from the middle Brazil. Bulletin of the American Museum of Natural History, 112,
Miocene Wood Mountain Formation, Saskatchewan. Canadian Jour- 125-90.
nal of Earth Science, 22, 1345-50. Skinner, M., & Johnson, W. (1984). Tertiary stratigraphy and the Frick col-
Maglio, V. J. (1969). A shovel-tusked gomphothere from the Miocene of lection of fossil vertebrates from north-central Nebraska. Bulletin of
Kenya. Breviora: Museum of Comparative Zoology, 310, 1-10. the American Museum of Natural History, 178, 215-358.
Mawby, J.E. (1968). Megabelodon minor (Mammalia, Proboscidea), a new Tassy, P. (1988). The classification of Proboscidea: how many cladistic clas-
species of mastodon! from the Esmeralda Formation of Nevada. Pa- sifications? Cladistics, 4, 43-57.
leobios, 4, 1-10. Tassy, P. (I 996). Who is who among the Proboscidea? In The Proboscidea:
May, S., & Repenning, C. (1982). New evidence for the age of the Mt. Eden The Evolution and Palaeoecology of Elephants and Their Relatives,
fauna, Southern California. Journal of Vertebrate Paleontology, 2, eds. J. Shoshani & P. Tassy, pp. 39-48. Oxford: Oxford University
109-13. Press.
McKenna, M. C. (1975). Towards a phylogenetic classification of the Mam- Tassy, P., & Shoshani, J. (1988). The Tethytheria and their relatives. In
malia. In Phylogeny of the Primates, eds. W. P. Luckett & F. S. Szalay, The Phylogeny and Classification of the Tetrapods, Vol. 2, ed. M. J.
pp. 21-45. New York: Plenum. Benton. Systematics Association Special Volume, 35B, pp. 283-315.
Mebrate, A. (1987). The Long-jawed Gomphotheres of North America. Un- Oxford: Clarendon Press.
published Ph.D. Dissertation. Lawrence: University of Kansas. Tedford,R. A., Skinner, M.,Fields,R. W.,Rensberger,R. W., Whistler,D.P.,
Miller, W. E. (1980). The late Pliocene Las Tunas Local Fauna from Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb, S. D. (1987).
southernmost Baja California, Mexico. Journal of Paleontology, 54, Fauna! succession and biochronology of the Arikareean through
762-805. Hemphillian North America. In Cenozoic Mammals of Norlh Amer-
(1990). A Rhynchotherium skull and mandible from southeastern Ari- ica, ed. M. 0. Woodburne, pp. 153-210. Berkeley: University of
zona. Brigham Young University Geology Studies, 36, 57-67. California Press.
Olsen, S. J. (1957). A new beak-jawed mastodon! from Florida. Journal of Tobien, H. (1972). The status of the genus Serridentinus Osborn, .1923
the Palaeontological Society of India, Lucknow, 2, 131-5. (Proboscidea, Mammalia) and related forms. Mainzer Geowis-
Osborn, H. F. (1922). Dibelodon edensis (Frick) of Southern California, senschaftliche Mitteilungen, 1, 143-91.
 Proboscidea 621

(1973). On the evolution of the mastodonts (Proboscidea, Mammalia), America. In Coevolution, ed. M. H. Nitecki, pp. 267-306. Chicago:
Part 1: The bunodont trilophodont groups. Hessisches Landesamt University of Chicago Press.
fuer Bodenforschung Notizblatt, 101, 202-76. (1992). A brief history of New World Proboscidea with emphasis on their
(1975). The structure of the mastodon! molar (Proboscidea, Mammalia), adaptations and interactions with man. In Proboscidean and Pale-
Part 2: The zygodont and zygobunodont patterns. Mainzer Geowis- oindian Interactions, eds. J. W. Fox, C. B. Smith, & K. T. Wilkins,
senschaftliche Mitteilungen, 4, 195-233. pp. 16-34. Waco, Tex.: Baylor University Press.
(1978). On the evolution of the mastodonts (Proboscidea, Mammalia), Webb, S. D., MacFadden B. J., & Baskin, J. A. (1981). Geology and pale-
Part 2: The bunodont tetralophodont groups. Geologisches Jahrbuch ontology of the Love Bone Bed from the late Miocene of Florida.
Hessen, 106, 159-208. American Journal of Science, 281, 513-44.
(1986). Mastodonts (Proboscidea, Mammalia) from the late Neogene Webb, S. D., & Perrigo, S. C. (1984). Late Cenozoic vertebrates from
and early Pleistocene of the People's Republic of China. Mainzer Honduras and El Salvador. Journal of Vertebrate Paleontology, 4,
Geowissenschaftliche Mitteilungen, 15, 119-81. 237-54.
Vacek, M. (1877). Uber Osterreichische Mastodonten und ihre Beziehun- Webb, S. D., & Tessman, N. (1968). A Pliocene vertebrate fauna from low
gen zu den Mastodontarten Europas. Abhandlungen Geologisches elevation in Manatee County, Florida. American Journal of Science,
Reichsanst. Wien, 7, 1-45. 266, 777-811.
Voorhies, M. R. (1990). Vertebrate paleontology of the proposed Norden Whistler, D. P., & Burbank, D. W. (1992). Miocene biostratigraphy and
Reservoir area: Brown, Cherry, and Keya Paha counties, Nebraska. biochronology of the Dove Spring Formation, Mojave Desert, Cal-
Division ofArcheological Research, University ofNebraska, Lincoln ifornia, and characterization of the Clarendonian mammal age (late
Technical Report, 82-9, 1-105. Miocene) in California. Geological Society ofAmerica Bulletin, 104,
Webb, S. D. (1977). A history of savanna vertebrates in the New World, 644-58.
Part 1: North America. Annual Review of Ecology and Systematics, Woodburne, M. 0. (1961). Upper Pliocene geology and paleontology of
8, 355-80. part of the Meade Basin, Kansas. Papers of the Michigan Academy
(1983). The rise and fall of the late Miocene ungulate fauna in North of Science andArls and Letters, 46, 61-101.
 Part VI: Eutheria incertae sedis

44 Eutheria incertae sedis: Mingotherium and Idiogenomys

SPENCER G. LUCAS and ROBERT M. SCHOCH

INTRODUCTION A B

Mingotherium and Idiogenomys of the late Paleocene and late

Eocene, respectively, are two unusual North American mammalian
genera, each based on isolated (but highly distinctive) tribosphenic
teeth. The affinities of these taxa are enigmatic, and given our scant
knowledge of these forms, they are not referable with certainty to
any known order. However, they are a component of the Tertiary D E H
mammal faunas of North America, so for the sake of completeness
they must be included in the present volume. Here, we describe the
basic characteristics of these genera.

F J
INCLUDED GENERA

The locality numbers listed for each genus refer to the list of unified
localities in Appendix I. The acronyms for museum collections are
listed in Appendix III.

Mingotherium Schoch, 1985

Type species: Mingotherium holtae Schoch, 1985.
Type specimen: ChM PV4113, right Ml or M2 (Figure Figure 44.1. Teeth of Mingotherium and ldiogenomys. A-C, holotype of
44.lA-C). Mingotherium holtae, right Ml or M2: A. Occlusal view. B. Anterior view.
Characteristics (after original diagnosis of the genus by C. Posterior view. D-K, Idiogenomys ozziei. D. Holotype, left upper cheek
Schoch, 1985, p. 5): Moderate-sized epitheres (sensu tooth, occlusal view. E. Left upper cheek tooth, occlusal view. F. Right
lower cheek tooth, occlusal view. G. Right lower cheek tooth, labial view.
McKenna, 1975) in which Ml or M2 bear distinct, well-
H-K are the same teeth as D-G with wear facets highlighted in black (D-K
separated, connate, subequal, far labially placed paracones after Ostrander, 1983, Figs. 5-6). Scale bar for A-C = 1 cm; for D-K =
and metacones and large, far lingually placed protocones; 1 mm.
parastyles, metastyles, and hypocones absent; Ml-2 bear
heavy, well-developed, subequal anterior and posterior cin-
gula that are connected by poorly developed labial cin- Mingotherium may have affinities with Pseudictops, uintatheres,
gula, but do not contact lingually; Ml-2 each bear two and xenungulates (see discussion in Schoch, 1985).
moderate-sized, labially placed roots below the paracone Dimensions of sole known tooth, an Ml or M2: Maximum
and metacone, respectively, and a single large, lingually length= 14.8 mm; maximum width= 24.2 mm.
placed root below the protocone; ·Ml-2 characterized by Included species: M. holtae only, known from locality
moderate crown hypsodonty in which enamel extends far- GC19 only (Torrejonian 2 [early Paleocene] of the Gulf
ther up the single lingual root than the labial roots. Coast).

623
 Spencer G. Lucas and Robert M. Schoch 624

Idiogenomys Ostrander, 1983 ARCHAIC UNGULATES AND

Type species: Idiogenomys ozziei Ostrander, 1983. UNGULATELIKE MAMMALS
Type specimen: SDSM 10422, right upper cheek tooth (Fig-
Deltatherium (Eutheria incertae sedis, possibly sister taxon to the
ure 44.10-K).
Pantodonta) and Cyriacotherium (Eutheria incertae sedis, ?Der-
Characteristics (original diagnosis of the genus from
moptera) have been included in Lucas's chapter on Pantodonta
Ostrander, 1983, p. 136): "Extremely small [mammals];
(Chapter 18). H)iolestes (Eutheria incertae sedis, ?Didymoconidae
upper cheek teeth square in outline; brachydont; bunodont;
or ?Cimolestidae) is included in Archibald's chapter on Archaic
cuspate; one large lingual cusp, two or three small buccal
ungulates (Chapter 20).
cusps; two accessory cuspules present on unworn teeth.
Lower cheek teeth squarish in occlusal outline; brachy-
dont; bunodont; cuspate; no more than three cusps, one ARTIODACTYLA
large lingual cusp and two smaller buccal cusps; cusps con-
Apriculus (sometimes included in the Anthracotheriidae) is included
ical on upper and lower cheek teeth." Judged by the size of
in Stucky's chapter on primitive Artiodactyla (Chapter 23). Brachy-
the teeth, Idiogenomys ozziei was an extremely small mam-
hyops (Suoidea incertae sedis) is included in Effinger's
mal. The known teeth appear to be of a basic tribosphenic
chapter on Entelodontidae (Chapter 24). Camelodon (selenodont
pattern, and thus are probably those of a therian, but show
artiodactyl incertae sedis) is included in Prothero's chapter on
no clear similarities to the teeth of any known order.
Oromerycidae (Chapter 28). Delahomeryx (?Ruminantia incertae
Dimensions of known teeth: Length of upper cheek teeth:
sedis) and Hidrosotherium (Selenodontia incertae sedis) are in-
0.66-1.04 mm. Width of upper cheek teeth: 0. 79-1.16 mm.
cluded in Webb's chapter on hornless ruminants (Chapter 31).
Length of lower cheek teeth: 0.83-1.03 mm. Width of
lower cheek teeth: 0.82-0.94 mm.
Included species: /. ozziei only, known from locality CP98B PERISSODACTYLA
only (middle Chadronian [late Eocene] of the Central
Lambdotherium (Perissodactyla incertae sedis) is included in
Plains).
Mader's chapter on Brontotheriidae (Chapter 36). Problematical
perissodactyl taxa Paleomoropus, Schizatherium, and Toxotherium
have been incorporated into the chapter on Tapiroidea and other
OTHER PROBLEMATICAL EUTHERIAN TAXA moropomorphs (Colbert & Schoch, Chapter 39). Paleomoropus is
also briefly discussed in Coombs's chapter on Chalicotherioidea
Various other North American Tertiary mammals are of questionable
(Chapter 38). Procadurocodon (indeterminate amynodontid) is in-
taxonomic position, but they have been housed within the framework
cluded in Wall's chapter on Amynodontidae (Chapter 40).
of other chapters in this volume, in chapters where the average reader
might expect to find them. A brief guide to these taxa is presented
here (added to the chapter by the senior editor).
REFERENCES

CARNIVORA McKenna, M. C. (1975). Toward a phylogenetic classification of the Mam-

malia. In Phylogeny of the Primates, eds. W. P. Luckett & F. S. Szalay,
Nothocyon (arctoid carnivore incertae sedis) is included in Munthe's pp. 21-46. New York: Plenum.
chapter on Canidae (Chapter 7). Palaeogale and Stenogale (Car- Ostrander, G. E. (1983). New early Oligocene (Chadronian) mammals from
the Raben Ranch local fauna, northwest Nebraska. Journal of Pale-
nivora incertae sedis) are included in Baskin's chapter on
ontology, 57, 128-39.
Mustelidae (Chapter 9). Brachyrhyncocyon has been incorporated Schoch, R. M. ( 1985). Preliminary description of a new late Paleocene land-
within the Amphicyonidae (Hunt, Chapter 11), synonymized (as the mammal fauna from South Carolina, U.S.A. Postilla, Peabody Mu-
senior name) with Daphoenocyon. seum, Yale University, 196, 1-13.
 Appendixes

Appendix I: Tertiary mammal localities

WEST COAST MARINE WM13. Round Mountain Silt: (E.L. Barstovian)

Sharktooth Hill Local Fauna.
ALASKA Refs: Tedford et al., 1987; Barnes, 1991 (pers. comm.).
WM14. Briones Formation: (L. Barstovian)
WMl. Unalaska Formation: (L. Oligocene) WM15. Monterey Shale: (Barstovian-Clarendonian)
Unalaska Island. WM16. Santa Margarita Formation: (L.L. Barstovian-E.
Clarendonian)
BRITISH COLUMBIA Scotts Valley, Santa Cruz Co.
Refs: Barnes, 1991 (pers. comm.).
WM2. Sooke Formation: (L. Oligocene)
WM17. Monterey Formation (=Sisquoc Formation): (Clar-
Vancouver Island.
endonian)
WM18. Modelo Formation: (Clarendonian)
WASHINGTON
WM19. Towsley Formation: (E. Hemphillian)
WM3. Pysht Formation: (L. Oligocene) WM20. Purisima Formation (=Drakes Bay Formation):
Clallam Co. (E. Hemphillian)
WM3II. Clallam Formation: (L. Arikareean) Inc. Lava Mountain Fauna.
Merrick Bay and Slip Point, Clallam Bay, Clallam Co. WM21. San Mateo Formation: (E.-L. Hemphillian)
WM4. Astoria Formation: (L. Arikareean-E. Barstovian) A. San Luis Rey River Local Fauna.
Refs: Barnes et al., 1981.
B. Lawrence Canyon Local Fauna.
OREGON Refs: Barnes et al., 1981.
WMS. Yaquina Formation: (Arikareean) WM22. Imperial Formation: (Hemphillian)
WM6. Nye mudstone: (L. Arikareean-E. Hemingfordian) WM23. Paso Robles Formation: (Hemphillian)
Near Newport, Lincoln Co. WM24. Capistrano Formation: (Hemphillian-Blancan)
WM7. Astoria Formation: (L. Hemingfordian) A. Lower part (Hemphillian).
Inc. "Iron Mountain beds." B. Upper part (Blancan).
WM8. Empire Formation: (Hemphillian) WM25. Pismo Formation: (E. Blancan)
WM26. San Diego Formation: (Blancan)
CALIFORNIA Chula Vista District.
Refs: Demere, 1989 (pers. comm.).
WM9. Skooner Gulch Formation: (L. Arikareean)
WMlO. Jewett Sand: (E.L. Arikareean-L.L. Arikareean)
Woody Local Fauna (L.L. Arikareean). BAJA CALIFORNIA, MEXICO
Refs: Barnes, 1991 (pers. comm.). WM27. San Gregorio Formation: (L. Oligocene)
WMl 1. Vaqueros Formation: (L.L. Arikareean) WM28. Ysidro Formation: (Hemingfordian)
Unnamed Local Fauna. WM29. Unnamed formation: (Hemphillian)
Refs: Barnes, 1991 (pers. comm.). El Rosario Arriba.
WM12. Unnamed formation: (Barstovian) WM30. Almejas Formation: (Hemphillian)
625
 Appendix I: Tertiary mammal localities 626

CENTRAL AMERICA Refs: Wood and Wood, 1937; J. Wilson, 1960; Tedford
et al., 1987.
B. Garvin Farm Local Fauna, near Navasota, and
MEXICO GULF COAST Hildago Bluff (E. Hemingfordian).
CAI. Suchilquintongo Formation: (E. Barstovian) Refs: Hesse, 1943; J. Wilson, 1960; Patton, 1969a;
Valle de Oaxaca Local Fauna, Oaxaca Valley (E. Barsto- Forsten, 1975; Tedford et al., 1987.
vian). GC311. Unknown formation: (?L. Arikareean or E. Hem-
Refs: Ferrusquia-Villafranca, 1984, 1990. ingfordian)
CA2. Unnamed unit: (E. Barstovian) Independence, Washington Co.
El Gramal Local Fauna, Oaxaca. GC4. Fleming Formation: (L.L. Arikareean-L.L. Barstovian)
Refs: Ferrusquia-Villafranca, 1984. A. Aiken Hill Local Fauna (L.L. Arikareean).
CA3. Unnamed unit: (? L. Clarendonian) Refs: Tedford et al., 1987.
Ixtapa Faunule, Chiapas. B. Point Blank Fauna (inc. Bennett Farm and Gay Hill)
Refs: Ferrusquia-Villafranca, 1984, 1990. (E. Barstovian).
CA4. Unnamed unit: (E.E. Hemphillian-Blancan) Refs: Hesse, 1943; J. Wilson, 1960; Patton, 1969a;
A. Tehuichila, Hidalgo (E.E. Hemphillian). Forsten, 1975; Hinderstein and Boyce, 1977; Prothero
Refs: Hulbert, 1988a; Carranza-Casteneda and Miller, and Sereno, 1982; MacFadden, 1984; Prothero and
1993. Manning, 1987; Tedford et al., 1987.
B. Techuichila, Hildago (Blancan). C. Trinity River Local Fauna (inc. Trinity River Pit 1, and
CA6. Carrillo Puerto Formation: (L. Hemphillian) MacMurray Pits 1 and 2) and Moscow Local Fauna (E.
A. Rancho Chapas, Yucatan. Barstovian).
Refs: Domning, 1989. Refs:Woodburne, 1969;PattonandTaylor, 1971, 1973;
B. Noc Ac, near Merida, Yucatan. Taylor and Webb, 1976; Baskin, 1982; Prothero and
Refs: Domning, 1989. Sereno, 1982; MacFadden, 1984; Prothero and Man-
ning, 1987; Tedford et al., 1987; Hulbert, 1988a; D.
Bryant, 1991.
OTHER CENTRAL AMERICAN AREAS
D. Burkeville Fauna (E.-L. Barstovian)
CA 7. Cucaracha Formation: (E. Hemingfordian) Refs: Patton, 1969a; Tedford et al., 1987.
Galliad Cut Local Fauna, Canal Zone, Panama. E. Cold Spring Fauna (inc. Noble Farm Fauna, Grime's
Refs: Ferrusquia-Villafranca, 1984. Prairie, Sam Houston Local Fauna and Goodrich Local
CA8. Unnamed unit: (E. Barstovian) Fauna) (L.L. Barstovian).
Carbonaras Creek, Guetamala. Refs: Hesse, 1943; J. Wilson, 1960; Patton, 1969a;
CA9. Gracias Formation: (E.E. Hemphillian) Forsten, 1975; Prothero and Sereno, 1982; Prothero
Gracias Fauna, Honduras. and Manning, 1987; Tedford et al., 1987; Hulbert,
Refs: Webb and Perrigo, 1984. 1988a.
CAIO. Unnamed unit: (E.E. Hemphillian) GC5. Lower Fleming Formation: (E. Hemingfordian)
Corinto Fauna, El Salvador. Carnahan Bayou Member, Toledo Bend.
Refs: Webb and Perrigo, 1984. Refs: Manning, 1990
GC6. Goliad Formation (E. Clarendonian-latest Hemphil-
lian)
GULF COAST A. Lapara Member: Bridge Ranch and Buckner Ranch
Local Faunas (E. Clarendonian).
Refs: Patton, 1969a; Forsten, 1975; J. Wilson, 1978;
TEXAS GULF COAST
Webb and Hulbert, 1986; Hulbert, 1987b, 1988a,
GCI. Yegua Formation: (Duchesnean) 1988b.
Houston Co. B. Lapara Member: Lapara Creek Fauna (inc. Farrish
GC2. Unnamed formation: (L.L. Arikareean) Ranch Fauna and Navasota) (E. Clarendonian).
Dinohyus Site. Refs: Patton, 1969a; Forsten, 1975; J. Wilson, 1978;
Refs: Tedford et al., 1987. Webb and Hulbert, 1986; Hulbert, 1987b; Tedford
GC3. Oakville Formation: (L.L. Arikareean-E. Heming- et al., 1987; Hulbert, 1988a, 1988b.
fordian) C. Labahia Member: Labahia Mission Local Fauna
Garvin Gully Fauna: (E.E. Hemphillian-L.E. Hemphillian).
A. Derrick Farm Local Fauna (=Cedar Run Local Fauna) D. Nueces River Valley (inc. Fordyce Quarry and Sor-
and Cedar Creek (L.L. Arikareean). ensen Ranch Sand and Gravel Pit) (latest Hemphillian).
 Appendix I: Tertiary mammal localities 627

FLORIDA GCI 1. Alachua Formation: (L. Clarendonian-L.E. Hem-

phillian)
GC7. Ocalala Limestone: (Whitneyan or E.E. Arikareean) A. Love Bone Bed Local Fauna (L. Clarendonian).
I-75 Local Fauna, south of Gainesville, plus Cow House Refs: Webb, MacFadden, and Baskin, 1981; Tedford
Slough Local Fauna, west of Thonotosassa. et al., 1987.
Refs: Patton, 1969b. B. Mixon's Bone Bed Local Fauna, McGehee Local
GC8. Hawthorn Formation: (E.L. Arikareean-L.E. Hem- Fauna, and Haile 19A, Coffrin Creek (E.E. Hemphi-
phillian) llian).
A. Buda Local Fauna and SB-lA Local Fauna (E.L. Refs: Simpson, 1930; Webb, 1969b; Webb, 1983;
Arikareean). MacFadden, 1984; Tedford et al., 1987; Hulbert,
Refs: Frailey, 1978, 1979. 1988b.
B. Brooksville Local Fauna and Martin Anthony Local C. Moss Acres Racetrack site (L.E. Hemphillian).
Fauna (=Oreodont Site) (=Unit 9, Tampa Limestone) Refs: Hulbert, 1988a.
(E.L. Arikareean). GC12. Uncertain Formation: (?L. Clarendonian-E. Hem-
Refs: Frailey, 1979; MacFadden, 1980b; Tedford et al., phillian)
1987. Waccasassa River.
C. Franklin Phosphate Co. Pit 2 (L.L. Arikareean). Refs: Domning, 1990.
Refs: Tedford et al., 1987. GCl2II. Uncertain Formation (?Alachua Formation): (L.E.
D. Thomas Farm Local Fauna (E. Hemingfordian). Hemphillian)
Refs: Maglio, 1966; Tedford and Frailey, 1976; Withlacoochee Local Fauna (=Units 4A, 4X, and SE).
MacFadden and Webb, 1982; Tedford et al., 1987. Refs: Becker, 1985; Tedford et al., 1987.
E. Ashville Site and Cummer Lumber Co. Site (E.L. GCl3. Upper part of the Bone Valley Formation: (LE.-
Barstovian). latest Hemphillian)
Refs: MacFadden and Webb, 1982; Tedford et al., 1987; A. Inc. Manatee Local Fauna (L.E. Hemphillian).
Hulbert, l 988a. Refs: Webb and Tessman, 1968; Hulbert, 1988b.
F. Leisey IC Locality (L.E. Hemphillian). B. Upper Bone Valley Fauna (inc. Palmetto Mine, Fort
Refs: Hulbert and Morgan, 1989. Green Mine, near Pierce, near Kingsford, Tiger Bay
GC9. Torreya Formation: (L.L. Arikareean-E. Barstovian) Mine, Swift Mine, Payne Creek Mine, Kingsford Mine,
A. "Lower Hawthorn" (inc. Seaboard Local Fauna, City PhosphoriaMine, Chicora Mine, Nichols [Mobil] Mine,
Waterworks site, Suwannee Springs, Jim Woodruff Lakeland Mine, Bartow Mine, Mulberry Mine, Agrico
Pam site, and Griscom Plantation Local Fauna) Mine, Fort Meade Mine, and Whidden Creek) (latest
(E. Hemingfordian). Hemphillian).
Refs: Tedford et al., 1987. Refs: Webb, 1969b, 1973; MacFadden and Waldrop,
B. Dogtown Member (inc. Midway Fauna and Quincy 1980; MacFadden and Galiano, 1981; Berta and
Local Fauna) (L. Hemingfordian). Galiano, 1983; Wright and Webb, 1984; Berta and
Refs: MacFadden and Webb, 1982; Tedford et al., 1987. Morgan, 1985; MacFadden, 1986; Hulbert, 1987b,
C. Willacoochee Creek Local Fauna (E. Barstovian). 1988a, 1988b; Tedford et al., 1987.
Refs: D. Bryant, 1991. C. Lockwood Meadows Local Fauna, Sarasota Co.
GCIO. Lower Part of the Bone Valley Formation: (latest Hemphillian).
(E. Barstovian-L.E. Hemphillian) Refs: MacFadden, 1986.
A. Units 1-3 sites (inc. Sweetwater Branch site [Arcadia GC14. Unnamed river bottom deposits: (L. Blancan)
Formation, E. Barstovian] and lower Agricola Fauna) Santa Fe River IA and 1B sites.
(E. Barstovian to L.L. Barstovian). Refs: Morgan and Ridgeway, 1987.
Refs: Tedford et al., 1987; Hulbert, I 988a. GC15. Unnamed fissure fill: (L. Blancan)
B. Lower Bone Valley Fauna (inc. Bradley Local Fauna, Haile XVA Sinkhole site.
Noralyn Mine, Kingsford Mine, Nicols Mine, and Refs: Kurten and Anderson, 1980; Morgan and Ridgeway,
Brewster Mine) (L. Barstovian-E. Clarendonian). 1987.
Refs: MacFadden, 1982; MacFadden and Webb, 1982; GC16. Unnamed formation: (L. Blancan)
Berta and Galliano, 1984; Hulbert, 1988a. St. Petersburg Times Site.
C. Unit 5 site (inc. Agricola Fauna, Phosphoria Mine) Refs: Morgan and Ridgeway, 1987.
(E. Clarendonian). GC17. Pinecrest beds: (L. Blancan)
Refs: Hulbert, l 988a. Macasphalt Shell Pit and Sommers Pit, Sarasota Co., and
D. Brewster Phosphate Mine (L.E. Hemphillian). Acline Shell Pit.
Refs: Hulbert, 1987a, 1987b, 1988a. Refs: Morgan and Ridgeway, 1987.
 Appendix I: Tertiary mammal localities 628

GCl8. Tamiami Formation: (L. Blancan) CC2. Delmar Formation: (Bridgerian)

Bass Point Waterway 1 and Northport, Port Charlotte, El Swami's Point.
Jobean, and Mule Pen Quarry. Refs: SDMNH collections.
Refs: Morgan and Ridgeway, 1987. CC3. Scripps Formation: (L. Bridgerian/E. Uintan)
Black's Beach Local Fauna.
SOUTHERN EAST COAST Refs: Walsh, 1991.
CC4. Friars Formation: (E.E. Uintan)
GCl 9. Unnamed formation, Louisiana: (Torrejonian 2)
Poway Fauna (=Friars-Mission Valley Fauna).
Junior Oil Company Beard No. I Well.
Refs: Golz and Lillegraven, 1977; Golz, 1979; Krishtalka
Refs: Archibald et al., 1987.
and Stucky, 1983b; Walsh, 1991.
GC20. Williamsberg Formation, South Carolina: (Clarkfor-
CC5. "Cypress Canyon Unit" (formally part of Mission Val-
kian)
ley Formation): (L.E. Uintan)
Black Mingo Group, Santee Rediversion Canal, near St.
Refs: Walsh, 1991.
Stephen."
CC6. Stadium Conglomerate: (E.-L. Uintan)
Refs: Schoch, 1985.
A. Lower Fauna (E. Uintan).
GC21. Bashi Formation, Mississippi: (E. Wasatchian)
Refs: Golz and Lillegraven, 1977; Schiebout, 1977;
Refs: Beard and Tabrum, 1991.
Walsh, 1991.
GC21II. TuscahomaFormation, Mississippi: (E. Wasatchian)
B. Eastview Local Fauna (Scripps Ranch) and
Red Hot Local Fauna
Stonecrest Local Fauna (L. Uintan).
Refs: Beard, Dawson, and Tabrum, 1995.
Refs: Walsh, 1991.
GC22. Twiggs Clay, Georgia: (L. Duchesnean or E.E. Chad-
CC7. Santiago Formation: (E. Uintan-Duchesnean)
ronian)
A. Mesa Drive Local Fauna (E. Uintan).
Twiggs Co.
B. Jeff's Discovery Local Fauna and Rancho de! Oro Lo-
GC23. Jackson Group, Arkansas: (Uintan/Duchesnean)
cal Fauna (E.L. Uintan).
Crow Creek Local Fauna.
Refs: Walsh, 1991.
Refs: Westgate and Emry, 1985.
C. Laguna Riviera (L. Uintan).
GC24. Byram formation, Mississippi: (Chadronian)
Refs: Golz and Lillegraven 1977; Golz, 1979; Walsh,
GC24II. Unnamed formation, Alabama: (L.E.-E.L. Arika-
1991.
reean)
Conecuh River (E.L. Arikareean). D. Chestnut Avenue Local Fauna (inc. Water Tower Lo-
cality, Blue Bone Locality, and Mission de! Oro)
Refs: Westgate, 1992.
(L. Uintan/E. Duchesnean).
GC25. Hawthorn Formation, South Carolina: (Heming-
Refs: Golz, 1979; SDMNH collections.
fordian)
Dorchester Co. E. Camp San Onofre Local Fauna (E. Duchesnean).
Refs: Golz and Lillegraven, 1977, Walsh, 1991.
GC27. Citronelle Formation, Alabama: (E.E. Hemphillian)
Mobile Local Fauna. CC8. Mission Valley Formation: (L. Uintan)
Refs: Tedford and Hunter, 1984. La Mesa Local Fauna (inc. La Mesa, San Carlos, Fletcher
Hills, East San Diego, and Murphy Canyon).
GC28. Yorktown Formation, North Carolina: (latest Hem-
phillian) Refs: Golz and Lillegraven, 1977; Walsh, 1991.
Lee Creek Mine. CC9. Sespe Formation, Simi Valley area: (L. Uintan-
Refs: Tedford and Hunter, 1984; Eshelman and Whitmore, L.E. Arikareean)
A. Myton Member: Tapo Canyon ( =Tapo Ranch) Local
1997.
Fauna (L. Uintan).
GC29. Citronelle Formation, Louisiana: (latest Hemphillian)
Refs: Golz and Lillegraven, 1977; Krishtalka and
Tunica Hills.
Stucky, 1983b; Kelly, 1990; Kelly and Whistler,
Refs: Manning and MacFadden, 1989.
GC30. ?Yorktown Formation, North Carolina: (L. Blancan) 1994.
Smith Mill Run. AA. Brea Canyon Local Fauna (L. Uintan).
Refs: Kelly, 1990; Kelly and Whistler, 1994.
B. Strathem Local Fauna (latest Uintan/earliest Duch-
CALIFORNIA CENTRAL REGION AND esnean).
COASTAL RANGES Refs: Kelly, 1990; Kelly and Whistler, 1994.
BB. Pearson Ranch Local Fauna (inc. Alamos Canyon
CALIFORNIA sites) (E. Duchesnean).
CCI. Unnamed formation: (Wasatchian) Refs: Golz and Lillegraven, 1977; Krishtalka and
Morena Boulevard Local Fauna. Stucky, 1983b; Kelly, 1990; Lucas, 1992; Kelly and
Refs: Walsh, 1991. Whistler, 1994.
Appendix I: Tertiary mammal localities 629

B2. Simi Valley Landfill Local Fauna (L. Duchesnean). Refs: Kelly and Lander, 1988.
Refs: Kelly and Whistler, 1994. F. Doe Spring Fauna (L.L. Barstovian).
C. Kew Quarry Local Fauna, Ventura Co. (E.E. Refs: James, 1963; Kelly and Lander, 1988.
Arikareean). G. Mathews Ranch Fauna (inc. Cuyama Sites, Apache
Refs: Lander, 1983; Emry, Russell, and Bjork, 1987. Canyon, and Hedgehog Quarry) (E. Clarendonian).
D. Lower and Upper South Mountain Local Fauna, Refs: James, 1963.
Grimes Canyon Local Fauna, Unnamed Local Fauna H. Nettle Spring Fauna (inc. Remnant Hill) (L. Claren-
(Little Simi Valley), Shiells Canyon Local Fauna, donian).
Alamos Canyon Local Fauna, Sycamore Canyon Lo- Refs: James, 1963; UCB collections.
cal Fauna and East Fork of Maria Ygnacio Creek I. "Hh" sites and Sequence Canyon Fauna (L.E. Hem-
(E.E. Arikareean). phillian to L. Hemphillian).
Refs: Lander, 1983. Refs: James, 1963; Tedford et al., 1987; UCB collec-
E. Canton Canyon, Piru Creek - Santa Clara River drain- tions.
age (L.E. Arikarrean). CCl8. Topanga Formation: (L. Hemingfordian-E. Bars-
CC9II. Sespe Formation, Sespe Gorge Area (Uintan) tovian)
Hartman Ranch Local Fauna. Santa Ana Mountains (E. Barstovian).
Refs: Golz, 1979. Refs: Raschke, 1984.
CClO. Sweetwater Formation: (E. Duchesnean) CC19. Branch Canyon Sandstone: (L. Hemingfordian)
Bonita Local Fauna. Vedder Local Fauna.
Refs: Walsh, 1991. Refs: Lindsay, 1974; Munthe, 1979; UCB collections.
CCl 1. Alegria Formation: (E.E. Arikareean) CC20. Olcese Sand: (L. Hemingfordian-E. Barstovian)
Canada de la Gaviota Local Fauna, Santa Ynez Mountains. Barkers Ranch Local Fauna (L. Hemingfordian).
CCl2. Otay Formation: (E.E. Arikareean) Refs: Tedford et al., 1987.
Eastlake fossil localities, San Diego Area. CC21. Bopesta Formation (=Kinnick Formation): (L. Hem-
Refs: Demere, 1988. ingfordian-E.L. Barstovian)
CC13. Tecuya Formation sites: (E.E.-L.E. Arikareean) A. Phillips Ranch Local Fauna (Merychippus carrizaensis
Tecuya Canyon Local Fauna (L.E. Arikareean). Zone) (L. Hemingfordian).
Refs: Stock, 1920; Lander, 1983; Emry, Russell, and Bjork, Refs: Buwalda, 1916; Quinn, 1987.
1987. B. Cache Peak Fauna (Merychippus stylodontus Zone)
CC14. Jewett Sand: (E.L. Arikareean-L.L. Arikareean) (E. Barstovian).
Pyramid Hill Local Fauna (E.L. Arikareean). Refs: Buwalda, 1916; Quinn, 1987; UCB collections.
Refs: Lander, 1983. C. Cache Peak Fauna (Merychippus cf M. intermontanus
CCl5. Tick Canyon Formation: (L.L. Arikareean) Zone) (E.-E.L. Barstovian).
Tick Canyon Local Fauna. Refs: Quinn, 1987.
Refs: Jahns, 1940; Whistler, 1967. CC22. Crowder Formation (=Punchbowl Formation or Ca-
CC16. Vaqueros Formation: (L.L. Arikareean-L. Heming- jon Formation): (L. Hemingfordian-E.L. Barstovian)
fordian) A. U.C.R. RV-6989, Cajon Valley (Crowder Unit 1)
"A:' sites (L.L. Arikareean). (L. Hemingfordian).
Refs: Repenning and Vedder, 1961; Tedford et al., 1987. Refs: Woodbume and Golz, 1972; Reynolds,
CCl7. Caliente Formation: (E. Hemingfordian-L. Hem- Reynolds, and Korth, 1991.
phillian) B. B4, Cajon valley (Crowder Unit 2) (E.L. Barstovian).
A. Below Lower Triple Basalt, Unit 2 (E. Hemingfor- Refs: Reynolds, Reynolds, and Korth, 1991.
dian). CC23. Temblor Formation: (L.E. Arikareean-E.L. Barsto-
B. Hidden Treasure Spring Local Fauna (L. Heming- vian)
fordian). North Coalinga Local Fauna and Domengine Creek (E.L.
Refs: Kelly and Lander, 1988. Barstovian).
C. West Dry Canyon Local Fauna, Jerd Spring Fauna, Refs: Merriam, 1915; Bode, 1935; Woodbume, 1969;
Padrones Spring Local Fauna, Road Gap Canyon 2 LACM collections.
and CIT site 315 (L. Hemingfordian). CC24. Santa Margarita Formation: (L.L. Barstovian-Clar-
Refs: James, 1963; Kelly and Lander, 1988; LACM endonian)
collections. Comanche Point Local Fauna (E. Clarendonian).
D. Upper Dry Canyon Fauna, Lower Dome Spring Fauna, Refs: Savage, 1955.
CIT site 322-323, and Kent Quarry (E. Barstovian). CC25. Merhten Formation: (L.L. Barstovian-L. Hemphillian)
Refs: Kelly and Lander, 1988. A. Two Mile Bar Local Fauna (L.L. Barstovian).
E. Upper Dome Spring Fauna (E.-E.L. Barstovian). Refs: Stirton and Goeriz, 1942.
Appendix I: Tertiary mammal localities 630

B. Turlock Lake and Modesto Reservoir Local Fauna CC41. Hungry Valley Formation and Peace Valley beds:
(L. Hemphillian). (L. Hemphillian)
Refs: Wagner, 1976; LACM and UCB collections. Inc. Kinsey Ranch Local Fauna and Coyote Canyon.
CC26. Green Valley Formation: (E. Clarendonian-E. Blan- Refs: LACM collections.
can) CC41II. Petaluma Formation: (L. Hemphillian)
A. Sycamore Creek Local Fauna (inc. Green Valley site) Sonoma Co.
(E. Clarendonian). CC42. San Joaquin Formation: (E. Blancan)
B. Black Hawk Ranch Local Fauna (L. Clarendonian). Inc. Buttonwillow and Kettleman Hills.
Refs: Macdonald, 1948; Ritchie, 1948; Savage, 1955; CC43. Tehama Formation: (E. Blancan)
Baskin, 1981; UCB collections. Tehama Local Fauna (=Black Ranch).
C. Maxum Local Fauna, Contra Costa Co. (E. Blancan). Refs: Kurten and Anderson, 1980.
CC27. Orinda Formation: (L.L. Barstovian-E. Clarendonian) CC44. Sonoma Volcanics: (E. Blancan)
Orinda -sites (E. Clarendonian). Ritchie Creek.
CC28. Moraga Formation: (E. Clarendonian) Refs: Woodbume, 1966.
CC30. Contra Costa Group: (L. Clarendonian) CC45. Tulare Formation: (E. Blancan)
A. Kendall-Mallory Local Fauna. Asphalto.
B. Layfayette Ridge and Wills Pit. Refs: Kurten and Anderson, 1980.
CC31. San Pablo Group: (E. Clarendonian) CC46. Unnamed unit: (Blancan)
Inc. Tesla, UCMPV5019, Las Trampas Ridge, andingram Aguanga Horizon, Riverside Co.
Creek. CC47. San Timoteo beds: (E. Blancan)
CC32. Chanac Formation: (E. Clarendonian-E.E. Hemphi- Circle C Ranch.
llian)
A. South Tejon Hills Local Fauna (E. Clarendonian).
BAJA CALIFORNIA
Refs: Drescher, 1942; Savage, 1955; Tedford et al.,
CC50. Tepetate Formation: (E. Wasatchian and E. Bridge-
1987.
B. North Tejon Hills Local Fauna (L. Clarendonian). rian)
Refs: Drescher, 1942; Tedford et al., 1987; LACM col- Lomas Las Tetas de Cabra Fauna (=Punta Prieta - Rancho
Rosarito Locality and Occidental Buttes) (E. Wasatchian).
lections.
CC33. Mint Canyon Formation: (L. Clarendonian) Refs: Novacek et al., 1991.
Mint Canyon Local Fauna. CC51. Unnamed unit: (E? Hemingfordian)
La Purisima, Baja California Sur.
CC34. Lake Mathews Formation: (L.E. Clarendonian)
Cajalco Local Fauna. Refs: Ferrusquia-Villafranca, 1984.
CC35. Siesta Formation: (L. Clarendonian-E.E. Hem- CC52. Rosarito Beach Formation: (E. Hemingfordian-Bar-
phillian) stovian)
A. Siesta sites (L. Clarendonian). La Mission, Baja California Norte.
Refs: UCB collections. Refs: Ferrusquia-Villafranca, 1984.
B. Norris Canyon (E.E. Hemphillian). CC53. Unnamed unit: (L. Blancan)
Refs: Webb, 1983. Las Tunas Local Fauna, Baja California Sur.
CC36. Mulholland Formation: (E.E.-L.E. Hemphillian) Refs: Miller and Carranza-Casteneda, 1984.
Mulholland Fauna (L.E. Hemphillian).
Refs: Tedford et al., 1987; UCB collections.
NORTHERN GREAT BASIN
CC37. Pinole Formation: (L. Hemphillian)
Inc. Hemme Hills Local Fauna and Pinole Local Fauna.
Refs: Stirton, 1939; Tedford et al., 1987; UCB collec-
CALIFORNIA GREAT BASIN
tions. NB!. Goler Formation: (Torrejonian 2)
CC38. Kem River Formation: (E.E.-L.E. Hemphillian) Laudate, Southern California.
Kem River sites (L.E. Hemphillian). Refs: McKenna, 1960a, 1960b; Archibald et al., 1987.
Refs: LACM collections. NB2. Lower Red Beds: (Duchesnean)
CC39. Etchegoin Formation sites (E.-L. Hemphillian) Titus Canyon, Death Valley, Grapevine Mountains.
A. Kettleman Hills (E.L. Barstovian). Refs: Stock, 1949.
B. North Dome, Kettleman Hills (L.E. Hemphillian). NB3. Hector Formation: (E.L. Arikareean-L. Hemingfor-
Refs: Merriam, 1915; Tedford et al., 1987. dian)
CC40. Eden beds: (L. Hemphillian) A. Black Butte Mine Local Fauna (E.L. Arikareean).
Mount Eden Local Fauna. Refs: Woodbume et al., 1974.
Refs: Frick, 1921; May and Repenning, 1982; UCB col- B. Logan Mine Fauna (L.L. Arikareean).
lections. Refs: Woodbume et al., 1974; Lander, 1985.
Appendix I: Tertiary mammal localities 631

C. Lower Cady Mountain Local Fauna (E. Heming- Refs: Whistler and Burbank, 1992.
fordian). D. Dove Spring Fauna (Epicyon aphobus/Hipparion
Refs: Lander, 1985. forcei Zone) (L. Clarendonian).
D. Upper Cady Mountain Local Fauna (L. Heming- Refs: Whistler and Burbank, 1992.
fordian). E. Dove Spring Fauna (Paronychomys!Osteoborus dia-
Refs: Lander, 1985; Tedford et al., 1987. bloensis Zone) (E.E. Hemphillian).
NB4. Upper part of the Tropico Group: (E. Hemingfordian) Refs: Whistler and Burbank, 1992.
Boron Local Fauna. NBS. Avawatz Formation: San Bemadino Co. (E. Claren-
Refs: Whistler, 1984; LACM collections. donian)
NBS. Diligencia Formation: (E. Hemingfordian) Unit 2 Avawatz Mountain Fauna.
NB6. Barstow Formation: (L. Hemingfordian-L.L. Bars- Refs: Henshaw, 1939; LACM collections.
tovian) NB9. Bedrock Springs Formation: (E.E. Hemphillian)
A. Owl Conglomerate Member: Red Division Fauna Lava Mountains Fauna.
(L. Hemingfordian). Refs: Tedford et al., 1987.
Refs: Lander, 1985; Woodbume, Tedford, and Swisher, NB 10. Homed Toad Formation: (L. Hemphillian)
1990. Warren Local Fauna.
8. Middle Member: Rak Division Fauna (=Barstow, Refs: Tedford et al., 1987.
Third Division Fauna) (inc. Ledge 3, Coon Canyon, NBll. Alturus Formation: (Hemphillian)
and Unknown Quarry) (L. Hemingfordian). Alturus sites.
Refs: Woodbume, Tedford, and Swisher, 1990. Refs: UCB collections.
C. Middle Member: Green Hills Fauna (inc. Steepside, NB12. Coso Formation: (L. Hemphillian-L. Blancan)
Sunset, Rak, Turbin, Oreodon, Camp, Deep, Raven, A. Coso Mountains. (L. Hemphillian).
Sunrise, Sunder Ridge, and Yermo Quarries, Rain- Refs: Kurten and Anderson, 1980.
bow Canyon, and Daggett Ridge Local Fauna, Alvord B. Coso Mountains. (E. Blancan).
Mountain, Slippery Slide, and Doc's Steepside) (E.E. NB13. Palm Springs Formation: (E. Blancan-E. Irving-
Barstovian). tonian)
Refs: Lander and Reynolds, 1985; Reynolds and A. Layer Cake Local Fauna (E. Blancan).
Lander, 1985; Woodbume, Tedford, and Swisher, Refs: Downs and White, 1968; Kurten and Anderson,
1990; AMNH collections. 1980.
D. Middle Member: Second Division Fauna (inc. Val- B. Arroyo Seco Local Fauna (E. Blancan).
leyview, Sandstone, Mayday, Hailstone, ?Leader, Refs: Downs and White, 1968; Kurten and Anderson,
?Hidden Hollow, and Skyline Quarries, and Saucer 1980.
Butte) (E. Barstovian). C. Anza-Borrego Desert/Vallecito Creek Fauna (L. Blan-
Refs: Kelly and Lander, 1988; Woodbume, Tedford, can).
and Swisher, 1990. Refs: Kurten and Anderson, 1980; MacFadden and
E. Upper Member: Barstow Fauna (=Barstow Fauna, Skinner, 1979.
First Division) (inc. Starlight, New Year, Hemicyon
and Easter Quarries, East of Owl Canyon, Rodent
Hill Basin, and Rodent Hill) (E.L. Barstovian).
NEVADA
Refs: Merriam, 1919; Woodbume, Tedford, and
Swisher, 1990. NB 14. Sheeps Pass Formation: (E. Bridgerian)
F. Cronese Local Fauna (L.L. Barstovian). Elderberry Canyon Local Fauna.
Refs: Lander, 1985. Refs: Emry, 1990.
NB7. Dove Spring Formation (=Ricardo Formation): (E.L. NB15. Unnamed formation: (E.E. Arikareean)
Barstovian-E.E. Hemphillian) Rizzi Ranch Local Fauna, Allegheny Creek, Wild Horse
A. Iron Canyon Fauna (Ustatochoerus profectus/Cope- Range.
mys russelli Zone), Last Chance Gulch and Nightmare NB16. Unnamed formations: (L. Hemingfordian)
Gulch Fauna (E.L.-L.L. Barstovian). Inc. Horse Camp Springs Local Fauna, Red Mountain Lo-
Refs: Whistler and Burbank, 1992. cal Fauna, and Southwestern Fish Creek Mountain.
B. Iron Canyon Fauna (Cupidinimus avawatzensis/Para- NB 17. Unnamed formation: (L. Hemingfordian)
cosoryxfurlongi Zone) (L.L. Barstovian-E. Clarendo- Massacre Lake Fauna.
nian). Refs: Morea, 1981.
Refs: Whistler and Burbank, 1992. NB 18. Virgin Valley Formation: (E. Barstovian)
C. Ricardo Fauna (Epicyon aphobus!Hipparion forcei Virgin Valley Local Fauna.
Zone) (inc. S. burnhami sites) (E.-L. Claren- Refs: Merriam, 1911; Merriam and Stock, 1928; LACM
donian). collections; UCB collections.
Appendix I: Tertiary mammal localities 632

NB19. Carlin Formation (Humboldt Group): (L. Heming- NB28. Unnamed formation: (L. Clarendonian)
fordian-E. Hemphillian) Roglove Local Fauna.
A. Humbolt Group (inc. Carlin Area [Hadley Ranch]) Refs: Nelson and Madsen, 1987.
(E. Barstovian). NB29. Truckee Formation: (L. Clarendonian)
B. UCMP Loe. V4709 (=Fossil Hill) (L.L. Barstovian). Inc. Brady Pocket Local Fauna, Nightingale Road Local
C. Chalk Springs Fauna (inc. Trilolite, North Carlin, Fauna, Kawosh Mountain locality, Hazen locality, Kate
Reese River Crossing, South Reese River, Chalk Peak locality, and Verdi locality.
Spring, South Pine Valley, Mount Lewis, and Elko Refs: Macdonald, 1956; Macdonald and Pelletier, 1958;
Wells localities) (L. Clarendonian). UCB collections.
Refs: Macdonald and Pellatier, 1958. NB30. Unnamed unit: (E. Hemphillian)
D. South Pine Valley (E. Hemphillian). Rabbit Hole, U.C. Loe. V73117.
NB20. Washoe Formation: (E. Barstovian) NB31. Thousand Creek Formation: (E.E.-L. Hemphillian)
High ~ock Canyon Local Fauna. Thousand Creek Fauna (L.E. Hemphillian).
Refs: Stirton, 1940; UCB collections. Refs: Merriam, 1911; Macdonald and Pelletier, 1958;
NB21. Raine Ranch Formation: (E. Barstovian) LACM collections.
Inc. Lone Mountain Local Fauna, Camp Creek Local NB32. Muddy Creek Formation: (Clarendonian-E. Blancan)
Fauna, and Woodruff Creek Local Fauna. A. Mormon Mesa, Clark Co. (E.E. Hemphillian).
Refs: Macdonald, 1966. B. Inc. Double Butte, Limestone Comer, Rodent Quarry,
NB22. Monarch Mill Formation: (E. Barstovian-? Hem- and near Reno (E. Blancan).
phillian) NB33. Panaca Formation: (L. Hemphillian)
A. Near Eastgate (E. Barstovian). Panaca Local Fauna.
B. Eastgate (?Hemphillian). Refs: Macdonald and Pelletier, 1958.
NB23. Esmeralda Formation: (E. Barstovian-L. Claren- NB34. Unnamed unit: (L. Hemphillian)
donian) Golgotha Watermill Pothole Quarry.
A. Early Barstovian site. NB35. Hay Ranch Formation: (L. Hemphillian)
B. Stewart Spring Fauna, and Tonopah Local Fauna Carlin High Quarry.
(E.L. Barstovian). NB36. Unnamed formation: (Blancan)
Refs: Henshaw, 1942; AMNH, LACM, and UCB col- Wichman faunule.
lections. Refs: Macdonald and Pelletier, 1958; Kurten and Ander-
C. Cedar Mountain Local Fauna, Fish Lake Valley Local son, 1980.
Fauna, and near Warrior Mine (E. Clarendonian). NB36II. Unnamed formation: (E.? Blancan)
Refs: Clark, Dawson, and Wood, 1964; Macdonald and Peavine Creek, near Reno.
Pelletier, 1958; UCB collections.
NB24. Stewart Valley Formation: Mineral Co. (E.L. Barsto-
vian) SOUTHERN GREAT BASIN
UCMP Loe. V6020.
NB25. Unnamed unit: (E.L. Barstovian-L. Clarendonian)
ARIZONA
A. Jersey Valley Local Fauna (E.L. Barstovian).
B. Jersey Valley Local Fauna (Dacies Pass) (L. Clarendo- SB 1. "Chalk Canyon" Formation: (E. Orellan)
nian). Cottonwood Creek, Cave Creek area.
NB26. Unnamed Unit: (L. Barstovian-E. Clarendonian) Refs: Lindsay and Tessman, 1974.
Antelope Valley Site SB2. Unnamed formation: (Whitneyan/E.E. Arikareean)
A. Antelope Creek Local Fauna (Barstovian). Atravesada, Pima Co.
B. L.E. Clarendonian site. Refs: Lindsay and Tessman, 1974.
NB27. Coal Valley Formation: (L.L. Barstovian-L.E. SB3. Unnamed unit: (E.?L. Arikareean)
Hemphillian) Muggins Mountains, near Wellton, Yuma Co.
A. Coal Valley Fauna (E. Clarendonian). Refs: Lindsay and Tessman, 1974.
Refs: Axelrod, 1956; Macdonald and Pelletier, SB4. Unnamed formation: (L. Arikareean-Hemingfordian)
1958. Anderson Mine, Yavapai Co. (Hemingfordian).
B. Smiths Valley Fauna (inc. Upper Petrified Tree Refs: Lindsay and Tessman, 1974.
Canyon, Wilson Canyon, and Yerington locality) (L.E. SB5. Unnamed formation: (E. Hemingfordian)
Hemphillian). Black Canyon, Aqua Fria River Drainage.
Refs: Macdonald and Pelletier, 1958; Macdonald, SB6. Milk Creek Formation: (L.L. Barstovian)
1959; Tedford et al., 1987. Inc. Shields Ranch Quarry, Walnut Grove, Hazelwood
NB27II. Unnamed formation: (Clarendonian) Ranch, Iron Springs, and Prescott Southeast.
Seven Miles west of Blair Junction, Esmeralda Co. Refs: Lindsay and Tessman, 1974.
Appendix I: Tertiary mammal localities 633

SB7. Unnamed formation: (E. Clarendonian) Refs: Lindsay and Tessman, 1974; Tomida, 1987.
Greywater Wash, Apache Co. SB16. Unnamed formation: (E. Blancan)
Refs: Lindsay and Tessman, 1974. Cosomi Wash, Santa Cruz Co.
SB8. Unnamed formation: (E. Hemphillian) Refs: Lindsay and Tessman, 1974; Kurten and Anderson,
Bartlette Reservoir, Maricopa Co. 1980.
Refs: Lindsay and Tessman, 1974. SB 17. Fissure fillings in Kaibab Limestone: (L. Blancan-
SB9. Big Sandy Formation: (L. Hemphillian) E. Irvingtonian)
Wikieup Local Fauna (inc. Clay Bank, Birdbone, Horse- Anita Local Fauna (inc. Cal Verde Mine).
shoe, Concretion and Split Cliff Quarries, and Ward's Refs: Hay, 1921; Lindsay and Tessman, 1974; Kurten and
Prospect). Anderson, 1980.
Refs: MacFadden, Johnson, and Opdyke, 1979; MacFad- SB 18. 111 Ranch beds: (L. Blancan)
den and Skinner, 1979. A. Flat Tire Local Fauna.
SBIO. Quiburis Formation: (L. Hemphillian) Refs: Lindsay and Tessman, 1974.
Inc. Old Cabin Quarry, Camel Canyon, Redington Local B. Tusker Local Fauna
Fauna (Reddington Quarry =Bingham Ranch), San Pedro Refs: Lindsay and Tessman, 1974; Kurten and
Valley, and Turtle Pocket. Anderson, 1980.
Refs: Lindsay, Opdyke, and Johnson, 1984. C. Henry Ranch and South Red Knoll Quarry.
SBll. Bidahochi Formation: (L. Hemphillian) Refs: Lindsay and Tessman, 1974.
Inc. White Cone Local Fauna, Jeddito Local Fauna, Taylor SB19. Unnamed formation: (L. Blancan)
Gravel, and Keams Canyon Local Fauna. Snowflake, Navajo Co.
Refs: Baskin, 1979; Lindsay, Opdyke, and Johnson, 1984. Refs: Lindsay and Tessman, 1974.
SB12. Verde Formation: (E. Blancan)
Verde Local Fauna (inc. Clarkdale, Buckboard Wash,
House Mountain, and Sacred Mountain). SOUTHERN COLORADO
Refs: Lindsay and Tessman, 1974; Lindsay, Opdyke, and SB20. Animas Formation, San Juan Basin: (Tiffanian 4-
Johnson, 1984; Czaplewski, 1987. Clarkforkian 2)
SB13. Unnamed formation: (E. Blancan) A. Mason Pocket near Tiffany (Tiffanian 4).
Bear Springs and Matthew Wash. Refs: Archibald et al., 1987.
Refs: Lindsay and Tessman, 1974; Tomida, 1987. B. Bayfield and others (Tiffanian 5).
SB14. St. David Formation Sequence: (E. Blancan- Refs: Archibald et al., 1987.
E. Irvingtonian) C. North of Mason Pocket (Clarkforkian 2).
A. Benson Local Fauna (inc. Gomphothere 6947, Rabbit, SB21. Cuchara Formation: (L.? Wasatchian)
Rat Fink, Carnivore, Betonite, Gray Point, Post Ranch, Huerfano Basin.
East Side Island, Devil's Lawn and Green Saddle Quar- Refs: Robinson, 1963.
ries) (E. Blancan). SB22. Huerfano Formation: (M. Wasatchian-E. Bridgerian)
Refs: Lindsay and Tessman, 1974; Opdyke et al., 1977; A. Lower Huerfano A, Localities VIII, IX, XII (M.
Dalquest, 1978; Harrison, 1978; Kurten and Anderson, Wasatchian).
1980. Refs: Robinson, 1966; Krishtalka and Stucky, 1985.
B. Mendeville Ranch Local Fauna, Cochise Co., and B. Upper Huerfano A, Localities IV, VI, XI (L.
Noye's Bonanza (E. Blancan). Wasatchian).
Refs: Lindsay and Tessman, 1974; Opdyke et al., 1977. Refs: Robinson, 1966; Bown and Kihm, 1981; Schoch
C. McCrae Wash, Honey's Hummock and Horsey Green and Lucas, 1981; Stucky, l 984b; Krishtalka
Bed (E. Blancan). and Stucky, 1985.
Refs: Lindsay and Tessman, 1974; Opdyke et al., 1977. C. Huerfano B and Uppermost Huerfano A, Localities I,
D. Wolf Ranch and Cal Tech (E. Blancan). II, III, (?IV), V, VII (E. Bridgerian).
Refs: Lindsay and Tessman, 1974; Opdyke et al., 1977; Refs: Robinson, 1966; McKenna, 1976; Stucky, 1984b.
Harrison, 1978. SB22II. Farasita Formation (Bridgerian)
E. California Wash and Johnson Pocket (L. Blancan).
Refs: Lindsay and Tessman, 197 4; Opdyke et al., 1977;
NEW MEXICO
Kurten and Anderson, 1980.
F. Curtis Ranch Glyptotherium and Gidley localities SB23. Nacimiento Formation, San Juan Basin: (Puercan 2-
(L. Blancan). Tiffanian 1)
Refs: Lindsay and Tessman, 1974; Opdyke et al., 1977; A. Ectoconus Zone (=Hemithaleus facies) (inc. De-na-
Harrison, 1978; Kurten and Anderson, 1980. Zin Wash (Sinclair and Granger Locality 2, lower
SB15. Unnamed formation: (E. Blancan) part) (Williamson and Lucas Locality 7); and Lower
Duncan, Greenlee Co. Kimbetoh and Betonnie-Tsosie Washes (inc. Sinclair
Appendix I: Tertiary mammal localities 634

and Granger Localities 5, 6, and 7, Black Toe, Black Refs: Tsentas, 1981; Archibald et al., 1987; William-
Stripe, Old Dolan Ranch, Tsosie, Eduardo Arroyo, son and Lucas, 1993; Williamson, 1993.
Mammalon Hill, plus others]) (Williamson and Lu- GG. Animas River Valley (inc. Cedar Hill); Upper Kim-
cas Locality 11) (Puercan 2). beto and Betonnie-Tsosie Washes (inc. Sinclair and
Refs: Tsentas, 1981; Archibald et al., 1987; William- Granger Localities 8 and 9 [lower part], and KU Lo-
son, 1993; Williamson and Lucas, 1993. cality 9 [= Little Pocket]) (Williamson and Lucas
B. Taeniolabis Zone: West Fork of Gallegos Canyon Localities 14 and 15) (Torrejonian ?3 - may span
(inc. Sinclair and Granger Locality 4) (Williamson whole range).
and Lucas Locality 6), andDe-na-Zin Wash (inc. Bar- Refs: Archibald et al., 1987; Williamson, 1993;
rel Spring Arroyo, Coal Creek, Sinclair and Granger Williamson and Lucas, 1993.
Locality 2, upper part) (Williamson and Lucas Lo- H. Panto lambda Zone, in part: (inc. Upper Kutz Canyon
cality 8) (Puercan 3 ). sites [Williamson and Lucas Locality 5], Head of
Refs: Lucas, 1984; Archibald et al., 1987; Williamson, Escaveda Wash [Sinclair and Granger Locality 14]
1993; Williamson and Lucas, 1993. [Williamson and Lucas Locality 17], upper part of
C. Lower Kutz Canyon sites (Williamson and Lucas Lo- West Flank Torreon Wash [=Sinclair and Granger
cality 1); De-na-Zin Wash (Sinclair and Granger Lo- Locality 10, "west branch of Arroyo Torreon"]
cality 3) and Periptychus Site (Williamson and Lucas [Williamson and Lucas Locality 21], upper part of
Locality 9); Kimbetoh and Betonnie-Tsosie Washes East Flank Torreon Wash [=Sinclair and Granger
(inc. Upper Kimbetoh Arroyo Head ofKimbetoh Ar- Locality I, "east branch of Arroyo Torreon"]
royo and Dick's Dig, and upper Betonnie-Tsosie Ar- [Williamson and Lucas Locality 24], plus ?microsite
royo [Power Line]) (Williamson and Lucas Localities ofTsentas) (Torrejonian 3-Tiffanian 1).
12 and 13) (Torrejonian 1). Refs: Archibald et al., 1987; Williamson, 1993;
Refs: Archibald et al., 1987; Williamson, 1993; Williamson and Lucas, 1993.
Williamson and Lucas, 1993. SB24. San Jose Formation, San Juan Basin: (M. Wasatchian)
D. Mesa de Cuba: Type area of Cope's "Puerco marls" Regina and Tapicitos Members, Almagre and Largo beds
(=Simpson's Locality 222) (Williamson and Lucas (inc. Meniscotherium Quarry).
Locality 25) (Torrejonian 1-2). Refs: Lucas et al., 1981; Krishtalka and Stucky, 1986.
Refs: Simpson, 1959; Archibald et al., 1987; SB25. Galisteo Formation: (M. Wasatchian-Duchesnean)
Williamson, 1993; Williamson and Lucas, 1993. A. Cerillos Local Fauna (M. Wasatchian).
DD. Mesa de Cuba (Simpson's Localities 222, 226, and Refs: Lucas and Kues, 1979; Lucas et al., 1981.
229) (Williamson and Lucas Locality 26) (Torrejo- B. Tonque Local Fauna and Windmill Hill (Duchesnean).
nian 2). Refs: Lucas and Kues, 1979; Lucas et al., 1981; Lucas,
Refs: Simpson, 1959; Archibald et al., 1987; 1992.
Williamson and Lucas, 1993; Williamson, 1993. SB26. Baca Formation: (Bridgerian-Duchesnean)
E. Deltatherium Zone: (inc. Kutz Canyon [Williamson A. Carthage Coal Field and east of Rio Grande (Bridge-
and Lucas locality 2] and Head of Gallagos Canyon rian).
[Sinclair and Granger Locality 1, "Chico Springs"] Refs: Lucas et al., 1981; Lucas, 1983; Lucas, 1992.
[Williamson and Lucas Locality 1OJ) (Torrejonian 2). 8. White Mesa and Mariano Mesa, west of Rio Grande
Refs: Archibald et al., 1987; Williamson, 1993; (Uintan-Duchesnean).
Williamson and Lucas, 1993. Refs: Lucas et al., 1981; Lucas, 1983.
F. West Kutz Canyon sites: (inc. Bob's Jaw, Taylor SB26II. Hart Mine Formation: (Bridgerian)
Mound, "Angel's Peak Pocket" [=Big Pocket, KU Refs: Lucas and Williamson, 1993.
locality 13], Bob's Basin, and others) (Williamson SB27. Rubio Peak Formation: (Duchesnean-E. Chadronian)
and Lucas Locality 3) (Torrejonian 2). A. Black Range (Duchesnean).
Refs: Taylor, 1981; Archibald et al., 1987; William- Refs: Lucas 1986a, 1986b.
son, 1993; Williamson and Lucas, 1993. B. Black Range (E. Chadronian).
G. South Kutz Canyon sites: (inc. O'Neill Pantolambda, Refs: Lucas l 986a, l 986b.
Coprolite Point, and others [Williamson and Lucas SB27II. Palm Park Formation: (E. Chadronian)
Locality 4], lower horizon of Sinclair and Granger Refs: Lucas and Williamson, 1993.
Locality 14 ["cliffs at the head of Escaveda Wash"] SB28. Abiquiu Formation: (L.L. Arikareean)
[Williamson and Lucas Locality 16], lower part of Abiquiu Sites.
West Flank Torreon Wash [Williamson and Lucas Lo- Refs: Tedford, 1981.
calities 18-20], and lower part of East Flank Torreon SB29. Zia Sand: (L.L. Arikareean-L. Hemingfordian)
Wash [Sinclair and Granger Locality 11, lower hori- A. Piedra Parada Member (inc. Standing Rock Quarry)
zon] [Williamson and Lucas Localities 22 and 23]) (L.L. Arikareean).
(Torrejonian 3). Refs: Tedford, 1981.
Appendix I: Tertiary mammal localities 635

B. Charnisa Mesa Member (inc. Sandoval Co. site and SB33. Santa Fe Formation: (L.L. Barstovian-L. Blancan)
Blick and Cynarctoides Quarries) (E. Herningfordian). A. Early Site (L.L. Barstovian-E. Clarendonian).
Refs: Tedford, 1981. Refs: Lozinsky and Tedford, 1991.
C. Pilares Member (inc. Straight Cliff Prospect and Cha- B. Middle Red Member and Santa Fe sites (L. Clarendo-
rnisa Mesa Member, Jeep Quarry, and Mesa Prospect) nian).
(E. Herningfordian). Refs: Tedford, 1981; Lozinsky and Tedford, 1991.
Refs: Tedford, 1981. C. Ceja Member and Los Lunas sites (L. Blancan).
D. Kiva Quarry and Arroyo Charnisa Prospect, Charnisa Refs: Tedford, 1981.
Mesa Member (L. Herningfordian). SB33II. Popotosa Formation: (E.E. Hemphillian)
Refs: Tedford, 1981. Gabaldon Badlands site.
SB30. "Upper Zia Sand": (E. Barstovian-E. Clarendonian) Refs: Lozinsky and Tedford, 1991.
A. Rakomylus Site (E. Barstovian). SB34. Charnita Formation: (E. Clarendonian-L. Hemphil-
Refs: Tedford, 1981. lian)
B. Rincon Quarry (E.L. Barstovian). A. Round Mountain Quarry and Battleship Mountain, San
Refs: Tedford, 1981. Ildefonso sites (E. Clarendonian).
C. Middle Red Sites (E. Clarendonian). Refs: Tedford, 1981; Tedford and Barghoorn, 1993;
Refs: Tedford, 1981. USNM collections.
SB31. Gila Group: (L. Arikareean-L. Blancan) B. Lower Charnita sites (E.E. Hemphillian).
A. Black Range, head of Taylor Creek (L. Arikareean). Refs: Tedford and Barghoorn, 1993.
Refs: Tedford, 1981. C. San Juan and Rak Camel Quarries, Osbomoceros
B. Magnus Trench and Dry Creek (L. Hemphillian). Quarry, and Leyden Quarry (L.E. Hemphillian).
Refs: Tedford, 1981. Refs: MacFadden, 1977; MacFadden et al., 1979;
C. Magnus Trench and Duncan Basin (L. Blancan). Tedford and Barghoorn, 1993.
Refs: Tedford, 1981. SB34II. Puye Formation: (E. Blancan)
SB32. Tesuque Formation: (L. Hemingfordian-Hemphillian) Refs: Tedford and Barghoorn, 1993.
A. Nambe Member (L. Hemingfordian). SB35. Palomas Formation: (L. Blancan)
Refs: Tedford, 1981; Tedford and Barghoorn, 1993. Las Palomas Creek sites, Rio Cuchillo Negro Creek, and
B. Skull Ridge Member: Skull Ridge sites (inc. White Hot Springs.
Operation Quarry) (E. Barstovian). Refs: Tedford, 1981.
Refs: Galusha and Blick, 1971; Tedford, 1981; Tedford SB36. Sierra Ladrones Formation: (L. Blancan)
and Barghoorn, 1993. Arroyo de la Parida, Lemitar 1 and 2, Person Power Plant,
C. Chama-El Rito Member: Rio de! Oso-Abiquiu sites and Laguna.
(E. Barstovian). Refs: Tedford, 1981.
Refs: Tedford, 1981; Tedford and Barghoorn, 1993. SB37. Camp Rice Formation: (L. Blancan)
D. Pojoaque Member (inc. Santa Cruz sites and Jacona Chamberino/Canutillo sites (inc. E. Anthony, Vicinity La
Microfossil Quarry) (E.L. Barstovian). Union, Vicinity Anapra, UTEP 33).
Refs: Galusha and Blick, 1971; Tedford, 1981; Tedford Refs: Tedford, 1981; Harris, 1993.
and Barghoorn, 1993. SB38. Axial River Facies: (L. Blancan)
E. Chama-El Rito Member (inc. Rinconada sites) Santo Domingo sites and Elephant Butte.
(E.L. Barstovian). Refs: Tedford, 1981.
Refs: Tedford and Barghoorn, 1993.
EE. Dixon Member: Dixon sites (L.L. Barstovian).
TEXAS BIG BEND AREA
Refs: Tedford and Barghoorn, 1993.
F. Chama-El Rito Member: Chama-El Rito sites SB39. Black Peaks Formation: (Torrejonian-E. Wasatchian)
(E.L. Barstovian). A. Schiebout-Reeves Quarry (Western Tornillo Flats)
Refs: Tedford, 1981; Tedford and Barghoorn, 1993. (Tiffanian 1).
FF. Chama-El Rito Member (inc. Conical Hill Quarry) Refs: Schiebout, 1974.
(L.L. Barstovian). B. Eastern Tornillo Flats Washing Site, Ray's Bone Bed,
Refs: Tedford and Barghoorn, 1993. and Ray's Annex (Tiffanian 3).
G. Pojoaque Member (E. Clarendonian). Refs: Schiebout, 1974; Archibald et al., 1987.
Refs: Tedford and Barghoorn, 1993. C. Joe's Bone Bed (Tiffanian 5).
H. Ojo-Caliente Member (E. Clarendonian). Refs: Schiebout, 1974; Archibald et al., 1987.
Refs: Tedford and Barghoorn, 1993. D. New Taeniodont Site (Tiffanian 6-Clarkforkian).
I. Ojo-Caliente Member: Black Mountain Site (E.E. Refs: Schiebout, 1974.
Hemphillian). E. Southwall (Clarkforkian/Wasatchian).
Refs: Tedford and Barghoorn, 1993. Refs: Schiebout, 1974.
Appendix I: Tertiary mammal localities 636

F. Big Bend National Park (E.E. Wasatchian). S846. Delaho Formation: (E.L. Arikareean-E. Heming-
Refs: Rose, 1981. fordian)
SB3911. Lower Tomillo Formation: (Puercan 2 or 3) Castolon Local Fauna (E.L. Arikareean).
Dogie Locality Refs: Stevens et al., 1969; Stevens and Stevens, 1989.
Refs: Standhardt, 1995. SB47. Rawls Formation: (L.L. Arikareean)
SB40. Hannold Hill Formation: (E. Wasatchian) Santana Mesa Local Fauna, Member 9, Rio Grande Valley.
Jack's Locality, Big Bend National Park. Refs: Stevens, Stevens, and Dawson, 1969; Stevens and
Refs: Rapp, MacFadden, and Schiebout, 1983. Stevens, 1989.
SB41. Canoe Formation: (Bridgerian) SB48. Banta Shut-In Formation: (L. Clarendonian)
Big Yellow Sandstone Member, Big Bend National Park. Screw Beam Local Fauna.
Refs: J. Wilson, 1967; Wilson and Schiebout, 1984. Refs: Stevens, Stevens, and Dawson, 1969; Stevens and
SB42. Laredo Formation: (Uintan) Stevens, 1989.
A. SpHlway Quarry SB49. Camp Rice Formation: (L. Blancan)
B. Casa Blanca Local Fauna (L. Uintan). Huspeth Local Fauna.
Refs: Westgate, 1990. Refs: Akersten, 1972; Kurten and Anderson, 1980.
SB43. Devil's Graveyard Formation, Buck Hill Group: SB50. Love Formation: (L. Blancan)
(Uintan-E. Chadronian) Red Light Local Fauna.
A. Whistler Squat Local Fauna (inc. "Pruett Formation," Refs: Akersten, 1972; Kurten and Anderson, 1980.
Hen Egg Mountain, Wax Camp, and Boneanza)
(E. Uintan).
NORTHERN AND CENTRAL MEXICO
Refs: Gustafson, l 986a; J. Wilson, 1986.
B. Serendipity Local Fauna (inc. Purple Bench Locality, SB51. Unnamed formation: (?Duchesnean)
Titanothere Channels, Stone Creek, and Agua Fria- Marfil Faunule, Guanajuato.
Green Valley Area) (L. Uintan). Refs: Ferrusquia-Villafranca, 1984.
Refs: J. Wilson, 1986. SB52. Prietos Formation: (E. Chadronian)
C. Skyline Channels and Cotter Channels, Bandera Mesa Rancho Gaitan Local Fauna, Chihuahua.
Member (M. Duchesnean). Refs: J. Wilson, 1978; Ferrusquia-Villafranca, 1984, 1993.
Refs: Gustafson, 1986a; J. Wilson, 1986; Wilson and SB53. Unnamed unit: (?Hemingfordian)
Stevens, 1986. Yecora, Sonora.
D. Montgomery Bone Bed, Bandera Mesa Member (L. Refs: Ferrusquia-Villafranca, 1984.
Duchesnean). SB54. Unnamed unit: (L. Hemingfordian)
Refs: J. Wilson, 1986. Tubutama, Sonora.
E. Bandera Mesa Member: Coffee Cup Local Fauna, Refs: Ferrusquia-Villafranca, 1984.
Trans-Pecos (E. Chadronian). SB55. Unnamed unit: (L. Hemingfordian)
Refs: J. Wilson 1984, 1986. Zoyatal Local Fauna, Arroyo Cedazo, Aguascalientes.
SB44. Vieja Group (=Chambers Formation): (Uintan- Refs: Dalquest and Mooser, 1974; Ferrusquia-Villafranca,
M. Chadronian) 1984.
A. Candelaria Local Fauna, Colemena Tuff (L. Uintan). SB56. Unnamed unit: (L. Hemphillian)
Refs: J. Wilson, 1978; Wilson and Schiebout, 1984; Tlaxcala.
Gustafson, l 986a. Refs: Miller and Carranza-Casteneda, 1984.
B. Porvenir Local Fauna, Chambers Tuff (L. Duchesnean ). SB57. Unnamed unit: (L. Hemphillian)
Refs: J. Wilson, 1978; Wilson and Schiebout, 1984; Basuchil, Chihuahua.
Gustafson, 1986a, Lucas, 1992. SB58. Unnamed unit: (L. Hemphillian-E. Blancan)
C. Little Egypt Local Fauna, Chambers Tuff, Trans-Pecos Ocote Local Fauna, Rancho El Ocote, Guanajuato
(inc. Reeves Bone Bed and Chalk Gap Draw) (E. A. L. Hemphillian unit.
Chadronian). Refs: Dalquest and Mooser, 1980; Miller and Carranza-
Refs: J. Wilson, 1977a, 1977b, 1978; Gustafson, 1986a. Casteneda, 1984; Montellano, 1989.
D. Airstrip Local Fauna, Presido Co. (inc. Red Mound B. Latest Hemphillian unit.
North) (E. Chadronian). Refs: Dalquest and Mooser, 1980; Miller and
Refs: J. Wilson, 197la, 197lb, 1974, 1977a, 1977b; Carranza-Casteneda, 1984; Montellano, 1989.
Emry, 1978; Gustafson, 1986a. C. E. Blancan unit.
E. Ash Springs Local Fauna (M. Chadronian). Refs: Dalquest and Mooser, 1980; Millerand Carranza-
Refs: J. Wilson, 1971a, 197lb, 1974, 1977a, 1977b, Casteneda, 1984; Montellano, 1989.
1978, 1984; Emry, 1978. SB59. Unnamed unit: (latest Hemphillian)
SB45. Unnamed formation: (E. Arikareean) Coecillo Fauna, Rancho Viejo area.
University of Texas site 40225. Refs: Miller and Carranza-Casteneda, 1984.
 Appendix I: Tertiary mammal localities 637

SB60. Unnamed unit: (latest Hemphillian) SP2. Clarendon beds (TX): (E.-L. Clarendonian)
Yepomera/Rincon Local Fauna, Chihuahua. Clarendon Local Fauna:
Refs: Furlong, 1941; Lindsay, 1984; Lindsay and Jacobs, A. Inc. McAdams Quarry, Bromley Ranch, Spade Flats,
1985. Charles and Rowe-Lewis Ranch, Stanton Ranch, Dilli,
SB61. Unnamed unit: (E. Blancan) Shannon Ranch, Grant Quarry, Rizley Quarry, Adam
Matachic, Chihuahua. Rizley Farm, and Skillet Creek (E. Clarendonian).
Refs: Lindsay, 1984. Refs: Schultz, 1977; AMNH collections.
SB62. Unnamed unit: (E. Blancan) B. Inc. Gidley Horse Quarry (L. Clarendonian).
Concha Local Fauna, Chihuahua. SP3. Hemphill beds (TX): (L.E.-L. Hemphillian)
Refs: Lindsay, 1984; Lindsay and Jacobs, 1985. A. Box T Local Fauna (L.E. Hemphillian).
SB63. Unnamed unit: (E. Blancan) Refs: Schultz, 1977; Tedford et al., 1987.
Minaca Mesa and Hearst Ranch, Chihuahua. B. Coffee Ranch Local Fauna (=Miami Quarry) (L.
Refs: Kurten and Anderson, 1980; Lindsay, 1984. Hemphillian).
SB64. Unnamed unit: (E. Blancan) Refs: Schultz, 1977; Dalquest, 1983; Tedford et al.,
La Goleta, Michoacan. 1987; AMNH collections.
Refs: Kurten and Anderson, 1980; Miller and Carranza- SP4. Goodnight beds (TX): (L.E.-latest Hemphillian)
Casteneda, 1984. A. Goodnight Local Fauna (L. Hemphillian).
SB65. Unamed unit: (Blancan) Refs: Schultz, 1977; AMNH collections.
Rancho Viejo, Guanajuato. B. Axtel (=Palo Duro Canyon) (TX), Buis Ranch (OK),
Refs: Miller and Carranza-Casteneda, 1984. Currie Ranch (TX), Smart Ranch (TX), and Christian
Ranch (TX) Local Faunas (latest Hemphillian).
Refs: Hibbard, 1954; Schultz, 1977.
SOUTHERN GREAT PLAINS SP5. Blanco Formation (TX): (L. Blancan-Irvingtonian)
Blanco Local Fauna (inc. Low, Marmot, Carter, Mead, and
Red Quarries) (L. Blancan).
TEXAS/OKLAHOMA PANHANDLE
Refs: Dalquest, 1975; Schultz, 1977; Kurten and Ander-
SPl. Ogallala Formation (L. Clarendonian-L. Blancan) son, 1980.
A. Laverne (Beaver) Local Fauna (OK), ExellLocal Fauna
(TX), Durham Local Fauna (OK), Coetas Creek Local
Fauna (TX), T.M.M. 1176, Red River Drainage (TX), CENTRAL GREAT PLAINS
Cole Highway Pit Fauna, and Staked Plains (TX) (L.
Clarendonian).
UTAH
Refs: Dalquest and Hughes, 1966; Schultz, 1977.
B. Capps Pit Local Fauna (George New Pitts) (OK), Ar- CPL North Hom Formation, Wasatch Plateau: (Puercan 2-
nett Local Fauna (=Adair Ranch Quarry) (OK), and Torrejonian 1)
Port of Entry Pit (TX) (E.E. Hemphillian). A. Gas Tank Hill Local Fauna (=Flagstaff Peak) (Puer-
Refs: Schultz, 1977; Tedford et al., 1987. can 2).
C. Higgins Local Fauna (=Sebits Ranch) (TX), Canadian Refs: Robison, 1980.
River sites (TX), and Burson Pit (TX) (L.E. B. Wagonroad (=lower part of Locality 4 of Gazin)
Hemphillian). (Puercan 3).
Refs: Schultz, 1977; MacFadden and Skinner, 1979; Refs: Robison, 1980; Tomida and Butler, 1980.
Tedford et al., 1987. C. Dragon Canyon (=Locality 2 and upper part of Local-
D. Optima (Guymon) Local Fauna (OK), Ogallala Group ity 4 of Gazin), and Sage Flats (Torrejonian 1).
(OK), Turkey Track Canyon (TX), Parcell Ranch (TX), Refs: Robison, 1980; Tomida and Butler, 1980;
and Canadian River Sites (TX) (L. Hemphillian). Archibald et al., 1987.
Refs: Schultz, 1977. CP2. Flagstaff Formation: (Tiffanian)
E. Virgil Clark Pit, Campell Pit, Miller Pit, and Nation Pit Spanish Fork Canyon.
(OK) (latest Hemphillian). CP3. Flagstaff Limestone: (? Wasatchian)
Refs: Tedford et al., 1987. Wasatch Plateau.
F. Beck Ranch Local Fauna (TX) (E. Blancan). Refs: Rich and Collinson, 1973.
Refs: Dalquest, 1978; Madden and Dalquest, 1990. CP4. Colton Formation: (L. Wasatchian)
G. Red Corral Local Fauna (TX).and Channing, Oldham Book Cliffs Area, Uinta Basin.
Co. (TX) (E. Blancan). Refs: Archibald et al., 1987.
Refs: Kurten and Anderson, 1980. CPS. Green River Formation: (Wasatchian-Duchesnean)
H. Cita Canyon Local Fauna (TX) (L. Blancan). A. Powder Wash, Uinta Basin (E. Bridgerian).
Refs: Schultz, 1977; Kurten and Anderson, 1980. Refs: Krishtalka and Stucky, 1984.
 Appendix I: Tertiary mammal localities 638

B. Central Utah, Sanpete Co. (Duchesnean). Refs: Van Valen, 1978; Gingerich, Rose, and Krause,
Refs: Nelson, Madsen, and Stokes, 1980. 1980; Rose, 1981; Gingerich, 1983.
CP6. Uinta Formation: (E.-L. Uintan) F. Airport Locality (Long Draw Quarry), Lower Sand
A. Wagonhound Member, Uinta Basin (inc. White River Draw, Witter(= Croc Tooth) Quarry, Divide Quarry,
Pocket and Kennedy's Basin) (E. Uintan). SandDraw,AnthillLocality, UMSC-243, UMSC-261
Refs: Kay, 1957. (Tiffanian 4).
B. Myton Member, Uinta Basin (inc. Devil's Playground, Refs: Archibald et al., 1987.
Leland Bench Draw, Myton Pocket, Leota Ranch Quarry, G. Princeton Quarry, Schaff Quarry, Fossil Hollow, Brice
and Protylopus Quarry) (L. Uintan). Canyon, Princeton Storm Locality, Fritz Quarry, Jepsen
Refs: Kay, 1957. Valley, Middle Sand Draw, Sunday Locality, and vari-
CP7. Duchesne River Formation, Uinta Basin: (L. Uintan- ous UM localities (Tiffanian 5).
L. Duchesnean) Refs: Gingerich, Rose, and Krause, 1980; Rose, 1981;
A. Bren.nan Basin Member, Randlett Fauna (L. Uintan). Schaff, 1985; Archibald et al., 1987.
Refs: Anderson and Pickard, 1972. H. Reis Locality, Rough Gulch, Foster Gulch Cleopatra
B. Dry Gulch Member, Halfway Fauna (L. Uintan). Reserve Locality, Foster Gulch Oil Well no. l. (Clark-
Refs: Kay, 1957; Anderson and Pickard, 1972. forkian 2).
C. LaPoint Member, Twelve Mile Creek (L.? Duches- Refs: Rose, 1981.
nean). CP14. Fort Union Formation, Washakie Basin: (Torrejonian
Refs: Anderson and Pickard, 1972; Emry, 1981; Lucas, 3-Clarkforkian 2)
1992. A Swain Quarry (Torrejonian 3).
CPS. Norwood Formation: (L. Duchesnean) Refs: Rigby, 1980; Archibald et al., 1987.
Morgan Valley. B. Eastern Rock Springs Uplift: UW localities V-77009-
CP9. Keetley Volcanics: (E. Chadronian) 10, 77012, 77014, 78055 (Torrejonian 3).
Peoa Local Fauna. Refs: Winterfield, 1982.
Refs: Nelson, 1976. C. Eastern Rock Springs Uplift (Plesiadapis churchilli
CPI 0. Salt Lake Formation: (L. Hemingfordian) Zone): UW localities V-7705-08, 7713, 77015-16,
Goose Creek Valley. 77061 (Tiffanian 4).
CPll. Unnamed formation: (L. Blancan) Refs: Winterfield, 1982.
Beaver Local Fauna. D. Eastern Rock Springs Uplift (Plesiadapis simpsoni
Zone): UW localities V-76008, 79059-60, 78052-54
(Tiffanian 5).
WYOMING
Refs: Winterfield, 1982.
CP12. Fort Union Formation, Bighorn Basin: (Puercan) E. Big Multi Local Fauna (=Bitter Creek) (Clarkforkian
A. Mantua Lentil Local Fauna (Polecat Bench) (Puer- 2).
can I). Refs: Rose, 1981; Archibald et al., 1987.
Refs: Van Valen, 1978; Gingerich, Rose, and Krause, CP15. Fort Union Formation, Shotgun Member, Wind River
1980; Archibald et al., 1987. Basin: (Tiffanian 1-Clarkforkian 2)
B. Leidy Quarry (Puercan 1). A. Shotgun Local Fauna (Keefer Hill = Twin Buttes)
Refs: Van Valen, 1978. (Tiffanian 1).
CP13. Fort Union (Polecat Bench) Formation, Bighorn and Refs: Sloan, 1987.
Clark's Fork Basins: (Puercan-Clarkforkian 2) B. "Malcolm's Locality"(= the Badwater Locality) (Tif-
A. UW!ocalities V-81084, 82007, 82010, 82046 (Puercan fanian 4).
undifferentiated). Refs: Krishtalka, Black, and Riedal, 1975; Gingerich,
B. Rock Bench Quarry, plus Hunt Creek Area and Eagle 1979.
Quarry (Montana) (Torrejonian 3). C. West side of Shotgun Butte (Clarkforkian 2).
Refs: Gingerich et al., 1980; Rose, 1981; Butler, Refs: Gazin, 1971.
Krause, and Gingerich, 1987. CP16. Fort Union Group, Bison Basin: (Tiffanian 2-5)
C. UW localities V81050-51, 81053, 81055-63, 82004-6 A Saddle Locality (Tiffanian 2).
(Torrejonian 3). Refs: Gazin, l 956a, 1971.
Refs: Hartman, 1986. B. Ledge Locality, Saddle Annex, and West End (Tiffa-
D. UWlocalitiesSC-263, V-81047, V-82026(Tiffanian2). nian 3).
Refs: Hartman, 1986. Refs: Gazin, 1956a, 1971.
E. Seaboard Well, Cedar Point Quarry, Jepsen Quarry, C. Titanoides Locality (Tiffanian 5).
and various UM localities (Tiffanian 3). Refs: Gazin, 1956a, 1971.
Appendix I: Tertiary mammal localities 639

CP17. Fort Union and Willwood Formations, Bighorn- B. Twin Creek (Tiffanian 3).
Clark's Fork Basins: (Clarkforkian 1-Clarkforkian 2) Refs: Gazin, 1969.
A. Plesiadapis gingerichi Zone (inc. Little Sand Coulee, CP22. Hoback Formation, Hoback Basin: (Tiffanian 3-
Various UM localities [inc. Eastern Clark's Fork Basin Clarkforkian 2)
and Badger Hill], and Bear Creek [Montana] [inc. Sil- A. Battle Mountain, UM-Sub-Wy localities 21 and 22
ver Coulee beds]) (Clarkforkian 1). (Tiffanian 3).
Refs: Rose, 1981; Archibald et al., 1987. Refs: Dorr, 1978; Gingerich, 1983.
B. Plesiadapis cooki Zone (inc. Franimys Hill, Phil's Hill, B. Dell Creek, UM-Sub-Wy locality 1(Tiffanian5).
Paint Creek, Krause Quarry, Holly's microsite, Upper Refs: Dorr, 1978; Gingerich, 1983; Gingerich and
Sand Draw, and various UM localities [inc. Discovery Winkler, 1985.
site, SC-29] Carpolestes skull site, Ectocion site, and C. UM-Sub-Wy localities 7, 10, 20 (Clarkforkian 2).
Unionid Coquina site) (Clarkforkian 2). Refs: Dorr and Gingerich, 1980; Rose, 1981.
Refs: Rose, 1981; Archibald et al., 1987. CP23. Wasatch and Pass Peak Formation, Hoback Basin:
CP18. Willwood Formation, Bighorn-Clark's Fork Basins: (E.-M. Wasatchian)
(Tiffanian 6-Clarkforkian 3) A. UM-Sub-WY localities 2, 4, 16, 28, 29 (E. Wasatchian).
A Various UM localities (Tiffanian 6). Refs: Dorr, 1978.
Refs: Archibald et al., 1987. B. UM-Sub-Wy localities 23 and 27 (M. Wasatchian).
B. Granger Mountain, Rainbow Valley, and various UM Refs: Dorr, 1978.
localities (Phenacodus/Ectocion Zone) (Clarkfork- CP24. Wasatch Formation, Chappo Member: (Tiffanian 3-
ian 3). E.E. Wasatchian)
Refs: Rose, 1981; Archibald et al., 1987. A. Chappo Type Locality ( =Chappo-17) (Tiffanian 3).
CP19. Willwood Formation, Clark's Fork Basin area: Refs: Gingerich, 1979, 1983; Dorr and Gingerich, 1980.
(E.E. Wasatchian) B. Buckman Hollow Locality (=Chappo-I and Chappo-
Inc. Bighorn River area near Worland, Foster Gulch, Hole- 12) (Clarkforkian 2).
in-the-Ground, McCullough Peaks, Sand Coulee area. Refs: Dorr and Gingerich, 1980.
Refs: Gingerich, 1987, 1989. C. Chappo Oil Well (E.E. Wasatchian).
CP19II. Unknown formation, Clark's Fork Basin: Refs: Dorr and Gingerich, 1980.
(M. Wasatchian) CP25. Wasatch Formation: (E. Wasatchian-E. Bridgerian)
UM locality SC-265, three miles south of Jepsen's A. Bitter Creek Area, Zonules 2 and 0, Washakie Basin,
Camp 1. Bitter Creek Promontory section, plus Four Mile Area
CP20. Willwood Formation, Bighorn Basin: (Clarkforkian- (also extends into Colorado) (E. Wasatchian).
E. Bridgerian) Refs: Savage, Waters, and Hutchison, 1972.
A. "Gray Bull beds," Haplomylus-Ectocion Range Zone B. Powder River Basin (inc. Dry Well and Bozeman Lo-
(inc. Sand Creek facies [=No Water Creek area] and calities, Monument Blowout, and Reculusa Blowout)
Elk Creek facies, Foster Gulch area) (E.-M. Wasat- (E. Wasatchian).
chian). Refs: Delson, 1971.
Refs: Bown, 1979, 1980; Bown et al., 1993. C. Bitter Creek and Red Desert Local Faunas, Knight
B. "Gray Bull beds," Bunophorous Interval Zone (L.M. Member, Washakie Basin (E. Wasatchian).
Wasatchian). Refs: Gazin, 1962.
Refs: Shankler, 1980. D. West of Elk Monument Local Fauna, Knight
C. Ly site beds, Heptodon Range Zone (E. L. Wasatchian). Member, Fossil Basin (E. Wasatchian).
Refs: Shankler, 1980. Refs: Gazin, 1962.
D. Lost Cabin beds, Elk Creek Section, Lambdotherium E. Knight Station and Fossil Butte Local Faunas, Knight
Range Zone (L. Wasatchian). Member, Fossil Basin (M. Wasatchian).
Refs: Shankler, 1980. Refs: Gazin, 1962.
E. Wapiti I and II Faunas: North Fork of Shoshone River, F. Dad Local Fauna, Knight Member, Washakie Basin,
Absaroka Range (L. Wasatchian). Niland Tounge (L. Wasatchian).
Refs: Torres, 1985; Gunnell et al., 1992. Refs: Gazin, 1962; Stucky, 1984a.
F. Wapiti III Fauna: North Fork of Shoshone River, Ab- G. La Barge Local Fauna, Knight Member, Bridger Basin
saroka Range (E. Bridgerian). (L. Wasatchian).
Refs: Torres, 1985; Gunnell et al., 1992. Refs: Gazin, 1962.
CP21. Evanston Formation, Fossil B\jsin: (Tiffanian 1-3) H. New Fork Tongue, La Barge area, Green River Basin
A Little Muddy Creek faunule (Tiffanian 1). (inc. Arkosic facies and Western facies) (L. Wasat-
Refs: Gazin, 1969. chian).
Appendix I: Tertiary mammal localities 640

Refs: Gazin, 1962; West, 1973a; Stucky, 1984a. CP29. Wagonbed Formation, Wind River Basin: (M.
I. Cathedral Bluff Tongue, Washakie Basin (?L. Bridgerian-Duchesnean)
Wasatchian). A. Localities 17 and 18, Bad water area (inc. Shirley Basin)
Refs: West and Dawson, 1973; Schoch and Lucas, (E. Bridgerian).
1981. B. Locality 1, Badwater area (E. Uintan).
J. Cathedral Bluff Tongue, Green River Basin (inc. C. Hendry Ranch Member, Localities 5, SA, 6, 7, and as-
Streckfus Draw, Cottonwood Creek area) (E. Bridge- sociated localities, Badwater area (L. Uintan).
rian). Refs: Black and Dawson, 1966; Krishtalka and Se-
Refs: West and Dawson, 1973; Honey, 1988. toguchi, 1977;Black, 1978;Krishtalka, 1979;Dawson,
CP26. Togwotee Pass Area, Wind River Basin: (Tiffanian 3- 1980; Krishtalka and Stucky, 1983b.
E. Wasatchian) D. Hendry Ranch Member, Wood and Rodent Localities
A. Love Quarry (Tiffanian 3). and Locality 20, Badwater area (Uintan/Duchesnean).
Refs: McKenna, 1980b; Archibald et al., 1987. Refs: Black and Dawson, 1966; Krishtalka and Se-
B. "Low Locality" and "Rohrer Locality" (Tiffanian 6). toguchi, 1977; Black, 1978; Krishtalka, 1979; Dawson,
Refs: McKenna, 1980b; Archibald et al., 1987. 1980; Krishtalka and Stucky, 1983b; Maas, 1985.
C. Lower variagated sequence: Purdy Basin, Sheridan Pass CP30. Tatman Formation: (L. Wasatchian)
Quadrangle and Red Creek (Clarkforkian 2). Western Big Hom Basin (="Pitchfor k Formation").
Refs: McKenna, 1980b; Archibald et al., 1987. CP31. Aycross Formation: (L. Wasatchian-M. Bridge-
D. Lower variagated sequence: Hardscrabble Creek, Lava rian)
Mountain, and Tripod Peak Quadrangle, north bank of A. L-41, Love's Locality, Togwotee Pass Area (L.
Papoose Creek (E. Wasatchian). Wasatchian-E. Bridgerian).
Refs: McKenna, 1980b. Refs: McKenna, 1980b.
CP27. Wind River Formation: (E. Wasatchian-E. Bridgerian) B. Togwotee Summit Fauna (=Landslide Locality),
A. Cooper Creek Basin, Laramie Basin (E. Wasatchian). Togwotee Pass (L. Wasatchian/E. Bridgerian).
Refs: Princhinello, 1971; Davidson, 1987. Refs: McKenna, 1980b.
B. Lysite Member (inc. Shirley Basin) (M. Wasatchian). C. Northwest Wind River Basin (E. Bridgerian).
Refs: Guthrie, 1967; Harshman, 1972; Stucky, 1984a. Refs: Stucky, l 984b.
C. Lost Cabin Member, lower part: Boysen Reservoir area D. Wapiti IV Fauna: North Fork of Shoshone River
(M. Wasatchian-E. Bridgerian). (E. Bridgerian).
Refs: White, 1952. Refs: Torres, 1985; Gunnell et al., 1992.
D. Lost Cabin Member, lower part = Lambdotherium E. Owl Creek - Grass Creek Area, Big Hom Basin, plus
Range Zone, Wind River Basin (inc. Red Creek facies, Wapiti V Fauna (M. Bridgerian).
[lower] Deadman Butte section, Arminto Unit, Buck Refs: Bown, 1982; Eaton, 1982; Gunnell et al., 1992.
Spring Quarries, Thirty-One Cent Quarry, Pot Luck CP32. Wapiti Formation: (L. Wasatchian/E. Bridgerian)
Quarry, Bunny Foot Quarry, Halfway Draw Quarry, A. West of Cody (?L. Wasatchian).
Pavillion Butte, and Viverravus locality) (L. Wasat- B. Carter Mountain (Bridgerian).
chian). Refs: Eaton, 1982.
Refs: Guthrie, 1971; Korth and Evander, 1982; Stucky CP33. Green River Formation: (L. Wasatchian)
and Krishtalka, 1982; Stucky, 1984a, 1984b; Stucky, Fossil Lake, Upper Fossil Butte Member.
Krishtalka, and Redline, 1990. CP34. Bridger Formation, Green River Basin: (E. Bridgerian-
E. Lost Cabin Member, upper part = Paleosyops E. Uintan)
Assemblage-Zone (Huerfano B Fauna) (inc. [upper] A. Bridger A localities (E. Bridgerian).
Deadman Butte, Rainbow Butte, and Davis Ranch Refs: McGrew and Sullivan, 1970; Krishtalka and
[=Sullivan Ranch] sections) (earliest Bridgerian). Stucky, 1983b; Gunnell and Bartels, 1994.
Refs: Stucky, 1984a, 1984b. B. Blacks Fork Member (inc. New Fork-Big Sandy Area)
CP28. Indian Meadows Formation: (M.-L. Wasatchian) (M. Bridgerian).
A. Lysite Member: Northwest Wind River (M. Refs: West, 1973a.
Wasatchian). C. Blacks Fork Member (inc. East Hill and Trap-Hutch
Refs: Keefer, 1965. Quarries, Trap 72, George's Gorge, Notharctus Knob,
B. Lost Cabin Member: Northwest Wind River (L. Fort Bridger, Grizzly Buttes, Smith's Fork, Millers-
Wasatchian). ville, Cottonwood Creek, Granger Station, and Sugar
Refs: Keefer, 1965. White) (M. Bridgerian).
C. Shotgun Butte, Eastern Wind River Basin (L. Refs: Gazin, 1976; Stucky, 1984b.
Wasatchian). D. Upper Bridger Formation, Twin Buttes Member,
Refs: Keefer, 1965; Archibald et al., 1987. Tabernacle Butte (inc. Henry's Fork, Hyopsodus Hill,
Appendix I: Tertiary mammal localities 641

Morrow Creek Locality, Sage Mountain and Stuck F. Flagstaff Rim, upper part (Ash F to J) (?inc. Dishpan
Truck, plus Behunin Locality and Misery Quarry) Butte) (M. Chadronian).
(L. Bridgerian). Refs: Emry, 1992.
Refs: McGrew et al., 1959; McKenna, Robinson, and G. Bates Hole/Ledge Creek (M. Chadronian).
Taylor, 1962; West and Atkins, 1970; Gazin, 1976; Refs: Skinner and Gooris, 1966.
West and Hutchinson, 1981; Krishtalka and Stucky, H. Bates Hole (Dilts Ranch, upper section) (Orellan).
1983a. I. Bates Hole, inc. Hall Ranch (="Phinney Butte") (Whit-
CP35. Fowkes Formation: (L? Bridgerian) neyan).
Bulldog Hollow Member, State Line Quarry, Thomas CP39II. White River Formation: (L. Chadronian)
Canyon Local Fauna (also extends into N.E. Utah). A. Red Fauna and West Canyon Creek, (E.E. Chadro-
Refs: Nelson, 1973; Eaton, 1982. nian).
CP36. Tepee Trail Formation, North Fork of Owl Creek Refs: Skinner and Gooris, 1966; Lillegraven and
(=Late Basic Breccia): (L. Bridgerian-E. Uintan) Tabrum, 1983.
A. Lower Tepee Trail Formation (inc. Lower Holy City B. Big Sandstone Draw Lentil, Beaver Divide Area (E.
beds) (L. Bridgerian). Chadronian).
Refs: Eaton, 1985. Refs: Emry, 1975.
B. Upper Tepee Trail Formation (=Bone Bed A, East Fork C. Beaver Divide Conglomerate (?inc. near Toltee,
River, Wind River Basin) (inc. Upper Holy City beds Laramie Mountains). (L. Chadronian).
and Foggy Day beds) (E. Uintan). CP40. White River Formation, Douglas Area, Converse Co.:
Refs: McKenna, 1980b; MacFadden, 1980a; Eaton, (L. Chadronian-Orellan)
1985. A. North Platte River drainage (inc. southeast of Douglas
CP37. Wiggins Formation, North Fork of Owl Creek, and Irvine Bridge area) (L. Chadronian).
Absaroka Mountains: (E. Uintan-L. Uintan) B. North Platte River drainage (inc. southeast of Douglas,
A. Lower Wiggins Formation (E. Uintan). south of Pine Bluffs, Irvine Quadrangle, Reno Ranch,
Refs: Eaton, 1985. Morton Ranch, W.R. Silver Ranch, and Herman Wulff
B. Upper Wiggins Formation (L. Uintan). Ranch) (Orellan).
Refs: Eaton, 1985. CP41. White River Formation, Lusk Area, Niobrara Co.:
CP38. Washakie Formation, Washakie Basin: (M. (L. Chadronian-Whitneyan)
Bridgerian-E. Uintan) A. Seaman Hills Area, Hat Creek drainage, Nuttal Rhino
A. Kinney Rim Member (M. Bridgerian). Quarry, Shack Draw Indian Creek, Whitman Post Off-
Refs: Granger, 1909. ice, Emerson Hines Ranch, Keel Ranch, Kelinke
B. Lower Adobe Town Member, Washakie A (L. Ranch, McGinnis Ranch, Kraft Ranch buildings, Lance
Bridgerian). Creek Basin, Boner Brothers Ranch, Piper Ranch,
Refs: Turnbull, 1972; Roelher, 1973; Archibald et al., Spring Draw, Hat Creek Store, Titanotherium channels
1987. (inc. north of Whitman and Thompson Ranch), plus
C. Middle Adobe Town Member, Washakie B (E. Uintan). (some of these may also extend into Orellan) Road-
Refs: Turnbull, 1972; Roelher, 1973. stop, Anderson Ranch, Boner Bros. Ranch, Jim Chris-
D. Upper Adobe Town Member (E. Uintan). tian Hill, Hermit Ranch, and Lost Springs (L. Chadro-
Refs: Granger, 1909. nian).
CP39. White River Formation, Bates Hole Area, Fremont B. Lusk Area/Hat Creek Drainage (inc. northeast of
Co., and Natrona Co.: (Duchesnean-Whitneyan) Walker Ranch, Niobrara Co., Kraft Ranch, S-Bar Creek,
A. Flagstaff Rim, 95-175 m below Ash A (Duchesnean/ Bald Butte, Indian Creek, East Shack Draw, Spring
E. Chadronian). Draw, Highway 85, Old Woman Creek), DeGerring
Refs: Emry, 1992. Ranch, Ant Hill Locality, Road Stop Locality
B. Flagstaff Rim, below Ash B (inc. Lone Tree Gulch/Dry (Orellan).
Hole Quarry, Dilts Ranch [lower section]) (E. Chad- C. Sherrill Hills (=Seaman Hills), Niobrara Co., plus Tea
ronian). Kettle Rock, Three Tubs Locality, Phinney Ranch
Refs: Emry, 1978, 1979, 1992. (Whitneyan).
C. Flagstaff Rim, Ash B to F (E. Chadronian). CP42. Goshen Hole Area, Goshen Co.: (E. Chadronian -
Refs: Emry, 1975, 1992. E.E. Arikareean)
D. Shirley Basin (west flank of Laramie Range) (E. (inc. Yoder, Brule, and Arikaree Formations)
Chadronian) A. Yoder Local Fauna (E. Chadronian).
E. Cameron Springs Local Fauna, Beaver Rim (L.E. Refs: J. Wilson, 197la, 197lb, 1974; Kihm, 1987.
Chadronian). B. Manganese Pocket, Goshen Hole (E. or M. Chadro-
Refs: Hough, 1955; Wilson, 197la,1971b. nian).
Appendix I: Tertiary mammal localities 642

Refs: Emry, 1978. miles southeast of Chugwater (Platte Co.), near

C. Goshen Hole (inc. Brule Equivalent, Fort Rawhide Buttes, Lay Ranch beds, Spoon Butte (L.L.
Laramie, and Harvard Fossil Reserve) (Orellan). Arikareean).
D. 66 Mountain (inc. Brule Equivalent, Fort Laramie, and B. Arikaree Group, North Platte River drainage and Van
Dog Skull Locality) (Whitneyan/E.E. Arikareean). Tassel (E. Hemingfordian).
CP43. Unnamed formation: (M. Chadronian) CP52. Harrison Formation: (E.L. Arikareean)
Mink Creek Local Fauna. Inc. 77 Hill Quarry, Niobrara Co. (Harrison equivalent),
Refs: McKenna, 1972; Prothero and Shubin, 1989. Riggs Rawhide Creek, Keeline area, Harrison equivalent
CP44. Unnamed formation: (M.Chadronian) (inc. Z Quarry), and Van Tassel Creek.
Pilgrim Creek Local Fauna. Refs: LACM collections.
CP45. Colter Formation, Jackson Hole: (L.E. Chadronian- CP53. Upper Harrison equivalent: (E. Hemingfordian)
L.L. Barstovian) Inc. 7 miles south of Chugwater, Roll Quarry, and Jay Em
A. Emerald Lake Fauna, Washakie Range (could be in District.
White River Formation) (L.E. Chadronian). CP54. Split Rock Formation: (L. Hemingfordian-E.
Refs: Love, McKenna, and Dawson, 1976; Prothero Clarendonian)
and Shubin, 1989. A. Split Rock local Fauna, Sweetwater drainage, Lower
B. Emerald Lake Fauna, Washakie Range (L.E. Porous Sandstone Sequence, and Silty Sandstone Se-
Arikareean). quence (E. Hemingfordian).
Refs: Barnosky, 1986. B. Split Rock Local Fauna, Sweetwater drainage, Gran-
C. East Pilgrim 11 Local Fauna (L. Arikareean). ite Mountain and Devil's Gate Fauna, Arikaree Group,
Refs: Barnosky, 1986. Upper Porous Sandstone Sequence (L. Hemingfordian).
D. East Pilgrim 5 Local Fauna and Saunders Local Fauna Refs: Munthe, 1988.
(E. Hemingfordian). C. Aleurodon Locality and Love's 23V (E. Clarendonian).
Refs: Barnosky, 1986. Refs: Munthe, 1988.
E. Cunningham Hill Fauna, North Pilgrim 2, and Two CP55. Unnamed unit: (Barstovian)
Ocean Lake (L.L. Barstovian). 10 miles northeast of Saratoga, Carbon Co.
Refs: Barnosky, 1986. CP56. Ash Hollow Formation: (L.L. Barstovian)
CP46. Tepee Trail or Wiggins Formation: (Whitneyan) Inc. Trail Creek Quarry Local Fauna and Escarpment
Cedar Ridge Local Fauna. Quarry.
Refs: Setoguchi, 1978; Lillegraven and Tabrum, 1983. Refs: Forsten, 1970; Cassiliano, 1980.
CP46II. Brule Formation: (Whitneyan) CP58. Unnamed formation: (E.E. Hemphillian)
Indian Stronghold, White River Group. Kelley Road Local Fauna.
CP47. Unnamed formation: (E.E. Arikareean) CP59. Teewinot Formation: (L.E. Hemphillian-L. Blancan)
Darton 's Bluff Local Fauna A. Teewinot Local Fauna (L.E. Hemphillian).
Refs: McKenna, 1980c. B. Thayne Local Fauna, Lincoln Co. (L. Blancan).
CP48. Gering Formation (or Arikaree Formation): (E.E. CP60. Unnamed formation: (L. Blancan)
Arikareean) Boyle Ditch Local Fauna.
Inc. Little Muddy Creek, Willow Creek area, North Platte
River drainage, Horse Creek drainage, and Spanish Dig-
gings.
NORTHERN COLORADO
Refs: McKenna, 1972.
CP49. Arikaree Formation: (E. Arikareean-E.L. Barstovian) CP61. Denver Formation, Denver Basin: (Puercan 1-Puercan
A. Split Rock Area, Hypsiops Locality (E.E. Arikareean). 2-3)
B. Bear Creek Mountain, Niobrara Co., and Fox Creek A. Littleton Local Fauna, Alexander and South Table
Gap (L.E. Arikareean). Mountain Localities (Puercan 1).
C. Joe's Quarry, Laramie Co. (L. Hemingfordian/ Refs: Archibald et al., 1987.
E. Barstovian) B. Corral Bluffs, Jimmy Camp Creek, and West Bijou
D. Gravel Quarry, Goshen Hole (E.L. Barstovian). Creek 1 (Puercan 2-3).
CP50. Monroe Creek Formation: (L.E. Arikareean) Refs: Archibald, 1982; Archibald et al., 1987.
Big Muddy Creek (inc. north of Keeline Locality) CP62. Wasatch or Debeque Formation, Atwell Gulch Mem-
CP51. Upper Harrison beds (="Marsland Formation"): ber: (Tiffanian 3/4-Clarkforkian 3)
(L.L. Arikareean-E. Hemingfordian) A. Hummer Locality, Piceance Creek Basin (Tiffanian
A. Inc. Marsland equivalent (Niobrara Co.), south of Lusk, 3/4).
16 Mile District, 18 Mile District, 25 Mile District Refs: Rose, 1981.
(Goshen Co.), Royal Valley District (Niobrara Co.), 5 B. Plateau Valley, and Dry Rock Local Fauna, Piceance
Appendix I: Tertiary mammal localities 643

Creek Basin (Clarkforkian 1). CP69. ?Rosebud Formation: (L. Arikareean)

Refs: Kihm, 1984; Archibald et al., 1987. Gerry's Ranch.
C. Late Clarkforkian site. CP70. Arikaree Formation: (E.E. Arikareean)
Refs: Kihm, 1984. Pawnee Buttes.
CP63. Wasatch Formation, Sand Wash Basin: (E. Wasatchian) CP71. Martin Canyon beds: (E. Hemingfordian-E. Bars-
Four Mile Area, Washakie Basin (also extends into tovian)
Wyoming): Martin Canyon Local Fauna (inc. Quarry A Local Fauna
Inc. Sand, Timberlake, Despair, Anthill, and Kent Quar- [=Runningwater equivalent], Clay Quarry, Grover area,
ries, East Alheit Pocket and West Alheit Pocket. Weld Co., and unnamed Local Fauna) (E. Hemingfordian).
Refs: McKenna, 1960b; Archibald et al., 1987. Refs: Galbreath, 1953; Martin and Mengle, 1984; Tedford
CP64. DeBeque Formation, Piceance Creek Basin (=Shire et al., 1987.
Member of Wasatch Formation): (E. Wasatchian- CP72. North Park Formation: (E.-L. Hemingfordian)
Bridgerian) A. Big Creek Park (E. Hemingfordian).
A. Rifle area, Piceance Creek Basin (E. Wasatchian). B. Piney River Valley (USGS D532b), plus Rickstrew
Refs: Kihm, 1984; Krishtalka and Stucky, 1986. Ranch (USGS D523a) (L. Hemingfordian).
B. Piceance Creek Basin (M. Wasatchian). CP73. Brown's Park Formation: (E.L. Arikareean-L.L. Bars-
Refs: Kihm, 1984. tovian)
C. Piceance Creek Basin (L. Wasatchian). A. Near Sunbeam, lower part of Brown's Park Formation
Refs: Kihm, 1984. (E.? L. Arikareean).
D. Piceance Creek Basin (? Bridgerian). B. Cross Mountain Local Fauna (E.-L. Hemingfordian).
Refs: Kihm, 1984. Refs: Honey and Alzett, 1988.
CP64II. Uinta Formation, Piceance Creek Basin: (Uintan) C. Weller Ranch (L. Barstovian).
CP65. Washakie Formation: (E. Uintan) D. Cedar Springs Draw Local Fauna (L.L. Barstovian).
Sand Wash Basin. Refs: Honey and Alzett, 1988.
Refs: West and Dawson, 1975; Eaton, 1982. CP74. Troublesome Formation: (E. Hemingfordian-E. Cla-
CP66. Antero Formation and Tallahasse Creek Conglomer- rendonian)
ate: (E.L. Chadronian) A. Middle Part (E. Hemingfordian).
South Park, Rocky Mountains, Middle Member. B. Upper Part (inc. UCM Highway 40 site) (E. Barsto-
CP67. Florissant Formation and Castle Rock Conglomerate: vian).
(M. Chadronian) C. Independence Mountain (E. Clarendonian).
Upper' Part, South Park, Rocky Mountains, and Denver CP75. Pawnee Creek Formation: (E. Hemingfordian-L. Bars-
Basin. tovian)
CP68. White River Formation: (?E. Chadronian-Whitneyan) A. KU locality 48 (E. Hemingfordian).
A. Kings Canyon, Medicine Bow Mountains (M. Chadro- B. Lower Pawnee Creek Fauna (inc. Pawnee Quarry, Eu-
nian) banks Fauna, Big Springs Quarry, and Pawnee Buttes
B. Horsetail Creek Member (inc. Horsetail Creek Mem- Area) (E. Barstovian).
ber Kings Canyon, Pawnee Buttes [inc. Trigonias Refs: Galbreath, 1953; Tedford et al., 1987; UCB col-
Quarry = Titanotherium beds, Horizon A], Chalk lections.
Bluffs, Two Miles Creek, Peevy Ranch, George Creek) C. Upper Pawnee Creek Fauna (Pawnee Buttes area) (inc.
(L. Chadronian). Horse and Mastodon Quarries, and West Quarry) (E.L.
Refs: Galbreath, 1953. Barstovian).
C. Cedar Creek Member (inc. Pawnee Buttes [lower part] Refs: Tedford et al., 1987.
[Lower Oreodon beds] Pawnee Creek [lower and upper CP76. Ogallala Formation, lower part: (L.L. Barstovian)
parts], Chimney Canyon, ?Chalk Bluffs, Hirsh Small Upper Pawnee Creek Fauna (inc. Sand Canyon Fauna,
Jaw Locality, Mellinger Locality, Greenley/Pawnee Kennesaw Local Fauna, Vim-Peetz Local Fauna, Tapir
Buttes area, Stone Ranch, Casement Ranch, Clyde Hill, and Uhl Pit).
Ward Ranch, Jack Casement Ranch, Micro Locality, Refs: Galbreath, 1953; Tedford et al., 1987.
Beaver Locality, Chimney Rock, Deadman's Draw, CP77. Unnamed formation: (E. Hemphillian)
Oscar Olander Ranch) (Orellan). Sunny Side Ranch.
Refs: Galbreath, 1953. CP78. Ogallala Group: (L.E. Hemphillian)
D. Vista Member (Whitneyan): Wray Fauna.
Inc. Pawnee Buttes (Leptauchenia beds, Horizon C), Refs: Cook, 1922; Frye, Leonard, and Swineford, 1956;
plus Castle Rock. Tedford et al., 1987.
Refs: Galbreath, 1953. CP80. Unnamed formation: (L. Blancan)
 Appendix I: Tertiary mammal localities 644

Donnelly Ranch. A. Sharps Fauna A and Blue Ash Local Fauna (Whit-
Refs: Hager, 1974; Kurten and Anderson, 1980. neyan).
Refs: Martin, 1974; Tedford et al., 1985.
B. Sharps Fauna B (inc. Quiver Hill localities) (E.E.
SOUTH DAKOTA Arikareean).
CP82. Slim Buttes Formation: (Duchesnean) Antelope Creek Refs: Tedford et al., 1985; AMNH collections.
Local Fauna. C. Wounded Knee/Sharps Fauna (=Sharps Fauna C)
Refs: Bjork, 1967; Lucas, 1992. (E.E. Arikareean).
CP83. Chadron Formation: (E.-L. Chadronian) Refs: Macdonald, 1970; Martin and Green, 1984.
A. Ahearn Member, Big Badlands (inc. Battle Creek D. Godsell Ranch Channel and Blue Ash Channel
Breaks) (E. Chadronian). (E. Arikareean).
Refs: Clark, 1937; Clark, Beerbower, and Kietzke, CP86. Rosebud Formation ( =Arikaree Formation):
1967; J. Wilson, 1974. (E.E. Arikareean-E. Hemingfordian)
B. Crazy Johnson Member, Big Badlands (inc. Indian A. Rosebud Fauna, lower part (E.E. Arikareean).
Creek, Phinney Breaks) (M. Chadronian). Refs: Skinner, Skinner, and Gooris, 1968; LACM col-
Refs: Clark, 1937; Clark, Beerbower, and Kietzke, lections.
1967. B. Upper part of Rosebud Formation and Wounded Knee/
C. Peanut Peak Member, Big Badlands (inc. Granath Monroe Creek Fauna (inc. Wounded Knee/Porcupine
Quarry, Flat Top Buttes, Slim Buttes, south of Lead, Creek area) (L.E. Arikareean).
Black Hills, Big Corral Draw, Little Corral Draw, Boh- Refs: Macdonald, 1970; Martin and Green, 1984.
ling Ranch, Quinn Draw, Stebbins Ranch, Hutenmacher C. Wounded Knee/Harrison Fauna (inc. Eagle Nest Butte
Table, Sect. 12, T42N, R45W, and Upper Titanoth- and Squaw Butte Area) (E.L. Arikareean).
erium beds) (L. Chadronian). Refs: Macdonald, 1970; Martin and Green, 1984.
Refs: Clark, 1937; Clark, Beerbower, and Kietzke, D. Wounded Knee/Rosebud Fauna (=Upper Rosebud
1967; J. Wilson, 1974; Prothero and Shubin, 1989. Fauna) (inc. Porcupine Buttes) (L.L. Arikareean).
CP84. Brule Formation: (Orellan-E.E. Arikareean) Refs: Macdonald, 1970; Martin and Green, 1984.
A Scenic Member, Big Badlands (inc. Short Pine E. Wounded Knee (E. Hemingfordian).
Hills, Lower and Middle Oreodon beds [=Meta- CP87. Valentine Formation: (L.E. Arikareean-L.L. Barsto-
mynodon channels], northwest of Slim Buttes [Unit A], vian)
Slim Buttes Units A-E, Big Corral Draw, Little Corral A. Wewela Fauna, Turtle Butte (L.E. Arikareean).
Draw, "Mauvais Terraces," Cedar Pass, Harris Ranch Refs: Skinner, Skinner, and Gooris, 1968.
Badlands [Unit A], Red Shirt Table, Spring Creek, B. Burge Member: Turtle Butte (L.L. Barstovian).
Sheep Mountain area, south Arrow Wound Table, Cain Refs: Skinner, Skinner, and Gooris, 1968.
Creek, Bear Creek Basin, Chamberlain Pass, Cotton- CP88. Batesland Formation: (E. Hemingfordian)
wood Pass area, Quinn Draw Basin, Cuny Table, Flint Hill Local Fauna.
Harney Spring Range, Babby Butte, and "Lower Refs: Harksen and Macdonald, 1967; Macdonald, 1970;
Nodular Zone") (Orellan). Martin and Green, 1984; Tedford et al., 1987.
Refs: Clark, Beerbower, and Kietzke, 1967; CP89. Fort Randall Formation: (L. Barstovian)
Lillegraven, 1970; Lemley, 1971; Clark and Guens- Bijou Hills Local Fauna.
berg, 1972; Bjork and Macdonald, 1981; Damuth, Refs: Green and Holman, 1977; Green, 1985.
1982; W. Simpson, 1985; Prothero and Shubin, 1989. CP90. Ash Hollow Formation (=Thin Elk Formation):
B. Poleslide Member, Big Badlands (inc. Upper Oreodon (E.-L. Clarendonian)
beds, northwest of Slim Butte [Units B and C], Slim A Inc. Little White River, Big Spring Canyon Local
Buttes [Units F-H], Protocems channels, Indian Stro- Fauna, Bennett Co., Rosebud Agency Quarry, Hol-
nghold, Big Corral Draw, Cedar Pass, Red Shirt Ta- low Hom Bear Quarry, and Joe Thin Elk Gravel Pit
ble, Rockyford, Bloom Ranch, Harris Ranch Badlands (=Mission Local Fauna) (E. Clarendonian).
[Units B-D], Palmer Creek area, Sheep Mountain Ta- Refs: Gregory, 1942; Harksen and Green, 1971;
ble, Leptauchenia clays, Big Coral Draw, Cottonwood UCB collections.
Creek, Battle Creek Draw, Wolff Ranch Badlands, B. Wolf Creek area (E. Clarendonian).
Wolff Lake, Hay Creek) (Whitneyan). Refs: Skinner and Johnson, 1984.
Refs: Lillegraven, 1970; R. W. Wilson, 1984; W. Simp- CP94. Hieb Sand Pit: (L. Blancan)
son, 1985; Prothero and Shubin, 1989. Delmont Local Fauna.
C. Harris Ranch Local Fauna (=Harris Ranch Badlands Refs: Pinsof, 1985.
Unit E) (Whitneyan/E.E. Arikareean). CP95. Herrik Formation: (L. Blancan)
Refs: W. Simpson, 1985. Burke Gravel Pit Locality and Irene Gravel Pit Locality.
CP85. Sharps Formation: (Whitneyan-E. Arikareean) Refs: Pinsof, 1985.
 Appendix I: Tertiary mammal localities 645

CP96. Medicine Root Gravels: (L. Blancan) Refs: Tedford et al., 1985.
Medicine Root Gravel Locality. CPlOO. Whitney Formation: (Whitneyan)
Refs: Pinsof, 1985. A. West of White Clay (lower Whitney Formation).
CP97. Western Alluvium: (L. Blancan) B. Crow Butte Northeast.
Pickstown Locality and Wood Sand Pit Locality. CPlOOII. Arikaree Formation: (Whitneyan or E. Arikareean)
Refs: Pinsof, 1985. Birdcage Gap (Mo- I 06).
CP101. Gering Formation: (E.E. Arikareean)
Inc. Horse Creek Area, Helvas Canyon Member sites
NEBRASKA
(North Platte River drainage), Dunlap Ranch Quarry, Dur-
CP98. Chadron Formation: (E. Chadronian-L. Chadronian) nal Ranch Quarry, George's Rhino Wash, Lone Skull
A. Chadron A (inc. Chadron area, Sioux Co., and Dawes Quarry, No-Name, Mike's Main Channel, Nerud' s Thrasher
Co.) (E. Chadronian). Quarry, Nipple Butte Quarry, Pumice Channel Prospect,
B. Chadron B (inc. Norman Ranch, Raben Ranch Lo- Redding Gap Ant Hill I and 2, Round Rock Gap, Black
cal Faunas, and Bone Cove, Dirty Creek Flats, Dirty Hanks Canyon, Court House Rock, Pumpkin Seed Creek,
Ridge Creek, Twin Buttes, and Ant Hill Local Redington Gap, and Middlebranch Local Fauna.
Faunas) (M. Chadronian). Refs: Voorhies, 1973; Swisher, 1982.
Refs: Hough and Alf, 1956; Ostrander, 1985. CP102. Monroe Creek Formation: (L.E. Arikareean)
BB. Middle-Late Chadronian sites. Inc. Squaw Butte area, Niobrara River drainage, Warbon-
Arner Ranch (Sioux Co.), Warbonnet Creek (Sioux net Creek, Wildcat Hill, and Tunnell Hill Locality.
Co.), and Sect. 2., T32N, R52W (Dawes Co.). CP103. Harrison Formation: (E.L. Arikareean)
C. Chadron C (inc. Chadronia Pocket Local Fauna, Harrison Fauna (inc. Peterson's Quarry A, Syndyoceros
Brecht Ranch Local Fauna [inc. Bartlett West 1, Rook Quarry, Eagle Crag Local Fauna, and Stenomylus Quarry).
Ranch (lower levels), White's south end, Hawthorne Refs: Hunt, 1981, 1985; Skinner and Johnson, 1984; Skol-
A andB, Schomner Ranch (lower level)], Henry Mor- nick, 1985; Tedford et al., 1987; Stevens, 1991.
gan Quarry, Pine Ridge [Toadstool Park], and Geike CP104. Marsland Formation (=Upper Harrison beds): (L.L.
Ranch) (L. Chadronian). Arikareean-E. Hemingfordian)
Refs: Ostrander, 1985; Prothero, Denham, and Far- A. Agate Spring Local Fauna (inc. University Quarry,
mer, 1982; Gustafson, 1986b. Harper Quarry, Quarry 1, Quarry 3, Agate Spring
CP99. Brule Formation, White River Group: (Orellan- Quarry, American Museum-Cook Quarry, and Dicer-
E.E. Arikareean) athere Quarry) (L.L. Arikareean).
A. Orella Member (inc. Everson Ranch, Warbonnet Creek, Refs: Hunt, 1985; Forsten, 1991; AMNH collections.
Meng's Ranch [=Harrison Ranch], Oreodon beds, B. Morava Ranch Quarry and Niobrara Canyon Local
Roundtop area, Hat Creek Basin [inc. Badland Creek, Fauna (L.L. Arikareean).
Prairie Dog Creek, Cedar Creek, Dry Creek], Dout Refs: Hunt, 1981, 1985; Coombs, 1984; Tedford et al.,
Ranch, Whitehead Creek, Brecht Ranch area, War- 1987.
bonnet Creek, Geike Ranch, Munson Ranch, Dead C. Marsland Local Fauna (E. Hemingfordian).
Horse Local Fauna [inc. Schomner Ranch (upper lev- Refs: Hunt, 1985.
els), Bartlett Ranch, White Ranch, Rook Ranch (upper CP105. Runningwater Formation: (E. Hemingfordian)
levels), Pasture 34, Hawthorn Island, Horse Island and Inc: Runningwater, Marsland, Hemingford, Dunlap Camel,
Central Turtle level], Rabbit Graveyard Local Fauna Apollo, Potter, Felton Dam, Woods Canyon, Aletomeryx
[inc. Bartlett south central, Rabbit Graveyard, Scratch- (=Antelope Creek), and Brown Quarries, Middle Cotton-
ing Post, Central Marker bed, Anderson's lumpy level, wood Creek, South of Antelope Valley, Warren Barnum
Bartlett's lumpy level, Midwest Valley], and Bartlett Ranch, Hovarka Locality, Cephalogale Ash Locality, Gre-
High Local Fauna [inc. Bartlett High, Bartlett Dike gory Ranch, Upper Loup Fork beds, Agate Area, and Whis-
Valley (highest level), and Anderson High]) (Orellan). tle Creek-Niobrara River Divide.
Refs: Gustafson, 1986b; Korth, 1989. Refs: Cook, 1965; Hunt, 1981; Tedford et al., 1987.
B. Whitney Member (inc. North Platte River drainage CP106. Running Water Formation Equivalent: (E. Heming-
[Lower and Upper Ash], Faunas I and II [below Non- fordian)
pariel Ash], Eagle Nest Butte, Ruby Ranch, Scottsbluff Inc: Bridgeport Quarries, Concretion Channel Prospect,
Monument, Roberts Draw, Roundhouse Rock, Crow Site 2 Prospect, Pepper Creek, and J.L. Ray Ranch.
Butte, Chimney Rock, and Weitzel Ranch) (Whitneyan). Refs: Hunt, 1981; Skinner and Johnson, 1984.
Refs: Tedford et al., 1985. CP107. Box Butte Formation: (L. Hemingfordian)
C. Lower and Upper Fauna III (=lower Sharps Forma- Red Valley Member: Box Butte Fauna (inc. Foley Quarry,
tion) (inc. Hat Creek-White River drainages, Brown Dry Creek Prospect A, Creek Prospect B, Dry Creek
Siltstone beds, Niobrara River drainage, and White Prospect D, Middle of the Road Quarry and Sand Canyon
River Valley) (E.E. Arikareean). Region).
Appendix I: Tertiary mammal localities 646

Refs: Galusha, 1975; Tedford et al., 1987. Refs: Skinner and Johnson, 1984; Tedford et al., 1987;
CP108. Sheep Creek Formation: (L. Hemingfordian) Voorhies, 1990a, 1990b; Voorhies et al., 1987.
A. Lower Sheep Creek Fauna (inc. Greenside and Long B. Crookston Bridge Member: Niobrara River Local
Quarries, Antelope Draw and Stonehouse Draw). Fauna (inc. Crookston Bridge [= Merycodus], Rail-
Refs: Skinner, Skinner, and Gooris, 1977. way A, Railway B, Devil's Jump Off, Nenzel, Rip-
B. Upper Sheep Creek Fauna (inc. Hilltop, Thistle, Target, ple, Runlofson, Sawyer, Schoettger, West Valentine,
Buck, Ashbrook, Vista, Rhino, Conference, Ravine Yale, Jamber and Hottell Ranch Horse Quarries, Haz-
and Thomson Quarries, Above Long Quarry, Above ard Homestead [= Driftwood Creek, and "Republican
Greenside Quarry, Stonehouse Draw, Aphelops Draw River faunas"], Annies Geese Crossing, Deer Fly Lo-
and Pliohippus Draw). cality, Myers Farm, Forked Hills of Hayden, Fairfield
Refs: Skinner, Skinner, and Gooris, 1977; Tedford Creek IV and Jones Canyon Site [inc. Loup Fork beds,
et al., 1987. and Merriman Bridge]) (L.L. Barstovian).
CP109. Hay·Springs Area: (L. Hemingfordian-L.L. Barsto- Refs: Skinner and Johnson, 1984; Tedford et al., 1987;
vian) Voorhies, l 990b.
A. Temporal equivalent of Sheep Creek Formation C. Devil's Gulch Member (inc. Elliot, Deep Creek, Devil's
(inc. Ginn Quarry, Antelope Valley) (L. Hemingfor- Gulch, Devil's Gulch Horse, Logan, Dutch Creek,
dian). Mizner Slide, Fairfield Falls, Fairfield Creek, Horse
Refs: Galusha, 1975. Thief Canyon, Rattlesnake Gulch, and Verdigree Quar-
B. Temporal equivalent of Valentine Formation (L.L. ries) (L.L. Barstovian).
Barstovian). Refs: Skinner and Johnson, 1984; Voorhies, 1990b.
CPI 10. Olcott Formation: (E. Barstovian) D. Burge Member: Burge and Penny Creek Faunas (inc.
Lower Snake Creek Fauna (inc. Camel, Snake, Trojan, June, Lucht, Burge, Gordon Creek, Buzzard Feather,
North Wall, East Wall, Douglas, Sinclair, East Sand, West Ewert, West June. South Lucht, Midway, Paleo Chan-
Sand, Sand, Jenkins, East Jenkins, West Jenkins, AMNH nel, 379, Quinn Mastodon, Swallow, Tetrabelodon,
1908, Mesoceras, East Surface, West Surface, Far Sur- Skull, White Point, White Face and Yale Mastodon
face, New Surface, Version, New Sand, Echo [=Antelope Quarries, and UCMP Joe. V-3312 [inc. "Ainsworth"
Draw], Humbug, South Humbug, Mill, Boulder, Grass near Burge and Forest Reservation and Williams
Roots and Prosynthetoceras Quarries, Princeton Locality Canyon]) (Barstovian/Clarendonian boundary).
lOOOc, Pocket 34, East Sinclair Draw, and West Sinclair Refs: Webb, 1969a; Skinner and Johnson, 1984;
Draw [plus Olcott equivalent, Hay Springs Area]). Voorhies, 1990b.
Refs: Skinner, Skinner, and Gooris, 1977; AMNH collec- CPl 15. Snake Creek Formation: (E. Clarendonian- L. Hem-
tions. phillian)
CPl 11. Sand Canyon beds: (E. Barstovian) A. Murphy Member: Unnamed Local Fauna (E. Claren-
Observation Quarry and Survey Quarry. donian).
Refs: Tedford et al., 1987; AMNH collections. Refs: Skinner, Skinner, and Gooris, 1977.
CP112. Unnamed formation: (E. Barstovian) B. Laucomer Member: Snake Creek Fauna (inc. Kilpa-
Lighthill Locality. trick Qarry, Olcott Quarry [=Hipparion Channel],
CPI 13. Ogallala Formation: (L.L. Barstovian-latest Hem- A.M.N.H. Quarry 7, and Hesperopithecus site[= Ol-
phillian) cott Hill]) (L. Clarendonian).
A. UNSM locality Wt-15A (Webster Co.) and UNSM lo- Refs: Skinner, Skinner, and Gooris, 1977.
cality Fr-20, (Franklin Co.) (L.L. Barstovian). C. Johnson Member: Aphelops Draw Fauna (inc. Aph-
B. UNSM locality Bn-103 (Banner Co.) (Clarendonian). elops Quarry and The Pits) (E.E. Hemphillian-L.E.
C. UNSM locality Sh-101 (Shannon Co.) (latest Hem- Hemphillian).
phillian). Refs: Skinner, Skinner, and Gooris, 1977; Tedford
CPI 14. Valentine Formation: (E.L. Barstovian-E. Clarendo- eta!., 1987.
nian). D. Johnson Member: (inc. ZX Bar Local Fauna and Plio-
A. Cornell Dam Member: Norden Bridge Quarry Local hippus Draw) (L. Hemphillian).
Fauna (inc. Egelhoff, Achilles, Kuhre, Norden Bridge Refs: Skinner, Skinner, and Gooris, 1977.
[=Tunnel Rock Locality], Penbrook, Rocky Ford, Lost CP116. Ash Hollow Formation: (E. Clarendonian-latest
Chance, Immense Journey, Zippy Lizard, Carrot Top Hemphillian)
and Hottell Ranch Rhino Quarries, Quarry without a A. Cap Rock Member: Minnechaduza Fauna: (inc. Lit-
name, Site Cr 121, Lost Duckling Site, Rossetta Stone tle Beaver B, Rock Ledge Mastodon, Poison Ivy,
Locality, Welke Locality, High in the Saddle Locality, Chokecherry, Big Beaver A and Wilson Quarries,
Small Falls Prospect, Miller Creek and Fairfield Creek McGinley's Stadium, Fairfield Creek no. 3, Horse
no. l) (L.L. Barstovian). Thief Canyon nos. 1and2, Johnson Rhino Site, Timm
Appendix I: Tertiary mammal localities 647

Ranch Site [=Prospect Locality 288], North Rim Lo- Mullen Local Fauna.
cality and Crazy Locality [inc. Box Butte Prospect]) CP121. Long Pine Formation: (L. Blancan)
(E. Clarendonian). Big Springs Local Fauna.
Refs: Webb, 1969a; Skinner and Johnson, 1984; Refs: Voorhies, 1987.
Tedford et al., 1987; Voorhies, 1990b.
AA. Ash Hollow Formation/Ogallala Formation: (inc.
KANSAS
Lessig Canyon [Brown Co.], Dutch Canyon [Brown
Co.], and Turtle Canyon [Sheridan Co.]) (Clarendo- CP123. Ogallala Formation: (L.L. Barstovian-latest
nian undifferentiated). Hemphillian)
B. Merritt Dam Member: (inc. Clayton, East Clayton, A. Wakeeney Fauna and North Fork of Solomon Creek
Egger, Emry, Hom, Hurlburt, Jonas Wilson, Platy- (=Sappa Creek) (E. Clarendonian).
belodon [UNSM Cr22], Pratt, Quinn Rhino, Rhino B. North Fork of Solomon Creek (L. Clarendonian).
Horizon 111, Wade, Blue Jay, Kepler, White Point, C. Jack Swayze Quarry, John Dakin Quarry, Long Island
Fat Chance, Serendipity, Precarious, Bear Creek, Quarry, Arens Quarry, and Sebastian Place
Gallup Gulch, Miller, and Mesinger Quarries, Jim (E.E. Hemphillian).
Lessig Camel Site, Eli Ash Pit, Shoreline Local Fauna, Refs: Hulbert, 1987b; Nelson and Madsen, 1987.
Xmas-Kat Channels Fauna [Balanced Rock, Xmas, D. Edson Quarry, Sherman Co., Lost Quarry, Found
Hans Johnson, Machaerodus, Leptarctus, Connec- Quarry, and Rhinoceras Hill Quarry (L. Hemphillian).
tion Kat, East Kat, Kat, West Line Kat, Quarter Line Refs: Frye, Leonard, and Swineford, 1956; Bennett,
Kat, Trailside Kat and West Kat Quarries] [inc. Broad- 1979; Harrison, 1983.
water and Harrisburg Localities]) (L. Clarendonian). E. Saw Rock Canyon Local Fauna, Seward Co. (latest
Refs: Skinner and Johnson, 1984; Leite, 1990; Hemphillian).
Voorhies, 1990b. Refs: Hibbard, 1949, 1952a, 1952b; Kurten and
C. Inc. Feltz Ranch, Lemoyne Local Fauna, and Ogal- Anderson, 1980.
lala Beach Local Fauna (E.E. Hemphillian). CP124. Unnamed formation: (L. Clarendonian)
Refs: Bown, 1980; Tedford et al., 1987; Leite, 1990; Ashland, Clark Co. Site.
Voorhies, l 990b. CP125. Unnamed formation; (E. Hemphillian)
D. Inc. Kimball Fauna, Cambridge Local Fauna Smith Co. Site and Trail Canyon, south fork of Driftwood
(=UNSM locality Ft.-40, Frontier Co.) Oshkosh Lo- Creek.
cal Fauna, Potter Quarry, Amebelodonfricki Quarry, CP126. Ash Hollow Formation: (L.E. Hemphillian)
and Greenwood Canyon (L.E. Hemphillian). Minium Quarry.
Refs: Tedford et al., 1987; Voorhies, 1990b. Refs: Thomasson et al., 1990.
E. Inc. Bear Tooth Local Fauna, Mailbox Local Fauna, CP127. Delmore Formation: (latest Hemphillian)
Honey Creek, Dalton Local Fauna, and Uptegrove Kinkerman's sand pit and Moundridge gravel pit.
Local Fauna (E.L. Hemphillian). Refs: Hibbard, 1952a.
Refs: Tedford et al., 1987; Voorhies, 1990b. CP128. Rexroad Formation: (E. Blancan)
F. Santee and Devil's Nest Airstrip Local Faunas (latest A. Fox Canyon Local Fauna.
Hemphillian). Refs: Hibbard, 1950; Dalquest, 1978; Kurten and
Refs: Voorhies, 1990b. Anderson, 1980.
CP116II. Unknown formation: (?Hemphillian) B. Keefe Canyon Local Fauna.
Sand Canyon. Refs: Hibbard, 1950.
CPl 17. Broadwater Formation: (E. Blancan) C. Rexroad Local Fauna and Wendell Fox Pasture, Meade
A. Lisco Local Fauna. Co.
Refs: Schultz and Stout, 1948; Kurten and Anderson, Refs: Hibbard, 194la, 194lb, 1950; Bjork, 1973;
1980; Voorhies and Comer, 1986. Dalquest, 1978; Kurten and Anderson, 1980.
B. Broadwater Local Fauna (L. Blancan). D. Bender Local Fauna.
Refs: Schultz and Stout, 1948; Hager, 1974; Kurten CP129. Unnamed formation: (E. Blancan)
and Anderson, 1980. Hart Draw Local Fauna.
CPl 18. Kiem Formation: (L. Blancan) CP130. Ballard Formation: (E. B!ancan)
Sand Draw Local Fauna (=Stegomastodon Quarry). A. Missler Member: (inc. Deer Park Local Fauna and
Refs: Hager, 1974; Kurten and Anderson, 1980; Voorhies Spring Creek Local Fauna).
and Comer, 1986. Refs: Hibbard, 1950; Skinner and Hibbard, 1972;
CP119. Unnamed formation: (L. Blancan) Kurten and Anderson, 1980.
Seneca Local Fauna. B. Sanders Local Fauna.
CP120. Unnamed formation: (L. Blancan) Refs: Hibbard, 1950.
 Appendix I: Tertiary mammal localities 648

CP 131. Belleville Formation (=Meade Formation): (L. Blan- B. Courtney Pit and Russell Pit (L. Hemphillian).
can) Refs: Bums and Young, 1988.
White Rock Local Fauna, Republic Co. (inc. Hibbard,
Sandpit, and South Localities) and Dixon Local Fauna
SASKATCHEWAN
(Belleville equivalent).
Refs: Hibbard, 1956; Eshelman, 1975; Kurten and Ander- NP6. Frenchman Formation: (Puercan 0)
son, 1980. Fr-1 site.
CP132. Crooked Creek Formation: (L. Blancan- Refs: Fox, 1990.
Irvingtonian) NP7. Ravenscrag Formation: (Puercan 0-Tiffanian 4)
A. Seger Local Fauna (L. Blancan). A. Medicine Hat Brick and Tile Quarry (Long Fall Hori-
Refs: Hager, 1974; Kurten and Anderson, 1980. zon) (Puercan 0).
B. Atwater Member (L. Blancan-E. Irvingtonian): Refs: Fox, 1990.
Borchers Local Fauna. B. RAV W-1 and Pine Cree Park (Puercan 2).
Refs: Hibbard, 1941c; Hager, 1974; Kurten and Refs: Russell, 1974; Archibald et al., 1987; Fox, 1990.
Anderson, 1980. C. Croc Pot (Puercan 2).
Refs: Fox, 1990.
D. Roche Percee (Tiffanian 4).
NORTHERN GREAT PLAINS Refs: Fox, 1990.
NP8. Swift Current Creek beds: (L. Uintan)
Swift Current Creek Local Fauna, Cypress Hills.
ALBERTA Refs: Russell, 1965a; Storer, 1984a.
NPl. Porcupine Hills Formation: (Puercan ?2-Tiffanian 1) NP9. Cypress Hills Formation, Lac Pelletier assemblages:
A. Alberta Corehole 66-1 (=Balzac West) (Puercan ?2). (Duchesnean-E. Chadronian)
Refs: Fox, 1990. A. LacPelletier Lower Fauna (Duchesnean).
B. Calgary 2E and Calgary 7E, Elbow River (Torre- Refs: Storer, 1988, 1990; Bryant, 1992.
jonian 3). B. LacPelletier Upper Fauna (Duchesnean).
Refs: Russell, 1929, 1958; Fox, 1990. Refs: Storer, 1993.
C. Cochrane I and II (Tiffanian 1). C. Simmie Local Fauna (latest Duchesnean/earliest
Refs: Russell, 1929; Fox, 1990. Chadronian).
NP2. Coldspur Formation: (Torrejonian 1-2?) Refs: Storer, 1993.
Diss Locality. D. Blumenort Local Fauna (?E. Chadronian).
Refs: Fox, 1990. Refs: Storer, 1993.
NP3. Paskapoo Formation: (Tiffanian 1-Tiffanian 4) NPlO. Cypress Hills Formation: (Duchesnean-L. Heming-
A. Saunders Creek Locality, Nordegg Aaron's Locality, fordian)
Hand Hills East, and Hand Hills West (lower level) A. Southfork (Cypress Hills Local Fauna) (L. Duches-
(Tiffanian 1). nean).
Refs: Fox, 1990. Refs: Storer, 1984b.
B. Hand Hills West (upper level) (Tiffanian 3). B. Cypress Hills Fauna (inc. Calf Creek Local Fauna
C. Blindman River localities: UADW-1, UADW-2, and Hunter Quarry) (E. Chadronian).
UADW-3, Mel's Place and Burbank (Tiffanian 3). Refs: Russell, 1972, 1975, 1978, 1980, 1982, 1985;
Refs: Archibald et al., 1987; Fox, 1990. Hunt, 1974; Storer, 1981, 1984a, Emry and Storer,
D. Erikson's Landing, Red Deer, and Joffres Bridge sites 1981; Kihm, 1987.
(Tiffanian 3). B2. Kealy Springs West Local Fauna (L. Chadronian).
Refs: Fox, 1990. Refs: Storer, 1994.
E. Canyon Ski Area, Crestomere School, and One-Jaw BB. Fossil Bush Local Fauna, Cypress Hills Plateau (Orel-
Gap (Tiffanian 4). lan).
Refs: Fox, 1990. Refs: Storer, 1993.
F. Swan Hills Site 1 (Tiffanian 4). C. Anxiety Butte (Whitneyan).
Refs: Russell, 1967; Fox, 1990. Refs: Storer and Bryant, 1993.
NP4. Ravenscrag Formation: (Tiffanian 3 or 4) CC. Kealy Springs Local Fauna, Eastend Area (E.
Police Point. Arikarrean ).
Refs: Krishtalka, 1973; Fox, 1990. Refs: Storer, 1993.
NP5. Hand Hills Formation: (L. Barstovian-Pleistocene) D. Anxiety Butte (Hemingfordian).
A. Sinclair Site (L. Barstovian/E. Clarendonian). Refs: Storer and Bryant, 1993.
Refs: Bums and Young, 1988. E. Topham Local Fauna (L. Hemingfordian).
 Appendix I: Tertiary mammal localities 649

Refs: Skwara, 1988. Hartman Quarry, and Azzara Quarry (Torrejonian 3).
NPl l. Wood Mountain Formation: (E.L. Barstovian) Refs: Simpson, 1937; Rose, 1981; Archibald et al.,
Wood Mountain Local Fauna (inc. Yost Farm, Kleinfolder 1987; Gingerich and Winkler, 1985; Sloan, 1987.
Farm, Quantock and Four Comers localities). NP19II. Melville Formation: (Torrejonian 3-Tiffanian 3)
Refs: Storer, 1975; Madden and Storer, 1985. A. Douglas Quarry (=Simpson's Locality 63) (Torrejo-
nian 3).
Refs: Krause and Gingerich, 1983; Krause and Maas,
BRITISH COLUMBIA
1990.
NP12. Unnamed formation: (Bridgerian) B. Bangtail Locality, Bangtail Plateau (Tiffanian 1).
Princeton. Refs: Gingerich, Houde, and Krause, 1983.
Refs: Krishtalka et al., 1987. C. Scarritt Quarry (=Simpson's Locality 56)(Tiffanian 1).
NP13. Kishenehn Formation: (Bridgerian to Duchesnean, Refs: Simpson, 1937; Krishtalka, 1973; Krause and
and Chadronian) Maas, 1990.
Flathead River Valley. D. Simpson's Localities 11 and 13 (=Melville Locality)
Refs: Russell, 1965b. (Tiffanian 3).
Refs: Archibald et al., 1987.
NP20. Tongue River Formation, Powder River Basin: (Tor-
NORTH WEST TERRITORIES
rejonian 3-Tiffanian 4)
NP14. Eureka Sound Formation: (L. Wasatchian-E. Bridge- A. Medicine Rocks 1, Mehling Site (Torrejonian 3).
rian) Ellesmere Island. Refs: Archibald et al., 1987.
Refs: West and Dawson, 1977; McKenna, 1980a. B. Newell's Nook (USGS D-2003) (Tiffanian 1).
Refs: Robinson and Honey, 1987.
C. White Site, 7-UP Butte, and Highway Blowout (Tiffa-
MONTANA
nian 2).
NP15. Hell Creek!fullock Formation: (Puercan 0) Refs: Archibald et al., 1987.
A. Bug Creek Anthills. D. Circle Locality (Tiffanian 3).
Refs: Archibald, 1982. Refs: Wolberg, 1979.
B. Bug Creek West and Scmenge Point. E. Olive Locality (Tiffanian 4).
Refs: Van Valen, 1978; Archibald, 1982. Refs: Wolberg, 1979.
C. Harbicht Hill, Chris's Bone Bed, and Ferguson Ranch. NP21. Sage Creek Formation, Sage Creek Basin: (Bridge-
Refs: Van Valen, 1978; Archibald, 1982. rian)
NP16. Tullock Formation: (Puercan 1-Torrejonian 1) A. Fields Draw localities (E. Bridgerian).
A. McKeever Ranch localities, Hell's Hollow Local Fauna B. Type Sage Creek Formation (M. or L. Bridgerian).
(including Worm Coulee 5) (Puercan 1). NP22. Dell beds, Sage Creek Basin: (L. Uintan)
Refs: Archibald, 1982. Dunlap Draw Local Fauna, Hough Draw Local Fauna, and
B. Garbani Quarry, Biscuit Butte, Biscuit Spring, and Yel- "Sage Creek beds."
low Sand Hill Localities (Puercan 3). Refs: Hough, 1955; Tabrum, 1990 (pers. comm.).
Refs: Archibald et al., 1987. NP23. Climbing Arrow Formation: (L. Uintan-E.Chadro-
C. Purgatory Hill (Puercan 3?). nian)
Refs: Van Valen, 1978; Archibald et al., 1987. A. Shoddy Springs Local Fauna, Mud Spring Gulch, Three
D. Horsethief Canyon and Mosquito Gulch Localities (Tor- Forks Basin (L. Uintan or E. Duchesnean).
rejonian 1). Refs: Black, 1967; Krishtalka et al., 1987.
Refs: Archibald et al., 1987. B. Black Butte and Lone Mountain, Bozeman Group (L.
NPl 7. Bear Formation, Crazy Mountain area: (Puercan 2) Duchesnean).
Simpson Quarry. C. Thompson Creek Fauna (Three Forks Basin) (in part)
Refs: Hartman et al., 1989. and Toston Fauna (Townsend Valley or Towstow area)
NP18. Ludlow Formation: (Puercan 3) (in part) (E. Chadronian).
Bechtold Site, Powder River Basin. Refs: Wilson, 197la, 197lb.
Refs: Archibald et al., 1987. NP24. Renova Formation, Climbing Arrow Member:
NP19. Lebo Formation, Crazy Mountain Area: (Torrejonian (L. Uintan-L. Chadronian)
1-3) A. Lower Conley Ranch Local Fauna and Burnt Hills Lo-
A. Simpson's Localities 9 and 78 (Torrejonian 1). cal Fauna, Lower Ruby River Basin (L. Uintan or E.
B. Simpson's Locality 25 and Lebo No. 2 (Torrejonian 2). Duchesnean).
C. Silberling Quarry (Simpson's Locality 1), Gidley B. Type locality of Macrotarsius montanus, North Boul-
Quarry (Simpson's Locality 4), Simpson's Locality 9, der Valley (L. Uintan or E. Duchesnean).
Appendix I: Tertiary mammal localities 650

C. Pipestone Springs Local Fauna, Jefferson River Basin C. Dog Town Mine Local Fauna, Canyon Ferry Reservoir,
(M. Chadronian). Douglass Creek Local Faunas (Douglass Creek Basin),
Refs: Tabrum and Fields, 1980; Garcia, 1990 (pers. and Lower Winston-Hadcock Ranch Local Fauna (M.
comm.). or L. Chadronian).
D. Little Pipestone Creek Local Faunas, Jefferson River Refs: White, 1954; Konizeski, 1961; Ostrander, 1985;
Basin (M-L. Chadronian). Prothero and Shubin, 1989.
Refs: Garcia, 1990 (pers. comm.). D. Townsend Valley sites (inc. Upper Winston-Hadcock
E. Miscellaneous Chadronian sites (precise age un- Ranch Local Faunas [Canyon Ferry Sites]) (L.
known) (inc. Bull Mountain [Jefferson River Basin], Chadronian-Orellan ).
Williams Creek North Local Fauna [Lower Ruby River Refs: White, 1954.
Basin], Upper Conley Ranch Local Fauna [Upper Ruby E. Canyon Ferry area (Orellan)
River Basin], Colbert Creek Local Fauna, Lower Madi- F. Black Butte Local Fauna, Gravelly Range (Orellan)
son River Valley, Sage Creek, and Beaverhead Rock NP29II. Unnamed formation: (E?. Chadronian)
Local Fauna [Beaverhead Basin]). Capitol Rock Fauna, Capitol Rock, Custer National Forest,
Refs: Garcia, 1990 (pers. comm.). Carter Co.
NP25. Renova Formation (undifferentiated), Beaverhead Refs: Mader and Alexander, 1995.
Basin: (Duchesnean-L. Chadronian) NP30. Dunbar Creek Formation: (M. Chadronian-Orellan)
A. Lower Mantle Ranch Local Fauna (E. Duchesnean). A. Thompson Creek Fauna (in part), Three Forks Basin
Refs: Tabrum, 1990 (pers. comm.). (M. Chadronian).
B. West McCarty's Mountain Local Fauna and Diamond- B. Dry Hollow Local Fauna (=Upper Toston Local
0 Ranch Local Fauna (L. Duchesnean). Fauna), Toston Area (Orellan).
Refs: Tabrum, 1990 (pers. comm.). NP3 l. Kishenehn Formation: (?M. Chadronian)
C. McCarty's Mountain Local Fauna (E. Chadronian). Paola Sliding Locality.
Refs: Tabrum, 1990 (pers. comm.). Refs: McKenna, 1990.
D. Upper Mantle Ranch Local Fauna (M. or L. Chadro- NP32. Cook Ranch Formation, Sage Creek Basin:
nian). (L. Chadronian-Orellan)
NP26. Renova Formation, Bone Basin Member: (Chadro- A. Little Spring Gulch Local Fauna (L. Chadronian).
nian) B. Matador Ranch Local Fauna and Cook Ranch Local
Jefferson River Basin. Fauna (Orellan).
NP27. Renova Formation, Dunbar Creek Member: (E. Refs: Tabmm, 1990 (pers. comm.).
Chadronian-Orellan) NP33. Blacktail Deer Creek Formation, Sage Creek Basin:
A. Monforton Ranch Local Fauna, North Boulder Valley (Whitneyan-E.L. Arikareean)
(E. Chadronian). A. White Hills Local Fauna (Whitneyan).
B. Mud Spring Gulch, Hossfeldt Hills (L. Chadro- B. Blacktail Deer Creek Local Fauna, Blacktail Deer Creek
nian). Basin (E.L. Arikareean).
C. Easter Lily Mine West Local Fauna, Jefferson River Refs: Hibbard and Keenman, 1950; Tedford et al.,
Basin (L. Chadronian). 1987.
Refs: Garcia, 1990 (pers. comm.). NP34. Fort Logan Formation (=Deep River beds):
D. Jefferson River Basin sites (inc. Palisade Cliff and (Chadronian-E. Barstovian)
Easter Lily Mine East Local Faunas) (Orellan). A. White Sulphur Springs (?Whitneyan).
Refs: Garcia, 1990 (pers. comm.). B. White Sulphur Springs (inc. Spring Creek 1, Crab Tree
NP28. Renova Formation, Passamari Member: (Whitneyan- Bluff, and Thompson Gulch North) (E.E. Arikareean).
L.L. Arikareean) Refs: Douglass, 1901.
A. Williams Creek South Local Fauna, Lower Ruby River C. "Deep River beds" (=Fort Logan beds), (L. Arika-
Basin (Whitneyan). reean).
B. Belmont Park Ranch Local Fauna and Chalk Buttes, Refs: Koerner, 1940; Tedford et al., 1987; UWA col-
Upper Ruby River Basin (L.L. Arikareean). lections.
NP29. Unmapped beds temporally equivalent to Renova For- D. Deep River Local Faunas (=Smith Creek Area, Spring
mation: (E. Chadronian-Orellan) Creek 2) (L. Hemingfordian).
A. Prickly Pear Creek Local Fauna (E. Chadronian). Refs: UWA collections.
Refs: Ostrander, 1985. E. Deep River Local Faunas (=Smith Creek Area)
B. Lower Thompson Creek Local Fauna (inc. North Boul- (E. Barstovian).
der River Valley and Three Forks Basin) (M. Chadro- NP35. Medicine Lodge beds, Horse Prairie-Grasshopper
nian). Creek Basin: (Chadronian?-L.L. Arikareean)
Appendix I: Tertiary mammal localities 651

A. Mill Point Local Fauna, Grasshopper Creek Basin B. Chalk Cliffs Local Fauna (Units 16-23), Paradise Val-
(E.E. Arikareean). ley (E.L. Barstovian).
B. Everson Creek Local Fauna, Horse Prairie Basin Refs: Barnosky and Labar, 1989.
(L.E. Arikareean). NP40II. Unknown formation: (? Hemingfordian)
C. Grasshopper Creek Local Fauna (L.L. Arikareean). Red Rock Creek.
NP36. Cabbage Patch beds (E.E. Arikareean-E.L. Arika- NP41. Madison Valley Formation, Three Forks Basin: (L.
reean) Hemingfordian-E. Clarendonian)
A. Lower Cabbage Patch Fauna (E.E. Arikareean). A. Middle part (L. Hemingfordian or E. Barstovian).
Refs: Rassmussen, 1989. B. Anceney Local Fauna (="Madison Valley Fauna")
B. Middle and Upper Cabbage Patch Faunas (inc. Bert (E.L. Barstovian).
Creek 2) (L.E. Arikareean). Refs: Dorr, 1956; Sutton, 1977.
Refs: Rich and Rassmussen, 1973; Rassmussen, 1989. C. Upper part (E. Clarendonian).
C. Pikes Peak Creek Local Fauna (L.E. or E.L. Arika- NP42. Flint Creek beds, Flint Creek Basin: (E. Barstovian)
reean) Flint Creek Local Fauna.
D. Tavenner Ranch Local Faunas (E.E.-L.E. or E.L. Refs: Tedford et al., 1987; Pierce and Rasmussen, 1989.
Arikareean ). NP43. Barnes Creek beds, Flint Creek Basin: (E.L. Barsto-
Refs: Rich and Rassmussen, 1973. vian)
NP37. Canyon Ferry beds: (L.E. Arikareean) Barnes Creek Local Fauna (=East of New Chicago, plus
Canyon Ferry Local Fauna, Townsend Valley. Bert Creek Local Fauna).
Refs: White, 1954; Leiggi, 1991. Refs: Tedford et al., 1987.
NP38. Sixmile Creek Formation: (L.E. Arikareean-E. NP45. Unnamed formation, Deer Lodge Basin: (E.E.
Hemphillian) Hemphillian)
A. Type Sixmile Creek Formation (Townsend Valley) and Deer Lodge Local Fauna, plus Dempsey Creek Gravel Pit
Roy Gulch-Sixteenmile Creek Drainage (Clarkstow and Drewsey Creek.
Basin) (L.E. Arikareean). Refs: Konizeki, 1957.
B. North Boulder Valley Fauna(= Nigger Hollow Local
Fauna) (L.L. Arikareean).
NORTH DAKOTA
Refs: Tedford et al., 1987.
C. Frank Ranch Local Fauna, Sant Ranch Local Fauna NP46. Bullion Creek Formation (=Fort Union Formation):
(Jefferson River Basin), McKanna Spring Local Fauna (Torrejonian) Billings Co.
(Ndrth Boulder Valley), and Sweetwater Creek Local NP47. Tongue River Formation, Williston Basin: (Torrejo-
Fauna (Upper Ruby River Basin) (L. Hemingfordian nian 3-Tiffanian 4)
or E. Barstovian). A. Donnybrook, Lloyd and Hares Site (=Heart Butte)
Refs: Tedford et al., 1987. (Torrejonian/Tiffanian).
D. Antelope Hills Local Fauna (Jefferson River Basin) (E. Refs: Archibald et al., 1987.
Barstovian). B. Judson and Brisbane (Tiffanian 3).
E. Williams Ranch Local Fauna (Upper Ruby River Basin) Refs: Holtzman, 1978.
(=Dorr and Wheeler Locality 5) (E?-E.L. Barstovian). C. Wannagan Creek Quarry and Riverdale (Tiffanian 4).
Refs: Dorr and Wheeler, 1964. Refs: Holtzman, 1978; Archibald et al., 1987;
F. Old Windmill Local Fauna(? Clarendonian). Erickson, 1991.
G. Ruby River Fauna (L. Clarendonian-E. Hemphillian). NP48. Sentinel Butte Formation: (Tiffanian 3?-5?)
H. Mayflower Mine Local Fauna, Old Wagon Road Local A. Type of Titanoides primaevus area near Fort Buford
Fauna (Jefferson River Basin), and North Silver Star (?Tiffanian 3).
Triangle Local Fauna (E. Hemphillian). Refs: Russell, 1974.
I. Inc. Tostow area sites, Townsend Valley (E. Hemphil- B. Red Spring Locality (?Tiffanian 4).
lian). Refs: Kihm, Hartman, and Krause, 1993.
NP39. Divide Basin: (L. Arikareean-E. Barstovian) NP49. Golden Valley Formation: (E. Wasatchian)
Divide Local Faunas, Cabbage Patch beds (E.L.- Camel Buttes Member, Williston Basin (inc. White Buttes
L.L. Arikareean). and Turtle Valley Localities).
NP40. Hepburn's Mesa Formation: (L. Hemingfordian-E.L. Refs: West, 1973b.
Barstovian) NP50. Brule Formation, White River Group: (Orellan-
A. Chalk Cliffs Local Fauna (Units 3-13), Paradise Valley Whitneyan)
(E. Barstovian). A. Dickinson Member, Unit 3 (L. Chadronian).
Refs: Barnosky and Labar, 1989. B. Dickinson Member, Units 4 and 5 (inc. White Butte,
 Appendix I: Tertiary mammal localities 652

Little Badlands Locality, Kostelecky Ranch, Fitterer D. Turtle Cove and Kimberly Members, Between Picture
Ranch, and Meduna Ranch, Stark Co., and 7 miles Gorge Ignimbrite and Deep Creek Tuft (="Lower John
south of South Heart) (Orellan). Day"?) (inc. Blue Basin, Sheep Rock area, and sites
C. Schefield Member, Unit 6, and Fitterer Ranch (Whit- UCMP 806, 833) (E.E.-L.E. Arikareean).
neyan). Refs: Merriam and Sinclair, 1907; Osborn, 1909; Lull,
NP51. Kildeer Formation: (Whitneyan-L. Hemingfordian) 1922; Macdonald, 1970; Fisher and Rensberger, 1972;
A. Whitneyan sites (inc. Rainey Butte). LACM collections.
B. Kildeer mountains (L. Hemingfordian). E. KimberlyMember,MeniscomysZone(="MiddleJohn
Day") (inc. sites UCMP 802, 858, 869, 904. V-4850,
?? V-6632) (L.E. Arikareean).
PACIFIC NORTHWEST Refs: Merriam and Sinclair, 1907; Lull, 1922; Fisher
and Rensberger, 1972; Prothero and Shubin, 1989.
WASHINGTON F. Kimberly Member, Entoptychus-Gregorymys Zone
(="Upper John Day") (E.L.-L.L. Arikareean).
PNI. Keechelus Formation: (E.E.-L.E. Arikareean) Refs: Merriam and Sinclair, 1907; Osborn, 1909; Lull,
A. Lower Wildcat Creek Local Fauna, Rimrock Lake area, 1922; Fisher and Rensberger, 1972.
Tieton River drainage (E.E. Arikareean). G. Haystack Valley Member, Mylagaulodon Zone
B. Upper Wildcat Creek Local Fauna, Rimrock Lake area, (="Uppermost John Day") (E. Hemingfordian).
Tieton River drainage (L.E. Arikareean). Refs: Merriam and Sinclair, 1907; Fisher and Rens-
PN2. Ellensburg Formation: (L. Clarendonian-E. Hemphil- berger, 1972; Tedford et al., 1987; LACM, UCB and
lian) UWA collections.
A. Granger Clay Pit, LACM 6431 (L. Clarendonian). H. Warm Springs Local Fauna and Shitike Creek (inc.
Refs: Martin and Tedrow, 1988. Mecca localities) (L. Arikarrean-E. Hemingfordian).
B. Yakima Canyon (E.E. Hemphillian). Refs: Woodburne and Robinson, 1977; Tedford et al.,
PN3. Ringold Formation: (latest Hemphillian-E. Blancan) 1987.
A. Lind Coulee Local Fauna (latest Hemphillian). PN7. Mascall Formation: (E. Barstovian)
B. River Road Local Fauna (latest Hemphillian). Mascall Fauna (inc. "John Day area," Picture Gorge as-
C. White Bluffs Local Fauna (inc. "Ringold fauna" and semblage, Crooked River assemblage, Cottonwood Creek,
?Bluff Top Local Fauna) (E. Blancan). Carbonaras Creek, and Gateway assemblage).
Refs: Kurten and Anderson, 1980; Lindsay, Opdyke, Refs: Downs, 1956; Tedford et al., 1987; LACM and UCB
and Johnson, 1984; Gustafson, 1978, 1985. collections.
PN4. Ringold Formation Equivalent: (L. Blancan) PN8. Butte Creek Volcanic Sandstone: (E.-E.L. Barstovian)
Inc. Taunton Local Fauna and Haymaker's Orchard Local A. Beatty Buttes (E. Barstovian).
Fauna Refs: Wallace, 1946.
Refs: Morgan and Morgan, 1995. B. Skull Springs Fauna and Red Basin Local Fauna (E.L.
Barstovian).
Refs: Shotwell, 1968; LACM collections.
OREGON
PN9. Sucker Creek Formation: (E.L.-L.L. Barstovian)
PN5. Clarno Formation: (Bridgerian-Duchesnean) A. Sucker Creek Fauna (E.L. Barstovian).
A. Clarno Nut Bed (Bridgerian). Refs: Scharf, 1935; LACM collections.
Refs: Krishtalka et al., 1987. B. Inc. Owyhee River drainage and Quartz Basin locali-
B. Hancock Quarry (Duchesnean). ties (="Deer Butte Formation") (L.L. Barstovian).
Refs: Krishtalka et al., 1987. Refs: Shotwell, 1968.
PN6. John Day Formation: (Chadronian-E. Hemingfordian) PNlO. JunturaFormation: (L.L. Barstovian-L. Clarendonian)
A. Big Basin Member, lower part (Chadronian). Black Butte Local Fauna (inc. Ironside sites) (L. Claren-
B. Big Basin Member, upper part (inc. Camp Creek, Sil- donian).
ver Wells) (Orellan). Refs: Shotwell, 1970.
Refs: Merriam and Sinclair, 1907 PNl I. Drewsey Formation: (E.E. Hemphillian)
C. Turtle Cove Member, Below Picture Gorge Ignimbrite Inc. Bartlett Mountain Local Fauna, Rome Local Fauna,
(="Lower John Day") (inc. John Day River Valley, Antelope Reservoir, Drinkwater Local Fauna, Warm
Camp Watson, Morgan's Place, Turtle Cove, and Springs, and Otis Basin Local Fauna.
Haystack Valley) (Whitneyan-E.E. Arikareean). Refs: Shotwell, 1956, 1970.
Refs: Merriam and Sinclair, 1907; Eaton, 1922; Lull, PN12. Rattlesnake formation: (E.E.-L.E. Hemphillian) ·
1922; Fisher and Rensburger, 1972. Rattlesnake Fauna, Cottonwood Creek (L.E. Hemphillian).
 Appendix I: Tertiary mammal localities 653

Refs: UCB collections. Hemingfordian-E.L. Barstovian)

PN13. Unnamed unit: (L.E. Hemphillian) A. Farmingdale Fauna (E. Hemingfordian).
Juniper Creek Canyon. Refs: Tedford and Hunter, 1984.
Refs: Shotwell, 1970. B. Shiloh Local Fauna (E.-E.L. Barstovian).
PN14. Shutler Formation: (L. Hemphillian) Refs: Tedford and Hunter, 1984.
Inc. McKay Reservoir Local Fauna, Krebs Ranch, South NC2. Lower Calvert Formation, Delaware: (E. Hemingfor-
of Arlington, and Ordinance (West End Blowout). dian)
Refs: Shotwell, 1956, 1958; LACM and UCB collections Pollack Pit Local Fauna, Cheswold Sand.
PN15. Chalk Butte Formation: (L. Hemphillian) Refs: Emry and Eshelman, 1997.
Inc. Little Valley and Christmas Valley Local Fauna. NC3. Calvert Formation, Maryland: (E.-E.L. Barstovian)
Refs: Shotwell, 1970; LACM collections. A. Plum Point Marl Member, Zone 10 (E. Barstovian).
Refs: Tedford and Hunter, 1984; Wright and Eshelman,
1987.
IDAHO
B. Plum Point Marl Member, Zones 13-15 (=Chesapeake
PN16. Peterson Creek beds, Salmon-Lehmi Basin: Bay Fauna, in part) (E.L. Barstovian).
(E.E.-L.E. Arikareean) Refs: Tedford and Hunter, 1984; Wright and Eshelman,
Peterson Creek Local Fauna (E.L. Arikareean). 1987.
Refs: Nichols, 1976. NC4. Choptank Formation, Maryland: (E.L. Barstovian)
PNI 7. Starlight Formation: (L.E. Arikareean) Zone 17 (=BG and E Power Plant and Lisby) (=Chesa-
Cedar Ridge, Rockland Valley. peake Bay Fauna, in part).
PNl 8. Mollie Gulch beds, Salmon-Lehmi Basin: (L. Heming- Refs: Tedford and Hunter, 1984; Wright and Eshelman,
fordian-E. Barstovian) 1987.
Mollie Gulch Local Fauna (inc. Boulder Creek). NC5. St. Mary's Formation, Maryland: (L.L. Barstovian-
PN19. Railroad Canyon beds, Medicine Lodge Basin: (L. L.E. Clarendonian)
Hemingfordian-E. Barstovian) Little Cove Point.
A. Railroad Canyon Local Fauna (=Maiden Creek Local Refs: Tedford and Hunter, 1984.
Fauna) (L. Hemingfordian). NC6. Yorktown Formation, Maryland: (L. Blancan)
B. Upper Railroad Canyon Local Fauna (E. Barstovian). Charles Co.
PN20. Salt Lake Group: (E. Barstovian)
Marsh Creek, near Downata Hot Springs.
Refs: MacFadden and Nelson, 1980.
PN21. Chalk Butte Formation: (L. Clarendonian-E. Hem- EAST COAST MARINE
phillian)
A. Chalk Basin East (L. Clarendonian). NORTHERN EAST COAST
B. Starr Valley (E. Hemphillian).
PN22. Unnamed formation: (E. Hemphillian) EM!. Shark River Marls, New Jersey: (Bridgerian)
Stroud Claim. Monmouth Co.
PN23. Glenns Ferry Formation: (E.-L. Blancan) Refs: Krishtalka et al., 1987.
A. Hagerman Local Fauna (inc. Horse Quarry) (E. Blan-
can).
Refs: Bjork, 1970; Kurten and Anderson, 1980; SOUTHERN EAST COAST
Lindsay, Opdyke, and Johnson, 1984; Morgan and EM2. Lisbon Formation, Mississippi: (Uintan/Duchesnean)
Morgan, 1995. Lutetian Marine series.
B. Sand Point and Flat Iron Butte (E. Blancan). Refs: Krishtalka et al., 1987.
Refs: Morgan and Morgan, 1995. EM3. Castle Hayne Formation, North Carolina: (M. Eocene)
C. Grand View Fauna (inc. Nine Foot Rapids Local Fauna, EM4. Ashley Formation, South Carolina: (L. Oligocene)
Castle Butte, Wild Horse Butte and Jackass Butte) (L. EMS. Chandler Bridge Formation, South Carolina: (L.
Blancan). Oligocene)
Refs: Shotwell, 1970; Kurten and Anderson, 1980; EM6. Charleston phosphate beds, South Carolina: (L.
Morgan and Morgan, 1995. Oligocene-Pleistocene)
A. (L. Oligocene/E. Miocene).
B. (Arikareean-Hemingfordian).
NORTHERN EAST COAST EM7. Pungo River Formation, North Carolina: (L. Heming-
fordian-E.L. Barstovian)
NCI. Kirkwood Formation, Monmouth Co., New Jersey: (E. Late Barstovian site.
 Appendix I: Tertiary mammal localities 654

EM8. Yorktown Formation: (latest Hemphillian-Blancan) CENTRAL AMERICA

A. North Carolina Site (latest Hemphillian).
EM12. Yellow Limestone Formation, Jamaica: (M. Eocene)
B. Virginia Site (latest Hemphillian-Blancan).
Freeman's Hall
EM13. San Sebastian Formation, Puerto Rico: (E?-M.
Oligocene)
GULF COAST
EM14. Juana Diaz Formation, Puerto Rico: (M.-L.
EM9. Avon Park Limestone, Florida: (M. Eocene) Oligocene)
EMIO. Inglis Formation, Florida: (M. Eocene) EM15. Guines Formation, Cuba: (Hemingfordian or Barsto-
EMl 1. Parachuchla Formation, Florida: (E. Arikareean) vian)
 Appendix II: References for locality listings

Akersten, W. A. ( 1972). Red Light Local Fauna (Blancan) of the Love For- Beard, K. C., & Tabrum, A. R. (1991). The first early Eocene mammal
mation, southeastern Huspeth County, Texas. Bulletin of the Texas from eastern North America: an omomyid primate from the Bashi
Memorial Museum, 20, 1-52. Formation, Lauderdale County, Mississippi. Mississippi Geology,
Anderson, D. W., & Picard, M. D. (1972). Stratigraphy of the Duchesne River 11, 1-6.
Formation (Eocene-Oligocene), northern Uinta Basin, northeastern Becker, J. J. (1985). Fossil herons (Aves: Ardeidae) oftheLateMioceneand
Utah. Bulletin of Utah Geological and Mineral Survey, 97, 1-29. Early Pliocene of Florida. Journal of Paleontology, 5, 24-31.
Archibald, J. D. (1982). A study of Mammalia and geology across the Bennett, D. K. (1979). The fossil fauna of Lost and Found Quarries
Cretaceous-Tertiary boundary in Garfield County, Montana. Uni- (Hemphillian, latest Miocene), Wallace County, Kansas. Occasional
versity of California, Publications in Geological Sciences, 122, Papers of the Museum of Natural History, University of Kansas
1-286. (Lawrence), 79, 1-24.
Archibald, J. D., Gingerich, P. D., Lindsay, E. H., Clemens, W. A., Krause, Berta, A., & Galiano, H. (1983). Meganteron hesperus from the late
D. W., & Rose, K. D. (1987). First North American Land Mammal Hemphillian of Florida with remarks on the phylogenetic relation-
Ages of the Cenozoic Era. In Cenozoic Mammals of North America, ships of machairodonts (Mammalia, Felidae, Machairodontinae).
ed. M. 0. Woodburne, pp. 24-76. Berkeley: University of California Journal of Paleontology, 57, 892-9.
Press. , ( 1984). A Miocene amphicyonid (Mammalia, Carnivora) from the Bone
Axelrod, D. I. (1956). Mio-Pliocene floras from West-Central Nevada. Uni- Valley Formation of Florida. Journal of Vertebrate Paleontology, 4,
versity ofCalifornia, Publications in Geological Sciences, 33, 1-322. 122-5.
Barnes, L. G., Howard, H., Hutchinson, J. H., & Welton, B. J. (1981). The Berta, A., & Morgan, G. S. (1985). A new sea otter (Carnivora: Mustelidae)
vertebrate fossils of the marine Cenozoic San Mateo Formation at from the late Miocene and early Pliocene (Hemphillian) of North
Oceanside, California. In Geologic Investigation of the San Diego America. Journal of Paleontology, 59, 809-19.
Coastal Plain, eds. P. L. Abbott& S. O'Dunn, pp. 53-70. San Diego: Bjork, P. R. (1967). Latest Eocene vertebrates from northwestern South
San Diego Association of Geologists. Dakota. Journal of Paleontology, 41, 227-36.
Barnosky, A. D. (1986). Arikareean, Hemingfordian, and Barstovian mam- (1970). The carnivores of the Hagerman Local Fauna (late Pliocene)
mals from the Miocene Colter Formation, Jackson Hole, Teton of southwestern Idaho. Transactions of the American Philosophical
County, Wyoming. Bulletin of the Carnegie Museum of Natural His- Society, 60, 1-54.
tory, 26, 1-69. (1973). Additional carnivores from the Rexroad Formation (upper
Barnosky, A. D., & Labar, W. J. (1989). Mid-Miocene (Barstovian) envi- Pliocene) of southwestern Kansas. Transactions of the Kansas
ronmental and tectonic setting near Yellowstone Park, Montana and Academy of Science, 76, 24-8.
Wyoming. Geological Society of America Bulletin, 101, 1448-56. Bjork, P.R., & Macdonald, J. R. (1981). Geology and paleontology of the
Baskin, J. A. (1979). Small mammals of the Hemphillian age White Cone Lo- Badland and Pine Ridge area, South Dakota. In Geology of the Black
cal Fauna, northeastern Arizona. Journal of Paleontology, 53, 695- Hills, South Dakota and Wyoming, ed. F. J. Rich, pp. 211-21. Rapid
708. City, S. Dak.: Geological Society of America Rocky Mountain Sec-
(1981). Barbourofelis (Nimravidae) and Nimravides (Felidae), with a tion Annual Meeting Fieldbook Guidebook.
description of two new species from the late Miocene of Florida. Black, C. C. (1967). Middle and late Eocene mammal communities: A major
Journal ofMammalogy, 62, 122-39. discrepancy. Science, 156, 62-4.
( 1982). Tertiary Procyonidae (Mammalia: Carnivora) of North America. (1978). Paleontology and geology of the Badwater Creek area, central
Journal of Vertebrate Paleontology, 2, 71-93. Wyoming, Part 14: The Artiodactyls. Annals of the Carnegie Mu-
( 1991 ). Early Pliocene horses from late Pleistocene flu vial deposits, Gulf seum, 47, 223-59.
Coastal Plain, South Texas. Journal of Paleontology, 65, 995-1006. Black, C. C., & Dawson, M. R. (1966). Paleontology and geology of the
Beard, K. C., Dawson, M. R., & Tabrum, A. R. (1995). First diverse Bad water Creek area, central Wyoming, Part 1: History of field work
land mammal fauna from the early Cenozoic of the southeastern and geological setting. Annals ofthe Carnegie Museum, 38, 297-307.
United States: the early Wasatchian Red Hot Local Fauna, Lau- Bode, F. C. (1935). The fauna of the Merychippus Zone, North Coalinga
derdale County, Mississippi. Journal of Vertebrate Paleontology, 19 district, California. Carnegie Institute of Washington Publications,
(suppl. to no. 3), p. l 8A. 453, 66-96.

655
 Appendix II: References for locality listings 656

Bown, T. M. (1979). Geology and mammalian paleontology of the Sand Czaplewski, N. J. (1987). Middle Blancan vertebrate assemblage from the
Creek facies, lower Will wood Formation (Lower Eocene), Washakie Verdi Formation, Arizona. Contributions to Geology, University of
County, Wyoming. Geological Survey of Wyoming, Memoir no. 2, Wyoming, 25, 133-55.
1-151. Dalquest, W.W. (1975). Vertebrate fossils from the Blanco Local Fauna of
(1980). The Willwood Formation (Lower Eocene) of the southern Texas. Occasional Papers of the Museum of Texas Tech University,
Bighorn basin, Wyoming, and its mammalian fauna. In Early 30, 1-52.
Cenozoic Paleontology and Stratigraphy of the Bighorn Basin, (1978). Early Blancan mammals of the Beck Ranch Local Fauna of
W)>oming, ed. P. D. Gingerich. University of Michigan Papers in Texas. Journal ofMammalogy, 59, 269-98.
Paleontology, 24, 127-36. (1983). Mammals of the Coffee Ranch Local Fauna, Hemphillian of
(1982). Geology, paleontology, and correlation of Eocene volcanistic Texas. The Pearce-Sellards Series, Texas Memorial Museum, 38,
rocks, southeast Absaroka Range, Hot Springs County, Wyoming. 1-41.
United States Geological Survey, Professional Paper, 1201-A, Dalquest, W.W., & Hughes, J. T. ( 1966). A new mammalian local fauna from
Al-A75. the lower Pliocene of Texas. Transactions of the Kansas Academy of
Bown, T. M., & Kihm, A. J. (1981). Xenicohippus, an unusual new Hyra- Sciences, 69, 79-87.
cotherine (Mammalia, Perissodactyla) from lower Eocene rocks of Dalquest, W.W., & Mooser, 0. (1974). Miocene vertebrates from Aguas-
Wyoming, Colorado, and New Mexico. Journal of Paleontology, 55, calientes, Mexico. Pearce-Sellards Series, Texas Memorial Museum,
257-70. 21, 1-10.
Bown, T. M., Rose, K. D., Simons, E. L., & Wing, S. L. (1993). Distribution (1980). The Late Hemphillian mammals of the Ocote Local Fauna, Gua-
and stratigraphic correlation of upper Paleocene and lower Eocene najualo, Mexico. Pearce-Sellards Series, Texas Memorial Museum,
fossil mammals and plant localities of the Fort Union, Willwood 32, 1-25.
and Tatman Formations, southern Bighorn basin, Wyoming. United Damuth, J. (1982). Analysis of the preservation of community structure in
States Geological Survey Professional Paper, 1540, 1-269. assemblages of fossil mammals. Paleobiology, 8, 434-46.
Breyer, J. A. (1985). Vertebrate biostratigraphy of the Late Miocene rocks, Davidson, J. R. (1987). Geology and mammalian paleontology of the Wind
western Nebraska. National Geographic Society Research Reports, River Formation, Laramie basin, southeastern Wyoming. Contribu-
18, 177-88. (General stratigraphic reference.) tions to Geology, University of W)>oming, 25, 103-32.
Bryant, D. (1991). New early Barstovian (middle Miocene) vertebrates from Dawson, M. R. (1980). Geology and paleontology of the Badwater Creek
the upper Torreya Formation, eastern Florida panhandle. Journal of area, central Wyoming, Part 20: The Late Eocene Creodonta and
Vertebrate Paleontology, 11, 472-89. Camivora. Annals of the Carnegie Museum, 49, 79-91.
Bryant, H. D. ( 1992). The Carnivora of the Lac Pelletier lower fauna (Eocene: Delson, E. (1971). Fossil mammals of the Early Wasatchian Powder Basin
Duchesnean), Cypress Hills Formation, Saskatchewan. Journal of Local Fauna, Eocene of northeast Wyoming. Bulletin ofthe American
Vertebrate Paleontology, 66, 847-55. Museum of Natural History, 146, 309-64.
Burns, J. A., & Young, R. R. (1988). Stratigraphy and palaeontology of Demere, T. A. (1988). Early Arikareean (late Oligocene) vertebrate fos-
the Hand Hills Region. Occasional Papers of the Tyrill Museum of sils and biostratigraphic correlations of the Otay Formation at East-
Paleontology (Society of Vertebrate Paleontology 48th. AGM Field lake, San Diego County, California. In Paleogene Stratigraphy, West
Trip), 1-13. Coast ofNorth America, Pacific Section, eds. M. V. Filewicz & R. L.
Butler, R. F., Krause, D. W., & Gingerich, P. D. (1987). Magnetic polarity Squires, Society of Economic Paleontologists & Mineralogists, West
and biostratigraphy of middle-late Paleocene continental deposits of Coast Paleogene Symposium, 58, 35-43.
south-central Montana. Journal of Geology, 95, 647-57. Domning, D. P. (1989). Fossil Sirenia of the West Atlantic and Caribbean
Buwalda, J. P. (1916). New mammalian faunas from Miocene sediments region, III: Xenosiren yucateca gen. et sp. nov. Journal of Vertebrate
near Tehachapi Pass in the southern Sierra Nevada. University of Paleontology, 9, 429-37.
California Publications Bulletin of the Department of Geological (1990). Fossil Sirenia of the West Atlantic and Caribbean region, IV:
Sciences, 10, 75-85. Corystosiren varguezi, gen. et sp. nov. Journal of Vertebrate Paleon-
Carranza-Casteneda, 0., & Miller, W. E. (1993). Hemphillian and Blancan tology, 9, 429-37.
equids from Hildago, Mexico. Journal of Vertebrate Paleontology, Dorr, J. A., Jr. (1956). Anceny local mammal fauna, latest Miocene, Madison
13 (suppl. to no. 3), p. 29A. Valley Formation, Montana. Journal of Paleontology, 30, 62-74.
Cassiliano, M. (1980). Stratigraphy and vertebrate paleontology of the Horse Dorr, J. A. (1978). Revised and amended fossil vertebrate fauna! lists, early
Creek - Trail Creek area, Laramie county, Wyoming. Contributions Tertiary, Hoback basin, Wyoming. Contributions to Geology, Uni-
to Geology, University ofW)>oming, 19, 25-68. versity of W)>oming, 16, 79-84.
Clark, J. (1937). The stratigraphy and paleontology of the Chadron Forma- Dorr, J. A., & Gingerich, P. D. (1980). Early Cenozoic mammalian paleon-
tion in the Big Badlands of South Dakota. Annals of the Carnegie tology, geologic structure, and tectonic history in the overthrust belt
Museum, 25, 261-350. near LaBarge, western Wyoming. Contributions to Geology, Univer-
Clark, J.B., Beerbower, J. R., & Kietzke, K. K. (1967). Oligocene sedimen- sity ofW)>oming, 18, 101-15.
tation, stratigraphy, paleoecology and paleoclimatology in the Big Dorr, J. A., Jr., & Wheeler, W. H. (1964). Cenozoic paleontology, stratig-
Badlands of South Dakota. Fieldiana, Geology Memoirs, 5, 1-158. raphy, and reconnaissance geology of the Upper Ruby River Basin,
Clark, J.B., Dawson, M. R., & Wood, A. E. (1964). Fossil mammals from southwestern Montana. Contributions of the Museum of Paleontol-
the lower Pliocene of the Fish Lake Valley, Nevada. Bulletin of the ogy, University of Michigan, 13, 297-339.
Museum of Comparative Zoology, 131, 27-63. Douglass, E. ( 1901 ). Fossil Mammalia of the White River Beds of Montana.
Clark, J.B., & Guensburg, T. E. (1972). Arctoid genetic characters as related Transactions of the American Philosophical Society, 20, 237-79.
to the genus Parictis. Fieldiana (Geology), 26, 1-73. Downs, T. (1956). The Mascall fauna from the Miocene of Oregon. Univer-
Cook, H.J. (1922). A Pliocene fauna from Yuma County, Colorado, with sity of California Publications in Geological Sciences, 31, 199-354.
notes on the closely related Snake Creek beds from Nebraska. Pro- Downs, T., & White, J. A. (1968). A vertebrate fauna! succession in super-
ceedings of the Colorado Museum of Natural History, 4, 1-29. posed sediments from late Pliocene to middle Pleistocene in Califor-
(1965). Runningwater Formation, middle Miocene of Nebraska. Amer- nia. XXIII International Geological Congress, 10, 41-7.
ican Museum Novitates, 2227, 1-8. Drescher, A. B. (1942). Later Tertiary Equidae from the Tejon Hills, Cali-
Coombs, M. C. (1984). Excavation of a late Arikareean vertebrate assem- fornia. In Studies of Cenozoic Vertebrates of Western North America
blage in northwest Nebraska. National Geographic Research Re- and of Fossil Primates, eds. A. B. Drescher, E. L. Furlong, I. S.
ports, 16, 145-52. Demay, P. C. Henshaw, R. W. Wilson, H. Howard & G. H. R. Von
 Appendix II: References for locality listings 657

Koenigswald. Carnegie Institute of Washington Publication, 530, Rocky Mountain Section - Society of Economic Paleontologists and
1-23. Mineralogists. (General stratigraphic reference.)
Eaton, G. F. (1922). John Day Felidae in the Marsh Collection. American Fisher, R. V., & Rensberger, J.M. (1972). Physical stratigraphy of the John
Journal of Science, 5, 425-52. Day Formation, central Oregon. University of California Publica-
Eaton, J. G. (1982). Paleontology and correlation of Eocene volcanic rocks tions in Geological Sciences, 101, 1-45.
in the Carter Mountain area, Park County, southeastern Absaroka Forsten, A. (1970). The late Miocene Trail Creek mammalian fauna. Con-
range, Wyoming. Contributions to Geology, University of W)>oming, tributions to Geology, University ofW)>oming, 9, 39-51.
21, 153-94. (1975). The fossil horses of the Texas Gulf Coastal Plain: a revision.
(1985). Paleontology and correlation of the Eocene Tepee Trail and Wig- Pearce Sellards Series, Texas Memorial Museum, 22, 1-86.
gins Formations in the North Fork of Owl Creek area, southeastern (1991). Size trends in Holarctic anchitherines (Mammalia, Equidae).
Absaroka range, Hot Springs County, Wyoming. Journal of Verte- Journal of Paleontology, 65, 147-59.
brate Paleontology, 5, 345-70. Fox, R. C. (1990). The succession of Paleocene mammals in western Canada.
Emry, R. J. (1975). Revised Tertiary stratigraphy and paleontology of the Geological Society of America Special Paper, 243, 51-70.
western Beaver Divide, Fremont County, Wyoming. Smithsonian Frailey, D. ( 1978). An early Miocene (Arikareean) fauna from North Central
Contributions to Paieontology, 25, 1-20. Florida (the SB-IA Local Fauna). Occasional Papers of the Museum
(1978). A new hypertragulid (Mammalia, Ruminantia) from the early of Natural History, University of Kansas (Lawrence), 75, 1-20.
Chadronian of Wyoming and Texas. Journal of Paleontology, 52, (1979). The large mammals of the Buda Local Fauna (Arikareean,
1004-14. Alachua County, Florida). Bulletin of the Florida State Museum,
(1979). Review of Toxotherium (Perissodactyla, Rhinoceratoidea) with Biological Sciences, 24, 123-73.
new material from the early Oligocene of Wyoming. Proceedings of Frick, C. ( 1921 ). Extinct vertebrate faunas of the Badlands of Bautista Creek
the Biological Society of Washington, 92, 28-41. and San Timoteo Canyon, southern California. University of Califor-
(1981). Additions to the mammalian faunas of the type Duchesnean, nia Publications, Department of Geological Sciences, 12, 277-424.
with comments on the status of the "Duchesnean" age. Journal of (1937). Homed ruminants of North America. Bulletin of the American
Paleontology, 55, 563-76. Museum of Natural History, 69, 1-669. (General taxonomic refer-
( 1990). Mammals of the Bridgerian (middle Eocene) Elderberry Canyon ence.)
Local Fauna of eastern Nevada. Geological Society of America Spe- Frye, J. C., Leonard, A. B., & Swineford, A. (1956). Stratigraphy of the
cial Paper, 243, 187-209. Ogallala Formation (Neogene) of northern Kansas. State Geological
(1992). Mammalian range zones in the Chadronian White River Forma- Survey of Kansas Bulletin, 118, 1-92.
tion at Flagstaff Rim, Wyoming. In Eocene-Oligocene Climatic and Furlong, E. L. (1941). A new Pliocene antelope from Mexico with remarks
Biotic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 106-15. on some known antilocaprids. In Studies of Cenozoic Vertebrates of
Princeton: Princeton University Press. Western North America and of Fossil Primates, eds. A. B. Drescher,
Emry, R. J., & Eshelman, R. E. (1997). The Pollack Pit Local Fauna E. L. Furlong, I. S. Demay, P. C. Henshaw, R. W. Wilson, H. Howard
(early Hemingfordian, early Miocene): first Tertiary land mammals & G. H. R. Von Koenigswald. Carnegie Institute of Washington Pub-
from Delaware. Delaware Geological Survey Publications (in lication, 530, 25-33.
press). Galbreath, E. C. (1953). A contribution to the Tertiary geology and paleon-
Emry, R. J., Russell, L. S., & Bjork, P.R. (1987). The Chadronian, Orellan, tology of northeastern Colorado. Paleontological Contributions, 4,
and Whitneyan North American Land Mammal Ages. In Cenozoic 1-120.
Mammals of North America, ed. M. 0. Woodburne, pp. 118-52. Galusha, T. (1975). Stratigraphy of the Box Butte Formation, Nebraska.
Berkeley: University of California Press. Bulletin of the American Museum of Natural History, 56, 1-68.
Emry, R. J., & Storer, J. E. (1981). The hornless protoceratid Pseu- Galusha, T., & Blick, J.C. (1971). Stratigraphy of the Santa Fe Group, New
doprotoceras (Tylopoda: Artiodactyla) in the early Oligocene of Mexico. Bulletin of the American Museum of Natural History, 144,
Saskatchewan and Wyoming. Journal of Vertebrate Paleontology, 1-127.
1, 101-10. Gazin, C. L. (1956a). Paleocene mammalian faunas of the Bison Basin
Erickson, B. R. (1991). Flora and fauna of the Wannagan Creek Quarry: late in south-central Wyoming. Smithsonian Miscellaneous Collections,
Paleocene of North America. Scientific Publications of the Science 131, 1-57.
Museum of Minnesota, New Series, 7, 5-19. (l 956b ). The occurrence of fossil mammalian remains in the Fossil Basin
Eshelman, R. E. (1975). Geology and paleontology of the early Pleistocene of southwestern Wyoming. Journal of Paleontology, 30, 707-11.
(late Blancan) White Rock Fauna from north central Kansas. Claude ( 1962). A further study of the lower Eocene mammalian faunas of south-
Hibbard Memorial, Vol. 4; Papers on Paleontology, Museum of Pa- western Wyoming. Smithsonian Miscellaneous Collections, 144,
leontology, University of Michigan, 13, 1-60. 1-98.
Eshelman, R. E., & Whitmore, R. C., Jr. (1997). Pliocene (late Hemphillian) (1969). A new occurrence of Paleocene mammals in the Evanston For-
land mammals from the Lee Creek Mine, Aurora, North Carolina. mation, southwestern Wyoming. Smithsonian Contributions to Pa-
In Geology and Paleontology of the Lee Creek Mine, North Car- leontology, 2, 1-16.
olina, Vol. 3, ed. C. F. Ray. Smithsonian Institution Contributions to (1971). Paleocene primates from the Shotgun Member of the Fort Union
Paleobiology (in press). Formation in the Wind River Basin, Wyoming. Proceedings of the
Ferrusquia-Villafranca, I. (1984). A review of the early and middle Tertiary Biological Society of Washington, 84, 13-37.
faunas of Mexico. Journal of Vertebrate Paleontology, 4, 187-98. (1976). Mammalian fauna! zones of the Bridger middle Eocene. Smith-
(1990). Biostratigraphy of the Mexican continental Miocene. Paleonto- sonian Contributions to Paleontology, 26, 1-25.
logica Mexicana, 56, 1-149. Gingerich, P. D. (1979). Lambertocyon eximus, a new arctocyonid (Mam-
(1993). Contributions to the knowledge of Mexico's Oligocene mam- malia, Condylarthra) from the late Paleocene of western North
mals: additions and revisions of the Chadronian Rancho Gaitan local America. Journal of Paleontology, 53, 524-29.
fauna, northeastern Chihuahua. Journal of Vertebrate Paleontology, (1983). New Adapisoricidae, Pentacodontidae, and Hyposodontidae
3 (suppl. to no. 3), 34A-35A. (Mammalia, Insectivora and Condylarthra), from the late Paleocene
Fields, R. W., Tabrum, A. R., Rasmussen, D. L., & Nichols, R. (1985). Ceno- of Wyoming and Colorado. Contributions from the Museum of Pale-
zoic rocks of the intermontane basin of western Montana and eastern ontology, University of Michigan, 26, 197-225.
Idaho: A summary. In Cenozoic Paleogeography of West Central ( 1987). Early Eocene bats (Mammalia, Chiroptera) and other vertebrates
United States, eds. R. M. Flores & S. S. Kaplan, pp. 9-36. Denver: in freshwater limestones of the Willwood Formation, Clark's Fork
 Appendix II: References for locality listings 658

Basin, Wyoming. Contributions from the Museum of Paleontology, Harksen, J.C., & Green, M. (1971). Thin Elk Formation, lower Pliocene,
University of Michigan, 27, 275-320. South Dakota. South Dakota Geological Survey Report of Investiga-
( 1989). New earliest Wasatchian mammalian fauna from the Eocene of tion, 100, 1-7.
northwestern Wyoming: composition and diversity in a rarely sam- Harksen, J.C., & Macdonald, J. R. (1967). Miocene Batesland Formation
pled high-floodplain assemblage. University of Michigan Papers in named in Southwestern South Dakota. South Dakota Geological Sur-
Paleontology, 28, 1-97. vey, Reports of Investigation, 96, 1-10.
Gingerich, P. D., Houde, P., & Krause, D. W. (1983). A new earliest Tiffa- Harris, A.H. (1993). Quaternary vertebrates of New Mexico. In Vertebrate
nian (late Paleocene) mammalian fauna fromBangtail Plateau, west- Paleontology in New Mexico, eds. S. G. Lucas & J. Zikek. Bulletin
ern Crazy Mountain Basin, Montana. Journal of Paleontology, 57, of the New Mexico Museum of Natural History and Sciences, 2,
957-70. 179-97.
Gingerich, P. D., Rose, K. D., & Krause, D. W. (1980). Early Cenozoic Harrison, J. A. (1978). Mammals of the Wolf Ranch Local Fauna, Pliocene
mammalian fauna of the Clark's Fork Basin - Polecat Bench Area, of the San Pedro Valley, Arizona. Occasional Papers of the Museum
northwestern Wyoming. In Early Cenozaic Paleontology and Stratig- of Natural History, University of Kansas, Lawrence, 73, 1-18.
raphy ofthe Bighorn Basin, Wyoming, ed. P. D. Gingerich. University (1983). The carnivores of the Edson Local Fauna (late Hemphillian),
of Michigan Papers in Paleontology, 24, 51-68. Kansas. Smithsonian Contributions to Paleontology, 54, 1-42.
Gingerich, P. D., & Winkler, D. A. (1985). Systematics of Paleocene Viver- Harshman, E. N. (1972). Geology and uranium deposits, Shirley Basin area,
ravidae (Mammalia, Carnivora) in the Bighorn Basin and Clark's Wyoming. United States Geological Survey Professional Paper, 745,
Fork Basin, Wyoming. Contributions from the Museum of Paleon- 1-82.
tology, University of Michigan, 27, 87-128. Hartman, J.E. ( 1986). Paleontology and biostratigraphy oflower part Polecat
Golz, D. J. (1979). Eocene Artiodactyla of southern California. Natural Bench Formation, southern Bighorn Basin, Wyoming. Contributions
History Museum of Los Angeles County Science Bulletin, 26, 1-85. to Geology, University of Wyoming, 24, 11-63.
Golz, D. J., & Lillegraven, J. A. (1977). Survey of known occurrences of Hartman, J. H., Buckley, G. A., Krause, D. W., & Kroeger, T. J. (1989).
terrestrial vertebrates from Eocene strata of southern California. Con- Paleontology, stratigraphy, and sedimentology of Simpson Quarry
tributions to Geology, University of Wyoming, 15, 43-64. (early Paleocene), Crazy Mountains Basin, south central Montana.
Granger, W. ( 1909 ). Fauna! horizons of the Washakie Formation of southern In 1989 Field Conference Guidebook: Montana Centennial Edition,
Wyoming. Bulletin of the American Museum ofNatural History, 26, eds. D. E. French & R. F. Grabb. Geologic Resources of Montana,
13-23. I, 173-85.
Green, M. (1985). Micromammals from the Miocene Bijou Hills Local Hay, 0. P. (1921 ). Description of species of Pleistocene Vertebrata, types and
Fauna. Dakoterra, 2, 141-54. specimens of most of which are preserved in the United States Na-
Green, M., & Holman, J. A. (1977). A late Tertiary stream-channel deposit tional Museum. Proceedings of the United States National Museum,
from South Bijou Hills, South Dakota. Journal of Paleontology, 51, 59, 599-642.
543-7. Henshaw, P. C. (1939). A Tertiary mammalian fauna from the Avawatz
Gregory, J. T. (1942). Pliocene vertebrates from Big Spring Canyon, South Mountains, San Bernadino County, California. In Studies of Ceno-
Dakota. University of California Publications, Bulletin of the De- zoic Vertebrates and Stratigraphy of Western North America., eds. P.
partment of Geological Sciences, 26, 307-446. C. Henshaw, R. W. Wilson, H. Howard, A.H. Miller, J. F. Dougherty,
Gunnell, G. F., & Bartels, W. S. (1994). Early Bridgerian (middle Eocene) & R. H. Jahns. Carnegie Institute of Washington Publication, 514,
vertebrate paleontology and paleoecology of the southern Green 1-30.
River Basin, Wyoming. Contributions to Geology, University of (1942). A Tertiary mammalian fauna from the San Antonio mountains
Wyoming, 30, 57-70. near Tonopah, Nevada. In Studies of Cenozaic Vertebrates of West-
Gunnell, G. F., Bartels, W. S., Gingerich, P. D., & Torres, V. (1992). The ern North America and of Fossil Primates, eds. A. B. Drescher, E. L.
Wapiti valley faunas: early and middle Eocene fossil vertebrates from Furlong, I. S. Demay, P. C. Henshaw, R. W. Wilson, H. Howard &
the North Fork of Shoshone River, Park County, Wyoming. Contri- G. H. R. Von Koenigswald. Carnegie Institute of Washington Publi-
butions from the Museum of Paleontology, University of Michigan, cation, 530, 77-168.
28, 247-87. Hesse, C. J. (1943). A preliminary report on the Miocene vertebrate fauna
Gustafson, E. P. ( 1978 ). The vertebrate fauna of the Pliocene Ringold Forma- of southeast Texas. Transactions of the Texas Academy of Sciences,
tion, south-central Washington. Bulletin of the Museum of Natural Proceedings, 26, 157-79.
History, University of Oregon, 23, 1-62. Hibbard, C. W. (194la). Mammals of the Rexroad fauna from the up-
(1985). Soricids (Mammalia, Insectivora) from theBlufftop Local Fauna, per Pliocene of southwestern Kansas. Transactions of the Kansas
Blancan Ringold Formation of central Washington, and the correla- Academy of Science, 44, 265-313.
tion of Ringold Formation Faunas. Journal of Vertebrate Paleontol- (1941 b ). New mammals of the Rexroad fauna, upper Pliocene of Kansas.
ogy, 5, 88-92. American Midland Naturalist, 26, 357--68.
( l 986a). Carnivorous mammals of the late Eocene and early Oligocene (194lc). The Borchers fauna, a new Pleistocene interglacial fauna from
of Trans-Pecos, Texas. Bulletin of the Texas Memorial Museum, 33, Meade County, Kansas. State Geological Survey of Kansas Bulletin,
1-66. 38, 197-220.
(1986b). Preliminary biostratigraphy of the White River Group (1949). Pliocene Saw Rock Canyon fauna in Kansas. Contributions from
(Oligocene, Chadron and Brule Formation) in the vicinity of the Museum of Paleontology, University of Michigan, 7, 91-105.
Chadron, Nebraska. Transactions of the Nebraska Academy of Sci- (1950). Mammals of the Rexroad Formation from Fox Canyon, Kansas.
ences, 14, 7-19. Contributions from the Museum ofPaleontology, University ofMichi-
Guthrie, D. A. (1967). The mammalian fauna of the Lysite Member, Wind gan, 8, 113-92.
River Formation (early Eocene) of Wyoming. Memoirs of the South- ( l 952a). Vertebrate fossils from late Cenozoic deposits of central Kansas.
ern California Academy of Sciences, 5, 1-53. University of Kansas Paleontological Contributions (Vertebrata), 2,
(1971). The mammalian fauna of the Lost Cabin Member, Wind River 1-14.
Formation (Lower Eocene) of Wyoming. Annals of the Carnegie ( l 952b ). The Saw Rock Canyon fauna and its stratigraphic significance.
Museum, 43, 47-113. Papers of the Michigan Academy of Science, Arts and Letters, 38,
Hager, M. W. (1974). Late Pliocene and Pleistocene history of the Don- 387-411.
nelly Ranch vertebrate site, southeastern Colorado. Contributions to (1954). A new Pliocene vertebrate fauna from Oklahoma. Papers of the
Geology, University of Wyoming, Special Paper 2, 2-62. Michigan Academy of Science, Arts and Letters, 34, 339-59.
 Appendix II: References for locality listings 659

(1956). Vertebrate fossils from the Meade Formation of southwestern Basin, Wyoming. United States Geological Survey, Professional Pa-
Kansas. Papers of the Michigan Academy of Science, Arts and Let- per, 495-A, Al-A77.
ters, 41, 145-201. Kelly, T. S. (1990). Biostratigraphy of Uintan and Duchesnean land mam-
Hibbard, C. W., & Keenman, K. A. (1950). New evidence of the Miocene mal assemblages from the middle member of the Sespe Formation,
Age of the Blacktail Deer Creek Formation in Montana. Contri- Simi Valley, California. Contributions in Science, the Natural His-
butions to the Museum of Paleontology, University of Michigan, 8, tory Museum of Los Angeles County, 419, 1-42.
193-204. Kelly, T. S., & Lander, E. B. (1988). Biostratigraphy and correlation of
Hinderstein, B., & Boyce, J. (1977). The Miocene salamander Batra- Hemingfordian and Barstovian Land Mammal assemblages, Caliente
cosauroides dissimulans from east Texas. Journal of Herpetology, Formation, Cuyama Valley area, California. In Tertiary Tectonics
II, 369-72. and Sedimentation in the Cuyama basin, San Luis Opisbo, Santa
Holtzman, R. C. (1978). Late Paleocene mammals of the Tongue River Barbara and Ventura Counties, California. Pacific Section Society
Formation, western North Dakota. North Dakota Geological Survey of Economic Paleontologists and Mineralogists, 59, 1-19.
Report of Investigation, 65, 1-68. Kelly, T. S., & Whistler, D. P. (1994). Additional Uintan and Duchesnean
Honey, J. G. (1988). A mammalian fauna from the base of the Eocene Cathe- (middle and late Eocene) mammals from the Sespe Formation, Simi
dral Bluffs Tongue of the Wasatch Formation, Cottonwood Creek Valley, California. Contributions in Science, the Natural History Mu-
area, Southeast Washakie Basin, Wyoming. United States Geologi- seum of Los Angeles County, 439, 1-29.
cal Survey Bulletin, 1669-C, Cl-Cl4. Kihm, A. J. (1984). Early Eocene Mammalian Fauna of the Piceance Creek
Honey, J. G., & Alzett, G. (1988). Paleontology, taphonomy, and stratigra- Basin, Northwestern Colorado. Unpublished Ph.D. Thesis. Boulder:
phy of the Browns Park Formation (Oligocene and Miocene) near University of Colorado.
Maybell, Moffat Co., Colorado. United States Geological Survey (1987). Mammalian paleontology and geology of the Yoder Member,
Professional Paper, 1358, 1-52. Chadron Formation, east-central Wyoming. Dakoterra, 3, 28-45.
Hough, J. (1955). An upper Eocene fauna from the Sage Creek area, Beaver- Kihm, A. J., Hartman, J. H., & Krause, D. W. (1993). A new late Paleocene
head County, Montana. Journal of Paleontology, 29, 22-36. mammal Local Fauna from the Sentinel Butte Formation of North
Hough, J., & Alf, R. (1956). A Chadron mammal fauna from Nebraska. Dakota. Journal of Vertebrate Paleontology, 13 (suppl. to no. 3),
Journal of Paleontology, 30, 132-40. p.44A.
Hulbert, R. C., Jr. (1987a). A new Cormohipparion (Mammalia, Equidae) Koerner, H. E. (1940). The geology and vertebrate paleontology of the Fort
from the Pliocene (latest Hemphillian or Blanc an) of Florida. Journal Logan and Deep River Formations of Montana, Part 1: New verte-
of Vertebrate Paleontology, 7, 451-68. brates. American Journal of Science, 238, 837-62.
(1987b). Late Neogene Neohipparion (Mammalia, Equidae) from the Konizeski, R. L. (1957). Paleoecology of the middle Pliocene Deer Lodge
Gulf Coastal Plain of Florida and Texas. Journal of Paleontology, fauna, western Montana. Bulletin of the Geological Society of
61, 809-30. America, 68, 131-50.
(1988a). Calippus and Protohippus (Mammalia, Perissodactyla, (1961). Paleoecology of an Early Oligocene biota from Douglass Creek,
Equidae) of the Gulf Coastal Plain. Bulletin of Florida State Mu- Montana. Geological Society of America Bulletin, 72, 1633-42.
seum, Biological Sciences, 32, 221-340. Korth, W.W. (1989). Stratigraphic occurrence of rodents and lagomorphs
(1988b). Cormohipparion and Neohipparion (Mammalia, Perisso- in the Orella Member, Brule Formation (Oligocene), northwestern
dactyla, Equidae) from the late Neogene of Florida. Bulletin of Nebraska. Contributions to Geology, University of Wyoming, 27,
Florida State Museum, Biological Sciences, 33, 229-338. 15-20.
Hulbert, R. C., Jr., & Morgan, G. S. (1989). Stratigraphy, paleoecology, Korth, W. W., & Evander, R. L. (1982). A new species of Orohippus (Peris-
and vertebrate faunas of the Leisey Shell Pit Local Fauna, early sodactyla, Equidae) from the early Eocene of Wyoming. Journal of
Pleistocene (lrvingtonian) of southwestern Florida. Papers in Florida Vertebrate Paleontology, 2, 167-71.
Paleontology, Florida Museum of Natural History, 2, 1-19. Krause, D. W., & Gingerich, P. D. (1983). Mammalian fauna from Dou-
Hunt, R. M., Jr. (1974). Daphoenictis, a cat-like carnivore (Mammalia, Am- glas Quarry, earliest Tiffanian (late Paleocene) of the eastern Crazy
phicyonidae) from the Oligocene of North America. Journal of Pa- Mountain Basin, Montana. Contributions from the Museum of Pale-
leontology, 48, 1030-47. ontology, University of Michigan, 26, 157-96.
(1981 ). Geology and vertebrate paleontology of the Agate Fossil Krause, D. W., & Maas, M. C. (1990). The biogeographic origins of late
Beds National Monument and surrounding region, Sioux County, Paleocene-early Eocene mammalian immigrants to the Western Inte-
Nebraska. National Geographic Society Reports, 13, 263-85. rior of North America. Geological Society ofAmerica Special Paper,
(1985). Fauna! succession, lithofacies, and depositional environments in 243, 71-105.
Arikaree rocks (lower Miocene) of the Hartville Table, Nebraska and Krishtalka, L. (1973). Late Paleocene mammals from the Cypress Hills, Al-
Wyoming. Dakoterra, 2, 335-52. berta. Special Publications of the Museum ofTexas Tech UniversitJi,
Jahns, R. H. (1940). Stratigraphy of the easternmost Ventura Basin, 2, 1-77.
California, with a description of a new lower Miocene mammalian (1979). Paleontology and geology of the Badwater Creek area, central
fauna from the Tick Canyon formation. In Studies of Cenozaic Ver- Wyoming, Part 18: Revision of Late Eocene Hyopsodus. Annals of
tebrates and Stratigraphy of Western North America, eds. P. C. Hen- the Carnegie Museum, 48, 377-89.
shaw, R. W. Wilson, H. Howard, A. H. Miller, J. F. Dougherty, Krishtalka, L., Black, C. C., & Riedal, D. W. (1975). Paleontology and
& R.H. Jahns. Carnegie Institute of Washington Publication, 514, geology of the Badwater Creek area, central Wyoming, Part 10: A
145-94. late Paleocene mammal fauna from the Shotgun Member of the Fort
James, G. T. (1963 ). Paleontology and nonmarine stratigraphy of the Cuyama Union Formation. Annals of the Carnegie Museum, 45, 179-212.
Valley Badlands, California, Part 1: Geology, fauna! interpretation, Krishtalka, L., & Setoguchi, T. (1977). Paleontology and geology of the
and systematic descriptions of Chiroptera, Insectivora and Rodentia. Badwater Creek area, central Wyoming, Part 13: The Late Eocene
University of California Publications in Geological Sciences, 45, Insectivora and Dermoptera. Annals of the Carnegie Museum, 46,
191-234. 71-99.
Kay, J. L. (1957). The Eocene vertebrates of the Uinta Basin. In Eighth An- Krishtalka, L., & Stucky, R. K. (1983a). Revision of the Wind River faunas,
nual Field Conference, Intermontain Association of Petroleum Ge- early Eocene of central Wyoming, Part 3: Marsupialia. Annals of the
ologists, ed. 0. G. Seal, pp. 110--14. Carnegie Museum, 52, 205-27.
Keefer, W.R. (1965). Stratigraphy and geologic history of the uppermost ( 1983b ). Paleocene and Eocene marsupials of North America. Annals of
Cretaceous, Paleocene, and lower Eocene rocks in the Wind River the Carnegie Museum, 52, 229-63.
 Appendix II: References for locality listings 660

(1984). Middle Eocene marsupials (Mammalia) from northeastern Utah Love, J. D., McKenna, M. C., & Dawson, M. R. (1976). Eocene, Oligocene
and the mammalian fauna from Powder Wash. Annals ofthe Carnegie and Miocene rocks and vertebrate fossils at the Emerald Lake locality,
Museum, 53, 31-45. three miles south of Yellowstone National Park, Wyoming. United
(1985). Revision of the Wind River faunas, early Eocene of central States Geological Survey Professional Paper, 932-A, Al-A28.
Wyoming, Part 7: Revision of Diacodexis (Mammalia, Artiodactyla). Lozinsky, R. P., & Tedford, R.H. (1991). Geology and paleontology of the
Annals of the Carnegie Museum, 54, 413-86. Santa Fe Group, southwestern Albuquerque Basin, Valencia County,
(1986). Early Eocene artiodactyls from the San Juan Basin, New Mexico. New Mexico. New Mexico Bureau of Mines and Mineral Resources
Contributions to Geology, University of Wyoming, 3, 183-96. Bulletin, 132, 1-36.
Krishtalka, L., West, R. M., Black, C. C., Dawson, M. R., Flynn, J. R., Lucas, S. G. (1983). The Baca Formation and the Eocene-Oligocene bound-
Turnbull, W. D., Stucky, R. K., McKenna, M. C., Bown, T. M., Golz, ary in New Mexico. In Socorro Region II, eds. C. E. Chapin &
D. J ., & Lillegraven, J. A. (1987). Eocene (Wasatchian throughDuch- F. Callender, pp. 187-92. New Mexico Geological Society.
esnean) biochronology of North America. In Cenozoic Mammals of (1984). Early Paleocene vertebrates, stratigraphy, and biostratigraphy,
North America, ed. M. 0. Woodburne, pp. 77-117. Berkeley: Uni- West Fork of Gallegos Canyon, San Juan Basin, New Mexico. New
versity of California Press. Mexico Geology, 6, 56-60.
Kurten, B., & Anderson, E. (1980). Pleistocene Mammals ofNorth America. (1986a). The first Oligocene mammal from New Mexico. Journal of
New York: Columbia University Press. Paleontology, 60, 1274-6.
Lander, E. B. (1983). Continental vertebrate faunas from the upper member ( 1986b ). Oligocene mammals from the Black Range, southwestern New
of the Sespe Formation, Simi Valley, California, and the terminal Mexico. In Truth or Consequences Region, Socorro, eds. R. E.
Eocene event. In Cenozoic Geology ofthe Simi Valley Area, Southern Clemons, W. E. King, & G. H. Mack, pp. 261-63. Albuquerque:
California, eds. R. R. Squires & M. V. Filewicz. Pacific Section New Mexico Geological Society.
Society of Economic Paleontologists and Mineralogists Fall Trip ( 1992). Redefinition of the Duchesnean land-mammal "age," late Eocene
Volume and Guidebook, 124-44. of western North America. In Eocene-Oligocene Climatic and Bi-
( 1985). Early and middle Miocene continental vertebrate assemblages, otic Evolution, eds. D.R. Prothero & W. A. Berggren, pp. 88-105.
central Mohave Desert, San Bernadino County, California. In Ge- Princeton: Princeton University Press.
ological Investigations along Interstate 15, Cajon Pass to Manix Lucas, S. G., & Kues, B. S. (1979). Vertebrate biostratigraphy of the Eocene
Lake, California, ed. R. E. Reynolds, pp. 127-44. Redlands, Calif.: Galisteo Formation, north-central New Mexico. New Mexico Geo-
San Bernadino County Museum. logical Society Guidebook, 30th Field Conference, Santa Fe County,
Lander, E. B., & Reynolds, R. E. (1985). Fossil vertebrates from the Cal- 225-9.
ico Mountain area, central Mohave Desert, San Bernadino County, Lucas, S. G., Schoch, R. M., Manning, E., & Tsentas, C. (1981). The Eocene
California. In Geological Investigations along Interstate 15, Cajon biostratigraphy of New Mexico. Geological Society of America Bul-
Pass to Manix Lake, California, ed. R. E. Reynolds, pp. 153-6. Red- letin, 92, 951-67.
lands, Calif.: San Bernadino County Museum. Lucas, S. G., & Williamson, T. E. (1993). Eocene vertebrates and late
Leiggi, P. (1991). Report In Society of Vertebrate Paleontology News Bul- Laramide stratigraphy of New Mexico. In Vertebrate Paleontology
letin, 151, 65. in New Mexico. eds. S. G. Lucas & J. Zidek. Bulletin of the New
Leite, M. (1990). Stratigraphy and mammalian paleontology of the Ash Mexico Museum of Natural History and Science, 2, 145-58.
Hollow Formation (Upper Miocene) on the north shore of Lake Mc- Lull, R. S. (1922). Primitive Pecora in the Yale Museum. American Journal
Conaughy, Keith County, Nebraska. Contributions to Geology, Uni- of Science, 5, 111-19.
versity of Wyoming, 28, 1-29. Maas, M. C. (1985). Taphonomy of a late Eocene microvertebrate locality,
Lemley, R. E. (1971). Notice of new finds in the Badlands. Proceedings of Wind River Basin, Wyoming (U.S.A.). Palaeogeography, Palaeocli-
the South Dakota Academy of Sciences, 50, 70-4. matology, Palaeoecology, 52, 123-42.
Lillegraven, J. A. (1970). Stratigraphy, structure and vertebrate fossils of Macdonald, J. R. (1948). The Pliocene carnivores of the Black Hawk Ranch
the Oligocene Brule Formation, Slim Buttes, northwestern South fauna. University of California Publications, Bulletin of the Depart-
Dakota. Bulletin of the Geological Society of America, 81, 831-50. ment of Geological Sciences, 28, 53-80.
Lillegraven, J. A., & Tabrum, A. K. (1983). A new species of Centetodon (1956). A new Clarendonian mammalian fauna from the Truckee For-
(Mammalia, Insectivora, Geolobididae) from southwestern Montana mation of western Nevada. Journal of Paleontology, 30, 186-202.
and its biogeographical implications. Contributions to Geology, Uni- (1959). The middle Pliocene mammalian fauna from Smiths Valley,
versity of Wyoming, 22, 57-73. Nevada. Journal of Paleontology, 33, 872-87.
Lindsay, E. H. (1974). The Hemingfordian mammal fauna of the Ved- (1966). The Barstovian Camp Creek Fauna from Elko County, Nevada.
der locality, Branch Canyon Formation, Santa Barbara County, Contributions in Science, Los Angeles County Museum, 92, 1-18.
California, Part II: Rodentia (Eomyidae and Heteromyidae). Pale- ( 1970). Review of the Miocene Wounded Knee faunas of southwestern
oBios, 16, 1-19. South Dakota. Bulletin ofthe Los Angeles County Museum ofNatural
Lindsay, E. H. (1984). Late Cenozoic mammals from northwestern Mexico. History, Science, 8, 1-82.
Journal of Vertebrate Paleontology, 4, 208-15. Macdonald, J. R., & Pelletier, W. J. (1958). The Pliocene mammalian faunas
Lindsay, E. H., & Jacobs, L. L. (1985). Pliocene small mammals from Chi- of Nevada. U.S.A. International Geological Congress, 20th Session,
huahua, Mexico. Paleontologica Mexicana, 51, 1-50. Section VII, 365-88.
Lindsay, E. H., Johnson, N. M., Opdyke, N. D., & Butler, R. F. (1987). MacFadden, B. J. (1977). Magnetic polarity stratigraphy of the Chamita
Mammalian chronology and the magnetic polarity time scale. In Formation stratotype (Mio-Pliocene) of north-central New Mexico.
Cenozoic Mammals of North America, ed. M. 0. Woodburne, American Journal of Science, 277, 769-800.
pp. 267-84. Berkeley: University of California Press. (General strati- (1980a). Eocene perissodactyls from the type section of the Tepee Trail
graphic reference.) Formation of northeastern Wyoming. Contributions to Geology, Uni-
Lindsay, E. H., Opdyke, V. D., & Johnson, N. M. (1984). Blancan- versity of Wyoming, 18, 135-43.
Hemphillian land mammal ages and late Cenozoic mammal dispersal (1980b). An early Miocene land mammal (Oreodonta) from a marine
events. Annual Review of Earth and Planetary Science, 12, 445-88. limestone in northern Florida. Journal of Paleontology, 54, 93-101.
Lindsay, E. H., & Tessman, N. T. (1974). Cenozoic vertebrate localities and (1982). New species of primitive three-toed horse from the Miocene
faunas in Arizona. Journal of the Arizona Academy of Science, 9, phosphate mining district of central Florida. Florida Scientist, 45,
1-24. 117-25.
 Appendix II: References for locality listings 661

(1984). Systematics and phylogeny of Hipparion, Neohipparion, Elk mountain and Tabernacle Butte area, Wyoming. Bulletin of the
Nannippus, and Cormohipparion (Mammalia, Equidae) from the AmericanMuseumofNaturalHistory, 117, 117-76.
Miocene and Pliocene of the New World. Bulletin of the American McGrew, P. 0., & Sullivan, R. (1970). The stratigraphy and paleontology
Museum of Natural History, 179, 1-195. of Bridger A. Contributions to Geology, University of Wyoming, 9,
(1986). Late Hemphillian monodactyl horses (Mammalia, Equidae) from 66-85.
the Bone Valley Formation of central Florida. Journal of Paleontol- McKenna, M. C. (1960a). A continental Paleocene vertebrate fauna from
ogy, 60, 466-75. California. American Museum Novitates, 2024, 1-20.
MacFadden, B. J., & Galiano, H. (1981). Late Hemphillian cat (Mammalia, (1960b). Fossil Mammalia from the early Wasatchian Four Mile fauna
Felidae) from the Bone Valley Formation of central Florida. Journal of northwest Colorado. University of California, Publications in Ge-
of Paleontology, 55, 218-26. ological Sciences, 37, 1-130.
MacFadden, B. J., Johnson, N. M., & Opdyke, N. D. (1979). Magnetic po- (1972). Vertebrate paleontology of the Togwotee Pass area, northwestern
larity stratigraphy of the Mio-Pliocene mammal-bearing Big Sandy Wyoming. In Guidebook, Field Conference on Tertiary Biostratigra-
Formation of western Arizona. Earth and Planetary Science Letters, phy of Southern and Western Wyoming, ed. R. M. West, pp. 80-101.
44, 349-64. (1976). Esthonyx in the upper fauna! assemblage, Huerfano Formation,
MacFadden, B. J., & Nelson, M. E. (1980). Miocene three-toed horse from Eocene of Colorado. Journal of Paleontology, 50, 354-61.
the Salt Lake group of southeastern Idaho. Transactions ofthe Kansas (1980a). Eocene paleolatitude, climate, and mammals of Ellesmere
Academy of Sciences, 83, 20-5. Island. Palaeogeography, Palaeoclimatology, Palaeoecology, 30,
MacFadden, B. J., & Skinner, M. F. ( 1979). Diversification and biogeography 349-62.
of the one-toed horses Onohippidium and Hippidium. Postilla, 175, (l 980b ). Late Cretaceous and early Tertiary vertebrate paleontological
1-10. reconnaissance, Towotee Pass area, northwestern Wyoming. In As-
MacFadden, B. J., & Waldrop, J. S. (1980). Nannippus phlegon (Mammalia, pects of Vertebrate History, ed. L. L. Jacobs, pp. 321-43. Flagstaff:
Equidae) from the Pliocene (Blancan) of Florida. Bulletin of the Museum of Northern Arizona Press.
Florida State Museum, Biological Sciences, 25, 1-37. ( l 980c ). Remaining evidence of Oligocene sedimentary rocks previously
MacFadden, B. 1., & Webb, S. D. (1982). The succession of Miocene (Arika- present across the Bighorn basin, Wyoming. In Early Cenozoic Pa-
reean through Hemphillian) terrestrial mammal localities and faunas leontology and Stratigraphy of the Bighorn Basin, Wyoming, ed.
in Florida. In Miocene of the Southwestern United States, eds. T. M. P. D. Gingerich. University of Michigan Papers in Paleontology, 24,
Scott & S. B. Upchurch, pp. 186-99. Tallahassee: Florida Depart- 143-6.
ment of Natural Resources, Bureau of Geology Special Publication (1990). Plagiomenids (Mammalia, ?Dermoptera) from the Oligocene of
No. 25. Oregon, Montana, and South Dakota, and middle Eocene of north-
Madden, C. T., & Dalquest, W. W. (1990). The last rhinoceros in North western Wyoming. Geological Society of America Special Paper,
America. Journal of Vertebrate Paleontology, 10, 266-7. 243, 211-34.
Madden, C. T., & Storer, J. E. (1985). The Proboscidea from the mid- McKenna, M. C., Robinson, P., & Taylor, D. W. (1962). Notes on
dle Miocene Wood Formation, Saskatchewan. Canadian Journal of Eocene Mammalia and Mollusca from Tabernacle Butte, Wyoming.
Earth Science, 22, 1345-50. American Museum Novitates, 2102.
Mader, B. J., & Alexander, J. P. (1995). Megacerops kuwagatarhinus n. Merriam, J. C. (1911). Tertiary mammal beds of Virgin Valley and Thou-
sp., an unusual brontothere (Mammalia, Perissodactyla) with distally sand Creek in northwestern Nevada. University of California Pub-
forked horns. Journal of Paleontology, 69, 581-7. lications, Bulletin of the Department of Geological Sciences, 6,
Maglio, V. J. (1966). A revision of the fossil selenodont artiodactyls from 199-304.
the middle Miocene of Thomas Farm, Gilchrist County, Florida. (1915). Tertiary vertebrate faunas of the North Coalinga region of
Breviora, 225, 1-27. California. Transactions of the American Philosophical Society, 22,
Manning, E. (1990). The late early Miocene Sabine River. In Transactions of 191-234.
the Gulf Coast Association of Geological Societies, eds. E. Kinsland (1919). Tertiary mammalian faunas of the Mojave Desert. University of
& T. Cagle, XL, 531-49. California Publications, Bulletin of the Department of Geological
Manning, E., & MacFadden, B. J. (1989). Pliocene three-toed horses from Sciences, 11, 437-585.
Louisiana, with comments on the Citronelle Formation. Tulane Stud- Merriam, J. C., & Sinclair, W. J. (1907). Tertiary faunas of the John Day
ies in Geology and Paleontology, 22, 35-46. Region. University of California Publications, Bulletin of the De-
Martin, J. E., & Green, M. (1984). Insectivora, Sciuridae, and Cricetidae partment of Geological Sciences, 5, 171-205.
from the early Miocene Rosebud Formation in South Dakota. Special Merriam, J. C., & Stock, C. (1928). A further contribution to the mam-
Publication, Carnegie Museum of Natural History, 9, 28-40. malian fauna of the Thousand Creek Pliocene northwestern Nevada,
Martin, L. D. (1974). New rodents from the Lower Miocene Gering For- Publications of the Carnegie Institute of Washington, 393, 5-21.
mation of western Nebraska. Occasional Papers of the Museum of Miller, W. E., & Carranza-Casteneda, 0. (1984). Late Cenozoic mammals
Natural History, University of Kansas (Lawrence), 32, 1-12. from Central Mexico. Journal of Vertebrate Paleontology, 4, 216-36.
Martin, L. D., & Mengel, R. (1984). A new cuckoo and chachalaca from Montellano, M. (1989). Pliocene Camelidae of Rancho El Ocote, central
the early Miocene of Colorado. Special Publication of the Carnegie Mexico. Journal ofMammalogy, 70, 359-69.
Museum of Natural History, 9, 171-7. Morea, M. F. ( 1981 ). Massacre Lake Local Fauna (Miocene, Hemingfordian)
Martin, L. E., & Tedrow, A. R. (1988). Carnivora from the Ellensburg For- from Northwestern Washoe County, Nevada. Unpublished Ph.D. The-
mation (Miocene) of central Washington. Proceedings of the North sis. Riverside: University of California.
Dakota Academy of Sciences, 42, 15. Morgan, G. S., & Ridgeway, R. B. (1987). Late Pliocene (Late Blancan) ver-
Matthew, W. D., & Cook, H. J. (1909). A Pliocene fauna from western tebrates from the St. Petersburg Times site, Pinellas County, Florida,
Nebraska. Bulletin of the American Museum of Natural History, 26, with a brief review of Florida Blancan faunas. Papers in Florida
361-414. Paleontology, Florida Museum of Natural History, I, 1-22.
May, S. R., & Repenning, C. A. (1982). New ~vidence for the age of the Morgan, J. K., & Morgan, N. H. (1995). A new species of Capromeryx
Mount Eden fauna, southern California. Journal of Vertebrate Pale- (Mammalia: Artiodactyla) from the Taunton Local Fauna of
ontology, 2, 109-13. Washington, and the correlation with other Blancan faunas of
McGrew, P. 0., Berman, J. E., Hecht, M. K., Hun;imell, J. M., Simpson, Washington and Idaho. Journal of Vertebrate Paleontology, 15,
G. G., & Wood, A. E. (1959). The geology and paleontology of the 160-70.
 Appendix II: References for locality listings 662

Munthe, J. (1979). The Hemingfordian mammalian fauna of Vedder Local- tion, southern Sierra Nevada, California. Contributions in Science,
ity, Branch Canyon Sandstone, Santa Barbara County, California, Natural History Museum of Los Angeles County, 393, 1-31.
Part III: Carnivores, perissodactyls, artiodactyls and summary. Pale- Rapp, S. D., MacFadden, B. J., & Schiebout, J. A. (1983). Magnetic polarity
obios, 29, 1-22. stratigraphy of the early Tertiary Black Peaks Formation, Big Bend
(1988). Miocene mammals of the Split Rock area, Granite Mountain National Park, Texas. Journal of Geology, 91, 555-72.
Basin, central Wyoming. University of California, Publications in Raschke, R. E. (1984). Early and Middle Miocene vertebrates from the
Geological Sciences, 126, 1-136. Santa Ana Mountains, California. In The Natural Science of Orange
Nelson, M. E. (1973). Age and stratigraphic relationships of the Fowkes County: a Collection of Occasional Papers Concerning the Natural
Formation, Eocene, of southwestern Wyoming and eastern Utah. Science of Orange County, California, in Celebration of the Tenth
Contributions to Geology, University of Wyoming, 12, 27-31. Anniversary of the Natural History Foundation of Orange County,
(1976). A new Oligocene fauna from northeastern Utah. Transactions of California, ed. B. Butler. Memoirs ofthe Natural History Foundation,
the Academy of Sciences, 79, 7-13. Orange County, 1, 61-7.
Nelson, M. E., & Madsen, J. H. (1987). A new Clarendonian (late Miocene) Rasmussen, D. L. (1989). Depositional environments, paleoecology, and
fauna from eastern Wyoming. Contributions to Geology, University biostratigraphy of Arikareean Bozeman Group Strata west of the
of Wyoming, 25, 23-8. Continental Divide in Montana. 1989 MGS Field Conference, Mon-
Nelson, M. E., Madsen, J. H., Jr., & Stokes, W. L. (1980). A titanotherefrom tana Centennial, 205-13.
the Green River Formation, central Utah: Teleodus uintensis (Peris- Repenning, C. A. (1987). Biochronology of microtine rodents. In Ceno-
sodactyla, Brontotheriidae). Contributions to Geology, University of zoic Mammals ofNorth America, ed. M. 0. Woodburne, pp. 236-68.
Wyoming, 12, 27-31. Berkeley: University of California Press. (General stratigraphic ref-
Nichols, R. ( 1976). Early Miocene mammals from the Lemhi Valley ofldaho. erence.)
Tebiwa, 18, 9-48. Repenning, C. A., & Vedder, J. G. (1961). Continental vertebrates and
Novacek, M. J., Ferrusquia-Villafranca, I, Flynn, J. J., Wyss, A. R., & Norell, their stratigraphic correlation with marine mollusks, eastern Caliente
M.A. (1991). Wasatchian (early Eocene) mammals and other verte- Range, California. United States Geological Survey Professional Pa-
brates from Baja California, Mexico: the Lomas Las Tetas de Cabra per, 424-C, C235-39.
Fauna. Bulletin of the American Museum of Natural History, 208, Reynolds, R. E., & Lander, E. B. (1985). Preliminary report on the Miocene
1-88. Daggett Ridge Local Fauna, central Mohave Desert, San Bernadino
Opdyke, N. D., Lindsay, E. H., Johnson, N. M., & Downs, T. (1977). The County, California. In Geological Investigations Along Interstate
paleomagnatism and magnetic polarity stratigraphy of the mammal- I 5, Cajon Pass to Manix Lake, California, ed. R. E. Reynolds,
bearing section of the Anza-Borrego State Park, California. Quater- pp. 105-10. Redlands, Calif.: San Bernadino County Museum.
nary Research, 7, 316-29. Reynolds, R. E., Reynolds, R. L., & Korth, W. W. (1991). Late Heming-
Osborn, H.F. (1909). Cenozoic mammal horizons of western North America. fordian and early Barstovian Faunas from the Crowder Formation,
Bulletin of the United States Geological Survey, 361, 1-90. Cajon Pass, San Bernadino County, California. Journal of Vertebrate
Ostrander, G. E. (1985). Correlation of the early Oligocene (Chadronian) in Paleontology, 11 (suppl. to no. 3), p. 52A.
northwestern Nebraska. Dakoterra, 2, 205-31. Rich, T. H. V., & Collinson, J. W. (1973). First mammalian fossil from the
Patton, T. H. (1969a). Miocene and Pliocene artiodactyls of Texas Gulf Flagstaff Limestone, central Utah. Vulpavus australis (Carnivora,
Coastal Plain. Bulletin ofFlorida State Museum Biological Sciences, Miacidae). Journal of Paleontology, 47, 854-66.
14, 115-226. Rich, T. H. V., & Rasmussen, D. L. (1973). New North American erinaceine
(1969b). An Oligocene land vertebrate fauna from Florida. Journal of hedgehogs (Mammalia, lnsectivora) in North America. University of
Paleontology, 43, 543-6. Kansas Museum of Natural History, Occasional Papers, 21, 1-54.
Patton, T. H., & Taylor, B. E. (1971 ). The Synthetoceratinae (Mammalia, Ty- Rigby, K. J., Jr. (1980). Swain Quarry of the Fort Union Formation, mid-
lopoda, Protoceratidae). Bulletin ofthe American Museum ofNatural dle Paleocene (Torrejonian), Carbon County, Wyoming. Geologic
History, 145, 119-218. Setting and mammalian fauna. Evolutionary Monographs, 3, 1-162.
(1973). The Protoceratinae (Mammalia, Tylopoda, Protoceratidae) and Ritchie, K. A. (1948). Lower Pliocene horses from Black Hawk Ranch,
the systematics of the Protoceratidae. Bulletin of the American Mu- Mount Diablo, California. University of California Publications,
seum of Natural History, 150, 351-413. Bulletin of the Department of Geological Sciences, 28, 1-44.
Pierce, H. G., & Rasmussen, D. L. (1989). New land snails (Archaeogas- Robinson, L. M., & Honey, J. G. (1987). Geology and setting of a new
tropoda, Helicinidae) from the Miocene (early Barstovian) Flint Paleocene mammal locality in the northern Powder River Basin,
Creek beds of western Montana. Journal of Paleontology, 63, Montana. Palaios, 2, 87-90.
646-51. Robinson, P. (1963). Fossil vertebrates and age of the Cuchara Formation
Pinsof, J. D. (1985). The Pleistocene vertebrate localities of South Dakota. of Colorado. University of Colorado Studies, Series in Geology, 1,
Dakoterra, 2, 233-64. 1-9.
Princhinello, K. A. (1971). Earliest Eocene mammalian fossils from the (1966). Fossil Mammalia of the Huerfano Formation, Eocene, of Col-
Laramie Basin of southeastern Wyoming. Contributions to Geology, orado. Yale Peabody Museum, Bulletin of the Museum of Natural
University of Wyoming, IO, 73-87. History, 21, 1-95.
Prothero, D.R., Denham, C.R., & Farmer, H. G. (1982). Oligocene calibra- Robison, S. F. (1980). Paleocene (Puercan-Torrejonian) Mammalian Fau-
tion of the magnetic polarity time scale. Geology, 10, 650-3. nas of the North Horn Fonnation, Central Utah. Unpublished M.S.
Prothero, D.R., & Manning, E. M. (1987). Miocene rhinoceroses from the Thesis. Provo, Utah: Brigham Young University.
Texas Gulf Coastal Plain. Journal of Paleontology, 61, 388-423. Roehler, H. W. (1973). Stratigraphy of the Washakie Formation in the
Prothero, D. R., & Sereno, P. C. (1982). Allometry and paleoecology of Washakie Basin, Wyoming. Geological Survey Bulletin, 1369,
medial Miocene dwarf rhinoceroses from the Texas Gulf Coastal 1-40.
Plain. Paleobiology, 8, 16-30. Rose, K. D. (1981). The Clarkforkian land-mammal age and mammalian
Prothero, D. R., & Shubin, N. ( 1989). The evolution of Oligocene horses. In composition across the Paleocene-Eocene boundary. University of
The Evolution ofPerissodactyls, eds. D.R. Prothero & R. M. Schoch, Michigan Papers in Paleontology, 26, 1-197.
pp. 142-75. Oxford: Oxford University Press. Russell, L. S. (1929). Paleocene vertebrates from Alberta. American Journal
Quinn, J. P. (1987). Stratigraphy of the middle Miocene Bopesta Forma- of Science, 217, 162-176.
 Appendix II: References for locality listings 663

(1958). Paleocene mammal teeth from Alberta. Bulletin of the National Oligocene). Bulletin of the Carnegie Museum of Natural History, 9,
Museum of Canada, 147, 96-103. 1-61.
(1965a). Tertiary mammals of Saskatchewan, Part I: The Eocene fauna. Shankler, D. M. (1980). Fauna! zonation of the Willwood Formation in the
Contributions of the Royal Ontario Museum, Life Sciences, 67, 1-33. central Bighorn basin, Wyoming. In Early Cenozoic Paleontology
(1965b). The continental Tertiary of western Canada. In Vertebrate Pa- and Stratigraphy of the Bighorn Basin, Wyoming, ed. P. D. Gingerich.
leontology in Alberta, eds. R. E. Folinsbee & D. M. Ross. University University of Michigan Papers in Paleontology, 24, 99-114.
of Alberta, Department of Geology Bulletin, 2, 41-52. Shotwell, J. A. (1956). Hemphillian mammalian assemblage from North-
(1967). Paleontology of the Swan Hills area, north-central Alberta. Life eastern Oregon. Bulletin of the Geological Society of America, 67,
Sciences Contributions, Royal Ontario Museum, 71, 1-30. 717-38.
(1972). Tertiary mammals of Saskatchewan, Part II: The Oligocene (1958). Inter-community relationships in Hemphillian (Mio-Pliocene)
fauna, non-ungulate orders. Life Sciences Contributions, Royal On- mammals. Ecology, 29, 271-82.
tario Museum, 84, 1-63. (1968). Miocene mammals of southeast Oregon. Bulletin of the Museum
( 197 4 ). Fauna and correlation of the Ravenscrag Formation (Paleocene) of Natural History, University of Oregon, Eugene, 14, 1-67.
of southwestern Saskatchewan. Life Sciences Contributions, Royal (1970). Pliocene mammals of southeast Oregon and adjacent Idaho. Bul-
Ontario Museum, 102, 1-52. letin of the Museum of Natural History, University of Oregon, Eu-
(1975). Revision of the fossil horses from the Cypress Hills Formation gene, 17, 1-103.
(lower Oligocene) of Saskatchewan. Canadian Journal of Earth Sci- Simpson, G. G. (1930). Tertiary land mammals of Florida. Bulletin of the
ences, 12, 636-48. American Museum of Natural History, 59, 149-211.
(1978). Tertiary mammals of Saskatchewan, Part IV: The Oligocene Simpson, G. G. (1937). Additions to the upper Paleocene fauna of Crazy
anthracotheres. Life Sciences Contributions, Royal Ontario Museum, Mountain Field. American Museum Novitates, 940, 1-15.
115, 1-16. ( 1959). Fossil mammals from the type area of the Puerco and Nacimiento
(1980). Tertiary mammals of Saskatchewan, Part V: The Oligocene en- strata, Paleocene of New Mexico. American Museum Novitates,
telodonts. Life Sciences Contributions, Royal Ontario Museum, 122, 1957, 1-22.
1-42. Simpson, W. D. (1985). Geology and paleontology of the Oligocene Harris
(1982). Tertiary mammals of Saskatchewan, Part VI: The Oligocene Ranch Badlands, southwestern South Dakota. Dakoterra, 2, 303-33.
rhinoceroses. Life Sciences Contributions, Royal Ontario Museum, Skinner, M. F., & Hibbard, C. W. (1972). Pleistocene pre-glacial and glacial
113, 1-58. rocks and faunas of north-central Nebraska. Bulletin of the American
(1985). Revision of the fossil horses from the Cypress Hills Formation Museum of Natural History, 148, 1-148.
(Lower Oligocene) of Saskatchewan. Canadian Journal of Earth Skinner, M. F., & Johnson, F. W. (1984). Tertiary stratigraphy and the Frick
Sciences, 12, 636-48. collection of fossil vertebrates from north-central Nebraska. Bulletin
Savage, D. E. (1955). Nomnarine lower Pliocene sediments in California. of the American Museum of Natural History, 178, 215-368.
University of California Publications, Bulletin of the Department of Skinner, M. F., Skinner, S. M., & Gooris, R. J. (1968). Cenozoic rocks and
Geological Sciences, 31, 1-26. faunas of Turtle Butte, south-central South Dakota. Bulletin of the
Savage, D. E., Waters, B. T., & Hutchison, J. H. (1972). Wasatchian suc- American Museum of Natural History, 138, 379-436.
cession at Bitter Creek Station, northwestern border of the Washakie (1977). Stratigraphy and biostratigraphy of late Cenozoic deposits in
basin, Wyoming. In Guidebook: Field Conference on Tertiary Bios- central Sioux County, western Nebraska. Bulletin of the American
tratigraphy of Southern and Western Wyoming. ed. R. M. West, Museum of Natural History, 158, 265-371.
pp. 32-9. Skinner, S. M., & Gooris, R. J. (1966). A note on Toxotherium (Mammalia,
Schaff, C. R. (1985). Paleocene mammals from the Beartooth region of Rhinoceratoidea) from Natrona County, Wyoming. American Mu-
Wyoming and Montana. National Geographic Research Reports, seum Novitates, 2261, 1-12.
I979 projects, 589-95. Skolnick, R. (1985). Geology and paleontology of the Lay Ranch beds.
Scharf, D. W. (1935). A Miocene mammalian fauna from Sucker Creek, Proceedings of the National Academy of Sciences, 95, 54.
southeastern Oregon. Contributions to Paleontology, Carnegie Insti- Skwara, T. (1988). Mammals of the Topham Local Fauna: early Miocene
tute of Washington Publications, 453, 97-118. (Hemingfordian), Cypress Hills Formation, Saskatchewan. Natural
Schiebout, J. A. (1974). Vertebrate paleontology and paleoecology of Pa- History Contributions, Museum ofNatural History, Regina, 9, 1-169.
leocene Black Peaks Formation, Big Bend National Park, Texas. Sloan, R. E. ( 1987). Paleocene and latest Cretaceous mammal ages, biozones,
Bulletin of the Texas Memorial Museum, 24, 1-88. magnetozones, rates of sedimentation, and evolution. Geological So-
( 1977). Eocene Perissodactyla from the La Jolla and Poway Groups, San ciety of America Special Paper, 209, 165-95.
Diego County, California. Transactions of the San Diego Society of Standhardt, B. R. (1995). Early Paleocene (Puercan) vertebrates of the Dogie
Natural History, 18, 217-28. locality, Big Bend National Park, Texas. In National Park Service
Schoch, R. M. ( 1985). Preliminary description of a new late Paleocene land- Technical Report NPS/NPRO/NRTR-95116, eds. V. L. Santucci &
mammal fauna from South Carolina, U.S.A. Postilla, 196, 1-13. L. McClelland, pp. 46-8.
Schoch, R. M., & Lucas, S. G. (1981). New conoryctines (Mammalia: Tae- Stevens, M. S. (1991). Osteology, systematics, and relationships of earliest
niodonta) from the middle Paleocene (Torrejonian) from western Miocene Mesocyon venator (Cook), Camivora, Canidae. Journal of
North America. Journal of Mammalogy, 62, 683-91. Vertebrate Paleontology, 11, 45-66.
Schultz, G. B., & Stout, T. M. (1948). Pleistocene mammals and terraces in Stevens, M. S., & Stevens, J. B. (1989). Neogene-Quaternary deposits
the Great Plains. Bulletin of the Geological Society of America, 59, and vertebrate faunas. In Vertebrate Paleontology, Biostratigraphy
553-88. and Depositional Environments, Latest Cretaceous and Tertiary,
Schultz, G. E. (ed.). (1977). Guidebook: Field Conference on Late Cenozoic Big Bend Area, Texas, eds. A. B. Busbey, III & T. M. Lehman,
Biostratigraphy of the Texas Panhandle and Adjacent Oklahoma. pp. 67-90. Austin: Guidebook of the 49th Annual Meeting of the
Canyon, Tex.: Special publication of the Killgore Research Center, Society of Vertebrate Paleontology, University of Texas,
Department of Geology and Anthropology, West Texas State Uni- Stevens, M. S., Stevens, J.B., & Dawson, M. R. (1969). New early Miocene
versity, 1, 1-160. · Formation and Vertebrate Local Fauna, Big Bend National Park,
Setoguchi, T. (1978). Paleontology and geology of the Badwater Creek Brewster County, Texas. Pearce Sellards Series, Texas Memorial
area, central Wyoming, Part 16: The Cedar Ridge Local Fauna (late Museum, 15, 1-53.
 Appendix II: References for locality listings 664

Stirton, R. A. (1939). Cenozoic mammal remains from the San Francisco Taylor, B. E., & Webb, S. D. (1976). Miocene Leptomerycidae (Artiodactyla,
Bay region. University of California Publications, Bulletin of the Ruminantia) and their relationships. American Museum Novitates,
Department of Geological Sciences, 24, 339-410. 2596, 1-22.
(1940). The Nevada Miocene and Pliocene mammalian faunas as fau- Taylor, L. H. (1981). The Kutz Canyon Local Fauna, Torrejonian (middle
na! units. Proceedings of the Sixth Pacific Science Congress, 2, Paleocene) of the San Juan Basin, New Mexico. In Advances in San
627-40. Juan Basin Paleontology, eds. S. G. Lucas, J. K. Rigby, Jr., & B. S.
Stirton, R. A., & Goeriz, H. F. (1942). Fossil vertebrates from the superaja- Kues, pp. 242-63. Albuquerque: University of New Mexico Press.
cent deposits near Knights Ferry, California. University ofCalifornia Tedford, R. H. (1981). Mammalian biochronology of the late Cenozoic
Publications, Bulletin of the Department ofGeological Sciences, 26, basins of New Mexico. Bulletin ofthe Geological Society ofAmerica,
447-72. 92, 1008-22.
Stock, C. (1920). An early Tertiary vertebrate fauna from the southern coast Tedford, R. H., & Barghoorn, S. (1993). Neogene stratigraphy and mam-
range of California. University of California Publications, Bulletin malian biochronology of the Espanola Basin, northern New Mexico.
of Geological Sciences, 12, 267-76. In Vertebrate Paleontology in New Mexico, eds. S. G. Lucas & J.
(1949). Mammalian fauna from the Titus Canyon Formation, California. Zidek. Bulletin of the New Mexico Museum of Natural History and
Publicatiof! of the Carnegie Institution of Washington, 584, 229-44. Science, 2, 159-68.
Storer, J. E. (1975). Tertiary mammals of Saskatchewan, Part III: The Tedford, R.H., & Frailey, D. (1976). Review of some Carnivora (Mammalia)
Miocene fauna. Life Sciences Contributions, Royal Ontario Museum, from the Thomas Farm Local Fauna (Hemingfordian, Gilchrist
103, 1-134. County, Florida). American Museum Novitates, 2610, 1-9.
(1981). Leptomerycid artiodactyls of the Calf Creek Local Fauna (Cy- Tedford, R. H., & Hunter, M. E. (1984). Miocene marine-non-marine cor-
press Hills Formation, Oligocene, Chadronian) Saskatchewan. Nat- relations, Atlantic and Gulf Coastal Plains, North America. Palaeo-
ural History Contributions, Museum of Natural History, Regina, 3, geography, Palaeoclimatology, Palaeoecology, 47, 129-51.
1-32. Tedford, R. H., Skinner, M. S., Fields, R. S., Rensberger, J. M., Whistler,
(1984a). Mammals of the Swift Current Creek Local Fauna (Eocene: D. P., Galusha, T., Taylor, B. E., Macdonald, J. R., & Webb,
Uintan, Saskatchewan). Natural History Contributions, Museum of S. D. (l 987). Fauna! succession and biochronology of the Arika-
Natural History, Regina, 7, 1-158. reean through Hemphillian (late Oligocene through earliest Pliocene
(l 984b). Fossil mammals of the Southfork Local Fauna (early Chadro- epochs) in North America. In Cenozoic Mammals of North
nian) of Saskatchewan. Canadian Journal of Earth Sciences, 21, America, ed. M. 0. Woodbume, pp. 153-210. Berkeley: University
1400-5. of California Press.
(1988). The rodents of the Lac Pellatier lower fauna, late Eocene (Duch- Tedford, R.H., Swinehart, J.B., Hunt, R. M. Jr., & Voorhies, M. R. (1985).
esneau) of Saskatchewan. Journal of Vertebrate Paleontology, 8, Uppermost White River and lowermost Arikaree rocks and faunas,
84-101. White River Valley, northwestern Nebraska, and their correlation
( 1990). Primates of the Lac Pellatier lower fauna Eocene, (Duchesneau), with South Dakota. Dakoterra, 2, 335-52.
Saskatchewan. Canadian Journal of Earth Sciences, 27, 520-4. Thomasson, J. R., Zakrzewski, J. R., Lagarry, H. E., & Mergen, D. C.
(1993 ). Additions to the mammalian paleofauna of Saskatchewan. Mod- (1990). Late Miocene (late early Hemphillian) biota from northwest-
ern Geology, 18, 475-87. ern Kansas. National Geographic Research, 6, 231-344.
(1994). A latest Chadronian (Late Eocene) mammalian fauna from the Tomida, Y (1987). Small mammal fossils and correlation of continental de-
Cypress Hills, Saskatchewan. Canadian Journal of Earth Science, posits, Safford and Duncan basins, Arizona, U.S.A. National Science
31, 1335-41. Museum, Tokyo, 1-141.
Storer, J.E., & Bryant, H. N. (1993). Biostratigraphy of the Cypress Hills Tomida, Y., & Butler, R. F. (1980). Dragonian mammals and Paleocene
Formation (Eocene to Miocene), Saskatchewan: Equid types (Mam- magnetic polarity, North Hom Formation, central Utah. American
malia: Perissodactyla) and associated fauna! assemblages. Journal Journal of Science, 280, 787-81 l.
of Paleontology, 67, 660-9. Torres, V. (1985). Stratigraphy of the Eocene Willwood, Aycross, and Wapiti
Stucky, R. K. (1984a). Revision of the Wind River Faunas, Early Eocene of Formations along the North Fork of the Shoshone River, north-central
central Wyoming, Part 5: Geology and biostratigraphy of the upper Wyoming. Contributions to Geology, University of ii'yoming, 23,
part of the Wind River Formation, northeastern Wind River Basin. 83-97.
Annals of the Carnegie Museum, 53, 231-94. Tsentas, C. (1981). Mammalian biostratigraphy of the middle Paleocene
(l984b). The Wasatchian-Bridgerian land mammal age boundary (early (Torrejonian) strata of the San Juan Basin: notes on Torreon Wash
to middle Eocene) in western North America. Annals of the Carnegie and the status of Pantolambda and Deltatherium fauna! "zones." In
Museum, 53, 347-82. Advances in San Juan Basin Paleontology, eds. S. G. Lucas, J. K.
Stucky, R. K., & Krishtalka, K. L. (1982). Revision of the Wind River Faunas, Rigby, Jr., & B. S. Kues, pp. 264-92. Albuquerque: University of
Early Eocene of central Wyoming, Part I: Introduction and multitu- New Mexico Press.
berculates. Annals of the Carnegie Museum, 53, 231-94. Turnbull, W. D. (1972). The Washakie Formation of Bridgerian-Uintan ages,
Stucky, R. K., Krishtalka, K. L., & Redline, A. D. (1990). Geology, vertebrate and the related fauna. In Guidebook: Field Conference on Tertiary
fauna, and paleoecology of the Buck Springs Quarries (early Eocene, Biostratigraphy of southern and western ii'yoming, ed. R. M. West,
Wind River Formation), Wyoming. Geological Society of America pp. 20-31.
Special Paper, 243, 169-86. Van Valen, L. (1978). The beginning of the age of mammals. Evolutionary
Sutton, J. F. (1977). Mammals of the Anceney Local Fauna (Late Miocene) Theory, 4, 45-80.
of Montana. Unpublished Ph.D. Thesis. Lubbock: Texas Tech Uni- Voorhies, M. R. (1973). Early Miocene mammals from northeast Nebraska.
versity. Contributions to Geology, University ofii'yoming, 12, 1-10.
Swisher, C. C. (1982). Stratigraphy and Biostratigraphy of the Eastern por- (1987). Fossil armadillos in Nebraska: the northernmost record. The
tion of Wildcat Ridge, Western Nebraska. Unpublished Ph.D. Thesis. Southwestern Naturalist, 32, 237-43.
Lincoln: University of Nebraska. ( l 990a). Vertebrate paleontology of the proposed Norden Reservoir area,
Tabrum, A. R., & Fields, R. W. (1980). Revised mammalian fauna! list for Brown, Cherry, and Keya Paha Counties, Nebraska. Technical Report
the Pipestone Springs Local Fauna (Chadronian, Early Oligocene), 82-09, Division of Archeological Research, pp. 1-593. Denver: U.S.
Jefferson County, Nebraska. Northwest Geology, 9, 45-5 l. Bureau of Land Reclamation.
 Appendix II: References for locality listings 665

(1990b). Vertebrate biostratigraphy of the Ogallala group in Nebraska. Wyoming, Part 6: The fauna and correlations of Bridger E. Milwau-
In Geologic Framework and Regional Hydrology: Upper Cenozoic kee Museum Publications, Contributions in Biology and Geology,
Blackwater Draw and Ogallala Formations, Great Plains, ed. T. 46, 1-8.
C. Gustavson, pp. 115-5 I. Austin: Bureau of Economic Geology, Westgate, J. W. (1990). Uintan land mammals (excluding rodents) from an
University of Texas. esturine facies of the Laredo Formation (Middle Eocene, Claibourne
Voorhies, M. R., & Corner, R. G. (1986). Megatylopus (?) cochrani (Mam- Group) of Webb County, Texas. Journal ofPaleontology, 64, 454-64.
malia: Camelidae). A re-evaluation. Journal of Vertebrate Paleon- (1992). Dinohyus aff. D. hollandi (Mammalia, Entelodontidae) in Al-
tology, 6, 65-75. abama. Journal of Paleontology, 66, 685-7.
Voorhies, M. R., Holman, J. A., & Xiang-Xu, X. (1987). The Hotell Ranch Westgate, J. W., & Emry, R. J. (1985). Land mammals of the Crow Creek
rhino quarries (basal Ogallala, medial Barstovian), Banner County, Local Fauna, Late Eocene, Jackson Group, St. Francis County,
Nebraska. Part 1: Geologic setting, fauna! lists, lower vertebrates. Arkansas. Journal of Paleontology, 59, 242-8.
Contributions to Geology, University ofii'yoming, 25, 55-69. Whistler, D. D. (1967). Oreodonts of the Tick Canyon Formation, southern
Wagner, H. M. ( 1976). A new species of Pliotaxidea (Mustelidae, Carnivora) California. Paleobios, I, 1-14.
from California. Journal of Paleontology, 50, 107-27. (1984). An early Hemingfordian (early Miocene) fossil vertebrate fauna
Wallace, R. E. (1946). A Miocene mammalian fauna from Beatty Buttes, from western Mohave Desert, California. Contributions in Science,
Oregon. Contributions to Paleontology, Carnegie Institute of Wash- Natural History Museum of Los Angeles County, 355, 1-36.
ington, 551, 113-34. Whistler, D. D., & Burbank, D. W. (1992). Miocene biostratigraphy
Walsh, S. L. (1991). Eocene Mammal Faunas of San Diego County. In and biochronology of the Dove Spring Formation, Mojave Desert,
Eocene Geologic History of San Diego, eds. P. L. Abbott & J. A. California, and characterization of the Clarendonian mammal age
May. Pacific Section of the Society of Economic Paleontologists and (late Miocene) in California. Bulletin of the Geological Society of
Mineralogists, 68, 161-78. Santa Fe Springs, Calif.: Comet Printing America, 104, 644-58.
Services. White, T. E. ( 1952). Preliminary analysis of the vertebrate fossil fauna of the
Webb, S. D. (1969a). The Burge and Minnechaduza Clarendonian mam- Boysen Reservoir area. Proceedings of the United States National
malian faunas of north-central Nebraska. University of California Museum, 102, 185-207.
Publications in Geological Sciences, 78, 1-191. (1954). Preliminary analysis of the fossil vertebrates of the Canyon Ferry
( l 969b ). The Pliocene Canidae of Florida. Bulletin of the Florida State Reservoir area. Proceedings of the United States National Museum
Museum, Biological Sciences, 14, 273-308. (Smithsonian Institution), 103, 395-438.
(1973). Pliocene pronghorns of Florida. Journal of Mammalogy, 54, Williamson, T. E. (1993). The Beginning of the Age of Mammals in the
203-21. San Juan Basin: Biostratigraphy and Evolution of Paleocene Mam-
(1983). A new species of Pediomeryx from the late Miocene of Florida, mals of the Naciemiento Formation. Unpublished Ph.D. Thesis. Al-
and its relationships within the subfamily Cranioceratinae (Rumi- buquerque: University of New Mexico.
nantia: Dromomerycidae). Journal ofMammalogy, 64, 261-76. Williamson, T. E., & Lucas, S. G. ( 1993). Paleocene vertebrate paleontology
Webb, S. D., & Hulbert, R. C., Jr. (1986). Systematics and evolution of of the San Juan Basin, New Mexico. In Vertebrate Paleontology in
Pseudhipparion (Mammalia, Equidae) from the late Neogene of the New Mexico, eds. S. G. Lucas & J. Zidek. Bulletin of the New Mexico
Gulf Coastal Plain and the Great Plains. Contributions to Geology, Museum of Natural History and Science, 2, 105-35.
University ofii'yoming, Special Paper 3, 237-72. Wilson, J. A. (1960). Miocene carnivores, Texas coastal plain. Journal of
Webb, S. D., MacFadden, B. J., & Baskin, J. A. (1981). Geology and pa- Paleontology, 34, 983-1000.
leontology of the Love Bone Bed from the late Miocene of Florida. ( 1967). Early Tertiary mammals. In Geology of Big Bend National Park,
American Journal of Science, 281, 513-44. eds. R. A. Maxwell, J. T. Lonsdale, R. T. Hazzard, & J. A. Wilson.
Webb, S. D., & Perrigo, S. C. (1984). Late Cenozoic vertebrates from Hon- University ofTexas Publications, 6711, 157-69.
duras and El Salvador.Journal of Vertebrate Paleontology, 4, 237-54. ( 1971 a). Early Tertiary vertebrate faunas, Vieja Group, Trans-Pecos,
Webb, S. D., & Tessman, N. (1968). A Pliocene vertebrate fauna from low Texas: Agriochoeridae and Merycoidodontidae. Bulletin of the Texas
elevation in Manatee County, Florida. American Journal of Science, Memorial Museum, 18, 1-83.
266, 777-81 l. (197lb). Early Tertiary vertebrate faunas, Vieja Group, Trans-Pecos,
West, R. M. ( l 973a). Geology and mammalian paleontology of theNewFork- Texas: Entelodontidae. Pearce Sellards Series, 17, 1-17.
Big Sandy area, Sublette County, Wyoming. Fieldiana (Geology), 29, (1974). Early Tertiary vertebrate faunas, Vieja Group and Buck Hill
1-193. Group, Trans-Pecos, Texas: Protoceratidae, Camelidae, Hypertrag-
(1973b). New records of fossil mammals from the early Eocene ulidae. Bulletin of the Texas Memorial Museum, 23, 1-34.
Golden Valley Formation, North Dakota. Journal ofMammalogy, 54, (1977a). Stratigraphic occurrence and correlation of Early Tertiary ver-
749-50. tebrate faunas, Trans-Pecos, Texas, Part I: Vieja Area. Bulletin of the
West, R. M., & Atkins, E. G. (1970). Additional middle Eocene (Bridge- Texas Memorial Museum, 25, 1-42.
rian) mammals from Tabernacle Butte, Sublette County, Wyoming. (l 977b ). Early Tertiary vertebrate faunas, Big Bend area, Trans-Pecos,
American Museum Novitates, 2404, 1-26. Texas: Brontotheriidae. Pearce Sellards Series, 25, 1-17.
West, R. M., & Dawson, M. R. (1973). Fossil mammals from the upper (1978). Stratigraphic occurrence and correlation of early Tertiary verte-
part of the Cathedral Bluffs Tongue of the Wasatch Formation (early brate faunas, Trans-Pecos, Texas, Part I: Vieja area. Bulletin of the
Bridgerian), northern Green River Basin, Wyoming. Contributions Texas Memorial Museum, 25, 1-42.
to Geology, University ofii'yoming, 12, 33-41. (l 984 ). Vertebrate faunas 49 to 36 million years ago and additions to the
(1975). Eocene fossil Mammalia from the Sand Wash Basin, northwest- species of Leptoreodon (Mammalia: Artiodactyla) found in Texas.
ern Moffet County, Colorado. Annals of the Carnegie Museum, 45, Journal of Vertebrate Paleontology, 4, 199-207.
231-53. (1986). Stratigraphic occurrence and correlation of early Tertiary ver-
(1977). Mammals from the Palaeogene of the Eureka Sound Forma- tebrate faunas, Trans-Pecos Texas: Agua Fria-Green Valley areas.
tion: Ellesmere Island, Arctic Canada. Geobios Memoir Special, I, Journal of Vertebrate Paleontology, 6, 350-73.
107-24. Wilson, J. A., & Schiebout, J. A. (1984). Early Tertiary vertebrate faunas,
West, R. M., & Hutchison, J. H. (1981). Geology and paleontology of Trans-Pecos, Texas: Ceratomorpha less Amynodontidae. Pearce Sel-
the Bridger Formation, southern Green River basin, southwestern lards Series, 39, 1-47.
 Appendix II: References for locality listings 666

Wilson, J. A., & Stevens, M. S. (1986). Fossil vertebrates from the lat- hyus (Tayassuidae). Bulletin of the American Museum of Natural
est Eocene, Skyline Channels, Trans-Pecos, Texas. Contributions to History, 141, 271-356.
Geology, University of Wyoming, Special Paper 3, 221-35. Woodburne, M. 0., & Golz, D. J. (1972). Stratigraphy of the Punchbowl For-
Wilson, R. W. (1984). The National Geographic Society - South Dakota mation, Cajon Valley, southern California. University of California
School of Mines and Technology expeditions into the Poleslide Mem- Publications in Geological Sciences, 92, 1-57.
ber of the Big Badlands of South Dakota in 1969: A program of Woodburne, M. 0., & Robinson, P. T. (1977). A new late Hemingfordian
conservation collecting. National Geographic Research Reports, I 0, mammal fauna from the John Day Formation, Oregon, and its strati-
637-42. graphic implications. Journal of Paleontology, 51, 750-57.
Winterfield, G. F. ( 1982). Mammalian paleontology of the Fort Union Forma- Woodburne, M. 0., Tedford, R.H., Stevens, M. S., & Taylor, B. E. (1974).
tion (Paleocene), eastern Rock Springs Uplift, Sweetwater County, Early Miocene mammalian faunas, Mohave Desert, California. Jour-
Wyoming. Contributions to Geology, University of Wyoming, 21, nal of Paleontology, 48, 6-26.
73-112. Woodburne, M. 0., Tedford, R.H., & Swisher, C. C., III. (1990). Lithostratig-
Walberg, D. L. (1979). Late Paleocene (Tiffanian) mammalian fauna raphy, biostratigraphy, and geochronology of the Barstow Formation,
of two localities in eastern Montana. Northwest Geology, 8, Mojave Desert, southern California. Bulletin of the Geological Soci-
83-93. ety of America, 102, 459-77.
Wood, H. E., & Wood, A. E. (1937). Mid-Tertiary vertebrates from the Wright, D. B., & Eshelman, R. E. (1987). Miocene Tayassuidae (Mammalia)
Texas Coastal Plain: Fact and fable. American Midland Naturalist, from the Chesapeake Group of the Mid-Atlantic Coast and their
18, 129-46. bearing on marine-nonmarine correlation. Journal of Paleontology,
Woodburne, M. 0. (1966). Equid remains from the Sonoma Volcanics, Cali- 61, 604-18.
fornia. Bulletin of the Southern California Academy of Sciences, 65, Wright, D. B., & Webb, S. D. (1984). Primitive Mylohyus (Artiodactyla:
185-9. Tayassuidae) from the late Hemphillian Bone Valley of Florida. Jour-
(1969). Systematics, biogeography, and evolution of Cynorca andDyseo- nal of Vertebrate Paleontology, 3, 152-9.
 Appendix III: Museum Acronyms

AM Amherst College Museum, Amherst, MA PMUM Paleontological Museum Uppsala, University of

AMNH Department of Vertebrate Paleontology, American Uppsala, Sweden
Museum of Natural History, New York, NY PU/YPM-PU Princeton University, now at Yale Peabody Mu-
ANSP Academy of Natural Sciences, Philadelphia, PA seum, New Haven, CT
ChM Charleston Natural History Museum, Charleston, SDSM South Dakota School of Mines, Rapid City, SD
SC SMU Southern Methodist University, Dallas, TX
CM Carnegie Museum of Natural History, Pittsburgh, PA SPSM St. Paul Science Museum, St. Paul, MN
CRL Collection Lemoine, Museum National d'Histoire TMM Texas Memorial Museum, University of Texas at
Naturelle, Paris, France Austin, TX
DMNH Denver Museum of Natural History UA University of Alberta, Edmonton, Alberta, Canada
F:AM Frick American Mammals, American Museum of UCMP University of California, Museum of Paleontology,
Natural History, New York, NY Berkeley, CA
FMNH Field Museum of Natural History, Chicago, IL UF University of Florida (Florida Museum of Natural
KUVP University of Kansas, Museum of Vertebrate Pale- History), Gainesville, FL
ontology, Lawrence, KS UM University of Michigan, Museum of Paleontology,
LACM/CIT Los Angeles County Museum, California Institute Ann Arbor, MI
of Technology collection, Los Angeles, CA UMVP University of Minnesota Vertebrate Paleontology,
MACN Museo Argentina de Ciencias Naturales, Buenos Minneapolis, MN
Aires, Argentina. UNSM University of Nebraska State Museum, Lincoln,
MCZ Museum of Comparative Zoology, Harvard Univer- NE
sity, Cambridge, MA UOMNH University of Oregon Museum of Natural History,
MHNH Museum National d'Histoire Naturelle, Paris, Eugene, OR
France USGS United States Geological Survey, Denver, CO
MPM Milwaukee Public Museum, Milwaukee, WI USNM United States National Museum, Washington,
NMC National Museum of Canada DC
NMMNH New Mexico Museum of Natural History, Albu- UT BEG Bureau of Economic Geology, University of
querque, NM Texas, Austin, TX (now housed with the collections
NMNH National Museum of Natural History, Smithsonian of the Texas Memorial Museum)
Institution, Washington, DC uw University of Wyoming, Laramie, WY
OMSI Oregon Museum of Science and Industry, Portland, Wa Walbeck Collection, Geologisch-Palaontologisch-
OR es, Institute, Halle, Germany
OUSM Stovall Museum, University of Oklahoma, Norman, YPM Peabody Museum of Natural History, Yale Univer-
OK sity, New Haven, CT

667
 Index

Boldface type is used throughout this index to indicate charts and illustrations.

Acarictis, 75-7, 97-8, 101 elrodi, 453 Alcidedorbignya, 281

ryani, 97-8, 108 giraffinus, 453, 454 Aletocyon, 80--1, 129, 134, 139
Aceratheriinae, 597, 600-1, 603 major, 453 multicuspis, 134
newgenus,516-7,600,603 proceras, 453 Aletodon, 255-6, 305, 306, 325
Aceratherium, 601, 604 robustus, 453 conardae, 306
Achaenodon,343-4,360,361,369,371,377 stocki, 453 gunnelli, 306
insolens, 370 sp., 454 mellon,306,319
robustus, 370 "Aepycamelus" priscus, 351-2, 452, 453, 459 quadravus,306
uintensis, 370 Agriochoeridae, 88, 337, 347-8, 353, 355, 406, Aletomerycinae, 355, 480, 481-2, 487
sp., 370 407,417 Aletomeryx, 156, 345, 349, 355, 467, 469, 478,
Achaenodontidae, 273, 338 Agriochoerinae, 406, 408-9, 417 479,480,481,482,483,487,488
Acinonyx, 237 new genus E 347-8, 406, 409, 417, 421 gracilis, 482
Aclistomycter, 343, 347-8, 405, 409-10, 417 new genus F 347-8, 406, 409, 417 lugni, 482
middletoni, 407, 409-10 Agriochoerus, 343, 347-8, 402, 404, 405, 406, marshi, 482
sp. indet., 410 408,409,416,417,418,421,422,425 marslandensis, 482
Aclistomycteriinae, 407, 409-10 antiquus, 403, 407, 408-9 occidentalis, 482
Acrocodia, 577 bullatus, 409 pristinus, 484
Actiocyon, 82-3, 149, 150 intermedius, 409 scotti, 482
Leardi, 149 latifrons, 408 sp., 482
Adapidium, 13, 269, 271, 272 matthewi, 409 Alforjas, 351-2, 453, 459, 460
Adelphailurus, 86-7, 237, 240, 241 maximus, 409 taylori, 454-5
kansensis, 239 minimus, 408 sp., 455
Aelurocyon, 82-3, 86, 152, 153, 155, 156, pumilus, 409 Allocyon, 84-5, 127, 178, 179-80, 186, 188,
166,167 transmontanus, 408 191
brevifacies, 156, 169 trifons, 409 loganensis, 179, 193
ferox, 156, 169 n.sp., 408-9 Allomeryx, 470-1
Aelurodon, 80--1, 88, 129, 136-7, 139, 140, 197 sp., 408-9 Allophaiomys, 26
ferox, 129, 136 Agriomeryx, 408-9, 421 A/lops, 529, 532
francisi, 136 Agriotherium, 84-5, 89, 175, 176, 177-8, 180, Alopecocyon, 128, 149
marshi, 136 182, 183, 184,186,187,192, 197,216,421, Alopex, 127
meandrinus, 216 422,423 Alostera, 299
saevus, 129 africanum, 184 Alticamelinae, 441
taxoides, 129, 136, 246 insignis, 184 Alticamelus, 447, 452, 453
wheelerianus, 136 intermedium, 184 Altilambda, 281
sp., 136 paradoxicum, 408 Altungulata, 249
Aeluroidae, 78, 114, 126, 127, 167 schneideri, 184, 194 Ambloctonus, 73, 75-7, 81, 100, 101
Aelurotherium, 106 sivalensis, 183-4 coloradensis, 106
Aenocyon, 127, 133 Aguascalientia, 351-2, 440, 446, 459 hyaenoides, 95, 106
Aepinacodon,342,345-6,381,382,384,385, wilsoni, 446 major, 95, 107
386,387 sp., 446 priscus, 95, 107
americanus, 385 Ailuraena, 245, 246 sinosus, 95, 106
defiectus, 385 Ailuridae, 74, 75, 77, 146, 148-9, 150, 208 Amblypoda, 250, 274
sp., 385 Ailurinae, 144, 145-6, 149, 150, 151 Amebelodon, 521, 522, 606, 607, 609, 611,
Aepycamelus, 348, 351-2, 440, 447, 450, 451; Ailuropoda, 145, 176 613,614,615,616-8,617
453-4,456,457,458,459,460 Ailurus, 148, 149, 151 britti, 522, 613, 617
alexandrae, 454 Albanosmilus, 231 fioridanus, 613, 614
bradyi, 453 Alces, 510 fricki, 613, 614

669
 Index 670

Amebelodon (cont.) Ancylopoda, 512, 513, 514 Aquascalientia, 445

hicksi, 613 Ancylotherium, 562, 563, 565 Archaelurus, 230
paladentatus, 613 Andrewsarchus, 293 Archaeohippus, 513--5, 519,540,544, 545,546,
sp., 613, 614 Anisonchinae, 313 547,554
Ammodon, 378 Anisonchus, 251, 252, 255-6, 299, 313, 314, blackbergi, 546
Amphibians, 82 323,324 equinanus, 544
Amphicaenopus, 516-7, 518, 595, 598, 599, athelas, 313 minimus, 546
602,603 dracus, 313 minutus, 546
platycephalus, 598-9 eowynae, 313 moumingi, 546
Amphicticeps, 153, 155, 178, 180 fortunatus, 313 penultimatus, 546
Amphictis, 82-3, 145, 146-7, 150, 153 gillianus, 313 stenolophus, 471, 472, 546
ambiguus, 146-7 oligistus, 313 ultimus, 545-6
antiquus, 147, 149 onostus, 313 sp., 546
sp., 146 sectorius, 313 Archaeolambda, 274, 276, 281
Amphicynodon, 127, 145, 176, 178, 179-80 willeyi, 313 Archaeomeryx, 341, 465, 467, 469
Amphicynodontinae, 78, 84-5, 127, 176, 178-82, sp., 313 Archaeoryctes, 319
186, 188, 193 Ankalagon, 253-4, 318, 325 Archaeotherium, 345-6, 375, 376-8, 379
Amphicyon, 84-5, 127, 197, 198, 200, 201, 202, saurognathus, 318 clavum, 378
203, 209, 210, 213, 214, 215, 217, 218-9, Anomerycoidodon, 412, 423 coarctatum, 378
226,380 Anoplotheriidae, 337, 338 crassum, 378
frendens,212,213 Anoplotherium, 440 latidens, 378
giganteus, 213 Anthracobune, 606 lemleyi, 378
ingens, 196, 199, 212, 213 Anthracotheriidae, 16, 88, 337, 338, marshi, 378
major, 213-4 339-40,381-8,624 mortoni, 376, 377, 378
steinheimensis, 212 biogeographic ranges, 342-4, 345-6, potens, 378
n.sp., 213 354-5 ramosum, 378
Amphicyonidae, 74, 75, 77-8, 82, 83, 84-5, 87, biology, 385, 387 scotti, 378
89, 127, 128, 196-227,624 defining features, 381-2 trippensis, 378
biochronology, 196, 197-8, 209 evolutionary patterns, 385-7 wanlessi, 378
biogeographic ranges, 84-5, 84-9, 209, 218-9 included North American genera, 384-5 sp., 378
biology, 216, 218-19 systematics, 382-4, Archaic ungulates, 73, 247, 255-6, 292-331
defining features, 198-200 temporal ranges, 386 biogeographic ranges, 250-3, 253-6
evolutionary patterns, 196, 197-8, 216-9 Anthracotheriinae, 383, 384, 386 biology, 322-4, 326
included North American genera, 202-16 Anthracotherium, 384 defining features, 294-7
localities of fossils, 222-7 Antiacodon, 339, 343-4, 360, 364, 371 evolutionary patterns, 322-6
systematics, 200-2 crassus, 362 included North American genera, 299-322
temporal ranges, 217 pygmaeus,359, 364,365 systematics, 297-9
Amphicyoninae,84-5, 87, 88, 127, 175, 196, 197, undescribed species, 364 temporal ranges, 325
199,200,201,202,203,209-16,217, vanvaleni, 364 Arctocyon, 253-4, 293, 295, 296, 303, 304,
218-9,218,219 venustus, 364, 366 323,324
Amphicyonine-amphicyonid carnivoran associa- Antiacodontinae, 342, 343--4, 364-6, 371 ferox, 294, 304
tion, 88 Antilocapra, 349, 491, 492, 494, 495, 497, 499, montanensis, 304
Amphicyonopsis, 214 500,501,503,504,505 mumak, 304
Amphimerycidae, 338, 340, 341 americana, 491, 492 primaevus, 304
Amphitragulus, 479 garciae, 499 sp., 304
Amynodon, 516-7, 583, 584, 585, 586, 587 Antilocapridae, 26, 338, 339, 341-2, 491-507, Arctocyonidae, 73, 74, 24 7, 248, 250, 251, 253-4,
advenus,584-5 595 255,257,275,293,295,298,300-5,322 -3,
reedi, 585 biogeographic ranges, 355, 346-6, 501 324
Amynodontidae, 583-8 biology, 501-5 Arctocyonides, 304
biogeographic ranges, 516-7, 516-8, 523 defining features, 491-3 Arctodus, 175, 176, 180, 182, 184, 185
biology, 587 evolutionary patterns, 501-5 Arctoidea, 74, 76, 77, 78, 80-1, 114, 126, 127,
defining features, 583-4 included North American genera, 495-500 144, 169
evolutionary trends, 587 systematics, 493-5 Eurasian immigration hypothesis, 176
included North American genera, 584-5, temporal ranges, 502 indeterminate, 128
587 Antilocaprinae, 355, 491, 492, 493, 494, 497, 502, problematical, 137-8, 139
systematics, 584 505 Arctonasua, 82-3, 150, 151
temporal ranges, 586 Apataelurus, 75-7, 96, 100, 101 eurybates, 148
Amynodontopsis, 516-7, 584, 585, 586, 587 kayi, 96, 107 fioridana, 145, 147-8
bodei, 585 Apheliscinae, 311 fricki, 148
Amynodonts, 16 Apheliscus, 109, 255-6, 257, 309, 311, 324 gracilis, 148
Anacodon,253-4,303-4,324 insidiosus, 311 minima, 148
nexus, 303-4 nitidus, 311 sp., 148
ursidens, 303-4 wapitiensis, 311 Arctonyx, 159
sp., 304 sp., 311 Arctostylopida, 247, 255-6, 332-6
Anancus, 613 Aphelops, 516-7, 520, 521, 523, 595, 596, 600, biogeographic ranges, 255, 257
Anatolostylops, 334, 335 601,602,603,604 biology, 335-6
Anchippodus, 251-2, 271, 272 malacorhinus, 600 defining features, 332
riparius, 271 megalodus, 600, 601, 604 evolutionary patterns, 335-6
Anchippus, 546 montanus, 601 included North American genera, 335
Anchitheriinae, 88, 513-5, 538, 539, 544-7, 554 mutilus, 600 systematics, 33 2--5
Anchitherium, 540, 544, 545, 553 sp., 600 Arctostylops, 255-6, 257, 335
Ancodon, 385 Apriculus, 343-4, 360, 361, 371, 384, 624 steini, 332-5
Ancodus, 384, 385 praeteritus, 368-9 Ardynictis, 319
 Index 671

Arfia,75-7,97, 100, 101, 107 trigonocomeus, 479, 484 lynchi, 412

junnei, 97, 107 sp., 484 Blickomylus, 347, 351-2, 457, 459
opisthotoma, 91, 97, 107 (see also Probarbouromeryx, galushai, 444, 445
shoshoniensis, 97, 107 Protobarbouromeryx) Blickotherium, 612
zele, 97, 107 Barylambda, 251-2, 257, 274, 278, 279, 281, 282 Bomburia, 309-10, 311
sp., 97 churchilli, 279 Boochoerus, 378
Arretotherium, 88, 344, 345-6, 381, 385, 386 faberi, 279 Bootherium, 66
acridens, 385 jackwilsoni, 279 Borissiakia, 563
fricki, 385, 387 sp., 279 Borocyon, 129
leptodus, 385 Barylambdidae, 276, 279, 282 Borophaginae, 80-1, 86-7, 88, 89, 124, 127, 128,
sp., 385 Barytheriidae, 606 129, 134-7, 138-40,139
Artiodactyla, 93, 247, 248, 249, 257, 273, 275, Basalina, 269-70, 272 Borophagus,67,80-1,89, 129, 139, 140
293,298,337,338-9,473,511,624 Bassaricyon, 148 diversidens, 137
biogeographic ranges, 342-356 Bassariscops, 80-1, 129, 134, 139 dudleyi, 137
Clarendonian chronofauna, 354-5 achoros, 134 pachyodon, 137
Eocene, 352-3 willistoni, 134 solus, 137
evolutionary patterns, 352-6 Bassariscus, 89, 82-3, 129, 147, 150, 151 sp., 137
ideas of interrelationships, 338 astutus, 144, 147 Bos, 66
Mio-Pliocene chronofauna, 355 casei, 145, 147 Bothriodon, 345-6, 381, 383, 384, 385, 386, 387
primitive, 338-9, 340-1 lycopotamicus, 147 advena, 384, 385
Running Water chronofauna, 353-4 ogallalae, 147 americana, 383
suiformes, 260, 267, 339-41, 342-3 parvus, 147 rostratus, 384, 385
systematics, 338-52 rexroadensis, 148 velaunus, 385
White River chronofauna, 353 sp., 147 sp., 385
Artiodactyla, Eocene bunodont and Bathmodon, 274 Bothriodontinae, 383, 385, 386
bunoselenodont, 358-74 Bathygeniinae, 407, 409 Bothriostylops, 334, 335
biogeographic ranges, 342, 343-4, 354-5 Bathygenys, 343, 347-8, 405, 409, 417, 418, 421, Bothrolabis, 392
biology, 370, 372 422,424 rostratus, 394
defining features, 359-60 alpha, 407, 409 siouxensis, 395
evolutionary patterns, 370-2 hedlundae, 409 socialis, 395
included North American genera, 361-70 reveesi, 405, 409 Bouromeryx, 349, 355, 480, 481, 487
systematics, 360-1 sp., 409 americanus, 485, 488
temporal ranges, 371 Bathyopsis, 251-2, 257, 285, 286, 287, 289, 290 madisonius, 485
Artionyx, 404, 405, 408-9, 421 fissidens, 287 milleri, 484-5
Asiagenes, 479, 481 middleswarti, 287 nebrascensis, 485
Asiocoryphodon, 280, 281 sp., 287 parvus, 485, 488
Asiostylops, 334, 335 Bathyopsoides, 285, 286, 287 pawniensis, 485
Astrohippus, 513-5, 521, 541, 548, 550, 551, 554 Beckia, 163 pseudonebrascensis, 485
albidens, 551 Bemalambda, 281 submilleri, 485
ansae, 551 Bemalambdoidea, 275 supemebrascensis, 485
curtivallum, ,548 Benthic foraminiferans, 15, 18, 20, 50 sp., 485
martini, 550 Bison,66,68,377,475,534 Bovidae, 73, 337, 338, 339, 341-2, 353, 355, 356,
sellardsi, 550 Blancocamelus, 351-2, 454, 455, 459 493,501, 508-10
stockii, 551 meadei, 454 biogeographic ranges, 353, 355
sp., 551 sp., 454 evolutionary patterns, 509-10
Auchenia, 454, 455 Blastomerycinae, 337, 341-2, 465-6, 468, included North American genera, 509
Aumelasia, 366 469-70,472-3,474 Bovoidea, 341-2
Australocamelus, 346, 351-2, 452, 459 biogeographic ranges, 345, 355-6, 353-4 Brachycrus,347-8,405,406,415,416,4 17,418,
orarius, 452 biology, 495 421,423,425
sp., 452 defining features, 465-6 buwaldi, 415
Auxontodon,343-4,360,364,365,371, 372 evolutionary patterns, 495 laticeps, 415, 418
pattersoni, 365 included genera, 492-3 rusticus, 415
processus, 365 systematics, 467, 469-70 sweetwaterensis, 416, 407
sp., 365 temporal ranges, 474 vaughani, 415-16
Aybelodon, 612 Blastomeryx, 349, 353-4, 463, 467, 472, 474, 483 wilsoni, 415
Azygonyx, 271 advena, 472 Brachyerix, 567
antilopinus, 483 Brachyhyops,342,345-6,375-7,378,3 79,624
Baioconodon,253-4,300,315,324 borealis, 478, 483 viensis, 380
antiquus, 315 elegans, 472 wyomingensis, 378-80
denverensis, 315 francesca, 472 sp., 380
Baluchitherium, 589 francescita, 471 Brachymeryx, 410, 421
Barbourochoerus,412,424 gemmifer, 469, 472 Brachyopsigale, 82-3, 160, 166, 168, 170
Barbourofelinae, 231-32, 233 marshi, 472, 482 dubius, 160
Barbourofelis, 79, 86-7, 88, 89, 169, 228, 233, medius, 472 Brachypotherium, 516-7, 520, 595, 601, 602,
236,241 mejferdi, 472 603,604
fricki, 228, 229, 231, 232, 234 mollis, 472 americanum, 601, 602
lovei, 232 olcotti, 472 brachypus, 601
morrisi, 232 primus, 472 Brachyprotoma, 159-60, 170
whitfordi, 231 vigoratus, 472 Brachypsalis, 82-3, 88, 155, 156-7, 161, 166,
sp., 232 wellsi, 473 167, 170
Barbouromeryx, 345, 347, 355, 479, 480, 481, sp., 472 angustidens, 157
484-5,487,488 Blickohyus, 347-8, 412, 417, 422, 423, 424, 459 hyaenoides, 157
cursor, 484 auritus, 412 matutinus, 157
 Index 672

Brachypsalis (cont.) jepseni, 278, 279 lupus, 197

modicus, 157 pattersoni, 278 sp., 133
obliquidens, 157 sp., 278 Canoidea, 114, 127
pachycephalus, 156-7 Caenolophus, 584, 587 Capromeryx, 353, 355, 494, 495, 498, 499, 500,
pristinus, 157 Caenopus,597,598,599 501,502,503
sp., 157 Cainotheriidae, 337, 338 arizonensis, 500
Brachypsaloides, 157 Calcareous nannofossils, 21 furcifer, 492, 500
Brachyrhynchocyon,84-5, 127,201-2,203,204, Calippus,513-5,521,541,547,550,554 gidleyi, 500
217,218-9,222,624 anatinus, 550 mexicanus, 500
dodgei, 200, 203-4 castilli, 550 minimus, 500
douglassi, 131 cerasinus, 550 minor, 500
montanus, 204 circulus, 550 tauntonensis, 500
sesoni, 132 elachistus, 550 sp., 500
Brasiliochoerus, 393 hondurensis, 550 Carcinodon, 253-4, 305, 324
Breameryx, 500 maccartyi, 550 aquilonius, 305
Bretzia, 355, 508--9 martini, 550 filholianus, 305, 316
pseudalces, 509 . optimus, 550 Cardiolophus, 515, 518--20, 542, 573, 579
Brevidontidae, 512 placidus, 550 radinskyi, 573
Broiliana, 145, 147 proplacidus, 550 semihians, 573
Brontops, 517, 518--20, 526, 527, 528, 529, 532, regulus, 550 Camassidentia, 114
533,534 sp., 550 Camivora, 73, 74, 248, 249, 252, 624
amplus, 532 Camelidae, 1, 337, 339, 340, 343-56, 439-62 incertae sedis, 82-3, 114, 119, 120, 165, 166,
bicomutus, 532 biogeographic ranges, 343-56, 351-2 167,624
brachycephalus, 532 · biology, 457-8, 460 body mass estimation, 168
dispar, 532 defining features, 440, 442-3 competition among, 110-1, 120
robustus, 532 evolutionary patterns, 457-60 Carnivora, early Cenozoic ("Miacoidea"),
selwynianus, 532 included North American genera, 441-57 110-23, 257
tyleri, 532 primitive, 88, 351-2, 441, 443-4 biogeographic ranges, 78--9, 80-3, 89
validus, 532 systematics, 440-1 biology, 119-20, 122
Brontotheriidae, 10, 15, 27, 511, 512, 525-36, temporal ranges, 459 defining features, 111, 112-3
595,624 Camelinae, 351-2, 441, 442-3, 452-3, 459 evolutionary patterns, 119-22
biogeographic ranges, 515-8, 518--20, 523 Camelini, 455-7, 459 included North American genera, 114-9
biology, 532, 534 Camelodon, 427, 441, 624 systematics, 111, 114, 115
defining features, 525-7 Camelomeryx, 434 temporal ranges, 121
evolutionary trends, 532, 534 Camelops, 351-2, 440, 447, 453, 454, 455, 458, Carnivoramorpha, 75
included North American genera, 529-32 459,460 Carnivorous archaic ungulates, 247, 248, 253-4,
systematics, 528-9 hestemus, 439, 455 257
temporal ranges, 533 huerfanensis, 455 Carodnia, 281
Brontotheriinae, 518--9, 525, 528, 530-2, 533 kansanus, 455 Carpocyon,80-1,88, 129, 138-40,139
Brontotherium, 532 minidokae, 455 cuspidatus, 136
Bryanictis, 78--9, 80, 82-3, 112, 113, 114-5, 115, sulcatus, 455 limosus, 13 6
120 traviswhitei, 455 sp., 136
microlestes, 114-5 vacondae, 455 Carsioptychus, 255-6, 299, 312-3, 325
vanvaleni, 116 sp., 455 coarctatus, 312
Bubogonia,252,255-6,309,325 Camels, 26, 67 hamaxitus, 312
saskia, 309 Camelus, 352, 455, 456, 457 matthewi, 312
Buisnictis, 159-60, 166, 168, 170 Campylocynodon, 145, 177, 179 Carya, 110
breviramus, 159-60 Canidae, 73, 74, 75, 77, 82, 84, 85, 86, 88, 89, Catagonus,345-6,351,393,394,395,396,397,
burrowsi, 160 102, 120, 124-43, 169, 176, 197, 199, 200, 398,400
meadensis, 159 203,205,218 brachydontus, 396-7
schoffi, 160 biogeographic ranges, 80--1, 82-9 metropolitanus, 396
sp., 160 biology, 138, 140-1 stenocephalus, 396
Bulbulodentata, 255-6, 308-15, 325 defining features, 124-6 wagneri, 389, 396, 399
Bunaelurus, 165 evolutionary patterns, 124, 138-41 sp., 396, 397
Bunodontia, 338, 339 included North American genera, 129-38 Cebochoeridae, 337, 341
Bunomerycidae, 339, 366-9 synonymy, 127-8 Cebochoerus, 416
Bunomeryx,343-4,358,360,361,367,368,371 systematics, 126-9 Cephalogale, 84-5, 87, 88, 127, 174, 180, 181,
elegans, 368 temporal ranges, 139 186,187, 191
montanus, 368 Caniformia, 74, 75, 76-8, 78, 111, 114, 115, geoffroyi, 180
Bunophorus,343-4,358,360,362,371,372 117-9, 127 minor, 181
cappettai, 362 Canimartes, 82-3, 163-4, 166, 168 ursinus, 181
etsagicus, 362 cumminsi, 163-4 n.sp. I, 194
gazini, 362 Caninae, 80--1, 87, 88, 89, 124, 127, 128, 129, n.sp. II, 194
grangeri, 362 132-4,139, 140-1 n.sp. III, 194
macroptemus, 362 Canis, 80--1, 89, 126, 127, 127, 130, 132, 133, Cephalophini, 73
pattersoni, 362 134, 136-8,139, 141 Ceratomeryx, 349, 355, 494, 495, 499, 500, 502
robustus, 362 condoni, 133 prenticei, 500
sinclairi, 362 cuspigerus, 206 Ceratomorpha, 511, 513, 514, 518-20, 571,
davisi, 133 573-6,579
Cadurcodon, 587 dirus, 246 Ceratotherium, 66, 377, 387
Caedocyon,80--1, 129, 131,139 familiaris, 133 Cercidiphyllum, 49
tedfordi, 131 latrans, 197 Cerdocyon,80--1, 89, 133, 134, 139
Caenolambda, 251-2, 278, 281, 282 lepophagus, 133 avius, 133
 Index 673

thous, 133 Coelodonta, 66 cerrosensis, 497

n. sp., 133 Collared lemming, 69 ilfonensis, 497
Cemictis, 82-3, 164, 166, 168 Colodon,516,518-20,572,575,576,578,579 furcatus, 492, 493, 497
hesperus, 164 cingulatus, 576 trilateralis, 497
sp., 164 dakotensis, 576 sp., 497
Cervavus, 483 hancocki, 576 Cotylopinae, 410-1
Cervidae, 26, 338, 339, 341-2, 493, 508 kayi, 576 Cotylops, 405, 411, 421
biogeographic ranges, 353, 355, 356 luxatus, 576 Cranioceras, 355, 478, 479, 480, 480, 481, 484,
evolutionary patterns, 509-10 occidentalis, 576 485,486,487,488,489,509
included North American genera, 509 procuspidatus, 576 clarendonensis, 485
Cervoidea, 339, 341-2, 493 stovalli, 576 dakotensis, 485
Cervalces, 510 woodi, 576 granti, 485
Cervus,483,496,510 sp., 576 kinseyi, 483
Cetacea, 247, 248, 249, 257, 299 Colonoceras, 574-5 mefferdi, 485-6
Cete,248,298,299,315-20,323,324 Coloreodon, 408-9, 421 pawniensis, 485
Cetungulata, 299, 315-20 Colpoclaenus, 253-4, 304, 315, 324 teres, 485
Chadronia, 434 keeferi, 304 unicomis, 483, 485-6
Chaenohyus,392,394 procyonoides, 304 n.sp., 485
Chalicotheriidae, 511, 560-8, 566 silberlingi, 304 sp., 485
Chalicotheriinae, 562-3 sp., 304 Cranioceratini, 355, 480, 484-6, 487
Chalicotherioidea, 512, 513, 560-8, 562, 569 Compositae, 49 Craterogale, 82-3, 158, 166, 168, 170
biogeographic ranges, 518--20, 520-1, 523 Conacodon,255-6,314,325 simus, 158
biology, 565, 567 cophater, 314 Creodonta, 73,74,91-109,248,257
defining features, 560-1 entoconus,295,314 biogeographic ranges, 75-7, 80-5, 89
evolutionary patterns, 565, 567 kohlbergeri, 314 biology, 100, 102
included North American genera, 562-5 utahensis, 314 defining features, 92-3
systematics, 561-2 sp., 314 evolutionary patterns, 100, 102
temporal ranges, 566 Conacodontinae, 314 extinction, 120
Chalicotherium, 562, 563, 565, 567 Condylarthra, 247, 275. See also Archaic ungu- included North American genera, 94-9
Chasmaporthetes, 86-7, 89, 245-6 lates non-creodont taxa described as, I 08-9
exilus, 243 Conepatus, 170 synonyms, I 06-9
johnstoni, 246 Conoryctella, 251-2, 261, 262, 263, 266 systematics, 93-4
ossifTagus,243,244,245,246 dragonensis, 262-3 temporal ranges, 101
sp., 246 pattersoni, 263 Cricetid rodents, 5, 26
Chasmotheriodes, 574 Conoryctes, 251-2, 261, 262, 263, 264, 266 Crocuta, 66, 243, 245, 246
Chasmotherium, 574-5 comma, 263 Cuon, 125, 127
Chelonocephalus sp., 410 Conoryctidae, 251-2, 253, 260, 265-7, 262-3, Cupressaceae, 47
Chelopoda, 513 266 Cuvieronius, 522, 608, 609, 612, 613, 617, 618-9
Chenopodiaceae, 49, 52 Copecion,255-6,257,320,321,322,325 hyodon,612
Chilotherium, 604 brachyptemus, 321 sp., 612
Chiroptera, 249 , davisi, 321 Cuyamacamelus, 351-2, 447, 448-50, 459
Choeroclaenus, 255-6, 309, 311, 325 sp., 321 jamesi, 448-9
turgidunculus, 309 Copemys, 5 Cyclopidius, 410, 421
Choerodon,342,345-6,378,379 Cordillerion, 612 Cymbalophus, 542
caninus, 378 Cormocyon,80--1,86-7,86, 129, 131, 135, 137, Cynarctinae, 129
sp., 378 138,139 Cynarctoides, 80--1, 87, 88, 128, 129, 134, 138,
Choeromeryx, 384 copei, 135 139
Choeropotamids, 341 sp., 135 acridens, 134
Choeropotamus, 416 Cormohipparion, 513-5, 521, 541, 54 7, 548, 549, cuspidatus, 134
Chorotherium, 408, 421 554,555 mustelinus, 134
Chriacus, 253-4, 257, 275, 292, 297, 301-2, emsliei, 549 sp., 134
322-3,324 goorisi, 547 Cynarctus,80--1,88, 129, 134, 136, 138-40,139
badgleyi, 302 ingenuus, 549 crucidens, 135
baldwini, 302, 303 occidentale, 549 fortidens, 135
calenancus, 302 plicatile, 549 saxatilis, 135
gallinae, 302 sphenodus, 549 sp., 135
hyattianus, 302 sp., 549 Cynelos, 84-5, 197, 199, 201, 202, 209, 211-2,
katrinae, 302 Coryphodon,251-2, 257,274-5,276,280,281, 213,2!4,215-6,217,218-9,225
metacometi, 302 282 bohemicus, 212
oconostotae, 302 anthracoideus, 280--1 caroniavorus, 211-2
orthogonius, 302 eocaenus, 280-1 crassidens, 212
pelvidens, 301-2 lobatus, 281 helbingi, 212
schlosserianus, 302 molestus, 281 idoneus, 211-2
truncatus, 302 oweni, 281 lemanensis, 211-2
n. sp., 302 proterus, 281 piveteaui, 212
sp., 302 simus, 280-1 rugosidens, 212
Chrysocyon,80--1,89, 134,139 subquadratus, 281 schlosseri, 212
brachyurus, 134 Coryphodontia, 275 sinapius, 211-2
n. sp., 134 Coryphodontidae, 276, 279-81, 282 steinheimensis, 216
Cimolestes, 74, 93, 261, 320 Cosoryx,349,355,485,493,494,495,496,497, n.sp. I, 211
Cimolestidae, 624 502,503 n.sp. II, 211
Claenodon, 300, 302, 304, 323 agilis, 497 sp., 211
Coatis, 144 burgensis, 497 Cynocephalus, 277
 Index 674

Cynodesmus,80-1, 127, 129, 131,139 Desmatippus, 513-5, 520, 540, 546, 554 Didymictidae, 114, 115
brachypus, 131, 132 avus,546 Didymictis, 73, 74, 78-9, 81, 109, 112, 113, 114,
canavus, 135 crenidens, 546 l!5-6, 120
cooki, 13], 135 integer, 546 altidens, 116
cuspidatus, 136 nebrascensis, 546 protenus, 115-6
euthos, 135 texanus, 546 microlestes, 114-5
iamonensis, 131, 132 tyleri, 546 tenuis, 116
martini, 131 sp., 546 vancleveae, 116
minor, 135 Desmatochoerinae, 412-3 sp., 116
nobilis, 132 Desmatochoerus, 412-3, 414, 424 Didymoconidae, 624
thompsoni, 135 Desmatoclaenus, 253--4, 301, 324 Didymoconus, 319
thooides, 131 dianae, 301 Dilophodon, 516, 518-20, 572, 574, 575-6, 579
vulpinus, 131, 133 hermaeus, 301 leotanus, 576
Cynodictis, 127, 128, 130, 137-8, 154, 198, 202, mearae, 301 minusculus, 572, 574, 576
203,206,209,219,206 paracreodus, 301 sp., 574, 576
Cynoidea, 76--8, 114, 126, 127 protogonioides, 301 Dinaelurus, 86-7, 228, 229, 231, 233, 234
Cynorca proterva, 393, 396 sp., 301 crassus, 231
. "Cynorca" occidentale, 345-6, 390, 395, 396, Desmatotherium, 512, 516, 518-20, 572, 574, Dinictis, 86-7, 102, 228, 230, 232, 233, 234
398 575,576,579 bombifrons, 230
"Cynorca" sociale, 345-6, 390, 391, 392, 393, guyotti, 574 cyclops, 230
394,395,398,399 intermedius, 574 eileenae, 230-1
Cyonasua, 147-8, kayi, 576 felina, 230
Cyriacotheriidae, 251-2, 27 5 woodi, 576 fortis, 230
Cyriacotherium, 251-2, 255, 257, 275, 276, 624 Desmostylia, 248, 250 mega/aides, 230
argyreum, 277 Deuterogonodon, 253-4, 301, 302, 324 paucidens, 231
psammium, 277 montanus, 302 platycopsis, 230
sp., 277 noletil, 302 sp., 230
sp., 302 Dinobastis, 237, 241
Daeodon, 378 Diacodectidae, 338, 339-40, 355 Dinocerata,250,256,257,274,275,277,284-91
Daphoenictis, 84-5, 201-2, 204, 204, 217, 218-9, Diacodexis, 342, 343-4, 360-1, 362, 370, 371 biogeographic ranges, 251-8, 251-2
222 antunesi, 362 biology, 288, 290
tedfordi, 203 gazini, 362 defining features, 284
Daphoeninae, 84-5, 87, 196, 197, 199, 200, gracilis, 361 evolutionary patterns, 288-90
201-7,217,218-9,219 ilicus, 361, 372 included North American genera, 286--8
Daphoenocyon,201-2,203,624 metsiacus, 360 systematics, 285-6
Daphoenodon,84-5,87,88, 197,198, 199,202, minutus, 361 temporal ranges, 289
203,206--7,211,217,218-9,219,225 pakistanensis, 339, 362 Dinocyon, 127, 180, 216
periculosis, 206 secans,359,360-2,363 Dinofelis, 86-7, 89, 237, 239, 240, 241
robustum, 207 secans-kelleyi, 361-2, 364 abeil,239
superbus,200,206-7 secans-metsiacus, 361 palaeoonca, 239
n.sp. I, 207 secans-primus, 360, 372 Dinogale, 82-3, 162, 166, 168, 170
n.sp. II, 207 secans-secans, 361, 362 siouxensis, 162
n.sp. III, 207 varleti, 362 Dinohippus, 513-5, 521, 541, 549, 550, 551, 554
Daphoenus,82,84-5, 128, 197,201-2,203,204, sp. A, 361 edensis, 551
205,208,217,218-9,219,222,223,224 sp. B, 361 interpolatus, 551
hartshomianus, 200, 204-5 "Diacodexis" woltonensis, 343--4, 361, 363, 371 leidyanus, 551
lambei, 205 Diatrymiform birds, 73, 81 mexicanus, 551
minimus, 206 Diceratheriinae, 597, 599, 603 osbomi, 551
socialis, 205 Diceratherium, 516-7, 518, 520, 521, 584, 595, pachyops, 551
transversus, 205 596,597-8,599,600,601,602,603 subvenus, 551
vetus, 198, 200, 204-5 advenum, 584 sp., 551
n.sp., 204-5 annectens, 599 Dinohyus, 88, 344, 345-6, 377, 378, 379
Daptophilus, 230 armatum, 599 hollandi, 376, 378
Dasyuroidea, 73 cooki, 595, 600 mento, 378
Dawn redwood, 49, 53 jamberi, 601 sp., 378
Dayohyus, 412, 421 matutinum, 599 Dinotomus, 231
Deinotheriidae, 606 niobrarense, 599 Diplacodon,518-20,527,528,531,533
Delahomeryx, 353--4, 469, 473, 474, 624 tridactylum, 599 progressum, 526, 528, 531
browni, 473 n. sp., 599 Diplarthra, 275
Delphodontines, 275 sp., 599 Diplobune, 405
Deltatheridia, 74 Dicerorhinus, 604 Diplobunops, 405, 408-9, 421, 422
Deltatheridiidae, 74, 251-2 Diceros, 66, 521, 604 Diplobunosinae, 408-9
Deltatheridium, 275 Dichobune, 358 Diploceras, 531
Deltatherium, 251-2, 253, 254, 275, 276, 293, Dichobunidae, 338, 340 Diploclonus, 529, 532
624 Dichobunids (see Artiodactyla, Eocene) bun- Dipoides, 66
fi.mdaminis, 275 odont and bunoselenodont Dipsalidictis, 73, 75-7, 80, 94-5, 100, 101, 106
Deperetellidae, 512 Dicotyles, 389, 391, 393, 395, 396, 397, 399-400 aequidens,95, 106
Dermoptera, 275-6, 624 Dicrocerus, 485 amp/us, 106
Desmathyus, 392, 394 Dicrostonyx, 69 amp/us, 106
brachydontus, 393 Didelphis, 68, 265 krausei, 95, 106
pinensis, 395 Didelphodontines, 250 platypus, 94-5, 106
vagrans, 393 Didelphodus, 268, 276 transiens, 95, 106
validus, 395 Didolodontidae, 293 sp., 95
 Index 675

Dipsalodon, 73, 75-7, 95-6, 101, 107 Earendil, 311-2 Entelodontidae, 16, 86, 88, 337, 338, 339-40,
churchillorum, 96, 107 undomiel, 312 375-80
matthewi, 95, 107 Ectocion, 253, 255, 255-6, 258, 320, 321-2, 325 biogeographic ranges, 342-55, 345-6
n. sp., 107 cedrus, 322 biology, 380
sp., 96 co llinus, 322 defining features, 375-7
Discritochoerus, 343-4, 360, 369, 370, 371 ignotum, 322 evolutionary patterns, 380
lapointensis, 370 major, 322 included North American genera, 378, 380
Dissacus, 253-4, 255, 257, 258, 317-18, 318, mediotuber, 322 systematics, 377-8
319,324,326 montanensis, 322 temporal ranges, 379
argenteus, 318 osbomianus, 296, 321-2 Entelodontoidea, 341
leptognathus, 317 parvus, 322 Entomodon, 108
navajovius, 318 silberlingi, 322 Eobasi/eus,251-2,285,286,287-8,289,290
p raenuntis, 318 superstes, 322 comutus, 287-8
saurognathus, 318 wyomingensis, 322 Eoconodon,252,253--4,316--17,324
serior, 318 sp., 322 comma, 316
willwoodensis, 318 Ectoconus,255,299,312,323,326,325 copanus, 317
sp., 318 ditrigonus, 312 coryphaeus,316--7
Dolichohippus, 67, 552 symbolus, 312 gaudrianus, 317
Dolichorhininae, 525, 528, 530, 533 sp., 294, 297, 312 nidhoggi, 317
Dolichorhinoides, 530 Ectoganus, 73, 251-2, 255-6, 257, 258, 261, 263, sp., 317
Dolicorhinus, 518-19, 530 265,266 Eoentelodon, 377
Doliochoerus, 392 copei, 265 Eoequus, 549-50
Dorcameryx, 498 gliriformis, 265 Eohippus, 542
Dorcatherium, 440
lobdelli, 265 Eomellivora, 82-3, 162, 163, 166, 168, 170
Dorraletes, 255-6, 305, 306, 325 simplex, 265 wimani, 163, 169
diminutivus, 306
sp., 265 sp., 163
Dracoclaenus, 310
Ectopocynus, 80-1, 86, 129, 130-1, 139 Eomeryx, 405, 408-9, 422
Drassonax, 179
antiquus, 130 Eometarhinus, 529
Dremotherium, 342
intermedius, 130 Eomoropidae, 516, 566
Drepanodon, 231
simplicidens, 130-1 Eomoropus,518-20,562,563,566,567,578
Drepanomeryx, 355, 478, 480, 487, 479, 482-3
Edaphocyon,82-3, 147,150 amarorum, 563, 567
488
lautus, 147 annectens, 563
falciformis, 482-3, 488
pointblankensis, 147 minimus, 563
sp., 483
(see also Matthomeryx) Edentata, 26, 249, 261 quadridentatus, 563
Dromocyon, 318 Egmoiteptecela, 231 Eomyids, 16, 595
Dromomerycidae, 26, 196, 337-8, 339, 341-2, Eira, 162, 164 Eotitanopinae, 529, 533
477-90,501,595 Elaeagnaceae, 49 Eotitanops, 515, 518-20, 525, 526, 527, 528, 529,
biogeographic ranges, 345-56, 355 Elasmognathus, 577 533
biology, 486, 488-9 Elephantidae, 511, 606 borealis, 529
defining featu~es, 477-8 Elephantoidea, 609 brownianum, 529
evolutionary patterns, 486--9 Elephas, 606 gregoryi, 529
included genera, 481-6 Ellipsodon, 255-6, 309-10, 325 major, 529
systematics, 478-81 grangeri, 310 minimus, 529
temporal ranges, 487 inaequidens, 309-10 princeps, 529
Dromomerycinae, 355, 480, 482-6, 487 priscus, 310, 311 Eotitanotherium, 518-20, 526, 527, 528, 531, 533
Dromomerycini, 355, 480, 482-4, 487 witkoi, 310 osbomi, 531
Dromomeryx, 349, 355, 478, 479, 480, 482, 483, yotankae, 310 Eotylopus,349-50,426,427,429,430
484,487 sp., 310 reedi, 427, 428
americanus, 485 Elomeryx, 342, 345-6, 385, 386, 387 n. sp. A, 428
antilopina, 483 armatus,381,382,385 n. sp. B, 428
borealis, 483-4, 478 garbantii, 385 Ephedra,45,49,52
cursor, 484 sp., 385 Ephyrachyus, 574, 575
madisonius, 485 Elotherium, 377, 378, 384 Epichriacus, 301
parvus, 485 Embrithopoda, 248 Epicyon, 80-1, 88, 129, 136--7, 139
pawniensis, 478, 484 Engelhardia, 49 aphobus, 136
texanus, 486 Enhydra, 167 diabloensis, 136
trigonocomeus, 484 Enhydritherium, 82-3, 161-2, 166, 168, 170 haydeni, 136, 140
whitfordi, 478, 484, 488 terraenovae, 161-2 injl.atus, 136
sp., 484 sp., 162 littoralis, 136
(see also Subdromomeryx) Enhydrocyon,80-1,86, 102, 127, 129, 130, 132, mortifer, 136
Duchesnehippus, 543 138, 139 saevus, 124, 136--7, 140
Duchesneodus, 517, 518-20, 525, 526, 528, 531, basilatus, 13 2 validus, 137, 140
533,534 crassidens, 132 sp., 137
primitivus, 531 geringensis, 132 Epigenetochoerus, 414, 422
uintensis, 531 oregonensis, 132 Epihippus, 14,513-5,516,538,540,543,554
Dusicyon, 133 pahinsintewakpa, 132 gracilis, 543
Dyseohyus,344,345-6,395,398 sectorius, 132 intermedius, 543
fricki, 393, 396 stenocephalus, 132 parvus, 543
serus, 392 venator, 131 uintensis, 543
stirtoni, 396 wallovianus, 131 sp., 543
Dyseomeryx, 482, 483 sp., 132 Epitriplopus, 516-7, 591-3, 592
Dyseotylopus, 444 Entelodon, 376, 377, 384 uintensis, 591-3
 Index 676

Eporeodon, 343, 347-8, 412-3, 417, 421, 422, montanus, 412 Gashatostylops, 333, 335
423,424,425 socialis, 412 Gazinocyon, 96, 107
condoni, 412 trigonocephalus, 407, 412 vulpeculus, 96, 107
davisi, 412 Eudinoceras, 280, 281 Gelocidae, 341, 463-76, 465, 469, 471-2, 474
jacksoni, 412 Euoplocyon,80-1,86, 127, 129, 136,139 biogeographic ranges, 350, 355, 356
major, 412-3 praedator, 136 biology, 475
meagherensis, 412 spissidens, 136 defining features, 465
minor, 413 Eupantodonta, 275, 276 evolutionary patterns, 4 75
occidentalis, 412-3 Eupecora, 342 included North American genera, 471-2
perbullatus, 412 Euprotogonia, 320 systematics, 467, 469
re/ictus, 413 Euryboas, 245 temporal ranges, 474
thurstoni, 413 Eusmilus, 86-7, 228, 229, 231, 232, 233, 234 Gelocus, 341
trigonocephalus, 414 belli, 231 Genetochoerus, 411, 412, 422
sp., 413 bidentatus, 231 Gentilicamelus, 351-2, 444, 447, 448, 457, 459
Eporeodontinae, 412-3 cerebra/is, 231 sternbergi, 447, 451, 458
Equidae, 337, 511, 512, 537-59 dakotensis, 231 Gidleyina, 321
biogeographic ranges, 513-5, 515-23 olsontau, 231 Gigantocamelus, 351-2, 447, 456, 457, 459
biology, 553-6 sicarius, 231 spatula, 457
defining features, 537-9 whitfordi, 231 sp., 457
evolutionary trends, 554--6 sp., 231 Gillisonchus, 252, 255-6, 299, 313, 323, 325
included North American genera, 541-51, Eutheria, incertae sedis, 623-4 gillianus, 313
553-4 biogeographic ranges, 251-2 Giraffidae, 338, 339, 345, 347, 348, 350, 493, 511
systematics, 539-41 Giraffoidea, 347
temporal ranges, 554 Felid-ursid-canid carnivoran association, 88 Glyptostrobus, 41
Equinae,540,547-53,554 Felidae, 73, 74, 75, 79, 86-7, 88, 120, 126, 127, Gnathabelodon, 522, 608, 611, 612, 616, 617
Equini, 513-5, 540, 549-52, 554 169, 199,218,236-42 buckneri, 612
Equisetum, 110 biogeographic ranges, 86-7, 88-9 thorpei, 612
Equoidea, 512 biology, 241 Gobiatherium, 285-6, 290
Equus,67,513-5,521,523 ,538,541,545,550, defining features, 236-7 Gobicyon, 129, 208
551,552,554,555,556 evolutionary patterns, 241 Gobiohyus, 358, 369, 383-4
achates, 552 included North American genera, 238-41 Gomphotheres, 67
cabal/us, 552 systematics, 23 7 Gomphotheriidae, 521, 522, 606, 609, 610-5, 617
calobatus, 552 temporal ranges, 240 Gomphotherium, 522, 607, 608, 609, 610--1, 612,
cumminsi, 552 Feliformia, 74, 75, 76, 78-9, 111, 114--17, 127 613,614,615,617,618-9
giganteus, 552 Felinae, 86-7, 238-9, 240 anguirivalis, 611
idahoensis, 552 Felis, 100, 237, 238-9, 241, 246 angustidens, 610
laparensis, 550 Feloidea, 74, 78, 79, 86-7, 88, 89, 114, 127 calvertensis, 611
occidentalis, 552 Fennecus, 127 cimarronis, 611
pons, 552 Ferae, 74, 110, 111, 250 fricki, 611
scotti, 552 Ferinestrix, 82-3, 166, 168, 170 giganteum, 611
shoshonensis, 552 vorax, 164 gregori, 611
simplicidens, 552, 556 Fimbrethil, 314 nebrascensis, 611
tau, 552 Fissipeda, 75, 114 obscurum, 611
sp., 552 Floridaceras, 516-7, 520, 595, 596, 600, 603 osborni, 611
Eremotherium, 68 whitei, 600 perfectum, 611
Erethizon, 68 Floridachoerus, 345-6, 392, 394, 395, 398 phippsi, 611
Escatepos, 255-6, 314, 324 olseni, 395 productum, 611
campi, 314 sp., 395 republicanum, 611
Esthonychidae, 269-71, 272 Floridatragulinae, 351-2, 440, 441, 442-3, simpsoni, 611
Esthonychinae, 270-1, 272 445-7,459,460 willistoni, 611
Esthonyx, 251-2, 257, 269, 270, 271, 272, 273 Floridatragulus, 346, 348, 350, 351-2, 354, 440, n.sp., 611
acer, 270 439,445,446-7,449,458, 459 sp., 607, 611
acutidens, 271 barbouri, 446 Goniacodon,252,253-4,3 16,324
ancylion, 270--1 dolichanthereus, 446 crassicuspis, 316
bisulcatus, 270--1 hesperus, 446 hiawathae, 316
burmeisterii, 270 nanus,446 levisanus, 316
grangeri, 271 texanus, 446 sp., 316
gunnelli, 271 sp., 446, 460 Grangeria, 518-20, 560, 562, 563, 566, 567, 578
latidens, 271 Forstercooperia, 516-7, 589, 590-1, 592 anarsius, 563, 567
spatularius, 271 grandis, 591 canina, 560, 563
xenicus, 270-1 minuta, 591 Gregorychoerus, 412, 424
sp., 271 totadentata, 590 Griphippus, 547
Eubelodon,522,608,611, 611,618 Fouchia,518-20,572,574 ,579 Gula, 149, 162, 164, 169, 170
morrilli, 611 elyens is, 57 4 Gypsonictops, 296
Eubrontotheres, 526, 528, 533
Euceratherium, 66, 67, 68 Galecynus, 137-8 Hadrocyon, 129, 216
Eucrotaphinae, 407, 411-2, 417 Galecyon,15-7,98,101, 108 Hadroleptauchenia, 410, 422
Eucrotaphus, 347-8, 412, 417, 421, 422, 423, mordax, 98, 108 Hantkenina, 16
424,425 Calera, 164 Hapalodectes, 248, 253-4, 257, 317, 326, 324
auritus, 412 Galictines, 89 anthracinus, 317
dickinsonensis, 412 Galictini, 154, 164--5, 166, 170 compressus, 317
helenae, 412 Galictis, 164 leptognathus, 296, 317
jacksoni, 405, 412 Ganatherium, 285 sp., 317
 Index 677

Hapalodectidae, 248, 253-4, 317-8, 324 cookei, 108 Hidrosotherium, 353-4, 473, 474, 624
Haplaletes, 255-6, 305, 306, 325 grandis, 98, 108 transpecoensis, 473
andakupensis, 306 viejaensis, 98, 108 Hipparion, 16, 183, 513-4, 521, 541, 547, 547,
diminutivus, 306 sp., 98 548,549,554
disceptatrix, 295, 306 Hemithlaeus, 252, 255-6, 312, 314, 325 forcei, 548
pelicatus, 306 kowalevskianus, 312 lenticulare, 548
serior, 306 sp., 312 lenticularis, 549
sp., 306 Hendryomeryx, 352-4, 469, 471, 474 mohavense, 549
Haploconus, 255-6, 299, 313, 314, 325 defordi, 471 prostylum, 548
angustus, 313 esulcatus, 471 sanfondensis, 547, 548
corniculatus, 312, 313 wilsoni, 471 shirleyae, 548
elachistus, 313 sp., 471 tehonense, 548
entoconus, 314 Heptacodon, 345-6, 381, 384, 385, 386, 387 trampasense, 548
gillianus, 313 curtus, 382, 384 sp., 548
inopinatus, 313 gibbiceps, 384 Hipparionini, 513-5, 540, 547-549, 554
lineatus, 313 occidentale, 384 Hippidion, 513-5, 541, 551-3, 554
sp., 313 pellionis, 384 bonaerense, 551
Haplocyon, 208 quadratus, 384 neogaeum, 551
Haplocyoninae, 197, 199, 201, 208 n. sp., 384 principale, 551
Haplohippus, 513-5, 517, 538, 543, 554 sp., 384 sp., 552
texanus, 543 Heptodon, 514, 515, 516, 518-20, 570, 571, Hippomorpha, 512, 513, 514
Haplolambda, 251-2, 278, 281, 282 573-4,575,578,579 Hippopotamidae, 339-340
quinni, 279 brownorum, 574 Hippopotamus, 66, 340, 577
simpsoni, 279 calciculus, 573-4 Hippotherium, 548, 549-50
Haplomastodon, 612, 613 posticus, 574 Hippotigris, 550
Haplomylus, 255-6, 257, 305, 306, 308, 325 ventorum, 574 Homacodon, 343-4, 360, 361, 366, 367, 369, 371
simpsoni, 306 sp., 574 priscus, 362
speirianus, 306 Herpestidae, 73, 74, 75, 79 pucillus, 362
sp., 306 Hesperhys-"Cynorca" sociale clade, 394-5, 398 vagans,358,359,364,366 -7
Harpagolestes, 253-4, 257, 258, 318-9, 323, 324, Hesperhys-Tayassu clade, 394-7 n. sp. A, 367
326 Hesperhys, 344, 345-6, 347, 391, 392, 393, 395, n. sp. B, 367
brevipes, 319 397,399 n. sp. C, 367
immanis, 319 pinensis, 394, 395 n. sp. D, 367
leotensis, 319 vagrans, 395 Homacodontidae, 338, 339, 340, 343-4, 366-9,
macrocephalus, 318-9 sp., 395 371
uintensis, 294, 319 Hesperocamelus, 351-2, 452, 453, 459 new genus A 343-4, 366, 371
sp., 319 alexandrae, 453, 454 new genus B 343-4, 368, 371
Harpyodus, 276, 281 stylodon, 453 Homo, 66,68
Hayoceros,494,495,497,4 99 sp., 453 Homogalax, 515, 518-20, 538, 539, 572-3, 578,
Hecubides, 212 Hesperocyon, 80-1, 82, 127, 129, 130, 139, 147, 579
Hedgehogs, 16 , 178, 179 aureus, 573
Helaletes, 515, 518-20, 569, 570, 572, 573, 574, coloradensis, 130, 138 bridgerensis, 573
575,576,579 gregarius, 125, 126, 130, 131, 138 primaevus, 573
boops,575 lippincottianus, 130 protapirinus, 572-3
intermedius, 574 paterculus, 130 semihians, 573
latidens, 573 pavidus, 130, 138 uintensis, 573
nanus,510,575 temnodon, 130, 131 sp., 573
sp., 575 wilsoni, 130 Homotherium, 86-7, 89, 237, 237, 239-41, 240
Helarctos, 183 sp., 130 crenatidens, 239
Heloderma, 460 Hesperocyoninae, 80-1, 84, 86, 87, 88, 124, 127, crusafonti, 239
Helohyinae, 337, 339, 343-4, 360, 369-70, 128, 129, 130-2, 138,139 idahoensis, 239
371 Hesperohipparion, 548 ischyrus, 239
new genus,343-4, 370,371 Hesperohippus, 548 johnstoni, 239
Helohyus, 343-4, 360, 361, 366, 368-70, Heteraletes, 518-20, 572, 575-6, 579 sp., 241
371, 378 leotanus, 575-6 Hoplictis, 82-3, 163, 166, 168, 170
lentus, 369-70 Heterocoryphodon, 280, 281 jl.orancei, 163
milleri, 359, 369 Heteromeryx, 349-50, 431, 432, 433, 434, 435, grangerensis, 163
plicodon,359,369-70 436,437 sp., 163
validus, 369 dispar, 434 Hoplitomeryx, 338
Hemiauchenia, 348, 351-2, 452, 453, 454, 455, Heterotitanops, 530 Hoplophoneus, 86-7, 102, 169, 228, 229, 230,
459,460 Hexacodus, 343-4, 360, 361, 362, 365, 366, 370, 231, 232, 233
blancoensis, 454 371,372 belli, 231
macrocephala, 439, 454 pelodes, 359, 362, 366, 367 cerebra/is, 231
minima, 454 uintensis, 366, 369 dakotensis, 231
paradoxa, 454 sp., 366 insolens, 231
vera, 454 Hexameryx, 355, 494, 495, 499, 501, 502, 504 latidens, 231
sp., 454 elmorei, 499 marshi, 231
Hemicyon, 127, 174, 180, 181, 182, 380 simpsoni, 499 mentalis, 231
Hemicyoninae, 84-5, 87, 88, 102, 174, 175, 176, Hexaprotodon, 281 molossus, 231
177,186, 188, 197,200 Hexobelomeryx, 355, 494, 495, 499, 501, 502, occidentalis, 231, 234
new genus, 182, 186, 187 504 oharrai, 231
Hemiganus, 264 fricki, 499 oreodontis, 231
Hemipsalodon, 75-7, 98, 101, 108 sp., 499 platycopsis, 230
 Index 678

Hoplophoneus (cont.) latidens, 307 hypostylus, 575

primaevus, 234 Lepidus, 307 imperialis, 575
robustus, 231 loomisi, 307 intermedius, 574
sicarius, 231 marshi, 307 modestus, 574--5
sp., 231 mentalis, 307 nanus,575
Hornless ruminants, 463-76 minor, 307 princeps, 575
see Blastomerycinae, Gelocidae minusculus, 307 sp., 575
Hypertragulidae, Leptomerycidae miticulus, 307 Hyracodon,516--7,518,519,587,589,590,591,
Houyanotherium, 285 paulus, 297, 307 592,593,599
Huetfanodon,251-2,261,263,266 powellianus, 295, 307 leidyanus, 591, 593
polecatensis, 263 sholemi, 307 nebraskensis, 590, 591, 593
torrejonius, 263 simplex, 307 petersoni, 591
sp., 263 speirianus, 306 primus, 591
Hunanictis, 319 tonksi, 307 priscidens, 591
Hyaena,243,245,246 uintensis, 307 selenidens, 591
Hyaenictis, 245 wolcottianus, 307 sp., 591
Hyaenictitherium, 245 wortmani, 307 Hyracodontidae, 516--7, 589-594
Hyaenidae, 73, 74, 75, 79, 89, 126, 129, 199, 207, sp., 307 biogeographic ranges, 516--7, 516-9, 523
243-6 Hyotherium, 392, 394 biology, 593
biogeographic ranges, 86--7, 89 Hypercoryphodon,280,281 defining features, 589-90
biology, 246 Hypermekops, 446 evolutionary trends, 593
defining features, 243-4 Hypertragulidae, 88, 337, 339, 341, 463-76, 464, included North American genera, 590--1,
evolutionary patterns, 246 470--1, 474 593
included North American genera, 245-6 biogeographic ranges, 342--4, 353-4, 354 systematics, 590
systematics, 244-5 biology, 475 temporal ranges, 592
Hyaenocyon, 132 defining features, 464 Hyracodontinae, 88, 591, 592, 593
Hyaenodon, 75-7,85,98,101, 102, 108, evolutionary patterns, 475 Hyracoidea, 248, 249, 250, 514
380 included North American genera, 470--1 Hyracotheriinae, 513-5, 540, 542-3, 555
brevirostris, 98, 98, 101, 100, 102, 108 systematics, 467, 468-9 Hyracotherium, 46, 512, 513-5, 515, 516, 537,
crucians, 98, 108 temporal ranges, 474 538,539,540,542,543,553,554,555,556,
cruentus, 108 Hypertragulus, 353-4, 367, 368, 464, 465, 466, 573,578
exiguus, 108 467,468,469,470,474,475 aemulor, 542
horridus, 98, 108 calcaratus, 470 angustidens, 542
leptocephalus, 108 heikeni, 470 borealis, 542
leptorhynchus, 98 hesperius, 470 craspedotum, 542
megaloides, 98, 108 minutus, 470 cristatum, 542
microdon, 98, 108 sp., 470 grangeri, 542
minutus, 108 Hypisodus, 353--4, 417, 463, 467, 469, 470, 474, index, 542
montanus, 98, 108 475 leporinum, 542
mustelinus, 98, 108 minimus, 470 pemix, 542
paucidens, 108 sp., 470 sandrae, 542
raineyi, 98, 108 Hypoconifera, 340 seekinsi, 542
venturae, 98, 108 Hypohippus,513-5,520,521,540,544,545,554 tapirinum, 542, 555
vetus, 98, 108 affinis, 545 vasacciense, 542
sp., 98, 108 chico, 545 venticolum, 542
Hyaenodontid-nimravid carnivoran association, equinus, 545 sp., 542, 573
83 nevadensis, 545 Hystricognath rodents, 68
Hyaenodontidae, 73, 74, 75-7, 81-2, 84, 85, 87, osbomi, 545
88, 91, 93, 94, 96--8, 100, 101, 107-8, 98, pertinax, 545 lbarus,343-4,360,363,371
108 tehonense, 545 ignotus, 363
Hyaenodontinae, 75-7, 83, 98, 101 wardi, 545 lcticyon, 151
Hyaenognathus, 137 sp., 545 Ictidopappus, 78-9, 80, 112, 113, 114, 119,
Hydrochaerus, 577 Hypolagus, 67, 138 120
Hydrochoerus, 577 Hypotemnodon, 131 mustelinus, 119
Hydropotes, 342 Hypselochoerus, 412-3, 422 sp., 119
Hyemoschus, 475 Hypsilolambda, 281 Idiogenomys, 251-2, 623--4
Hylochoerus, 399 Hypsiops, 347--8, 405, 406, 413, 425 ozziei, 623, 624
Hylomeryx, 343-4, 358, 360, 361, 362, 367-8, bannackensis, 414, 418 Ignatiolambda, 251-2, 256, 279, 281, 282
371 brachymelis, 414 bamesi, 279
annectans, 367 breviceps, 414 Ilingoceros, 355, 494, 495, 497, 501, 502, 504
breviceps, 408 erythroceps, 413 alexandrae, 497
quadricuspis, 367 johndayensis, 414 schizoceras, 497
sp., 367 leptoscelos, 413 sp., 497
Hyomeryx, 422 luskensis, 413 Ilingocerotini, 494
Hyopotamus, 382, 384, 385, 422 Hypsoparia, 158 Indarctos, 84--5, 89, 176, 177, 182, 183, 184, 186,
Hyopsodontidae, 247, 248, 250, 251, 255--6, 257, Hyrachyus, 515, 518--20, 569, 570, 571, 574, 575, 187, 192, 197,213,216
292,298,305-8,323,325 579,583,589-91 anthracitis, 183
Hyopsodus,255-6,257,258,294,305,307,323, agrarius, 575 arctoides, 183
325,367 agrestis, 575 atticus, 183
acolytus, 310 bairdianus, 575 lagrelii, 183
despiciens, 307 crassidens, 575 oregonensis, 183, 195
egressus, 364 eximius, 575 salmontanus, 183
fastigatus, 307 grandis, sinensis, 183
 Index 679

Siwalik species, 183 Lamini, 453-5, 459 edwardsi, 434

vireti, 183 Lampadophorus,265 leptolophus, 434
Indricotheriinae, 590-1, 592 Laopithecus, 363 major, 434
Indricotherium, 589 Laphictis, 163 marshi, 434
Insectivora, 73, 74 Laredochoerus,343-4,360,363--4,371 pusillus, 434
lntyrictis, 78--9, 80, 112, 113, 114, 115, 116, 120 edwardsi, 363--4 stocki, 434
vanvaleni, 116 Leptarctinae, 152, 154, 157-8, 166, 168, sp., 434
Ischnognathus, 75-7, 98, 101 170 Leptotragulus, 349-50, 432, 433, 434, 439
savagei, 98, 108 Leptarctini, 154, 157-8 clarki, 433
Ischyrichtini, 88, 154, 162--4, 166, 168, 170 Leptarctus, 82-3, 158, 165, 166, 168, 170 medius, 433
lschyrictis, 163 ancipidens, 158 proavus, 433
Ischyrocyon, 84--5, 88, 197, 199, 201, 213, oregonensis, 158 sp., 434
214-5,217,218-9,227 primus, 157-8 "Leptotragulus" profectus, 349-50, 433, 439
gidleyi, 215-6 progressus, 158 Lepus, 67
hyaenodus, 216 wortmani, 158 Leurocephalus,530
Ischyromyid rodents, 17 sp., 158 Lichnocyon, 148
Ischyrosmilus, 239 Leptauchenia, 343, 347-8, 402, 405, 410, 411, Limnenetes, 405, 410, 422
crusifonti, 239 417,418,421,422,424,425 Limnocyon, 75-7,99, 100,101, 107
idahoensis, 239 decora, 410 agilis, 107
johnstoni, 239 densa, 410 douglassi, 108
osbomi, 231 eiseleyi, 410 dysodus, 108
Isectolophidae, 511, 512, 514, 518-20, 571, harveyi, 410 potens, 99, 108
572-3,579 major, 410 protenus, 109, 115-6
Isectolophus, 515, 518-20, 571, 572, 573, 579 margeryae, 410 riparius, 108
latidens, 573 martini, 410 velox, 107, 108
annectens, 573 minora, 410 velox, 108
latidens, 573 nitida, 410 vera, 108
radinskyi, 573 parasimus, 410 verus, 92, 99, 108
scotti, 573 platyceps, 410 sp., 99
sp., 573 spp. indet., 410 Limnocyonidae, 75-7, 82, 83, 91, 94, 98-9, 100,
lthyogrammodon, 369, 370 Leptaucheniinae, 345, 347-8, 407, 410, 417 101, 108
Leptictids, 74 Limnofelis, 106
Jepsenia, 310 Leptictimorphs, 260, 261 Limnohyus,529
Jiaoluotherium, 285 Leptochoerinae, 338, 339, 343-4, 362--4, 371 Limnonyx,82-3, 161,166,168, 170
Juniper, 45 new genus,343--4, 364 sinerizi, 161
Leptochoerus,342,343-4,359,360,363,371 sp., 161
Kalakotia, 514 elegans, 363 Lions, 66
Kalobatippus, 513-5, 519, 540, 544--5, 554 emilyae, 363 Lipodectes, 275, 301
agatensis, 545 spectabilis, 363 Lipotyphla, 74
australis, 545 supremus, 363 Litaletes, 252, 255--6, 309, 310, 325
avus,545 sp., 363 disjunctus, 310
clarenci, 545 Leptocyon, 80--1, 87, 88, 127, 131, 132-3, 138, gazinensis, 310
condoni, 545 139, 140--1 mantiensis, 310
gracilis, 545 gregorii, 133 ondolinde,310
navasotae, 545 shermanensis, 133 stembergi, 310
praestans, 544--5 vafer, 132-3 sp., 310
sp., 545 vulpinus, 133 Litolestes, 306
Kazachostylops, 335 sp., 133 Litolophus, 562, 563
Kobus, 488 Leptolambda, 251-2, 279, 281, 282 Litomylus, 255-6, 305, 306, 307, 308, 325
Kolponomos,84--5, 127, 178, 180,186, 188, 191 schmidti, 279 aequidens, 308
clallamensis, 180, 194 Leptomerycidae, 17, 337, 339, 341, 463-76, alphamon, 308
newportensis, 180, 194 465,468,469,471,474 dissentaneus, 307-8, 309
Konobelodon (see Amebelodon britti) biogeographic ranges, 342-6, 353-4, 354-5 ishami, 308
Kuanchuanius, 269 biology, 475 orthronepius, 308
Kukusepasutanka, 345--6, 384, 386, 387 defining features, 465 osceolae, 308
schultzi, 385 evolutionary patterns, 475 scaphiscus, 308
Kyptoceras,349-50,352,435,437,438 included North American genera, 4 71 n.sp., 308
amatorum, 435 systematics, 467, 469 sp., 308
Kyptoceratini, 433, 435, 437 temporal ranges, 474 Lofochaius,269-70,271
Leptomeryx, 341, 353-4, 464, 465, 466, 467, 469, Longirostromeryx, 349, 353-4, 469-70, 472,
Lagomorpha, 285 471,473,474,475 473,474,475
Lama, 453, 454, 455 agatensis, 471 blicki, 473
Lambdoceras,346,349-50,431,435-6,437 blacki, 471 clarendoniensis, 473
hessei, 435-6 esulcatus, 471 merriami, 473
siouxensis, 436 evansi, 471 novomexicanus, 473
trinitiensis, 436 mammifer, 471 serpentinus, 473
sp., 436 obliquidens, 471 vigoratus, 472
Lambdotherium, 515, 518-20, 528-9, 533, 624 semicinctus, 471 wellsi, 473
popoagicum, 529 speciosus, 471 sp., 473
Lambertocyon, 253--4, 301, 303, 324 transmontanus, 471 Lophialetidae, 512
eximius, 303 yoderi, 471 Lophiaspis, 561, 562, 563
ischyrus, 303 sp., 471 Lophiodon,561,573-5,576,577
sp., 303 'Leptoreodon,349-50,432,433,434,437 Lophiodontidae, 512
 Index 680

Lophiohyus, 369, 370 sp., 610 priscum, 322

Loxodonta, 66, 606 Mammuthus, 66, 67, 68, 522, 609, 615, 616, 617 robustus, 322
Loxolophodon, 287 Mammutidae, 5, 522, 606, 609, 610, 616, 617 tapiacitum, 322
Loxolophus, 253-4, 294, 301-2, 302, 316, 321, Manteoceras, 530 terraerubrae, 322
324 Mantuan fauna, 258 sp., 322
adapinus, 302 Martes, 82-3, 162-3, 165, 166, 168, 170 Menoceras, 516-7, 520, 595, 596, 597-8, 600,
criswelli, 302 foina, 162 602,603
kimbetovius, 302 glareae, 162 arikarense, 595, 600
pentacus, 302 oregonensis, 162 barbouri, 600
priscus, 302 stirtoni, 162 cooki, 600
schizophrenus, 302 sp., 162 falkenbachi, 600
spiekeri, 302 Martinogale, 82-3, 158, 159-60, 166, 168, 170 marslandensis, 600
sp., 302 alveodens, 159 Menoceratinae, 600
Lutra, 157, 161, 163 nambiana, 159 Menodus, 529, 532
Lutravus, 82-3, 165, 166, 168, 170 Mastodons, 26 Menops, 518-20, 526, 527, 528, 529, 532, 533,
halli, 165 Mastodontidae, 595 534
Lutrinae, 88, 152 . Matthomeryx, 355, 479, 483, 487 crassicomis, 532
Lutrini, 154, 161-2, 166, 168 matthewi, 483 heloceras, 532
Lycaon, 125, 126, 127,246 (see also Drepanomeryx) marshi, 532
Lycyaena, 245 Mazama,469 serotinus, 532
Lynx,86-7,89,237,238,240 Mediochoerus,347-8,415,417,422,425 trigonoceras, 532
kurteni, 238 blicki, 415 varians, 532
lynx, 238 johnsoni, 415 walcotti, 532
proterolyncis, 238 mohavensis, 415, 418 Menotherium, 363
rexroadensis, 238 n.sp., 415 Mentoclaenodon, 253-4, 301, 303, 324
sp., 238 Megabathygenys, 409, 422 acrogenius, 303
Megabelodon, 522, 609, 611, 616, 617, 618 walbeckensis, 303
Machaeroides, 75-7, 100, 101, 107 joraki, 611 Mephitinae, 88, 89, 152
eothen,91,96, 101, 107 lulli, 611 Mephitini, 154, 159-60, 166, 168, 170
simpsoni, 96, 107 minor, 611 Mephitis, 82-3, 160, 166, 168, 170
Machaeroidinae, 75-7, 82, 93, 96, 101, 107 Megacamelus, 351-2, 443, 456, 457, 459 interrupta, 160
Machaeromeryx, 353-4, 465, 469, 469, 473, 474, merriami, 457 mephitis rexroadensis, 160
475 Megaceratops, 532 Merriamoceros, 355, 493, 494, 496, 502
gilchristensis, 473 Megacerops, 518-20, 526, 528, 529, 532, 533, coronatus, 491, 496
tragulus, 473 534 sp., 496
sp., 473 coloradensis, 532 Meryceros,355,493,494,496,502,503
Machairodontinae, 86-7, 239, 240, 241 curtus, 532 crucensis, 496
Machairodus,86-7,89, 197,237,239,240,241 kuwagatarhinus, 526, 532, 534 hookwayi, 496
ischyrus, 239 platyceras, 532, 534 joraki, 496
aphanistus, 239 Megachoerus, 342,345-6,378,379 major, 496
brachyops, 230 praecursor, 378 nenzelensis, 496
catacopsis, 238 zygomaticus, 378 warreni, 492, 496
coloradensis, 232, 236, 239 Megahippus,513-5,521,540,545,553,554 sp., 496
oreodontis, 231 matthewi, 545 Merychenia, 448
primaevus, 231 mckennai, 545 Merychippus, 521, 513--5, 520, 538, 541, 546,
tanneri, 239 sp., 545 547,549-50,553,554,555,556
sp., 239 Megalamynodon, 516-7, 584, 585, 586, 587 brevidontus, 54 7
Macrogenis, 345-6, 347, 391, 395, 396, 398 regalis, 585 calamarius, 547
crassigenis, 389, 390, 392, 396, 399 Megalesthonyx, 251-2, 257, 271, 272 califomicus, 547
n. sp., 396 hopsoni, 271 campestris, 550
Macrotarsius, 367 Megalictis, 156 carrizoensis, 54 7
Macrotherium, 565 Meganteron, 86-7, 237, 239, 240, 241 circulus, 550
Madoqua, 475 cultridens, 239 coloradense, 54 7
Mahonia, 45 hesperus, 239 francisi, 550
Maiorana, 250, 255-6, 314--5, 324 Megasespia, 410, 422 goorisi, 547
noctiluca, 314--5 Megatapirus, 577 gunteri, 547
Malaquiferus, 349-50, 428, 429 Megatylopus, 348, 351-2, 443, 456-7, 459, 460 insignis, 547
tourteloti, 428 cochrani, 456 intermontanus, 547
Mammacyon, 84--5, 208, 209, 217, 218-9, 224 gigas, 456 isonesus, 547
obtusidens, 209 major, 453, 456-7 mirabilis, 550
n.sp. I, 209 matthewi, 456 perditus, 550
n.sp. II, 209 primaevus, 456-7 primus, 547, 555
Mammut, 521, 522, 523, 607, 608, 610, 616, 617, sp., 456 proplacidus, 550
618-9 Megoreodon, 413, 422 re !ictus, 54 7
adamsi, 610 Meiostylodon, 269, 270, 271 republicanus, 547
americanum, 607, 610 Melinae, 152, 154, 158, 159-62, 166, 168 sejunctus, 547
cosoensis, 610 Melini, 154 seversus, 547
furlongi, 610 Mellivora, 162 stylodontus, 547
matthewi, 610 Mellivorinae, 152 sty/onus, 547
nevadanus, 610 Meniscotheriinae, 321-2 sumani, 547
praecursor, 610 Meniscotherium, 255-6, 257, 293, 294, 320-1, tehachapiensis, 547
sellardsi, 610 322,325,326 te rtius, 54 7
vexillianus, 610 chamense, 292, 297, 322 sp., 547
 Index 681

Merychyinae, 414--6 Mesaxonia, 250, 514 uintensis, 532

Merychyus,345,347-8,350,352,414,415,417, Mesocyon,80--1, 127, 129, 130, 131, 139 ultimum, 532
418,422,423,425 baileyi, 131 Metaxyomys, 67
arenarum, 414 brachyops, 131 Metoreodon, 414-5, 423
calaminthus, 415 coryphaeus, 131 Miacidae, 73, 74, 76, 110, 111, 120
calimontanus, 414 drummondanus, 131 Miacinae, 78-9, 81, 83, 111
crabilli, 415 geringensis, 132 Miacis, 81, 109, 112, 113, 114, 115, 117, 119,
curtus, 414 hortulirosae, 131 120, 121
delicatus, 414 iamonensis, 132 australis, 117
elegans, 414 josephi, 131 cognitus, 117
euryops, 414 robustus, 131 deutschi, 117
jahnsi, 415 temnodon, 131 exiguus, 117
major, 415, 418 venator, 131, 138 gracilis, 109, 117, 130, 138
medius, 415 sp., 131 hargeri, 117
paniensis, 415 Mesohippus, 10, 88, 513-5, 517, 520, 538, 539, hookwayi, 117
pariogonus, 413 540,543,544,546,554--5,554,556,576 jepseni, 117
re/ictus, 415 assiniboiensis, 544 latidens, 117
siouxensis, 413 bairdii, 544 medius, 117
smithii, 415 barbouri, 544 parvivorus, 117, 119
verrucomalus, 414 brachystylus, 544 petilus, 117
spp. indet., 415 celer, 544 sylvestris, 117, 138
Merycobunodon,349-50,427,428,429 equiceps, 544 uintensis, 117
littoralis, 427 exoletus, 544 washakius, 117
Merycochoerinae, 407, 412-3, 417 gidleyi, 544 winkleri, 117
Merycochoerus, 343,347-8,405,413,414,417, grallipes, 544 sp., 117
418,422,423,424,425 grandis, 544 Miacoidea, 73, 74, 75, 76, 81, 83, 114, 127, 128.
altiramus, 415 hypostylus, 544 See also Carnivora, early Cenozoic
buwaldi, 415 intermedius, 544 Michenia, 351-2, 440, 450, 451, 458, 459, 460
chelydra latidens, 544 agatensis, 451
coenopus, 415 longipes, 576 australis, 451
compressidens, 415 montensis, 544 deschutensis, 451
elrodi, 415 obliquidens, 544 exilis, 451
grandis, 413 planidens, 544, 546 yavapaiensis, 451
laticeps, 415 portentus, 544 sp., 451
leidyi, 413 praecocidens, 544 Microclaenodon, 253-4, 317-8, 324, 326
macrostegus, 413 propinquus, 544 assurgens, 317-8
madisonius, 415 proteulophus, 544 Microsus, 343-4, 360, 361, 365, 366, 367, 368,
magnus, 413 stenolophus, 546 371
matthewi, 413 texanus, 544 cuspidatus, 359, 367
montanus, 413 trigonostylus, 544 sp., 367
obliquidens;415 westoni, 544 Microtine rodents, 26, 67, 68
proprius, 407, 413, 418 sp., 544 Microtus, 26
rusticus, 415 Mesomephitis, 170 Mictomys, 67
superbus, 413 Mesomeryx, 343-4, 358, 360, 361, 368, 369, 371 Mimatuta, 26, 250, 255-6, 299, 311-2, 315,
temporalis, 413 grangeri, 368 325
Merycodesmus, 434 Mesonychidae,73,74,120,247,248,253-4,255, makpialutae, 312
Merycodontinae, 492, 493, 494, 495, 502, 504--5 257,257,293,318-9,324 minuial, 312
Merycodus, 349, 355, 472, 494, 495, 496, 497, Mesonyx, 253-4, 257, 292, 297, 319, 324, 326, morgoth, 311-2
498,502 423 sp., 312
nectatus, 496 dakotensis Mimotricentes, 253-4, 256, 301, 302, 303, 324
sabulonis, 496 lanius, 318 angustidens, 303
savaronis, 496 navajovius, 318 e[assus, 303
sp., 496 obtusidens, 319 fremontensis, 303
Merycoides, 343, 347-8, 406, 413-4, 417, 421, sp., 319 latidens, 303
422,423,424,425 Mesoreodon, 412-4, 423 mirielae, 303
cursor, 413 Metachriacus, 301-2 subtrigonus, 302-3
giganteus, 413 Metahyrachyus, 574-5 sp., 303
harrisonensis, 407, 413-4 Metailurus, 238 Mingotherium, 251-2, 623-4
longiceps, 414 Metamynodon, 385, 516-7, 518, 583, 584, 585, holtae, 623
nebraskensis, 413 586,587,599 Miniochoeriinae, 407, 410-1, 417
pariogonus, 413 chadronensis, 585, 587 Miniochoerus, 410-1, 423
Merycoidodon,404,405,405,411,412,423 mckennyi, 585 Mioclaenidae, 248, 251, 252, 255-6, 257, 298,
Merycoidodontidae, 337, 338, 347-8, 343, 347, planifrons,583,585,587 308-11, 323, 325
350,355,356,407,409,416,417 Metarhinus, 518-9, 526, 528, 530, 533 Mioclaenus, 255-6, 309, 310, 325
Merycoidodontinae, 410-11 abbotti; 530 angustus, 313
Merycopater, 408-9, 423 diploconus, 530 cuspidatus, 316
Merycopotamus, 384 jl.uviatilis, 530 etsagicus, 362
Mesagriochoerus, 408, 423 riparius, 530 filholianus, 305
Mesamynodon, 585 pater, 530 jl.overianus, 320
Mesatirhinus, 518-20, 526, 528, 530, 533 Metasequoia, 41, 47, 49, 53 opisthacus, 310
Junius, 530 Metatelmatherium, 518-20, 526, 528, 529, 532, sectorius, 313
megarhinus, 530 533 subtrigonus, 302
petersoni, 530 cristatum, 532 turgidunculus, 309, 311
 Index 682

Mioclaenus (cont.) elatus, 561, 564, 565, 567 Nanomeryx, 364

turgidus, 296, 310 hollandi, 564 Nanosmilus, 86-7, 231, 233
sp., 310 matthewi, 564, 565 kurteni, 23 l, 234
Miocyon, 78-9, 114, 120 merriami, 564, 565, 567 Nanotragulus, 88, 344, 353-4, 470-1, 474
magnus, 78-9, 117 oregonensis, 561, 564 albanensis, 471
major, 117 senex, 564 fontanus, 471
scotti, 117 sp., 564, 565 intermedius, 471
Miohippus, 88, 513-5, 518, 520, 521, 540, 544, Moschidae,26,339,341-2,465-6,469- 70, loomisi, 470-1
546,554 472-3,474 lulli, 471
acutidens, 544 Moschina, 342 matthewi, 471
anceps, 544 Moschoedestes, 600 ordinatus, 469, 471
antiquus, 544 Moschus,337,349-50,351,467,469,47 5 planiceps, 471
assiniboiensis, 544 Mustela, 82-3, 162, 164, 166, 168, 170 sp., 471
crassicuspis, 544 erminea, 164 Nasua,82-3,89, 147, 148,150
equinanus, 544 foina, 162 nasua, 148
eulophus, 544 frenata, 164 pronarica, 148
gemmarosae, 544 furlongi, 159 Navahoceras, 509
gidleyi, 544 gazini, 164 Nebraskomys, 66
intermedius, 544 meltoni, 164 Neoclaenodon, 304
meteulophus, 544 minuta, 165 Neocynodesmus, 132-3
obliquidens, 544 nivalis, 167 Neodiacodexis, 343-4, 360, 364, 365-6, 371
validus, 544 rexroadensis, 164 emryi, 365--6
sp., 544 sp., 164 sp., 365
Miolabine sp., 351-2, 441, 442-3, 449, 459 Mustelavinae, 145, 152, 153-5, 166 Neohipparion, 513--5, 521, 541, 547, 548, 549,
Miolabis, 351-2, 440, 443, 447, 448, 449, 450, Mustelavus, 82-3, 84, 145, 152, 153-5, 166, 167, 554
451,453,459 169, 176 affine, 548
califomicus, 448 priscus, 153-4, 155 arellenoi, 548
jissidens, 448 sp., 154 coloradense, 54 7
fricki, 448 Mustelictis, 145, 153, 155, 156 do liehops, 548
longiceps, 445, 448 Mustelidae, 74, 75, 76-7, 86, 87, 88, 89, 120, 126, eurystyle, 548
montanus, 448 127, 144, 145, 152-73,200,208 jl.oresi, 548
princetonianus, 448 biogeographic ranges, 82-83, 85-9 gidleyi, 548
tenuis, 448 biology, 165, 167-70 ingenuus, 549
transmontanus, 448 biostratigraphy, 170 lenticularis, 549
sp., 448 defining features, 152 leptode, 548
Miomastodon, 610 evolutionary patterns, 165-70 molle, 548
Miomustela, 82-3, 158, 159, 165, 166, 168, included North American genera, 152-65 manias, 548
170 systematics, 152 occidentale, 549
minor, 158 temporal ranges, 166 otomi, 548
madisonae, 158 Mustelinae, 88, 89, 152, 154, 158, 162-5, 166, phosphorum, 548
sp., 158 168 trampasense, 548
Mionictis, 82-3, 88, 157, 161, 166, 168, 170 Mustelini, 154, 164, 166 whitneyi, 548
angustidens, 161 Musteloidea, 74, 77, 144, 145, 152 sp., 548
elegans, 161 Mydaus, 159 Neohyaenodon, 108
incertus, 161 Mylagaulid rodents, 26 Neomustelidae, 82-3, 87, 88, 153, 154, 157-65,
letifer, 161 Mylagaulids, 595 166,168, 169, 170
pristinus, 161 Mylohyus, 66, 345-6, 351, 395, 396, 397, 398, Neoselenodontia, 339, 340
Miotapirus, 518-20, 521, 570, 572, 577, 578, 579 399 Neotragoceras, 353, 355, 356, 486, 508-9
harrisonensis, 577 elmorei, 393 improvisus, 509
marslandensis, 577 elmori, 397 lingreni, 509
sp., 577 fossilis, 397 Nestoritherium, 562
Miotylopus, 351-2, 444, 451, 459 longirostris, 397 Nimravidae, 74, 76, 78, 84, 85, 88, 102, 114, 127,
gibbi, 444 sp., 397 169,228-35
leonardi, 444 Myrmecophaga,68 biogeographic ranges, 83-6, 86-7, 89
taylori, 444 Mytonomeryx, 343-4, 358, 360, 361, 368, 371 biology, 232-34
wilsoni, 446 scotti, 368 defining features, 228-9
Miracinonyx, 86-7, 237, 239, 240, 241 evolutionary patterns, 232-4
studeri, 239 Nannippus,67,513-5,541,547,548-9,5 54,555 included North American genera, 229-32
trumani, 239 aztecus, 549 systematics, 229-30
Mithrandir, 313 beckensis, 549 temporal ranges, 233
Mixotoxodon, 333 gratus, 548 Nimravides, 86-7, 88, 197, 234, 238, 239, 240,
Mixtotherium, 467 hesperides, 549 241
Moeritheriidae,606 ingenuus, 549 catacopsis, 238
Mongoloryctes, 319 lenticularis, 549 galiani, 238
Mongolotherium, 285 minor, 549 lovei, 238
Mongooses, 73 montezuma, 549 nebraskensis, 238
Montanatylopus, 349-50, 426, 428-30, 429 morgani, 549 pediomus, 238
matthewi, 428 peninsulatus, 548-9 sp., 238
Moropomorpha,512,513,514,569-82,5 70,571, phlegon, 549 Nimravinae, 79, 230-1, 233
573,578,579 tehonense, 548 Nimravus, 86-7, 102, 228, 229, 230, 233, 234
Moropus,518-20,561,562,563,564,5 65,566, westoni, 549 altidens, 230
567 sp., 549 brachyops, 230
distans, 564, 565 Nanochoerus, 363 bumpensis, 230
 Index 683

confertus, 230 synonyms, 421-5 josephi, 314

gomphodus, 230 systematics, 404-6 marshater, 314
meridianus, 230 temporal ranges, 417 priscilla, 314
Nonticholeptine, 354 Oreodontoides, 347-8, 414, 417, 421, 422, 423 n. sp., 314
Notharctus, 109 curtus, 414 Oxyaena, 73, 75-7,81,82, 100,101, 106
Nothocyon, 80--1, 128, 132-3, 137-8, 139, 624 oregonensis, 414 aequidens, 106
geismarianus, 135, 137-8 Oreomunnea, 50 dysclerus, 99, 108
gregorii, 133, 138 Oreonetes, 347-8, 410-1, 417, 418, 423, 424, dysodus, 99, 108
latidens, 138 425 forcipata, 95, 106
lemur, 138 anceps, 410-1 gulo, 92, 95, 106
vulpinus, 133, 138 gracilis, 411 huerfanensis, 106
sp., 137 nicholsae, 411 intermedia, 95, 106
"Nothocyon" annectans, 80--1, 129, 134-5, 139 Oreonetinae, 410-1 lichna, 106
Nothokemas, 346, 351-2, 440, 447, 449, 451, 452, Orohippus,513--5,538,540,542,543,5 54 lupina, 95, 106
459,469,473 agilis, 543 morsitans, 106
jl.oridanus, 447 major, 543 pardalis, 95, 106
hidalgensis, 447 progressus, 543 platypus, 106
waldropi, 447 proteros, 543 simpsoni, 95, 106
sp., 447 pumulis, 543 transiens, 106
Nothokematinae, 351-2, 447 sylvaticus, 543 ultima, 106
Nothotylopus, 351-2, 447, 449-50, 459 sp., 543 wortmani, 99, 108
camptognathus, 449-50 Oromerycidae, 337, 339, 340, 426-30 spp., 95, 106
sp., 450 biogeographic ranges, 342-3, 349-50, Oxyaenidae, 4, 73, 74, 75-7, 80, 81, 82, 91, 93,
Nothrotheriops, 68 354-5 94-6, 100,101, 106,257
Notiomastodon, 612 biology, 428, 430 Oxyaeninae, 94-5, 101
Notiotitanops, 518-20, 526, 528, 531, 533 defining features, 426-7 Oxyaenodon,75-7,83,99, 100,101, 108
mississippiensis, 531 evolutionary patterns, 428-30 dyseodus,99
Numidotherium, 606 included North American genera, 427-8 Oxyclaenus, 253-4, 300, 316, 324
Nyctereutes, 127, 133, 170 systematics, 427 corax, 316
Nypa, 42, 44 temporal ranges, 429 cuspidatus, 316
Oromeryx, 349-50, 426, 427, 428, 429 pearcei, 316
Ocalientinus, 610, 611 plicatus, 428 pugnax,316
Octacodon, 384 Orthaspidotherium, 322 simplex, 305, 316
Odocoileus, 355, 508, 509 Orthocynodon, 584 sp., 316
brachydontus, 509 Orycteropus, 323 Oxydactylus, 351-2, 439, 444, 448, 451, 452, 453,
Ogmodontomys, 66 Osbomoceros,355,494,495,497,501,5 02 456,458,459,460
Okapia, 567 osbomi, 497 benedentatus, 451
Oligobuninae, 85, 152, 154, 155-7, 166, 167, 169, sp., 497 brachyodontus,450
170 Osbomodon,80--1,86,87,88,127,129,1 30,131, campestris, 451
Oligobunis, 82-3, 155, 166, 167, 169 132, 138,139 jl.oridanus, 447
crassivultus, 156 brachypus, 132 lacota, 451
jl.oridanus, 156 fricki, 132 longipes, 451
sp., 156 iamonensis, 132 longirostris, 451
Oligotomus, 321 renjiei, 132 lulli, 451
Omomyid primates, 13 sesoni, 132 wyomingensis, 451
Onohippidium, 513, 541, 551, 552, 554 Osbomohyus,412,422 sp., 451
galushai, 552 Osmotherium, 170 Oxyprimus,251,255-6,305,308,325
munizi, 552 Osmunda, 110 erickseni, 308
Onychodectes, 251-2, 260, 261, 262, 263, 265, Osteoborus, 80--1, 89, 129, 137, 139, 140, 151, galadrielae, 308
266 197 putorius, 308
rarus, 262 cyonoides, 125, 137 simplex, 305
tisonensis, 262 diabloensis, 136 sp., 308
Oodectes, 109, 112, 113, 114, 115, 117-8, 120, direptor, 137 Oxytomodon, 253-4, 305, 308, 324
121 galushi, 137 perissum, 305
herpestoides, 109, 117-8, 119 hilli, 137
jepseni, 109 littoralis, 136 Pachyaena,253-4,257, 318,319,324, 326
proximus, 118 ore, 137 gigantea, 318
sp., 118 progressus, 137 gracilis, 318
Ophiomys, 67 pugnator, 137 ossifraga, 318
Oracodon, 106 secundus, 137 sp., 318
Oreamnos, 534 sp., 137 Pachycynodon, 127, 178
Oreinotherium, 567 Otarohyus, 424 Pachynolophidae, 512
Oreocyon, 106 Otionohyus, 411-2, 424 Pachynolophus, 573, 574
Oreodon, 404, 411-2, 423 Otocyon, 125, 128-9, 170 Paenemarmota, 66, 246
Oreodontinae, 410-1 Otocyoninae, 127, 128-9 Paeneprolimnocyon, 107-9, 117-18
Oreodontoidea, 15, 17, 338, 339, 340, 402-25, Ottoceros, 355, 494, 497, 502 Paenungulata, 247, 248, 249, 275, 285, 298, 299.
423 peacevalleyensis, 497 See also Ungulatelike mammals
biogeographic ranges, 342-8, 347-8, 350, Ovibos, 66 Palaearctonyx, 78-9, 113, 114, 115, 119, 120
354-6 Oxetocyon,80-1, 129, 134,139 meadi, 118
biology, 416, 418 cuspidatus, 134 morsitans, 106
defining features, 403-4 Oxyacodon,255-6,293,299,305,314,3 25 Palaeictops, 109
evolutionary patterns, 416-8 agapetillus, 314 Palaeocastor, 169
included genera, 406-16 ferronensis, 314 Palaeochoerus,392, 394,395
 Index 684

Palaeodonta, 338, 339, 340 Pantolambda, 251-2, 253, 274-5, 278, 281, 282, Paranasua,82-3, 148,150
Palaeogale, 82-3, 84, 114, 115, 165, 166, 167, 282 biradica, 148
624 bathmodon, 278 Parapromerycochoerus, 424
dorothiae, 165 cavirictum, 278 Pararamoceros, 495
infelix, 165 intermedium, 278 Parastenopsochoerus, 410-1, 413, 424
lagophaga, 165 sp., 278 Paratoceras, 345, 349-50, 431, 433, 434, 435,
minuta, 165 Pantolambdidae, 275, 276, 277-9, 282 437
Pantolambdodon, 281 macadamsi, 435
sanguinarius, 165
Pantolambdoidea, 275 wardi, 435
sectoria, 165
Pantolestes, 361, 362 sp., 435
sp., 165
Pantolestids, 74, 100, 260 Paratylopus, 351-2, 443-4, 447, 448, 457, 458,
Palaeohypsodontus, 501
Pappictidops, 114 459
Palaeolama, 351-2, 453, 454, 455, 459, 460
Parabathygenys, 409, 424 labiatus, 444, 352
aequatorialis, 454
Parablastomeryx, 349, 353-4, 464, 469, 472, primaevus, 443-4, 451
guanajuatensis, 454
474 "Paratylopus" came/aides, 351-2, 444, 459
mirifica, 439, 454 Parenhydrocyon, 127, 129, 130, 131, 139
weddelli, 454 fioridanus, 472
galushi, 472 josephi, 131
Palaeomastodon, ·606, 607
gregorii, 472 robustus, 131
Palaeomerycidae, 338, 339, 341-2, 345, 347, 350
sp., 472 wallovianus, 131, 138
Palaeomeryx, 342, 478, 479, 481, 482, Pareporeodon, 412, 422
483-4,486,488 Paracamelus, 457
Paraceratherines, 589 Parictis, 84-5, 127, 176, 177, 178, 179, 180, 186,
Palaeonictinae, 75-7, 95-6, 101 187, 190
Palaeonictis, 75-7, 80, 95, 100, 101, 106 Paraceratherium, 589
Paracosoryx, 345,355,492,493,494,495,502, dakotensis, 179, 193
gigantea, 95 gilpini, 178-9, 193
503
occidentalis, 95, 106 major, 179, 193
alticornis, 495
peloria, 95, 106 montanus, 179, 193
dawesensis, 495
peloria parvus, 178-9, 193
furlongi, 495
sp., 95, 106 personi, 179, 193
loxoceros, 495
Palaeoryctidae, 74, 93, 100, 111, 260, 261 primaevus, 178-9, 193
nevadensis, 495
Palaeostylops, 333, 335 wilsoni, 493, 495 sp., 193
Palaeosyopinae, 529-30, 533 Parisectolophus, 573
sp., 495
Palaeosyops, 515, 51S-20, 525, 526, 527, 528, Paroodectes, 114
Paracotylops, 413, 424
529,530,532,533 Paradaphoenus, 84-5, 202, 205-6, 217, 21S-9, Paroligobunis, 82-3, 86, 155, 156, 166, 167, 169
copei, 530 224 frazieri, 156
diaconus, 530 cuspigerus, 205-6 petersoni, 156
fontinalis, 530 minimus, 206 simplicidens, 156
grangeri, 530 transversus, 205 sp., 156
huerfanensis, 530 Paradesmatochoerus, 412-3, 421 Paroreodon,347-8,414,417,422,423,424
humilis, 529 Paradoxodon, 302 marshi, 414
laevidens, 529 Paraenhydrocyon, 80-1, 86 parvus, 414
leidyi, 530 Paragale, 157 stocki, 414
longirostris, 529 Parahippus,513-5,520,537,540,545,546,547, Paroxyclaenidae, 293
major, 529 553,554 Parvitragulus, 353-4, 469, 470, 474
matthewi, 530 agrestis, 546 priscus, 4 70
monoconus, 530 barbouri, 546 sp., 470
montanus, 529 brevidens, 546 Pastoralodon, 281
paludosus, 529 cognatus, 546 Patriofelis, 75-7, 82, 100, 101, 106
priscus, 529 coloradensis, 546 coloradensis, 95, 106
robustus, 530 leonensis, 537, 546, 555 compressa, 106
Palaeotheriidae, 250, 512 pawniensis, 546 ferox, 93, 95, 106
Paleomoropus, 512, 513, 51S-20, 561, 562, 563, pristinus, 546 leidyanum, 106
570,571,573,578,579,624 vellicans, 546 leidyanus, 106
jepseni, 573 wyomingensis, 546 tigrinus, 106
Paleomustelidae, 82-3, 84, 86, 87, 88, 153, sp., 546 ulta, 92, 95, 106
155-7,166,167 Parahyus,343-4,360,361,370,371,378 sp., 95
Paleonictidae, 93, 95-6, 100, 106-7 aberrans, 369 Paup Fiction, 342
Paleoryctes, 74 vagus,370 Pecora, 16,26,66,338,339,341-2
Panthera, 237 Parailurus, 82-3, 149, 150, 188 Pediohippus, 544
atrox, 232, 341 anglicus, 149 Pediohyus, 392
lea, 66, 232 sp., 149 Pediomeryx, 353, 355, 479, 480-1, 485, 486, 487,
onca augusta, 241 Paralabis, 351-2, 443, 444, 459 488-9,509
palaeoonca, 239 cedrensis, 444 hemphillensis, 481, 486, 489
spelaeus, 241 Paramerychyus, 413-4, 424 sp., 486
tigris, 241 Paramerycochoerus, 413, 424 (see also Yumaceras)
Pantodonta, 13,247, 250, 253,255, 256,257,268, Paramerycoidodon, 411-2, 424 Pegomylus, 444
274-83,293,624 Paramiacis, 114 Pelonax, 378
biogeographic ranges, 251-2, 251-8 Paraminiochoerus, 410-1, 423 Pelycodus, 109
biology, 281 Paramiolabis, 351-2, 447, 448, 459 Penetrigonias, 516-7, 518, 595, 598, 603
defining features, 274 singularis, 448 dakotensis, 598
evolutionary patterns, 281-2 taylori, 448 hudsoni, 598
included North American genera, 277-81 tenuis, 448 sagittatus, 598
systematics, 274-7 sp., 448 Pentacemylus, 343-4, 358, 360, 361, 365, 368,
temporal ranges, 282 Paramynodon, 587 369,371,416
 Index 685

leotensis, 368 minor, 135 Pliogale, 82-3, 158, 159, 166, 168, 170
progressus, 368 sp., 135 furlongi, 159
Pentacodontidae, 293 Phoberocyon, 84-5, 88, 174, 180, 181, 182, 186, manka, 159
Peraceras,516-7,595,596,600,601,602,603, 187, 191 sp., 159
604 aurelianensis, 181 Pliogulo, 137
crassus, 601 huerzeleri, 181 Pliohippus, 513-5, 521, 541, 547, 549-50, 551,
hessei, 595, 601 johnhenryi, 181, 194 542,554
profectum, 601 Phocids, 145 bakeri, 550
superciliosum, 595, 596, 601, 604 Pholidotans, 74 castilli, 550
troxelli, 601 Phyllotillon, 562, 563 circulus, 550
sp., 601 Pinnipedia, 74, 75, 114 coalingensis, 550
Perchoeru!, 342, 345-6, 390, 391, 393, 394, 398, Pinnipedimorpha, 144 fairbanksi, 550
397-99 Pithecistes, 410, 424 fossulatus, 550
minor, 394 Planktonic foraminiferans, 18, 21 francesca, 550
probus,392,394,397 Platybelodon, 521, 522, 607, 608, 609, 613, 614, hondurensis, 550
socialis, 395 615,617,618,619 leardi, 550
sp., 394 barnumbrowni, 615 minor, 550
Pericyon, 205 danovi, 615 mirabilis, 550
Periptychidae, 248, 250, 251, 252, 255-6, 275, grangeri, 615 nobilis, 550
292,298,311-5,323,325 Platycarya pollen, 13 pernix, 550
Periptychinae, 312-3 Platygonus, 66, 345-6, 351, 391, 393, 395, 396, profectus, 550
Periptychus, 255-6, 294, 299, 312-3, 323, 397,398,399,400 robustus, 550
325 brachyrostris, 397 simus, 550
carinidens, 295, 312-3 compressus, 389, 397, 390 spectans, 550
coarctatus, 312 cumberlandensis, 397 supremus, 550
ditrigonus, 312 pearcei, 397 tantalus, 550
gilmorei, 313 oregonensis, 397 tehonense, 550
superstes, 313 Edson sp., 392 sp., 550
Perissodactyla, 93, 247, 248, 249, 250, 257, 274, sp., 397 Pliolemmus, 66
275,292,337,511-2,523 Platymastus, 252, 253-4, 319, 324 Pliomastodon, 610
biogeographic ranges, 515-23 mellon, 306, 319 Plionarctos, 84-5, 89, 182, 184-5, 186, 187, 192
Clarendonian chronofauna, 521 palantir, 319 edensis, 184, 195
Eocene,515-16 Platyochoerus, 410-1, 424 Plionictis, 82-3, 162-3, 163, 164, 165, 166, 168,
evolutionary patterns, 515-22 Plesiadapids, I 0, 11, 13 170
Mio-Pliocene chronofauna, 521-2 Plesictis, 82-3, 155, 156, 166, 167 buwaldi, 162, 164
Running Water chronofauna, 519-21 pygmeus, 155 gazini, 162
systematics, 512-4 genettoides, 155 kinseyi, 162
White River chronofauna, 517-9 julieni, 165 ogygia, 162
Perunium, 168 palmidens, 153 parviloba, 162
Phacochoerus,399 priscus, 155 sp., 162
Phenaceras, 285 pygmaeus, 153 Pliopotamys, 26, 66
Phenacocoelinae, 407, 413-4, 417 robustus, 153 Pliotaxidea, 82-3, 160-1, 166, 168, 170
Phenacocoelus,347-8,414,417,424 sicaulensis, 165 garberi, 161
kayi, 413 sp., 155 nevadensis, 160-1
monroensis, 414 Plesiesthonyx, 269, 270, 272 sp., 161
stouti, 414 Plesiocolopirus, 576 Plithocyon, 84-5, 88, 177, 180, 181-2, 186, 187,
typus, 414 Plesiogale, 157 191
Phenacodaptes,255-6,257,309,311,325 Plesiogulo, 82-3, 162, 163, 164, 166, 168, 170 armagnacensis, 181
sabulosus, 311 brachygnathus, 163 barstowensis, 181, 194
Phenacodonta,253,255-6,257,320-2,325 lindsayi, 163, 168-9 bruneti, 181
Phenacodontidae, 248, 292, 298, 299, 320-1, 325, marshalli, 163 statzlingii, 181
326 Plesiomiacis, 118 ursinus, 181, 194
Phenacodus,255-6,253,255,321,322,325, Plesippus, 552 n. sp., 194
326 Pleuraspidotherium, 295, 321, 322 sp. I, 181
almiensis, 321 Pleuroceros, 598 sp. II, 181
bisonensis, 321 Pliauchenia, 351-2, 447, 455, 459 sp., 181, 194
gidleyi, 321 humphresiana, 455 Poabromylus,349-50,432,434,436,437
grangeri, 321 magnifontis, 455 golzi, 434
intermedius, 321 merriami, 457 kayi, 434
magnus, 321 singularis, 448 minor, 434
primaevus, 294, 297, 321 spatula, 457 robustus, 434
puercensis, 320 Plioceros, 349, 355, 494, 495, 497-8, 501, 502, sp., 434
trilobatus, 321 504 Poatrephes, 415, 424
vortmani, 321 blicki, 497-8 Podocarpus, 50
matthewi, 321 dehlini, 497 Poebrodon,351-2,441,443,473,459
sp., 321 foblairi, 497, 505 californicus, 441
Phenacolophidae, 275, 293 sp., 497 kayi, 441
Phenacomys, 26 Pliocyon, 84-5, 88, 129, 197, 212-13, 217, sp., 441
Philotrox, 80-1, 127, 129, 132, 139 218-9,225,226 Poebrotheriinae, 441
condoni, 132 medius, 212 Poebrotherium, 88, 343, 351-2, 440, 443, 444,
Phiomia, 606, 607, 608 robustus, 212 451,457,458-60,459,466
Phlaocyon,80-1, 129, 134, 135, 138, 139 walkerae, 216 chadronense,443,446
leucosteus, 135 sp., 212 eximium, 443
 Index 686

Poebrotherium (cont.) Procynodictis, 78, 83, 112, 113, 114, 115, 118, sheperdi, 310
franki, 348-9, 353 120, 121, 128, 130 wilsoni, 310
sternbergi, 447 progressus, 118 sp., 310
wilsoni, 440, 443 vulpiceps, 118 Proneofiber, 26
sp., 443 sp., 118 Pronodens,344,353-4,465,469,471,474
"Poebrotherium" franki, 343, 351-2, 445-6, Procyon,82-3,89, 147, 148,150 silberlingi, 4 71
459 lotor, 148 sp., 471
Pogonodon,86--7,228,229,230-1,233,234 rexroadensis, 148 Pronomotherium, 415, 423, 425
cismontanus, 230 sp., 148 Proputorius, 159
davisi, 230 Procyonidae, 73, 74, 75, 76-7, 88, 89, 120, 126, Prosthecion, 321
eileenae, 230 127, 128, 144-51, 152,200 Prosthennops, 345-6, 391, 392-3, 396, 397,
paucidens, 231 biogeographic ranges, 82-3, 88-9 398
platycopsis, 230-1 biology, 149, 151 crassigenis, 392-3
serrulidens, 230 defining features, 144 elmorei, 397
Potamotherium, 82-3, 155, 157, 166, 167, 170 evolutionary patterns, 149-51 kernensis, 3 97
valetoni, 157 included North America genera, 146-9 serus, 392-3, 397
sp., 157 systematics, 144-6 "Prosthennops" niobrarensis, 345-6, 392, 393,
Potos, 148, 151 temporal ranges, 150 395,396,398,399
Pratifelis, 86--7, 238, 240 Procyoninae, 88, 129, 144, 145-6, 147-8, 149, "Prosthennops" xiphidonticus, 345-6, 393, 395,
martini, 238 150, 151 396,398,399
Pratilepus, 66 Prodaphaenus,78-9,83,112,113,114,115,118, Prosynthetoceras, 349-50, 435-6, 437, 486
Primates, 11, 93, 249 120 australis, 436
Princetonia, 253-4, 301, 324 scotti, 117, 118 francisi, 436
yalensis, 301 vulpinus, 118 rileyi, 436
Priscocamelus, 351-2, 452, 458, 459 Prodesmatochoerinae, 344 texanus, 436
wilsoni, 452 Prodesmatochoerus, 347-8, 405, 411, 412, 417, sp., 436
Pristinictis, 74, 78-9, 114, 120 418,421,422,423,424 Protagriochoerus, 408, 425
connata, 117 macrorhinus 411 Protapirus, 512, 518-20, 519, 521, 570, 572, 576,
Proailurus, 236 meekae,411 577,578,579
Proantilocapra, 355, 494, 495, 497, 501, periculorum, 405, 403, 407, 411 hancocki, 576
502 Prodinoceras, 251-2, 255, 256, 257, 281, 285, obliquidens, 576
platycornea, 497 286-7,286,288-90,289 priscus, 576
Probarbouromeryx, 355, 484, 487 harrisorum, 287 robustus, 576
sweeti, 484 hobackensis, 287 simplex, 576
(see also Barbouromeryx) lysitensis, 287 validus, 570, 576
Probassiriscus, 82-3, 147, 150 martyr, 286 sp., 576
antiquus, 147 newbilli, 287 Proteles, 243
matthewi, 147 praecursor, 287 Protelotherium, 370
Probathyopsis, 285, 286 successor, 287 Proteutherians, 74
Problastomeryx, 353-4, 469, 472, 474 sp., 287 Prothryptacodon, 253-4, 301, 324
olcotti, 472 Prodinoceratidae, 286-7, 289 albertensis, 301, 305
primus, 472 Prodipodomys, 67 ambiguus, 301
sp., 472 Prohesperocyon, 124, 129, 130, 139 hilli, 301
Proboscidea, 5, 247, 248, 249, 250, 274, 511-2, wilsoni, 130 furens, 301
523,525,606-622,617 Prohyaena, 129, 136 Protictis, 78-9, 81, 112, 113, 114, 115, 116, 117,
biogeographic ranges, 521-3, 522 Prolibytherium, 479 120
biology, 616, 618 Prolimnocyon, 75-7, 98, 99, 100, 101 agastor, 116
defining features, 607 antiquus, 99, 108 dellensis, 116
evolutionary trends, 616, 618-9 atavus,99, 100, 107, 108 haydenianus, 116, 119
included genera, 609-16 eerius, 99, 108 laytoni, 116
systematics, 608-9 elisabethae, 99, 108 para/us, 116
temporal ranges, 617 haematus, 99, 108 schaffi, 116
Procadurocodon,516--7,585,586,624 iudei, 99, 109 vanvaleni, 116
Procamelus, 347, 348, 351-2, 440, 447, 452, macfaddeni, 109 n. spp., 116
455-6,458,459,460 robustus, 107 sp., 116
coartatus, 456 sp., 99, 108 Protictoides, 78-9, 81, 116, 120
coconinensis, 456 Promartes, 82-3, 85, 153, 155, 156, 167, 166, aprophatos, 116
edensis, 454 167, 169, 170 Protitanops, 518-20, 528, 531, 533
gracilis, 451 darbyi, 155 curryi, 531
gracilis, 456 gemmarosae, 155 Protitanotherium, 518-20, 526, 528, 530-1, 533
grandis, 456 Lepidus, 155 emarginatum, 530-1
heterodontus, 450 olcotti, 155 Protobarbouromeryx, 355, 484, 487
inaequidens, 451, 456 vantassalensis, 155 marslandensis, 484
leptocolon, 456 sp., 155 (see also Barbouromeryx)
leptognathus, 456 Promephitis, 160, 170 Protoceras, 343, 349-50, 381, 385, 431, 432, 434,
longurio, 454 Promerycochoerinae, 412-3 435,436,437,438
occidentalis, 455-6 Promerycochoerus, 412-3, 424, 425 celer, 432, 435
sp., 456 Promesoreodon,412-3,424,425 neatodelpha, 435
Procastoroides, 67 Promioclaenus, 251, 255-6, 309, 310, 311, 325 skinneri, 435
Procerberus, 261 acolytus, 310 sp., 435
Procoileus, 481, 486, 509 aquilonius, 310 Protoceratidae, 26, 337, 338, 339, 340, 341, 431-
Procranioceras, 355, 480, 485, 487 lemuroides, 310 8, 509, 595
skinneri, 485, 488 pipiringosi, 310 basal, 433-4
 Index 687

biogeographic ranges, 343-8, 349-50, 352, secundaria, 107 Pseudoleptauchenia, 410, 425
354,356 secundarius, 96, 107 Pseudomesoreodon, 414, 425
biology, 436, 438 strenuus, 107 Pseudomerychochoerus, 425
defining features, 431-2 viverrinus, 96, 107, 108 Pseudoparablastomeryx, 346, 347, 353-4, 469,
evolutionary patterns, 436-8 vulpeculus, 107 471,474
included genera, 433-6 spp., 96, 107 francescita, 471
systematics, 432-3 Protungulatum, 248, 250, 253-4, 299, 300, 309, scotti, 465, 471
temporal ranges, 437 334 sp., 471
Protoceratinae, 349-50, 353, 433, 434-5, 437 donnae,294,295,300,308,312,322,324 Pseudopromerycochoerus, 413, 425
Protochriacus, 302 gorgun, 300 Pseudoprotoceras, 343, 349-50, 433, 434, 435,
Protocyon, 127 mckeeveri, 300 436,437
Protodichobune, 366 sloani, 300 longinaris, 434-5
Protogonia, 320 sp., 300 semicinctus, 435
Protogonodon, 301, 302, 321 Protylopus,349-50,426,427,428,429 taylori, 435
Protohippus, 513-5, 521, 541, 547, 549, 550, 554, annectens, 429 sp., 435
555,556 minor, 428 Pseudopterodon, 108
profectus, 550 parvus, 428 Pseudostenopsochoerus, 410-1, 425
gidleyi, 550 pearsonensis, 428 Psittacotherium, 251-2, 253, 256, 261, 263,
lenticularis, 549 petersoni, 428 264-5,266
martini, 550 robustus, 428 aspasiae, 265
perditus, 549-50, 555 stocki, 428 multifragum, 264-5
placidus, 550 sp., 428 Pterocarya, 50
proplacidus, 550 Prouintatherium, 285 Pterodon, 96, 108
simus, 550 Proviverra, 107, 109 Pteroneura, 161
supremus, 550 Proviverrinae, 75-7, 82, 94, 96-8, 101, 107-8 Puma,86--7,89,238-9,240
vetus, 550 Proviverroides, 75-7, 97, 101 concolor, 237, 238
sp., 550 piercei, 97, 107 idahoensis, 238
Protohyaenodon, 108 sp., 97, 107 lacustris, 238-9
Protolabinae, 351-2, 441, 442-3, 450-1, 459 Pseudailurus, 86--7, 88, 236, 238, 240 Purgatorius, 296
Protolabis, 351-2, 440, 447, 448, 450-1, 453, aeluroides, 238 Pyrocyon, 75-7,97, 101
455,458,459,460 hibbardi, 238 dioctetus, 97, 107
barstowensis, 451 intrepidus, 238 Pyrodon, 285
coartatus, 451, 456 marshi, 238 Pyrotheria, 250, 284, 285, 290
gracilis, 451 quadridentatus, 238
heterodontus, 440, 450-1 sinclairi, 238 Quercygale, 114
inaequidens, 451 stouti, 241, 238
notiochorinos, 451 thinobates, 238 Radinskya, 539
transmontanus, 448 sp., 238 Ragnarok,250,253-4,315,324
sp., 451 Pseudhipparion, 513-5, 521, 541, 547-8, 550, engdahli, 315
Protomeryx, 444 554 harbichti, 315
Protoprocyon, 82-3, 148, 150 curtivallum, 548 nordicum, 315
savagei, 148 gratum, 548 wovokae, 315
Protopsalis, 75, 95, 101 hessei, 548 sp., 315
tigrinus, 95, 106 retrusum, 547-8 Rakomeryx,355,478,479,480,481,483,487
Protoreodon, 342, 347-8, 368, 405, 408, 417, simpsoni, 548 americanus, 483
421,422,423,425 skinneri, 548 gazini, 483
annectens, 405, 408 sp., 548 jorakianus, 483
medius, 408 Pseudictops, 285, 290, 333, 623 kinseyi, 488
minor, 408 Pseudictops-like anagalids, 250 raki, 483
parvus, 405 Pseudobassaris, 153 sinclairi, 483, 486
petersoni, 405, 408 Pseudoblastomeryx, 353-4, 353, 469, 472, 474 yermonensis, 483
n. sp., 405, 408, 416 advena, 472 sp., 483
sp., 408 falkenbachi, 472 Rakomylus, 348, 351-2, 445, 457, 459
Protoreodontinae, 406, 408, 417 schultzi, 472 raki, 445
new genus A, 347-8, 406, 408, 417 Pseudoceras, 339, 341, 350, 353, 353-4, 356, sp., 445
new genus B, 347-8, 406, 408, 417, 425 467,469,471-2,474,475 Ramoceros, 347, 349, 355, 493, 494, 495, 496,
new genus C, 347-8, 406, 408, 417, 421 potteri, 472 501,502,503
new genus D, 347-8, 406, 408, 417, 421 skinneri, 471-2 grevicornis, 496
Protoselene, 255-6, 309, 310, 325 wilsoni, 472 hitchcockensis, 495
bombadili, 310 sp., 472 howardae, 495
griphus, 309, 310 Pseudoceratinae, (see also Gelocidae) 465, 469, kansanus, 495
novissimus, 310 471-2 marthae, 496
opisthacus, 309, 310 Pseudocyclopidius, 410, 425 osborni, 495
sp., 310 Pseudocynodictis, 127 palmatus, 496
Prototomus, 75-7, 96, 97, 100, 101, 107 Pseudocyon, 84-5, 88, 197,201,202, 209,213, ramosus, 495
deimos, 96, 107 214,215-16,217,218-19,226 sp., 496
insidiosus, 109 gerandianus, 210 Ramocerotini, 493, 494
jarrovii, 109 sansaniensis, 214 Raphictis, 78-9, 80, 112, 113, 114, 115, 116,
martis, 96, 107 Pseudocyonopsis, 209 120
mordax, 108 Pseudodesmatochoerus, 412-4, 425 gausion, 116
multicuspis, 107 Pseudogenetochoerus, 412, 425 sp., 116
phobos, 96, 107 Pseudolabis, 351-2, 444, 459 Rangifer, 504
robustus, 96, 107 dakotensis, 444 Ratel, 152
 Index 688

Ravenictis, 74, 78-9, 80, 114, 117, 120 hudsoni, 470 sp., 497
krausei, 117 minutus, 470 Spilogale, 82-3, 159-60, 166, 168, 170
Rhabdobunus, 613 sp., 470 interrupta, 160
Rhadinorhinus, 518-9, 530 Simocyon,82-3, 127,144, 149,150, 151 microdens, 160
Rhapsodus, 147 marshi, 145, 149 putorius, 160
Rhinoceros, 66, 572 primigenius, 149 rexroadi, 160
Rhinocerotidae, 511, 512, 595-605 robustus, 149 sp., 160
biogeographic ranges, 515-23, 516--7 Simocyoninae, 77, 127, 144, 146, 149 Stegodontidae, 606
biology, 602, 604 Simpsonictis, 78-9, 112, 113, 114, 116, 120 Stegomastodon, 522, 607, 608, 609, 613, 616,
defining features, 596 jaynanneae, 116 617,618-9
evolutionary trends, 602, 604 pegus, 116 mirificus, 613
included genera, 598-602 tenuis, 116 primitivus, 613
systematics, 597-8 sp., 116 rexroadensis, 613
temporal ranges, 603 Simpsonodus,343-4,360,362,371,372 sp., 613
Rhinocerotoidea, 16, 26, 516--7, 569, 571, chacensis, 362 Stegotetrabelodon, 606
574-5,589 sp., 362 Stelocyon, 253-4, 315-6, 324
Rhinochoerus, 571 Sinclairomeryx, 349, 355, 477, 480, 481, 482, arctylos, 315-6
Rhodopagidae, 512 487,488 Stenogale,82-3, 158, 165, 166, 167,624
Rhynchotherium, 67, 522, 608, 609, 612, 613, riparius, 482 gracilis, 165
615,617,618-9 sinclairi, 482 robusta, 163
blicki, 612 tedi, 482 sp., 165
edense, 612 sp., 482 Stenomylinae, 351-2, 353, 441, 442-3, 444-5,
euhyphodon,612,615 Sinohippus, 540 459
praecursor, 612 Sinopa, 75-7,97,101, 107 Stenomylini, 444-5, 459
simpsoni, 612 agilis, 107 Stenomylus, 343, 345, 347, 351-2, 354, 444, 458,
tlasca[ae, 612 eximia, 109 459,460
sp., 612 grangeri, 97, 107 gracilis, 444
Rostriamynodon, 584, 587 hians, 107 hitchcocki, 444
Ruminantia, 338, 339, 341-2 insectivorus, 109 tubutamensis, 444
biogeographic ranges, 342-56 major, 97, 107 sp., 444
incertae sedis, 473, 474 minor, 97, 107 Stenopsochoerus, 410--1, 425
systematics, 341-2, 466-68, 624 mordax, 108 Sthenictis, 82-3, 162, 163, 166, 168, 170
multicuspis, 107 bellus, 163
Saiga, 68 opisthotoma, 107 campestris, 163
Salamanders, 18 pungens, 97, 107 dolichops, 163
Sangamona, 509 rapax, 93, 97, 107 junturensis, 163
Sarcobatus, 52 secundaria, 107 lacota, 163
Sarcolemur, 343-4, 360, 364, 365, 371 shoshoniensis, 107 robustus, 163
furcatus, 365 strenua, 107 sp., 163
Sarcophilus, 100 vera, 108 Sthenodectes,518-20,526,528,529,532,533
Sarcothraustes, 316 viverrina, 107 incisivum, 532
Satherium, 82-3, 153, 161, 166, 168 vulpecula, 96, 107 Stibarus, 343-4, 360, 363, 371
ingens, 161 whitiae, 107 obtusilobus, 363
piscinarium, 161 sp., 97, 107 montanus, 363
sp., 161 Sirenia, 248, 249, 250 quadricuspis, 363
Scaptohyus, 378 Sivacyon, 128 yoderensis, 363
Schizolophodon, 573 Smilodon, 67, 169, 234, 236, 237, 241 sp., 363
Schizotheriinae, 560, 561, 562 Sminthosinis, 82-3, 165, 166, 168 Stipa, 45
Schizotheriodes, 518-20, 571, 573, 575, 578, 579 bowleri, 165 Stockocerotini, 494, 499
jackwilsoni, 575 sp., 165 Stockoceros, 494, 495, 497, 499, 500, 501, 503,
Schizotherium, 563, 624 Soergelia, 67, 68 504
Schochia,251-2,261,263,266 Sorex, 69 Strobodon, 80-1, 129, 136, 137, 139
sullivani, 263 Spano.xyodon,301-2 stirtoni, 136
Sciadopitys, 50 Speothos, 125, 127 sp., 136
Selenaletes, 512, 516, 518-20, 572, 574, 579 Spermophillus, 69 Stylinodon, 251-2, 257, 260, 261, 263, 265, 266,
scopaeus, 574 Sphenocoelus,518-20,525,526,528,530,533, 273
sp., 574 534 cylindrifer, 265
Selenodontia, 339 angustidens, 530 inexplicatus, 265
incertae sedis, 366-9, 473, 624 blairi, 530 mirus, 264, 265
Serbelodon, 339, 521, 522, 609, 614, 617, 618, bridgeri, 530 sp., 265
619 comutum, 530 Stylinodontidae, 251-2, 253, 255, 260, 261, 263,
barbourensis, 614, 618 fiuminalis, 530 266,267
barbouri, 614 harundivorax, 530 Stylohipparion, 540
bumhami, 614 heterodon, 530 Stypolophus, 107, 109
Serridentinus, 610, 611 hyognathus, 530 Subantilocapra, 355, 499, 502
Sespia,343,347-8,410,416,417,422,425 intermedius, 530 Subcosoryx, 497
califomica, 410 longiceps, 530 Subdesmatochoerus, 412, 425
marianae, 410 superior, 530 Subdromomeryx, 347, 355, 479, 483, 487, 488
nitida, 407, 410 uintensis, 530 antilopinus, 483
ultima, 410 Sphenomeryx, 367 scotti, 483
Sianodon, 587 Sphenophalos,350,355,494,495,497,499,502 wilsoni, 483
Simimeryx, 342, 344, 353-4, 368, 468-9, 470, middlewartsi, 497 sp., 483
474 nevadanus, 497 (see also Dromomeryx)
 Index 689

Subhyracodon, 516--7, 518, 519, 595, 596, 598, systematics, 570--2 Telmatheriinae, 525
599,602,603 temporal ranges, 579 Telmatherium, 516, 518-20, 526, 527, 528, 530,
kewi, 599 Tapirus, 418, 518-20, 521, 523, 570, 572, 575, 533,534
mitis, 599 576,577-8;579,580 cultridens, 530
occidentalis, 596, 599 bairdii, 578 manteoceras, 530
sagittatus, 598 califomicus, 577 validus, 530
tridactylum, 599 copei, 577 washakiensis, 530
sp., 599 excelsus, 577 Telmatocyon, 108
Submeryceros, 352-3, 496-7, 502 haysii, 577 Temnocyon, 84-5, 86, 127, 208-9, 217, 218-9,
crucianus, 496-7 indicus, 578 224
minimus, 497 johnsoni, 577 altigenis, 208
minor, 497 merriami, 577 ferox, 208
Submerycochoerus, 414, 425 pinchaque, 578 percussor, 208
Subparictis, 145, 179 priscus, 576 venator, 131
Suidae, 337, 338 simpsoni, 577 wallovianus, 131
Suiformes, 339-40 terrestris, 577-8 n.sp. I, 208
biogeographic ranges, 342-56, 345 veroensis, 577 n.sp. II, 208
systematics, 339-40 sp., 577 Ternnocyoninae,84-5,86,88, 197, 199,200,201,
Suina, 339-340 Tapochoerus,342,343-4,360,364,365,371 207-9,217,218-9,218,219
indeterminate, 378, 380 egressus, 364-5 Ternnocyonine-amphicyonid--entelodont
Sunkahetanka,80-1, 129, 131-2,139 Tapocyon, 78-9,83, 112, 113, 114, 115, 118, carnivoran association, 86
geringensis, 131-2 118-9, 120 Tephrocyon, 127, 135-6
pahinsintewakpa, 132 occidentalis, 118 Tetheopsis, 251-2, 257, 285, 286, 287, 289, 290
sheffleri, 131, 132 robustus, 118 ingens, 287
Suoidea, 339-40, 354 sp., 118 speirianus, 287
incertae sedis, 624 Tapyra, 577 Tethytheria, 248, 250
Superdesmatochoerus, 413, 425 Tatabelodon, 610, 611 Tetraclaenodon, 248, 253, 255-6, 258, 301, 320--
Sus, 391, 392 Taucanamo, 392 1, 322, 325, 326
Synaptomys, 26 Taxaceae,47 puercensis, 320
Syndyoceras,346,349-50,435,437 Taxeopoda,320--2 septentrionalis, 320
cooki, 435 Taxidea,82-3, 161, 166, 167, 168, 169, 170 sp., 294
Synoplotherium, 253-4, 318, 319, 324, 326 mexicana, 161 Tetralophodon, 522, 608, 609, 615-6, 617
canius, 318 nevadensis, 160 campester, 615-6
lanius, 318 taxus, 161 fricki, 615
vorax, 318 sp., 161 longirostris, 615
sp., 318 Taxidiini, 88, 154, 160-1, 166, 168, 170 Tetrameryx, 67, 355, 494, 495, 499, 500, 502, 504
Synthetoceras, 348, 349-50, 352, 431, 436, 437 Taxodiaceae, 41, 47 shuleri, 500
tricornatus, 436 Taxodium, 53 sp., 500
Synthetoceratinae, 349-50, 433, 435-6, 437 Tayassu,66,345-6,391,392,393,394,395,396, Texoceros,353,355,494,495,498-9,502,503,
Synthetoceratini, 433, 435-6, 437 397,398,400 504
Syspotamus, 571 edensis, 397 altidens, 498
Systemodon, 572-3 pecari, 389, 393, 397, 399 edensis, 498
Tayassuidae, 337, 339-40, 389-401 guymonensis, 498
Taeniodonta, 73, 253, 255, 256, 257, 260--7, 268, biogeographic ranges, 342-56, 345-6 minorei, 498-9
293 biology, 397, 399-400 vaughani, 499
biogeographic ranges, 251-2, 251-8 defining features, 389-91 sp., 499
biology, 265-7 evolutionary patterns, 397-400 Texodon,343-4,360,361,364,366,371
defining features, 260--1 included North American genera, 393-7 meridianus, 365, 366
evolutionary patterns, 265-7 systematics, 391-3 Thalassictis, 245
included North American genera, 261-5 temporal ranges, 398 Thangorodrim, 316
systematics, 261 Tayassuinae, 345-6, 395-7, 398 Theridomyids, 16
temporal ranges, 266 Black Hawk Ranch species, 345-6, 395, 396, Theriodictis, 127
Taligrada, 274, 275 398 Thinocyon, 75-7,99, 100, 101
Tanupolama, 454 Love Bone Bed species, 345-6, 395, 396, 398 cledensis, 99, 108
Tanymykter, 351-2, 450, 451, 458, 459, 460 Machaerodus Quarry species, 345-6, 392, 395, comptus, 99, 108
brachyodontus, 450 396,398,399 medius, 99, 108
Tanyops, 576 Teleoceras, 516--7, 521, 523, 595, 597, 600, 601, mustelinus, 99, 108
Tanyorhinus, 530 602,603,604 velox, 99, 108
Tapir, 577 fossiger, 601, 602 sp., 99, 108
Tapiravus,515-23,518-20,572,577,519 hicksi, 601, 602 Thinohyus,342,345-6,391,392,393,398,399
polkensis, 577 major, 601, 602 lentus, 392, 393, 394
rarus, 577 medicomutum, 601, 602 nanus,394
validus, 577 meridianum, 521, 595, 601, 602 osmonti, 394
sp., 577 ocotensis, 601 sp., 393
Tapiridae,511,512,518-20,571,576-8,579 proterum, 601, 602, 604 "Thinohyus" siouxensis, 345-6, 390, 392, 394,
new genus,518-20,576,579 n. sp. A, 601, 604 395,398
Tapiriella, 577 n. sp. B, 601, 602 Thryptacodon, 253-4, 257, 301, 303, 304-5,
Tapiroidea, 511, 512, 569-82, 579 sp., 601 324
biogeographic ranges, 515-23, 518-20 Teleoceratinae, 601-2 antiquus, 304-5
biology, 578, 580 Teleoceratini, 601-2, 603 australis, 305
defining features, 569 Teletaceras, 595, 603 barae, 305
evolutionary patterns, 578, 580 mortivallis, 598 belli, 305
included genera, 572-8 radinskyi, 598 demari, 305
 Index 690

Thryptacodon (cont.) loomisi, 614 vetulus, 271

loisi, 305 Toxotherium, 517, 518-20, 569, 572, 575, 579, sp., 269, 271
olseni, 305 624 Tubulidentata, 247, 248, 249, 250
pseudarctos, 305 hunteri, 575 Tylocephalonyx, 518-20, 520, 526, 560, 561, 562,
sp., 305 woodi, 575 563,565,566,567
Ticholeptinae, 345, 347-8, 407, 414-6, 417 sp., 575 skinneri, 561, 565
Ticholeptus, 347-8, 414-5, 417, 423, 424, 425 Tragelaphus, 498 sp., 565
bannackensis, 414 Tragulidae, 338, 339, 341 Tylopoda,337,338,339
brachymylis, 414 Tragulina, 339, 341 biogeographic ranges, 342-56, 347-52
breviceps, 414 Traguloidea, 88, 337, 341 systematics, 340
calimontanus, 414 Tragulus, 475, 511 Tytthaena, 75-7, 80, 93, 94, 101
hypsodus,415 Tremarctini, 84-5, 176, 186, 188 lichna,94, 106
major, 415 Tremarctos, 84-5, 89, 174, 180, 182, 184-5, 186, parrisi, 94, 106
obliquidens, 415 187, 193 Tytthaeninae, 75-7, 94, 101, 106
petersoni, 414 fioridanus, 185
rileyi, 415 ornatus, 184, 185 Uintaceras, 575, 591
tooheyi, 414 sp., 195 Uintacyon, 78-9, 112, 113, 114, 115, 118, 119,
zygomaticus, 415 Triacodon, 107, 108, 116-7 120, 121
sp., 415 Tricentes, 301, 302, 303, 309 acutus, 119
Tillodon, 251-2, 257, 271, 272-3 Triceromeryx, 479 asodes, 119
fodiens, 271 Tricuspidontidae, 293 bathygnathus, 119
Tillodontia, 250, 257, 261, 268-73, 275, 293 Trigenicus, 434 edax, 118-9
biogeographic ranges, 251-2, 251-8 Trigonias, 516-7, 518, 595, 596, 598, 599, 602, jugulans, 119
biology, 271-3 603 major, 117
defining features, 268 osborni, 598 massetericus, 119
evolutionary patterns, 271-3 wellsi, 598 promicrodon, 119
included North American genera, 269-71 sp., 598 pugnax, 119
systematics, 268-9 Trigonictis, 82-3, 164, 165, 166, 168 rudis, 119
temporal ranges, 272 cooki, 164 scotti, 118
Tillotherium, 271 idahoensis, 164 vorax, 119
Tinuviel, 255-6, 315, 325 kansasensis, 164 sp., 119
eurydice, 315 macrodon, 164 Uintatheriamorpha, 250
n. sp., 315 sp., 164 Uintatheriidae, 247, 257, 275, 285-6, 287-8, 289
Titanoideidae, 276, 277, 282 Trigonolestes, 361 (see also Dinocerata)
Titanoides,251-2,255,275,216,277,282 Triisodon,253-4,316,317,322,324 Uintatheriinae, 285, 286
faberi, 279 antiquus, 296, 317 Uintatherium, 251-2, 257, 275, 284, 285, 286,
gidleyi, 277 assurgens, 317 287,288,289,290
majus, 277 heilprinianus, 316 anceps,287
primaevus, 277, 278, 281 levisanus, 316 Ungulata, 285, 298
simpsoni, 277 guivirensis, 317 Ungulatelike mammals, 247
zeuxis, 277 sp.,317 systematics, 250
sp., 277 Triisodontines, 248, 250, 252, 255 Ungulates, 23, 83, 88
Titanops, 532 Trilophodon, 610 archaic, see Condylarthra and Archaic
Titanotheres (see Brontotheriidae) Triplopides, 516-7, 592, 593 Ungulates
Titanotheriomorpha, 512, 514 rieli, 593 Urocyon,80-1,89, 127, 133,139, 141
Titanotherium, 287 Triplopus,516-1,589,590,591,592,593 cinereoargenteus, 133
Titanotylopus,351-2,443,455,456,457,459 amorarum, 591 progressus, 133
nebraskensis, 457 cubitalis, 591 sp., 133
Tiznatzinia, 255-6, 309, 311, 325 implicatus, 591 Ursavinae, 84-5
turgidunculus, 311 obliquidens, 591 Ursavini, 176, 186
vanderhoofi, 311 rhinocerinus, 591 Ursavus, 84-5, 88, 176, 180, 182, 183, 185, 186,
Tomarctus, 127, 129, 131, 135-6, -138, 139, 140 woodi, 591 187, 188, 191
brevirostris, 135-6 sp., 591 brevirhinus, 183, 194
canavus, 135 Tritemnodon, 75-7, 97, 100, 101, 107 depereti, 183
confertus, 135, 136 agilis, 97, 107 ehrenbergi, 183
cuspidatus, 136 gracilis, 109 elmensis, 183
euthos, 135, 136 hians, 107 intermedius, 183
hippophagus, 135, 136 strenua, 107 orienta/is, 183
kellogi, 135, 136 strenuus, 92, 97, 107 pawniensis, 183, 194
marylandica, 135 whitiae, 107 primaevus, 183, 194
nobilis, 132 sp., 97, 107, 109 n. sp. I, 183, 194
optatus, 135, 136 Trobelodon, 610, 611 n. sp. II, 183, 194
paulus, 135 Trocharion, 158 Ursidae, 73, 74, 75, 78, 84-5, 88, 89, 120, 126,
propter, 135 Trochictis, 163, 164 127, 144, 146, 174-95, 197, 199, 200, 208,
rurestris, 125, 135 Trogosinae, 270, 271, 272 213,218,219
robustus, 135, 136 Trogosus,251-2,257,268,269,271,272-3 biogeographic ranges, 84-5, 84-9, 187
temerarius, 135 castoridens, 271 biology, 185, 188
thompsoni, 135 grangeri, 271 defining features, 17 4-6
sp., 135 hillsii, 271 evolutionary patterns, 185-8
Toromeryx, 349-50, 434, 437 hyracoides, 271 included North American genera, 178-85·
marginensis, 434 latidens, 271 localities of fossils, 190--5
Torynobelodon, 522, 608, 609, 613, 614-5, 617, minor, 271 systematics, 176-8
618,619 shantunensis, 269 temporal ranges, 186
 Index 691

Ursinae, 84-5, 88-, 89, 174, 175, 177-8, 182-5, minutus, 111, 117 sp., 542
186,200 politus, 117 Xenungulata, 285
Ursini, 84-5, 176, 185 sicarius, 117 Xinyuictis, 114
Ursoidea, 74, 77, 144, 145 vulpinus, 117 Xiphodontidae, 337, 338, 340
Ursus, 84-5, 89, 178, 182, 184, 185, 186, 187, sp., 117
188, 193 Viverridae, 73, 75, 79, 120, 126 Ysengrinia, 84-5, 197, 198, 201, 202, 209-11,
abstrusus, 185, 195 Vulpavus, 78-9,81,112,113, 114,115, 118, 119, 213,217,218-9
americanus, 174, 208 120, 121,139 depereti, 210
arctos, 174, 185, 197 australis, 119
gerandiana, 209
boeckhi, 185 canavus, 119
etruscus, 185 ovatus, 110, 119 tolosana, 210
lotor, 148 palustris, 119 Yumaceras, 349, 353, 355, 480, 481, 485, 486,
malayanus, 174 profectus, 110, 111, 119 487,488-9
minimus, 185 sp., 119 falkenbachi, 486
spelaeus, 185 Vulpes,80-1,88, 127, 132, 133, 141 figginsi, 486
taxus, 161 shermanensis, 133 hamiltoni, 486, 488
Ustatochoerus, 414-5, 425 stenognathus, 133 ruminalis, 486
Utemylus, 255-6, 305, 308, 325 velox, 197 sp., 486
latomius, 308 vulpes, 133 (see also Pediomeryx)
sp., 133
Vassacyon, 78-9, 81, 112, 113, 114, 119, 115, Zaisanamynodon, 587
120 Wanalia, 281 Zhelestidae, 292, 296
promicrodon, 119 Wasatchia, 360, 362 Zodiolestes, 82-3, 155-6, 157, 166, 167, 169
sp., 119 Wortmania, 251, 251-2, 261, 263, 264, 266
daemonelixensis, 155-6
Viverra, 148, 160 otariidens, 264
Zygolophodon, 5, 521, 522, 607, 610, 608, 617,
Viverravidae, 73, 74, 76, 84, 110, 114, 115 J.fyolestes, 253-4, 257, 319-20, 324, 624
619
Viverravinae, 78-9, 80, 81, 111 apheles, 319
Viverravus, 78-9, 81, 112, 113, 114, 116-7, 120 iglesius, 319-20 merriami, 610
acutus, 117 proavus, 607, 610
bowni, 117 Xenicohippus, 513-5, 542, 554 tapiroides, 610
gracilis, 116-7 grangeri, 542 turicensis, 610
lutosus, 117 osborni, 542 sp., 610