[Link]

org/

Bulletin of the Museum of Comparative Zoology at Harvard College.

Cambridge, Mass. :The Museum,1863-
[Link]

v.96 (1945-1946): [Link]

Page(s): Cover, Verso, Volume Title Page, Verso, Table of Contents, Blank, Title, Verso, Title
(No. 1), Page, Page [2], Article Start, Page [3], Page 4, Page 5, Page 6, Page 7, Page 8, Page
9, Page 10, Page 11, Page 12, Page 13, Page 14, Page 15, Page 16, Page 17, Page 18, Page
19, Page 20, Page 21, Page 22, Page 23, Page 24, Page 25, Page 26, Page 27, Page 28, Page
29, Page 30, Page 31, Page 32, Page 33, Page 34, Page 35, Page 36, Page 37, Page 38, Page
39, Page 40, Figure (diagram), Page 41, Figure, Page 42, Page 43, Page 44, Page 45, Page
46, Page 47, Page 48, Page 49, Page 50, Page 51, Page 52, Page 53, Page 54, Page 55, Page
[56], Title, Plates, Descr. of Plate 1, Plate 1

Contributed by: Harvard University, MCZ, Ernst Mayr Library

Sponsored by: Harvard University, Museum of Comparative Zoology, Ernst Mayr Library

Generated 5 June 2012 1:21 AM

[Link]
This page intentionally left blank.
Special CoJlections

e
..
,

HARVARD UNIVERSITY

LIBRARY
OF THE ,
,
MUSEUM OF COMPARATIVE ZOOLOGY

•
 OF THE

••

AT
•

HARVARD COLLEGE, IN CAMBRIDGE

VOL. 96
•

,,------- •

CAMBRIDGE, MASS., U. S. A.
. ......

•
1945 - 1946
 •

THE COSMOS PRESS., INc.

CAMBRIDGE, MASS., U. S. .4.~

• •

•
 •
•

r •

PAGE
No. 1.-A 8TUDY OF THE SNAKE, Tachyrnenis peruviana Wiegmann
AND ITS ALLIES. By Warren F. Walkel", JI', (5 plates.)
November, 1945 • • • • • • • • 1
No. 2. FIRST SUPPLEMENT TO TYPICAL REPTILES AND AMPHIBIANS.
By Thomas Barbour and Arthur Loveridge. February, 1946 57
No. 3. THE GENERA OF FOSSIL CONCHO:::;TRACA BI AN ORDER OF
VALVED CRUSTACEA. By Percy E. Ray111011d. (6 plates.)
J une, 1946 . . . . . . . . . 215
No. 4. STUDIES OF SOUTH AMERICAN PSAl\1l\10CHARIDAE. Part l.
By Nathan Banks, (3 plates.) December, 1946 . • • 309

•
•
...

\
 Bulletin of the MuseuIn of COlllparative Zoology
AT HARVARD COLLEGE ••

VOL. 96. No. 1

A STUDY OF THE SNAKE, TACHYMENIS PERUVIANA

WIEGMANN AND ITS ALLIES •
•
•
•

-
•

•
By WARREN F. WALKER, JR .

WITH FIVE PLATES

CAMBRIDGE, MASS., U. S. A.
• •

PRINTED FOR THE MUSEUM

•

NOVEMBER, 1945
 •

•
•

• •

PUBLICATIONS •

OF THE
••
MUSEUM OF COMPARATIVE ZOOLOGY
•

AT HARVARD COLLEGE

The BULLETINand MEMOIRSare devoted to the publication of

investigations by the Staff of the Museum or of reports by spec
ialists upon the Museum coIlections or explorations.
Of the BULLETIN,Vols. 1 to 95, and Vol. 96, No. 1, have appeared
and of the MEMOIRS,Vols. 1 to 55.
These publications are issued in numbers at irregular intervals.
Each number of the Bulletin and of the Memoirs is sold separately.
A price list of the publications of the Museum will be sent upon ap
plication to the Director of the Museum of Comparative Zoology,
Cambridge, Massachusetts.
Publication of Memoirs ceased with Vol. 55 .

•
-

•~

•
--~------------------~--------------------------~.~(

I
•

Bulletin of the MUSeUIll of Oomparatíve Zoology

AT HARVARD COLLEGE

VOL. 96. No. 1 1

..

,
•

A STUDY OF THE SNAKE, TACHYMENIS PERUVIANA

WIEGMANN AND ITS ALLIES

By WARREN F. WALKER, JR.

WITH FIVE PLATEB

•
•

CAMBRIDGE, MASS., U. S. A.
PRINTED FOR THE MUSEUM
N OVEMBER, 1945 •
 •
•

..

•
•

..

•
 No. 1. - A Study Di the Snake, Tachymenis peruoiana
• Wiegmann and Its Allies

CONTENTS
PAOE
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Histori cal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Geographic Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Composi tion of peruvia na. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Locali ties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Acknow ledgmen ts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 14
The peruviana Complex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 14
'I'acliumenis peruviana Wiegmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 14
'I'achumenis tarmensis spec. nov.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
•
Tachymenis affinis Boulenger. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22
The atten uata Complex . 24
Tachymenis attenuata auenuaia spec. et subsp. nov . 24
Tachymenis attenuata boliviana su bsp. nov . 26
The chilensis Complex. . . . . . . 28
Tachymenis chilensis chilensis (Schlegel) . 28
Tachymenis chilensis assimilis (Jan) : . 32
Tachymenis chilensis melanura subsp. nov . 35
· .
DlSCUSSlon. . . . . . . . . 38
Relationship of the Described Forms . 38
Ou tlying Species. . . . . 43
Redefini tion of Tachymenis. . . 44
Notes on the Habits of the Various Forms . 47
Key to Ttichumenis peruviana and AlIies . 52
Relationship of Tachymenis to other Boiginae . 53

Bibliography • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 54

Plates · · · . . . . . . . . . . .. 56

\
 4 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

INTRODUCTION ..

Historical
Tachqjmenis was established by Wiegmann in 1835 on the basis of a
specimen from Peru. The type was named peruoiama. Although
Wiegmann's characterization was adequate, many species from all
parts of the world were subsequently referred to the genus, which
naturally caused it to have very little real meaning. By the time of
Boulenger's Catalogue (1893-1896) Boettger, Boulenger, and Cope had
simplified the situation by placing the Old W orld forms in either
Tarbophis or Amplorhinus, and most of the New World ones in Ery
[Link], as follows:
Tachymenis dromiciformis Peters 1863 = Erythrolamprus dromiciformis
• - (Peters) 1863 .
Tachymenis vivax Günther 1864 = Tarbophis savignyi (Savigny) 1829.
Tachymenis fissidens Peters 1869 = Erythrolamprus imperialis (Baird
and Girard) 1859.
Tachymenis melanocephala Peters 1869 = Erythrolamprus lateritius (Cope)
1861.
Tachymenis taeniata Peters 1869 -= Erythrolam prus piceivittis (Cope)
1869.
Tachymenis piceivittis Günther 1872 =Erythrolamprus piceivittis (Cope)
1869.
Tachymenis nototaenia Peters 1882 = Amplorhinus nototaenia (Günther)
1864.
'I'achumenis bipunctata Garman 1883 = Erythrolamprus bipunctata (Gün-
ther) 1858.
Tachymenis fissidens Garman 1883 =Erythrolamprusfissidens (Günther)
1858.
Tachymenis imperialis Garman 1883 =Erythrolamprus imperialis (Baird
and Girard) 1859.
Tachymenis lateritia Garman 1895 = Erythrolamprus lateritius (Cope)
1863.
Tachymenis dicipiens Günther 1895 = Erythrolamprus dicipiens (Gün-
ther) 1895.
Tachymenis grammophrys Günther 1895 = Erythrolamprus qrammophrvs
(Dugés) 1888.

Boulenger completed the job in his Catalogue, and redefined Tachy

menas to include only peruciana, which he believed to range from cen
tral Peru through Bolivia to southern Chile, and affinis of central Peru.
The latter was a new species which he described. He included Coronella
•

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 5 •

chilensis Schlegel 1837 and Psammophylax assimilis Jan 1863 in

the synonymy of peruviana. AII of this was a significant contribution
toward an understanding of the genus.
However, in following years large numbers of peruviana (as rede
fined by Boulenger) were accumulated by the museums of the world. •

These were taken at elevations ranging from sea level to the high
Andean plateaus of 13,000 feet, and from localities extending from
central Peru to southern Chile a range in latitude equivalent to the
distance from N ew York to Panama 1 N aturally considerable geo
graphic variation was encountered, and similarly a great dealof varia
tion found in the specimens. Investigators began to feel that several
races or even species were involved. Werner (1898, 1901, 1904) recog- ~

nized five varieties of peruviana.· p. peruviana in Peru, p. vittata in cen-

tral Chile, p. coronellina in north central Chile, p. catenata in north

An indication of the climatic variation can easily be obtained from an examination of rainf'al
and temperature figures. Data from Carlson (1943) for a few selected stations follow:

Annual Period 01 Annual

Rainfall Heaviest Average Temperature
Station (Inches) Rain I Temp. (F) Ranoe
Cuzco, Peru. . . .......... 31.7 January 51.3 6.9
Areq uipa, Peru. . . . ...... 4.2 February 56.8 2.3
Antofagasta, Chile ....... 0.2 • • • • • • 63.0 12.4
Coquimbo, Chile ......... 4.5 June 58.3 10.4
Santiago, Chile .......... 13.8 JUlle 57.0 23.0
Valdivia, Chile .......... 105.5 June 52.9 16.4
Puerto Montt, Chile ...... 86.1 July 51.8 13.6

N.B. No specimens have been taken at Antofagasta. The record is merely included to indi
cate the aridity of northern Chile. Specimens have been taken at all other stations.
I

central and central Chile, and p. dorsalis in Bolivia. These varieties

were poorly defined and, since they only made matters confusing, they
have not heen generally recognized. Amaral (1930) followed Boulenger
in recognizing only peruviana and affinis in the are a from central Peru
through Bolivia to southern Chile. Hellmieh (1938), working primarily
with Chilean material, made another attempt to break up the apparent
vast array of forms ineluded in peruviana, hut carne to the conclusion
that only one widespread and very polymorphic species was involved.
The primary purpose of this work has been to review the situation
and make a more careful study of the variations of peruviana (as rede
fined by Boulenger) throughout its entire range, in order to deter
mine whether or not more than one species is involved, or, if only one,
whether or not there is any elinal sequence in the variation. In clarify
ing the status of peruviana, the generic type, 1have also considered its
•

6 BULLETIN: MUSEUM OF ••
ZOOLOGY
COMPARATIVE
 relation to affinis oí Peru and to the three outlying species which
have
heen added to the genus since Boulenger's Catalogue:
brasiliensis Gomes (1918) from Sao Paulo, BraziI; elonqaia Despax
(1910) frorn northern Peru; and surinamensis Dunn (1922) from
Surinam (Dutch Guiana). Thus, in a sense, this paper constitutes
a generic revision, hut, since no specimens of elongata and
brasiliensis are available and the descriptions are inadequate, my
conclusions regarding the outlying forms are tentative, and this
study centers around peruviana and its irnmediate alIies.
~

Geographic
Conditions
In order to understand the morphological variation found in
peru mana (sensu Boulenger) sorne knowledge of the geographic
conditions encountered by this snake in its wide range is necessary.
The bulk of the Peruvian and Bolivian specimens have come
from the Andean regio n from central Peru into Bolivia. Apparently
they are primarily highland snakes for they occur at Ieast as high
I

as 13,000 feet, and few have been taken at low elevations. The
lowest record on the west side
of the Andes is 1,630 meters (about 5,350 feet); on the east side a
few have worked their way farther down from the mountains and
have reached at least the edge of the tropical lowlands.
Specimens have been taken in the predominantly lowland
department of Madre de Dios in Peru and also from the city of
Santa Cl~UZ(elevation about 500
• meters, 1,640 feet) in Bolivia. Consequently many of these •

snakes have encountered alpine conditions with a vegetation

characterized by procumbent dwarf shrubs and low cushion or
rosette-shaped plants; others, the more favorable and moist
conditions found on descending the eastern side of the mountains,
where areas of grass steppes and scattered shrubs, savanna
and evergreen woods, and tropical rain forest are
encountered more or less in succession. Those from the western
slopes of the Andes in this region have found little precipita
tion due to prevailing south-easterly winds. Thus the climate at
the higher elevations is semiarid with a xerophytic vegetation
of cacti, low shrubs, and heath. Lower down, there is a belt of
desert crossed by fertile river valleys. This is complicated
along the narrow coastal plain by a rich lomas vegetation
dependant upon heavy fogs, hut Tachymenis does not extend this
low in Peru.
The Chilean specimens have been taken at lower elevations
along the coast and western slopes of the Andes in the central part
of the country from the province of Atacama to Chiloe. They are
apparently
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 7

absent frOID the extreme deserts of northern Chile where nitrate de

posits, salt basins, and intermittent rivers are the characteristic fea
tures. Neither are they found in the cold damp forests of the far south.
Thus these snakes have avoided the extremes, hut even in central
Chile they have encountered a wide range of topographic and climatic
conditions. The Andes tend to become progressively lower on going
south. At first the mountains ascend more or less directly from a nar
row coastal plain, hut to the south, a low coastal range of 1,000 feet
to 1,500 feet elevation and a broad central valley are interposed
between the Pacific Ocean and the Andes proper. In going south in
this region there is also an accompanying climatic change fully as
pronounced as that in topography. The temperature gradually de
creases, and the prevailing winds, which are off shore in the north,
shift to come in from the ocean. This causes a pronounced precipita
tion gradient. The northern deserts give way to semiarid conditions
•

in Atacama, and these to a mild mediterranean climate from the

southern part of the province of Coquimbo to Bio Bio. South of the
Rio Bio Bio forests appear and the climate is marine, a condition which
becomes very pronounced in the far south.

,
Cornposition of peruviana
Given the diversification of geographic conditions outlined aboye,
it is not surprising that peruviana, as hitherto considered, shows a
great deal of variation in coloration, squamation, dentition, and even
in penial characters. With the combined material of the American
Museum oí Natural History, the Chicago Museum oí Natural History,
the Museum of Comparative Zoology, and the United States National
Museum, enough specimens from all parts of the range were brought
together to be able to analyze the variation quite accurately. It was
soon obvious that peruviana was merely an omnium. qatherum,
With such a large amount of material (209 specimens from 59 locali
ties) , it has been relatively easy to separate the forma involved. Un
Ior.t. unately, however, sorne crucial regions where intergration might be
expected are poorly represented. Consequently, in breaking up the
array of forms previously included in peruviana, the question of which
are races and which species has been largely a matter of personal
judgment based on the degree of difference. 1 have no illusions as to
the permanency of my findings. Further revision will doubtless be
necessary as more material becomes available .
•

,
 •

8 BULLETIN: MUSEUM OF COMPARATIVE

••
ZOOLOGY I

As 1 see the situation, peruviana can be broken down into three

species complexes. One found in the Andean regions oí central and
southern Peru and Bolivia, a second along the edge of the tropical
lowlands of Peru and Bolivia, and a third along the coast of Chile.
These 1 have called the peruoiana, attenuata, and chilensis complexes
respectively. Names were available in the synonymy of peruviana for
many of the forms in these groups.
The peruviana complex includes peruciama of Wiegmann and tar
mensis to be described. 1 have also placed affinis of Boulenger in this
group. These species are characterized by a fairly stocky body forme
The body of the hemipenis is covered with spines of subequallength,
its base only has a few minute ones, and its tip is covered by small
spines which fuse at their bases to give a slightly calyculate appear
• ance. There are 5-12 solid maxillary teeth. Normally they have 1 pre
and 2 postoculars, and 8 supralabials with the fourth and fifth entering
the orbit.
The attenuata complex contains a new species with two races:
aitenuata attenuata and attenuata bolioiana. The body form of these is
quite elongate in comparison with members of the other complexes.
This is reflected in a greater number of caudals (60-69 as opposed to
37-55 in the other two groups), total ventrals (210-221 as opposed to
176-217), and ratio of taillength to totallength (0.22-0.24 as opposed
to 0.14-0.21). The spines on the body of the hemiphenis increase in
size posteriorly, the base has several very large spines as well as
scattered minute ones, and the tip is distinctly calyculate with only •

slight traces of spines. The solid maxillary teeth are more numerous
than in either other complex ranging from 12-16. The oculars are
variable, but the supralabials are still normally 8(4-5).
The chilensis complex contains three races arranged in north-south
sequence along the coast of Chile: chilensis assimilis \J an), chilensis
chilensis (Schlegel), and chilensis mclanura to be described. In all, the
body proportions, and consequently the ventral scales and ratio of tail
length to total. length, are close to those of perumarui, but the penis
resembles that of attenuata. The solid maxillary teeth range from 6-8.
There are typically 2 pre- and 2 postoculars, and the supralabials al"e
normally 7(3-4).
In summary peruviana as hitherto considered breaks up as follows:
The pe1·uvia'l~acomplex:
peruoiana Wiegmann
tarmensis spec. nov.
 ..

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 9

The attenuata complex:

attenuata attenuata spec. nov.
attenuata boliviana subsp. nov.
The chilensis complex:
chilensis chilensis (Schlegel)
chilensis assimilis (Jan)
chilensis melanura subsp. nov.
Boulenger's affinis appears to belong to the peruviana complex.
These three complexes will be described in detail. This will be fol
lowed by a discussion of their relationships to one another and to the
outlying forms of Tachymenis, my concept of the genus, and the habits
of the various forms.
Localities
In as much as South American collecting sites are frequently rather
unfamiliar towns, 1shall give a brief description of the localities from
which specimens have been examined before proceeding. In this list, 1
have followed the spelling on the American Geographical Society's
Map of Hispanic A7J1e1·icawhere possible in an attempt to standardize
the spelling of localities.' Unless specific and accurate elevations were
available, those of the American Geographical Society's maps have
similarly been followed. The writings of Bingham (1916), Bowman
(1916), Stiglich (1922-23), and Weberbauer (1936)have been extremely
useful in obtaining information about Peruvian localities, and Carlson
(1943) and Rudolf (1929) for Bolivian and Chilean places. The loca
tion of the department or province of any given Iocality can be deter
mined by consulting the accompanying map of the distribution of the
various forms (PI. 1).

Peruvian Localities:
Arequipa: The capital city of the coastal department of Arequipa.
Its elevation is about 7,500 feet. The region is semiarid except in
irrigated valley bottoms.
Chospiyoc: A town in the lower part of the Cuzco valley south oí
Ollantaytambo, department of Cuzco. Its elevation is 10,000
feet. The vegetation consists primarily of grassy steppes.
Chucuito: A town on the shore of Lake Titicaca just south of the
city of Puno, department of Puno. It is in a región of grass
, steppes. The elevation is 3,820 meterse
1 In accordance with present American trends, and in the interest of simnlicitv, 1 have dropped
all the Spanish accents from the place names, ..
•

10 BULLETIN: MUSEUM OF COMPARATIVE

••
ZOOLOGY

Collacachi: A town in the department of Puno, about five miles

southwest of Chucuito.
Cuajones: A mining camp located at an elevation oí 3,230 meters
in the upper tributaries of the Rio Moquegua, department of
Moquegua. The region is semiarid.
Cuzco: The capital city of the department of Cuzco. The city is
located at an elevation of 3,495 meters in a broad fertile valley
surrounded by grassy highlands.
Huacullani: A small town in the southern part oí the department
of Puno a bit northeast of Pisacoma. It is located at an elevation
of about 4,000 meters at the edge of the Titicaca basin very near
the Bolivian border. The vegetation is alpine .
•

Huanta: A large city in the northern part of the department of

Ayacucho. It is located at an elevation of 2,730 meters on a small
tributary of the lower Montaro in a region with a vegetation that
is transitional between the highland grass steppes and evergreen
# woods.
Huarocondo: A town in the central part of the Cuzco valley between
Chospiyoc and Cuzco, department oí Cuzco. lts elevation is
11,000 feet. The vegetation consists primarily of grassy steppes.
Huispang: A town in the department of Cuzco. It is probably in
the Urubamba valley northwest of the city of Cuzco. lts eleva-
tion is 12,175 feet. ,
Juliaca: A city in the department of Puno near the northwest end
of Lake Titicaca. lts elevation is 3,825 meterse The vegetation
•

consists primarily oí grassy steppes.

Macchu Picchu: Famous Inca ruins located aboye the deep Toron
toy gorge oí the Rio Urubamba, department of Cuzco. Elevations
• change rapidly in the region, but specimens from here are said to
have been taken at timberline at 12,000 feet elevation. The vege
tation is transitional between that oí the highlands and that of the
tropical lowlands.
Madre de Dios: .LL\. department oí southeastern Peru. Most of it
belongs to the tropicallowlands but there are a few foothills of the
Andes in the southwestern part which reach a height of 2,500
meterse It is not known from what part of the department the
speci•mens carne.
Ollantaytambo: A town in the Urubamba valley aboye the ruins
of Macchu Picchu, department oí Cuzco. lts elevation is 9,400
feet. The vegetation consists of grassy steppes.
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 11

Pachacayo: A town in the Montaro valley between Oroya and

Jauja at an elevation of about 3,500 meters, department of Junin.
The vegetation consists of grassy steppes.
Pisacoma: A small town located at an elevation of about 4,000
meters in the southern part of the department of Puno. It is very
near Huacullani. The vegetation is alpine.
Pucara: Inca ruins in the departrnent of Puno located near the
Cuzco railroad at an elevation of 3,880 meterse The vegetation is
alpine.
Puno: Capital city of the department of Puno located on the north
western shore of Lake Titicaca at an elevation of 3,820 meters.
The vegetation consists of grassy steppes.
Puquiura: A town in the Cuzco valley located at an elevation of
9,500 feet between uzco and Huarocondo, department of Cuzco.
The vegetation consists of grassy steppes.
San Anton: A town in the department of Puno located in the north
ern part of the Lake Titicaca basin at an elevation of 3,990 meterse
The vegetation is alpine.
Sicuani: A town in the departmen t of Cuzco along the railroad to ..
the capital. The elevation is 3,515 meters; the vegetation alpine.
Tarma: A town in the department of J unin encountered on the way
to the Chanchamayo región. lt is located near timber line at an
elevation of about 3,000 meterse
Toquepala: A mining camp located at an elevation of about 9,000
feet in the upper tributaries of the Rio Locumba, department of
Tacna. The climate is semiarid.
Urco: A hacienda in the upper part of the Urubamba valley aboye
the town of Urubamba, department of Cuzco. The elevation is
about 3,000 meters, The vegetation consists primarily of grassy
steppes.
Vitor: A town in the VitOI' valley below Arequipa, department of
Arequipa. lts elevation is 1,630 meters; the climate semiarid.
Yanama: A town in the Montaro valley just aboye Jauja, depart
ment of Junin. The elevation is about 3,500 meterse The vege
tation consists of grassy steppes.
Yauyos: A town in the upper tributaries of the Rio de Canete at an
elevation of 2,930 meters, department of Lima. The climate is
semiarid.
Yunguyo: A town in the department of Puno located along the
southern shore of Lake Titicaca at an elevation of 3,820 meterse
The vegetation consists of grassy steppes.
•
,
 •

1 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

Yura: A town in the department of Arequipa. It is located a bit

to the northwest of the city of Arequipa at an elevation of 2,575
meterse The climate is semiarid.
•

Bolivian Localities :
lncachaca: A town somewhere in the department oí Cochabamba.
Most of the department is below 3,000 meters, and has a vegeta
tion transitional between that of the highlands and that of the •

lowlands.
Santa Cruz de la Sierra: Capital city of the department of Santa
Cruz. The city is located at an elevation of about 500 meters
near the edge of the tropicallowlands.

Chilean Localities:
Angol: A city on the east side of the coastal range in the province
of Bio Bio. The mountains in the vicinity are under 1,000 meterse
The climate is slightly marine.
- Cerros de Aculio: A local range of mountains somewhere in the
southern part of the province of Santiago. Their elevation is
probably under 2,000 meterse The climate is mediterranean .
•
Chanarcillo: A town about 40 miles from the coast in an intermit
tent river valley of the semiarid province of Atacama. lts eleva
tion is between 500 and 1000 meterse
Chile: A few specimens have no data other than the country. One, •

with no data at all, is undouhtedly from Chile.

Chiquaihue: A town near Collipulli in the southern part of the
province oí Bio Bio. Its exact location is not known, hut Colli
pulli is located at an elevation of 200 to 500 meters on the main
Andean range just west of the central valley. The climate is
slightly marine.
Concepcion: 1t has been assumed that specimens with this data
have come from the vicinity oí the coastal city oí Concepcion
capital of the province oí Concepcion. The climate is transitional
between mediterranean and marine.
Coracol: A small town on the coastal range in the province of
Concepcion. The mountains are under 1,000 meters in elevation;
the climate transitional between mediterranean and marine.
Curacautin: A town on the western slopes oí the Andes in the
province of Cautin. lts elevation is about 500 meters; the climate
sligh tly marine .
•
 •
•

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 13

Domeyko: A semiarid range of mountains in the northern part oí

the province of Atacama. They reach a height of over 4,000
meterse It is not known from what part of the range the speci
mens carne.
El Tofo: A town located about 10 miles in from the ocean along the
semiarid coast of northern Coquimbo. lts elevation is between
500 and 1,000 meterse
El Vergel: A town located somewhere near Angol in the province
of Bio Bio. See Angol.
Fundo Hualpencillo: A town somewhere near Concepcion in the
province of Concepcion. See Concepcion.
Las Condes: A town in the province of Santiago, but its exact
location is not known. See Santiago.
Las Tundras: A small town near Chillan, the capital of the province
of Nuble. Chillan is located in the central valley at an elevation
of less than 200 meterse The cJimate is mediterranean.
Mafil: A town located in the central valley of the province of Val
divia at an elevation of less than 200 meterse It is along the rail
road connecting Valdivia with Cautin. The climate is marine.
Nahuelbuta Hills: Part of the coastal range west of Angol in the
province of Bio Bio. The mountains are under 1,000 meters
elevation. The climate is slightly marine. ..
Puerto Corral: An ocean port in the province of Valdivia just south
of the city of Valdivia. The climate is marine.
Puerto Montt: A port on the Gulf of Ancud in the northern part of
the province of Chiloe. Tbe climate is marine.
Punta Teatino: A small village located along the semiarid coast of
Coquimbo just north of La Serena.
Santiago: The capital of Chile, located in the province of Santiago.
The mountains in the province reach a height of about 5,000
meterse The climate is mediterranean. The city of Santiago is
located in the beginning of the central valley at an elevation of
about 500 meterse Sorne specimens with this locality are known
to have been taken near the city ; others from elsewhere in the
department.
Talcaguano: A coastal village just north of the city of Concepcion
in the province of Concepcion. The climate is transitional from
mediterranean to marine.
Traiguen: A town in the central valley in the northern part of the
province of Cautin. lts elevation is under 200 meterse The cli
mate is slightly marine .
•
•

1 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

Valparaiso: The chief sea port of Chile, being the port of entry for
Santiago. It is located in the southern part of the province of
Aconcagua. The climate is mediterranean.

Acknowledgments
At this time 1wish to thank Mr. C. M. Bogert, Dr. D. M. Cochran,
Mr. A. Loveridge, and Mr. C. H. Pope for the loan of the material
which made this study possible, and for allowing me a free hand in its
utilization. I also wish to express my deepest gratitude to Dr. T. Bar
bour, Mr. A. Loveridge, and Mr. K. P. Schmidt for suggestions and •

help during the course of these investigations, and to Dr. F. M. Car-

penter for assistance in making the photographs.

The PERUVIANA Complex

T ACHYMENIS PERUVIANA Wiegmann
Plate 2, fig. 2; Plate 3, figs. 5-9.
1835 Tachymenis peruviana Wiegmann, N. Acta. Ac. Leop.-Carol., 17, p. 252,
pl. 20, fig. 1 (type locality Peru). Ophis peruoiana Fitzinger, 1843,
Syst. Rept., pt. 1, p. 28. 'I'achsmienis peruuiarui Peters, 1863, Monatsb.
K. Preuss. Akad. Wissensch. Berlin, p. 275 (in partj.! Boulenger,
1896, Cato Sn. Brit. Mus. Nat. Hist., 3, p. 118 (in part: Aschiquiri,
Bolivia; Larecaja, Bolivia; Lake Titicaca). Barbour and Noble, 1921,
Pl'OC.U.S. Nat. MLlS.,68, p. 618 (in part: Pucyura- at 9,500 feet,
Chospiyoc at 10,000 feet, Huarocondo at 11,000 feet, Ollantaytambo •

at 9,400 feet all localities in the department of Cuzco, Peru).

Escomel, 1929, Fauna de Arequipa, p. 9. Amaral, 1930, Mem. Inst.
Butantan Sao Paulo, 4 (1929), p. 210 (in part). Hellmich, 1938,
Rev. Chilena Hist. Nat., 41, p. 108 (in part: Bolivian localities
Carquayeoleo near Rio Muleto in the department of Uyuni," Cerro
Sapo in the department of Cochabamba, Ayopayo at 3,200 meters
in the department of Cochabamba). Schmidt and Walker, 1943,
Zool. Ser. Field Mus. Nat. Hist., 24, pp. 290, 315 (in part: city of
Arequipa in Peru and sorne of the other specimens believed to be from
the department of Madre de Dios in Peru)."
1845 Ophis peruana Tschudi, Fauna Per., Herp., p. 58 (nomem nudum.
Tschudi had no specimens. He was merely verifying Fitzinger's
placement of peruuiana into Ophis, but misspelled the name) .
•

1After this had gone to press my attention was called to another reference to T. peruviana
as follows; Garman, 1875, Bull. l\1.C.Z., 3, p. 278 (Cuzco Valley).
2 This town is spelled "Puquiura" on the maps of the American Geographical Society.
3 See footnote 1, p. 20.
4 One other reference to 'I'achsrmenis peruviana in the literature results from a misidentification •

of the specimen: Amaral, 1925, Proe. U.S. Nat. Mus., 67, Art. 24, p. 14 (U.S.N.M. 12277 from
Guayaquil, Ecuador = Dryophylax nattereri Mikan):
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 15

1854 Dipsas chilensis Duméril and Bibron (not of Schlegel) Erpét. Gén., 7,
p. 1159 (in part).
1901 Tachymenis per1/¡vianavaro dorsalis Werner, Abhandl. und Berich. Kon.
Zool. Anthrop.-Ethnog. Mus. Dresden, 9(2), p. 8 (type locality =
Bolivia).
1915 Leimadophis andicolus Barbour, Pl~[Link]. Soco Washington, 28, p.
149 (type Iocality==Huispang at 12,175 feet, department of Cuzco,
Peru).
•

Wiegmann (1835) does not give a specific type locality for peruoiana,
so it is here restricted to the city of Puno in the department of Puno.
The choice of Puno is based on the fact that Dr. F. J. F. Meyen, who
collected the type specimen during his trip around the world in 1830-32,
mentions visiting this city in an account of his travels (1834). More
over, specimens from Puno agree very well with the figure and descrip
tion of the type. The type was probably a female judging by scale
count and tail shape. Because of the importance of perueiana as the
type of the genus, it is redescribed in detail on the basis of specimens
from Puno. These are then compared with specimens from other parts
of the range to determine the arnount of variation.
jfaterial Examincd. A total of 111 specimens (55 Q, 56 cf) refer
able to peruouma have been studied. A list of the localities from which
they carne follows:

Peruvian Localities:
Department of Arequipa
Arequipa: 1 Q, 1 el' (C.M.N.H. 34055, 34061) •
Vítor: 1 el' (C.M.N.H.34260)
Yura: 2 cf (C.M.N.R. 34201-2)
Department of Ayachucho
Huanta: 6 Q, 6 el' (C.M.N.H 39647-58)
Department of Cuzco
Chospiyoc: 1 el' (U.. N.M. 60740)1
Cuzco and the Cuzco Val1ey: 4 Q, 3 el' ((~.M.N.H. 40216-19,
M.C.Z. 3584-86)1
Huarocondo: 4 Q, 2 el' (M.C.Z. 12412-3; U.S.N.M. 60697-8,
60722)1
Huispang: 2 cf (M.C.Z. 10986-87)2
Ollantaytambo: 1 Q (U.S.N.M. 60743)1
Puquiura: 2 c , 1 cf (U.S.N.M. 60681-3)1
Sicuani: 1 ~ (A.M.N.H. 36154)
Urco: 4 ~,3 cf ( .M.N.H. 34105-07, 34131-33, 34142)

•
 ,

• 16 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

Department of Junin
Pachacayo: 2 v , 2 O' (M.C.Z. 42435-38)
Yanama: 1 O' (M.C.Z.45916)
Department of Lima
Yauyos:l~, 10' (M.C.Z.45914-15)
Department of Madre de Dios: 3 c , 4 O' (C.M.N.H. 40070,
40073-78)1
Department of Moquegua
Cuajones: 3 ~,4 O' (M.C.Z. 46651-57)
Department of Puno
Chucuito: 1 c , 1 O' (C.M.N.H.40214-15)
Collacachi: 2 ~,4 O' (C.M.N.H. 34156-61)
Huacullani: 1 ~ (C.M.N.H.40213)
Juliaca: 1 ~ (C.M.N.H.34086) •

Pisacoma: 1 ~ (C.1VI.N.H.40212)
Pucara: 1 ~,2 O' (M.C.Z.45903-05)
Puno: 2 c , 3 O' (A.M.N.R. 36155, C.M.N.H. 34259, U.S.N.M.
109015-17)a
San Anton: 2 v , 1 cJ (A.M.N.R. 5256, C.M.N.R. 4065-66)
Yunguyo: 11 ~,10 cJ (C.M.N.H.40174-94)
Department of Tacna
Toquepala: 1 cJ (M.C.Z.45941)
,
Bolivian Localities:
Department of Santa Cruz •

Santa Cruz de la Sierra: 1 ~ (A.M.N.R.8793) •

Diagnosis. This species is the type of the genus Tachymenis, and

forms the basis of the peruoiana complexo The characters of this com
plex have been discussed in the introduction. T. peruouma differs from
other members of the group chiefly in having 19 midbody scale rows,
dorsal scales with single apical pits, and generally a distinctly spotted
color pattern.
Descriptioii of topotypes. Five topotypes have been exarnined. AII
are from the city of Puno in the department of Puno. These specimens
al-e A.M.N.H. 36155, a male collected by Beck; C.M.N.H. 34259,
also a male collected by K. P. Schmidt on October 28,1939; U.S.N.M.
109015-17, one male (109016) and two females collected by J. Soukup.
All are snakes of moderate size. The body is rather stocky and cylin
drical with a more or less oval shaped head that is only slightly distinct
from the neck. The eye is of moderate size with a pupil that is sub-
•

•
• •

•
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 17

elliptical in shape. The hemipenis is bifurcate with a bifurcate ductus

sperrnaticus. Its tip is covered with small spines which fuse at their
bases giving a slightly calyculate appearance. The body of the hemi
penis is covered with larger spines of subequal size. There are no greatly
enlarged basal spines, but there are a few scattered minute ones. The
three topotypes whose dentition was examined have 6 solid maxillary
teeth which are subequal or increase slightly in size posteriorly. These
are followed after a gap by a pair of enlarged grooved fangs. The man
dibular teeth deerease in size posteriorly after the first few.
The rostral is broader than deep, slightly visible from aboye; inter
nasals a bit shorter than the prefrontals; frontal1Y2 times as long as
broad, longer than its distance from the end of the snout, as long as the
parietals; supraocular of normal size; nasal en tire or semi-divided;
loreal as deep as long or a bit deeper: 1 preocular reaching the top of the
head but not contacting the frontal; 2 postoeulars in three of the speci
mens, 1 in the other two; temporal s 2-3 in half the cases, 1-2 or 1-3 in
the others; supralabials 8(4-5); infralabials 9 in most of the specimens
with the first four in contact with the anterior pair of chin shields
which are as long' or longer than the posterior pair, but combinations
of 8(3) and 10(5) were also found.
The dorsal scales are smooth, have single apical pits, and a reduction
pattern that is typieally 19-19-]5, but one specimen has only 14 pre- .
anal rows. The abdominals are rounded, 141-147 in females and 144-
147 in males; anal divided; eaudals paired, 43-45 in females and 47-51
in males; total ventrals 186-190 in females and 191-198 in males.
The dorsal ground color of these snakes is light brown or grey dusted
with black. In addition, most of the dorsal scales bear a single blaek
apieal dot which is about the size of and in the general position oí the
apieal pite There is a short light rniddorsal line on the neek. This
narrows posteriorly and breaks up into a series of small middorsal
•
spots. There is a short black stripe on either side of the mideervical
linee Posteriorly this too breaks up into a series of spots. These spots
are of rnoderate size and dark. Those of opposite sides alternate a bit
and sorne fuse aeross the midline forming faint zigzag markings.
Laterally, on scale rows four and five, there is another series of dark
spots whieh fuse posteriorly to form a dark lateral linee The head

1 Sorne of these specimens have already been discussed in the literature which is listed in the
synonymy.
2 Type and paratype of Leimadophis arulicolus Barbour 1915. This specimen was reidentified
as T'achirmenie peruviana by Barbour and Noble in 1921.
3 Topotypes of peruviana.

•
 •

18 BULLETIN: MUSEUM OF COMPARATIVE Z

••
OGY
OOL

coloration is also brownish or grey with irregular dark markings. A

small black line runs from the snout through the eye to the angle of the
mouth. It is best developed behind the eye. A similar line extends
•

• from the parietals to the angle of the jaw, There is a small black mark
under each eye. The supralabials may also have a few other dark mark
ings, and the infralabials and gulars a few scattered ones as well. The
ventral ground color is light grey or yellow. The abdominals and
caudals are covered with irregular dark brown or black markings.
The markings in the center of the abdominals and caudals are scattered,
but laterally they tend to form a linee
The largest topotype, a female, has a totallength of 375 mm.; tail
65 mm. The ratio of tail length to total length is 0.16-0.17 in the
female topotypes; and 0.18 in the males.
TTaruüion, The remaining material agrees very well with the topo
types as regards major characters, but differences in, detail al'e com
mon. The body form, penial characters (fig. 2), and dentition are the
same in all. The number of solid maxillary teeth ranges from 6-10
and averages 8 in twenty specimens which have been examined from
various parts of the range. The type was said to have 5 such teeth.
There are also sorne differences in the proportions and number of
cephalic plates. The internasals are either as long as or shorter than the
prefrontals; frontal 1}.1 to 2 times as long as broad, longer than its
distance from the end of the snout, about as long as the parietals but it
deviates a bit in either direction; nasal single or semidivided, rarely
divided. Generally there is only 1 preocular, but 2 are present in 16 •

head halves. Typically there are 2 postoculars, but these fuse with
one another or with a supralabial to form 1 in 18 head halves. The
temporal formula is 2-3 in the majority of cases, but variations are
common. Combinations of 1-2, 1-3, 2-2, and 2-4 occur in 65 head
halves. Nearly all of these combinations could easily have been de
rived from a basic 2-3 pattern. The supralabials typically have a
formula of 8(4-5). However, two of the first labials have fused ill 13
head halves to give a combination of 7(3-4), one other specimen has a
count of 8(3-4-5) on one side of the head, and in two additional cases
a preorbitallabial has divided to give a formula of 9(5-6). There are
usually 9 infralabials, the first 4 or 5 of which contact the anterior
pair of chin shields, but variations of 8(3), 10(4), 10(5), and 10(6)
OCCUI' in 28 head halves.
The dorsal scales in all specimens are smooth and have a single
apical pit, but the pit is very poorly developed in certain populations.
Only a small amount of variation is encountered in the reduction
 •

•
WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 19
•

pattern of the dorsal scales. One specimen has 17 cervical rows and
two others have 20. Only one deviates at midbody with a count of 18.
Four specimens have 16 or 17 preanal rows, but in half oí these the
numher is reduced to the typical 15 a hit posteriorly. Two specimens
have as few as 14 preanal rows. The number of female abdominals
ranges from 134-152 with an average of 144; males 136-158, average
..
148. The anal plate is divided in all, and the left portion generally
overlaps the right, hut the reverse is the case in two mutant individuals,
and there is a partial fusion of the two halves in another specimen.
The female caudal range is 37-50, average 42; males 44-55, average 49.
These caudals are typically paired, hut in eight cases 2-6 were single.
The female total ventral range is 176-196, average 186; males 184-209,
average 197. Scale counts reported in the literature lower the male
• ahdominal and caudal range to 135 and 37 respectively, hut do not
otherwise affect the aboye ohservations.
The majority of the specimens are quite similar to the topotypes in
general color pattern, but may differ in a few details. The typical
colora tion can be seen in figures 5-7 . Varia tions occur in the row of
light middorsal spots which fuse to form a wavy line in many speci
mens. Other deviations are the failure of any of the large dorsal spots
to fuse aCI"OStShe midline; the reduction of the series of lateral
spots; more perfuse dark markings on the lahials; and a nearly
complete absence of dark markings on the abdominals and caudals.
y oung specimens have a coloration similar to that in typical adults.
Aside from the aboye departures, several distinct color phases are
present in the material at hand. About one half of the Madre de Dios
specimens are so melanistic that the typical pattern can only be de
tected with difficulty (fig. 9). The specimens from Puquiura are also
quite dark and in addition have a third dark diagonalline on the head.
This one extends from the first supralabial to the middle of the infra
labials. In specimens from Yauyos (Lima), Chospiyoc (Cuzco),
Huispang (Cuzco), and Bolivia the light middorsal line and dark
lateral lines are very pronounced, and the dark dorsal spots are a hit
Iarger than usual (fig. 8). As near as can be told from the description,
this race was first noticed [Link] by Werner (1901) and described
by him as dorsalis. However, its reappearance in scattered localities
shows that the race is not tenable. It would similarly be unwise to
name the other color varieties for they occur with or near typical
members, and are identical with them in other characters. The only
conclusion that can be reached is that peruviana is very polychromatic.
The females examined have an average total length of 368 mm.,
•

•
 •

- 20 BULLETIN: MUSEUM OF' COMPARATIVE ZOOLOGY

•

tail 59 mm.; in males the totallength averages 405 mm., tail 72 mm.
The largestfemale was taken at Urco and has a totallength of 626 mm.,
tail 99 mm. The largest male is from Vitor and has a total length of
656 mm., tailll1 mm. The ratio of taillength to totallength ranges
from 0.14-0.18 in the females, and from 0.16-0.21 in the males.
Ramqe. From the aboye data it is apparent that peruviana is a snake
of the .entral Andes. lts range extends from the departments of
Junin and Lima in central Peru south along the Andean chain to south
ern Bolivia.' Actuallocalities at which the species has been taken can
be found in the list of literature and material examined. lt is a fairly
eommon snake, but appears to be most abundant in the Andes of
southern Peru. This, however, may only be a result of more extensive
eollecting in that area. It has reached an elevation in southern Peru •

of nearly 4,000 meters at several points: Huacullani, Pisacoma, San •

Anton. It is thus essentially a highland form, but it does extend from

the upland valleys and plateau down on both sides of the Andes. The
lowest point it is known to have reached on the west coast is at Vitor
(1630 meters), but it has been taken as low as Santa Cruz in Bolivia
(c. 500 meters) on the east side. The vegetation in all of these regions
eonsists primarily of grass steppes as will be discussed later."
Remarks. As can be seen from the aboye discussion, Tnclujmenis
peru;via1~a is a wide spread form even when it is considered in the re
stricted sense used in this papero 1ts range extends along the Andean
ehain for nearly 1,000 miles. Most of the specimens in this area are
found in the highland valleys, but the species can apparently migrate •

to a certain extent from one valley to another. This seems too he

enough to scatter the variation and keep the species intacto Beeause
•
of the semi-isolation of each population, the variation is widely seat
tered and not arranged in any elinal sequenee. This is characteristic
of species in mountainous regions. Each population differs from neigh
boring ones noticeably in both squamation and coloration, and sorne,
such as those from Puquiura and Madre de Dios, differ enough to be
regarded as incipient races. However, it is dangerous to name such
forms. Tachumenis peruoiana would be an excellent form for the study
of speciation in mountainous areas. Although 111 specimens have

1 The southernmost record appears to be Carq uayeoleo (Hellmich, 1938). 1 can not find this
town, but it i said to be near Rio Muleto in the depar tment of Uyuni. N o department oí Uyuni
is shown on the American Geographical ociety's map . The only clue 1 can find to the location
oí Carquayeoleo i: a town of Uyuni in the department of Potosi. This town i near aRio Mulato .
2 After this was [Link], rny attention was called to a specimen from Bom Jesus de Lapa,
Bahia, Brazil (Carnegie Museum 347) which was identified by Griffin (1916) as 'I'achumenis
peruviana. 1 have reexamined this specimen and agree with the identification, but am skeptical
as to the accuracy of the locality record .
•
f

..

. WALKER: THE SNAKE TACHYMENIS PERUVIANA WIEGMANN 21

,
been examined, most are from southern Peru. The other parts oí the
range are poorly known.

TACHYMENIS TARMENSIS spec. nov.

•

Plate 4, figs. 10, 11.

Type. ..M.N.H. 5698, an aduIt female from Tarma in the depart
ment of Junin, Peru. lt was collected by M. P. Anderson on May 19,
1914.
Diaqnosis. It is a Tachymenis cIosely reIated to peruviana, but
differing in having a few more solid maxillary teeth (12 as opposed
to a maximum of 10 in peruviana), a few more caudals (55 as opposed
to a maximum of 50 in females of perusiosui) which causes the ratio of
taillength to totallength to be higher, absence of scale pits, and a very
light dorsal ground color with practically no trace of a pattern.
Description of type. The body form, pupil shape, and dentition are
quite similar to the corresponding characters of peruviana. There are
•

however, 12 sol id maxillary teeth.

The rostral is broader than deep, visible from aboye; internasals
about as long as the prefrontals; frontal twice as long as broad, longer
than its distance from the end oí the snout, as long as one and shorter
than the other parietal; supraocular about the size of the frontal;
- nasal entire; loreal a bit longer than deep; 1 preocular; 2 postoculars;
temporals 2-2; supralabials 8 (4-5); infralabials 9(5) on the right side
and 8(4) on the left; anterior chin shields longer than the posterior.
The dorsal scales are smooth, lack apical pits, and have a reduction
pattern of 19-19-15. The •abdominals are rounded, 135; anal divided;
caudals 55, 3 being single and the rest paired; total ventrals 190.
The top of the head and body is nearly uniform greyish brown in •

color (figs. 10-11). There is no trace of a pattern except for a bit of

black at the base of a few of the dorsal scales. The color of the under
side is grey stippled with black. This stippling becomes hea vier pos
teriorly.
The snake has a total length of 393 mm.; length oí tail 83 mm.;
ratio of tail length to total length 0.21.
Remarks and Ramqe. This form resembles peruoiana in general
structure. lt differs only a bit in such quantitative characters as num
ber of teeth, number of caudals, and in the ratio of taillength to total
length. lts chief departure is in the absence oí scale pits which are
often very poorly developed in peruoiana, and in its unusual coloration.
Using this degree of difference and its geographic position at the
 •

-
22 BULLETIN: MUSEUM OF COMP ARATIVE ZOOLOGY •

northern part of the range of peruviana as criteria, one would expect

tarmensis to be a race of peruviana. Perhaps it is, hut until males be
come available to check on penial characters it is hest considered as
a full species helonging to the peruviana complex. It is probably a
rather obscure form, for it is only represented by the type from Tarma,
and the Tarma region is quite well known .
•

T ACHYMENIS AFFINIS Boulenger

Plate 4, fig. 12.
1896 Tachymenis affinis Boulenger, Cat. Sn. Brit. Mus., 3, p. 319, pl. 7, fig. 1
(type locality = Muna-, Peru). Amaral, 1930, Mem. Inst. Butantan
Sao Paulo, 4 (1929), p. 209.
1921 Tachymenis peruoiana Barbour and Noble (not of Wiegmann), Proc.
U.S. Nat. Mus., 68, p. 618 (part: Cedrobamba ruins- at timber line,
12,000 feet, department of Cuzco, Peru).

M aierial. Tccliumenis affinis has previously heen known by the

•
single type in the British Museum. One additional specimen turned
up during an examination of material from the National Museum.
This specimen, U.S.N.M. 60721, an adult male, was collected at the
Inca ruins of Macchu Picchu in the department of Cuzco, Peru by
E. Heller in June 1915 while he was a member of the National Geo
graphic Society Yale University Peruyian Expedition. It was first
identified as pe1·uvia'l~aby Barbour and Noble (1921) in their report of
the Expedition's reptiles. But since it has 17 midbody scale rows and •

, a reduced color pattern, there is little doubt hut that it is affi'lLis. This
specimen enlarges our knowledge of this rare species. It is described
• and compared with the type below .
Diagnosis. It is a Tachumenis differing from all other members oí
the genus in having a dorsal scale reduction pattern of 17-17-15 in
stead oí 19-19-15. It further differs from peruviana in having a re
duced color pattern, hut resembles it so much in other features that
it has been included in the peruvioma complex.
Description. T. affinis is similar to peruoiama in body form, penial
characters, and general dental pattern. The specimen in question has
9 solid maxillary teeth that are subequal in size and are followed after
Stiglich (1922) cites two localities with this name in Peru
1 one along the northern coast of
Ancash, and one on the upper part of the Rio Huallaga in Huanuco. 1 am not sure which .is the
type locality but it is probably the latter since the latter is the larger of the two and 18 in the
highlands (2,000-2,500 meters).
This specimen (U.S.N.M. 60721) is discussed below. The Cedrobamba ruins are now better
2
known as the ruin s of Macchu Picchu.
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 23
•
a gap by a pair of enlarged grooved fangs. The type had 15 solid
maxillary teeth. This may seem like a great discrepancy, hut the
maxillary range in peruviana is at least 5-10.
The cephalic plates of the specimen examined are quite similar to
those of peruviana except that the frontal is only 1~ times as long as
broad, and is shorter than the parietals. The nasal is semidivided.
There is 1 preocular; 2 postoculars; temporals 2-2 on the right side,
shrivelled on the left; supralabials 8(4-5); infralabials 9(5). The
cephalic pIates in the type were essentially the same. lts temporals
were 2-3.
The dorsal scales are smooth, have single apical pits, and a reduc
tion pattern of 17-17-15. The abdominals are rounded, 147; anal
divided; caudals 50, 6 of these are single and the rest paired; total
ventrals 197. The type, also a male, had 153 ahdominals, 57 caudals,
and 210 total ventrals.
As can be seen, the specimen of affinis from Macchu Picchu agrees
very well with the type in the aboye characters, hut it does differ a bit
in coloration. Both are brownish aboye, but the dorsal scales are
darker than the lateral ones in the Macchu Picchu specimen (fig. 12)
while the lateral scales are said to be darker in the type. However, 1
do not believe that this is too significant for the color pattern of both
specimens seems to have been derived from a condition similar to that
in peruviana by a process of reduction of the prominent dark stripes
and spots. Both specimens have fair remnants of the head markings
of peruviana. The type has traces of the rows of dark dorsal spots.
These are obscure in the Macchu Picchu specimen, but traces of the
light middorsal line can be seen. Both lack the black apical dots tha t
are so prominent in the dorsal scales of peruviana. The ventral scales
of the Macchu Picchu specimen are darker than in the type.
The snake examined has a total length of 454 mm.; tail 84 mm.;
ratio of taillength to totallength 0.18. The type was a bit larger, and
had a ratio of 0.21.
Remar/es and Range. T. affi'nis appears to be a very rare form of the
highland valleys of Peru whose range is known to extend from Muna
in Huanuco to Macchu Picchu in Cuzco. It must consequently over
lap a large part of the range of perunnama although the two have not
yet been taken together. This fact together with the marked differ
ence in the reduction pattern of the dorsal scales precludes the possi
hility of the two forms heing races of the same species. There is no
doubt that affi1¿is is a separate species, hut 1 have placed it in the
perunnama complex since the two are so close in penial characters.
 •
• 24 BULLETIN:MUSEUMOF COMPARATIVZEOOLOGY ••

The ATTENUATA Complex

/

TACHYMENISATTENUAT ATTENUATAspec. et subsp.

nov.
Plate 2, fig. 3; plate 4, figs. 13-15.
1943 Tachymenis peruoiana Schmidt and Walker (not of Wiegmann), Zool.
•
Ser. Field Mus. Nat. Hist., 24, p. 290 (in part : sorne of Peruvian
specimens) .

Type. C.M.N.H. 40071, an adult male collected by Sr. Juan Perea,

and originally deposited in the University of Arequipa. The specimen
has no further data than Peru, but such material in the University's
collection is believed to be from the department of Madre de Dios
(Schmidt and Walker, 1943) ..
Diaqnosis. It is a Tachumenis differing from members of the per'll
viana complex primari1y in its more attenuate body proportions. 'I'his
results in a greater number of caudals and total ventrals, and in a
greater ratio of tail length to total length. Other differences can be
found in the condition of the hemipenis, and in the dentition. AII of
• these have been discussed in the introduction. This race differs from
boliciana, to be described, in having a slightly lower number of solid
maxillary teeth (12-14 as opposed to 14-16), and of abdominals and
caudal s (in males 148-150 and 60-64 as opposed to 152 and 69), and
in having a speckled instead of a checkered color pattern. ~
Description of type. The body form is quite attenuate in comparison •

with peruoiama; subeliptical in section. The head is moderately dis

tinct from the neck. The eye is of average size with a subeliptical
shaped pupil. The hemipenis (fig. 3) is bifurcate with a bifurcate
ductus spermaticus; tip distinctly calyculate with only slight indica
tions of spines at the top of the folds; body spinose with the length of
the spines increasing posteriorly; base with a row of 4 very Iarge spines
followed by a few scattered minute ones. There are 12 solid maxillary
teeth which increase slightly in size posteriorly and are followed after
a gap by a pair of enlarged fangs; mandibular teeth decrease in size
posteriorly after the first few.
The rostral is broader than deep, visible from aboye; internasals
shorter than the prefrontals; frontal 172 times as long as broad, longer
than its distance from the end of the snout, distinctly shorter than the
parietals; supraocular of normal size; nasal entire; loreal deeper than
long; 1 preocular extending to the top of the head but not contacting
the frontal; 2 postoculars; temporals 2-2; supralabials 8(4-5); 8 in-
..
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 25

fralabials with the first 5 in contact with the anterior pair oí chin
shields which are as long as or longer than the posterior pair.
The dorsal scales are smooth, lack apical pits, and have a reduction
pattern of 19-19-15; abdominals rounded, 150; anal divided; caudal s
paired, 60; total ventrals 210.
The ground coloration aboye and below is a light brown which is so
heavily stippled and flaked with hlack that the general appearance is
dark grey. However, upon careful examination traces oí a pattern
similar to that of peruoiana can be seen beneath all oí this. (See figs.
13-15). These traces al'e especially prominent on the head and neck,
for a diagonal black line runs from the eye to the cerner of the mouth,
a second one extends from the parietals to the angle of the jaw, and •

there is a distinct black stripe on either side of the midline on the

back of the neck. A small black line also runs along the sides of the
belly hut becomes obscure posteriorly for the general black stippling
becomes very dense.
. The specimen has a total length of 582 mm.; tail 126 mm.; ratio of
taillength to totallength 0.22.
Notes parat1Jpe. On paratype, C.M.N.H. 40072, also an adult
maleon has the same data as the type. It is also quite similar to the type
•

in genel·al hody form, penial characters, and dentition. lt has 7 large

basal spines' on the hemipenis, and 14 solid maxillary teeth. The •

cephalic plates are the same with the exception of minor differences:
2 preoculars; temporals 2-3 on the right side and 2-2 on the left; in
fralabials 8(5) on the right side and 9(5) on the left. The dorsal scales
have the same reduction pattern, but have traces of two apical pits I
There are 148 abdominals, 64 caudals, and 214 total ventrals. The
greatest departure is in coloration. The head, neck and belly are simi
lar to the type, hut the back is a nearly uniform light greyish brown
with a mínimum of black stippling and flaking. lts general appearance
is close to that of tarmensis. The parátype has a total length of 581
mID.; tail 131 mm.; ratio of taillength to totallength 0.23.
Remarlcs and Ranqe. The slight indication of scale pits in the para
type is puzzling. This fact together with the aberrant color might
warrant the recognition of this specimen as a separate form, hut this
does not seem advisable at this time for the two resemble each other in
all other characters, and Tachumenis is known to be very polychromatic
and to show sorne variation in terms of scale pits.
It is unfortunate that more specific locality data are not available
for this race. Both this form and peruouma are found in the depart
ment of Madre de Dios. lts more elongate body forro suggests the
•
 •

26 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

•

possibility that it is a tropicallowland type, while perueiama may come

frOID the mountains in the southern part of the department.
Morphologically, a. attenuata resembles tarmensis in the greater
number of solid maxillary teeth, lack of scale pits (at least in the type),
and greater number of caudals. But this does not necessarily indicate
• relationship. T. tarmensis is presumably a local northern form so the
two are separated by a large geographic hiatus. Furthermore, al
though the coloration of a. attenuata is variable it is nearer to that of
peruviana than to that of tarmensis. Its resemblances to tarmensis
couId easily have been arrived at through parallel evolution. It seems
more probable that both attenuata and tarmensis evolved indepen- I

dently from peruviana stock ..

TACHYMENISATTENUATABOLIVIANA subsp.
nov.
Plate 4, fig. 16.
Type. A.M.,N.H. 36020, an adult female taken at an elevation of
2,500 meters at Incachaca in the department of Cochabamba, Bolivia
by J. Steinbach.
Diagnosis.
•
It is a Tochumenis which resembles a. attenuata very •

closeIy. As mentioned aboye, it differs primarily in coloration, but

this is supplemented by slight differences in the number of solid maxiI-
lary teeth, abdominals, and caudals. '
Description of type. The general body form, pupil shape, and denti •

tion are the same as in a. attenuata. There are, however, 16 solid

maxillary teeth.
The proportions of the cephalic plates are also the same. There are
2 preoculars; 2 postoculars; temporals 2-3; supralabials 8(3-4-5); in
fralabials 10(5) on the right and 9(4) on the left.
The dorsal scales are smooth, lack apical pits, and have a reduction
pattern of 19-19-15; abdominals rounded, 149; anal divided; caudals
paired, 61; total ventrals 210.
The color pattern (fig. 16) is closer to the color of peruviana than is
that of a. aiienuaia. The dorsal ground color is light brown. On either
side of the light middorsalline, there is a row of fairly large uniformly
shaped dark spots. Those oí opposite sides alternate with one another,
fuse with one another at their corners across the midline, and also
alternate with and fuse with a less distinct row oí lateral spots. The
net result is a checkered pattern which is especially prominent an
teriorly. The various rows of spots can be easily homologized with

•
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 27
•

similar rows in peruoiama, but the spots are not as large and generally ,
do not fuse in peruoiama. The underside is a light brown flaked with
black. Anteriorly these black markings are arranged into four longi
tudinal lines, but posterior1y they merge with the darkened ground
color. The head and neck markings are the same as in typical peru-
•
m ama.
The type has a total length of 576 mm.; tail 125 mm.; ratio of tail
length to totallength 0.22.
Notes on paratypes. Two paratypes A.M.N.H. 36015 (female) and
A.M.N.H. 36017 (male) have the same data as 1he type. The female
specimen resembles the type very closely. It has 14 solid maxillary
teeth. There is 1 preocular; 2 postoculars on the left side and 1 on the
right; temporals 2-3; supralabials 9(5-6); infralabials 9(4) on the left
side and 9(5) on the right. There are 156 abdominals; 65 caudals, and
221 total ventrals. The coloration is the same as in the type. The
specimen has a total length of 630 mm.; tail 142 mm.; ratio of tail
length to total length 0.23.
The male specimen is also quite close. It has 16 solid maxillary
teeth. Its hemipenis is similar to that of a. attenuata, and has 7 large
basal spines, There is 1 preocular; 2 postoculars; temporals 2-3;
supralabials 8(4-5); infralabials 9(5). The dorsal scales, however, are
unusual in that there is an indication of a single apical pite There are
152 abdominals; 69 caudaIs; 221 total ventrals. This specimen is very
melanistic, so only traces oí the checkered pattern and head-neck mark-
~

ings can be seen. lts totallength is 661 mm.; tail162 mm.; ratio 0.24.
Remarks and Range. As in a. aitenauiia the indication of scale pits
and the aberrant coloration of one of the paratypes presents a problem.
It is probably merely an example of the variability of these characters,
but more material is needed to be sure, The presence of a pit in a
minority of the population may indicate the recent derivation of this
form from pitted ancestors.
There is little doubt in my mind that typical aiieniuüa and boliviana
are two races of the same species despite the present lack of inter
• grades for they have so many features in common. As a matter of
fact, their separation is merely a function of coloration (a variable
character) supplemented by minor differences in teeth and scale
counts. Both races are probably found along the edge of the tropical
lowlands of the Amazon basin attenuata being a northern race and
boliviana a southern one. •

•
28 BULLETIN: MUSEUM OF COMPARATtVE ZOOLOGY

The CHILENSIS Complex

TACHYMENIS CHILENSIS CHILENSIS (Schlegel)
Plate 2, fig. 4; plate 5, figs. 17-19, 21.
1837 Coronella chilensis Schlegel, Phys. Serp., 2, p. 70 (type locality = Chile).
Guichenot, 1854, in Gay, Hist. Chile, Zool. 2, p. 79, pl. 4, fig. 1
(Valdivia and Colchagua, Chile). Dipsas ohilensis, Duméril and
Bibron, 1854, Erpét. Gén., 7, p. 1159 (in part: Chile). Tachymenis
chilensis Girard, 1854, Proc. Acad. N ato Sci. Philadelphia, p. 226
• (Santiago, Chile). Girard, 1855, U. S. Naval Astron. Exped., 2, p.
213, pl, 37, figs. 1-6 (Santiago, Chile). Günther, 1858, Cato Colubr.
Sn., p. 34 (Colchagua and Tichitata in Chile, and Chile). Cope, 1860,
Proc. Acad. Nat. Sci. Philadelphia, p. 247 (Talcahuano! and Quin
quina Id., Chile). Mesotes chilensis Jan, 1863, Arch. Zool. Anat.
Fisiol., 2, p. 308 (Chile). Jan, 1866, Icon. Gén. des Ophid., 18, pI. 6,
fig. 2 (Chile).
1863 'I'achumenis peruviana Peters (not oí Wiegmann), Monatsb. K. Preuss.
Akad. Wissensch. Berlin, p. 275 (in part). Steindachner, 1867, Novara
Rept., p. 62 (Chile). Boulenger, 1896, Cato Sn. Bl"it. Mus. Nat. Hist.,
3, p. 118 (in part: Colchagua and Talcahuana! in Chile, southern
Chile, Chile). Werner, 1896, Verhandl. der K.K. Zool.-Bot. Gesellsch.
Wien, 46, p. 356 (Frutillar, Chile). Amaral, 1930, Mem. Inst. Butan
tan Sao Paulo, 4 (1929), p. 210 (in part). Hellmich, 1938, Rev.
Chilena Hist. Nat., 41, p. 108 (in part: Puerto Montt in Chile, and
sorne other Chilean specimens). Schmidt and Walker, 1943, Zool. Ser.
Field Mus. Nat. Hist., 24, p. 315 (in part). •

1898 Tachymenis peruoioma varo vittata2 Werner, Zool. Jahrb. Sup., 4, p. 259,
pI. 13, fig. 9c (type locality = Frutillar, Chile). Werner, 1904, Ergeb.
Hamburger Magalh. Sammelr.,inNaturhistorishes Mus. zu Harnburg,
1, Rept. und Batr., p. 15 (Coronel, Valdivia, and Puerto Montt, all in
Chile). ·
1898 Tachymenis affinis Werner (not of Boulenger), Zool. Jahrb. Sup., 4, p.
259, pl. 13, fig. 9 d (Chile).
1904 Tachymenis peruviana val'. catenatoi Werner, Ergeb. Hamburger Magalh.
Sammelr., in Naturhistorishes Mus. zu Hamburg, 1, Rept. und Batr.,
p. 14 (Estancilla in Valdi via, Chile).
Although Schlegel (1837) based his very brief description of chilensis
•

on two specimens collected by M. Lesson and M. Garnot from an un

known part of Chile, the scale count of these snakes and their essen-

1 This locality is spelled "Talcaguano' on maps oí the American Geographical Society.

2 These specimens were first reported by Werner in 1896 as peruviana. See aboye.
3 This variety was first described by Werner in 1898 on the basis of specimens frorn Coquimbo.
The type series, judging írom locality and coloration, should be referred to c. assimilis. The
present series, however, appears to be c. chilensis.
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 29

•
•

tially lineate color pattern warrant the restriction of this name as a

trinominal to the most cornmon race of the chilensis complex. This •

race occurs in central and part of southern Chile. In as much as it has

been poorly defined and is very cornmon, it is redescribed in detail on
the basis of the material examined.
•
Material examined. A total of 57 specimens (26 9 ,31 0') referable
to this race have been examined. AIl are from hile. A list of the
Iocalities from which they carne follows. The provinces are arranged
in a north-south sequence.
Chile no further data: 3 9,3 O' (C.M.N.H. 23815; M.C.Z. 1978,
2539,3692; U.S.N.M. 436,56442)
Province of Aconcagua
20 miles northeast of Valparaiso: 1 9 ( .M.N.H. 40117)
Province of antiago: 1 9, 2 O' (C.M.N.H. 9943, 9945-46)
In or near Santiago: 3 9, 1 O' (M.C.Z. 2058; U.S.N.M. 5514,
3 specs.)
Cerro de Aculeo: 1 9. 1 -r (M.C .Z. 19976-77)
Province of Nuble
Las Tundras: 1 9, 1 O' (M.C.Z. 6513, 2 specs.)
Province of e oncepcion
Concepción: 2 c , 5 O' (M. ,.Z. 16372-74, ('.M.N.H. 3863-66)
Coracol: 1 c , 1 ~ (A.M.N.H. 18328, 18333)
Fundo Hualpencillo: 1 ~ (e .M.N.H. 31616)
Talcaguano: 5 c , 7 O' (A.M.N.H. 21159; M.C.Z. 2065 2 specs.,
2066 2 specs., 2540 4 specs., 2766)
Province of Bio Bio •

Angol: 1 O' ( .M.N.H.23812)

Chiquaihue: 2 ~ (C.M.N.H. 23808-09)
El Vergel: 1 O' ( ,.M.N.H.31615)
Nahuelbuta Hills: 3 ~ ,2 O' (C.M.N.H. 23810-11,23813-14,31614)
Province of Cautin
Curacautin: 4 O' (C".l\1.N.H. 6221-24)
Traiguen: 1 ~ (C.M.N.H.31617)
Province of Valdivia
Puerto Corral: 1 o , 1 O' (A.M.N .H. 36149-50) •

Province of Chiloe
Puerto Montt: 1 O' (A.M.N,H. 27659)
Diaqnosis. A Tcclujmenis which resembles peruciama in general body
proportions and consequently in scale counts, but resembles cttenuaia
in penial characters. Its supralabials are normally 7(3-4). This forro
has been placed into a separate species complex fOI~ these reasons as
 I

30 BULLETIN: MUSEUM OF COMP ARATIVE ZOOLOGY

has been pointed out in the introduction. This particular race differs
from the other two in the chilensis complex primarily in having a dorsal
color pattern that is distinctly lineate,yet the dark dorsallinesarequite
narrow (not over 1 scale wide). T. c. chilensis also tends to have more
I

abdominals, caudals, and total ventrals than either assimilis or

melomura.
Descripium. These snakes are of the same order oí size as peruviana,
have similar body proportions, and pupil shape. The hemipenis
(fig. 4), however, resembles that of attenuata. It is bifurcate with a
bifurcate ductus spermaticus; tip distinctly calyculate with remnants
of minute spines on the tips of the folds: body spinose with the spines
increasing in length posteriorly; base with a row of very large spines
followed by a few scattered minute ones. The number of large basal
spines ranges from 4-6 in the specimens studied. The solid maxillary
teeth are subequal or increase slightly in size posteriorly. They are
followed after a gap by a pair of enlarged grooved fangs. The number
of solid maxillary teeth ranges only from 6-7 in the material at hand,
but one specimen figured in the literature (Jan, 1866) had 10 such
teeth. The mandibular teeth decrease in size posteriorly after the
first few. ,
The rostral is broader than deep, slightly visible from aboye; inter
nasals shorter than the prefrontals; frontal lVz-1% times as long as
broad, a bit longer than its distance from the end of the snout, shorter
than the parietals in the adults, hut as long as the parietals in the
young; supraocular of normal size; nasal entire, rarely semi-divided; •

Ioreal'Ionger than deep or as long as deep; preoculars 2, rarely 3, the

upper one reaching the top oí the head but not contacting the frontal;
postoculars 2, rarely 1 or 3; temporals usually 2-3 hut variations of 1-2,
1-3, and 2-2 are very cornmon, and combinations of 1-1, 2-4, and 3-2
OCCUl~; supralabials normally 7(3-4), but 9 head halves have 8(4-5) •

and one has 8(3-4); infralabials generally 9 with the first four or five
contacting the anterior pair of chin shields which are as long as or
longer than the posterior; but combinations of 8(3), 8(4), 10(4), and
•
] 0(5) are also found. Aside from one specimen with a single preocular,
reports in the literature do not increase the aboye variation.
The dorsal scales are smooth, have single well developed apical pits,
and a reduction pattern that is typically 19-19-15. A bit of variation
is encountered in the number of preanal rows. Ten specimens have 16
or 17 preanal rows which are reduced to 15 posteriorly, two specimens
have 17 rows which are not reduced, and one specimen has only 14
preanal rows. The abdominals are rounded, female range 148-158,
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 31

average 152; male range 152-169, average 159. The anal is divided
The caudals are typically paired, hut in two cases several are single,
female range 37-51, average 43; male range 45-53, average 49. The
total ventral range in fernales is 185-204, average 194; males 197-217,
•

average 208. Scale counts reported in the literature lower the female
caudal range to 33 and the male caudal range to 39, but do not other
wise affect the aboye observations.
Even though the coloration of c. chilensis is extremely variable,
certain common features can be detected, and similarities with the
pattern of peruvioma be seen. In the majority of cases (fig. 17 and 21)
the ground color is a light brownish-grey, and most of the dorsal scales
bear a black apical dot about the size of and in the position of the apical
pito The light middorsal line is very pronounced, but there are only
.. traces of the prominent rows of black spots which bordered it in pcru
ouina. The spots have been reduced and rnerely consi t of a series of
black streaks which are never more than one scale wide. These Iuse
into a narrow lineo In sorne specirnens they are completely lost in the
posterior regions. As in peruouuui, there is another black line on the
• fourth row of scales, but in c. chilensis, this is accentuated by a slight
lightening of the ground color immediately aboye it, and a general
darkening of the ground color below it. The head has the typical
Tachinnenis pattern consisting of a dark dash under the eye, a black
line extending from the eye to the angle of the mouth, and a similar •

line extending from the parietals to the angle of the jaw. However,
the latter line is somewhat reduced as compared with peruouuui. The
color on the under side is yellowish with numerous irregular black
marks. Sometimes these are arranged in a linear manner. Markings
are nearly absent on the underside of the chin and tail.
The coloration of very young specimens is essentially the sarne as
. just described. But the rows of black dorsal spots 01' streaks are a bit
more pronounced, and it is more apparent that the dark lateralline is
composed of a series of fused spots. Y oung from as far south as Puerto
Montt are spotted, hut this is not as obvious in their case as in speci
mens from central Chile.
In a number of specimens (fig. 18), the ground color above and
below is quite dark, and the light middorsal line is sharply set off by
very distinct black lines. These Iines are still only one scale wide.
This is the most striking departure, but it is connected to members with
the typical pattern by intermediate forros. In a few specimens, the
opposite has occurred (fig. 19). The ground color has become very light
aboye the dark lateralline, and all traces of the dark dorsallines have
•

32 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

been lost. Consequently the light middorsal line tends to blend with
the ground color and is not too distinct.
The females ha ve an average totallength oí 318 mm., tail 50 mm.
The largest female, from an unknown part of Chile, has a length of
550 mm., tail 90 mm. Males averages 378 mm. overall, tail 65 mm.
The largest male, taken at Concepcion, has a totallength of 597 mm.,
tail 95 mm. In the females, the ratio of taillength to totallength is
0.1¿l-O.17; in males 0.15-0.18. A couple of aberrant males, however,
extend the range of tbe latter to 0.13-O.33!
Ramqe. T. c. chilensis is one oí the more common snakes of central
Chile. Its range extends Irom the southern part of the province of
•

Aconcagua as far south as Puerto Montt in the province oí Chiloe.

This includes the mediterranean region oí central Chile and sorne oí the
forested regions of the south. The race seems to be more abundant in
the former area. From the little data available on the vertical distribu
tion of c. chilensis, it appears that it extends from sealevel to at least
2,000 meters elevation. It reaches the highest points in the northern
part oí its range.
Remarlcs. Geographically, this race occurs between assimilis to the •
north and melomura ill parts of the south. It is also intermediate in
characters. The cline in the number of abdominals, caudals, and total
ventrals is lowest in assimilis, reaches a maximum in chilensis, and falls
off a bit in melamura as will be seen. The counts of assimilis are the
closest to pC7·1lViaJ1CI,. Similar tendencies can be seen in the coloration
change; northern assimilis is closest to peruoiasic, chilensis has a •

pattern which merely carries the differences in assimilis a step further ,

and mclomura is the farthest removed from per1¿via11a. Evidence of this
sort strongly indicates that the en tire chilcusis complex has been de
rived from lJeruvia?La.

TACHYl\1ENIH CHILENSIS ASSIMILIS (Jan)

. Plate 5, fig. 20.
1858 Tachymenis chilensis Girard (not of Wiegrnann), U. S. Explor. Exped.,
20 Hel"p., p. 173 (Valparaiso, Chile).
1863 Psammophulax assimilis Jan, Arch. Zool. Anat. Fisiol., 2, p. 311 (type
Iocality unknown). Jan, 1866, Icon. Gén. des Ophid., 18, pI. 1, fig. 2
(Chile).
1896 Tcchinnenis perumarui Boulenger (not of Wiegmann), Cat. Sn. Brit.
Mus. Nat. Hist., 3, p. 118 (in part: Coquimbo, Chile). Amaral, 1930,
Mem. Inst. Butantan Sao Paulo, 4 (1929), p. 210 (in part). Hell-
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 33

mich, 1938, Rev. Chilena Hist. Nat., 41, p. 108 (in part: sorne of
Chilean specimens). Schmidt and Walker, 1943, Zool. Ser. Field
Mus. Nat. Hist., 24, p. 315 (in part).
1898 Tachymenis peruviana varo coronellina Werner, Zool. Jahrb. Sup., 4,
p. 259, pI. 13, fig. 9b (type locality = Coquimbo, Chile). Werner, 1904,
Ergeb. Hamburger Magalh. Sarnmelr., in Naturhistorisches Mus.
zu Hamburg, 1, Rept. und Batr., p. 14 (Valparaiso, Chile).
1898 Tachymenis peruviana varo catenata Werner, Zool. Jahrb. Sup., 4, p.
259, pl. 13, fig. 9 a (type locality==Coquímbo, Chile).

In the original citation of assimilis (1863) Jan did not know the type
locality, and gave a very brief description of the animal. Fortunately
however, he subsequently (1866) figured this form on the basis of a
specimen from Chile.' It is quite clear from this figure that the name
assimilis should be restricted to the most northern race of chilensis.
This race is redescribed below on the basis of specirnens examined.
Mate'i~'l·alExamincd, Sixteen specimens (7 ~, 9 if) have been
studied. A list of the localities from which they carne follows:
No data: 1 if (A.M.N.R.37942)2
Chile: 1 cf (C.M.N.H. 7004)
Province of Atacama
Chanarcillo: 1 ~ (M.C.Z. 6514)
Domeyko: 1 ~ (C.M.N.R.5771)
Province of Coquimbo
El Tofo: 3 c , 2 e (A.M.N.R. 36084-88)
Punta Teatino: 3 cf (A.M.N.R. 64939-41)
Province of Aconcaqua
Valparaiso: 2 ~ , le (A.M.N.R. 36146-47, U.S.N.M. 5531)
Province of Santiago
Las Condes: 1 cf (M.C.Z. 21211)
Diagnosis. It is a Tochsrmenns clearly helonging to the chilensis
complex, hut differing from the typical race in having a slightly lower
number of abdominals, caudals, and total ventrals; and in ha ving a
color pattern that is partly spotted and partly lineate.
Descl·1·ZJt7·01Z. This race resembles chilensis very closely as regards
general proportions, hemipenial characters, and dentition. The number
of large, basal penial spines ranges from 2-6. There are six or seven
solid maxillary teeth followed after a gap by a pair of large, grooved
fangs.
The proportions and the disposition of the cephalic plates is also the
1 The figured specimen may be the actual type, for it agrees with the type description as
regards certain abnorrnalities in the head sbields.
Z This specimen closely resembles those from Coquimbo.
 \

...
•
•

34 BULLETIN: MUSEUM OF COMP ARATIVE ZOOLOGY

•
same as in chilensis except that the frontal is a bit nearer the length oí
the parietal. There are two preoculars in every case except for one side
oí a specimen from Punta Teatino in which the lower preocular has
fused with the loreal; postoculars 2; temporals generally 2-3, but a
variation oí 2-2 is common; supralabials 7(3-4); infralahials generally
9 with the first 4 or 5 in contact with the anterior pair of chin shields
which are as long as or longer than the posterior pair, hut combinations
of 8(3) were found.
The dorsal scales are smooth, have single well developed apical pits, •

and a reduction pattern of 19-19-15. The ahdominals are rounded,

female range 141-148, average 146; male range 137-154, average 145.
Anal divided. Caudals paired, female range 38-44, average 41; male
range 40-49, average 46. The total ventral range in females is 181-190,
average 187; males 184-210, average 194.
Records in the literature extend the aboye variation a bit but not
significantly. The figure by Jan (1866) shows 9 solid maxillary teeth
and supralabials 8(4-5). The type had both a single preocular and
postocular. Counts of Boulenger (1896) would extend the female
caudal range to 49, and counts of Werner (1898) would extend their
abdominal range to 149.
The color pattern oí c. assimilis (fig. 20) is intermediate between the
spotted pattern oí p. peruviana and the lineate one of c. chilensis. As a
rule, the dorsal ground color is a light brownish-grey, and most of the
scales have a black apical doto The light middorsal line is present,
but it is frequently broken up into a series oí middorsal spots. Anter
iorly this line is bordered on each side by a l'"OW oí fairly large dark
spots. Each row begins as a dark streak on the back oí the neck. This
is the pattern of p. peruouma; however, these spots fuse into a line at
about the middle oí the body. In sorne cases they may even be fused
into a line for the entire Iength oí the body thus giving the lineate
appearance of c. chilensis. But if this is the case, the resulting line is
always more than one scale wide. There is also a distinct black lateral
•
line on the fourth row oí scales as is typical fOI' Tuclnjmrnis. The color
of the head and underside is the same as in c. chilcnsis. The most com
mon departure from the aboye was the blackening of the edges of many
of the dorsal scales. This partially obscured the typical pattern. In a
couple oí other cases, the dorsal ground color was very light and the
pattern reduced.
The females examined have an average totallength of 316 mm., tail
49 mm. The largest specimen, taken in Valparaiso, measures 510 mm.
overall, tail 80 mm. The males have an average total length of 372

•
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 35

mm., tail 66 mm. The largest specimen was from Las ondes. It has
a totallength of 484 mm. and a taillength of 81 mm. The ratio of tail
length to total length ranges from 0.15-0.16 in females; 0.16-0.19 in
males.
Remarks and Ranqe, The range of assimilis extends from the
Province of Atacama in northern hile south into the Province of
Santiago. Thus it is primarily a north-central race, hut it extends far
enough south to overlap slightly the range of chilensis. Both races are
found in the Provinces of Aconcagua and Santiago; however, assimilis
appears to he more cornmon in Aconcagua and chilensis in Santiago.
To my knowledge, these two races have not been taken at the same
locality in this area of overlap. At the northern part of its range,
assimilis seems to be separated from the Peruvian and Bolivia forms
by the severe deserts of the northern parts of Chile and southern
Bolivia, but more collecting in these regions is necessary to be certain.
The climate encountered by assimilis is primarily semiarid, but this
race do es enter the mediterranean regions of Chile. This race has been
collected primarily along the coast, but does extend into the moun
tains. 1t has been taken in the Domeyko range, which reaches an
elevation of 4,000 meters, but we do not know from what part.

TACHYMENISCHILENI. S MELANURAsubsp.
nov.
Plate 5, fig. 22. •

Type. .M.N.H. 3874, an adult male collected by C. C. Sanborn

between February 14 and 28, 1923 at Mafil, Province of Valdivia,
Chile.
Diag11 os1·S. A Taclurmenis belonging to the chilensis complex, but
I

differing from the typical rae e in having such a great development of

melanin that the ground color is very dark and the typicallineate pat
tern obscure. The race also tends to have a lower number of abdomin
als, but this difference is not too significant.
Dcscription. 01 type. The general body shape is the same as in most
of the forms of Taclumicnis of moderate size, cylindrical in section,
and an oval shaped head that is only slightly distinct from the neck.
The eye is oí moderate size with a slightly subeliptical-shaped pupilo
The hemipenis is similar to typical chilensis being bifurcate with a
bifurcate ductus spermaticus; tip distinctly calyculate with smalI
spines on the tips of the folds; body covered with spines that increase
in size posteriorly; and 5 very Iarge basal spines followed by a few
scattered minute ones. There al"e8 solid maxillary teeth which increase
 •

36 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

a bit in size posteriorly. These are followed after a gap by a pair of

enlarged grooved fangs. The mandibular teeth decrease slightly in
size posteriorly after the first few.
The rostral is broader than deep, visible from aboye; internasals
shorter than the prefrontals; frontal 1V2 times as long as broad, longer
than its distance from the end of the snout, distinctly shorter than the
parietals; supraocular of normal size; nasal entire; loreal a bit longer
•
than deep; 2 preoculars, the upper one reaching the top of the head but
not contacting the frontal; 3 postoculars; temporals 2-3 on the left
side, 2-2 on the right; supralabials 7(3-4); 9 infralahials with the first
4 in contact with the anterior pair of chin shields which are longer than
the posterior pair.
The dorsal scales are smooth, have a well developed apical pit, and
a reduction pattern of 19-19-15; abdominals rounded, 163; anal
divided; caudals paired, 52; total ventrals 215.
The dorsal ground color (fig. 22) is a uniíorm light hlack. This
naturally tends to obscure any pattern, but upon close examination
the remains of the typical chilensis one can be seen. There is a faint
indication of the light middorsal line bordered by a darker black, and
the narrow black line on the fourth row of scales. The latter still has a
slight lightening of the ground color just aboye it. The head too is very
dark, but the typical black streaks from the eye to the angle of the
mouth and from the parietals to the angle of the jaw, as well as a few
dark supralabial markings can still be seen. The ventral color is a dark
grey with irregular black markings showing through. These tend to be
arranged in a linear manner. The undersides of the ti p of the tail and
chin are much lighter. There are a few dark rnarkings on the infra
labials.
The type has a totallength of 394 mm., tail 67 mm.; ratio of tail to
total length 0.17.
ATO tes 011 paratJ/pes. Seventeen paratypes (8 c , 9 cf1) are available
([Link].N.H. 3867-73,3875-83, 5769). AIl are from Mafil and bear the
same data as the type. This series agrees very well with the type.
The number of large basal spines on the hemipenis ranges from 5-8;
the number of solid rnaxillary teeth Irom 7-8. Most have 7.
The frontal is 1Vz to 1% times as long as broad, generally shorter
than the parietals hut in the younger specimens the two plates are
subequal in length; nasal entire, rarely semidivided; loreal either longer
than deep 01' deeper than long; generally 2 preoculars, occasionally 3;
generally 2 postoculars, occasionally 3; ternporals usually 2-3 or 2-2,
but variations of 1-1,1-2,1-3, and 3-2 occur; supralabials 7(3-4) in
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 37

most cases, hut one specimen has 6(3-4) and another 8(4-5); infra
lahials 9\4) or 9(5) in most, hut comhinations oí 8\3), 10(4), and 10(5)
are present in this series; anterior chin shields generally longer than
the posterior, hut they are the same length in a few cases.
The dorsal scales have a well developed apical pit and a typical re
duction pattern of 19-19-15. Two specimens have 16 and 17 preanal
rows. In these cases the number is not reduced below 16. Ahdominals
147-155 in females, average 150; males 146-163, average 158. Anal
divided. Caudals typically paired, hut several specimens have 1-3
single; female range 41-51, average 45; males 40-54, average 49. The
total ventral range in females is 192-201, average 196; males 195-215,
average 207.
Ninety-five per cent of the paratypes have a coloration that is
practically identical with that of the type. These are all melanistic,
hut vary a bit in the degree with which the underlying chilensis pat •

tern shows through. The melanism is ahsent in only one specimen

(C.M.N.H. 5769). This one consequently has the typical chilensis
pattern, hut is referred to melamura on the basis of locality. No very
young specimens are present in the series, hut judging from a late
emhryo removed from C.M.N.H. 3878, the ground coloration would
he lighter and the lineate pattern more evidente
The females have an average totallength of 444 mm., tail 72 mm.:
the largest specimen measures 497 mm. and 90 mm. The males have
an average total length of 422 mm., tail 72 mm.; the largest measures
568 mm. and 77 mm. The ratio of tail length of total Iength ranges
from 0.15-0.18 in the females; 0.15-0.18 in most males, hut one excep
tional specimen has a ratio of 0.13!
Remarks omd Range. [Link] appears to he an extremely local
race found only in the vicinity of Mafil in central Valdivia. It has
undoubtedly been derived from c. chilensis, which it resembles very
closely, by a greater production of melanin. But why there should he
such a local melanistic race is hard to say, for the entire region of
southern Chile has a marine climate and typical chilensis is not only
found along the coast of Valdivia but also as far south as Puerto Montt
in Chiloe. •

This race has progressed a hit farther than c. chilensis in the evolu
tion of the color pattern from spotted to lineate, for there is no recapit
ulation of the spotted type. The late embryo removed from C.M.N.H.
3878 is light brown with a lighter middorsal line bordered by dark
lines, and a dark lateralline on each side. These dark lines are com
plete and not formed by small fusing spots as in young c. chilensis .
•
•

• •
 •

38 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

DISCUSS10N
•
Relationship of the Described Forms
Although the large series of Tachymenis discussed aboye readily
separates into three species complexes, there can be no doubt hut that
they form a cornpact phylogenetic group. T. peruoiama seems to be
the most primitive type and therefore the one closest to the ancestral
stock. Such an ancestral stock probably occupied the Andes of central
and southern Peru where peruoiama is now most abundante The source
of this stock cannot even be guessed at since the relationships of the
•
various genera of South American Boiginae is so poorly known. Per
haps it evolved from a lowland form which worked its way up into the
•
mountains, or perhaps from a form which followed the Andean range
south from sorne northern point. In any case, given an ancestral stock
resembling per1¿via1La and occupying the Andes of central and southern
Peru, lines of evolution to the other forms can he traced. These, how
ever, are necessarily hypothetical.
Probably ajfi1Lis was the first offshoot. It evolved in Peru primarily
by a reduction of the number of scale rows from 19-19-15 to 17-17-15. This
was accompanied by a marked reduction of the conspicuous peruvia11a
color pattern, hut not a complete loss (cf. figs. 7 and 12). Judging by
the present rarity of ajfi11,is, it was a small branch, and the • remaining
portion of the stock stayed more or less unchanged as
-peruniuma. The separation of affinis was complete. No affinis-peruciama •

intergrades are known even though both forms occupy the Andean
region of central and southern Peru. They have even been taken in the
same valley (Urubamba) although not at the same locality.
T. peruoiana spread widely to all parts of the Andes of central and
southern Peru, and into those of Bolivia. 1t even worked clown from
the highlands into sorne of the lower valleys. It has heen taken at Vitor
(1,630 meters) on the west side of the Peruvian mountains, and on the
east side in the department of Madre de Dios in Peru and the city of
Santa Cruz (c. 500 meters) in Bolivia. Despite this great spread, it has
remained more or less intacto Local populations differ slightly of
course, but the differences are widely scattered and not arranged in
any clinal sequence. Apparently peruoiana can reach high enough
elevations to cross mountain passes and allow the various populations
to intermingle to a certain extent. The capture of a number of speci
mens on the upland plateaus at nearly 4,000 meters elevation where the
vegetation is distinctly alpine indicates that this is true. Most speci-

•
•
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 39

mens, however, prefer the region of grassy steppes at elevations which

are a bit 10weI·.
But despite this ability to intermingle, per~¿viana did give rise to a
small offshoot iarmensis at the extreme northern part of its
range. T. tarmensis is known only from the vicinity of Tarma in Junin,
lts separation from peruoiana is not as well marked as in affinis. lts
evolution has been a function of an increase in the number of solid
maxillary teeth from a maximum of 10 in peruviana to 12, loss of the
apical pit in the dorsal scales which is already poorly developed in cer
tain pcruv1°an.a populations, a slight increase in the number of caudals
(female perucioma maximum is 50 ; female iarmensis 55), and a complete
loss of the marked peruoioma dorsal color pattern so that the snake is a
uniform light greyish brown aboye (cf. figs. 5 & 7, 10 & 11). lt is
possible that tarmensis is a race of peruoiana, but this can not be de
cided until more material becomes available.
Despite the differences mentioned aboye, affinis, peruvia1~a, and •

tarmensis have certain features in common as mentioned in the intro

duction so they have been placed into one complexo Two entirely
separate side lines have evolved from peruoiama in the center of this
complex atienuaui at the edge of the eastern tropicallowlands, and
chilensis along the coast of Chile. .
1". auenuata appears to have arisen from certain populations of
perueioma which migrated to the edge of the tropical lowlands and -
became somewhat adapted to the changing conditions by elongating
slightly. As mentioned in the introduction, the elongation of the body
caused an increase in the number of caudals and total ventrals, and in
the ratio of tail length to totallength. The separation was made more
complete by a change in the penial characters, and an increase in the
number of solid maxillary teeth. This has been pointed out above.
The color pattern of the ancestral peruoioma remained to a great extent.
T. atienuaia is now found along the edge of the tropical lowlands of
southern Peru and northern Bolivia. There was probably only one
uniform species at first, but during the course of time the Peruvian
and Bolivian populations have diverged a bit in their color patterns.
[Link] in Peru masked the typical dorsal pattern by the develop-
ment of a profuse black flaking on the back, while in a. bolieiana tbe
rows of dorsal black spots fused at their corners to give a checkered
effect (cf. figs. 5-7 & 13-16). Minor differences in the number of teeth
and number of abdominals and caudals also appeared.
One paradox presents itself in connection with this theory of the
evolution of attenuaia. That is, if aitenuaia arose from certain peruvioma

•
•
40 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

populations when they migrated down to the edge of the eastern tropi
cal Iowlands, how is it that other populations have reached quite low
elevations in this region without changing significantly? For example,
both forms have been taken in the Peruvian department of Madre de
Dios. In Bolivia perunnama has been taken at the city of Santa Cruz
(c. 500 meters), and aitetiuata in the department of Cochabamba at an
elevation of 2,500 meters! The problem can not be definitely answered
without a better idea of the environment in which the specimens were
taken, but it can be pointed out that there are mountains in parts of
Madre de Dios, that the tropical lowlands reach a high elevation in
Cochabamba', and that Santa Cl'UZ,despite its low elevation, is adja
cent not only to the Amazonian forests but also to a grazing and agri
cultural region. Thus the aboye concept of the evolution of aiienuaia
along the edge of the tropical lowlands of the Amazon Basin is not
necessarily thrown out despite a conflict in elevations.
T. chilensis, the second marked sideline arising from pcruviana
stock, evolved when that form spread down the western side of the
Andes to the coast of Chile. This sideline also became distinct from
peruviar~a in such features as penial characters, oculars, labials, and
coloration as pointed out in the introduction. As mentioned, the penis
resembles that of aitenuatá, hut this must represent a case of parallel
evolution since the two are completely isolated from one another by
the Andes. The southward migration of Tacluimenis into Chile must
have taken place sorne time ago when conditions in northern Chile
were not so extremely arid. ubsequently a hiatus appears to have
developed in this region. At least no specimens have been obtained
there, but more collecting is necessary to he certain.
As chilensis spread southward along the coast, three races appeared
more or less in north-soutb sequence. These differ from one another
in the number of abdominals and caudals, and in coloration. The num
ber of abdominals and caudals increases from north to central Chile,
and then falls off a bit in certain parts of the south. This cline shows
up best in the number of total ventrals as diagrammed on page 41.
The color difference is even more marked, and striking evolutionary
tendencies can be seen (figs. 20-22). [Link], the most northern
race, typically has a pattern which approaches that of peruvia'l¿a.
Both forms ha ve a narrow ligh t middorsal line or row of small spots
which is bordered by a row of very obvious large black spots. Laterally,
on the fourth scale row, each has a series of smaller dark spots which

1 Maps oí the American Geographic Society show "Las Yungas' reaching an elevation of at
least 3,000 meters in Cochabamba.
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 41

is tending to fuse into a line (cf. figs. 6 & 7 with 20). [Link]
differs in that the rows of dorsal spots fuse into lines posteriorly.
Thus assimilis has the spotted appearance of -peruouma anteriorly,
while posteriorly the pattern is more lineate.

220

T
I r
, t
• , I
210
• T I
I
I I
I t
I
, ,
I
I I
200
I
I
,
I
I
,
I _J_ I
•
t

190

180
c. assimilis c. chilensis c. melanura

In c. chilensis, this tendency is carried a step further. The light

middorsalline is very pronounced, the large bordering black spots are
reduced to very small ones which fuse into a very narrow line for the
entire length of the animal, and the lateral line is accentuated by a
Iightening of the ground color aboye it (fig. 21). Thus c. chilensis has a
conspicuous lineate pattern quite distinct from the spotted one of
perueioma hut easily derived from it. In fact there is a hit oí ontogene-
•
•

42 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

•

tic recapitulation, for the young of c. chilensis tend to be more spotted

than the adults.
T.c. mclanura is the most recent race. 1t has evolved from c. chilensis
by developing a great deal of dark pigmentation. This tends to ob
SCUl'ethe typical [Link] pattern, but traces of it can be seen in the
adults (fig. 22). The young, judging from a late embryo, have lost all
trace of a spotted pattern and merely recapitulate the adult pattern of
c. chilcnsis. At present melanura is a local race known only from the
vicinity of Mafil in Valdivia. Typical chilensis is found to the north
and south of it. The climate of this región is marine, and the recently
evolved mclamura would appear to be better suited for such condi
tions. If so, meltmura would be expected to gradually replace chilensis
•

in the forested south.

By way of surnmarizing the above relationships, a tentative phylo
genetic tree is presented below:
c. melanura

c. cJ1ilensis cl~ilensis complex

c. assimilis

a. aiienuaia a. boliviana
clltenuata
complex

tarmensis
~

pe1·uviana peruviana complex

affinis

Ancestral stock
resembli11g peruviana •

Tochumcrns migrated from Peru and Bolivia down the west side of
the Andes to the coast of Chile, and here gave rise to chilensis. It is
 I

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 43

•
ahsent in the higher mountains of Chile, hut this is not surprising when
one considers the ruggedness and severity of climatic conditions in the
higher parts of the Chilean Andes. But one problem remains. Why
has there not been a similar migration down the east side of the moun
tains into Argentina? The scrub woodlands and grass steppes of the
region should be ideal for Tachumenis, yet it is conspicuous by its
absence.' There is the possibility of course that more collecting would
reveal its presence. A real absence would be hard to explain unless
in terms of lack of time, or occupation of its ecological niche by other
forms. More field work is necessary.

Outlying Species
AII the snakes discussed aboye form a natural phylogenetic unit.
But in addition to these forms centering around peruoiana, three other
species have been added to the genus since Boulenger and others limited
it to include only pe,,<luviafl¿a and its allies. These are elongata Despax
(1910) from Tablazo de Paita, Peru; brasiliensis Gomes (1918) from
Pindamonhangaba in the State of Sao Paulo, Brazil; and surinamensis
Dunn (1922) from Surinam. Their relationship to pcruviana and allies
is a problem because of certain morphological differences and their
outlying geographic position. Perhaps they spread peripherally from
the Andes at sorne distant time, and alarge hiatus has since developed;
perhaps they and peruoiana had sorne comrnon lowland ancestor, or
perhaps they have been misassigned to Tochumenis. Dunn (1922) says
of surinamensis, "1 am in sorne doubt as to whether the generic assign
ment of this snake is correct, . .. lt is evidently closely allied to
Taclurmenis elonqaia:" Morphological characters should eventually
throw sorne light upon this problem, but the status of these snakes must
remain questionable until their anatomy and range is better known.
A reexamination of the types and accumulation of more material is
necessary. Existing descriptions omit certain important features such
as dentition, body proportions, penial characters, etc.

1 Amaral (1930) gives the range of peruviana (of Boulenger) as including Argentina and Para
guay but 1 have not found any specific records in the literature of its occurrence there. Speci
I

mens in the material 1 have examined from Argentina which were said to be Tachymenis have
proved to be Dryophaylax. Other reports in the literature of the occurrence of peruviana in such
places as Ecuador have also resulted from misidentification .

•
•

44 BULLETIN: MUSEUM OF COMP ARATIVE ZOOLOGY

Redefinition of Tachymenis: Synonymy, Morphological

Characters, Range, Habits
,
Genus TACHYMENIS

1835 Tachymenis Wiegmann, N. Acta. Ac. Leop.-Carol., 17, p. 252, pl. 20,
fig. 1 (type peruviana).
1837 Coronella Schlegel (not of Laurenti), Phys. Serp., 2, p. 70, (part).
1843 Ophis Fitzinger (not of Wagler), Syst. Rept., pt. 1, p. 28, (part).
1854 Dipsas Duméril and Bibron (not of Laurenti), Erpét. Gén., 7, p. 1159,
(part) .
•

1863 Mesotes Jan, Arch. Zool. Anat. Fisiol., 2, p. 308, (part).

1863 Psammophylax Jan (not of Fitzinger), Arch. Zool. Anat. Fisiol., 2, p.
311, (part).
1915 Leimadophis Barbour (not oí Fi tzinger), Proc. Biol. Soco Washington,
28, p. 149.
•

Since peruviana is the type of Tachymenis, peruciama and its irnme

diate alIies form the nucleus of the genus. It may be that they con
stitute a, subgenus or perhaps the entire genus. Consequently a knowl
edge of the questionable outIying forms is not absolutely necessary for
an understanding of the characters which must be used to define the
genus. It can be described in terms of the least common denominators
oí affinis, peruviana, iarmensis, oiienuata, and chilensis)
AII are snakes of moderate size. The largest individuals are only
about 26 inches long. The ratio of taillength to totallength is between
0.14-0.18 in the females of most of the forms, and 0.15-0.21 in most
males. Sorne forms, however, have a relatively longer tail so the ratio
is increased to 0.23 in females and to 0.24 in males. The body propor
tions of most of the snakes are moderately stocky. A few forms are a
bit more elongate. An oval shaped head is only .slightly distinct from
the neck. The body of all is more or less cyIindrical in section. The
eye is of rnoderate size with a subeliptical shaped pupilo
Their hemipenis is bifurcate, although not deeply so, with a bifurcate
ductus spermaticus. The tip is spinose becoming calyculate in sorne.
The body is spinose in all, but the length and distribution of these
spines vary from species to species. The base of all forrns bears a few

1 In discussing the range oí variation of such quantitative characters as nurnber oí teeth and
scales, 1 have only u ed my observations and those oí the original authors in the type descrip
tions. The slight extensions of this variation which have been reported in the literature have
already been discussed and tbey are not repeated bere because 1 have often found tbem to be
unreliable.
 WALKER: THE SNAKE, TACH1"MENIS PERUVIANA WIEGMANN 45
•
•

scattered minute spines. Sorne have several very large basal spines
as well.
The solid maxillary teeth are subequal or increase a bit in length
posteriorly. Their number ranges from 5-16. These are followed after
a gap by a pair of large, distinctly grooved fangs. The first few
mandibular teeth are quite small, but the length increases rapidly,
reaches a maximum in the front of the jaw, and then decreases grad
ually.
AII the typical cephalic plates are presento In all, the rostral is
broader than deep, slightly visible from aboye; internasals a bit shorter
than the prefrontals; frontal 1~ to 2 times as long as broad, longer
than its distance from the end of the snout, as long as or shorter than
the parietals; supraocular of normal size; nasal generally entire, occa
sionally semidivided; preoculars 1 or 2, rarely 3, the top preocular
reaching the top of the head but not contacting the frontal; postoculars
typically 2, rarely 1 or 3; temporals usually 2-2 or 2-3 but many vari
ations are cornmon; supralabials either 7(3-4) or 8(4-5), departures
from these are rare; infralabials variable but generally 9(4) or 9(5);
anterior pair of chin shields as long as or longer than the posterior.
The dorsal scales are smooth and generally have a single apical pit,
• although this is poorly developed in sorne and completely absentin other
forms. One individual had indications of two pits. The scales are
arranged in an odd number of parallel rows with a reduction pattern of
19-19-15 or 17-17-15. Only a small amount of variation is encoun
tered in the nurnber oí cervical or preanal rows.
The abdorninals are rounded and not extremely numerous. The
female range is 134-158; males 136-169. The anal is divided except •

in a few aberrant individuals. The caudals are paired in most, •

but sorne have a few entire ones. The female range is 37-65; rnale
40-69. The total ventral range in females is 176-221; males 184-221.
The color pattern in these snakes is quite diverse, but sorne common
features can be seen. In the basic pattern, there is a small dark streak
extending from the snout through the eye to the angle of the mouth.
A similar streak runs from the parietals to the angle of the jaw, The
labials are variably marked with black. Most specimens have such a
•
mark under each eye. There is a light line 01' series of spots on the
middle of the back. This is bordered on each side by a series of dark
spots which often fuse into a stripe or lineo There is a similar lateral .

•
 ..

••
46 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

series of spots or line on the fourth row of scales. Most of the dorsal
scales in the majority of forms have a dark apical dot about the size
of and in the position of the apical pito There is a large number of
irregular dark markings ventrally which may be arranged in a linear
manner. This pattern is quite fundamental and appears in practically
all the forms of Tachumenis. In a couple of forms it is masked by a
secondary black flaking or melanism but traces of the fundamental
pattern can be seen even in these. The pattern is, however, com
pletely lost in tarmensis. The spotted pattern is more primitive than •

the lineate, for it is found in the adults which are believed to be

closest to the ancestral type, and in the young of certain of the lineate
forms. The most advanced types such as chilensis melanura have lost
all trace of it even the young are lineate.
Sexual dimorphism is manifest in the genus in terms of the number of
abdominals and caudals. Oddly enough, the males have a greater
number of abdominals as well as caudals. Consequently there is only
, a slight overlap between the number of mal e and female total ventrals.
This tendency could be seen to a degree in all the forms studied if both
sexes were available. However, it is most apparent in analyses of
single populations. The specimens of perueiama from Huanta illustrate
this point very well. Their counts are summarized below in tabular
form: .

total
abdominal caudal ventral
range average range average range average
•

6 females 146-150 148 41-47 44 187-196 191

6 males 148-156 152 49-52 51 200-206 202 •

The aboye tendencies are also reflected in the ratio of tail length to
total length which is a bit higher in the males, and in the slightly
greater male size. Sexual dimorphism could not be detected in other
characters.
The genus seems to center in the Andes of the southern half of
Peru and northern Bolivia. It reaches elevations there as high as 4,000
•

•
 •

WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 47

meters, but is most abundant in the región oí grassy steppes at eleva

tions which are a bit lower. It has extended from tlle central Andes to
sorne oí the lower valleys on both sides oí the mountains. It has even
reached the edge of the tropical lowlands oí the Amazon basin, but
has not spread very widely there. It has also spread down the west
side of the Andes to the coast of Chile. 1t is found at lower elevations
}n Chile presumably because of the unfavorable nature of the higher
mountains. Members of the genus are absent in the severe deserts of
northern Chile, but do occur in the semiarid regions. They are prob
ably found in the valley bottoms in such places where there would be
sorne vegetation for cover. The snakes are very abundant in the medi
terranean portion of central Chile where they constitute one of the
cornmonest species. They have extended south from here into the
forested and lake region of Chile but have not been taken below Puerto
Montt. The genus apparently has not rnigrated down the east side of
the Andes to the plains of Argentina.

•
.. Notes on the Habits of the Various Forms

Since field notes have seldom been taken, little direct information is
available on the habits of Tachumenis. Nevertheless it is possible to •

make a few deductions on the basis of hábitat and morphology. In as

much as the members of the peruv1·a1¿a and chilensis complexes are
fairly stocky snakes, and the bulk of the specimens ha ve come from
alpine, semiarid, or mediterranean regions, they must be terrestrial,
Those from southern Chile were taken in forested country, but even
these have the body form of a terrestrial snake. The Montana and
Yungas range of members of the oiicnuaia complex, together with the
more elongate body forrn suggests the possibility that the terrestrial
life is supplemented by a semiarborial habit in these forms.
The subeliptical shaped pupil of all oí these snakes may indicate
that they are most active at night in their search for food, etc. Exam
ination of the stomach contents of ten specimens that had obviously
eaten recently revealed that their diet consists primarily oí small frogs,
but occasionally toads and terrestrial lizards are captured. Their
rear fanged poison apparatus would help to subdue the more active

•
48 BULLETIN: MUSEUM OF COMPARATIVE
••
ZOOLOGY

prey. The
stomach contents
of the specimens
examined is
tabulated
below:

Spe S LoStom
es c ach
M ity
Conte
T . •
nts 1
. C Yana
p . ma, Rem
. Z Junin, ains
p . Peru of a
e 4 frog
r 5
u 9 Huant
v 1 Buf
a,
i 6 Ayacu o
a
cho,
n
a C Peru s
3 •
p
9 i
6 n
T 4
. u
7 l
p
. o
p s
e u
r s
u tr
v if
i o
a
li
n
a u
m

T
s
c
h
u
d
i

T p
. . pe
 ruviana C.M.N.H. Juliaca, Liolaemus P u o
•
Puno, Peru m ultif l di n
34086 ormis•
e )
multifor u Lio ?
mis r la
( o e Re
C d m mai
o ns
e us
p of a
m pi
e
a ct frog
)
us
b pi
T. A.M Incachaca, Remains of
i ct Eups
.N.H Cochabamba, a
b us ophu
atte . Bolivia frog s
ii 36 and r (D
o u ta
uat 01 Valparaiso, lizard
n m en
a 7 Aconcagua, iat
boli Eupsophus i éri
Chile us
via C.M taeniatus l (G
na .N.H (Girard) T an ira
Coracol,
. s d rd
•
Concepcion, c
T. c. 40 Eupsophus )
ehilens 11 Chile taeniatus h Bi
is 7 , (Girard) br

..
T.e. A.M.N. 11 arn indebted to
H . Mr, B. Shreve for the
chilens identification of
is 18 certain of the
stomach
•
content
32 speci mens.
8

T.e. C.M.N.H. Curacautin, 3875

chilens Cautin,
is 6221 Chile

C.M.N.H. Traiguen,
T.c. Cautin, Chile
•

chilens 3
is 1
6
I 1
7
T.c.
melamu
ra C.M.N.H. Mafil,
Valdivia, Chile
3
T.c. 8
melanu 6
ra 7
•
C.M.N.H. Mafil,
Valdi,,·ia, Chile
 WALKER: TI1E ' T .. ~KE, rrACHY~IENI PERU,Tli\.NA '\TIEGMANN 49

No information is available on the reproduction of aiienuata or

affinis, but examination of the oviducts of adult females' of pcruviana
and, chilensis show that these forms are ovoviviparous and give birth
to from three to fourteen? young. Fertilization takes place in perueiana
about May and the young are born in September. The material on
chilensis is not as conclusive, hut seems to indicate that the reproduc
tive cycle extends from November to Fehruary or perhaps even latero
The data is tabulated below:
•

Tachumenis peruviana:

Corulitioii 01 the
Dat Locality Embryo

e January • • • • •

February • • <t • •

•
March C.M.N.H. Haunta, Ayachucho, Oviducts empty
•
39650 Peru
M.C.Z. Pachacayo, JUl1i11, Oviducts empty
•
42435 Peru

April • • • • •

May M.C.Z. Yauyos, Lima, Peru 4 eggs present; embryo

45914 not visible grossly
I, -
1 C.M.N.H. Yunguyo, Puno, 10 eggs present ; embryo
40179 Peru not visible grossly
1,
1 C.M.N.H. Yunguyo, Puno, 5 eggs present; embryo
40181 Peru not vi ible grossly
,,
1 C.M.N.H. Yunguyo, Puno, 5 eggs present; embry
40190 Peru not visible grossly
°
, •

1l 19 C.M.N.H. Tarma, Junin, Peru Oviducts empty

5698

1 AIl adult females with a known collecting date were examined by palpation for eggs. The
pregnant ones were opened, and one or more of the embryos studied.
2 Fourteen eggs were counted by palpation in U.S.N.I\1.. 56442. This specimen is from Chile
but since no further data is available, it has not been included in the accompanying tablea.
•

•
•
 •

50 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

• Condition 01 the
Date Specimen Locality Embryo

June • • • • •
•

July • • • • •

------ -----.-------------------
August 29 C.M.N.H. Pisa coma, Puno, Peru 6 eggs present; irnmature
40212 embryo; coiled length
15 mm.

September 14 C.M.N.H. Collacachi, Puno, 6 eggs present; irnmature

34156 Perll embryo; coiled length
8 mm.

'e 14 C.M.N.H. Collacachi, PlII10, . 8 eggs present; irnmature

34160 PeI'u embryo; coiled length
12 mm.

" 16 [Link].H. Urco, ClISCO,Peru Oviducts empty

34105
ce
16 C.M.N.H. UI·CO,Cusco, PelOtl 3 eggs present; immature
34107 I embryo; coiled length
10 mm .
•

" 19 C.M.N.H. Juliaca, Puno, PelOtl Oviducts empty

34086

" 19 C.M.N.H. Urco, Cusco, Peru 3 eggs present ; mature

•

34131 embryo; coiled length

,
32 mm., uncoiled 108
mm.

" 19 C.M.N.H. Urco, CtlSCO,Perll 8 eggs present; immature

34132 embryo; coiled length
17 mm.

October • • • • •

------------"._-------------------
November • • • • • •

December • • • • •

•
•

•
 1
WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN

Tacliumenis chilensis:

Condition of the
Date Specimen Locality Embryo
•

January • • • • •

February 14 C.M.N.H. Mafil, Valdivia, 7 eggs present; embryo

to 28 3868 Chile not visible grossly

C.M.N.H. Mafil, Valdivia, Oviducts empty

3872 Chile

C.M.N.H. Mafil, Valdivia, Oviducts empty

~ 3876 Chile

•
C.M.N.H. Mafil, Valdivia, 4 eggs present; mature
3878 Chile embryo; coiled length
21 mm., uncoiled 98
mm.

C.M.N.H. Mafil, Valdivia, Oviducts empty

3879 Chile
•

C.M.N.H. Mafil, Valdivia, Oviducts empty

3883 Chile
--

March 26 C.M.N.H. Valparaiso, Acon Ovíducts empty

40117 cagua, Chile

March 26 C.M.N.H. Concepcion, Oviducts empty

to April 20 3863 Concepcion, Chile
May C.M.N.H. Chiquahue, Bio Bio, Oviducts empty
•
23808 Chile
June I' • • • • •

July • • • • •

August 9 C.M.N.H. Domeyko, Atacama,

to 16 5771 Chile
September • • • • •

October · ...
,.

November C.M.N.H. Traiguen, Cautin, 8 eggs present ; embryo

31617 Chile not visible grossly
December • • • • •

•
 •
• ...
52 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

It is
noteworthythat
peruviana,
although
occurring a bit
south of the
equator, has
the reproductive
cycle
characteristic
of snakes of
the northern
hemisphere.
This is not too
surprising when
it is realized
that its remote
ancestors
probably carne
from the north,
and that the
seasonal
differences near
the equator
al'te not
pronounced
enough to
require a
change.
However, the
breeding period
of chilensis, or
at least the
southern
members of the
species, has
shifted to
conform to the
summer season
in southern
South America.
It would be of
interest to know
whether this
change has
become a specific difference, or whether it is 3 (2) Midbody
limited to the southern populations of scale rows 19
chilensis. This cannot be decided until more 4
material becomes available from northern 31 Midbody
Chile. scale rows
17. Found
in the
Keu io TACHYMENIS
Andean
PERUVIANA and region of
Allies Central and
As pointed out aboye, Tachirmenis Sou thern
peruouma and its irnmediate alIies form at Peru
least the nucleus of the genus. A key to their affinis
identifica tion follows: 4 (3) Dorsal
1 Body form moderately stocky. Correlated scales with
with this, the caudals single apical pits
are under 60 and the ratio of taillength to although these
alotesome
totallength is under 0.22. Solid maxillary
teeth 6-12 times poorly deve
2 color, female caud
11 Body form elongate. Correlated with this, Found in the
Andean
the caudals are 60 or over and the ratio region of
of tail length to total length is 0.22 or central and
overo Solid maxillary teeth 12-16. Found in southern
southeastern Peru or eastern Bolivia along Peru .
•
the edge of the tropical lowlands peruou ma
. •

The
aitenuat
a
comple
xo
5
2 (1) SupralabiaIs 8(4-5); preocular 1;
body of hemipenis covered with spines of
Andean
subequal region
size, no
of Iarge
Peru basal
and Bolivia
spines. .
Found in the The pel·uviana complex 3
21 Supralabials 7(3-4); preocuIars 2; body of the hernipenis covered

with spines which in crease in size posteriorly,

spines. Found along the coast of Chile
The
chilens
is
comple
x
6
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 1

41 Dorsal scales without apical pits, dorsal color a uniform grayish

brown with no trace or only a very slight trace of darker spots,
. female caudals over 50, 12 solid maxillary teeth. Found in or near
Tarma, Department of Junin, central Peru tarmensis
5 (11) Dorsal scales generally heavily flaked with black, obscuring
any other pattern; 12-14 solid maxillary teeth. Found in south-
eastern Peru attenuata
51 Aattenuata
checkered dorsal color pattern, 14-16 solid maxillary teeth.
Found in eastern Bolivia attenuata boliviana

6 (21) A distinctly spotted 011l lineate dorsal color pattern 7

61 Dorsal ground color so melanistic that an underlying lineate pattern
is nearly completely obscured. Found in or near Mafil, Provinee of
Valdivia, southern Chile chilensis melanura
7 (6) The rows of dorsal spots on eitherside of the light middorsalline
are very prominent being two or more seales wide. The spots fuse
into a stripe posteriorly, but are usually distinct anteriorly. Found
in northern Chile from the province of Atacma to the province of
San tiago. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . chilensis
assimilis
71 The rows of dark dorsal spots are much reduced in size and fuse to
form a line. They are not over one seaIe wide. Found in central
and southern Chile from the provinee of Aconcagua to the provinee
of Chiloe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ehilensis
ehilensis

Relationship between Tacliumenis and other Boiginae .

1 eonclude this diseussion of Tachqrmenis by emphasizing how little
we know about its relation to other genera of Boiginae in South Amer
ica. At one time there was eonfusion between Tachqmenis and Ery
throlamprus. Boulenger and others deeided that many species origi
nally described as Tachumenis were misassigned and that they should
be referred to Eruihrolamprus. A list of these can be found in the intro
duction. Boulenger pointed out in his Catalogue (1896) that members
of Erutholamprus .differ in having fangs which are only feebly enlarged
and sometimes not grooved, a round pupil, no seale pits, a variable
number of seale rows (15, 17, 19, 21, or 23), and ventrals which are
sometimes obtusely angulate laterally. These criteria should separate
the two genera, hut the situation has beeome confused because it has
, been realized that Eruthrolaniprus as redefined by Boulenger is an
omnium gatherum. Among other things, it included all of Coniophanes
(Cope, 1860) which has since been removed from the synonymy,
Dunn (1922) feels that Coniophomes is very close to Tachymenis, and
perhaps inseparable. Tacbumenis is the older name.
 •

54 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

BIBLIOGRAPHY
•

AMARAL,A. DO
1925. "South American Snakes in the Collection of the United States
National Museum." Proc. U. S. Nat. Mus., 67, Art. 24, pp. 1-30.
1930. "Estudos sobre Ophidios N eotropicos. XVIII Lista remissiva
dos Ophidios da Regiao Neotropico." Mem. Inst. Butantan Sao
Paulo, 4(1929), pp. 129-271.
AMERICANGEOGRAPHICALSOCIETYOF NEW YORK.
1922-1938. Peruvian Sheets of the "Map of Hispanic America,"
1 :] ,000,000.
1942. South American Sheets of the "Map of the Americas," 1 :5,000,000.
BARBOUR,T.
1915. "A new Snake from southern Peru." Proc. Biol. Soco Washington,
28, pp. 149-150.
BARBOUR,T., and NOBLE, G. K.
1921. "Amphibians and Reptiles from southern Peru collected by the
Peruvian Expedition of 1914-15 under the Auspices of Yale Uni
versity and the National Geographic Society." Proc. U. S. Nat.
Mus., 68, pp. 609-620.
[Link], .
1916. "Further Explorations in the Land oí the Incas. The Peruvian
Expedition of 1915 of the National Geographic Society and Yale
University.' Nat. Geog. Mag., 29, pp. 413-473, many figs.
,
BOULENGER,G. A.
1893-1896. "Catalogue of the Snakes in the British Museum oí
Natural History,' 1(1893), 2(1894), 3(1896). London.
BOWMAN,l.
1916. "The Andes of southern Peru.' Amer. Geog. Soco New York.
CARLSON,F. A.
1943. "Geography of Latin America." New York.
DESPAX, M. R.
1910. "Mission Géodésique de l'Éqllateur. Collections recueillies par
M. le Dr. Rivet. Liste des Ophidiens et Description des Espéces
nouvelles." Bull. Mus. Hist. N ato Paris, 16, pp. 368-376.
DUNN, E. R.
1922. "Two new Sou th American Snakes.' Pl'OC. Biol. Soco Washington,
35, pp. 219-20.
•
GOMES, J. F .
1918. "Contribuicao para o Conhecimento dos Ophidios do Brazil.
Pt. 11. Descricao de duas Especies novas." Mem. Inst. Butantan
Sao Paulo, 1(1), pp. 78-83, pl, 14, figs. 1-2 .
•
 WALKER: THE SNAKE, TACHYMENIS PERUVIANA WIEGMANN 55

GRIFFIN, L. E.
1916. "A Catalogue of the Ophidia from South America at present con
tained in the Carnegie Museum, with Descriptions oí sorne new
Species." Mem. Carnegie Mus., 7, pp. 163-228, pl. 28.
HELLMICH, W.
1938. "Anotaciones para el Conocimiento de las Culebras de Chile."
Rev. Chilena Hist. Nat. Santiago, 41, pp. 107-110.
.. JAN, G.
1863. "Enumerazione Sistematica degli Ofidi appartenente al Gruppo
Coronellidi.' Arch. Zool. Anat. Fisiol., 2, pp. 213-330.
1866. "Iconographie général des Ophidiens", 18, 6 pls.
MEYEN, F. J. F.
1834. "Reise um die Erde," 1. Berlin.
RUDOLPH, W. E. .
1929. "The new Territorial Divisions of Chile with Special Reference to
Chi16e." Geog. Rev., 19, pp. 61-77, 16 figs.
•

SCHLEGEL, H.
1837. "Essai sur la Physionomie des Serpens,' 2. Arnsterdam.
SCHMIDT, K. P., and W ALKER, W. F., JR.
1943. "Peruvian Snakes from the University of Arequipa." Zool. Ser.
Fjeld Mus. Nat. Hist., 24, pp. 279-296.
STIGLICH, G.
1922-1923. "Diccionario Geographico del Peru." Lima.
WEBERBAUER, A. .
1936. "Phytogeography of the Peruvian Andes," in MacBride, J. F.,
"Flora of Peru." Bot. Ser. FieId Mus. Nat. Hist., 13, pt. 1, pp.
13-37, map.
WERNER, F.
1898. "Die ReptiIien und Batrachier der Sammlung Plate." Zool.
Jahrb. Sup., 4, pp. 244-78, 2 text figs., pIs. 13-14.
1901. "ReptiIien und Batrachier aus Peru und BoIivien." Abhandl,
und Berich. Kon. Zool. Anthrop.-Ethnog. Mus. Dresden, 9(2), pp.
1-14, 8 text figs.
1904. "Ergebnise Hamburger Magalh. Sammelreise." Naturhistorishes
Mus. zu Hamburg, 1, Rept. und Batr.
WIEGMANN, A. F. A.
1835. "Beitrage zur ZooIogie, Gesammelt auf einer Reise um die Erde
von F. J. F. Meyen." Abhandl. 7, Amphibien. N. Acta. Ac. Leop.
Carol., 17, pp. 183-268, pls. 13-22 .

•
 •

..

- •

•
 •

PLATES

, •

•
 W ALKER Tachymenis peruviana Wiegrnann
•
•

•
PLATE 1
Fig. l. Map oí southwestern South America showing the approximate
range of the species of Tachymenis studied aboye. The outline for this map
was obtained through the courtesy of the American Geographical Society. The
key to the map follows. The range of T. tarmensis is in central J unin only.
Its symbol may not be clear.

.... . .- .
Tachymenis peruviana Wiegmann
'\ -,
•
• •

Tachymenis tarmensis WaIker.

Tachymenis affinis Boulenger
Tachymenis attenuata attenuata Walker
Tachymenis attenuata boliviana Walker
Tachymenis chilensis assimilis (Jan)
~ , Tachumenis chilensis chilensis (Schlegel) •
..•. ..•.•..'.... , ,
"'
.
• .• •• •••• ... -i I"'I'aclunnenis
. ,
.-'."
'
- chileneis melanura '\T alker

•
 BULL, MUS, COMP, ZOOL, WALKER, TACHYMENIS PERUVIANA WIEGMANN, PLATE 1,

••' •• • ""í
••
.' '0°
r.
1" \ "

- I
, -\ I

o~ \
I
I
-l
- ,

JUNIN \
\ I
,
I
I
I
o
, I
I -

- ' o·
J o.
I
o

-- •• o
{
o , I ",_,
,
.. ..,'
•
,, '.\ \

\ --,- I

\
o
-\.'
_
¡-.._-.--~'
(
\

- V\ I A
S~N_TA CRU~
\ -- \
\
.
o

t
_ }--'----i-----\ , / I!:""'-"',,-, .:
\
\\
- 1
~

PO
•
• .....•••\ I
\
L--- - -
\
TO
SI \
•
\
o

U -"
l -
-
-
•.
-
1
- "
-
-
t
\'~
, ,

A
N
T
O
\ F
A
G
A
S
T
A
;

1
\
o

<
\ I ,
'\
,
 -
,
\
_
I
. '.

-
O
. ,
-
' I
-
f -'1-
J
_- o
\ I

\--'\ - •

'
\
-
I

\
, - o
•
\ ,
I

-_.---\-
\
•
(
I
o
\ - \ I
•
'"
"
-
1
1
\

\
- -\
1
-
-
\
-
.
.
,
-
-

-
\ \
\ \ ,
I• ~-

\ - -- '4
- 0
°
\

\
-l

-
I

o
- 60°
-
\