Columbia University Press

Publishers Since 1893

New York Chichester, West Sussex
cup.columbia.edu

Copyright © 2018 Columbia University Press

All rights reserved
E-ISBN 978-0-231-54338-5

Library of Congress Cataloging-in-Publication Data

Names: McGhee, George R., author.
Title: Carboniferous giants and mass extinction : the late Paleozoic Ice Age world / George R.
McGhee Jr.
Description: New York : Columbia University Press, [2018] | Includes bibliographical references and
index.
Identi ers: LCCN 2017056021 (print) | LCCN 2017060808 (e-book) | ISBN 9780231543385 | ISBN
9780231180962 (cloth : alk. paper) | ISBN 9780231180979 (pbk.: alk. paper) | ISBN 9780231543385
(e-book)
Subjects: LCSH: Mass extinctions. | Extinction (Biology) | Climatic changes. | Glacial epoch. |
Paleobotany—Carboniferous. | Paleontology—Carboniferous. | Paleoclimatology—Paleozoic.
Classi cation: LCC QE721.2.E97 (e-book) | LCC QE721.2.E97 M388 2018 (print) | DDC 560/.175—dc23
LC record available at https://lccn.loc.gov/2017056021

A Columbia University Press E-book.

CUP would be pleased to hear about your reading experience with this e-book at cup-
[email protected].

Cover design: Lisa Hamm

Cover image: Richard Bizley / Science Source
Carboniferous insects. Artwork of a millipede (Arthropleura) and a dragon y (Meganeura) in the
forests of the Carboniferous Period (354–290 million years ago).
Illustrations: (pages i, ii, and xiii) Mary P. Williams
For Marae

Tha gràdh agam ortsa.

 Contents

Preface

1 Harbingers of the Late Paleozoic Ice Age

2 The Big Chill

3 The Late Carboniferous Ice World

4 Giants in the Earth …

5 The End of the Late Paleozoic Ice Age

6 The End of the Paleozoic World

7 The Legacy of the Late Paleozoic Ice Age

Notes

References

Index
 Preface

I
magine living in a world in which a dragon y-like insect with a
wingspan similar to that of a seagull zooms past your head;
multilegged millipedes as long as a small car and as wide as a truck tire
are crawling along on the ground; venomous scorpions, stingers held
curved over their backs, are as long as mid-sized dogs; and salamander-
like animals are neither cute nor tiny—they are as long as alligators and
have mouths full of sharp-pointed, deadly teeth.
Imagine living in a world with huge expanses of land covered by
tropical rainforests, a green band that stretches around the middle of the
planet across all of the continents. Rainforests in which you discover,
when you go hiking through them, very strange-looking trees: from a
distance they look as though they have no leaves, even though the upper
part of the trunk and the drooping branches of the tree are green. On
closer examination, through a pair of eld glasses, you can see that the
tops of these trees do have leaves, but they are peculiar small bladelike
leaves that partially overlap each other and wrap spirally around the
branches and upper trunks. The tops of the trees thus have a somewhat
fuzzy look, as though they were wearing green sweatshirts of leaves with
multiple sleeves-of-leaves extending out along all of their branches. These
peculiar sleeves-of-leaves look somewhat like those we see in our club
mosses, but our little club mosses stand only 12 to 13 centimeters ( ve
inches) high and these strange trees are gigantic, towering 50 meters (164
feet) into the sky.
Imagine living in a world in which a giant continent sits over the South
Pole, a continent over ve times larger than Antarctica—over twice the
size of Africa, our largest single free-standing landmass surrounded by
water—that is covered with vast rivers of ice sheets stretching away from
a huge, white, glacial polar cap.
That world is not imaginary; it is the Earth—not our modern Earth,
obviously, but rather the Earth as it existed in the Carboniferous Period of
the Paleozoic world, over 300 million years ago. All of the things described
here actually existed on that Earth—the giant animals, the giant trees, the
giant continent, the giant glacial ice cap. There were giants in the Earth in
those days.
This book is about that ancient Earth: how it came to be, and how it
ended. We will examine the various hypotheses as to how animals, insects
in particular, could have become so gigantic. We will examine the
evolution of the great tropical rainforests that at rst glance appear
similar to those of our modern Earth, but that turn out to have been very
di erent. The legacy of those great rainforests can be found around the
world today in the massive deposits of coal that are found in
Carboniferous strata—the huge volume of organic carbon that give the
geologic period Carboniferous its name. Those coal deposits have been one
of our major sources of energy since the beginning of the industrial era,
and the burning of those coal deposits is releasing 300-million-year-old
carbon dioxide back into our atmosphere and heating up the planet. The
ancient Carboniferous Earth is transforming the modern Earth.
The Paleozoic world was totally destroyed in the end-Permian mass
extinction, the greatest biodiversity crisis to occur since the evolution of
animal life on Earth. The great ice caps of the Late Paleozoic Ice Age
melted, the Earth began to warm, and the most catastrophic volcanic
eruptions in Earth history began to inject trillions of tons of greenhouse
gases into the Earth’s atmosphere. The resultant “Hot Earth” lasted for
ve million years, an apocalyptic world in which the tropics were lethally
hot—a world in which vertebrate land animals could only survive in cooler
latitudes higher than 30° in the Northern Hemisphere and higher than 40°
in the Southern Hemisphere. In stark contrast to our modern world—
where the equatorial tropics harbor the highest diversity of life on the
planet—the equatorial region of that Hot Earth was almost totally barren.
Even in the oceans, the equatorial temperatures were so high that sh and
other vertebrate animals could only exist in the cooler waters of the high
latitudes of the Earth. How could this have happened to the Earth? In this
book we will examine in detail the triggers of the end-Permian mass
extinction, and the nightmarish world that resulted.
This book is a summary of over four decades of my research into the
ecological consequences of the Late Paleozoic extinctions, and I thank
Columbia University Press, publishers of my previous books The Late
Devonian Mass Extinction (1996) and When the Invasion of Land Failed: The
Legacy of the Devonian Extinctions (2013), for making this third volume
possible. I thank three anonymous reviewers of the manuscript for
comments and suggestions that have helped me to improve the nal book.
And last, I thank my wife, Marae, for her patient love.
1 Harbingers of the Late Paleozoic Ice
Age

The late Paleozoic ice age lasted for ~67 m.y. [million years] in eastern
Australia, and as such, it was the longest-lived icehouse interval in the
Phanerozoic.
—Fielding, Frank, Birgenheier, et al. (2008, 55)

ICE AGES IN EARTH HISTORY

We know that at least seven ice ages have occurred in the past 4,560
million years, time periods during which the Earth partially—or almost
entirely—froze over. The history of the Earth is divided into four eons: the
Hadean, Archaean, Proterozoic, and Phanerozoic, from oldest to youngest
(for the geologic timescale, see table 1.1). These eons are themselves
divided into smaller time units, the eras, and they are divided into still
smaller time units, the periods (table 1.1), much as our calendar years are
divided into months, weeks, and days. We know from fossil evidence that
ancient bacterial life was present on the Earth 3,450 million years ago,
during the Paleoarchaean Era, and we have geochemical evidence that
indicates that life was present even earlier during the Eoarchaean Era,
some 3,830 million years ago (McGhee 2013; Nutman et al. 2016). The rst
four of the known ice ages occurred during the Proterozoic Eon, much
later in time than the appearance of life on Earth, and three more ice ages
occurred even later during the Phanerozoic Eon; thus life on Earth has
successfully survived them all—but at a cost. Many of the ice ages are
associated with periods of extinction and large losses of biological
diversity, a topic that will be explored in more detail in the next section of
this chapter.

TABLE 1.1 The geologic timescale and ice ages.

Eon Era Period Time of Onset (Ma) Ice Ages (ICE)

Phanerozoic Cenozoic Quaternary 2.59 ICE
Neogene 23.03 ICE
Paleogene 66.0 ICE
Mesozoic Cretaceous 145.0
Jurassic 201.3
Triassic 252.2
Paleozoic Permian 298.9 ICE
Carboniferous 358.9 ICE
Devonian 419.2 ICE
Silurian 443.8
Ordovician 485.4 ICE
Cambrian 541
Proterozoic Neoproterozoic Ediacaran 635 ICE
Cryogenian 850 ICE
Tonian 1,000
Mesoproterozoic Stenian 1,200
Ectasian 1,400
Calymmian 1,600
Paleoproterozoic Statherian 1,800
Orosirian 2,050
Rhyacian 2,300 ICE
Siderian 2,500
Archaean Neoarchaean 2,800
 Mesoarchaean 3,200
Paleoarchaean 3,600
Eoarchaean 4,000
Hadean 4,560

Source: Timescale modi ed from Gradstein et al. (2012).

Note: Ma = millions of years before the present for the start of each time unit listed.

A major di erence between the four Proterozoic and three Phanerozoic

glacial episodes is re ected in their formal names, given in table 1.2. The
four Proterozoic episodes are called snowball Earths, whereas the three
Phanerozoic episodes simply are called ice ages. The snowball-Earth
glaciations were much more geographically extensive than any seen in the
Phanerozoic in that continental-covering, sea-level-reaching ice sheets
extended from the poles of the planet all the way down to the equator.
From space, the entire planet may have looked like one giant snowball,
hence the name “snowball Earth.” The rst known snowball Earth, the
Huronian, occurred some 2,300 million years ago during the Rhyacian
Period (tables 1.1 and 1.2). Surprisingly, it is now thought that life may
have triggered not only this massive freezing of the Earth during the
Rhyacian but also the rst mass extinction in Earth history. How could
this be?

TABLE 1.2 Ice ages in Earth history.

Ice Age Position in Geologic Time (table 1.1)

A. Phanerozoic ice ages:
1. Cenozoic Ice Age Late Paleogene to Recent
2. Late Paleozoic Ice Age Late Devonian to Late Permian
3. End-Ordovician Ice Age Late Ordovician
B. Proterozoic ice ages:
1. Gaskiers Snowball Earth Ediacaran
2. Marinoan Snowball Earth Cryogenian
3. Sturtian Snowball Earth Cryogenian
4. Huronian Snowball Earth Rhyacian

About 200 million years earlier than the Huronian freezing, at the
beginning of the Siderian Period of the Paleoproterozoic Era (table 1.1), an
event of major importance in the evolution of life on Earth occurred—the
Great Oxygenation Event, or GOE for short. The very atmosphere of the
Earth has been radically transformed by the presence of life. The original
atmosphere of the Earth was probably very similar to that of its sister
rocky volcanic planets Mars and Venus—that is, composed mostly of
carbon dioxide. The atmosphere of Mars today is 95 percent carbon
dioxide and the atmosphere of Venus is 97 percent, whereas the
atmosphere of the pre-industrial-age Earth was only 0.03 percent carbon
dioxide.1 Both anaerobic and aerobic photosynthesizing bacteria2 actively
remove carbon dioxide from the atmosphere and use the carbon to form
complex hydrocarbons for food. Thus, on Earth, life has been removing
carbon dioxide from the atmosphere for the last 3,830 million years,
contributing to the transformation of the Earth’s atmosphere to its
present carbon-dioxide-depleted state.
The aerobic photosynthesizing cyanobacteria not only remove carbon
dioxide from the atmosphere but also add oxygen to the atmosphere.3
About 2,500 million years ago, the oxygen-producing activity of the
ancient cyanobacteria was nally to have its rst major impact on the
atmosphere of the Earth: it triggered the GOE. For the future evolution of
complex—and large—life-forms with aerobic metabolism, the GOE was
good news indeed as these organisms need free oxygen. For the ancient
anaerobic life-forms—the original inhabitants of Earth—the GOE was a
disaster because oxygen is a poison to them. The rst mass extinction in
the history of life on Earth probably occurred 2,500 million years ago,
when vast unknown numbers of species of anaerobic bacteria and archaea
perished by oxygen poisoning (McGhee 2013).
Di usion of oxygen into the atmosphere was most probably the trigger
of the Huronian Snowball Earth in the Rhyacian Period of the
Paleoproterozoic Era (table 1.1), which occurred about 200 million years
after the GOE (Lane 2002; Kopp et al. 2005). The steady drawdown of
carbon dioxide, caused by the photosynthetic activity of life, was already
reducing the greenhouse e ect of this gas in the atmosphere. The
presence of free oxygen in the atmosphere may then have begun to
remove an even more powerful greenhouse gas—methane—via oxidation.4
The resulting sharp drop in the greenhouse capacity of the atmosphere of
the Earth may have triggered this rst great snowball in Earth history.
Some 300 million years later, in the Orosirian Period of the
Paleoprotoerozoic Era (table 1.1), oxygen concentrations in the
atmosphere had reached high enough levels to begin oxidizing iron on
land, and the rst redbed strata appear in the terrestrial rock record.
Redbeds are just that: layered beds of red sandstones and shales in which
the iron has been oxidized, giving the strata their characteristic rusty-red
color. By this time, the aerobic photosynthetic activity of cyanobacteria
had raised the amount of oxygen in the atmosphere to something between
1 and 4 percent.5
Much later, the Earth froze in three snowball phases in quick
succession—quick on a geologic timescale (tables 1.1 and 1.2). Two of these
phases, the Sturtian and Marinoan Snowball Earths, occurred at 717 and
640 million years ago, respectively, during the Cryogenian Period of the
Neoproterozoic—an aptly named period as “cryogen” comes from the
Greek for “beginning of freezing cold.”6 The Sturtian Snowball Earth lasted
some 57 million years; the Marinoan, about ve million years; and the last
snowball Earth, the Gaskiers, began 580 million years ago but lasted only
340,000 years during the Ediacaran Age (table 1.1).7
The earliest soft-tissued marine animals evolved about 780 million
years ago (Erwin et al. 2011) during the Cryogenian, but it was only 541
million years ago that the rst marine animals with mineralized tissues—
skeletons—evolved. This evolutionary event marks the beginning of the
Phanerozoic Eon (table 1.1), the geologic eon of the “visible animals.”8 The
event is also often called the Cambrian Explosion, in recognition of the
evolutionary pulse in the diversi cation of animal life that occurred at the
beginning of the Cambrian Period of the Paleozoic Era ( g. 1.1). Large
macroscopic animals with skeletons made of calcium, silica, and
phosphorous compounds suddenly are present in rocks that previously
contained only the impressions made by a few types of soft-bodied
animals. All in all, almost 100 families of marine animals originated in the
pulse of diversi cation that occurred at the beginning of the Paleozoic
Era.

FIGURE 1.1 An evolutionary pulse in the diversi cation of large-bodied animals with skeletons
occurred in the early Cambrian (graph); illustrated are some of the characteristic skeletonized
marine animals of the Cambrian fauna (see text for discussion). Geologic timescale abbreviations:
V, Ediacaran; barred-C, Cambrian; Ө, Ordovician; S, Silurian; D, Devonian; C, Carboniferous; P,
Permian; TR, Triassic; J, Jurassic; K, Cretaceous; T, Tertiary (Paleogene and Neogene).

Source: From Paleobiology, volume 10, pp. 246–267, by J. J. Sepkoski Jr., “A Kinematic Model of
Phanerozoic Taxonomic Diversity: III. Post-Paleozoic Families and Mass Extinctions,” copyright ©
1984 The Paleontological Society. Reprinted with permission of Cambridge University Press.

Figure 1.1 illustrates some of the early skeletonized members of the

major evolutionary clades of animals that were to become characteristic
of the Paleozoic marine world: the Arthropoda (trilobites), Lophophorata
(inarticulate brachiopods, hyoliths9), Mollusca (monoplacophorans), and
Echinodermata (eocrinoids). We will examine the ecology and further
evolution of all of these animal groups in greater detail in the chapters
that follow. For example, the lophophorates are of particular interest in
that they were later to evolve gigantic brachiopod shell sh in the
Carboniferous.
Why did so many types of large animals with di erent skeletal
chemistries appear so suddenly in the fossil record? A clue may be seen in
some modern-day animals, such as bivalve molluscs that have the capacity
to switch from aerobic metabolism to anaerobic metabolism when the
oxygen content in water becomes depleted. While in the anaerobic-
respiration mode, these bivalves also produce acid metabolites as a
byproduct, and these acids actually begin to etch away and dissolve the
calcium-carbonate shell of the animal (Lutz and Rhoads 1977; Babarro and
De Zwaan 2008). Thus it has been proposed that the ability to grow and
maintain mineralized skeletal tissues is a trait that is found in the aerobic-
respiring organisms, meaning there has to be enough oxygen present in
the environment to allow organisms to respire aerobically. Atmospheric
modeling of the evolution of the Earth’s atmosphere indicates that around
540 million years ago, just before the dawn of the Cambrian, oxygen levels
in the atmosphere had nally risen to around 16 to 18 percent (Berner
2006), as compared to present-day levels of 21 percent. (We will consider
these models in detail in chapters 3 and 4.) Atmospheric oxygen levels of
16 to 18 percent may have been the nal trigger for animals belonging to
numerous disparate phylogenetic lineages to simultaneously achieve
sustained aerobic respiration, increase in size, and secrete mineralized
skeletal tissues of several di erent chemical compositions in the di erent
lineages. After the GOE and the Huronian Snowball Earth, the fact that the
planet was hit by at least three more snowball-Earth glaciations in the
latest Neoproterozoic (tables 1.1 and 1.2), immediately before the
Cambrian animal diversi cation, is evidence for the continued e ects of
the drawdown of the greenhouse-gas carbon dioxide from the atmosphere
and the injection of oxygen into the atmosphere by aerobic
photosynthetic life.
In contrast to the snowball Earths, only the poles and higher-latitude
regions of the Earth were covered by ice in the three Phanerozoic ice ages
(tables 1.1 and 1.2). Our modern world is the product of the Cenozoic Ice
Age, and polar ice and high-latitude continental glaciation still exist on
the Earth—but perhaps not for long. The Cenozoic Ice Age began about 34
million years ago in the late Paleogene,10 and its last glacial interval ended
about 10,000 years ago (Lewis et al. 2008). It remains to be seen whether
the Earth is still in the Cenozoic Ice Age or our warmer modern world is
merely a interglacial interlude and whether the great ice sheets will
return in the near future (on a geologic timescale). An alternative view is
that the Earth is now in a major and unusual warming phase, triggered by
the injection of carbon dioxide into the atmosphere by the human
burning of fossil hydrocarbons for fuel. Humans began to mine and burn
coal strata—the overwhelming majority of which are Carboniferous in age
—in a big way during the Industrial Revolution and later added oil and
natural-gas extraction and combustion through subsurface drilling. The
subsequent increase in the carbon dioxide content of the atmosphere
parallels an increase in the atmospheric temperature of the Earth,
resulting in the accelerated melting of the residual ice that still exists
from the last glacial phase of the Cenozoic Ice Age. Continued warming
may result in an ice-free hot Earth in the future, and the Cenozoic Ice Age
will be history.
The rst Phanerozoic glacial phase, the end-Ordovician Ice Age, was
peculiar in that it occurred during a period in geologic time when the
Earth was in a long-term greenhouse phase and was quite warm, with an
atmosphere still rich in carbon dioxide. The ice age was also quite short,
consisting of an intense glaciation phase that lasted only 1.9 million years
within a longer-term period of glacial advances and retreats in the latter
part of the Ordovician and early part of the Silurian (McGhee et al. 2012,
2013). The evolution of the rst land plants from water-dwelling plants
had occurred by the middle of the Ordovician, but by the end of the
Ordovician the continents of the Earth were still populated only by very
small liverworts and related simple plants. The Ordovician is sometimes
called the “Lilliputian plant world” because these tiny land plants—
averaging under three centimeters (a little over an inch) in height—
constituted the only plant cover of the planet Earth at that time. A few
multilegged marine animals occasionally ventured out of the oceans onto
the tidal ats and dry land near the end of the Ordovician—we have their
fossil footprints preserved in stone—but they quickly returned to water
and did not live on the land. (For an extensive discussion of the invasion of
land by plants and animals, see McGhee 2013.) And last, in the south polar
region of the planet, a vast ice sheet covered at least 30 million square
kilometers (almost 12 million square miles) of land (Sheehan 2001);
otherwise the land areas of the Earth were empty.
 This book is about the second of the Phanerozoic glacial episodes
(tables 1.1 and 1.2), the Late Paleozoic Ice Age—the longest ice age on
Earth for the past 541 million years since the pulse of diversi cation of
complex animal life in the Cambrian Explosion. If the world of the
Ordovician Ice Age seems strange, the world of the Late Paleozoic Ice Age
was far stranger, as we will see in this book.

ICE AGES AND EXTINCTIONS

All of the Phanerozoic ice ages (and probably all of the snowball Earths)
are associated with the extinction of large numbers of species and major
losses of biodiversity—but not all biodiversity crises in Earth history are
associated with ice ages. Of the eight largest Phanerozoic biodiversity
crises, ve are argued to have been related to ice ages—one with the short
end-Ordovician Ice Age and the other four with the longest glaciation in
the Phanerozoic, the Late Paleozoic Ice Age (table 1.3). Curiously, these are
also the rst ve biodiversity crises in Phanerozoic history (table 1.4)
since the Great Ordovician Biodiversi cation Event (GOBE) (Webby et al.
2004; Servais et al. 2010), the second massive diversi cation of animal life
in the oceans following the Cambrian Explosion (we will consider the
evolutionary signi cance of the GOBE in detail in chapter 6). Just as
curious is the fact that the end-Permian, end-Triassic (table 1.4), and end-
Cretaceous biodiversity crises were clearly not associated with ice ages.
We will consider the causes of the end-Permian mass extinction—the
largest in Earth history—in detail in chapter 6, along with the end-Triassic
extinction, as these two catastrophes appear to have had a similar causal
mechanism. The end-Cretaceous mass extinction—the event that
destroyed the dinosaur ecosystem—is now generally attributed to the
e ects of the impact of the massive asteroid that produced the 180-
kilometer-diameter (112-mile-diameter) Chicxulub Crater in Mexico and
thus had an extraterrestrial cause, not an Earthly one.

TABLE 1.3 Ecological-severity ranking from most severe (#1) to least severe (#7) of the eight
largest Phanerozoic biodiversity crises since the beginning of the Ordovician.

#1. End-Permian (Changhsingian Age)

#2. End-Cretaceous (Maastrichtian Age)
#3. End-Triassic (Rhaetian Age)
#4. Late Devonian (Frasnian Age)
#5. End-Middle Permian (Capitanian Age)
#6. Early Carboniferous (Serpukhovian Age)
#7. End-Devonian (Famennian Age), End-Ordovician (Hirnantian Age)

Source: From McGhee et al. (2004, 2013).

Note: The four biodiversity crises that are thought to be related to the Late Paleozoic Ice Age are
listed in bold type; see text for discussion.

TABLE 1.4 Epoch and age divisions of the geologic timescale in the critical time interval leading up
to and immediately following the Late Paleozoic Ice Age.

Period Epoch Age Time of Onset (Ma) Time of Crises

Jurassic (pars) Hettangian 201.3
Triassic Late Rhaetian 209.5 ← Biodiversity Crisis
Norian 228.4
Carnian 237
Middle Ladinian 241.5
Anisian 247.1
Early Olenekian 250.0
Induan 252.2
Permian Late Changhsingian 254.2 ← Biodiversity Crisis
(Lopingian) Wuchiapingian 259.8
 Middle Capitanian 265.1 ← Biodiversity Crisis
(Guadalupian) Wordian 268.8
Roadian 272.3
Early Kungarian 279.3
(Cisuralian) Artinskian 290.1
Sakmarian 295.5
Asselian 298.9
Carboniferous Late Gzhelian 303.7
(Pennsylvanian) Kasimovian 307.0
Moscovian 315.2
Bashkirian 323.2
Early Serpukhovian 330.9 ← Biodiversity Crisis
(Mississippian) Visean 346.7
Tournaisian 358.9
Devonian Late Famennian 372.2 ← Biodiversity Crisis
Frasnian 382.7 ← Biodiversity Crisis
Middle Givetian 387.7
Eifelian 393.3
Early Emsian 407.6
Pragian 410.8
Lochkovian 419.2
Silurian Late Pridolian 423.0
Ludfordian 425.6
Gorstian 427.4
Middle Homerian 430.5
Sheinwoodian 433.4
Early Telychian 438.5
Aeronian 440.8
Rhuddanian 443.8
Ordovician Late Hirnantian 445.2 ← Biodiversity Crisis
Katian 453.0
Sandbian 458.4
 Middle Darriwilian 467.3
Dapingian 470.0
Early Floian 477.7
Tremadocian 485.4

Source: Timescale modi ed from Gradstein et al. (2012).

Note: The temporal position of the seven major biodiversity crises (table 1.3) that occurred in this
time interval are indicated with arrows. Ma = millions of years before the present for the start of
each time unit listed.

The short-lived end-Ordovician Ice Age triggered a mass extinction of

marine species but had little e ect on life on land simply because there
was not much life on land to be a ected (McGhee 2013). My colleagues
Peter Sheehan, Dave Bottjer, Mary Droser, and Matthew Clapham and I
have conducted comparative paleoecological analyses that have revealed
that the environmental degradation produced by the end-Ordovician
glaciation precipitated a major loss of marine biodiversity, yet the
extinction failed to eliminate any key taxa or evolutionary traits and was
of minimal ecological impact (Droser et al. 2000; McGhee et al. 2004, 2012,
2013). In terms of ecological severity, the end-Ordovician extinction had
an impact equivalent to that of the end-Devonian extinction (table 1.3),
and both extinctions were triggered by an intense glacial phase that was
of a geologically short duration—1.9 million years in the end-Ordovician
Hirnantian Age and 1.4 million years in the end-Devonian Famennian Age
(McGhee et al. 2013; McGhee 2013).
In contrast to the end-Ordovician and end-Devonian biodiversity
crises, the glaciation in the Early Carboniferous Serpukhovian Age
triggered a precipitous drop in the speciation rate of marine species but
only moderate diversity losses. However, the ecological impact of those
diversity losses and ecosystem restructuring was an ecological level of
magnitude larger than that seen in the end-Ordovician or end-Devonian
extinctions (table 1.3) (McGhee et al. 2012). We will examine the sequence
of Early Carboniferous evolutionary events and glaciation phases in more
detail in chapter 2.
In contrast to all of the biodiversity crises considered thus far, the
extinctions that occurred in the Capitanian Age at the end of the Middle
Permian Epoch (tables 1.3 and 1.4) were associated with the waning of the
Late Paleozoic Ice Age, not with its onset or expansion. The end of the Late
Paleozoic Ice Age was triggered by the start of a major phase of global
warming, and at the end of the Paleozoic Era occurred the end-Permian
mass extinction, the most ecologically severe biodiversity crisis in the
entire Phanerozoic—a truly global “Category 1” ecological crisis in both
the Earth’s marine and terrestrial biota (McGhee et al. 2004; Cascales-
Miñana et al. 2015). We will consider the dire consequences of these two
Permian biodiversity crises in greater detail in chapters 5 and 6.
This leaves the enigmatic Late Devonian biodiversity crisis—so called
because it occurred within the Late Devonian Epoch at the end of the
Frasnian Age (tables 1.3 and 1.4). The Late Devonian crisis was a very
severe event, in terms of both the magnitude of biodiversity loss and the
ecological impact. It has the highest ecological severity of any of the
biodiversity crises associated with a Phanerozoic ice age—if it was
associated with a Phanerozoic ice age! We will consider the relationship of
the Late Devonian biodiversity crisis to the Late Paleozoic Ice Age in great
detail in the section “The Onset of the Late Paleozoic Ice Age?” later in
this chapter.
Finally, the reader will have noticed that the Cenozoic Ice Age (table
1.2) is not listed in the ecological-severity ranking of biodiversity crises in
table 1.3. Does this mean that no extinction or loss of biodiversity
occurred during the Cenozoic Ice Age? No, signi cant extinctions are
associated with the Cenozoic Ice Age, and we will also consider them in
more detail in the “The Onset of the Late Paleozoic Ice Age?” section. In
general, however, the biodiversity losses that occurred during the
Cenozoic Ice Age were of much less magnitude than those associated with
the Paleozoic crises. Why? Steve Stanley, a University of Hawaii
paleontologist, has proposed an answer to that question: the biota of the
Paleozoic world was more prone to extinction than the biota of the
Cenozoic, a phenomenon revealed in the classic analyses of the University
of Chicago paleontologists Dave Raup and Jack Sepkoski that demonstrate
that the mean extinction rate of marine animals has declined through
time; that is, Paleozoic marine species had a much higher rate of
extinction than Cenozoic marine species (Raup and Sepkoski 1982; Stanley
2007). This phenomenon is not unexpected: the theory of natural
selection would predict the evolution of increasing extinction resistance
with time. In the Paleozoic, the biota was dominated by ancient species
with higher extinction rates than modern species. For example, Stanley
has shown that in the Late Devonian extinction, the older species lineages
su ered extinction at a 20 percent higher rate than the more recently
evolved species lineages, and as the majority of the Late Devonian species
belonged to these older lineages, the total extinction rate was predictably
high (Stanley 2007). In contrast, Cenozoic marine species are much more
resistant to extinction, hence the total extinction rates seen in the
Cenozoic Ice Age are predictably lower than those seen in the Late
Paleozoic Ice Age. We will examine further evolutionary and ecological
di erences between the ancient Paleozoic species and the more modern
Cenozoic species in more detail in chapter 6.

THE MYSTERY OF THE ICE AGES

What triggered these ice ages and snowball Earths in geologic time? Why
did huge areas of the Earth become frozen in the past? Obviously, if the
entire planet gets colder, it must somehow either not be receiving enough
heat from the sun to maintain its previous temperature or be losing much
more of its heat to space than previously or both.
The Earth does generate some of its own heat by the decay of
radioactive minerals in its rocks, but the planet really depends upon
electromagnetic radiation—light—from the nearest star, our sun, for heat.
The sun is actually becoming hotter with time as it slowly burns more and
more of its original hydrogen gas in the fusion production of helium. A
cooler sun during the Proterozoic Eon may have contributed in part to the
intensity of the snowball-Earth glaciations. In the past 541 million years,
however, our models of the sun’s energy production predict no major
uctuations that could have triggered the Phanerozoic ice ages.
If the production of energy by the sun has been relatively constant for
the past 541 million years, then one way to change the amount of heat
that the Earth receives is to change the distance of the Earth from the sun:
a planet closer to the sun receives more energy per unit area than a planet
farther away. Although it is not apparent to us on human timescales, the
Earth’s distance from the sun actually does vary on geologic timescales.
The shape of the Earth’s orbit around the sun changes with time, from
being almost a perfect circle to being stretched out into an ellipse—a
shape variation that is known as orbital eccentricity (table 1.5). When the
Earth’s orbit is nearly circular, the planet absorbs about the same amount
of heat from the sun throughout the year, as the distance of the Earth
from the sun—the radius of the circular orbit—does not change during the
year. At the other extreme the Earth’s orbit is highly elliptical, with a long
axis and a short axis. Twice a year the Earth is located at the short axis of
the ellipse and warms up as it is closer to the sun, and twice a year the
Earth is located at the long axis of the ellipse and cools down as it is
farther from the sun. The stretching of the Earth’s orbit from a near circle
to an ellipse and its rounding back to a circle again is a function of the
gravitational in uence of the other planets in the solar system, and it
occurs periodically in cycles of 100,000 and 400,000 years of geologic time.

TABLE 1.5 Periodic orbital phenomena that a ect the Earth’s climate.

Phenomenon Periodicity of phenomenon

1. Orbital Eccentricity: variation in the shape of the Earth’s 100,000 years and 400,000 years
orbit, from near-circular at one extreme to elliptical at the
other.
2. Rotational Axis Obliquity: variation in the tilt of the 41,000 years
Earth’s spin axis, from a tilt of 22.1° at one extreme to 24.5°
at the other.
3. Rotational Axis Precession: circular variation in the 19,000 years and 23,000 years
orientation of the Earth’s spin axis, from pointing to
Polaris as the north pole star at one extreme to pointing in
the opposite direction, along an arc of 180°, at the other.

The Earth also experiences variation in solar radiation in two ways that
are caused by its orientation relative to the sun, not its distance. The rst
of these, rotational axis obliquity, refers to the fact that the degree of tilt
in the Earth’s rotational axis is not constant. The spin axis of the Earth is
not vertical; that is, the planet tilts over on its side as it rotates. This tilt is
responsible for the four seasons that we experience. For example, when
the Northern Hemisphere of the Earth is tilted toward the sun, it receives
more energy per unit area and is thus hotter—the summer season—than
when it is tilted away from the sun and is thus colder—the winter season.
The tilt in the rotational axis of the Earth varies between 22.1° and 24.5°.
When the spin axis of the Earth is tilted over at 24.5°, the seasonal
di erences in temperature are much more pronounced—the winters are
colder and the summers are hotter—than when the planet is tilted at only
22.1°. The degree of tilt of the Earth’s rotational axis oscillates in a 41,000-
year cycle in geologic time. Finally, the direction in which the rotational
axis of the Earth points varies with time—a phenomenon known as
rotational axis precession. Simply expressed, the Earth wobbles with time
like a toy top winding down, with its tilted-over rotational axis spinning
around and around in a circle. The tilt of the Earth relative to the sun
changes on 23,000- and 19,000-year cycles; thus, the Northern Hemisphere
is now tilted toward the sun (summer) in June, but 23,000 years from now
it will be tilted away from the sun (winter) in June.
To make things even more interesting, these three cycles—orbital
eccentricity, axial obliquity, and axial precession—interact to magnify or
diminish one another’s e ect. For example, when the Earth is located
farthest from the sun on an elliptical orbit (aphelion) and its axis is tilted
over at 24.5° and is positioned at one of the solstices, the planet will
become quite cold in one hemisphere and even the equator will cool. The
opposite e ect occurs when the Earth’s orbit is more circular, its axis is
tilted at only 22.1°, and the equinoxes occur at aphelion. These orbital
oscillations are called Milankovitch cycles, after the Serbian geophysicist
Milutin Milankovitć (usually anglicized to Milankovitch in the English
literature) who rst proposed them. It has been demonstrated that the
Earth’s climate did indeed vary in Milankovitch cycles of 400, 100, 41, 23,
and 19 thousand years during the Cenozoic Ice Age (Zachos et al. 2001),
and the 400- and 100-thousand-year eccentricity cycles have been
detected in Late Carboniferous strata during the Late Paleozoic Ice Age
(Heckel 2008; Horton et al. 2012).
However, the Milankovitch cycles in climate change became apparent
only after the Cenozoic Ice Age had started and glaciers had formed on the
Earth. Thus it appears that some stronger cooling mechanism is necessary
to chill the planet down to form glaciers, which then wax and wane on
weaker Milankovitch thermal frequencies. The next suspect in the ice ages
mystery is changes in the Earth’s atmosphere—speci cally, changes in the
amount of the greenhouse gases carbon dioxide (CO2) and methane (CH4)
in the atmosphere. These gases act much like the glass roof of a
greenhouse (hence the name greenhouse gas): the glass roof allows
higher-frequency sunlight to penetrate down into the greenhouse, where
the light is absorbed by the plants and then radiated back as lower-
frequency infrared light, but then the glass roof blocks the escape of the
infrared light and thus causes the temperature within the greenhouse to
become hotter and hotter. Likewise, given the same amount of energy
received from the sun, the Earth will retain more of that energy as heat
when the atmosphere of the planet is enriched with carbon dioxide and
methane and will lose more of that energy back to space when the
atmosphere contains little carbon dioxide and methane.
Thus, to cool the planet and initiate an ice age, the amount of
greenhouse gases in the Earth’s atmosphere must be reduced (Montañez
and Poulsen 2013). Carbon dioxide is removed from the atmosphere by
two principal mechanisms: biological photosynthesis11 and chemical
weathering of silicate rocks.12 Of these two processes, the chemical
weathering of silicate rocks appears to be the more important: it is
estimated that of the carbon dioxide that was removed from the Earth’s
atmosphere during the Phanerozoic, about 80 percent was removed by
silicate weathering and only about 20 percent by biological
photosynthesis (Raymo and Ruddiman 1992). Carbon-dioxide removal by
weathering is enhanced by the uplift and exposure of large surface areas
of silicate rock to the Earth’s atmosphere during mountain-building
events (Raymo and Ruddiman 1992) and by the exposure of silicate rocks
in the equatorial humid zone of the Earth where the combination of high
precipitation and high temperature intensi es the process of chemical
weathering (Kent and Muttoni 2008; Irving 2008). Methane is removed
from the atmosphere chie y by oxidation, but the oxidation of methane
then produces carbon dioxide,13 so we are back to the mechanisms of
atmospheric carbon-dioxide removal in order to trigger global cooling.
The role of tectonic-forcing mechanisms of widespread mountain-
building events (crustal buckling and uplift caused by collisions of
continental plates, accretion of island-arc terranes onto continental
plates, and subduction of oceanic plates under the margins of continental
plates) and continental positionings (large number of continental plates
located in the equatorial humid zone) in removing carbon dioxide from
the Earth’s atmosphere via chemical weathering has been implicated in all
of the known glaciations, from the four snowball-Earth glaciations in the
Proterozoic (Melezhik 2006) to the end-Ordovician (Lenton et al. 2012),
Cenozoic (Raymo and Ruddiman 1992; Zachos et al. 2001; DeConto and
Pollard 2003; Kent and Muttoni 2008), and Late Paleozoic ice ages in the
Phanerozoic. However, it has been argued that atmospheric carbon-
dioxide depletion by biological evolutionary processes was also a
signi cant contributing factor in triggering the Huronian Snowball Earth
(the evolution of aerobic photosynthesis) (Kopp et al. 2005), the end-
Ordovician Ice Age (the evolution of land plants) (Lenton et al. 2012), and
the Late Paleozoic Ice Age (the evolution of forests) (Algeo et al. 1995,
2001).
The topic of this book is the Late Paleozoic Ice Age, so let us take a
closer look at the proposed triggers of this climatic event. The Late
Devonian was clearly a time of major tectonic activity. In the short time
span (on geologic timescales) of only four million years during the
Frasnian Age, major mountain-belt deformation and uplift spread across
the Appalachian-Caledonian mountain chain in North America and Europe
(then joined together in a single continent called Laurussia; see g. 1.2)
and further to the south in the Variscide mountain chain extending into
northern Africa (on the giant southern supercontinent Gondwana; see g.
1.2). These crustal deformations and mountainous uplifts were driven by
the incipient collision of the southeastern margin of Laurussia (southeast
North America) and the northwestern margin of Gondwana (northwest
Africa). On the northeastern margin of Laurussia, mountainous uplifts
occurred in the Ural mountain belt (western Russia), driven by the
collision of the Kazakhstan crustal block with Laurussia, and to the east of
that, the Central Asian mountain belt buckled and uplifted in the collision
of the Siberian crustal block (eastern Russia) with the eastern margin of
the Kazakhstan block. In terms of tectonic plate collisions, the Frasnian
Age was a time of a multicar pileup on the geologic turnpike.

FIGURE 1.2 Paleogeography of the Earth in the late Frasnian Age. Lighter shaded areas
surrounding the continents are shallow continental marine waters, and deep oceanic waters are
black. The large continent straddling the equator in the left-center is Laurussia (modern-day North
America and Europe), the smaller continent to the northwest of Laurussia is northern Asia, the
islands east of Laurussia are pieces of eastern Asia, and the giant continent to the south of
Laurussia is Gondwana (modern-day South America, Africa, India, Antarctica, and Australia all
joined together).

Source: Global Paleogeography and Tectonics in Deep Time © 2016 Colorado Plateau Geosystems
Inc. Reprinted with permission.

But that was not all. On the northwestern margin of Laurussia, crustal
buckling and uplift occurred in the Antler mountain belt (western margin
of the United States and Canada) eastward into the Ellesmerian-
Svallbardian mountain belt (northern Canada). This deformation and
uplift was triggered not by continental block collisions but rather by
oceanic subduction and accretion of island-arc terranes along the
northwestern margin of Laurussia. Nor was that all. To the south, on the
giant continent Gondwana, similar oceanic-plate subduction and island-
arc terrane accretion occurred in the Bolivianide mountain belt on the
western margin of Gondwana (western South America) and in the Lachlan
mountain belt on the eastern margin of Gondwana (eastern Australia and
Antarctica). In summary, the Late Devonian Epoch was a period of intense
tectonic activity on a global scale. (For a detailed discussion of the
tectonic events that occurred during the Late Devonian, see McGhee
2013.)
Finally, there exists the possibility that yet another major tectonic
mechanism might have been active in the Late Devonian—mantle-plume
volcanism. In mantle-plume volcanism, a giant plume of magma rises
from deep in the Earth’s mantle and, when it intersects continental crust
at the Earth’s surface, erupts in huge volumes of basaltic lava—the
weathering of which could extract huge volumes of carbon dioxide from
the atmosphere. Possible evidence for mantle-plume volcanism comes
from the Viluy igneous province in eastern Siberia (Courtillot and Renne
2003; Courtillot et al. 2010; Bond and Wignall 2014). The original size of
the Viluy igneous province is unknown because it is highly eroded.
Estimates suggest that the extent of these lava ows may have been as
much as six million square kilometers (2.3 million square miles).
Radiometric dating of the Viluy lavas is not precise, but it does place the
eruptions in the interval of 377 to 350 million years ago—that is, sometime
from the middle Frasnian Age in the Late Devonian to the late Tournaisian
Age in the Early Carboniferous (table 1.4). Smaller igneous provinces,
some with kimberlite14 magmatism that demonstrates a mantle source for
the volcanism, are also known in other areas of the ancient Laurussian
continent from the same time interval; thus, there may have been several
areas of mantle-plume volcanism active in the Late Devonian (Racki and
Wignall 2005).
The uplift, deformation, and exposure to the atmosphere of huge
surface areas of silicate rocks that occurred in the Late Devonian tectonic
events should have triggered a massive drawdown of carbon dioxide from
the atmosphere caused by the chemical weathering of these newly
exposed rocks. In addition, the Laurussian continental block was
positioned directly on the equator within the equatorial humid zone of
the Earth in the Frasnian ( g. 1.2), thus enhancing the chemical
weathering of silicate rocks located in this region. Thus, it is not
unexpected that these potential carbon-dioxide-depleting tectonic
mechanisms for triggering global cooling have been proposed as causes
not only of the end-Famennian glaciation but also of a hypothesized
glaciation that may have occurred at the end of the Frasnian as well
(Averbuch et al. 2005). In addition, the University of Lille sedimentologist
Olivier Averbuch and his colleagues have o ered an independent
geochemical line of evidence in support of the chemical-weathering
hypothesis for global cooling in the Late Devonian: ratios of the isotopes
strontium-87 and strontium-86 in seawater. Strontium-87 is radiogenic
and is primarily found in continental rocks; strontium-86 is nonradiogenic
and is primarily found in sea oor rocks. Thus higher ratios of strontium-
87 to strontium-86 indicate higher rates of chemical weathering and
erosion of crustal rocks on the continents, with the delivery of more
strontium-87 to the oceans; lower ratios indicate the reverse (or increased
sea oor hydrothermal activity). Strontium-87 to strontium-86 ratios
increased during the Frasnian Age and spiked in the late Frasnian,
indicating an intensi cation of continental weathering at this critical
time. A second spike in strontium-87 to strontium-86 ratios occurred some
ve million years later in the early Famennian, indicating a second pulse
of intense continental weathering at this time (Averbuch et al. 2005). Did
these periods of intense weathering of continental silicate rocks deplete
the Earth’s atmosphere of enough carbon dioxide to trigger the formation
of glaciers in the late Frasnian and early Famennian?
Another continental positioning (other than equatorial) also
contributed to the formation of glaciers during the Late Devonian: the
giant continent Gondwana was de nitely positioned over the South Pole
of the Earth in the late Famennian, 360 million years ago ( g. 1.3). Over
the next 110 to 115 million years the geographic position of the South Pole
on Gondwana shifted eastward across the landmass, from South America
to Africa to Antarctica to Australia ( g. 1.3), as the giant tectonic plate
holding Gondwana slowly moved westward over the South Pole. In the
Cenozoic Ice Age the continent Antarctica was positioned over the South
Pole (where it still is today); it has an ice-covered surface area of 13.72
million square kilometers (5.30 million square miles). In the Late
Paleozoic, the giant continent Gondwana had a surface area of 74.23
million square kilometers (28.65 million square miles), almost ve-and-
one-half times bigger than Antarctica. Gondwana was so large that it was
never entirely covered by ice during the Late Paleozoic Ice Age; rather, the
centers of multiple glacial masses migrated across the giant continent in
concert with its movement over the South Pole ( g. 1.3).

FIGURE 1.3 Polar wandering path across Gondwana from 360 million years ago to 250 million years
ago. Also shown are the geologic outcrop areas of glacial strata, where Glacial I strata are Late
Devonian to Early Carboniferous in age, Glacial II strata are late-Early to Late Carboniferous in age,
and Glacial III strata are Early to Middle Permian in age.

Source: From Geological Society of America Special Paper 441, pp. 331–342, by T. D. Frank et al.,
“Paleozoic Climate Dynamics Revealed by Comparison of Ice-Proximal Stratigraphic and Ice-Distal
Isotopic Records,” copyright © 2008 Geological Society of America. Reprinted with permission.

Independent biological weathering events also may have contributed

to atmospheric carbon-dioxide depletion during the Devonian. A major
event in the evolution of land plants occurred in the Givetian Age of the
Middle Devonian—the rst forests on Earth—and the evolution of forests
certainly had to have added to the tectonic e ects triggering global
cooling. The Earth’s oldest known forest is the famous Gilboa fossil forest
of New York State; it consisted of trees that were large cladoxylopsids,
extinct relatives of our modern-day ferns. (Tree ferns still exist today,
such as the Australian tree fern Cyathea cooperi.) In the Givetian, the
cladoxylopsid fernlike tree was the species Wattieza (Eospermatopteris)
erianus, which stood more than eight meters (26 feet) tall and had trunks
that were a half-meter to a meter (1.6 to 3.3 feet) in diameter. Yet these
same large trees were anchored only by a broad, bulbous base that had
numerous small, short roots.
The next step in the evolution of trees was the appearance of forests of
giant Archaeopteris trees in the Frasnian Age of the Late Devonian. Unlike
the fernlike cladoxylopsid trees, Archaeopteris trees were lignophytes—
true woody trees with strong trunks; they towered 30 meters (100 feet)
into the air and had deep, branched root systems that penetrated over a
meter (3.3 feet) into the ground. These trees produced deep soil horizons
by both chemical and mechanical weathering of the rocks into which they
were rooted. However, the Archaeopteris trees still reproduced by spores,
not seeds like our modern lignophyte trees, and they were generally
restricted to the wetlands in the lowland areas of the Earth because they
needed a reliable source of water for their reproduction by spore.
The nal step in the evolution of trees was the appearance of the rst
seed-reproducing plants, the spermatophytes. The rst spermatophyte
lignophyte trees evolved in the Famennian Age and, not needing water in
reproduction, colonized the dry highlands and mountains that were out of
reach for non-spermatophyte Archaeopteris trees. Thus, huge areas of
exposed silicate rock were now within reach of the seed plants for
potential biological weathering.
The University of Cincinnati sedimentologist Thomas Algeo and his
colleagues have argued that the evolution of the rst woody forests on
Earth was a major contributor to atmospheric carbon-dioxide depletion
and a trigger for global cooling in the Late Devonian (Algeo et al. 1995).
They proposed that the evolution of forests triggered three di erent
mechanisms that acted in concert to remove signi cant amounts of
carbon dioxide from the atmosphere. First, they argued that the
photosynthetic productivity of the Archaeopteris trees that covered vast
coastal and lowland areas of the Earth by the late Frasnian removed
carbon from the atmosphere to form organic hydrocarbons. The increased
burial rate of organic carbon on land began the process of depleting
carbon dioxide in the atmosphere.
Second, they argued that the mechanical and chemical weathering of
silicate rocks by Archaeopteris root systems produced deep soil horizons
and removed carbon dioxide from the air by forming carbonates. Plant
roots not only fracture rocks mechanically, exposing more rock surface
area to contact with carbon dioxide and water and, thus, potential
weathering, they also introduce organic acids that directly weather the
rock chemically.
Third, they argued that enhanced soil formation by vascular plants
“resulted in elevated uxes of soil solutes (especially biolimiting
nutrients) as a consequence of (1) enhanced mineral leaching, (2) xation
of nitrogen by symbiotic root microbes, and (3) shedding of plant-derived
detrital carbon compounds … [and] elevated river-borne nutrient uxes
may have promoted eutrophication of semi-restricted epicontinental seas
and stimulated algal blooms” (Algeo et al. 2001, 233). Eutrophic blooms of
marine phytoplankton and algae in the seas would lead, in turn, to the
depletion of oxygen in the shallow seas and the accumulation of
unoxidized organic hydrocarbons in black shales, rather than returning
the carbon to the atmosphere as carbon dioxide in the normal process of
aerobic respiration by bacteria (Carmichael et al. 2016). The resultant
carbon-dioxide drawdown from the atmosphere is proposed to have led to
global cooling in a reverse greenhouse e ect.
 Algeo and colleagues thus proposed that major global cooling took
place in the late Frasnian, that that cooling was driven by the extraction
of large volumes of carbon dioxide from the atmosphere by the vast
forests of the newly evolved Archaeopteris trees, and that that cooling was
a trigger for the Late Devonian biodiversity crisis (table 1.3). This proposal
received widespread attention in the popular press, with one science
newsmagazine referring to the Earth’s rst widespread forests as “mass
murderers of the Devonian” (Flangan 1995). However, the Algeo and
colleagues model attributing the end-Frasnian extinctions to the spread of
Archaeopteris forests on land drew immediate criticism from other
Devonian workers. Tony Hallam, a University of Birmingham
paleontologist, and his colleague, Paul Wignall, a University of Leeds
paleontologist, commented: “Perhaps the only question arising from the
Algeo model lies in the degree to which chemical weathering [by plants]
increased in the Late Devonian. The Archaeopteris forests were restricted
to oodplain environments, whereas more upland areas may not have
been colonized until later in the Famennian, with the appearance of seed
plants. Increased chemical weathering [by plants] may not therefore have
become signi cant until the very end of the Devonian” (Hallam and
Wignall 1997, 91).
Thus a much stronger case may be made for global cooling triggered by
the biological weathering model of Algeo and colleagues for the late
Famennian extinctions at the end of the Late Devonian than the late
Frasnian ones within the Late Devonian (McGhee 2013). Evidence for the
existence of glaciers on the Earth in the late Famennian Age is
unequivocal (and will be discussed in detail in the next section of this
chapter), but did glaciers form in the late Frasnian Age as well? If so, were
the late Frasnian glaciers the result of carbon dioxide drawdown from the
atmosphere by the chemical weathering of vast areas of silicate rocks
exposed by numerous tectonic events in the late Frasnian? And were the
late Famennian glaciers the result of further carbon dioxide depletion in
the atmosphere exacerbated not only by the evolution of coastal and
lowland spore-reproducing Archaeopteris forests but also by the spread of
newly evolved highland and arid region seed-reproducing plants?

THE LATE PALEOZOIC ICE AGE

The University of Nebraska geologist Christopher Fielding and his
colleagues have demonstrated that the main pulses of the Late Paleozoic
Ice Age lasted for over 67 million years in eastern Australia (Fielding,
Frank, Birgenheier, et al. 2008). Sixty-seven million years is a long time.
The Phanerozoic Eon of geologic time is divided into three eras: the
Paleozoic, Mesozoic, and Cenozoic, from oldest to youngest (table 1.1), and
the main pulses of the Late Paleozoic Ice Age lasted longer than the entire
Cenozoic Era! That is, the entire Age of Mammals—the period of time in
which we, the mammals, have dominated the terrestrial ecosystems of the
Earth following the termination of the Age of Dinosaurs at the end of the
Cretaceous—has existed only for 66 million years, not quite as long as the
Late Paleozoic Ice Age. If we include the Late Devonian glacial episodes
that were the harbingers of the main glaciation interval (as will be argued
in the next section of the chapter), then the entire duration of the Late
Paleozoic Ice Age is even longer—some 115 million years in total.15 This
115-million-year duration estimate for the total span of the Late Paleozoic
Ice Age is more in accord with the fact that the giant continent Gondwana
was situated over the South Pole for more than 110 million years ( g. 1.3)
(Frank et al. 2008), just as the glaciated continent Antarctica has been
situated over the South Pole during the Cenozoic Ice Age. With a duration
of 115 million years, the Late Paleozoic Ice Age persisted for more than 20
percent of the entire Phanerozoic Eon.
Not only was the ice age of enormously long duration, but many
strange things happened during the Late Paleozoic Ice Age as well. About
90 percent of all of the coal-bearing strata on Earth were deposited during
the Late Paleozoic Ice Age; that is, almost all of the fossil-fuel coal reserves
of the entire world were deposited in an interval of time that constitutes
less than 2 percent of the total age of the planet (Lane 2002).16 Giant plants
grew in and around the coal swamps of the Late Paleozoic Ice Age. Our
modern little lycophytes (club mosses) and equisetophytes (horsetails)
grow to be only about 12 to 25 centimeters ( ve to ten inches) tall,
respectively; ancient relatives of the lycophytes towered 50 meters (164
feet) into the sky in the Late Paleozoic Ice Age, and the equisetophytes
produced trees 20 meters (65 feet) high.
In the Late Paleozoic Ice Age, giant animals roamed the grounds
beneath the unusual giant plants and ew in the skies above the forests—
dragon y-like insects with wingspans like a seagull’s, salamander-like
four-legged vertebrates and multilegged millipedes as long as alligators,
slithering silver sh as big as grasshoppers, spiders as large as lobsters. In
general, many of the ancient animals that lived during the Late Paleozoic
Ice Age were ve to seven times larger than their living relatives—and
sometimes 12 times larger!
How did such a world, such an Earth so unlike our own, come to be?
The initial environmental signals—harbingers of what was to come— rst
began to show up in the Late Devonian world.

THE ONSET OF THE LATE PALEOZOIC ICE AGE?

When did the Late Paleozoic Ice Age begin? Curiously, that is not an easy
question to answer. Let us begin by considering the events that occurred
at the onset of an ice age much closer in time to us—the Cenozoic. Three
million years ago, in the late Pliocene (table 1.6), glaciation of the Earth
had become bipolar: in the Southern Hemisphere, Antarctica was totally
covered by the Antarctic ice sheet; in the Northern Hemisphere, northern
North America was under the Laurentide ice sheet and northern Europe
was under the Fennoscandian ice sheet. The onset of glaciation in the
Northern Hemisphere coincided with extinction pulses rst in the marine
realm in the Pliocene (table 1.6) and then later on land in the Pleistocene
(Hayward 2002). While signi cant, the Plio-Pleistocene extinctions were
smaller in magnitude than the Paleozoic extinctions, in terms of the
amount of biodiversity loss and the ecological impact, and are not in the
list of the eight most severe biodiversity crises in the Phanerozoic (table
1.3). (For further discussion, see McGhee 2013, 203–212.)

TABLE 1.6 The timing of geologic and biotic events at the onset of the Cenozoic Ice Age.

Geologic Time Period Epoch Ice Sheets Biotic Events

Ma SH NH
Quaternary Holocene 00 ICE ICE
Pleistocene 01 ICE ICE ← Land extinctions
02 ICE ICE ← Marine extinctions
Neogene Pliocene 03 ICE ICE ← Marine extinctions
04 ICE
05 ICE
Late 06 ICE
Miocene 07 ICE
08 ICE
09 ICE
10 ICE
 11 ICE
12 ICE
Middle 13 Cold
Miocene 14 Cold ← Marine + land extinctions
15
16
Early 17
Miocene 18
19
20
21
22
23 ← Mi-1 Glacial Pulse
Paleogene Late 24
Oligocene 25
26
27 ICE
28 ICE
Early 29 ICE
Oligocene 30 ICE
31 ICE
32 ICE ← Land extinctions
33 ICE ← Marine extinctions
34 ICE ← Oi-1 Glacial Pulse
Late 35
Eocene (pars) 36

Source: Cenozoic timescale modi ed from Walker and Geissman (2009); geologic and biotic events
modi ed from McGhee (2013).
Note: Ma = millions of years before the present; SH = Southern Hemisphere; NH = Northern
Hemisphere.
 So the Cenozoic Ice Age began in the late Pliocene when glaciation
became bipolar, right? The answer is no, as the giant Antarctic ice cap is
much older than the late Pliocene. The question of the timing of the onset
of the Cenozoic Ice Age thus becomes: when did the Antarctic ice cap
form? Twelve million years ago, in the late Miocene (table 1.6), glaciation
of the Earth was unipolar; massive ice sheets had formed in Antarctica
that persist to the present day. So the Cenozoic Ice Age began in the late
Miocene when the Southern Hemisphere ice cap formed, right? The
answer is no, as the onset of continental glaciation in Antarctica is much
older than the late Miocene. The Earth began to cool rapidly 14 million
years ago, in the middle Miocene, and a pulse of extinction occurred in
both the marine and terrestrial realms at this same time (Sepkoski 1996;
Lewis et al. 2008). However, the onset of Miocene glaciation in Antarctica
was even older than that; it is dated from the Mi-1 glacial pulse in the
earliest Miocene, some 23 million years ago (table 1.6), which lasted for
about 200,000 years (Zachos et al. 2001).
So the Cenozoic Ice Age began in the earliest Miocene, right? The
answer is no, as the onset of glaciation in Antarctica is even older than
that. The rst pulse of continental glacier formation in Antarctica, and the
onset of the Cenozoic Ice Age, is dated to the Oi-1 glacial pulse in the
earliest Oligocene, some 34 million years ago (table 1.6). The Oi-1 glacial
pulse initiated the formation of ice sheets that lasted some eight million
years in eastern Antarctica, and it is estimated that the ice-sheet coverage
of the Oligocene glaciers was about 7.0 to 11.9 million square kilometers
(2.7 to 4.6 million square miles).17 The onset of the early Oligocene
glaciations coincided with two separate pulses of extinctions that
occurred in the oceans about 33 million years ago and with a third pulse of
extinction that occurred on land about 32 million years ago (table 1.6)
(McGhee 2001).
 Now the story becomes even more interesting: where is the
sedimentary proof that Antarctica was glaciated in the Oligocene? The
answer is that there is precious little sedimentary evidence of Oligocene
glaciation on the continent of Antarctica itself; possibly only two sites
preserve evidence of Oligocene glacial sediments (Strand et al. 2003; Ivany
et al. 2006). The problem is that the sedimentary evidence of the smaller
Oligocene glaciers was removed by the erosive action of the much larger
glaciers that formed in the Miocene. The Oligocene glaciers are estimated
to have covered between 7.0 and 11.9 million square kilometers (2.7 to 4.6
million square miles) of Antarctica, whereas it is estimated that the
Miocene glaciers covered 14 to 16.8 million square kilometers (5.4 to 6.5
million square miles).18 The much larger Miocene glaciers not only totally
covered the area once occupied by the Oligocene glaciers but also
extended much farther across the surface of Antarctica, eroding and
stripping away the glacial sedimentary deposits of the older Oligocene
glaciers in the process.
For this reason, sedimentary evidence for the existence of the Oi-1
glaciers is found o shore from Antarctica rather than on the mainland
itself. This evidence comes in the form of glacially derived, ice-rafted
debris found in o shore marine sediments. University of Michigan
geologist James Zachos and colleagues have documented the presence of
layers of angular quartz sands and heavy minerals at the Oi-1 stratigraphic
level on the Kerguelen Plateau in the southern Indian Ocean. These layers
contain medium-size19 and larger sand debris, debris that was too large to
have been transported o shore from Antarctica by wind and thus must
have been transported by ice (Zachos et al. 1992). Werner Ehrmann and
Andreas Mackensen, geologists at the Alfred Wegener Institute, have
reported the presence of gravel with pebbles at the same stratigraphic
horizon containing the ice-rafted sand deposits on the Kerguelen Plateau.
The presence of gravel in o shore marine deposits is unequivocal
evidence of ice rafting, and the presence of ice-rafted debris as far north
as 61°S on the Kerguelen Plateau is evidence of either a high frequency of
icebergs in the area or a few large debris-containing icebergs, both of
which evidence large-scale continental Oi-1 glaciation rather than small-
scale local glaciation in the Antarctic (Ehrmann and Mackensen 1992).
Other evidence for the Oi-1 glaciations on Antarctica comes from
models of glacioeustatic sea-level changes and from geochemical analyses
of Oligocene strata for evidence of changes in the temperature of sea-
surface waters. Sequence stratigraphic reconstructed sea-level curves
estimate a 67-meter (220-foot) drop in sea level at the Oi-1 stratigraphic
level (Katz et al. 2008), and independent models of ice-volume changes
with time give an average estimate of a 70-meter (230-foot) drop in sea
level (Pusz et al. 2011). Clearly the drop in sea level in the early Oligocene
was driven by the removal of water from the sea by the formation of
frozen-water deposits—glaciers—on land in Antarctica. Proof of a drop in
sea-surface temperature at the Oi-1 stratigraphic horizon is hampered by
the fact that the critical strata are missing in the rock record, further
evidence of a drop in sea level. Strata that have been preserved indicate
that a drop of 3 to 4°C (5 to 7°F) in sea-surface temperatures occurred
immediately before the Oi-1 glacial pulse, but the total decline in
temperature in the glacial pulse itself remains as yet unknown (Wade et al.
2012).
In summary, the onset of the Cenozoic Ice Age is dated to the earliest
Oligocene, the Oi-1 glacial pulse (table 1.6). Furthermore, the development
of glaciers in the Cenozoic Ice Age was not a continuous global cooling
process but rather took place in three glaciation steps in geologic time—
the early Oligocene, the middle Miocene, and the late Pliocene—and each
of these steps was associated with extinctions and biodiversity losses
(table 1.6).
I have proposed that the onset of the Late Paleozoic Ice Age was similar
in timing to that of the Cenozoic Ice Age, and that it also occurred in three
glaciation steps—the late Frasnian, the late Famennian, and the early
Serpukhovian (McGhee 2013, 203–212). It has been argued that glaciation
in the Early Carboniferous had become bipolar by the early Serpukhovian
Age: in the Southern Hemisphere, western Gondwana was covered in ice
sheets that would become comparable in size to those present in the
maximum phase of the Cenozoic Ice Age; in the Northern Hemisphere,
glaciers had formed on the Siberian crustal block in northern Asia
(González-Bonorino and Eyles 1995; Stanley and Powell 2003). Major
extinctions in the marine realm were also associated with the onset of the
Serpukhovian freezing; these glacial and biological events will be
considered in detail in chapter 2.
However, just as in the onset of the Cenozoic Ice Age, the beginning of
the Late Paleozoic Ice Age is not dated to the time when glaciation of the
Earth became bipolar, as the massive glaciers in Gondwana are much older
than the early Serpukhovian. Just as in the onset of the Cenozoic Ice Age,
the question of the timing of the onset of the Late Paleozoic Ice Age
becomes: when did the ice sheets form on Gondwana? In the late
Famennian Age of the Late Devonian, 363 million years ago (table 1.7),
massive ice sheets had already formed in Gondwana (Caputo et al. 2008).
These ice sheets covered at least 16 million square kilometers (6.2 million
square miles) of land in western Gondwana—present-day South America
and Africa ( g. 1.4) (Isaacson et al. 2008). And, as discussed previously, the
Famennian glaciation coincided with the seventh most ecologically severe
biodiversity crisis in the Phanerozoic (table 1.3).
FIGURE 1.4 Paleogeographic extent of the Famennian ice sheet (heavy dashed line) on the
continent of Gondwana, which covered parts of present-day South America and Africa. Lighter
dashed lines indicate continental shield regions, and numbered points indicate sedimentary basins
containing glacial strata.

Source: From Palaeogeography, Palaeoclimatology, Palaeoecology, volume 268, pp. 126–142, by P. E.

Isaacson et al., “Late Devonian–Earliest Mississippian Glaciation in Gondwanaland and Its
Biogeographic Consequences,” copyright © 2008 Elsevier. Reprinted with permission.

TABLE 1.7 The timing of geologic and biotic events in the proposed onset of the Late Paleozoic Ice
Age.

Age Geologic Time (Ma) Geologic and Biotic Events

TOURNAISIAN 347
 348
349 Cold Tournaisian Gap
350 ICE Tournaisian Gap
351 ICE Tournaisian Gap
352 ICE Tournaisian Gap
353 ICE Tournaisian Gap
354 Cold Tournaisian Gap
355 Cold Tournaisian Gap
356 Cold Tournaisian Gap
357 Cold Tournaisian Gap
358 Cold Tournaisian Gap
FAMENNIAN 359 ICE ← Land extinctions
360 ICE ← Marine extinctions
361 ICE
362 ICE
363 ICE
364
365
366
367 Cold Famennian Gap
368 Cold Famennian Gap
369 Cold Famennian Gap
370 Cold Famennian Gap
371 Cold Famennian Gap
372 ICE? ← Marine extinctions, Famennian Gap
FRASNIAN (pars) 373 ICE? ← Marine extinctions
374 ICE? ← Marine + land extinctions
375 ICE?
376
377
Source: Timescale modi ed from Gradstein et al. (2012); Famennian and Tournaisian glacial data are
from Caputo et al. (2008); biotic events modi ed from McGhee (2013).
Note: Ma = millions of years before the present.

At 16 million square kilometers, the late Famennian glaciers are

comparable in size to those that formed in the Miocene phase of the onset
of the Cenozoic Ice Age, which have been estimated to have been in the
range of 14 to 16.8 million square kilometers (5.4 to 6.5 million square
miles), as discussed earlier in this chapter.20 The sedimentary evidence of
glacial striated pavements and glacial tillites are still present in terrestrial
strata in South America and South Africa, and decimeter- to meter-size
(four-inch- to yard-size) ice-rafted glacial dropstones are found in
o shore marine strata of late Famennian age (Streel, Vanguestaine, et al.
2000; Isaacson et al. 2008; Brezinski et al. 2010). The formation of frozen
water masses on land the size of the Famennian glaciers should have
produced a major drop in sea level. When sea level falls, the coastlines of
the oceans retreat from the land. Land that was once under water is now
exposed and, more important, the mouths of rivers that used to empty
into the ocean are now far from the new coastline and at an altitude
higher than the new sea level. As a result, the rivers cut downward in their
river beds until they reach the new sea level, producing deeply incised
valleys. In the maximum phase of the Cenozoic Ice Age, valley incision
depths approached 100 meters (328 feet), indicating, of course, that the
level of the sea fell by 100 meters during the period of maximum ice
buildup on the land.
The late Famennian glaciation occurred over 359–363 million years ago
(table 1.7), but geological evidence of valley incision still exists in isolated
regions of the world. In both North America and Europe there are ancient
incised valleys with incision depths ranging from 75 to 90 meters (246 to
295 feet), almost as deep as those seen in the Pleistocene Epoch of the
Cenozoic Ice Age (Brezinski et al. 2010). Thus the severity of global cooling
in the late Famennian approached that seen in the maximum phase of the
Cenozoic Ice Age in terms of the amount of water frozen on the land.
So the Late Paleozoic Ice Age began in the late Famennian Age of the
Late Devonian, right? Many paleoclimatologists today would answer that
question with a “yes,” but many also have their doubts and think that the
onset of the Late Paleozoic Ice Age is even older than that. Sedimentary
evidence for glaciation and glacially driven changes in sea level in the
older Frasnian Age of the Late Devonian has increased steadily over time.
Around the world, cyclic patterns of sedimentation are known to exist in
Frasnian strata ( g. 1.5) that appear to represent patterns of sea-level rise
and fall that occur in cycles of around 10,000 to 100,000 years (Sandberg et
al. 1988; Montañez and Isaacson 2013; Becker et al. 2016; see also Elrick
and Witzke 2016).
FIGURE 1.5 Frasnian sedimentary cycles at Devils Gate, Nevada, that may be the result of
uctuations in sea level triggered by the expansion and contraction of glacial ice sheets in
Gondwana. The rock hammer spanning one such cycle in the center of the gure is 30 centimeters
(12 inches) tall.

Source: Photograph courtesy of Dr. Peter Isaacson, Department of Geological Sciences, University of
Idaho.

More recently, Jonathan Filer, a geologist at Towson University, has

demonstrated the existence of numerous sedimentary cycles in late
Frasnian subsurface strata that are continuous over 700 kilometers (435
miles) in eastern North America, revealed in the analysis of data from over
600 hydrocarbon test wells in the Appalachian Basin, and has argued that
these cycles are evidence of sea-level changes driven by the expansion and
contraction of ice sheets on land in Gondwana (Filer 2002). Wilson
McClung, a geologist at Chevron, and his colleagues have traced 12 of
Filer’s Frasnian sedimentary cycles from the subsurface to surface
outcrops and have argued that the ability to correlate these cycles in both
outcrop and subsurface, both parallel and perpendicular to the ancient
shoreline in the Appalachian Basin, con rms that the sedimentary cycles
are the product of global rises and falls in sea level produced by glacial
cycles of melting and freezing on land (McClung et al. 2013). Moreover,
McClung and colleagues have estimated that these 12 sedimentary cycles
had a temporal periodicity of around 375,000 years, and they have further
measured over 70 smaller-scale sedimentary cycles in outcrop with an
estimated periodicity of around 65,000 years, cyclic frequencies that are
similar in magnitude to the long-term orbital-eccentricity and rotational-
axis-obliquity climatic cycles (table 1.5) seen in the Cenozoic glaciations
(Zachos et al. 2001). Finally, McClung and colleagues have also
demonstrated the existence of incised-valley lls some 35 to 45 meters
(115 to 148 feet) thick in outcrop, evidence of a late Frasnian
glacioeustatic sea-level fall of about half the magnitude of the sea-level
fall that produced the incised valleys in the late Famennian glaciation.
(For an extensive discussion of the evidence for a major drop in sea level
during the late Frasnian—particularly the karsti cation of carbonate
platforms around the world—see McGhee 2013, 132–135.)
As yet, evidence for late Frasnian glaciation is found only outside of
Gondwana. On Gondwana itself, all e orts to nd late-Frasnian glacial
striated pavements and tillites, similar to those found in the late
Famennian, have failed. The western edge of the landmass of Gondwana
was positioned over the South Pole in the Frasnian as well as in the
Famennian ( gs. 1.2–1.4), so where were the Frasnian glaciers? I have
argued that a closer examination of the pattern of glacial onset seen in the
Cenozoic Ice Age may explain the enigma. In essence, I argue that the
absence of late Frasnian glacial sediments on Gondwana is to be expected,
and that Gondwana is the wrong place to search for such evidence
(McGhee 2014a, 2014b).
I predict that glacial sediments of late Frasnian age will never be
discovered on the landmass of Gondwana because they were removed by
the erosive action of the much larger glacier that formed in the late
Famennian, analogous to the removal of the Oi-1 glacial sediments on
Antarctica by the much larger Miocene glacier. For the same reason that
sedimentary evidence for the existence of the Oi-1 glaciers is found
o shore from Antarctica rather than on the mainland, I predict that
sedimentary evidence for the late Frasnian glaciers will be found o shore
from Gondwana rather than on the mainland. That evidence will come in
the form of glacially derived ice-rafted debris found in o shore marine
sediments (McGhee 2014a, 2014b).
The erosive e ects of the successive glaciations of the Late Paleozoic
Ice Age can be clearly seen in gures 1.3 and 1.4. Note that the existing
glacial strata of late Famennian age (Glacial I in g. 1.3) are found only in
northern Africa and northern South America, yet the position of the
South Pole in the late Famennian (360 Ma in g. 1.3) was to the south of
those outcrop areas. Clearly, glacial strata of late Famennian age once
existed all across middle and southern Africa and South America, a mirror
image about the South Pole to the existing glacial strata in northern
Africa and South America, yet those strata no longer exist—they were
removed by successive later phases of glaciation (Glacial II and Glacial III
in g. 1.3). A more detailed map of the distribution of late Famennian
glacial strata is shown in gure 1.4; note the question marks along the
southern and eastern margins of the heavy dashed line indicating the
geographic extent of the late Famennian ice sheet. The University of Idaho
geologist Peter Isaacson and his colleagues are clearly indicating here that
the Famennian ice sheet had to have extended further to the south and
east than the current existing glacial outcrops, but that those southern
and eastern Famennian glacial strata are no longer to be found.
As discussed above, McClung and his colleagues have demonstrated the
existence of Frasnian incised-valley lls that suggest a late Frasnian
glacioeustatic sea-level fall about 50 percent of the sea-level fall that
produced the incised valleys in the late Famennian. The minimum size of
the late Famennian glaciers has been measured to have been 16 million
square kilometers (6.2 million square miles) in western Gondwana
(Isaacson et al. 2008), and I have proposed that glaciers approximately 50
to 71 percent of the size of the Famennian ice sheet, or eight to 11 million
square kilometers (3.1 to 4.2 million square miles), were present in
western Gondwana in the late Frasnian (McGhee 2014a, 2014b). This
estimate is based on the scaling of the size range of the rst-step Oi-1
glaciers to the size range of the second-step Miocene glaciers at the onset
of the Cenozoic Ice Age, and the assumption that that scaling was similar
in the size ranges of the rst-step late Frasnian glaciers to the second-step
Famennian glaciers in the proposed onset of the Late Paleozoic Ice Age
(McGhee 2014a, 2014b).
Subsequent eldwork by McClung and his colleagues in the
Appalachian Mountains in North America has yielded further empirical
stratigraphic data that have allowed them to produce an independent
calculation for the possible size of the ice sheet proposed to have existed
in Gondwana during the late Frasnian. Their calculation falls in the range
of 5.6 to 7.4 million square kilometers (2.2 to 2.9 million square miles),
values that they noted are “similar to McGhee’s (2014) estimate for the
area of late Frasnian glaciation” (McClung et al. 2016, 139).
Based on either my larger estimate or the somewhat smaller estimate
of McClung and his colleagues, the late Frasnian ice sheet still would have
been totally covered by the much larger late Famennian glaciers. As in the
case of the Oi-1 glaciation in Antarctica, the late Famennian glaciers would
have erased the trace of the initial late Frasnian glaciers on Gondwana—
hence my prediction that glacial sediments produced by the late Frasnian
glaciers will only be found in marine sediments o shore from the
Gondwana landmass.
To test the hypothesis that glaciers formed in the late Frasnian, a
worldwide search should be initiated for the presence of ice-rafted debris
in late Frasnian marine strata deposited o shore from Gondwana. Rather
than the decimeter- to meter-size (four-inch- to yard-size) Famennian ice-
rafted debris, a search should be initiated for the presence of Frasnian ice-
rafted smaller debris of medium to coarse sand grains and gravels with
pebbles, similar to the ice-rafted debris found in the Oi-1 marine
sediments (Zachos et al. 1992; Ehrmann and Mackensen 1992).
Aside from the sedimentary evidence for glaciation, biological
evidence of global cooling during the late Frasnian biodiversity crisis has
been amassed by paleontologists and paleobotanists for over 36 years
(table 1.8). Biological evidence also exists for a six-million-year-long cold
interval following the late Frasnian biodiversity crisis in the early
Famennian, an interval of time I have called the Famennian Gap (table 1.7)
(McGhee 2013, 104–107).21 The late Famennian biodiversity crisis was
followed by a similar ten-million-year protracted cold interval in the Early
Carboniferous known as the Tournaisian Gap, which was punctuated with
an additional period of glaciation 350 to 353 million years ago (table 1.7).22
The names of both the Famennian and Tournaisian Gaps refer to gaps in
the fossil record in which very low diversity existed in both land animal
and plant species, and in marine species as well (McGhee 2013).

TABLE 1.8 Empirical biological observations in both marine and terrestrial ecosystems that have
been used to argue for global cooling in the late Frasnian.
A. MARINE ECOSYSTEMS
1. Di erential survival of high-latitude marine faunas:
• Brachiopods (Copper 1977, 1986, 1998)
• Microbial reef biota (Copper 2002)
2. Di erential survival of deepwater marine faunas:
• Glass sponges (McGhee 1996)
• Rugose corals (Oliver and Pedder 1994)
• Tornoceratid ammonoids (House 1988)
3. Migration of deepwater marine faunas into shallow waters:
• Glass sponges (McGhee 1996)
• Tornoceratid ammonoids (House 1988)
4. Blooms in cold-water plankton:
• Prasinophytes (Streel, Vanguestaine, et al. 2000)
• Radiolarians (Racki 1998, 1999; Copper 2002)
• Chitinozoans (Paris et al. 1996; Streel, Vanguestaine, et al. 2000; Grahn and Paris 2011)
5. Di erential survival of freshwater versus marine species:
• Acanthodian shes (Dennison 1979)
• Placoderm shes (Dennison 1978; Long 1993)
6. Latitudinal contraction of geographic range in surviving equatorial marine faunas:
• Foraminifera (Kalvoda 1990)
• Stromatoporoid and coral reefs (Stearn 1987; Copper 2002)
• Tentaculitoids (Wei et al. 2012)
• Trilobites (Morzadec 1992)
B. TERRESTRIAL ECOSYSTEMS
1. Di erential survival of high-latitude terrestrial biota:
• Land plants (Streel, Caputo, et al. 2000)
2. Latitudinal contraction of geographic range in surviving equatorial terrestrial biota:
• Land plants (Streel, Caputo, et al. 2000)
• Tetrapod vertebrates (McGhee 2013)

Source: Modi ed from McGhee (2014a).

 As discussed above, the onset of the Cenozoic Ice Age was stepwise,
with the rst step of glaciation taking place in the early Oligocene,
followed by a warming period, and the second step of glaciation taking
place in the middle Miocene (table 1.6). If the onset of the Late Paleozoic
Ice Age also took place in a stepwise fashion, with the rst step of
glaciation taking place in the late Frasnian and the second step of
glaciation taking place in the Famennian (table 1.7), then the timing of the
onset of the glaciations and associated extinction pulses in the Paleozoic
and Cenozoic Ice Ages is strikingly similar (table 1.9). In the onset of the
Cenozoic Ice Age, the time interval between the rst-step extinctions in
the Oligocene and the second-step extinctions in the Miocene was about
19 million years; in the proposed onset of the Late Paleozoic Ice Age, the
time interval between the onset of the Frasnian and Famennian
extinctions was about 14 million years. The sequential severity of the
extinctions was also the same: in the Cenozoic Ice Age, the rst-step
Oligocene extinctions were much more severe than the second-step
Miocene (Prothero 1994; Prothero et al. 2003), and in the proposed onset
of the Late Paleozoic Ice Age, the rst-step Frasnian extinctions were
much more severe than the second-step Famennian (table 1.3). Finally,
Cenozoic glaciers persisted for ve million years following the Oligocene
extinctions, and the Famennian Gap cold interval persisted for six million
years following the Frasnian extinctions. If the Earth cooled at a similar
rate in the Cenozoic and Late Paleozoic, then the temporal spacing and
magnitudes of extinction seen in these two time intervals may not be
coincidental (table 1.9).23

TABLE 1.9 Biological similarities between the stepwise onset of the Cenozoic Ice Age and proposed
stepwise onset of the Late Paleozoic Ice Age

Cenozoic Ice Age Late Paleozoic Ice Age

A. Glacial Step 2
Miocene extinction pulses 13–14 million Famennian extinction pulses 359–360 million
years ago years ago
B. Glacial Step 1
Oligocene extinction pulses 32–33 million Frasnian extinction pulses 372–374 million years
years ago ago
C. Severity of extinction pulses
Oligocene (Step 1) > Miocene (Step 2) Frasnian (Step 1) > Famennian (Step 2)
D. Time interval between onset of extinction pulses 1 and 2
19 million years 14 million years
E. Duration of Antarctic glaciers E. Duration of Famennian Gap cold interval
following Oligocene extinction pulses following Frasnian extinction pulses
ve million years six million years

Did the Late Paleozoic Ice Age begin in the late Frasnian or the late
Famennian? Current paleogeographic reconstructions of the Earth in the
late Famennian show a large ice cap at the South Pole, on the western
margin of Gondwana ( g. 1.6), caused by the presence of glacial deposits
in South American and Africa that are of late Famennian in age ( gs. 1.3
and 1.4). The western margin of Gondwana was also positioned on the
South Pole in the late Frasnian ( g. 1.2), but no ice cap is shown there at
present—will one be demonstrated to have existed there in the future?
Abundant circumstantial evidence exists for late Frasnian glaciation ( g.
1.5, tables 1.6–1.9); what remains to be discovered is a few remnant glacial
strata of late Frasnian age on Gondwana or ice-rafted glacial debris in
marine sediments o shore from the Gondwana landmass, similar to that
found for the Oligocene glaciation in Antarctica.
FIGURE 1.6 Paleogeography of the Earth in the late Famennian Age. Note the similarity of the
positioning of the continents in the Famennian world to that of the Frasnian ( g. 1.2), but the
presence of an ice cap at the South Pole in the Famennian world.

Source: Global Paleogeography and Tectonics in Deep Time © 2016 Colorado Plateau Geosystems
Inc. Reprinted with permission.

In either event, the glaciations of the Late Devonian were unipolar,

con ned to the giant southern continent of Gondwana ( g. 1.6), and thus
were only harbingers of the bipolar glaciation that was to come with the
formation of a polar sea-ice cap in the Northern Hemisphere of the Earth.
That freezing came in the Serpukhovian Age of the Early Carboniferous,
and we will explore its consequences in the next chapter.
2 The Big Chill

The long duration of the LPIA [Late Paleozoic Ice Age] sets it apart from earlier
Paleozoic intervals of massive glaciation…. Glaciation was bipolar: glacial
marine deposits of Serpukhovian age provide the earliest evidence of late
Paleozoic glaciation in Siberia, although ice volume was greatest in the
Southern Hemisphere, where glaciers expanded over broad areas of
Gondwana .
—Stanley and Powell (2003, 877–878)

THE STRANGE WORLD OF THE TOURNAISIAN GAP

The world at the dawn of the Early Carboniferous Epoch was very di erent
from that of the Late Devonian.1 On land, the great forests of the Late
Devonian were gone—the largest land plants stood only two meters (6.6
feet) high, in stark contrast to the towering 30-meter-high (100-foot-high)
Archaeopteris hibernica trees that were so widespread before the end-
Devonian extinctions (we will consider Archaeopteris trees in more detail
in chapter 3). The land plants that did survive into the Early Carboniferous
were mostly shrubby—the dense stands of trees and shady forests of the
Late Devonian had vanished from the Earth. Only in the Visean Age of the
Early Carboniferous would tall trees once again evolve, chie y the 40-
meter-tall (131-foot-tall) woody seed-tree Pitus primaeva.2
 In the oceans, the great reefs were gone. The Late Devonian reefs were
the largest, most widespread reef tracts ever to exist in Earth history—
covering some ve million square kilometers (almost two million square
miles) of shallow marine sea oor. They were hit hard in the Late Devonian
extinctions, and it took 130 million years before the corals recovered and
once again became major reef-building organisms in the Middle Triassic.
Even the composition of the shell sh in the seas had changed: the
Devonian had been the golden age of the ancient brachiopods,3 the
lampshells, but now the more modern shell sh species of the molluscs,4
the clams and oysters, had diversi ed following the extinction of over
three-quarters of all brachiopod species. The giant predatory armored
shes of the Devonian world were all extinct. Never again would the
oceans be populated by huge sh as big as modern-day killer whales but
with strange bony armor around their heads and no teeth! Instead of
teeth, they had sharp bone blades in their mouths that, similar to the
sharp bone beaks of modern-day snapping turtles, functioned very
e ciently in slicing up prey animals ( g. 2.1).

FIGURE 2.1 Bone shield of the Late Devonian armored sh Dunkleosteus.

Source: Redrawn from Moy-Thomas and Miles (1971) in Invertebrate Palaeontology, by M. J. Benton,
copyright © 2015 Michael J. Benton. Reprinted with permission.

Welcome to the world of the Tournaisian Gap, a peculiar world of

depauperate ecosystems and a gap in evolutionary innovation that was to
last some ten million years in the Tournaisian Age.5 For most of the
Tournaisian Gap, fossils of our ancestors, the tetrapod vertebrates, are
missing in the rock record. Only in the very latest Tournaisian Age, and
continuing into the following Visean Age, would fossils of the tetrapod
vertebrates once again be found in the rock record. Why? The answer
appears to be that the tetrapods existed in such small populations during
the Tournaisian Gap that the probability that any individual specimen
would be preserved in the fossil record was very low. The tetrapods had
been hit very hard by the Late Devonian extinctions and barely survived
into the Early Carboniferous.6 However, in the late Tournaisian and early
Visean Ages, the tetrapods began to recover and once again to spread
across the land areas of the Earth—and to show up in the fossil record.
The Tournaisian Gap world was the product of a classic evolutionary
phenomenon known as an evolutionary bottleneck—a period of time in
which major losses of biodiversity occur and only a small fraction of the
individuals and species that existed before the “bottleneck” survive. An
evolutionary bottleneck is produced when the population sizes of a given
species shrink almost to the critical minimum level from which a species
cannot recover. One of the immediately observable consequences of a
species’ surviving an evolutionary bottleneck is a sharp reduction of
genetic diversity in the population of survivors. Numerous examples of
this phenomenon are evident in nature. The cheetahs in Africa survived a
severe evolutionary bottleneck, but the remaining cheetahs have such low
genetic diversity that they are still endangered with extinction. They
came very close indeed to the minimum-viable-population limit.
European and Asian populations of humans, Homo sapiens , also show the
characteristic low genetic diversity that results from passing through an
evolutionary bottleneck. It is estimated that all European and Asian
humans could be the descendants of as few as 160 people, possibly a single
clan of hunter-gatherers, who managed to exit Africa by crossing the Bab
al-Mandab (Gate of Grief) of the Red Sea about 50,000 years ago (Wade
2006).
The Tournaisian Gap world was actually the product of two
bottlenecks, which I call the End-Frasnian Bottleneck and the End-
Famennian Bottleneck (McGhee 2013). Two bottlenecks are worse than
one, as we will see. For example, the giant armored sh of the Late
Devonian ( g. 2.1) were severely a ected by the End-Frasnian Bottleneck
and lost half of their phylogenetic lineages and genetic diversity—yet they
managed to survive. They survived in a weakened condition, however, and
the End-Famennian Bottleneck proved fatal. None survived that diversity
constriction, and the genetic lineage of the great armored shes was
terminated forever.
The End-Frasnian Bottleneck sharply reduced the morphological
diversity that was previously present in the tetrapod lineages. The low
morphological diversity seen in the survivors re ects their low genetic
diversity, and only the more conservative or plesiomorphic (close to the
ancestral condition) tetrapods survived the bottleneck constriction. In
particular, the diversity of sizes seen in Frasnian tetrapods was lost, as
only the midsize forms survived into the Famennian. The worldwide
geographic distribution of the Frasnian tetrapods was also lost in the
bottleneck constriction, as only the Laurussian tetrapods managed to
survive. The same evolutionary bottleneck phenomena of reduction in
morphological variance and geographic range are seen in tetrapod species
that survived the End-Famennian Bottleneck—but now evolutionary
reversals occurred in the survivors as well during the Tournaisian Gap.
The Famennian tetrapods had made major evolutionary advances in their
step-by-step emergence from the rivers and lakes of the Earth and their
invasion of the land: they had lost their shlike teeth and lateral-line
systems—used by sh to “hear” in water, a trait useless on dry land—and
they had lost their internal gills and could no longer breathe in water but
relied exclusively on their lungs to breathe air. When the tetrapods begin
to show up again in the fossil record in the late Tournaisian and the
Visean, following the Tournaisian Gap, it is shocking to see that many of
the tetrapod groups had abandoned dry land entirely and returned to an
aquatic mode of life. What could have happened during the span of the
Tournaisian Gap that would trigger a period of reverse evolution? What
caused many tetrapod species to retreat from the land and go back into
the water?7
Only the tetrapods of the family of the whatcheeriids seem to have
managed to begin to diversify in the late Tournaisian Age. Starting with
the only known Tournaisian species of the family, Pederpes nneyae from
Scotland ( g. 2.2), the whatcheeriids diversi ed and by the mid-Visean
had achieved a worldwide distribution.8 By the early Visean Age, other
tetrapod groups were recovering their population numbers, were
producing new evolutionary innovations and clades, and were
geographically on the move. No less than 23 new tetrapod species
appeared in the fossil record by the late Visean (see McGhee 2013, table
7.4), proof that the harsh climatic conditions—whatever they may have
been—of the Tournaisian Gap world had ended.
FIGURE 2.2 Reconstruction of the whatcheeriid tetrapod Pederpes nneyae from Scotland. Note the
presence of six toes on the forefoot and the laterally placed eyes in the head. The animal was about
one meter (3.3 feet) long.

Source: Illustration by Kalliopi Monoyios © 2013. Reprinted with permission.

THE RETURN OF THE GLACIERS

Evidence for the existence and timing of the glaciation in the Famennian
Age of the Late Devonian (table 1.7) is undisputed, but once we cross into
the Tournaisian Age of the Early Carboniferous, the data are more open to
question (table 2.1). In a comprehensive 2003 summary of the geologic
evidence for glaciation, University of Wisconsin geologist John Isbell and
his colleagues argued that three distinct glacial phases could be
recognized in the Late Paleozoic Ice Age: Phase GI, the time span from the
Famennian Age in the Late Devonian through the Tournaisian Age in the
Early Carboniferous; Phase GII, the time span from the Serpukhovian Age
in the Early Carboniferous through the Bashkirian Age of the Late
Carboniferous; and Phase GIII, the time span from the late Moscovian Age
in the Late Carboniferous to the middle of the Sakmarian Age in the Early
Permian (see table 1.4 for the geologic timescale in this interval of time).
Isbell and colleagues (2003) argue that glacial strata of the Famennian and
Tournaisian ages, their proposed glacial phase GI, are present in the
northwestern Gondwana basins in Brazil and Bolivia in South America and
in the Tim Mersoï basin in north-central Africa. Thus in the rst column
in table 2.1, the glaciation that started in the Famennian Age of the Late
Devonian is shown to continue throughout the entire Tournaisian Age of
the Early Carboniferous, and the entire Visean Age of the Early
Carboniferous is shown to be free of continental glaciers.

TABLE 2.1 The timing of geologic and biotic events in the Tournaisian and Visean Ages of the Early
Carboniferous.

Age Geologic Time (Ma) Geologic Events Biotic Events

(1) (2) (3)
VISEAN 331
332
333 ICE
334 ICE
335 ICE ICE
336 ICE ICE
337 ICE ICE
338 ICE ICE
339 ICE ICE
340 ICE ICE
341
342
343
344
345 ICE
346 ICE
TOURNAISIAN 347 ICE ICE
348 ICE ICE
349 ICE ICE Tournaisian Gap
350 ICE ICE ICE Tournaisian Gap
351 ICE ICE ICE Tournaisian Gap
 352 ICE ICE ICE Tournaisian Gap
353 ICE ICE ICE Tournaisian Gap
354 ICE ICE Tournaisian Gap
355 ICE Tournaisian Gap
356 ICE Tournaisian Gap
357 ICE Tournaisian Gap
358 ICE ICE Tournaisian Gap
FAMENNIAN (pars) 359 ICE ICE ICE ← Land extinctions
360 ICE ICE ICE ← Marine extinctions

Source: Geologic events column (1) from Isbell et al. (2003), column (2) from Caputo et al. (2008),
and column (3) from Frank et al. (2008). Biotic events from table 1.7; timescale modi ed from
Gradstein et al. (2012).

In contrast, the University of Pará geologist Mário Vicente Caputo and

his colleagues (2008) argue that glacial strata only from the middle
Tournaisian Age can be reliably dated biostratigraphically,9 and that the
early Tournaisian and late Tournaisian time intervals may have been cold
but free of continental ice sheets. A major fall in sea level did occur in the
middle Tournaisian in North America on the Laurussian continent to the
north of the giant continent Gondwana, leading to erosional valley
incisions as deep as 60 meters (200 feet) in some areas (Kammer and
Matchen 2008), and it has been argued that this sea-level drop was
triggered by the formation of glacial ice sheets on the land.10
Furthermore, Caputo and colleagues report the existence of glacial
strata in Brazil and Argentina that have been biostratigraphically dated to
the middle Visean Age of the Early Carboniferous,11 an interval of time
that was paleoclimatically reconstructed as glacier-free by Isbell and
colleagues (see table 2.1). The biostratigraphic dating of these glacial
strata to the middle Visean has been corroborated by radiometric dating
of volcanic strata that occur with the glacial strata in Argentina. Uranium-
lead dating gives an age of 335.99 ± 0.06 million years ago, a date that falls
in the middle of the Visean Age (Gulbranson et al. 2010).12 Milo Barham, a
geologist at the National University of Ireland, and his colleagues have
presented further geochemical evidence of a sharp 4.5°C (8.1°F) drop in
sea-surface temperatures that occurred 336 million years ago, consistent
with the onset of global icehouse conditions (Barham et al. 2012). The
sedimentary strata in scattered regions around the world also begin to
show cycles in sedimentary deposition that have been argued to have
Milankovitch periodicities in the middle to late Visean. It is argued that
cyclic-deposited sedimentary strata in Europe and North America were
driven by periodic uctuations in global sea level as continental ice sheets
waxed and waned on Gondwana in the Visean (Schwarzacher 1989; Smith
and Read 2000; Wright and Vanstone 2001; Gastaldo, Purkyňová, Šimůnek,
and Schmitz 2009; Bishop et al. 2009; Giles 2009). Thus, in the second
column of table 2.1, only the middle of the Tournaisian Age is shown to be
glaciated, but, unlike in column 1, the middle of the Visean Age is also
shown to be glaciated.
Finally, the University of Nebraska geologist Tracy Frank and
colleagues (2008) have conducted geochemical analyses of oxygen-isotope
ratios in ancient brachiopod shells in an e ort to reconstruct
paleoclimatic uctuations in the Early Carboniferous (see McGhee 2013,
196–199). Collecting brachiopod isotopic data from North America,
Europe, Iran, and China—all o shore from the giant continent Gondwana
—they reconstructed the earliest Tournaisian to have been glacial in
Gondwana, followed by a warming episode and ice-free conditions until
the middle Tournaisian, global cooling and a return to glaciation from the
middle Tournaisian through the earliest Visean, another warming episode
and ice-free conditions in the early Visean, and a return to glacial
conditions in the middle Visean (column 3 in table 2.1).
 In summary, hard empirical data from dated glacial strata in Brazil and
Argentina demonstrate that Gondwana was glaciated in the middle
Tournaisian and in the middle Visean. Geochemical analyses suggest an
expansion in the duration of the middle Tournaisian glaciation from four
to ten million years (extending into the earliest Visean), and an expansion
in the duration of the middle Visean glaciation from six to eight million
years (table 2.1). However, regardless of the precise pattern of the
development of the Tournaisian and Visean glaciers, those glaciers were
still unipolar, con ned to the giant southern continent of Gondwana.

THE ONSET OF BIPOLAR CONTINENTAL ICE?

The freezing of the Earth began in earnest in the Serpukhovian Age.
Following a two- to four-million-year warmer interval in the late Visean
(table 2.1), the ice returned to western and central Gondwana with a
vengeance in the Serpukhovian and persisted through the remainder of
the Early Carboniferous and through the Bashkirian Age in the following
Late Carboniferous—a time span of some 15 million years of continental
glaciation (Frank et al. 2008). This time, however, the glaciers were not
con ned to western and central Gondwana but spread to eastern
Gondwana—present-day Australia—as well. The University of Nebraska
geologist Christopher Fielding and his colleagues (Fielding, Frank,
Birgenheier et al. 2008) document four distinct pulses of glacial advance
and retreat in eastern Gondwana, which they have labeled C1 (for rst
Carboniferous glacial pulse, the oldest) through C4 (for fourth
Carboniferous glacial pulse, the youngest; see column 2 in table 2.2). The
rst glacial pulse, C1, consisted of alpine glaciers that formed in highland
areas of eastern Gondwana. It was short-lived, but the next glacial pulse,
C2 in the Serpukhovian, was much colder, and continental glaciers formed
for the rst time in eastern Gondwana (table 2.2).

TABLE 2.2 The possible onset of bipolar glaciation during the Late Paleozoic Ice Age.

Age Geologic Time (Ma) South Pole Climatic Events North Pole
West Gondwana East Gondwana Siberia
MOSCOVIAN 308 ICE
309 ICE
310 ICE
311 ICE ICE?
312 ICE ICE?
313 ICE ICE?
314 ICE ICE?
315 ICE ICE?
BASHKIRIAN 316 ICE ICE-C4 ICE?
317 ICE
318 ICE ICE
319 ICE ICE
320 ICE ICE
321 ICE ICE-C3
322 ICE
323 ICE
SERPUKHOVIAN 324 ICE ICE ICE?
325 ICE ICE ICE?
326 ICE ICE-C2 Cold
327 ICE
328 ICE
329 ICE ICE
330 ICE ICE-C1
VISEAN (pars) 331
332
Source: Gondwana data from Isbell et al. (2003), Frank et al. (2008), and Fielding, Frank, Birgenheier
et al. (2008); Siberian data from Epshteyn (1981a), Chumakov (1994), and Raymond and Metz (2004).
Timescale modi ed from Gradstein et al. (2012).
Note: Bold type indicates the existence of continental glaciers and normal type indicates the
presence of alpine glaciers, where glaciations in eastern Gondwana are designated C1 through C4,
from oldest to youngest.

Independent geochemical analyses of oxygen isotope ratios in

brachiopod shells from the Russian Platform on the continent of Laurussia
—far to the north of Gondwana—reveal a major positive anomaly13 in the
late Serpukhovian, an anomaly that is argued to be evidence of a buildup
of ice volume on the Earth that was equivalent “to a full Pleistocene
glaciation” (Mii et al. 2001, 144). The analyses of the Universidad Nacional
de Salta glacial geologist Gustavo González-Bororino and his University of
Toronto colleague Nicholas Eyles agree with that glacial magnitude
assessment and indicate that, in the late Serpukhovian cooling of the
Earth, an ice cap formed on Gondwana that covered some 21 million
square kilometers (eight million square miles) of land, just a little less
than the 23.5-million-square-kilometer (9.1-million-square-mile) ice cover
that developed on the Earth in the Last Glacial Maximum (LGM) phase of
the Cenozoic Ice Age (González-Bororino and Eyles 1995).
But did the glaciers become truly bipolar in the Serpukhovian freezing
event—did ice sheets exist on the continental landmasses both in the
giant southern continent of Gondwana and in the high-latitude, Siberian
region of the northern continent Laurussia? In our present-day world,
only two large areas of continental glaciation still exist (and they are
rapidly dwindling in size). These are the glacial ice cap in Antarctica,
which covers 13,720,000 square kilometers (5,297,322 square miles) of
land, and the glacial ice cap in Greenland, which covers 2,166,097 square
kilometers (836,335 square miles). Together they total some 15,886,097
square kilometers (6,133,657 square miles) of the Earth that is covered by
ice today. But in the Pleistocene maximum glaciation phase of the
Cenozoic Ice Age, much of northern Europe was covered by the
Fennoscandian continental glacier, and much of northern North America
was covered by the Laurentide continental glacier. When these two giant
glaciers are added to the mix, then a total of 23.5 million square
kilometers (9.1 million square miles) of the landmasses of the Earth were
covered by ice in the LGM of the Cenozoic Ice Age (González-Bororino and
Eyles 1995). Was the continental coverage of ice on the Earth in the
Serpukhovian similar to that seen in the LGM in the Cenozoic, with a
continental ice cap at both poles?
Many paleontologists think that continental glaciation became truly
bipolar in the Serpukhovian, as evidenced by the epigraph at the
beginning of this chapter from the work of the paleontologists Steve
Stanley and Matthew Powell (see also Fielding, Frank, and Isbell 2008). The
analysis of faunal migration patterns in marine species by Anne Raymond
and Cheryl Metz, paleontologists at Texas A&M University, indicate that
the North Pole had cooled signi cantly in the late Serpukhovian, and it is
probable that massive sea ice had formed o shore from Siberia (column 3
in table 2.2) (Epshteyn 1981a; Chumakov 1994; Raymond and Metz 2004).
By the end of the Bashkirian Age, it has been argued that continental ice
sheets were present in Siberia in the Northern Hemisphere, evidenced by
the presence of ice-rafted debris and glacial-marine sediments in the rock
record.14 Thus by late Bashkirian and into the early Muscovian (table 2.2),
the Late Paleozoic Ice Age had de nitely become bipolar—or had it?
The data have been called into question by the University of Wisconsin
geologist John Isbell and his colleagues (2016), who argue that a sea-ice-
cap coverage of the North Pole (like today’s) is still not a continental, land-
ice coverage, and that the Bashkirian-Muscovian data are evidence only
for the presence of sea ice. They argue that continental glaciation on the
Earth was never truly bipolar in the late Paleozoic, and that all continental
glaciers were con ned to the Southern Hemisphere. We will return to
Isbell and colleagues’ arguments when we discuss the Permian, where the
bipolarity question again arises. In light of this ongoing debate, I have
placed question marks on the distribution of Northern Hemisphere
continental ice in table 2.2.
Whatever the distribution of continental ice on the Earth in the
Serpukhovian, the e ect of the Serpukhovian “Big Chill” was devastating
to marine life—precipitating the sixth most severe ecological crisis in the
Phanerozoic, greater even than that of the glacially triggered end-
Famennian extinctions (table 1.3). The biological e ects of that crisis will
be examined in detail in the next two sections of this chapter.

THE SERPUKHOVIAN CRISIS IN THE OCEANS

The Serpukhovian biodiversity crisis marks the boundary between the
Early and Late Carboniferous Epochs (or the Mississippian and
Pennsylvanian Epochs in the United States; see table 1.4) (Gradstein et al.
2012). Although the biodiversity crisis was large enough to allow a major
subdivision in the geologic timescale to be recognized, it has received
relatively little analysis until recently. Indeed, the University of
Birmingham paleontologist Tony Hallam and his colleague, the University
of Leeds paleontologist Paul Wignall, have commented (1997, 92): “The
importance ascribed to the event varies from author to author: Saunders
and Ramsbottom (1986) called it a major extinction, and others (Ziegler
and Lane 1987; Weems 1992), with undoubted exaggeration, ranked it as a
mass extinction comparable to the big ve.” The “big ve” that Hallam
and Wignall refer to are the end-Ordovician, Late Devonian (end-
Frasnian), end-Permian, end-Triassic, and end-Cretaceous mass
extinctions (see table 1.3), rst recognized by the University of Chicago
paleontologists Dave Raup and Jack Sepkoski in biodiversity analyses they
conducted back in 1982. These ve extinction events triggered the ve
largest losses of biodiversity in the Earth’s oceans in Phanerozoic history,
although the ecological impact of the end-Ordovician event was relatively
minor in contrast to the end-Cretaceous event, which triggered the least
loss of biodiversity of the big ve but precipitated the second most severe
ecological disruption in the entire Phanerozoic, as discussed in chapter 1
(table 1.3) (McGhee et al. 2004).
My colleagues Peter Sheehan, Dave Bottjer, and Mary Droser and I have
conducted comparative paleoecological analyses that have revealed that
the ecological impact of the Serpukhovian biodiversity crisis was greater
than that of the end-Ordovician (one of Raup and Sepkoski’s big ve) and
end-Devonian (end-Famennian) extinctions (table 1.3)—both of which
were triggered by brief but intense glaciations, as discussed in chapter 1.
Yet the magnitude of the biodiversity loss that occurred in the
Serpukhovian crisis was less than in any of the big ve—much less than in
the end-Ordovician crisis—so why did the Serpukhovian crisis cause a
larger ecological disruption? There is a signi cant di erence in the
mechanism by which biodiversity was lost in the Serpukhovian crisis as
opposed to the end-Ordovician crisis. In the end-Ordovician, as well as the
end-Permian and end-Cretaceous crises, diversity loss was driven by a
sharp major increase in the extinction rate (Bambach et al. 2004). In
contrast, the Serpukhovian biodiversity loss was driven by a precipitous
drop in the speciation rate at the beginning of the Serpukhovian during a
period of elevated extinction rates (Bambach et al. 2004, g. 6; Stanley
2007), as shown in gure 2.3. In this, the Serpukhovian biodiversity crisis
is more similar to the Late Devonian (end-Frasnian) biodiversity crisis
(McGhee 1988, 1996) and the end-Triassic biodiversity crisis (Bambach et
al. 2004) than to the end-Ordovician, end-Permian, and end-Cretaceous
mass extinctions.

FIGURE 2.3 The magnitude of Late Paleozoic marine extinction rates (dashed line and square data
points) and origination rates (solid line and diamond data points) is measured in percent change in
generic diversity per time interval (vertical axis). Geologic timescale abbreviations (horizontal
axis): ORD, Ordovician; SILUR, Silurian; DEVON, Devonian; E CARB, Early Carboniferous; L CARB,
Late Carboniferous; PERM, Permian; and SERP, Serpukhovian (within the gure); see text for
discussion.

Source: From McGhee et al. (2012), modi ed from Stanley and Powell’s (2003) original gure.

The University of Hawaii paleontologist Steve Stanley and his

colleague, the University of California (Santa Cruz) paleontologist
Matthew Powell, have argued that the Serpukhovian crisis initiated “a
new state of the global marine ecosystem” (Stanley and Powell 2003, 877)
and “reordered the marine ecosystem to one dominated by widespread
genera, which had intrinsically low macroevolutionary rates” (that is, low
speciation and extinction rates) (Powell 2008, 525). This phenomenon can
be seen clearly in gure 2.3: both extinction and speciation rates plummet
at the end of the Serpukhovian and remain low for the entire duration of
the major phase of the Late Paleozoic Ice Age, from the Early
Carboniferous through the Middle Permian. There was no signi cant
recovery of biodiversity in marine ecosystems after the Serpukhovian
crisis. Instead, both speciation rates and extinction rates remained
anomalously low for 50 million years ( g. 2.3), and the marine fauna of the
Late Paleozoic Ice Age were characterized by “faunal persistence, not
origination or extinction” (Bonelli and Patzkowsky 2011, 14) in a
protracted period of ecological and evolutionary stagnation in
paleotropical regions.15
The ecological dynamics of the global marine ecosystem have been
characterized by a series of evolutionary faunas during the Phanerozoic,
rst recognized in the biodiversity analyses of Jack Sepkoski (1984, 1990).
We will examine these evolutionary faunas in great detail in chapter 6, but
here let us brie y consider some characteristics of two of them: the
Paleozoic evolutionary fauna and the modern evolutionary fauna. In
general, the older Paleozoic evolutionary fauna was characterized by a
higher rate of extinction than the younger modern evolutionary fauna,
which is more extinction resistant, a phenomenon we considered brie y
in chapter 1. Steve Stanley’s analyses have revealed that the Serpukhovian
crisis was the most ecologically selective of the Paleozoic biodiversity
crises in that it preferentially eliminated species of the Paleozoic
evolutionary fauna, and spared the species of the modern evolutionary
fauna, to a much greater extent than any of the other Paleozoic crises
(Stanley 2007). For example, the Serpukhovian crisis eliminated genera of
the Paleozoic evolutionary fauna relative to the genera of the modern
evolutionary fauna in a ratio of 2.4 to 1, in contrast to the end-Ordovician
mass extinction, which had a Paleozoic-to-modern evolutionary fauna
extinction ratio of only 1.4 to 1. The modern evolutionary fauna
experienced a biodiversity loss of only 15.4 percent in the Serpukhovian
crisis, the smallest modern evolutionary fauna diversity loss in the any of
the Paleozoic biodiversity crises (Stanley 2007). In essence, the
Serpukhovian biodiversity crisis was a harbinger of the demise of the
dominant Paleozoic evolutionary fauna and its ecological replacement by
the modern evolutionary fauna in the Mesozoic; we will consider that
demise in great detail in chapter 6, “The End of the Paleozoic World.”
The Serpukhovian crisis was also ecologically selective in that it
preferentially spared marine genera with broad latitudinal geographic
ranges and eliminated those with narrow latitudinal distributions (Powell
2005, 2008). Matthew Powell has further argued that the key selective
factor underlying this ecological di erentiation was thermal tolerance; no
other ecological trait, such as niche breadth, geographic dispersal range,
species richness, body size, or habitat position in an onshore-o shore
gradient, played a signi cant role in survival in the Serpukhovian crisis
(Powell 2008). Because these other ecological traits are taken as selective
in times of background extinction (Jablonski 1986), as opposed to mass
extinction, Powell has proposed that the Serpukhovian biodiversity crisis
“adhered to a mass-extinction, not background, regime” (Powell 2008,
526). Thus, contrary to the assessment of Tony Hallam and Paul Wignall
(quoted at the beginning of this section of the chapter) that to consider
the Serpukhovian crisis as comparable to the big ve extinctions was an
exaggeration (Hallam and Wignall 1997, 92), Powell does consider the
Serpukhovian biodiversity crisis to have possessed the qualities of a mass
extinction comparable to the big ve.
Finally, although the great Late Devonian reefs—dominated by ancient
and now extinct stromatoporoid sponges, tabulate corals, and rugose
corals—had been destroyed in the Devonian extinctions, the rugose corals
(or horn corals, since they looked like the horns of a bull) had survived
into the Early Carboniferous. But those rugose coral survivors were
strongly impacted by the Serpukhovian biodiversity crisis: the
ecologically dominant rugose species of the Early Carboniferous were
decimated and were ecologically replaced by entirely new rugose coral
species in the following Late Carboniferous. This ecological-replacement
event has been called the “mid-Carboniferous rugose coral evolutionary
event” (Wang et al. 2006, 339) in which the Serpukhovian crisis triggered a
major ecological shift from a coral fauna dominated by large, solitary
corals to one of massive colonial corals that, it is argued, have been
adapted to the changed marine conditions associated with the onset of
glaciation.
It is interesting that both the end-Ordovician mass extinction and the
Serpukhovian biodiversity crisis were triggered by glaciations. Other than
that common trigger, the two events were very di erent. Glaciation in the
Ordovician triggered an enormous jump in the extinction rate of marine
organisms ( g. 2.3) and major biodiversity losses, yet the ecological
impact of those extinctions was minimal. Glaciation in the Serpukhovian
triggered a precipitous drop in the speciation rate ( g. 2.3) but only
moderate biodiversity losses, yet the ecological impact of those diversity
losses and ecosystem restructuring was an order of ecological magnitude
larger (table 1.3).
Why were these two events so di erent? First, in contrast to the
Serpukhovian glaciations, the primary Ordovician glaciations ended
abruptly, although there were signi cant brief glacial advances and
accompanying sea-level changes during the early Silurian (Finnegan et al.
2011; Harris and Sheehan 1998). These latter glacial events may have set
back recovery for brief intervals, but in general faunas recovered more or
less continuously during the early Silurian. Thus, in the post-Ordovician,
glacial changes were still ongoing but not of su cient strength to reset to
a new state in the global marine ecosystem as it did in the post-
Serpukhovian.
Second, unlike the post-Ordovician, the Serpukhovian glaciations
struck the descendants of a fauna that had already survived a series of
glaciation events (tables 1.7 and 2.1) in the end-Famennian, mid-
Tournaisian, mid-Visean Ages—and possibly in the end-Frasnian Age as
well. Thus, in the Serpukhovian, global cooling did not trigger a sharp
jump in extinction rates of marine organisms, as it did in the Ordovician
( g. 2.3). Rather, it was the loss of habitat diversity in the new globally
homogeneous marine environments that triggered a drop in speciation
rate and subsequent loss of biodiversity. Major glaciation continued in the
post-Serpukhovian, and the cool and globally similar marine
environments (only cool-water niches were available to be lled, warm
tropical niches did not exist; see Powell 2005) perpetuated the reset in
global marine ecosystems triggered by the Serpukhovian biodiversity
crisis. Ultimately, the vast expansion of forests in the Carboniferous may
have been an ecological trigger for some of the di erences between the
Serpukhovian and Ordovician glaciations. Without abundant land plants,
carbon dioxide could not be drawn down in the Silurian as easily as in the
Carboniferous, and global temperatures rebounded in the Silurian but not
in the Carboniferous (Sheehan 2001; Stanley 2007). We will examine the
vast—and strange—Carboniferous forests in great detail in the next
chapter.
THE SERPUKHOVIAN CRISIS ON LAND
Paleontologists Ann Raymond and Cheryl Metz have compared the
climatic e ects of the Cenozoic and Late Paleozoic Ice Ages and have
noted that climatic similarities of the two ice ages “include glacial onset in
the Southern Hemisphere and subsequent extension of glaciation to the
Northern Hemisphere; cyclic advance and retreat of continental glaciers
governed by Milankovitch orbital cycles, resulting in cyclothems [cyclic
sedimentary layers]; formation of an ever-wet, ever-warm climate zone
centered on the paleoequator; and strengthening of the pole-to-equator
temperature gradient” (Raymond and Metz 2004, 657). The formation of
an ever-wet, ever-warm equatorial climate zone produced a stable
environment in which “equatorial land-plant diversity increased
dramatically relative to mid-latitude and high-latitude diversity.” On the
other hand, the high-latitude poles became much colder with massive ice
buildup, and the mid-latitude regions centered on the 30°N and 30°S
latitudinal belts developed “strongly expressed arid climatic zones”
(Raymond and Metz 2004, 658).
The C1 glacial pulse at the beginning of the Serpukhovian was followed
by a warming period of about two million years before the much colder C2
glacial pulse began (table 2.2). During this warm interval, a major
extinction in land plants occurred in the mid-latitude regions of
Laurussia, in the present-day Silesian Basin of the Czech Republic. The
Colby College paleobotanist Robert Gastaldo and his colleagues report that
nearly all plant clades su ered extinctions—species of groundcover
plants, climbing vines, understory shrubs, and trees all were a ected,
whether they reproduced by spores or by seeds. This extinction and loss
of land-plant diversity occurred about 327 million years ago (table 2.3),
“several million years prior to the onset of maximum glaciation and sea-
level downdraw at the Mississippian-Pennsylvanian [Serpukhovian-
Bashkirian] boundary,” and appeared to be due to a climatic “shift toward
greater seasonality, with an increased number of dry months” (Gastaldo,
Purkyňová, and Šimůnek 2009, 351).

TABLE 2.3 Biotic events during the Serpukhovian “Big Chill”.

Age Geologic Time Climatic Events Biotic Events

(Ma)
(1) (2) (3)
MOSCOVIAN 308 ICE
309 ICE
310 ICE
311 ICE ICE?
312 ICE ICE?
313 ICE ICE?
314 ICE ICE?
315 ICE ICE?
BASHKIRIAN 316 ICE ICE- ICE?
C4
317 ICE
318 ICE ICE
319 ICE ICE
320 ICE ICE
321 ICE ICE-
C3
322 ICE
323 ICE
SERPUKHOVIAN 324 ICE ICE ICE? ← Marine biodiversity
minimum
325 ICE ICE ICE?
326 ICE ICE- Cold ← First lycophyte rainforests
C2
327 ICE ← Extinction in land plants
 328 ICE
329 ICE ICE
330 ICE ICE-C1 ← Marine biodiversity loss
begins
VISEAN (pars) 331
332

Source: Timescale modi ed from Gradstein et al. (2012).

Note: Column (1) is West and Central Gondwana and column (2) is East Gondwana (Australia), both
in the Southern Hemisphere; column (3) is Siberia (northeastern Russia) in the Northern
Hemisphere.

In great contrast, Gastaldo and colleagues report that the wetland

species were una ected. Here we begin to see a major di erence in
response of marine and land-dwelling species to the great climatic shift
that took place in the Serpukhovian. For marine species the news was
uniformly bad, as discussed in the previous section of the chapter, but
among land species some were driven to extinction while others thrived.
The winners during the Serpukhovian crisis on land were the wetland
species—and some of these winners were very strange indeed.
The strangest were the lycophyte trees. The lycophytes are still alive
today, although most people have never seen one. A fairly large living
lycophyte plant belongs to the stag’s horn clubmoss species Lycopodium
clavatum, which stands about 12 centimeters high—about ve times as tall
a quarter ( g. 2.4).16 In contrast, its ancient relatives during the
Carboniferous formed gigantic lycophyte trees that towered 50 meters
(164 feet) into the sky. Small lycophyte trees—thin, pole-like trees covered
with tiny leaves all over their trunks as well as their short branches at the
top of the trunk—are found as fossils in Earth’s oldest forest, the famous
Middle Devonian Gilboa fossil forest of New York State (Stein et al. 2012)
that we brie y considered in chapter 1. While other plant lineages evolved
adaptive innovations like woody tissue, which allowed the construction of
large trunks, and seed reproduction, which freed the plants from wetland
habitats, the ancient lycophytes remained behind in the wetlands. Their
time would come—and it came in the C2 freezing of the planet (table 2.3).
The rst forests that were actually dominated by species of lycophyte
trees appeared on the Earth about 326 million years ago in Central Asia
(Cleal and Thomas 2005). These strange trees ourished in the Karaganda
Basin in present-day northeastern Kazakhstan, spreading across marshy
land and river deltas formed as the sea level began to fall with the freezing
of large masses of ice at the poles. The lycophyte trees would eventually
form huge rainforests that spanned the globe in the ever-wet, ever-warm
equatorial climate formed by the Late Paleozoic Ice Age. We will examine
these rainforests in great detail in the next chapter.
FIGURE 2.4 The Stag’s horn clubmoss Lycopodium clavatum is a living lycophyte plant that stands
about 12 centimeters (4.7 inches) high.
Source: Modi ed and redrawn from Lecointre and Le Guyader (2006).

THE MYSTERY OF THE SERPUKHOVIAN LAND ANIMALS

How did the Serpukhovian crisis a ect the land animals—was there a
crisis at all? Or, like the wetland species of land plants, did the land
animals thrive during the chilling of the Earth during the Serpukhovian?
We have no de nitive answer to that question, at least not yet.
First, we have no known evidence of a major extinction event in land-
animal species during the Serpukhovian crisis.17 Second, we do know that
major evolutionary innovations occurred in the mid-Carboniferous land-
animal lineages—both tetrapod and arthropod—but it remains unclear
whether these innovations occurred in the Bashkirian Age at the
beginning of the Late Carboniferous or earlier, during the Serpukhovian
crisis period at the end of the Early Carboniferous. For example, the
tetrapods evolved the amniote egg and related morphological changes
that freed them from water in their reproduction. This was a crucial step
in the invasion of land by the vertebrate animals—prior to this innovation,
the tetrapods still laid their eggs in water, much like sh. Many
amphibians, which are not amniotes, still do so. In addition, the insect
lineage of the arthropods evolved the rst wings and capability of ight
among land animals. This also was a crucial step in the invasion of land by
the arthropods—prior to this innovation, insects were constrained to
walking on the ground, their dispersal rates limited by their walking
speed and small size. With the evolution of ight, the insects could take to
the skies and disperse over huge areas of land in a very short time.18
There are two evolutionary lineages of amniote animals: the reptilian
amniotes, the ancestors of living reptiles; and the synapsid amniotes, the
ancestors of living mammals (table 2.4). The evolutionary split between
the reptiles and the synapsids had already occurred in the Bashkirian as
the oldest known (as yet) skeletal fossils of a reptilian amniote, the species
Hylonomus lyelli, and the oldest known skeletal fossils of a synapsid
amniote, the species Protoclepsydrops haplous, are dated to about 317
million years ago in the late Bashkirian Age (table 2.4).19 Thus the amniote
ancestor to the reptilian and synapsid lineages had to be older than the
late Bashkirian.

TABLE 2.4 Amniote tetrapod lineages and species in the Early to Late Carboniferous transition.

TETRAPODA (limbed vertebrates)

– basal tetrapods
– Neotetrapoda
– – BATRACHOMORPHA (ancestors of living amphibians)
– – – basal batrachomorphs (“temnospondyls”)
– – REPTILIOMORPHA (ancestors of amniote tetrapods)
– – – basal reptiliomorphs (“anthracosaurs”)
– – – – Seymouriamorpha
– – – – AMNIOTA (amniote tetrapods, conquerors of the land)
– – – – – basal amniotes
– – – – – – Casineria kiddi? ← Visean
– – – – – SYNAPSIDA (ancestors of living mammals)
– – – – – – basal synapsids
– – – – – – – Protoclepsydrops haplous ← Bashkirian
– – – – – REPTILIA (ancestors of living reptiles)
– – – – – – basal reptiles
– – – – – – – Hylonomus lyelli ← Bashkirian

Source: Phylogenetic classi cation modi ed from Benton (2015).

Note: The geologic age of species is in bold; older paraphyletic tetrapod group names are in
quotation marks; major clades are in capitals.
 The late Bashkirian species Protoclepsydrops haplous is a basal member
of the clade of the synapsid amniotes—but we have trace fossil evidence
that much more highly evolved, derived species of the synapsid amniote
clade also existed in the Bashkirian! This evidence comes from a fossil
trackway in Germany that preserves the footprints of a large animal of an
as yet unknown species, but the trackway (which was given the trace-
fossil species name Dimetropus) is very similar to those produced by more
highly derived ophiacodontid, edaphosaurid, or sphenacodontid
synapsids, all of which were large animals (Voigt and Ganzelewski 2010;
McGhee 2013; we will examine all of these synapsid groups in detail in
chapter 4). If highly derived synapsid species were also present in the
Bashkirian Age, then the ancestors of the synapsid amniote lineage must
be older than the Bashkirian.
The oldest known (as yet) possible candidate for the rst amniote is a
small animal known only from its postcranial skeleton; that is, its body
skeleton is pretty well preserved but the fossil is missing its head. This
species is Casineria kiddi, and its body possesses several traits that are
usually taken to be amniote traits, but it cannot be proved that the animal
is de nitely an amniote because its skull is missing. The Casineria kiddi
fossil is from Cheese Bay, Scotland, and is dated to the mid-Visean Age (see
the detailed discussion of Casineria kiddi in McGhee 2013). Even though it
is impossible to prove that Casineria kiddi is the rst amniote, the
phylogenetic analyses of the University of Bristol vertebrate
paleontologist Michael Benton have led him to predict that the amniotes
evolved about 340 million years ago, in the middle of the Visean (Benton
2015, 123–124 and text box 5.1). If the amniotes evolved 340 million years
ago, in the Visean (table 2.1), why did they diversify and attain population
numbers large enough to appear in the fossil record only in the
Bashkirian, some 17 million years later? That is, did the chilling climatic
shift in the intervening Serpukhovian Age have a negative impact on the
evolution of the earliest amniotes?
A similar puzzle exists concerning the evolution of the rst winged
insect. It was long thought that the oldest known member of the clade of
the winged insects, the Pterygota (table 2.5), was the species Eugeropteron
lunatum from Bashkirian-aged strata in Argentina (Grimaldi and Engel
2005). Eugeropteron lunatum is an odonatopteran, the clade of the
dragon ies (table 2.5), which is almost the basal clade of the winged
insects—but not quite. That honor belongs to the may ies, the
ephemeropterans (table 2.5), so Eugeropteran lunatum cannot have been
the rst ying insect, but it was close to the origin of ight. However, the
University of Exeter biologist Robin Wootton and his colleagues (1998)
noted that Eugeropteron lunatum was surprisingly advanced in its wing
morphology and suggested that there had to be earlier, less highly
adapted, odonatopteran ying species that were the ancestors of
Eugeropteron lunatum.

TABLE 2.5 Flying insect lineages and species in the Early to Late Carboniferous transition.

INSECTA (insects)
– basal insects
– PTERYGOTA (winged insects, conquerors of the land)
– – Ephemeroptera (may ies)
– – Metapterygota
– – – Odonatoptera (dragon ies, damsel ies)
– – – – Eugeropteron lunatum ← Bashkirian
– – – Palaeodictyopterida
– – – – Delitzschala bitterfeldensis ← Serpukhovian
– – – NEOPTERA (folding-wing insects)
– – – – Protoptera
– – – – – Kemperala hagensis ← Serpukhovian
– – – – Polyneoptera
– – – – – Anartioptera
– – – – – – Orthopterida
– – – – – – – Orthoptera
– – – – – – – – Archaeorthoptera
– – – – – – – – – Ampeliptera limburgica ← Serpukhovian

Source: Phylogenetic classi cation modi ed from Grimaldi and Engel (2005) and Lecointre and Le
Guyader (2006).
Note: The geologic age of species is in bold; major clades are in capitals.

Indeed, in 2005, the Czech Charles University zoologist Jakub Prokop

and his colleagues announced the discovery of four winged-insect fossil
species from Serpukhovian-age strata in the Czech Republic (Prokop et al.
2005). These four species are Delitzschala bitterfeldensis (a
palaeodictyopteran), Kemperala hagensis (a protopteran), Ampeliptera
limburgica (an archaeorthopteran), and fragments of another
archaeorthopteran species as yet unidenti ed. These fossils thus provide
de nitive corroboration that the winged insects evolved on Earth in the
Serpukhovian Age of the Early Carboniferous. Or do they? The
protopteran and archaeorthopteran species are highly derived neopteran
insects (table 2.5)—insects that had evolved the capability of folding their
wings, an adaptation not present in the basal clades of the may ies and
dragon ies. Thus the presence of highly derived neopteran species in
Serpukhovian strata at the end of the Early Carboniferous argues for the
evolution of ying insects even earlier in the Early Carboniferous. Yet
even if the winged insects evolved in the Visean Age—along with the rst
amniote, Casineria kiddi, if it is an amniote—they did not become
numerous enough, or have population sizes large enough, to appear in the
abundance seen in the fossil record until the Bashkirian Age at the start of
the Late Carboniferous. Why? What delayed the diversi cation of species
possessing these new adaptive innovations?20 Did the Serpukhovian crisis
negatively a ect the land animals after all?
Regardless of whether the evolution of the crucial adaptive innovations
of the tetrapod amniote egg or the insect wing occurred in the dawn of
the Late Carboniferous or the twilight of the Early Carboniferous, it is
undisputed that a major diversi cation of both tetrapod and insect species
started in the Bashkirian Age. Five new species of basal reptiliomorphs
(table 2.4) appeared in the fossil record in the Bashkirian Age at the
beginning of the Late Carboniferous, seven more new species in the
following Moscovian Age, two more in the Kasimovian Age, and one in the
Gzhelian Age at the close of the Late Carboniferous (see McGhee 2013,
table 7.6). In the synapsid amniotes, ancestors to the mammals, the
Bashkirian species Protoclepsydrops haplous (table 2.4) was joined by three
new species in the Moscovian, four more in the Kasimovian, and three in
the Gzhelian. And, in addition to the Bashkirian species Hylonomus lyelli
(table 2.4), ve new species of reptilian amniotes appeared in the
Moscovian, one in the Kasimovian, and two in the Gzhelian.
The ying insects and other arthropod groups show a similar pattern
of diversi cation beginning in the Bashkirian Age (see McGhee 2013, 237–
249). Fourteen new species of millipedes, centipedes, and arachnids
(spiders, scorpions, and kin) appeared in the Late Carboniferous—
including the gigantic anthropleurid millipedes, the largest arthropods
ever to exist on land. In the ying insects, the gigantic gri en ies (ancient
relatives of modern dragon ies) also evolved in the Late Carboniferous.
These were the largest ying insects ever to inhabit the Earth; we will
consider the evolution of giant Carboniferous arthropods in detail in
chapter 4. The ying insects also experienced a burst of ecological and
morphological innovation in the Late Carboniferous, evolving the ability
to fold their wings and the ability to eat and digest living plants—the
evolution of the rst herbivorous insects.
All of the newly evolved tetrapods and arthropods appeared on an
Earth very di erent from our modern-day world. Viewed from space, the
Carboniferous Earth had brilliant white ice caps at both poles, patches of
brown and reddish arid deserts scattered around the planet in the
temperate latitudes, and a broad green band of lush rainforests in the
equatorial zone. The trees that inhabited those rainforests were
fantastically di erent from our modern rainforest ora, as we will see in
chapter 3.
3 The Late Carboniferous Ice World

The Late Paleozoic Ice Age is the only time in Earth history, other than the
Neogene, when vegetated land masses were subjected to climate uctuations
associated with extended intervals of polar glaciation.
—Gastaldo, Purkyňová, Šimůnek, and Schmitz (2009, 336)

THE STRANGE CARBONIFEROUS TREES

Our modern rainforests are dominated by owering plants (angiosperms)
and conifers (pinophytes), both of which are spermatophyte lignophyte
euphyllophytes—that is, “seed-reproducing woody leafy trees” in the
precise language of modern evolutionists (table 3.1). Like the rainforests
in the cool Carboniferous world, our modern rainforests evolved in a cool
Neogene world, where the South Pole has been glaciated for 12 million
years, since the late Miocene, and the North Pole has been variously
glaciated for three million years, since the late Pliocene (table 1.6).
However, in stark contrast to the trees in our modern rainforests, the
giant trees of the Carboniferous rainforests did not reproduce with seeds
(they were not spermatophytes), did not have woody trunks (they were
not lignophytes), and did not have large leaves (they were not
euphyllophytes). In addition, there were no owers of any kind in the
Carboniferous rainforests—the angiosperms, the owering plants, would
only evolve some 140 million years later in the Early Jurassic.1 To
understand just how strange the giant Carboniferous rainforest trees
were, we need rst to examine the sequence of the evolution of plants on
land (table 3.1).

TABLE 3.1 Land-plant lineages of the Late Paleozoic Ice Age.

EMBRYOPHYTA (land plants)

– Marchantiophyta (liverworts)
– STOMATOPHYTA (stomate plants)
– – Anthocerophyta (hornworts)
– – Hemitracheophyta
– – – Bryophyta sensu stricto (mosses)
– – – POLYSPORANGIOPHYTA (branched-sporophyte plants)
– – – – Horneophyta†
– – – – TRACHAEOPHYTA (vascular plants)
– – – – – Rhyniopsida†
– – – – – Eutracheophyta
– – – – – – LYCOPHYTA (microphyll-leafed plants)
– – – – – – – Zosterophyllopsida†
– – – – – – – Asteroxylales
– – – – – – – – Drepanophycales†
– – – – – – – – Lycopodiales
– – – – – – – – – Protolepidodendrales†
– – – – – – – – – Lycopodiaceae (club mosses)
– – – – – – – – – – Selaginellales (spike mosses)
– – – – – – – – – – – Lepidodendrales† (scale trees)
– – – – – – – – – – – – Lepidodendraceae†: Lepidodendrid scale trees
– – – – – – – – – – – – Sigillariaceae†: Sigillarian scale trees
– – – – – – – – – – – Selaginellaceae
– – – – – – – – – – Isoetales (quillworts)
– – – – – – – – – – – Chaloneriaceae†
– – – – – – – – – – – Isoetaceae
– – – – – – EUPHYLLOPHYTA (megaphyll-leafed plants)
– – – – – – – Moniliformopses
– – – – – – – – Cladoxylopsida†
– – – – – – – – – Pseudosporochnaceae†: Wattieza (Eospermatopteris) erianus trees
– – – – – – – – Equisetophyta (horsetails)
– – – – – – – – – Calamitaceae†: Calamitean horsetail trees
– – – – – – – – – Sphenophyllaceae†
– – – – – – – – – Equisetaceae
– – – – – – – – Filicophyta (true ferns)
– – – – – – – – – Marattiales: Marattialean tree ferns
– – – – – – – LIGNOPHYTA (woody plants)
– – – – – – – – Aneurophytales†
– – – – – – – – Archaeopteridales†
– – – – – – – – – Archaeopteridaceae†: Archaeopterid spore trees
– – – – – – – – SPERMATOPHYTA (seed plants)
– – – – – – – – – Lyginopteridales†
– – – – – – – – – – Lyginopteridaceae†: Lyginopterid seed-fern trees
– – – – – – – – – Medullosales†
– – – – – – – – – – Medullosaceae†: Medullosan seed-fern trees
– – – – – – – – – Gigantopteridales†
– – – – – – – – – – Gigantopteridaceae†: Gigantopterid seed-fern trees
– – – – – – – – – Core seed plants
– – – – – – – – – – Ginkgophyta (ginkgos)
– – – – – – – – – – Pinophyta (conifers)
– – – – – – – – – – – Cordaitales†
– – – – – – – – – – – – Cordaitaceae†: Cordaitean conifer trees
– – – – – – – – – – Cycadophyta (cycads)
– – – – – – – – – – Gnetophyta (gnetophytes)
– – – – – – – – – – [ANGIOSPERMAE ( owering plants) not present in Paleozoic!]

Source: Phylogenetic classi cation modi ed from Kenrick and Crane (1997a, 1997b), Donoghue
(2005), and Lecointre and Le Guyader (2006).
Note: Important tree groups are in bold italics; extinct lineages are marked with a dagger (†); major
clades are in capitals. The classi cation given in this table is a phylogenetic one, where the taxa
listed are monophyletic clades. In the older literature the reader will encounter non-phylogenetic
classi cations of plants containing groupings of plants that are now recognized as paraphyletic. To
help the reader translate the older plant classi cations, here is a helpful list of the major
paraphyletic groups that have been used in the past: “Pteridophytes” = non-spermatophyte
tracheophytes, “Progymnosperms” = non-spermatophyte lignophytes, “Gymnosperms” = non-
angiosperm spermatophytes, “Pteridosperms” = non-core seed plants (“seed ferns”); for
discussions, see Niklas (1997) and Donoghue (2005).

Land plants, the Embryophyta, evolved from freshwater lamentous

green algae, and it appears that the rst tiny land plants emerged from
the aquatic world in the Middle Ordovician (Lecointre and Le Guyader
2006; McGhee 2013). The most primitive of the land plants are the
marchantiophytes, the liverworts (table 3.1). The oldest known spores and
cuticular fragments that are believed to be from liverworts are found in
strata dated to the Dapingian Age of the Middle Ordovician (see table 1.4
for the Ordovician Age divisions of the geologic timescale), around 470
million years ago (Rubinstein et al. 2010; Wellman 2010). About 15 million
years later, de nite proof that the liverworts had evolved comes from
fossils of Katian Age of the Late Ordovician, about 453 million years ago
(Wellman et al. 2003). These fossils include not just isolated spores but also
tissue fragments of the sporangia, the structures that held the spores
together in life. Some of these fossil sporangia hold as many as 7,450 spore
tetrads together, and the shape of the spherical sporangia fragments
suggest that they could have held as many as 95,000 spore tetrads in the
living plant. The ability to produce large numbers of spores is considered
to be an adaptation to living in harsh terrestrial environments and, apart
from the terrestrial strata in which they are found, is further evidence
that these fossils were from land plants. Finally, the microscopic
ultrastructure of the walls of the spores reveals parallel-arranged lamellae
of a type that is found only in the spores of living liverworts, providing
further evidence that the plants that produced them were indeed
liverworts, the most primitive of the land plants (Wellman et al. 2003;
Wellman 2010). Today, these simple liverwort land plants resemble lichen,
the terrestrial algal-fungal symbionts that are not really plants, and like
lichen, they can encrust rocks or tree trunks.
The liverworts have no roots, no vascular system, no true stomata
(these plant structures will be discussed shortly), and are con ned to
humid environments. The next step in the evolution of land plants was
the evolution of the Stomatophyta (table 3.1), plants so named because
they possess stomata to regulate gas exchange in dry air. Stomata are
openings in the wall tissues of the plant that can be opened and closed by
two guard cells; they allow the plant to regulate gas and water vapor
exchange between it and the atmosphere around it. It is critical to a land
plant (as opposed to a water plant) to prevent as much water loss as
possible to the surrounding dry atmosphere of the land regions, and the
stomata are key anti-dehydration structures in this e ort. The most
primitive or least derived members of the stomatophyte clade are the
anthocerophytes, the hornworts, and fossil Stomatophytes hornwort spores
demonstrate that the stomatophytes had evolved by the Katian Age of the
Late Ordovician, about 453 million years ago (Lecointre and Le Guyader
2006).
The bryophytes, the mosses, are more advanced stomatophytes of the
hemitracheophyte clade (table 3.1). Both the hornworts and the mosses
have upright sporophytes; that is, the sporangium is held up in the air on
a thin vertical stem. Upright sporophytes are another adaptation to life in
dry air, but it is debated whether this adaptation evolved once in the
hornworts and was simply inherited by the mosses, or whether the
mosses independently evolved upright sporophytes and thus this
adaptation is convergent in the two groups (Lecointre and Le Guyader
2006; for an extensive discussion of convergent evolution in plants, see
McGhee 2011). The next step in the evolution of land plants was the
evolution of plants with sporophytes that were branched—that is, vertical
stems that supported more than one sporangium. These plants are the
Polysporangiophyta (table 3.1), which evolved in the Early Silurian. The
most primitive polysporangiophytes were the now extinct horneophytes
and include the fossil species Horneophyton lignieri and Aglaophyton major
from the Early Devonian Emsian Age (for more information about these
fossil species, see McGhee 2013).
Following the evolution of plants with erect, multibranched stems, the
next step in the evolution of land plants was the evolution of the vascular
plants—that is, plants possessing water-conducting tubes that transport
water up the stem against the force of gravity. These tubes are called
tracheids, hence the name Tracheophyta (table 3.1) for these plants. The
tracheophytes also evolved in the Early Silurian, and the most primitive
members of the tracheophyte clade are another extinct group, the
rhyniopsids, such as the fossil species Rhynia gwynne-vaughanii from the
Early Devonian Pragian Age (for more information about these fossil
species, see McGhee 2013).
The basal, lea ess tracheophytes gave rise to the more derived
Eutracheophyta, the rst land plants to evolve leaf-like structures. The
eutracheophytes are divided into two major clades: the microphyll-leafed
vascular plants, the Lycophyta, which have numerous tiny leaves; and the
megaphyll-leafed vascular plants, the Euphyllophyta, which have less
numerous big leaves (table 3.1). The clade of the tiny-leafed lycophyte
plants includes modern-day small club mosses, spike mosses, and
quillworts. It also includes the gigantic Carboniferous trees of the Late
Paleozoic Ice Age. The evolutionary split between the lycophytes and the
euphyllophytes appears to have occurred by the Middle Silurian
Homerian Age, as fossils of the basal lycophyte species Cooksonia pertoni
and Baragwanathia longifolia are found in strata of that age (for more
information about these fossil species, see McGhee 2013).
The clade of the large-leafed plants, the euphyllophytes, includes most
of the land plants that are familiar to us in our modern world. This large
clade is itself divided into two major subclades: the Moniliformopses and
the Lignophyta (table 3.1). The clade of the moniliformopses includes the
cladoxylopsids, an extinct group of fernlike plants; the equisetophytes, the
modern-day horsetails; and the licophytes, the modern-day true ferns.
All of these plants still reproduce by spores and have very little, if any,
woody tissue.
The other clade of the euphyllophytes, the Lignophyta, are known as
the woody plants as they contain substantial amounts of wood tissue. The
lignophytes also evolved “bifacial cambium,” an advanced trait that will be
discussed in more detail below. The most primitive or least derived
members of the lignophyte clade, the Aneurophytales and the
Archaeopteridales (table 3.1), still reproduced by spores, and both groups
of plants are now extinct. The more derived lignophytes, the
Spermatophyta, evolved the rst seeds in the Late Devonian Famennian
Age. The evolution of seed-reproducing plants was a major innovation in
the geologic history of plants; the implications of this evolutionary
innovation in reproductive type will be discussed in more detail below.
The last major innovation in plant evolution was the evolution of the
owering plants, the Angiospermae (table 3.1), but that did not occur until
the early Jurassic in the Mesozoic Era. Life in the ancient Paleozoic world
saw no owers.
Now let us consider the evolution of trees. Trees are a characteristic
adaptation of plants to life on land (McGhee 2011, 2013). Simple plant
growth in a two-dimensional plane soon leads to crowding and
overgrowth, with one plant shading out another in the competition for
light from the sun, the energy source for plant survival. When crowding
occurs in the two-dimensional plane of the land surface, the solution is to
move into the third dimension above the land surface, to evolve plants
with single vertical stems, then branched stems, and then branched stems
with leaves—hence the evolution of the Stomatophyta, the
Polysporangiophyta, and the Euphyllophyta (table 3.1).
Even now, however, crowding will again become a problem when these
types of plants cover the surface of the land. And, just as in human cities
when crowding occurs in a region of low buildings, the solution is to move
even higher into the third dimension above the land surface—to construct
towering skyscraper buildings or, in the case of plants, to evolve trees. The
force of gravity has to be reckoned with, and new support structures had
to be evolved in order to create a massive central structure, a tree trunk,
that rises vertically from the land surface. At some distance above the
ground, branches extend out from the tree trunk in order to capture as
much sunlight as possible for the survival of the tree and, in order to
ensure the survival of the species, to facilitate fertilization and dispersal
of the tree’s o spring. All of these structures are heavy, and the tree trunk
must be strong enough to support them without breaking or bending.
Given these di culties, one might think that only one advanced clade of
plants would successfully evolve a tree. Thus it is astonishing that not one,
but no less than nine separate phylogenetic lineages of plants
independently, convergently, evolved the tree form, each with its own
di erent solution to the problem of growing a su ciently strong central
trunk structure (Niklas 1997; McGhee 2011). The lycophytes, cladoxylopsid
moniliformopses, and lignophytes all separately evolved the tree form in
the Middle Devonian Givetian Age. Then the equisetophyte and
 licophyte moniliformopses independently evolved the tree form in the
Late Devonian and Early Carboniferous. The other four convergently
evolved tree forms occurred later in geologic time, after the end of the
Paleozoic Era (twice more in the licophytes, and twice more in the
lignophytes) (Niklas 1997; McGhee 2011).
The oldest tree fossils yet known are of the extinct cladoxylopsid
Wattieza (Eospermatopteris) erianus trees (table 3.1), found in the famous
Gilboa fossil forest in New York State (Stein et al. 2007). These ancient
trees stood eight meters (26 feet) tall and looked like thin, tall poles
topped with a shaving brush ( g. 3.1)! They had no horizontal branches
and no woody tissue—in contrast, the trunk of the tree was probably
hollow like a reed. More signi cantly, down in the understory of the
Gilboa forest were other smaller, thin, pole-like trees covered with
numerous tiny leaves—the rst lycophyte trees, ancestors of the giant
Carboniferous trees (Stein et al. 2012).
FIGURE 3.1 The extinct cladoxylopsid-lineage fernlike tree Wattieza (Eospermatopteris) erianus: (a)
the reassembled tree fossils; (b) a reconstruction of the living tree. The living tree was about eight
to ten meters (26 to 33 feet) tall.

Source: From Macmillan Publishers Ltd: Nature (Stein et al., 2007), copyright © 2007. Reprinted with
permission.

The Gilboa fossil forest is of Givetian Age in the Middle Devonian, and
in the later Givetian, the woody lignophytes also convergently evolved a
tree form—the archaeopteridalean Archaeopteris (table 3.1). An
Archaeopteris tree was much more like the trees that are familiar to us in
the modern world ( g. 3.2) in that it had a trunk with a core of heartwood,
and it had numerous horizontal branches with leaves. However, it was
unlike any modern lignophyte trees in that it still reproduced with spores,
not seeds. Unlike the short-lived Wattieza-type cladoxylopsid trees, the
Archaeopteris spore trees could tower 30 meters (100 feet) into the sky.
Archaeopteris trees became very numerous in the Late Devonian, and by
the middle Frasnian Age vast areas of the Earth were covered with
Archaeopteris forests. Then, mysteriously, all of the Archaeopteris trees died
out in the end-Devonian biodiversity crisis (McGhee 2013).

FIGURE 3.2 The extinct archaeopterid-lineage spore tree Archaeopteris was about 30 meters (100
feet) tall, here compared to a four-meter- (13-foot-) tall male African elephant.
Source: Illustration by Mary Persis Williams.

In the Early Carboniferous Tournaisian Gap (table 2.1), following the

End-Famennian Bottleneck, the tallest trees on the Earth were only two
meters (6.6 feet) high (DiMichele and Hook 1992; McGhee 2013). The
lignophytes had not lost the ability to produce trees, however, and in the
later Visean Age of the Early Carboniferous, the rst of the lignophyte
seed-fern trees evolved (table 3.1), such as the 40-meter-tall (131-foot-tall)
Pitus primaeva, discussed in chapter 2. The lycophytes also would begin to
produce large trees—the gigantic tropical scale trees—starting in the
Serpukhovian Age of the Early Carboniferous (table 2.3).
Having summarized the evolutionary events that led to the origin of
trees on Earth, we now turn speci cally to the types of trees that were
present in the great rainforests of the Carboniferous. Five types of trees
were dominant: the lycophyte scale trees, the equisetophyte horsetail
trees, the licophyte tree ferns, the medullosan seed-fern trees, and the
cordaitean pinophyte trees (table 3.1) (Pfe erkorn et al. 2008). In that
order, let us examine each of these ve types of tree in detail. Only the last
two tree types were seed plants with woody core tissue like our modern
trees—and the rst two were unlike any trees alive on Earth today.

The Lycophyte Scale Trees

The strangest trees in Carboniferous rainforests—from our modern

perspective—were the towering, pole-like lepidodendrid lycophytes ( g.
3.3). First, their gigantic size is di cult to understand. The lycophytes are
still alive today, but they are quite small plants. A fairly large modern
lycophyte, the stag’s horn club moss Lycopodium clavatum, stands only 12
centimeters ( ve inches) high ( g. 2.4). In contrast, an average
Lepidodendron tree could tower 50 meters (160 feet) into the sky (Cleal and
Thomas 2005). How could a lycophyte plant in the Carboniferous grow to
be over 400 times taller than any living lycophyte on the Earth today?

FIGURE 3.3 The lycophyte scale tree Lepidodendron was about 50 meters (160 feet) tall, here
compared to a four-meter- (13-foot-) tall male African elephant.

Source: Illustration by Mary Persis Williams.

Second, the lycophyte trees had determinate growth, a mode of plant

growth that is unusual in plants in our modern world. Consider, for
example, the mode of growth of a modern dandelion weed, Taraxacum
o cinale, in your lawn. For many days the dandelion plant has bladelike
leaves that are low to the ground, and eventually it develops a bright
yellow ower that is also located low to the ground. Then, suddenly, the
plant shoots up a long stem as much as 15 centimeters (six inches) high,
crowned with a fuzzy white globe of tufted seeds. This is the seed-
dispersal phase in the reproductive cycle of the dandelion in which it
elevates its seeds above ground level in order to more e ciently expose
the seeds to wind currents that will carry the tufted seeds considerable
distances away from the parent plant.
The mode of growth of the Lepidodendron tree was even stranger than
that of the modern dandelion: after it reproduced, it died. Like a young
dandelion plant, the young Lepidodendron tree grew low to the ground—
but for many years, not just for a few weeks. During this time it resembled
a low stump, but it was green rather than bark-brown as it was covered
with the tiny scalelike leaves of the lycophytes. Below ground, however, it
was constantly expanding its peculiar rootlike stigmarian system: in
Lepidodendron trees even the stigmarian rootlets could photosynthesize,
and they often protruded above ground in order to catch sunlight.
When the Lepidodendron tree reached its reproductive phase, similar to
the development of a stem by a dandelion plant, it began to grow a pole-
like trunk that stretched into the sky ( g. 3.3). This trunk could reach
heights of 50 meters (160 feet) but was quite slender, only about one meter
(3.3 feet) in diameter—in essence a thin, towering pole covered with tiny,
green, scalelike leaves. As the trunk grew taller and taller, it would begin
to shed the leaves in its lower reaches, leaving only the leaf-cushion bases
while thickening the trunk. Fossil trunks of these trees are covered with
leaf-cushion bases that look like the pattern of scales seen on the body of a
snake, hence these trees are often given the common name of “scale
trees.” But the scalelike leaf-cushion bases could still photosynthesize—
even without leaves—and hence produce food locally on the trunk.
Only when the trunk reached its maximum height would it then begin
to produce branches in pairs, drooping on either side of the central trunk,
and to develop cones on the branches. Like the upper trunk, these
branches were covered with sleeves of leaves, but they were smaller than
those found on the trunk. The Lepidodendron tree would then rapidly
produce spores in its cones, which would be dispersed by winds at a
height of 50 meters above ground level. The tree’s microspores were small
and lightweight; thus they could easily be carried by the wind and did not
need structures like the tufts of the much larger dandelion seed, which
increase the surface area available to catch wind current for support in
transporting the seed through the air. After its spores were all shed, the
Lepidodendron tree would die. The towering dead trunks of the
Lepidodendron trees would remain standing for a while before they began
to rot and then, one by one, the great trunks would fall to the swamp
waters below.
Astonishingly, the trunks of the gigantic lycophyte trees were not
supported by woody tissue like our modern giant trees, such as the giant
redwood Sequoiadendron giganteum, which can stand 90 meters (300 feet)
high. Most modern trees belong to the clade of the lignophytes, the woody
plants (table 3.1). Woody trees possess a bifacial cambium, with secondary
xylem tissue toward the center of the trunk and secondary phloem tissue
toward the outside (Donoghue 2005). The secondary phloem of a modern
woody tree transports food produced by the photosynthesizing leaves in
the crown of the tree down to the trunk and underground root regions,
and the woody tissue supports the mass of the trunk, branches, and
photosynthetic needles or leaves against the downward pull of gravity.
 In contrast, the trunks of the lycophyte trees were constructed in a
most peculiar manner. They had only a unifacial cambium with no
secondary phloem. Without secondary phloem, the tree could not
transport food and nutrients for long distances; thus the tree developed
photosynthesizing structures—tiny leaves and even the bases holding the
leaves—all over its branches and trunk to provide living tissues with food
that was locally produced. Even the peculiar stigmarian rootlike tissues of
the tree could photosynthesize food. Thus, unlike modern woody trees
with food production con ned to the crown of the tree with its
photosynthesizing leaves, the lycophyte tree produced food over just
about its entire surface (Donoghue 2005).
The unusual growth mode of the trunk of the lycophyte tree was
probably due to the absence of secondary phloem and lack of su cient
woody tissue to support the trunk against the force of gravity. The weight
of the tree was held up by an outer layer, called the periderm, of cortex
and leaf bases—in essence, the external bark of the tree, rather than an
internal wood core, was its structural support. The trunk of the tree was
produced only when it was needed, in the reproductive phase, when the
spores were developed high above the surface of the Earth to be dispersed
by wind. Once the spores of the tree were dispersed, the trunk was no
longer needed, and it died and fell to the Earth. In contrast, modern-day
woody trees go through decades of reproductive cycles before dying.

The Equisetophyte Horsetail Trees

The second strange type of trees present in the Carboniferous rainforests

that do not exist today were the equisetophyte trees ( g. 3.4). Like the
lycophytes, equisetophyte plants still exist today, but most are quite small.
The modern eld horsetail Equisetum arvense stands 25 centimeters (ten
inches) high and is abundant in many moist environments today, such as
roadside ditches (Lecointre and Le Guyader 2006). In contrast,
Carboniferous horsetail trees were gigantic—some Calamites trees stood 20
meters (65 feet) tall—80 times taller than our modern small horsetail
rushes ( g. 3.4) (Taylor and Taylor 1993).

FIGURE 3.4 The equisetophyte horsetail tree Calamites was about 20 meters (65 feet) in height,
here compared to a four-meter- (13-foot-) tall male African elephant.

Source: Illustration by Mary Persis Williams.

 Although the calamitean horsetail trees belonged to the clade of the
euphyllophytes (table 3.1), they still possessed microphylls—small leaves
that were whorled about jointed branches attached to trunks that were
also jointed. These jointed trunks grew upwards from large rhizomes
buried in the soil, rather than from rootlike stigmarian systems as in the
lycophyte trees ( g. 3.4). Like the modern-day eld horsetail, these
underground rhizomes had rootlet clusters at the nodes where the erect
stems, or trunks as in the case of Calamites, were produced.2
Like the lycophyte trees, the ancient equisetophyte trees had a
unifacial cambium and no secondary phloem; also like the peculiar
lycophyte trees, they possessed the unusual trait of determinate growth
(Donoghue 2005). Unlike the lycophyte trees, the segmented trunks of the
equisetophyte trees were structurally supported by internal wedges of
wood (Niklas 1997). Although the equisetophyte trees possessed more
woody support than the lycophyte trees, the inward-pointing wedges of
wood that were arranged around the circular interior of the trunk of the
equisetophyte trees were very di erent from the massive amount of
heartwood and sapwood found in the core of the trunk of a lignophyte
tree.

The Filicophyte Fern Trees

The third major type of tree found in Carboniferous forests, the tree ferns,
are not so strange to us as the extinct lycophyte and equisetophyte trees
because some tree ferns, such as the Australian tree fern Cyathea cooperi,
still exist today. However, most living ferns are small herbaceous plants, or
more rarely shrub sized. They are easily recognizable by their
characteristic large fronds of leaves that unfurl from ddlehead-shaped
spirals and the clusters of dark spores that periodically develop on the
undersides of their leaves. In contrast, the Carboniferous tree ferns were
large: the marattialean tree fern Psaronius ( g. 3.5) stood ten meters (33
feet) high (Taylor and Taylor 1993).

FIGURE 3.5 The marattialean true tree fern Psaronius was about ten meters (33 feet) tall, here
compared to a four-meter- (13-foot-) tall male African elephant.

Source: Illustration by Mary Persis Williams.

The true ferns, the licophytes (table 3.1), reproduce by spores and are
not lignophyte plants. Thus the living tree ferns still exhibit those traits
that they share with their extinct Carboniferous relatives, the
equisetophyte trees. Unlike the equisetophytes, which evolved only a
single type of tree, the licophytes have convergently evolved tree forms
no less than three separate times: Psaronius-type tree ferns in the
Carboniferous, Tempskya -type tree ferns in the Cretaceous, and the tree
ferns that survive today (McGhee 2011). Each group invented a di erent
type of trunk in their evolution: the Carboniferous tree ferns had trunks
supported by an outer mantle of adventitious roots; the Cretaceous tree
ferns had trunks consisting of interwoven stems bound together by
adventitious roots; and living tree ferns have a lower columnar base of
adventitious roots and an upper trunk supported by an outer layer of
cortex and external layer of leaf bases (Niklas 1997). Thus all three types
of tree ferns construct their trunks in a very di erent manner than either
the lycophyte or equisetophyte trees: they modify their root-growth
systems for usage as support structures above ground.

The Medullosan Seed-Fern Trees

Only two of the ve major tree types found in the great Carboniferous
rainforests were spermatophyte lignophytes—seed plants with woody
core tissue like our modern trees. These were the seed-fern trees and the
pinophyte conifer trees (table 3.1). The evolution of the lignophyte plants
in the Middle Devonian introduced a typical tree-trunk construction—
having a structural support core of heartwood in the center of the trunk—
that is very familiar to us in the modern world, but it could be argued that
this construction is less mechanically e cient than the tree-trunk
construction of the ancient non-lignophyte lycophyte, equisetophyte, and
licophyte trees in the Carboniferous forests. Karl Niklas, a paleobotanist
at Cornell University, notes: “Engineers have long known that the best
location for mechanically supportive materials is just beneath the surface
of a vertical column, where mechanical bending and torsional forces reach
their maximum intensities. This strategy is particularly important when
the quantity of the sti est building material is limited, perhaps for
economy in design. It is no coincidence that the sti est tissues in a variety
of plants tend to develop just beneath or very near the external surface of
vertical stems … the sti lycopod periderm produced by the
lepidodendrids … and the pertinacious mantle of adventitious roots
around the stems of Psaronius are but a few examples of this mechanical
strategy” (Niklas 1997, 331–333).
On the other hand, the evolution of bifacial cambium—one in which
both secondary xylem and phloem tissues are produced—in the
lignophytes was an adaptive innovation (Donoghue 2005). The secondary
phloem in lignophytes makes possible the long-distance transport of food
from the leaves high up in the crown of the tree down to the roots located
far underground at the other end of the tree. The lycophyte and
equisetophyte trees had a unifacial cambium and did not possess
secondary phloem (Donoghue 2005). The growth consequences of
possessing only a unifacial cambium for the lycophyte and equisetophyte
trees have been noted previously.
The evolution of the seed plants, the spermatophyte lignophytes, in
the Late Devonian Famennian Age was a major adaptive innovation in the
geological history of plants. The key adaptation of the seed plants is their
freedom from needing water in reproduction, unlike the spore-
reproducing plants, and the spermatophytes could now colonize the dry
highlands and mountains that were out of reach for non-spermatophyte
plants. The evolution of the seed in plants was the ecological equivalent of
the evolution of the amniote egg in vertebrates (Niklas 1997). a key
adaptation that also freed the synapsid and reptilian amniotes (ancestors
of modern-day mammals and reptiles, table 2.4) from needing water in
their reproduction. These two groups, the seed plants and the amniote
vertebrates, would become the victorious plant and vertebrate conquerors
of the terrestrial realm of the Earth in the Carboniferous (McGhee 2013).
The ancient seed-fern trees were not true ferns (table 3.1), the spore-
reproducing licophytes ( g. 3.5), but rather were seed-reproducing
plants that possessed foliage that looked very similar to that of a true fern.
Seed ferns were also not a monophyletic clade in that three separate
clades of spermatophyte lignophytes are often called “seed ferns”—the
lyginopteridales, the medullosales, and the gigantopteridales (table 3.1)—
all three of them extinct. The lyginopterids produced some fairly large
seed-fern trees in the Early Carboniferous, such as Pitus, but the most
important seed-fern trees in the Late Carboniferous were the
medullosans, trees such as Medullosa and Sutcli a. Medullosa was a small
seed-fern tree that stood only about ve meters (16 feet) tall, about half as
tall as the true fern tree Psaronius ( g. 3.5). Smaller seed ferns like
Neuralethopteris and Alethopteris also contributed to ground cover in the
Carboniferous rainforests, as will be discussed shortly.

The Cordaitean Conifer Trees

The only other modern-type lignophyte trees found in the great

Carboniferous rainforests were the pinophyte conifer trees (table 3.1). It is
in the clade of the pinophytes that we nally nd modern trees that are
even more gigantic than the Carboniferous giants. The giant redwood
trees of western North America, such as Sequoiadendron giganteum, can
reach a height of 90 meters (300 feet)—almost twice the 50-meter (160-
foot) height of the peculiar Carboniferous Lepidodendron scale trees.
The important conifer trees in the Late Carboniferous were the
cordaitean pinophytes, trees such as Cordaites and Walchia. Cordaites was a
small conifer tree that stood only about ve meters (16 feet) tall, about
half as tall as Psaronis ( g. 3.5).3 Some cordaitean trees had a tree-on-stilts
appearance much like a modern mangrove tree, such as Rhizophora mangle,
in that the trunk of the tree was elevated above ground by an extensive
root system, the upper part of which was subaerial. Also like a modern
mangrove tree, Cordaites trees are thought to have lived in wet
environments on the margins of swamps and in coastal areas.

The Peculiar Carboniferous Rainforests

In summary, walking or boating in most of the great Carboniferous

rainforests of the ancient Earth would have been quite a di erent
experience than walking or boating in a modern rainforest. In a modern
rainforest, the forest oor is in perpetual gloom as sunlight is shaded out
by the canopy of leaves of the giant woody seed trees. A distinct canopy
ecosystem exists up in the air of the modern rainforest, an ecosystem
inhabited by vertebrate and arthropod carnivores and herbivores that live
their lives in the sunlit canopy of leaves high above the Earth’s surface.
In contrast, in most of the Carboniferous rainforests the forest oor
was illuminated, not shrouded in gloom, as the towering lycophyte and
equisetophyte spore trees did not have large leaves that would shade out
the forest oor below ( g. 3.6). The lycophytes also developed their
canopies of branches only in the reproductive phase of their life cycle, so
that any shading e ect produced by their canopies of branches existed
only for a fraction of the lifetime of the tree. Thus, much of the time, the
rainforest may have had the peculiar appearance of being lled with many
tall, pod-topped green poles without branches.
FIGURE 3.6 In situ reconstruction of a lycophyte tropical rainforest of Bashkirian Age, early Late
Carboniferous, in present-day Alabama in North America. Characteristic ora are labeled, and the
vertical scale in meters (right margin of the gure) gives the height of the plants in the foreground;
see text for discussion.

Source: From Gastaldo, R. A., I. M. Stevanović-Walls, W. N. Ware, and S. F. Greb, “Community

Heterogeneity of Early Pennsylvanian Peat Mires,” Geology vol. 32, pp. 693–696, copyright © 2004
Geological Society of America. Reprinted with permission.

Figure 3.6 illustrates a painstaking, labor-intensive reconstruction of a

Late Carboniferous, Bashkirian Age, North American rainforest in
Alabama (Gastaldo et al. 2004). The Colby College paleobotanist Robert
Gastaldo and his colleagues descended into old coal mines in Alabama and
took a careful census of in situ trees and plants embedded in the roofs of
the mines, located immediately above the mined-out coal layer. This
sampling procedure not only allowed the paleobotanists to census the
species composition of the rainforest plants that existed at a single instant
in time, but it also allowed them to spatially map the geographic
distribution of the plants within the rainforest.
Rainforest reconstructions from three of their 17 eld sites are shown
in gure 3.6, labeled “9/2”, “9/3”, and “9/23”. Site 9/3, in the middle of
the gure, shows the rainforest region with the least number of trees in
canopy stage and greatest number of juvenile pole-like trees (94 percent).
Taller Sigillaria lycophyte scale trees and lower Psaronis Marattialean tree
ferns (Pecopteris leaf fossils) and Sphenopteris seed-fern trees are labeled in
this region of the gure. Site 9/2, on the left in the gure, shows the
rainforest with taller, mature Lepidophloios scale trees that have gone into
canopy stage, with branches at the top of the pole-like trunk that itself is
covered with a sleeve of leaves. Below the Lepidophloios and Sigillaria
lycophyte trees are shown lower Calamites equisetophyte horsetail trees
and a ground cover of Sphenopteris pottsvillea seed ferns. On the right in
gure 3.6, at site 9/23, the rainforest is shown with a 53 percent canopy
cover with still taller, mature Lepidodendron lycophyte trees—note again
the fuzzy appearance of the trunk of the tree, which is covered in a sleeve
of leaves. Below the canopy of Lepidodendron and Lepidophloios lycophyte
trees—a canopy that covers only half of the sky in the rainforest—vinelike
Lyginopteris seed ferns and Sphenophyllum equisetophytes are shown on
the left, and ground cover Neuralethopteris and Alethopteris seed ferns are
shown on the right.
This Alabama rainforest was overwhelmingly dominated by juvenile
trees, with 95 percent juveniles in the youngest forests and 89 percent
juveniles in the eld sites with the greatest number of mature lycophyte
trees. The youthfulness of the forest may have been a function of its
origin: the forest formed when the mire basin subsided, water and
sediment ooded in, and the new trees took root directly on top of the
older, buried forest debris that would become the massive coal bed that
was mined out by humans some 320 million years later. The lycophyte
trees in this Alabama rainforest are also smaller than those in many other
lycophyte-dominated forests, as the Lepidodendron scale trees are only 23
meters (75 feet) high—in other rainforests, these lycophyte trees could be
twice that height. Unlike in a modern rainforest, there was no vertical
strati cation of speci c lycophyte and equisetophyte tree species adapted
to capturing light at di erent height zones in the ancient Carboniferous
rainforests (Cleal 2010). Given their determinate mode of growth, the
di ering heights of the lycophyte and equisetophyte trees within the
forest re ected the di ering ages of the trees instead ( g. 3.6).
It has generally been argued that geographic variation in the
composition of the lycophyte and equisetophyte trees species within the
Carboniferous rainforests were a function of the areal distribution of
di erent soil types and conditions such as waterlogged regions, muddy-
margin regions along riverbanks, better-drained regions between rivers,
dry upland regions, and so on (Cleal 2010). In particular, it has been
argued that Lepidophloios trees preferred waterlogged regions,
Lepidodendron trees preferred better-drained regions, and Sigillaria trees
preferred drier upland regions. In contrast, the Alabama rainforest
contained species of all three of these lycophyte tree genera coexisting in
the same area ( g. 3.6). Gastaldo and colleagues suggest that the
previously argued soil-moisture gradient for lycophyte tree species could
be a sampling artifact, as previous reconstructions of Carboniferous
rainforests were made from coal-ball assemblages—assemblages that are
time averaged, with oras that “represent the resistant biomass
contribution from several plant generations to the peat,” so that “resistant
plant parts that accumulate and are buried may represent a century or
more of biomass concentration in any one coal ball” (Gastaldo et al. 2004,
693). Alternatively, Gastaldo and colleagues suggest that the absence of a
soil-moisture gradient in lycophyte tree distributions in the Alabama
rainforest may be a function of the early geologic age of the forest in the
Bashkirian, at the dawn of the Late Carboniferous. They suggest that a
soil-moisture preference by the lycophyte trees may have evolved later, a
result of “increasing habitat specialization as peat mires became more
extensive in the Late Pennsylvanian” (Gastaldo et al. 2004, 696).

THE SPREAD OF THE GREAT RAINFORESTS

The lycophytes rst evolved tree forms in the Middle Devonian Givetian
Age, but it was not until the Early Carboniferous Serpukhovian Age that
the rst of the lycophyte-dominated tropical rainforests formed (table
2.3). The most extensive of these early tropical rainforests formed in what
is now central Asia and northwest China, with smaller rainforests
scattered in northern Europe and in North America. Best estimates of the
original geographic extent of these rainforests indicate that they probably
covered almost a half-million square kilometers (almost 200,000 square
miles) of the tropical region of the Early Carboniferous Earth (table 3.2). In
the Late Carboniferous Bashkirian and Moscovian Ages, the lycophyte-
dominated rainforests progressively expanded in the Earth’s tropical
region until they reached a maximum geographic extent of almost 2.5
million square kilometers (almost a million square miles; table 3.2) in the
late Moscovian ( g. 3.7) (Cleal and Thomas 2005).

TABLE 3.2 Climatic events, area of rainforest coverage, and biological events during the
Carboniferous phase of the Late Paleozoic Ice Age.

Age Geologic Time Climatic Events Areal Extent of Biological Events

(Ma) Rainforests
(1) (2) (3)
(Permian) 298 ICE ICE- 1,255,000 km2 ← Recovery in the
P1 rainforests
GZHELIAN 299 1,087,000 km2
300 1,095,000 km2
301 1,110,000 km2
 302 1,111,000 km2
303 1,120,000 km2
KASIMOVIAN 304 1,131,000 km2
305 1,131,000 km2
306 1,131,000 km2
307 1,131,000 km2 ← Extinction in the
rainforests
MOSCOVIAN 308 ICE 2,395,000 km2
309 ICE 2,170,000 km2
310 ICE 1,945,000 km2
311 ICE ICE? 1,721,000 km2
312 ICE ICE? 1,840,000 km2
313 ICE ICE? 1,822,000 km2
314 ICE ICE? 1,804,000 km2
315 ICE ICE? 1,786,000 km2
BASHKIRIAN 316 ICE ICE- ICE? 1,456,000 km2
C4
317 ICE 1,258,000 km2
318 ICE ICE 1,060,000 km2
319 ICE ICE 826,000 km2
320 ICE ICE 664,000 km2
321 ICE ICE- 467,000 km2
C3
322 ICE 467,000 km2
323 ICE 467,000 km2
SERPUKHOVIAN 324 ICE ICE ICE? 450,000 km2
325 ICE ICE ICE? 450,000 km2
326 ICE ICE- Cold 450,000 km2 ← First rainforests
C2
327 ICE
328 ICE
329 ICE ICE
 330 ICE ICE-
C1
VISEAN (pars) 331
332

Source: Gondwana data from Isbell et al. (2003), Frank et al. (2008), and Fielding, Frank, Birgenheier
et al. (2008); Siberian data from Epshteyn (1981a), Chumakov (1994), and Raymond and Metz (2004).
Rainforest data are from Cleal and Thomas (2005); timescale modi ed from Gradstein et al. (2012).
Note: Climatic events column (1) is West and Central Gondwana and column (2) is East Gondwana
(Australia), both in the Southern Hemisphere; column (3) is Siberia (northeastern Russia) in the
Northern Hemisphere. Bold type indicates the existence of continental glaciers, and normal type
indicates the presence of alpine glaciers; glaciations in eastern Gondwana are designated C1
through C4, from oldest to youngest.

FIGURE 3.7 Paleogeographic distribution of lycophyte rainforests in the Late Carboniferous late
Moscovian Age (upper gure) contrasted with the Early Permian Asselian Age (lower gure).
Rainforest areas are shown in black, highland areas are shown in dark gray, lowland areas are
shown in light gray, and areas under marine waters are shown in white; see text for discussion.

Source: Thanks to Dr. Christopher Cleal, Department of Biodiversity and Systematic Biology,
National Museums and Galleries of Wales, for creating this gure for this book (modi ed from Cleal
and Thomas, 2005).

In the late Moscovian Age, huge expanses of lycophyte-dominated

tropical mires stretched across the continental-interior and Appalachian-
basin regions of the present-day United States ( g. 3.7). To the east,
lycophyte rainforests stretched continuously from eastern Canada in
North America across southern Ireland, through Wales and England,
across northern Europe, to vast tropical mires in Kazakhstan. Along the
way, large isolated basins lled with lycophyte-dominated mires existed in
northern Africa, Spain, and middle Europe. Further to the east, lycophyte
rainforests existed in basins in western Asia and in vast mires on the Sino-
Korean continental block (present-day Korea and North China; g. 3.7).
The spread of the great rainforests is clearly correlated with the onset
of the major Carboniferous phase of the Late Paleozoic Ice Age. Massive
continental ice sheets formed rst in western and central Gondwana in
the early Serpukhovian (glacial phase GII of Frank et al. 2008), with alpine
ice sheets in the highlands of eastern Gondwana (table 3.2) (glacial phase
C1 of Fielding, Frank, Birgenheier et al. 2008). By the late Serpukhovian,
lowland continental ice sheets had formed in eastern Gondwana (glacial
phase C2 of Fielding, Frank, Birgenheier et al. 2008), and the northern
hemisphere of the Earth had become glaciated as well, with a polar sea-ice
cap in Laurasia (o modern-day Siberia, northeastern Russia) (Epshteyn
1981a; Frakes et al. 1992; Stanley and Powell 2003; Raymond and Metz
2004).
Why was the spread of the great rainforests correlated with the onset
of the massive-ice-formation phase of the Carboniferous glaciation? The
huge accumulation of frozen water on land in massive glaciers had the
e ect of lowering global sea levels, and as the seas drained o of the land,
vast areas of once ooded shallow marine shelf were exposed to subaerial
conditions in the tropics. Thus vast areas of at, miry, wetland habitats
were created, in what the Kentucky Geological Survey geologist Stephen
Greb and colleagues call the “largest tropical peat mires in Earth history”
(Greb et al. 2003, 127)—perfect habitats for the lycophyte trees in
particular.
Not only did the falling sea level expose vast areas of land that was
once under water, it also had the e ect of lowering the base level of
erosion in the uplands. Rivers that once had been near sea level were now
at an elevation considerably higher than sea level, and they began to
incise their valleys, cutting downward toward the new sea level. This
increased erosion transported huge amounts of sediment outward over
the newly exposed atlands, creating large river deltas like our modern-
day Amazon delta—again, perfect habitats for the lycophyte trees in
particular.
Plants are photoautotrophic organisms; that is, they produce their own
food by using energy from the sun. In this process, they extract carbon
dioxide from the atmosphere, split hydrogen atoms from water molecules
in their leaves, and combine the hydrogen and carbon dioxide molecules
to produce hydrocarbon molecules like sugar. Once the hydrogen is split
from water molecules, the oxygen in the water molecules is left over as a
waste product and the plants simply dump it into the atmosphere.4
The spread of the great rainforests had profound consequences for the
Earth’s atmosphere. Huge amounts of carbon dioxide were being
extracted from the atmosphere to form food for the land plants in the
forests. At the same time, huge amounts of oxygen were being produced
by the plants and dumped into the atmosphere. Hour after hour, day after
day, year after year, for millions and millions of years, this process took
place in the great rainforests. The carbon-dioxide content of the Earth’s
atmosphere must have declined as a result, and the oxygen content of the
Earth’s atmosphere must have increased—but how can we measure these
changes in the atmosphere of the Earth in the Carboniferous, over 300
million years ago?

THE GREAT RAINFORESTS AND ATMOSPHERIC OXYGEN

We know that the concentration of oxygen in the Earth’s atmosphere was
very low in the Frasnian Age of the Late Devonian but that the amount of
oxygen in the atmosphere began to increase following the Famennian
Gap. How do we know this? The empirical data that support this
conclusion come from the distribution of charcoal deposits in the
sedimentary rock record. Charcoal is produced by wild res, and wild res
do not occur unless quite speci c levels of oxygen are present in the
atmosphere. Despite extensive stratigraphic searches, the Frasnian Age is
known for the rarity of charcoal deposits in its strata. The absence of
wild res in the Frasnian is quite striking, given the spread of Earth’s rst
forests during this same time interval. The geologists Andrew Scott, of the
University of London, and Ian Glasspool, of the Field Museum of Natural
History, comment: “Archaeopteris, the rst large woody tree, evolved in the
Late Devonian and spread rapidly. By the Mid-Late Frasnian, monospeci c
archaeopterid forests dominated lowland areas and coastal settings over a
vast geographic range. Despite this extensive wood biomass, charcoal
occurrences are rare with only isolated fragments of charred Callixylon
(archaeopterid) wood reported from this interval and small amounts of
inertodetrinite (microscopic charcoal fragments) preserved in early Late
Devonian Canadian coals” (Scott and Glasspool 2006, 10863). In fact, the
entire interval of the Eifelian Age through the Frasnian Age has been
termed the “charcoal gap,” as there is very little evidence of any wild res
anywhere during this period (Scott and Glasspool 2006, 10862, 10864).
In contrast, the Famennian Age is well known for numerous charcoal
deposits in its strata, deposits that demonstrate that wild res were
common. The oldest Famennian charcoal deposits yet discovered come
from strata in Pennsylvania dated to about 362.7 million years ago.5
Charcoal data from around the world indicate that frequent wild res
occurred in widely separated regions of the Earth in the last 3.5 million
years of the Famennian (Scott and Glasspool 2006, g. 1; Marynowski and
Filipiak 2007; Marynowski et al. 2010). An oxygen level of at least 13
percent has to be present in the atmosphere in order for wild res to
ignite (Scott and Glasspool 2006). Thus hard empirical data exist to
demonstrate that the atmosphere of the Earth contained 13 percent
oxygen, or more, in the last 3.5 million years or so of the Famennian. The
absence of charcoal in strata in the Famennian Gap, and in the mid to late
Frasnian, can be used to argue that oxygen levels present in the Earth’s
atmosphere were below 13 percent during this span of time. Woody plant
material was abundant in the Frasnian; if wild res occurred, then charcoal
should be present in Frasnian strata—but it is not.
In addition to the empirical charcoal data, another type of evidence
supports the previous conclusion concerning oxygen levels in the Earth’s
atmosphere during the Late Devonian: the “empirical model,” a model
whose predictions depend not only on the mathematics of the model’s
assumptions but also on input from empirical data. The geochemist
Robert Berner of Yale University and his colleagues have constructed two
empirical models, Rock-Abundance and Geocarbsulf, in an attempt to
predict atmospheric oxygen concentrations throughout the span of the
Phanerozoic on Earth (Berner et al. 2003; Berner 2006). Both models
predict that the concentration of oxygen in the Earth’s atmosphere
increased from the Frasnian into the Famennian—thus matching the
charcoal data—and then continued to increase throughout the
Carboniferous into the Permian ( g. 3.8). Speci cally, both models predict
minimum oxygen concentrations in the atmosphere in the early Frasnian
(380 million years ago): 17 percent in the Rock-Abundance model and 13
percent in the Geocarbsulf model, and higher oxygen concentrations in
the Famennian: 21–23 percent in the Rock-Abundance model and 16–17
percent in the Geocarbsulf model.

FIGURE 3.8 Modeled uctuations in atmospheric oxygen content from the Late Devonian through
the Triassic. The square data points and solid line show the predictions of the Rock-Abundance
model, and the diamond data points and dotted line show the predictions of the Geocarbsulf model;
see text for discussion. Geologic timescale abbreviations: L DEV, Late Devonian; CARB,
Carboniferous; PERM, Permian; TRI, Triassic.

Source: Data from Berner et al. (2003) and Berner (2006).

 However, Scott and Glasspool urge some caution in accepting either
model’s predictions: “Predictions of the degree to which O2 uctuated [in
the Late Devonian] are based on data-driven models. However, the
complexity of the feedback mechanisms that govern O2 levels results in a
large degree of uncertainty in these models, as is evident in their frequent
re nement. Additional data sets are invaluable to these re nements”
(Scott and Glasspool 2006, 10861). They note that the empirical models’
predictions of lower oxygen concentrations in the Frasnian atmosphere t
with the evidence of the “charcoal gap” (Scott and Glasspool 2006, 10862)
in the strata of this time period. However, given the frequency of charcoal
occurrences in late Famennian strata, they suggest that oxygen
concentrations in the atmosphere in the late Famennian may have been
higher than the 17 percent predicted by the Geocarbsulf model and more
in accord with the 23 percent concentration predicted by the Rock-
Abundance model. The discrepancy between the charcoal data set and the
Geocarbsulf model becomes even more marked in the Carboniferous:
“Collectively, these data suggest levels of O2 modeled for this interval
rising from 17 percent to 23.5 percent [in the Geocarbsulf model] are
inappropriate and instead favor prior, higher levels modeled at ≈23–31.5
percent [in the Rock-Abundance model], values further supported by the
occurrence of very large arthropods at this time” (Scott and Glasspool
2006, 10863). The appearance of gigantic arthropods on Earth in the
Carboniferous will be considered in detail in chapter 4.
Thus we have two lines of evidence that oxygen was in short supply in
the atmosphere of the Frasnian world: the charcoal-distribution data and
the empirically based geochemical models. This evidence leads
immediately to the question: why was oxygen in short supply? One
obvious way to address this question is to consider the organisms that
were producing the oxygen in the rst place—the plants. As outlined in
chapter 1, it is known that land plants su ered a major loss in diversity in
the end-Frasnian biodiversity crisis. The number of known macrofossil
genera dropped to 13, a minimum diversity level that persisted unchanged
through the Famennian Gap. Taking a look at longer timescales, Anne
Raymond and Cheryl Metz, paleobotanists at the University of Texas, have
shown that land plants were losing diversity through the entire Middle
Devonian into the Frasnian (Raymond and Metz 1995), though the loss was
not as severe as the diversity loss that occurred in the late Frasnian. Land-
plant macrofossil diversity was at its maximum in the Early Devonian
Emsian, with 38 genera known. Diversity dropped to 31 genera in the
Middle Devonian Eifelian, dropped further to 24 genera in the middle
Frasnian, and then dropped precipitously to a low of 13 genera in the late
Frasnian. In summary, the decline in land-plant diversity seen in the fossil
record roughly parallels the decline in oxygen levels in the atmosphere
predicted by the geochemical models for the Frasnian, suggesting a causal
relationship between land-plant diversity and atmospheric oxygen levels.
The reverse of this argument is also true: the increase in land-plant
diversity and the geographic spread of the great rainforests in the
Carboniferous and their persistence in the Permian should have produced
a major increase in atmospheric oxygen levels on the Earth. And indeed,
the geochemical models predict that just such an increase did in fact
occur ( g. 3.8). But, as Scott and Glasspool have cautioned, the two
geochemical models give di erent predictions: the Rock-Abundance
model predicts that a maximum concentration of 35 percent oxygen in the
atmosphere occurred about 290 million years ago, in the Artinskian Age of
the Early Permian, whereas the Geocarbsulf model predicts that a
maximum concentration of 31 percent oxygen in the atmosphere
occurred much later, about 270 million years ago, in the Wordian Age of
the Middle Permian. More important, the Geocarbsulf model predicts that
the oxygen content of the Earth’s atmosphere did not increase to 30
percent or higher until the Artinskian Age of the Early Permian—
atmospheric oxygen concentrations for the entire Late Carboniferous are
predicted to have been below 30 percent ( g. 3.8).6 The biological
implications of these model di erences in the predicted oxygen content of
the Earth’s atmosphere in the Carboniferous will be explored in detail in
chapter 4, where we will consider the evolution of animal gigantism in the
Carboniferous.

THE EARTH UP IN SMOKE?

Wild res are common in our present-day world with its atmosphere of 21
percent oxygen—we are all familiar with scenes of re ghters battling
blazes in California or Arizona in western North America, or in Spain or
Greece in Mediterranean Europe. However, truly massive wild res are
con ned to these generally arid regions of the Earth. In contrast, in a
world with an atmosphere of 25 percent oxygen, wild res will become
widespread even in wet climatic regions (Scott and Glasspool 2006). And in
a world with an atmosphere of 30 percent oxygen, almost every single tree
struck by lightning during a storm will catch re—regardless of how wet it
may be. A large thunderstorm with multiple lightning strikes in a
localized area would result in a raging wild re, burning even swamp
foliage down to the water level.
The Rock-Abundance model predicts that the oxygen content of the
Earth’s atmosphere ( g. 3.8) exceeded 25 percent around 340 million years
ago in the Visean Age of the Early Carboniferous, and the Geocarbsulf
model predicts a somewhat later date of 320 million years ago in the
Bashkirian Age of the Late Carboniferous. Moreover, the Rock-Abundance
model further predicts that oxygen in the Earth’s atmosphere reached the
30 percent level around 330 million years ago, in the Serpukhovian Age of
the Early Carboniferous, and remained above 30 percent until 260 million
years ago, in the Late Permian.
Were wild res almost constantly present in the great rainforests of the
Late Carboniferous Earth? Two lines of empirical evidence suggest that
they were. First, the University College London biochemist Nick Lane
notes that over 15 percent of the volume of some Late Carboniferous coals
is charcoal, “an extraordinary amount if we consider that coal beds are
formed in swamps, which under modern conditions virtually never catch
re” (Lane 2002, 95).7 The abundant presence of charcoal in Carboniferous
coal strata is hard fossil evidence that the wet, tropical mires of the Earth
were frequently on re. At present atmospheric levels of 21 percent
oxygen, our modern rainforests and swamps in Indonesia and Malaysia
are constantly wet and virtually re free; thus the atmosphere of the
Earth must have contained much more oxygen in order for tropical mires
to catch re in the Carboniferous.
Second, Lane notes that empirical evidence exists that Late
Carboniferous wild res burned exceedingly hot compared to our present-
day wild res, a point that further explains how wet foliage could burn so
frequently:

Coals that formed during periods of hypothetically high oxygen, such

as the Carboniferous and Cretaceous, contain more than twice as much
charcoal as the coals that formed during low-oxygen periods like the
Eocene (54 to 38 million years ago). This implies that res raged more
frequently in time of high oxygen and were not related to climate
alone. Some of the properties of the charcoal support this
interpretation. The shininess of charcoal depends on the temperature
 at which it was baked. Charcoals formed at temperatures about 400°C
are shinier than those that cooked at lower temperatures, and so
re ect back more of the light directed at them. The di erence can be
detected with great accuracy using a technique known as re ection
spectroscopy. The shininess of fossil charcoals from both the
Carboniferous and Cretaceous implies that they formed at searing
temperatures, almost certainly above 400°C and perhaps as high as
600°C, in res of exceptional intensity. (Lane 2002, 95–96)

Fueled by a rich oxygen atmosphere, ames burning at 400°C (752°F) to

600°C (1,112°F) could sweep through even the soggiest rainforest foliage in
the Late Carboniferous. These two empirical data—the amount of charcoal
present in Carboniferous coals and the high temperature at which that
charcoal was formed—strongly suggest that the Rock-Abundance model
for the oxygen content of the Earth’s atmosphere in the Carboniferous
and Permian is more accurate than the Geocarbsulf model ( g. 3.8).
The Late Carboniferous sky was probably almost always yellow-beige in
color, somewhat similar to the present-day orange-beige color of the sky
on Mars that we see in the numerous photographs taken by our robot
rovers on the Martian surface. The Martian sky is colored by dust; the
Carboniferous sky was colored by smoke. If we could have viewed the
Earth from space in the Late Carboniferous, we would have seen thick
black plumes of smoke rising from numerous spots in the green band of
the great rainforests, smearing out to long streaks of brown where the
rising smoke plumes encountered higher-level atmospheric winds. The
faint smell of smoke was probably almost constantly detectable in the air
around the world, even in higher-latitude localities far away from the
tropical band straddling the equator that contained the great rainforests
and their numerous wild res.
WHY DID SO MUCH COAL FORM IN THE LATE CARBONIFEROUS?
Nick Lane points out that 90 percent of the Earth’s coal strata were
deposited in the time span from the Serpukhovian Age in the Early
Carboniferous to the Wuchiapingian Age in the Late Permian, and marvels
at the “riddle posed by this 70-million-year period, which lasted from
around 330 to 260 million years ago. [This means that] 90 percent of the
world’s coal reserves date to a period that accounts for less than 2 percent
of the Earth’s history. The rate of coal burial was therefore 600 times faster
than the average for the rest of geological time” (Lane 2002, 84).
Why did so much coal form in the Late Carboniferous? One key to this
mystery may be the type of coal that is found in these strata—namely, its
unusually high charcoal content. Lane points out that “charcoal is
virtually indestructible by living organisms, including bacteria. No form of
organic carbon is more likely to be buried intact” (Lane 2002, 79) and that
some Carboniferous coal beds “contain over 15 per cent fossil charcoal by
volume…. The closest modern equivalents to Carboniferous coal swamps,
the swamps of Indonesia and Malaysia, are almost charcoal-free” (Lane
2002, 95). Scott and Glasspool further point out that we now know that the
mineral inertinite is fossil charcoal, and that inertinite is “a constituent
component of many coals” (Scott and Glasspool 2006, 10861) in the
Paleozoic. Thus part of the “riddle posed by this 70-million-year period,
which lasted from around 330 to 260 million years ago” (Lane 2002, 84) in
which the rate of coal burial was astoundingly high, may be attributable
to the high oxygen content of the Earth’s atmosphere during this same
period of time. If the Rock-Abundance model is correct, then the oxygen
content of the Earth’s atmosphere was above 30 percent for the entire
period from 330 to 260 million years ago ( g. 3.8). The resulting wild res
in the tropical mires of the Earth produced enormous amounts of nearly
indestructible charcoal to be buried in the rock record.
 A second key to the mystery of the massive amounts of coal found in
Carboniferous strata may have been the general absence of wood
consumers in Carboniferous ecosystems. Even if 15 percent of
Carboniferous coal is near-indestructible charcoal, what accounts for the
preservation of the other 85 percent that is not? Scott Richard Shaw, a
paleoentomologist at the University of Wyoming, asks: “Why does most of
our coal and much of our petroleum date from the Carboniferous period?
The orthodox view is simply that the Carboniferous swamps provided
optimal conditions for fossil fuel formation. After that time, the world
became drier, and conditions were not as favorable. But is that all there is
to it? Forests didn’t go away after the Carboniferous. If anything, there
were even more trees…. Something other than a change in the weather
must have occurred” (Shaw 2014, 74). Shaw argues that that “something
other” was the evolution of numerous wood consumers in post-
Carboniferous forests: “In the Early Carboniferous, most of today’s
macroscopic and microscopic consumers of dead wood had not yet
evolved. There were no birds, mammals, bees, wasps, bark beetles, wood-
boring beetles, bark lice, termites, or ants.” Thus, “the Late Devonian and
Carboniferous really were special for their excess production of plant
materials … because the plants were able to produce more biomass than
the herbivores could consume, for millions of years. The rst important
insect wood consumers—the wood roaches—did not appear until the Late
Carboniferous. They were followed by the appearance of bark lice and the
diversi cation of wood-boring beetles in the Permian” (Shaw 2014, 75). In
contrast, in the Early Carboniferous, the fungus consumers of trees were
joined only by the oribatid mites and a few detritus-feeding millipedes
and primitive wingless insects (Shaw 2014, 205, note 5).
A third key to the mystery of the Carboniferous coal strata might be
found in the peculiar growth mode of the ancient lycophyte and
equisetophyte trees. Did these trees simply produce more biomass than
existing Carboniferous wood consumers could destroy, or was there
another component to their excess biomass production? One possibility is
that the lycophyte trees in particular had extremely fast determinate
growth. It has been estimated that a 50-meter-high (160-foot-high)
mature tree could be produced in ten years—a tree that would then
promptly die after producing its spores (Phillips and DiMichele 1992).
However, this argument has been challenged by the Stanford University
paleobotanist Kevin Boyce and reconsidered by the Smithsonian
paleobotanist William DiMichele, who maintain that “such rapid growth
would violate all known physiological mechanisms” and that the giant
lycophyte trees grew at a pace comparable to that of a modern-day palm
tree (Boyce and DiMichele 2016). In any event, a single dead Lepidodendron
tree falling down into the surrounding mire could contain as much as 3.2
tonnes (3.5 tons) of carbon (estimated by Cleal and Thomas, 2005, using
the carbon-biomass estimates of Baker and DiMichele, 1997).
In a Carboniferous rainforest containing 500 to 1,800 such giant, fast-
growing lycophyte trees per hectare (2.5 acres) (DiMichele et al. 2001),
Cleal and Thomas calculate that these trees could extract between 160 and
578 tonnes (176 and 636 tons) of carbon from the atmosphere per year.
Unlike in modern lignophyte trees, this xed carbon did not then
continue to be present in living biomass as the trees lived on for decades
or even centuries. The mature lycophyte trees died in only ten years or so,
and the carbon contained in the dead tree would be taken out of the living
ecosystem and transferred to the swamp waters below. The water in the
tropical swamps appears to have been acidic and with low fungal activity
(DiMichele et al. 1985; Robinson 1990), and Cleal and Thomas calculate
that only about 25 percent of the carbon in the dead trees in the mires was
returned to the atmosphere by fungal consumption of the trees. Another 5
percent of the carbon was probably lost in runo water from the swamp,
resulting in a net accumulation of from 108 to 390 tonnes (119 to 429 tons)
of carbon in the swamp sediments per year.
These estimates of carbon accumulation in the rock record for the Late
Carboniferous rainforests are radically di erent from those for modern
rainforests. The lignophyte trees in a modern rainforest grow slowly and
live for decades or even centuries, and most of the carbon in the dead
vegetation is recycled back into the atmosphere by aerobic-respiring
fungal and bacterial consumers. In contrast to the estimated 108 to 390
tonnes (119 to 429 tons) of carbon stored in the sediments per year in the
lycophyte-dominated rainforest, it is estimated that in a modern
lignophyte-dominated rainforest, between two and ten tonnes (2.2 and 11
tons) of carbon goes into long-term sequestration per year (Cleal and
Thomas 2005).
Further evidence that the peculiar growth mode of the lycophyte trees
was a major factor in the accumulation of such major amounts of coal in
the rock record comes from comparing the amount of coal produced by
the Bashkovian-Moscovian rainforests, which were dominated by
lycophyte trees, with that of the Permian rainforests, which were
dominated by tree ferns. Cleal and Thomas note that both rainforests
produced a similar tonnage of coal—about 200 billion tonnes (220 billion
tons)—but that that amount was produced in about 11 million years by the
lycophyte-dominated rainforests in the Bashkovian-Moscovian span of
time, whereas it took the tree-fern-dominated rainforests about 35 million
years in the Permian to produce the same amount of coal. That is, the
lycophyte-dominated rainforests, with their fast-growing, determinate-
growth-mode trees, produced coal at a rate three times higher than the
slower-growing, non-determinate-growth-mode rainforests dominated by
tree ferns (Cleal and Thomas 2005).
 A fourth key to the mystery of the Carboniferous coal deposits might
be found in the atmosphere of the Carboniferous world: there simply may
have been much more carbon dioxide in the atmosphere that could be
extracted and xed in plant tissues than in our modern world. That is, net
plant productivity may have been much higher in the Carboniferous world
than in our modern world, where plants are carbon deprived in our
atmosphere containing 0.04 percent carbon dioxide. Even with a
hyperoxic atmosphere, atmospheric modeling by the University of
She eld botanist David Beerling and geochemist Robert Berner of Yale
University suggests that in an atmosphere with a carbon-dioxide content
of 0.06 percent, the “CO2 fertilization e ect is larger than the cost of
photorespiration, and ecosystem productivity increases leading to the net
sequestration of 117 Gt C [117 billion tonnes of carbon] into the vegetation
and soil carbon reservoirs. In both cases, the e ects result from the strong
interaction between pO2 [partial pressure of oxygen in the atmosphere],
pCO2 [partial pressure of carbon dioxide], and climate in the tropics”
(Beerling and Berner 2000, 12428). And the tropics were where the great
rainforests were located in the Carboniferous.
In summary, all of these factors—excess wild res and charcoal
production in a world with a hyperoxic atmosphere, the lack of numerous
wood consumers in Carboniferous ecosystems, rapid determinate growth
in the peculiar lycophyte and equisetophyte trees, and more carbon
dioxide in the atmosphere for the Carboniferous plants to x into more
plant biomass—acted in concert to produce 90 percent of the world’s coal
reserves in a period of time that accounts for less than 2 percent of the
history of the Earth.

THE KASIMOVIAN CRISIS IN THE GREAT RAINFORESTS

Something happened to the great rainforests in the Kasimovian,
something that would have been noticeable even from space. The
rainforests still existed in the Kasimovian, but the area of the Earth
covered by their green expanse had shrunk by 53 percent (table 3.2); that
is, over half the area of the green band that circled the Earth in the
Moscovian was gone in the Kasimovian (Cleal and Thomas 2005).
Descending into the rainforests from space, we would immediately
notice that the trees were di erent—the towering, peculiar lepidodendrid
scale trees that were so common in the Bashkirian and Moscovian
rainforests were gone. Instead, the majority of the trees surrounding us
were now tree ferns, marattialean licophytes (table 3.1). What happened
to the great lycophyte trees?
The University of Pennsylvania paleobotanist Hermann Pfe erkorn
and colleagues argue that the fossil record shows this marked changeover
in the great Carboniferous rainforests occurred in three steps: First came
the abrupt disappearance of the towering, pole-like lepidodendrid scale
trees. Species of the once-abundant lycophyte tree genera Lepidodendron
( g. 3.3), Lepidophloios, Paralycopodites, Hizemodendron, Diaphorodendron,
and Synchysidendron all vanish from the fossil record. Second, there was a
short interval of time in which the wetlands were populated by smaller
understory seed-reproducing trees—medullosan seed-fern trees and
cordaitean pinophyte trees—and a ground cover consisting of spore-
reproducing plants—sphenophylleacean horsetails and chaloneriacean
quillworts (table 3.1), both of which groups are now extinct. This brief
interval is sometimes represented by a single coal bed in the fossil record.
Third, the wetlands then began to be repopulated with spore-reproducing
trees—large marattialean tree ferns ( g. 3.5) and the surviving sigillarian
lycophyte scale trees ( g. 3.9). The sigillarian scale trees were smaller
than the lepidodendrids, averaging about 20 to 25 meters (66 to 82 feet) in
height (Taylor and Taylor 1993), with crowns that did not branch as
profusely as the lepidodendrids (compare gs. 3.3 and 3.9).

FIGURE 3.9 The lycophyte scale tree Sigillaria was about 20 to 25 meters (66 to 82 feet) tall, here
compared to a four-meter- (13-foot-) tall male African elephant.

Source: Illustration by Mary Persis Williams.

All in all, 87 percent of the Moscovian trees species and 33 percent of

the ground cover and vine species went extinct in the Kasimovian. In
addition, the proportion of trees to ground cover and vine species also
changed radically in the oral changeover. In the Moscovian, the ratio of
trees to ground cover and vines was 30 to 18; in the Kasimovian, that ratio
changed to 18 to 25, an almost complete reversal of the oral structure of
the rainforest (Pfe erkorn et al. 2008). Why were there so many more
vines in the Kasimovian rainforests? Michael Krings, a paleobotanist at the
Bavarian State Museum of Paleontology, and his colleagues argue that the
evolution and proliferation of new climbing plant species in the
Kasimovian was driven by a radical change in the canopy structure of the
tropical rainforests during the Late Carboniferous (Krings et al. 2003). The
Bashkirian and Moscovian tropical rainforests were dominated by the
peculiar arborescent lycophytes, polelike scale trees that did not possess
large leafy crowns even during their reproductive phase, when they grew
multiple drooping branches at the tops of their trunks. Moreover, their
canopy branches were covered by small, microphyll-type leaves that
would cast relatively little shadow on the ground below the tree. The
combination of these two structural characteristics of the lycophyte trees
resulted in a rainforest with a very open canopy, with a great deal of
sunlight reaching the ground surface below the trees.
That canopy structure of tropical rainforests changed in the
Kasimovian with the extinction of most of the lycophyte trees and their
replacement with euphyllophyte tree ferns. These trees possessed larger,
macrophyll-type leaves that would catch more sunlight in the canopy and
hence extensively shade the ground below the tree. In addition, the
canopy of the marattialean licophyte tree ferns existed throughout the
life of the tree, not just during a limited reproductive phase as with the
lycophyte trees, so the shading of the ground below was a more
continuous phenomenon of the new-style rainforest canopies. For the rst
time in geologic time, terrestrial forests evolved relatively closed canopies
much more similar to those seen in our modern forests today (DiMichele
and Hook 1992). For understory plant species, the evolution of closed
canopies was a disaster. Sunlight became a much more limited resource,
and thus selective pressure existed for the evolution of plants that could
climb trees and reach the canopy level of the forest where sunlight was
available. The proliferation of vines in the Kasimovian rainforests was the
result (Krings et al. 2003).
What could have caused the great rainforests of the Earth to shrink by
53 percent in area and, furthermore, caused the demise of the towering,
polelike lepidodendrid scale trees? One hypothesis to explain the
Kasimovian crisis in the great rainforests invokes a climatic kill
mechanism: the interpulse-drying model (table 3.3). The great ice cap that
had covered much of central and western Gondwana in the Serpukhovian
and Bashkirian retreated in the Moscovian. In eastern Gondwana, present-
day eastern Australia, the late Bashkirian C4-glacial-pulse continental ice
sheets began to wane from the middle to the end of the Moscovian (table
3.2, g. 2.5) (Fielding, Frank, Birgenheier et al. 2008), and in the Northern
Hemisphere the ice melted about 310 million years ago. The Earth was
entering a major interpulse period (end of C4 to beginning of P1 glacial
pulses; table 3.2), a period of warmer and drier climates that would last for
some nine million years in the Kasimovian and Gzhelian.

TABLE 3.3 Summary of the kill-mechanism models that have been proposed for collapse of the
great rainforest ecosystems in the early Kasimovian.

I. Climatic Kill Mechanism

A. Interpulse-drying model
B. Continental-positioning model
C. Refugia-constriction model
II. Tectonic Kill Mechanism
A. Continental-uplift model
III. Hyperoxic Kill Mechanism
A. Carbon-depletion model
IV. Fungal Kill Mechanism
A. Tree-blight model

Note: See text for model sources.

Hermann Pfe erkorn and his colleagues argue that the demise of the
great lepidodendrid scale-tree forests was triggered by the onset of these
warmer and drier interpulse conditions: “The transition from the middle
to late Pennsylvanian witnessed a major oristic change within the
wetland biome…. This involved a wholesale change from wetlands
dominated by lepidodendrid lycopsids to dominance by marattialean tree
ferns…. These patterns appear to be a response to a strong pulse of global
warming at or near the Westphalian-Stephanian boundary” (Pfe erkorn
et al. 2008, 309). The Westphalian and Stephanian are older geologic time
divisions of the Late Carboniferous that were widely used in Europe before
the acceptance of the modern epochs and ages (table 1.2). The
Westphalian spanned the interval of time from the middle Bashkirian
through most of the Moscovian, and the Stephanian spanned the interval
of time from the latest Moscovian through the middle Gzhelian (Carroll
2009, g. 4.1).8 The paleobotanists Christopher Cleal and Barry Thomas
(2005) date the collapse of the great rainforests to the Moscovian-
Kasimovian boundary. Their data indicate that the great rainforests
reached an all-time maximum extent of 2,395,000 square kilometers
(924,470 square miles) in the late Moscovian and shrank to only 1,131,000
square kilometers (436,566 square miles) in the early Kasimovian (table
3.2).
Pfe erkorn and colleagues argue that it was the interpulse drying-out
of many of the formerly extensive tropical mires that killed the
lepidodendrid scale trees. They point out that not all of the peculiar
lycophyte trees perished—the sigillarian scale trees survived ( g. 3.9).
Most of the lycophyte trees required standing water for fertilization and
germination of their spores, and thus they “preferred habitats in which
water tables were high, and only one group—the sigillarians—possessed
ecological tolerances that allowed them to colonize better-drained soils
(DiMichele and Phillips, 1996). It was the sigillarians that survived the
climatic perturbation and continued into the Stephanian [= Kasimovian],
where they co-occur with pteridosperms [= non-core seed plants] in
mineral substrate settings [= not in peat-accumulating wetlands]”
(Pfe erkorn et al. 2008, 313).
The lepidodendrid scale trees were ecologically replaced by
marattialean tree ferns, and although tree ferns also reproduce by spores,
they do not require standing-water conditions for germination and
fertilization. Rather, tree fern spores “require moist conditions for
germination and su cient rainfall during a season when gametes can be
produced and fertilization occurs. Tree ferns do not require standing-
water conditions for this to happen. Hence, the reproductive strategy of
this group involving high fecundity and airborne distribution of spores
allowed them to colonize soils developed under a climate characterized by
more seasonal rainfall” (Pfe erkorn et al. 2008, 313). The high fecundity of
ferns is due to their production of small spores in very large numbers.
Ecologically, high fecundity is a characteristic of weeds—they grow rapidly,
produce large numbers of o spring, and die quickly. This type of life-
history strategy is very useful to modern-day “disaster species,” enabling
them to rapidly colonize areas where the normal vegetation has been
destroyed, such as in a forest re. And in the Moscovian-Kasimovian
extinction, the Moscovian-style rainforests were destroyed forever.
 Finally, the paleobotanical e ect of the return of continental glaciers in
the Permian phase of the Late Paleozoic Ice Age (to be considered in detail
in chapter 5) gives further support to the interpulse-drying climatic kill
model. Massive ice sheets once again formed across Gondwana in the
Asselian and Sakmarian Ages in the early Permian, and extensive
rainforests once again formed as well. These rainforests were not as
geographically extensive as the Carboniferous rainforests and were
dominated by tree ferns rather than the peculiar lycophyte trees of the
Carboniferous, but it is notable that these new rainforests formed at the
same time as the return to colder, wetter climates in the Permian world.
Only in the far east, on the large islands of the Sino-Korean continental
block (present-day Korea and North China) and the Yangtze continental
block (present-day South China), did the lycophyte-dominated rainforests
ourish in the Asselian Age ( g. 3.7). The equatorial regions to the west,
covered by lush lycophyte-dominated tropical mires in the late Moscovian
of the Carboniferous ( g. 3.7), were now populated by plants that had
once been found mostly in highland regions or in savannah environments
in the lowlands (Cleal and Thomas 2005). As discussed above, the post-
Kasimovian-crisis forests of marattialean tree ferns, gigantopterid seed-
fern trees, and cordaitean conifers (table 3.1) were adapted to drier
climates and did not produce peat bogs that would become coal deposits.
In contrast to the interpulse-drying model is the alternative
continental-positioning climatic kill model (table 3.3): First, the gradual
tectonic assembly of all the continents of the Earth into the “all
continent” or “world continent” Pangaea9 resulted in the Earth’s
landmasses being more or less symmetrically arranged north and south of
the equator. Second, it is proposed that the presence of equivalent areas of
land straddling the equator caused monsoonal atmospheric circulation to
develop, shifting rainfall away from the tropics along the equatorial
region of Pangaea. Third, with less rainfall, the wetland habitats of the
equatorial regions were disrupted and eventually dried out, leading to the
extinction of the equatorial rainforests (Phillips and Peppers 1984;
DiMichele et al. 1985). This model does not necessarily negate the
interpulse-drying model; it could be argued that both climatic-drying
mechanisms operated simultaneously at the Moscovian-Kasimovian
boundary to trigger the demise of the hydrophyllic lepidodendrid scale
trees.
The continental-positioning model has been questioned by Cleal and
Thomas, who point out that “most of the terranes that eventually gave
rise to Pangaea were already assembled by the early Bashkirian Age”
(Cleal and Thomas 2005, 21) and that, rather than declining and shrinking,
tropical rainforests began to ourish and spread during the Bashkirian
(table 3.2). As an alternative to either the interpulse-drying or
continental-positioning climatic kill mechanism, they propose a tectonic
kill mechanism: “It has been suggested that the changes in the
geographical distribution of the palaeotropical coal forests were mainly
driven by climatic change, but our analysis instead suggests that the
underlying controlling-factor was plate tectonics” (Cleal and Thomas
2005, 21). They argue that the onset of the tectonic uplift of the Variscan
mountain chain in the middle Moscovian in the European region of
Laurussia lowered the water table throughout Europe. Lowering the water
table disrupted and drained many of the wetland habitats, and uplifted
mountainous regions shed more eroded sediment into lowland habitats
and thus also disrupted and lled in low-lying wetland regions. Tectonic
uplifting is proposed to have spread westward across the Laurussian
supercontinent, disrupting and draining wetland habitats in the North
American region by the end of the Moscovian. Likewise, in central Asia to
the east, Cleal and Thomas propose that “similar e ects of tectonics can
be seen in the central palaeotropical coal forests, where they disappeared
following the so-called ‘Hercynian’ tectonic activity resulting from the
suturing of the Kazakhstan and Angara plates” (Cleal and Thomas 2005,
22). They argue that only China saw declining tectonic activity in the
Moscovian, and note that in China the rainforests persisted into the
Kasimovian.
In rebuttal of the tectonic kill mechanism, Pfe erkorn and colleagues
argue that “the rate of change of tectonics and rate of vegetation change
at the Westphalian-Stephanian are simply not congruent. Tectonic uplift
requires durations of time on the order of millions of years,” but “[t]he
oral change at the Westphalian-Stephanian boundary … is nearly
instantaneous, implying a widespread and fast-acting causative agent”
(Pfe erkorn et al. 2008, 312). They argue that only global climatic change
would a ect the entire planet simultaneously, and the rapidity of global
climatic change from glacial to interglacial phases is well known to us:
only 11,000 years ago, huge glaciers covered northern Europe and North
America, yet now, in an interglacial period, northern continental glaciers
remain only in Greenland. And the rainforests did indeed persist into the
Kasimovian Age in China, but their oral composition changed radically:
the peculiar lepidodendrid scale trees vanished and were replaced by
marattialean tree ferns. If the lack of major tectonic uplift and the lack of
lowered water tables in China during the Kasimovian permitted the
rainforests to persist, then why did the lepidodendrid scale trees not
persist in China as well? Even in these remaining wetland regions, major
oral changes occurred, a phenomenon more compatible with global
climate change as a triggering mechanism.
Finally, a totally di erent climatic interpretation for the Carboniferous
tropical peat mires has been proposed by Howard Falcon-Lang, a
paleobotanist at the University of London, who argues that during Late
Carboniferous glacial episodes, tropical climates were cool and seasonally
dry, not wet, and in some cases even semiarid. He further argues that
interglacial episodes were warm and humid , not dry, and it was during the
interglacial episodes that the lepidodendrid-dominated tropical mires
formed (Falcon-Lang 2004). Falcon-Lang then teamed up with the
Smithsonian Institution paleobotanist William DiMichele to propose a
new potential climatic kill mechanism: the refugia-constriction model
(table 3.3) (Falcon-Lang and DiMichele 2010). The paper in which this
model is proposed is particularly interesting in that the two authors state
clearly in the introduction that they disagree on the paleoclimatic
reconstruction for the Late Carboniferous coal deposits: Falcon-Lang
argues that the coal strata represent transgressive sea-level to highstand
sea-level deposits, formed during the glacial-to-interglacial interval when
the ice sheets were melting on Gondwana and global sea level was rising
to its eventual maximum highstand. In contrast, DiMichele argues that the
coal strata represent early to late lowstand sea-level deposits, formed
during glacial episodes when ice sheets were spreading to their maxima
on Gondwana and global sea level was falling to its eventual minimum
lowstand (Falcon-Land and DiMichele 2010).10
Despite their diametrically opposing views on paleoclimate conditions
during the formation of the coal strata, Falcon-Lang and DiMichele agree
that glacial periods were cool and dry with globally low sea level and that
interglacial periods were warm and humid with globally high sea level.
They agree that the lepidodendrid-dominated rainforests were most
geographically widespread during interglacial humid periods, and most
geographically restricted during glacial dry periods. They note that
during the numerous smaller, short-term glacial-interglacial rhythms, or
cyclothems (Heckel 2008), that occurred in the Late Carboniferous, “coal
forests dominated during humid interglacial phases, but were replaced by
seasonally dry vegetation during glacial phases. After each glacial event,
coal forests reassembled with largely the same species composition,” in
that less than 10 percent species turnover occurred in each
disappearance-reappearance event in the rainforests (Falcon-Land and
DiMichele 2010, 611). They then conclude: “There is only one credible
solution to this paradox. Coal forests contracted into geographically
restricted refugia during glacial phases and surviving taxa then
repopulated tropical lowlands when climate became more humid at the
start of the next interglacial” (Falcon-Lang and DiMichele 2010, 613). In
this model, they propose that the obligate-wetland lepidodendrid scale
trees were con ned to wetland refugia within river valleys during glacial
periods: “While we cannot demonstrate unequivocally that dispersed
refugia survived in equatorial valleys during glacial periods, it remains a
viable hypothesis that is amenable to further testing. It is worth
emphasizing that incised valley systems acted as long-term refugia for
wetland plant taxa during periods of dry climate at other times in
geological history (Demko et al., 1998)” and that “valley drainages have
been proposed as refugia for tropical rainforest species during Pleistocene
glacial phases (Meave and Kellman, 1994)” (Falcon-Lang and DiMichele
2010, 614).
How, then, is the Kasimovian crisis in the lepidodendrid-dominated
rainforests to be explained in this di erent climatic kill model—
particularly in that the crisis occurred during the major climatic
transition from the end of the C4 glacial phase in eastern Gondwana to the
beginning of the nine-million-year-long C4-to-P1 interpulse phase in the
Kasimovian and Gzhelian (table 3.2)? Surely the lepidodendrid-dominated
rainforests should have experienced a resurgence in the reestablishment
of humid conditions in this major interpulse interval, if indeed humid
conditions did occur (and not dry conditions, as argued in the interpulse-
drying and continental-positioning climatic kill models). Instead, the
obligate-wetland lepidodendrid scale trees abruptly disappeared and were
replaced by dry-tolerant tree ferns during the C4-to-P1 interpulse interval
(table 3.2).
Falcon-Lang and DiMichele argue that the intensity of climatic swings
in the short-term glacial-interglacial cyclothem rhythms increased
steadily throughout the Moscovian C4 glacial interval; that the “e ect of
increasing glacial intensity through the late Middle Pennsylvanian [ = late
Moscovian] would have been to progressively reduce the size and
connectivity of tropical refugia from one glacial phase to the next”; and
that “[t]his trend culminated in maximum intensity at the Middle-Late
Pennsylvanian [ = Moscovian-Kasimovian] boundary, marked by an
extreme marine regression (Heckel, 1991) … and valley incision (Easterday,
2004)” (Falcon-Lang and DiMichele 2010, 614). That is, they argue that the
C4 glacial interval ended with an extreme ice-buildup pulse that
drastically lowered global sea levels (Heckel 1991), and hence river-
erosion base levels, triggering erosive downcutting and valley incision in
river valleys. On a global scale, with falling sea level, coastal river valleys
became progressively deeper and narrower—and the obligate-wetland
lycophyte species were con ned to refugia in these same river valleys.
They then conclude that “the abrupt step change from coal forests
dominated by lycopsids to those dominated by tree ferns around the
Middle-Upper Pennsylvanian [ = Moscovian-Kasimovian] boundary
ultimately resulted from hyperconstriction of equatorial refugia during an
extreme glacial phase. These changes presumably reduced the already-
diminished lycopsids to unsustainably low numbers that eliminated some
species and severely reduced the ability of survivors to recolonize
environmentally favorable areas following deglaciation at the start of the
Late Pennsylvanian [ = Kasimovian]” (Falcon-Lang and DiMichele 2010,
615).
In contrast to either climatic or tectonic kill mechanisms, both of
which invoke nonbiological physical factors, the University of Washington
paleontologist Peter Ward argues that the great rainforests may have been
responsible for their own demise! The rainforests are known to have
extracted an enormous amount of carbon dioxide from the atmosphere—
hence the worldwide distribution of coal strata of Carboniferous age and
the very name “Carboniferous” for this interval of geologic time—but did
the great rainforests draw down the carbon dioxide content of the
atmosphere too far? In essence, did the rainforests run out of carbon?
If carbon-dioxide partial pressures in the atmosphere are too low, and
oxygen partial pressures are too high, then the presence of oxygen can
interfere with the photosynthetic process that most land plants use to
produce sugars. In the middle Cenozoic, about 30 to 50 million years ago, a
critical threshold was reached in the evolution of the Earth’s atmosphere.
This event was caused partly by the activity of land plants themselves, as
they have been extracting carbon dioxide from the atmosphere and
releasing free oxygen back into the atmosphere for the past 425 million
years, since the rapid radiation of land plants in the Late Silurian. Another
factor was the formation of the gigantic Himalayan mountain chain,
produced by the plate-tectonic-driven collision of the Indian subcontinent
with Asia, and subsequent weathering processes that have removed a
substantial amount of carbon dioxide from the Earth’s atmosphere. In any
event, plants began to experience di culty in xing carbon using their
usual photosynthetic pathway.
The majority of land-plant lineages use an ancient photosynthetic
pathway known as C3, which goes all the way back to the evolution of the
rst cyanobacteria (Sage 1999). To deal with the problem of too little
carbon dioxide, and too much oxygen, in the atmosphere, the owering
plants (angiosperms) evolved a more e cient photosynthetic pathway,
known as C4, in the middle Cenozoic. Interestingly, no less than 34
di erent lineages of angiosperms have convergently, independently,
evolved the C4 photosynthetic pathway (McGhee 2011).
The catch is that C4 photosynthesis has only been evolved by the
angiosperms, and the angiosperms only evolved in the Early Jurassic—
they were not present in the Carboniferous world (table 3.1). In fact, none
of the living plant lineages that were present in the Carboniferous world
have evolved C4 photosynthesis today, and there is no evidence that any of
the Carboniferous members of these lineages ever evolved C4
photosynthesis (Sage 1999). C4 photosynthesis appears to be
developmentally constrained to the angiosperm plant lineage; that is, it
seems that only the owering plants have the developmental exibility to
evolve the biochemical and cellular changes necessary for C4
photosynthesis (McGhee 2011).
Atmospheric models indicate that oxygen contents reached very high
levels in the Late Carboniferous ( g. 3.8), and this same period
corresponds to a carbon-dioxide minimum in the atmosphere (Berner
2006). The Earth’s atmosphere became hyperoxic, and carbon depleted,
during the Late Carboniferous to Middle Permian interval of geologic
time. These observations led Peter Ward to argue: “As far as is known, C4
plants did not exist in the Permian and thus the drop in carbon dioxide
greatly a ected plant life, as shown by the presence of a plant extinction
at the time of minimum carbon dioxide levels, at about 305 [= 307,
beginning of the Kasimovian in table 3.2] million years ago … about two-
thirds of all plant species known from coal seams going extinct … most
authors blame this extinction on a drying of the many coal swamps … it
seems as likely that the extinction was at least partially caused by the
carbon dioxide minimum” (Ward 2006, 137). That is, the extinction of the
peculiar lepidodendrid scale trees may have been triggered by their own
excessive photosynthetic activity in producing a hyperoxic kill
mechanism—the carbon-depletion model for plant extinction (table 3.3).
The hyperoxic kill mechanism is one that would a ect the entire planet
immediately once the critical threshold ratio of carbon dioxide to oxygen
was reached. Thus the carbon depletion model is consistent with the
nearly instantaneous extinction of the lepidodendrid scale trees that
occurred at the Moscovian-Kasimovian boundary. Experiments by the
University of She eld botanist David Beerling and his colleagues show
that when two modern plant species—Hedera helix (English ivy) and Betula
pubescens (birch), both of which are angiosperms—were grown in
laboratory conditions with an atmospheric composition of 35 percent
oxygen, their leaf photosynthetic rates were 29 percent lower than those
of control plants grown in the present Earth’s atmosphere with 21 percent
oxygen. Model calculations then suggest that the global net primary
productivity of land plants may have decreased as much as 18.7 percent—a
decrease of 6,300 million tonnes (6,943 million tons) of carbon xed per
year—if atmospheric concentrations of oxygen reached 35 percent in the
Carboniferous (Beerling et al. 1998). Could this proposed oxygen-induced
suppression of photosynthetic e ciency have killed the lepidodendrid
scale trees?
Interestingly, the density of stomatal pores in the leaves of the plants
increased in the higher-oxygen-grown plants. Beerling and colleagues
point out that one possible explanation for this e ect might be the plants’
attempt to increase the di usion rate of carbon dioxide into the leaf
under high-oxygen conditions, and thus to acquire more of the
atmospherically depleted carbon for photosynthesis. They further note
that Carboniferous leaf fossils typically have much higher densities of
stomatal openings than do leaves of plants from other geologic time
periods, and that this observation would make sense if the Carboniferous
plants had grown in an atmosphere with much higher oxygen levels than
that of the present Earth’s atmosphere (Beerling et al. 1998).
However, Christopher Cleal and colleagues have reported an observed
decrease in stomatal density in fossil leaves of the medullosan seed fern
Neuropteris ovata during the late Moscovian, before the abrupt
disappearance of the lepidodendrid scale trees at the Moscovian-
Kasimovian boundary (Cleal et al. 1999). Given the experimental results of
Beerling and colleagues, the decrease in density of stomatal pores in plant
leaves could be taken as evidence of a decrease in oxygen content in the
Earth’s atmosphere before the extinction of the lepidodendrid scale trees,
not an increase. That is, if the lepidodendrid scale trees were su ering
from the e ects of high atmospheric concentrations of oxygen in the
Moscovian, then they should have begun to recover with a decrease in
oxygen content in the atmosphere in the late Moscovian, not suddenly
gone extinct instead.
Further di culties for the carbon-depletion model of land-plant
extinction arise from uncertainties in the actual amount of carbon dioxide
present in the Late Carboniferous atmosphere. Subsequent computer
modeling suggests that the roughly 20 percent drop in terrestrial primary
productivity predicted in plants living in an atmosphere of 35 percent
oxygen would have occurred only if the carbon-dioxide content of that
atmosphere was 0.03 percent or less (similar to that of the atmosphere of
the pre-industrial-age Earth). If the carbon-dioxide content of the
atmosphere is greater than 0.04 percent, then photosynthetic e ciency
actually increases in this more carbon-dioxide-rich atmosphere, o setting
the cost of photorespiration caused by the 35 percent oxygen content of
the atmosphere, and results in an increase in carbon xation, not a
decrease. Thus David Beerling and Robert Berner conclude in this later
study that the “Permo-Carboniferous rise in pO2 [= the oxygen content of
the atmosphere] was unlikely to have exerted catastrophic e ects on
ecosystem productivity” (Beerling and Berner 2000, 12428).
Another possible biological kill mechanism for the lepidodendrid scale
trees is tree blight—that the lepidodendrid trees were attacked by a clade-
speci c fungus that specialized in infesting these forms of lycophyte scale
trees (table 3.3). I have often wondered if the sudden disappearance of the
lycophyte trees of the family Lepidodendraceae at the end of the
Moscovian was equivalent to the sudden death of huge numbers of
American chestnut trees, Castanea dentata , in North American forests in
1904. In that year, the invasive fungus Endothia parasitica was introduced
inadvertently into North America, and within 40 years this parasitic
fungus had wiped out the magni cent tree species that had once been so
widespread in North American forests. The American chestnut trees used
to stand 30 meters (100 feet) tall, with trunks 1.2 meters (four feet) in
diameter, but now they are gone (Little 1980).
Forty years is but a blink of the eye in the geologic record, and it is easy
to visualize the evolution and rapid spread of a fungal species that
specialized in attacking trees of the Lepidodendraceae in the late
Moscovian, a scenario I call the “tree-blight model” (table 3.3). Even if it
took several hundred years for the fungal blight to spread globally in the
great rainforests, that interval of time would appear as instantaneous in
the geologic record. Tree blight would explain the sudden absence of the
lepidodendrid scale trees in the rainforests that still existed in the
Kasimovian. Surely some of the Moscovian lepidodendrid scale trees must
have survived the shrinking of the areal extent of the rainforests and lived
on in reduced numbers in their reduced swampy habitats in the
Kasimovian—but this was not so, they all vanished, and tree blight would
explain that empirical observation. The survival of lycophyte trees of the
family Sigillariaceae into the Kasimovian could perhaps have been
equivalent to the survival of the American-chestnut-related Ozark
chestnut, Castanea ozarkensis. It is attacked by the same fungal parasite
that caused tree blight in the American chestnut, but the Ozark species is
more resistant to the fungal infestation and is not threatened with
extinction (Little 1980).
Finally, it may well be that the extinction of the lepidodendrid scale
trees and the dramatic shrinkage in the geographic extent of the great
rainforests in the early Kasimovian were not triggered by any single kill
mechanism. That is, the catastrophe may have been the product of several
kill mechanisms operating simultaneously: water stress in land plants
triggered by a global climatic shift from a cold, wet glacial phase to a
warm, dry interpulse phase; major tectonic uplifting and wetland draining
in regions of Europe, North America, and Asia; all possibly exacerbated by
photosynthesis stress in land plants caused by carbon depletion and
hyperoxia in the Earth’s atmosphere; and possible tree blight in the
Lepidodendraceae. For whatever reason or reasons, the great
Serpukhovian-Moscovian lepidodendrid-dominated rainforests were
destroyed forever.
In the Permian, the ice would return for the second major phase of the
Late Paleozoic Ice Age, and the rainforests would begin to recover and to
spread once again (table 3.2). We will continue to examine the evolution of
the great rainforests of the Late Paleozoic Ice Age in chapter 5, but rst let
us explore the biological consequences of the Earth’s hyperoxic
Carboniferous atmosphere for animals—both land animals and marine
animals—in chapter 4.
4 Giants in the Earth …

There were giants in the Earth in those days …

—Genesis 6:4 (King James version)

THE STRANGE CARBONIFEROUS ANIMALS

The writer of the book of Genesis states that there were giants in the Earth
in the early days following the creation. In the chronology outlined in
Genesis, those giants existed about 6,000 years ago. In actual fact, the rst
giant land animals on the Earth existed in the Carboniferous, some 345
million years ago. Both on the land and in the seas, animals began
appearing in the Carboniferous that were much larger than any living
modern-day counterparts—in some cases, as much as ten to 12 times
larger! In this chapter we will examine the empirical evidence for the
evolution of animal gigantism in the Carboniferous and then explore the
possible causes of that gigantism.

Giant Arthropods

The Carboniferous giant arthropods (table 4.1) began to appear in the

Visean Age, not long after the end of the Tournaisian Gap. One notable
example is the fossil of a giant scorpion, Pulmonoscorpius kirktonensis,
discovered in East Kirkton in Scotland, that was 700 millimeters (28
inches) long. Here in North America, di erent living species of scorpions
range in length from 37 to 127 millimeters (1.5 to ve inches), and the
“giant desert hairy scorpion” of the Southwest, Hadrurus arizonensis,
reaches a length of 140 millimeters (5.5 inches) (Milne and Milne 1980). In
comparison with Pulmonoscorpius kirktonensis, we can see that our modern
“giant” scorpion is no giant at all—the Carboniferous scorpion was fully
ve times larger! Imagine hiking in the Arizona desert today and
encountering a real giant scorpion, one that is as long as a midsize dog
( g. 4.1). How could a scorpion have achieved such a gigantic size? Yet it
was only a harbinger of what was to come.

TABLE 4.1 Terrestrial arthropod lineages of the Late Paleozoic Ice Age.

ARTHROPODA (jointed appendages)

– CHELICERIFORMES
– – Chelicerata
– – – MEROSTOMATA (sea scorpions and kin)
– – – – – Water scorpion giants
– – – ARACHNIDA (scorpions and spiders)
– – – – – Scorpion, trigonotarbid, spider giants
– MANDIBULATA
– – MYRIAPODA (multi-legged)
– – – Diplopoda (millipedes)
– – – – Arthropleurid giants
– – – Chilopoda (centipedes)
– – PANCRUSTACEA
– – – Hexapoda (six-legged)
– – – – Entognatha
– – – – – Dipluran giants
– – – – INSECTA (insects)
– – – – – basal insects
– – – – – – Silver sh giants
– – – – – PTERYGOTA (winged insects, conquerors of the land)
– – – – – – Ephemeroptera (may ies)
– – – – – – – May y giants
– – – – – – Metapterygota
– – – – – – – Odonatoptera (dragon ies, damsel ies)
– – – – – – – – Meganisoptera
– – – – – – – – – Gri en y giants
– – – – – – – Palaeodictyopterida
– – – – – – – – Palaeodictyopteran sap-sucker giants
– – – – – – – NEOPTERA (folding-wing insects)
– – – – – – – – Cockroach giants

Source: Phylogenetic classi cation modi ed from Grimaldi and Engel (2005) and Lecointre and Le
Guyader (2006).
Note: Taxa containing giants are in italics; major clades are in capitals.

FIGURE 4.1 Some Late Paleozoic scorpions, such as the Early Carboniferous Pulmonoscorpius
kirktonensis, were as long as a midsize dog.

Source: Illustration by Mary Persis Williams.

 Giant arthropods really started becoming numerous on the Earth in the
Moscovian Age of the Late Carboniferous. Close relatives of the scorpions,
spiders and their ancient kin began to achieve gigantic sizes. The extinct
trigonotarbids were morphologically very similar to modern spiders, but
they did not possess silk glands to spin webs. They started out small in the
Devonian, from two to 15 millimeters (0.08 to 0.6 inch) long, not unlike
modern spiders. In the Late Carboniferous, however, trigonotarbids
evolved that were 50 millimeters (two inches) long—more than three
times larger than the largest Devonian trigonotarbids (Shear and
Kukalová-Peck 1990). A gigantic spider, Megarachne servinei from
Argentina, had a leg span of almost 500 millimeters (20 inches) (Shear and
Kukalová-Peck 1990; Lane 2002).1 For those who are arachnophobic, just
imagine encountering a spider over a foot and a half long!
Other giant Moscovian arthropods include the 2.5-meter- (eight-foot-)
long millipede species Arthropleura armata and A. mammata, animals that
were 150 millimeters (six inches) wide and weighed up to ten kilograms
(22 pounds) (Kraus and Brauckmann 2003).2 Most millipede species today
are tiny, slender animals about 50 millimeters (two inches) long; our
largest species today, the African “giant black millipede” Archispirostreptus
gigas, can reach 385 millimeters (15 inches) in length and 20 millimeters
(0.8 inch) in width. Once again, we see that our modern “giant” arthropod
species is not a giant—the Carboniferous millipede species were six and a
half times larger! In fact, the Carboniferous millipedes were fully as long
as a modern alligator ( g. 4.2).
FIGURE 4.2 Some Late Paleozoic millipedes, such as the Late Carboniferous Arthropleura armata,
were as long as a modern alligator.

Source: Illustration by Mary Persis Williams.

The hexapods and insects also began to evolve species with

unprecedented body sizes in the Late Carboniferous (table 4.1). The
forcipate diplurans (Entognatha; table 4.1) are usually very small cryptic
insects, hiding under logs and leaves on the ground, and are generally four
to six millimeters (about one-sixth to one-fourth of an inch) long (Milne
and Milne 1980). The gigantic forcipate dipluran species Testajapyx
thomasi, found in Moscovian-aged Mazon Creek strata in Illinois, was 58
millimeters (2.3 inches) long, including a 48-millimeter body and ten-
millimeter antennae (Kukalová-Peck 1987). On average, the Late
Carboniferous Testajapyx thomasi was 12 times larger than living forcipate
diplurans! In the same Mazon Creek strata are found gigantic silver sh
insects (basal insects; table 4.1), the species Ramsdelepidion schusteri, that
are 60 millimeters (2.4 inches) long. Modern silver sh species can often
become pests, living in moist regions of the house, such as under sinks or
around water pipes, and emerging at night to feed on book bindings or
clothes—but their silvery, attish bodies are usually small, around nine to
13 millimeters (one-third to one-half inch) long (Milne and Milne 1980).
On average, the Late Carboniferous silver sh were more than ve times
larger—imagine nding a silver sh the length of your little nger
munching on your favorite book!
The ground-dwelling millipede and insect giants were not alone. Giant
insects ew in the skies of the Late Carboniferous and Early Permian
(table 4.1). The most primitive of the living ying insects are the
Ephemeroptera, the may ies (table 4.1). Modern may ies live for only one
day, hence their scienti c name “ephemeral wings.” The may y itself is
merely the mobile, reproductive phase of the species—after mating and
laying their eggs, they die. The Carboniferous may ies, however,
possessed functioning biting mouthparts and clearly fed as adults, not just
as aquatic nymphs. And they were gigantic: the Moscovian Age may y
species Bojophlebia prokopi had a wingspan of 450 millimeters (18 inches)
(Kukalová-Peck 1985). In contrast, the largest living may ies in North
America, the golden may ies, can have a wingspan up to 60 millimeters
(2.4 inches)—the Carboniferous may y Bojophlebia prokopi was over seven
times larger. In fact, the wingspan of Bojophlebia prokopi was as wide as
that of a modern midsize bird, such as a blue jay, rather than an insect.
Astonishingly, even larger ying insects existed in the Late
Carboniferous and Early Permian. The most impressive of these were the
meganisopterans, the gri en ies (table 4.1), which were ying predators
similar in appearance to modern-day dragon ies—but much bigger. Giant
gri en ies, such as the Chinese gri en y Shenzhousia qilianshanensis,
which had a wingspan of 499 millimeters (20 inches), began to appear in
the Bashkirian.3 The Moscovian gri en y Bohemiatupus elegans in Europe
was larger, with a wingspan of 552 millimeters (22 inches), and the
Kasimovian-Gzhelian gri en y Meganeura monyi was even larger, with a
wingspan of 628 millimeters (two feet) (Clapham and Kerr 2012). In the
Early Permian, some gri en ies were larger still, the largest being the
Asselian-Artinskian Meganeuropsis permiana with a wingspan up to 720
millimeters (28 inches) (Grimaldi and Engel 2005; Clapham and Kerr 2012).
Giant gri en ies persisted into the beginning of the Middle Permian with
the presence of Arctotypus sp., with a wingspan of 489 millimeters (19
inches) (Clapham and Kerr 2012).
The largest dragon y in North America—again, like the “giant desert
hairy scorpion,” found in the Southwest—is the Walsingham’s Darner,
Anax walsinghami, with a wingspan of 150 millimeters (six inches) (Milne
and Milne 1980). The largest living dragon y, the Australian species
Petalura ingentissima, is slightly larger, with a wingspan of 168 millimeters
(6.6 inches) (Dorrington 2015). Similar to our comparison of modern and
Carboniferous scorpions, which were ve times larger, the wingspan of
the Carboniferous and Permian gri en ies was almost ve times that of
our largest living dragon ies. In fact, these gri en ies had a wingspan
more similar to that of a large modern bird, such as a magpie or a gull ( g.
4.3), than a living dragon y. Instead of being dive-bombed by hungry gulls
at the beach, eager to snatch a French fry from your hand, imagine
encountering similar-sized dragon ies!
FIGURE 4.3 Some Late Paleozoic dragon y-like insects, such as the Early Permian Meganeuropsis
permiana, had wingspans as wide as a modern seagull.

Source: Illustration by Mary Persis Williams.

Not only were there giant ying predatory insects in the Late
Carboniferous, but the ying herbivorous insects were gigantic as well.
These include the Moscovian palaeodictyopteran species Mazothairos
enormis (table 4.1) from Mazon Creek strata in Illinois, which had a
wingspan of up to 600 millimeters (two feet) and thus was aptly given the
species name “enormous.” These ancient insects did not eat energy-rich
animal protein or fat like a gri en y; they survived on tree sap. Yet these
Carboniferous sap-suckers still achieved gigantic sizes (Grimaldi and Engel
2005). Our largest living sap-sucking ying insect, the Malaysian cicada
species Pomponia imperatoria, has a wingspan of 200 millimeters (eight
inches)—the Carboniferous ying sap-suckers were three times larger.
Not all of the giant Carboniferous insects were larger than their
modern relatives. For example, the modern living cockroach Blaberus
giganteus of Central America and northern South America is fully 90
millimeters (3.5 inches) long, a size not exceeded by the large
Carboniferous roachoids, or cockroach-like insects. However, and in
summary, many of the Late Carboniferous and Early Permian arthropod
species had body sizes that were ve to seven times larger than their
modern-day counterparts, and an overall size range from three to as much
as 12 times larger.
What type of an Earth could have produced such huge arthropods?
How could such gigantic scorpions, millipedes, spiders, silver sh,
may ies, gri en ies, and cicada-like sap-suckers have existed? Gigantism
convergently evolved in multiple phylogenetic lineages of arthropods in
the Carboniferous: it appeared independently in the two clades of the
cheliceriform and the mandibulate arthropods, independently in the
myriapod and pancrustacean clades within the mandibulates, and
independently within multiple separate insect clades (table 4.1). To make
matters even more mysterious, there were other giant animals on the
Earth in this interval of geologic time, as we will see in the next section.

Giant Vertebrates

The giant Carboniferous arthropods were not alone—many of the Late

Carboniferous tetrapod vertebrates were also gigantic (table 4.2), at least
in comparison to their Early Carboniferous ancestors. The majority of the
tetrapod species of the Early Carboniferous were small animals, usually
only about 300 millimeters (one foot) in total length. In contrast, many of
the newly evolved tetrapods of the Late Carboniferous were ten times
larger: they were often three meters (3,000 millimeters, or ten feet) long.
And, as with the insects, the attainment of large body sizes in the Late
Carboniferous tetrapods was independent of diet: both carnivores and
herbivores were big animals.

TABLE 4.2 Terrestrial vertebrate lineages of the Late Paleozoic Ice Age.

TETRAPODA (limbed vertebrates)

– basal tetrapods
– Neotetrapoda
– – BATRACHOMORPHA (ancestors of living amphibians)
– – – basal batrachomorphs (“temnospondyls”)
– – – – Eryopid giants
– – REPTILIOMORPHA (ancestors of amniote tetrapods)
– – – basal reptiliomorphs (“anthracosaurs”)
– – – Seymouriamorpha
– – – Diadectomorph giants
– – – AMNIOTA (amniote tetrapods, conquerors of the land)
– – – – SYNAPSIDA (ancient ancestors of mammals)
– – – – – basal synapsids (“pelycosaurs”)
– – – – – – Caseid, ophiacodont, edaphosaur, sphenacodont giants
– – – – – THERAPSIDA (closer ancestors of mammals)
– – – – – – Biarmosuchia (basal therapsids)
– – – – – – Dinocephalian giants
– – – – – – Dicynodontia
– – – – – – Gorgonopsia
– – – – – – Therocephalia
– – – – – – – CYNODONTIA (very close ancestors of mammals)
– – – – REPTILIA (ancestors of living reptiles)
– – – – – PARAREPTILIA
– – – – – – Pareiasaur giants
– – – – – EUREPTILIA
– – – – – – Captorhinid giants
– – – – – – DIAPSIDA (ancestors of living lepidosaurs, archosaurs, and dinosaurs—the birds)

Source: Phylogenetic classi caton modi ed from Benton (2015).

Note: Taxa containing giants are in italics; older paraphyletic tetrapod group names are in
quotation marks; major clades are in capitals.

Vertebrates began the invasion of land in earnest in the Visean Age of

the Early Carboniferous, following the catastrophe of the end-Devonian
tetrapod extinctions and the long Tournaisian Gap (table 1.7) (McGhee
2013). The rst of the batrachomorphs, the ancestors of modern-day
amphibians, and of the reptiliomorphs, the ancestors of modern-day
amniote animals like mammals, reptiles, and birds, appear in the fossil
record at this time (table 4.2).4 The basal batrachomorphs are represented
in the Visean by a single species, Balanerpeton woodi, from East Kirkton in
Scotland. Individuals of this species were small, averaging about 200
millimeters (eight inches) long, and looked much like salamanders (Steyer
2012). Later, in the Early Permian, the batrachomorphs would evolve the
giant Eryops megacephalus—its species name means “very large headed”—
which, at 2,000 millimeters (6.6 feet) long, was ten times bigger than
Balanerpeton woodi. Eryops megacephalus looked nothing like a harmless
salamander—it had a massive body with thick legs ( g. 4.4), and its large
head had a mouth full of sharp teeth including extra fang teeth located in
the palate in the roof of its mouth—it was a deadly predator (Steyer 2012).

FIGURE 4.4 Some Late Paleozoic salamander-like amphibians, such as the Early Permian Eryops
megacephalus, were two meters (6.6 feet) long.

Source: Illustration by Mary Persis Williams.

The reptiliomorph clade is represented by several di erent species in

the Visean Age, of which Silvanerpeton miripedes, Eldeceeon rolfei, and
Westlothiana lizziae are found in the East Kirkton strata along with the
batrachomorph species Balanerpeton woodi. These reptiliomorphs were
also small, with body lengths, from snout to pelvis, of only about 200
millimeters (eight inches). The tail vertebrae are missing from the fossils,
so the precise length of their tails is unknown, but the total length of
these early reptiliomorphs was probably only around 300 millimeters (one
foot). Later, in the Late Carboniferous, the diadectomorph reptiliomorphs
would evolve the giant Diadectes maximus—its species name means “the
greatest”—that, as with the batrachomorph lineage, was ten times larger
than the early reptiliomorphs. Rather than 300 millimeters (one foot),
Diadectes maximus was fully 3,000 millimeters (ten feet) long ( g. 4.5).
Diadectes maximus is also very interesting in that it is the oldest known
fossil tetrapod species to have evolved herbivory—it was a plant-eater,
rather than a carnivore (Steyer 2012).

FIGURE 4.5 Some Late Paleozoic plant-eating reptile-like animals, such as the Late Carboniferous
Diadectes maximus, were three meters (ten feet) long.

Source: Illustration by Mary Persis Williams.

It is from the early reptiliomorphs that the two amniote clades evolved
—the synapsids, ancestors of modern-day mammals, and the reptilians,
ancestors of modern-day reptiles and birds (table 4.2). Within our own
phylogenetic lineage, the synapsids, giant predatory ophiacodonts began
to appear in the Late Carboniferous and Early Permian: Ophiacodon
uniformis was 2.5 meters (eight feet) long. In this same interval of time,
giant edaphosaurs and sphenacodonts began to appear: Edaphosaurus
cruciger was 3.5 meters (11.5 feet) long, and Dimetrodon grandis —its species
name means “great size”—was three meters (ten feet) long ( g. 4.6). The
ophiacodonts and sphenacodonts were large carnivores, but the
edaphosaurs were herbivorous; thus, as in the arthropod lineages, the
evolution of giant body sizes was not linked to diet. In the Middle
Permian, the herbivorous caseid synapsids (table 4.2) would evolve species
of Cotylorhynchus that were three meters (ten feet) long, and the
dinocephalian therapsids would produce the giant Moschops herbivores
that were fully ve meters (16.4 feet) long, the largest land vertebrates in
the Paleozoic (Benton 2015). The dinocephalians—the group name means
“terrible headed”—were bizarrely shaped with massive, barrel-chested
bodies but quite short, stocky legs ( g. 4.7). Given their massively built
skulls, it has been suggested that these animals butted their heads
together in ritual combat during the mating season, much like modern-
day goats and sheep (Benton 2015).

FIGURE 4.6 Some of our Late Paleozoic synapsid cousins, such as the carnivore Dimetrodon grandis,
were sail- nned giants more than three meters (ten feet) long.

Source: Illustration by Mary Persis Williams.

FIGURE 4.7 Some of our Late Paleozoic therapsid cousins, such as the Middle Permian Moschops
herbivores, were the largest land vertebrates in the Paleozoic, twice the size of a modern cow.

Source: Illustration by Mary Persis Williams.

Curiously, the other amniote clade, the reptilians, apparently did not
evolve giant individuals until the Late Permian. The parareptiles (table
4.2) produced the pareiasaurs with massive, elephant-like legs; muscular
humps on their backs associated with thick neck muscles; and odd knobby,
hornlike protrusions on their skulls. The African pareiasaur Bunostegos
akokanensis was three meters (ten feet) long (Steyer 2012). Also in the Late
Permian, the African captorhinid reptiles (table 4.2) produced giants like
Moradisaurus grandis—again, the species name means “great size”—which
were two meters (6.6 feet) long (Steyer 2012).

Giant Marine Invertebrates

Giant animals also appeared in marine and coastal ecosystems in the

Carboniferous. Not only did the giant 700-millimeter- (28-inch-) long
Pulmonoscorpius kirktonensis scorpions appear on land, but some of their
water-scorpion cousins (table 4.1) were over three times as large! In
coastal estuaries and rivers in the Devonian and Early Carboniferous, one
could encounter water scorpions up to 2.5 meters (eight feet) long, such as
the giant Jaekelopterus rhenaniae ( g. 4.8) (Braddy et al. 2007; Steyer 2012,
67; Palmer et al. 2012, 122).

FIGURE 4.8 Some Late Paleozoic water scorpions, such as the Devonian Jaekelopterus rhenaniae,
were over 2.5 meters (eight feet) long.

Source: Illustration by Mary Persis Williams.

Out in the open oceans of the world, the largest brachiopod shell sh
known in Earth history appeared—Gigantoproductus giganteus, an animal
aptly named the “gigantic giant Productus” (see table 4.3 for the
phylogenetic position of brachiopods). The shell of the Early
Carboniferous Gigantoproductus giganteus was 300 millimeters (one foot)
wide ( g. 4.9) (Newell 1949; Clarkson 1998; Qiao and Shen 2015). In
contrast, the shell of an average productid brachiopod in the Late
Devonian seas was only about the size of a quarter,5 fully 15 times smaller
than Gigantoproductus giganteus ( g. 4.9).

TABLE 4.3 Marine invertebrate lineages of the Late Paleozoic Ice Age.
EUKARYA (life with eukaryote cells)
– Bikonta (bi agellate single cells)
– – Chromoalveolata (“brown algae” and kin)
– – – Dinophyta (dino agellates)
– – Rhizaria
– – – Foraminifera
– – – – Fusulinid giants
– Unikonta (uni agellate single cells)
– – Choanozoa
– – – METAZOA (multicellular animals)
– – – – Cnidaria (diploblastic metazoans)
– – – – – Coral giants
– – – – BILATERIA (triploblastic metazoans with bilateral symmetry)
– – – – – PROTOSTOMIA (bilaterians with protostomous development)
– – – – – – Lophotrochozoa
– – – – – – – Lophophorata
– – – – – – – – Bryozoa
– – – – – – – – – Bryozoan giants
– – – – – – – – Brachiopoda
– – – – – – – – – Brachiopod giants
– – – – – – – Eutrochozoa
– – – – – – – – Mollusca
– – – – – – – – – Ammonoid giants
– – – – – – – – – Bivalve giants
– – – – – – Ecdysozoa (see table 4.1 for arthropod giants)
– – – – – DEUTEROSTOMIA (bilaterians with deuterostomous development)
– – – – – – Echinodermata
– – – – – – – Crinoid giants
– – – – – – – Echinoid giants
– – – – – – Chordata (see table 4.2 for vertebrate giants)

Source: Phylogenetic classi cation modi ed from Lecointre and Le Guyader (2006).
Note: Taxa containing giants are in italics; major clades are in capitals.

FIGURE 4.9 Some Late Paleozoic marine shell sh, such as the Early Carboniferous brachiopod
Gigantoproductus giganteus, had shells that were 300 millimeters (one foot) wide. In contrast,
average productid brachiopod shells in the Late Devonian were about the size of a quarter (lower
right).

Source: Modi ed and redrawn from Brunton, Lazarev, and Grant (2000).

Fossils of productid brachiopods are so numerous in Carboniferous

marine strata that the Carboniferous seas were often referred to as
“Productus seas” in older historical geology textbooks (Tasch 1973, 297),
and many of the Carboniferous species were giants. And gigantism did not
just occur in the ecological niche of the productid brachiopods, which
lived mostly buried and hidden in the sediment. The fully epifaunal
brachiopods—such as species of Punctospirifer—which lived exposed on the
sea oor (and thus more vulnerable to predator attack) exhibited vefold
increases in size in the Late Carboniferous to Middle Permian time span.
This size-increase trend is so notable that Norman Newell, a
paleontologist at the American Museum of Natural History, commented:
“The trend for increasing size during the late Paleozoic is well shown by a
large number of brachiopod genera. Most of these begin in the early
Pennsylvanian or late Mississippian and show progressive increase in
linear dimensions throughout the Pennsylvanian and Permian” (Newell
1949, 110–111).
The bryozoans— lter-feeding lophophorates (table 4.3) like the
brachiopod shell sh, but colonial rather than solitary—also experienced
size increases during the Late Carboniferous to Middle Permian time span.
Speci cally, the diameters of the zooecial openings in the colony skeletons
—a rough measure of the size of the actual tiny zooid individuals that lived
in the numerous branches of the colony—increased by 100 percent from
the Kasimovian to the Roadian (Newell 1949, 109). Gigantism was not
con ned to the clade of the lophophorates: in the lter-feeding
echinoderm sea lily (crinoid, table 4.3) genus Calceolispongia , the volume of
the basal plate of the calyx skeleton of the animal increased from two
cubic millimeters (0.08 cubic inches) to 800 cubic millimeters (2.6 cubic
feet) during the Artinskian Age in the Early Permian. If changes in the
volume of the basal plate of the calyx skeleton are taken as a rough
proportional measure of biomass-volume changes of the total animal,
then the biomass volume in these crinoid species increased by a factor of
400 during the Early Permian (Newell 1949, 112). Some mobile echinoderm
groups also experienced more than a tripling in size: species of the sea
urchin (echinoid, table 4.3) order Melonechinoida increased in size from a
diameter of 45 millimeters (1.8 inches) in the genus Palaeechinus to 155
millimeters (six inches) in the genus Melonechinus in the span of time from
the Tournaisian to Visean Ages in the Early Carboniferous (Newell 1949,
111–112).
 Predatory marine animals also experienced gigantism in the
Carboniferous. The sessile rugose coral species, such as Zaphrentis,
typically had horn- or conelike skeletons about 15 millimeters (0.6 inch)
high in the Devonian, whereas the Devonian-Carboniferous Canina species
had skeletons 100 millimeters (four inches) high and Siphonophyllia species
had even larger skeletons at 750 millimeters (30 inches) high—a ftyfold
increase in height from the Devonian Zaphrentis (Newell 1949, 108). The
swimming, squidlike ammonoid predators generally had phragmocones—
the section of the shell containing the main biomass of the animal—with a
maximum diameter of about 75 millimeters (three inches) in the Devonian
and Early Carboniferous, rarely ranging up to a diameter of 150
millimeters (six inches). In contrast, in the Permian, ammonoid species
with phragmocone diameters of 150 millimeters were common in several
di erent families of these predators, and species of Medliocottia and
Cyclolobus possessed phragmocones with diameters of 250 millimeters
(almost ten inches)—over three times larger than those in the Devonian
(Newell 1949, 114).
Probably the most astonishing examples of gigantism that occurred in
marine organisms during the Carboniferous and Permian are organisms
that were not even multicellular animals, but single-celled fusilinid
foraminifera (table 4.3). These single-celled organisms grew tiny,
lenticular coiled shells about 0.5 millimeters (0.02 inch) in axial diameter
—about the size of the period at the end of this sentence—in the
Serpukhovian Age of the Early Carboniferous. They then progressively
evolved numerous species and genera whose axial lengths increased to
the “acme in microscopic giants (Parafusulina kingorum Dunbar and
Skinner)” (Newell 1949, 106) in the Early to Middle Permian, a species that
had an axial length of 60 millimeters (2.4 inches)—an increase in size by a
factor of 120 from the Early Carboniferous. This organism—a single cell—
was almost three times the width of a quarter ( g. 4.10).

FIGURE 4.10 Some Late Paleozoic single-celled organisms, like the Early Permian marine
foraminifera Parafusulina kingorum, were 60 millimeters (2.4 inches) long and almost three times
the width of a quarter. In contrast, average foraminifers in the Early Carboniferous were about the
size of the period at the end of this sentence.

Source: Modi ed and redrawn from Hoskins, Inners, and Harper (1983).

A complicating factor in fusulinid gigantism is that these foraminifera

harbored symbiotic photosynthetic algae in their tissues—that is, they
harbored an extra source of food—and this trait may have contributed to
their ability to become giants. This trait is mirrored by another
astonishing example of invertebrate gigantism—the living Tridacna gigas
bivalve molluscs with shells over one meter (3.3 feet) in diameter. The
Permian gigantic bivalves (table 4.3) of the family Alatoconchidae
(Aljinovic et al. 2008; Isozaki and Aljinovic 2009) are also believed to have
harbored photosynthetic algae as symbionts in their tissues. We will
reconsider both the fusulinids and the alatoconchid bivalves in more
detail in the ecosystem evolution and animal gigantism section later in
the chapter.

THE CARBONIFEROUS OXYGEN PULSE AND ANIMAL GIGANTISM

The trigger for the evolution of animal gigantism in the Carboniferous had
to have been some globally pervasive change in the Earth’s environment
because that gigantism appeared independently, convergently, in the
separate phylogenetic lineages of both the arthropods (table 4.1) and the
vertebrates (table 4.2), and it occurred in the same interval of geologic
time for both of these separate phylogenetic lineages. Arthropods and
vertebrates are radically di erent types of animals—arthropods are
protostomous animals and vertebrates are deuterostomous—and you have
to go back in time all the way to the Neoproterozoic to nd a common
ancestor.6 That is, the lineages of the protostomous and deuterostomous
animals diverged back in the Neoproterozoic, meaning that the
arthropods and vertebrates are separated by a vast chasm of more than
660 million years of independent evolution. Yet both of these independent
lineages began to evolve giant animals in the Carboniferous—why?

Arthropods and Oxygen

The breathing system used by the insect species o ers the best clue to
solving the mystery of Carboniferous animal gigantism. Insects breathe
with a series of small tubes, the tracheae, that extend from pore-like
openings in their exoskeletons down into their body tissues. The insect
depends entirely upon di usion of oxygen in from these tubules (and
reverse di usion of carbon dioxide out of the tubules) to keep their tissues
alive. This system of breathing is strongly size limiting. Larger and larger
insects have larger and larger volumes of internal tissue, tissues that
depend upon the surface areas of the tiny tracheae for gas exchange. With
increasing size, the volume of any object increases as a cubic function of
linear dimension, whereas surface area increases only as a square
function. This constraint—the area-volume e ect—means that insects are
constrained to sizes small enough to have the proper ratio of internal
tissue volume to surface area of tracheal tubes (Harrison et al. 2010).
The area-volume problem is even more acute for ying insects— ying
is a highly energetic activity that requires a lot of oxygen. And larger,
heavier insects require even more energy—and oxygen—to y than
smaller insects. Thus how could a dragon y-like insect with a wingspan of
almost three-quarters of a meter ( g. 4.3) have existed in the Late
Paleozoic? It should have su ocated under its own mass, its tracheae
unable to aerate its large volume of internal tissues. But what if the
atmosphere had more oxygen in it than it presently does? An atmosphere
with a higher partial pressure of oxygen—more oxygen molecules
concentrated in a given volume of air—could potentially lift the constraint
of the area-volume e ect to higher ratios of tissue volume to tracheal
surface area, thus making possible the existence of much larger insects
(Dudley 1998; Harrison et al. 2010).
The mystery of giant-arthropod breathing deepens when it is revealed
that the gigantic Arthropleura millipedes ( g. 4.2) had no tracheae at all!
How could they breathe without tracheae? The zoologist Otto Kraus, of
the University of Hamburg, and the paleontologist Carsten Brauckmann,
of the Technical University of Clausthal, commented on this mystery:
“Despite the fact that many well-preserved fossils are known and
described, not even a trace of spiracles (stigmata) and tracheae has ever
been found…. The maximum length of the largest lepidopteran larvae is
approximately 15 cm. However, Arthropleura specimens exceeded this
dimension up to more than ten times. Provided that large arthropleurids
had really maintained tracheae, they should have been extremely stable
and strong tubes. They could hardly be overlooked” (Kraus and
Brauckmann 2003, 46–47).
One possibility is that the giant arthropleurid millipedes accomplished
gas exchange with the atmosphere directly across the surface area of their
bodies, and indeed Kraus and Brauckmann note: “In contrast to other very
large terrestrial arthropods … arthropleurids were apparently not
equipped with a heavily sclerotized or even mineralized exoskeleton”
(Kraus and Brauckmann 2003, 44–45; original emphasis). As anyone who
has stepped on a large beetle has experienced, an audible crunch is
produced as the beetle’s exoskeleton is crushed. The crunch sound is
produced because the beetle has a very rigid, sti exoskeleton whose
cuticular tissues contain a great deal of chitin (its tissues are “heavily
sclerotized”). Gas exchange across the rigid body surface area of a beetle is
minimal; it accomplishes gas exchange through its tracheae instead. The
exoskeleton of the giant Carboniferous millipedes was much softer, more
limber and exible, hence more permeable to gas exchange with the
atmosphere. However, even given this very unusual exoskeleton in the
arthropleurid millipedes, Kraus and Brauckmann argue that “gas
exchange simply via the body surface can certainly be excluded” (2003,
46). Instead, they argue that the giant millipedes must have been
semiaquatic, living mostly in water, and may have breathed through a
series of plates (“K plates”) along the bottom surface of their body, plates
that perhaps may have functioned somewhat like the gills in other aquatic
arthropods.
 Preserved fossil trackways made by giant arthropleurid millipedes on
land are well known and have even been given their own ichnotaxon
designation, or trace-fossil name—Diplichnites. Thus clearly the giant
Carboniferous millipedes spent a lot of time on land. Instead of being
semiaquatic and spending most of their time in water, what if the giant
millipedes spent most of their time on land (like normal millipedes) and
did indeed successfully breathe directly across their body surface areas
because there was more oxygen in the atmosphere than in our modern
world? The slowly crawling giant millipedes would not have required as
much oxygen as the highly energetic ying gri en ies, and gas exchange
across their body surface area may have been so e cient in a world with a
hyperoxic atmosphere that they did not even require tracheae in order to
breathe—hence these structures became vestigial and were later lost in
the evolutionary lineage of the arthropleurid millipedes.
It is di cult to see how such large insects could have existed without
having an atmosphere that contained much more oxygen than is present
in the modern atmosphere of the Earth. And indeed, it is exactly in this
period of time—the Late Carboniferous through the Middle Permian—that
the empirical models of the ancient atmosphere of the Earth, which we
considered in detail in chapter 3, predict that oxygen concentration levels
in the atmosphere began to exceed 30 percent and eventually reached a
high of around 35 percent ( g. 3.8). Imagine living in a world that has two-
thirds more oxygen in the atmosphere than in the air that we currently
are breathing.
The problem is, those same empirical models give di ering results on
exactly when oxygen began to exceed 30 percent in the atmosphere and
when the maximum oxygen content of the atmosphere was reached ( g.
3.8). Table 4.4 shows the known fossil occurrences of some of the giant
arthropod groups alongside the oxygen-content levels in the atmosphere
predicted by the Rock-Abundance model and the Geocarbsulf model. Note
that the known stratigraphic distribution of giant arthropod fossils
matches the oxygen predictions of the Rock-Abundance model much
better than those of the Geocarbsulf model: giant scorpions appear in the
Visean Age, and the Rock-Abundance model predicts that the Earth’s
atmosphere contained 25.8 percent oxygen in the Visean—4.8 percentage
points more oxygen than is present in the atmosphere of the Earth today.
Giant ying gri en ies appear in the fossil record in the Bashkirian Age,
when the Rock-Abundance model predicts that the oxygen content of the
atmosphere had risen to 32 percent, 11 percentage points more oxygen
than our present-day atmosphere of 21 percent. Giant millipedes appear
in the fossil record in the Moscovian Age, when the Rock-Abundance
model predicts that the oxygen content of the atmosphere of the Earth
had risen even further, to 32.8 percent.

TABLE 4.4 Known fossil occurrences of giant terrestrial arthropods in the Carboniferous and
Permian compared with the oxygen content in the Earth’s atmosphere.

Geologic Age Atmospheric O2 Giant Arthropod Fossil Occurrences

(1) (2)
Permian
Changhsingian 24.1% 23.6%
Wuchiapingian 26.5% 29.0%
Capitanian 28.3% 30.0%
Wordian 30.0% 31.0% ← Gri en ies
Roadian 30.6% 30.9% ← Gri en ies
Kungarian 33.0% 30.5% ← Gri en ies
Artinskian 35.0% 30.0% ← Gri en ies
Sakmarian 34.8% 29.3% ← Gri en ies
Asselian 34.5% 28.5% ← Gri en ies, millipedes
Carboniferous
Gzhelian 34.2% 27.9% ← Gri en ies, millipedes
Kasimovian 33.5% 26.5% ← Gri en ies, millipedes
Moscovian 32.8% 25.8% ← Gri en ies, millipedes, spiders, silver sh
Bashkirian 32.0% 25.0% ← Gri en ies
Serpukhovian 30.0% 23.0%
Visean 25.8% 18.8% ← Scorpions
Tournaisian 23.0% 17.0%

Source: Column (1) is the Rock-Abundance model (Berner et al. 2003); column (2) is the Geocarbsulf
model (Berner 2006). Oxygen data are extrapolated from gure 3.8 using the timescale of Gradstein
et al. (2012).
Note: Atmospheric oxygen contents of 25% or higher are underlined; those of 30% or higher are in
bold.

In contrast, the Geocarbsulf model (table 4.4) predicts that the

atmosphere of the Earth contained only 18.8 percent oxygen in the Visean
Age, 1.2 percentage points lower than that of the present Earth’s
atmosphere! How could a scorpion 700 millimeters (28 inches) long have
breathed in an atmosphere that contained less oxygen than our world
today, with its much smaller scorpions? The Geocarbsulf model predicts
that atmospheric oxygen had risen to only 25 percent in the Bashkirian—
just 4 percentage points higher than our present world—yet at this same
time giant, energetically ying gri en ies appear in the fossil record.
How could a dragon y-like insect with a wingspan of 720 millimeters (28
inches) have breathed in an atmosphere containing only 4 percentage
points more oxygen than in our world, where our large dragon ies have a
wingspan of only 150 millimeters (six inches). This mismatch between the
Geocarbsulf data and the fossil record data of giant arthropods (table 4.4)
in the Carboniferous supports Scott and Glasspool’s argument that we
previously considered in chapter 3: “Collectively, these data suggest levels
of O2 modeled for this interval rising from 17 percent to 23.5 percent [in
the Geocarbsulf model] are inappropriate and instead favor prior, higher
levels modeled at ≈23–31.5 percent [in the Rock-Abundance model], values
further supported by the occurrence of very large arthropods at this time”
(Scott and Glasspool 2006, 10863). Indeed, the Geocarbsulf model predicts
that atmospheric oxygen on the Earth did not reach the 30 percent level
until the Artinskian Age of the Early Permian (table 4.4), some 33 million
years after the appearance of the rst gri en ies!
We can also contrast the di ering predictions of the Rock-Abundance
and Geocarbsulf models by reversing our argument: if the appearance of
giant arthropods in the fossil record is linked to high oxygen levels in the
atmosphere, then the disappearance of giant arthropods in the fossil record
may be linked to lower oxygen levels. The giant gri en ies declined in
abundance during the Capitanian in the Middle Permian, were rare in the
Late Permian, and did not survive the end-Permian mass extinction (Nel
et al. 2009; Clapham and Kerr 2012). In the Rock-Abundance model,
oxygen levels in the atmosphere had dropped to 28.3 percent in the
Capitanian—almost as low as the 25.3 percent level predicted for the
Visean, a time during which no gri en ies were present on the Earth
(table 4.4). Indeed, the gri en ies had already begun to experience
di culties back in the late Early Permian, when one of the two
subfamilies of meganeurid gri en ies went extinct; the
paleoentomologist André Nel of the National Museum of Natural History
in Paris and his colleagues note that the “Meganeurinae apparently did
not survive after the Early Permian as the Middle to Late Permian
gri en ies are all Tupinae” (Nel et al. 2009, 115). In the Rock-Abundance
model, oxygen levels in the Early Permian rose from 34.5 percent to a
maximum of 35 percent, fell slightly to 33 percent and then dropped still
further to 30.6 percent in the Roadian Age at the beginning of the Middle
Permian—and the Meganeurinae became extinct.
 Although only meganeurid gri en ies of the subfamily Tupinae
survived in the Middle Permian, Nel and colleagues note that the
“diversity of gri en ies was very high during the Middle Permian, while
the more advanced groups of Protozygoptera and Triadophlebiomorpha
were also ourishing” (Nel et al. 2009, 115). The Protozygoptera and
Triadophlebiomorpha were smaller dragon y-like insects, and they did
not decline in abundance in the Late Paleozoic, nor did they go extinct in
the end-Permian mass extinction. Thus, the giant gri en ies went extinct
unlike the coexisting “contemporaneous Protozyogpera or
Triadophebiomorpha that were still ourishing during the Triassic” (Nel
et al. 2009, 115). Yet in the Geocarbsulf model, oxygen levels in the
atmosphere persisted as high as 30 percent in the Capitanian at the end of
the Middle Permian and as high as 29 percent at the beginning of the Late
Permian in the Wuchiapingian (table 4.4), a pattern that does not match
the demise of the giant gri en ies during this same interval of time. In
contrast, the Rock-Abundance model predicts a much lower oxygen level
of 28.3 percent in the Captianian and 26.5 percent in the Wuchiapingian,
which is more in accord with the demise of the giant gri en ies.

Vertebrates and Oxygen

In contrast to the Visean Age giant scorpions and Bashkirian Age giant
gri en ies (table 4.4), the rst of the giant vertebrates began to appear a
little later, in the Moscovian Age of the Late Carboniferous (table 4.5). Also
in contrast to the giant arthropods, numerous vertebrate giants persisted
into the late Middle Permian and Late Permian; examples include the
Capitanian ve-meter-long (16.4-foot-long) dinocephalian therapsid
Moschops capenis ( g. 4.7), the Late Permian three-meter-long (ten-foot-
long) pareiasaur parareptile Bunostegos akokanensis, and the two-meter-
long (6.6-foot-long) captorhinid reptile Moradisaurus grandis.

TABLE 4.5 Known fossil occurrences of giant terrestrial vertebrates in the Carboniferous and
Permian compared with the oxygen content in the Earth’s atmosphere.

Geologic Age Atmospheric O2 Giant Vertebrate Fossil Occurrences

(1) (2)
Permian
Changhsingian 24.1% 23.6% ← Pareiasaurs, captorhinids
Wuchiapingian 26.5% 29.0% ← Pareiasaurs, captorhinids
Capitanian 28.3% 30.0% ← Dinocephalians, caseids
Wordian 30.0% 31.0% ← Dinocephalians, caseids
Roadian 30.6% 30.9% ← Dinocephalians, caseids
Kungarian 33.0% 30.5% ← Batrachomorphs, edaphosaurs, sphenacodonts
Artinskian 35.0% 30.0% ← Batrachomorphs, edaphosaurs, sphenacodonts
Sakmarian 34.8% 29.3% ← Batrachomorphs, edaphosaurs, sphenacodonts
Asselian 34.5% 28.5% ← Batrachomorphs, edaphosaurs, sphenacodonts
Carboniferous
Gzhelian 34.2% 27.9% ← Diadectomorphs, ophiacodonts, edaphosaurs
Kasimovian 33.5% 26.5% ← Diadectomorphs, ophiacodonts
Moscovian 32.8% 25.8% ← Diadectomorphs, ophiacodonts
Bashkirian 32.0% 25.0%
Serpukhovian 30.0% 23.0%
Visean 25.8% 18.8%
Tournaisian 23.0% 17.0%

Source: Column (1) is the Rock-Abundance model (Berner et al. 2003); column (2) is the Geocarbsulf
model (Berner 2006). Oxygen data are extrapolated from gure 3.8 using the timescale of Gradstein
et al. (2012).
Note: Atmospheric oxygen contents of 25% or higher are underlined; those of 30% or higher are
also in bold.
 The e ect of a hyperoxic atmosphere on vertebrates is more subtle
than for arthropods with their tracheal breathing system (Graham et al.
1995, 1997). Increased body size is clearly one possible e ect, and, not only
in the Carboniferous, increased levels of oxygen in the atmosphere in the
early Cenozoic have been linked to the evolution of very large mammalian
vertebrates during this same interval of time (Falkowski et al. 2005). It is
surely no coincidence that very large vertebrates coexisted with very
large arthropods in the Carboniferous and Permian landscapes (tables 4.4
and 4.5).
Michael McKinney, a paleontologist at the University of Tennessee, has
conducted an extensive analysis of body-mass increases in the
ophiacodont, edaphosaur, and sphenacodont ( g. 4.6) synapsid clades
(table 4.2) during the Early Permian time span (McKinney 1990). During
the Asselian and early Sakmarian Ages, the largest sphenacodont in his
sample had a body mass of about 35 kilograms (77 pounds), the largest
ophiacodont had a body mass of about 25 kilograms (55 pounds), and the
largest edaphosaur had a body mass of about 15 kilograms (33 pounds). By
the late Sakmarian Age, the body mass of the largest edaphosaur in his
sample had increased to around 90 kilograms (198 pounds), and by the
Kungurian Age, at the end of the Early Permian, the body mass of the
largest edaphosaur was 330 kilograms (728 pounds). The body mass of the
largest ophiacodont increased to 120 kilograms (265 pounds) by the
Artinskian Age, and still further to 230 kilograms (507 pounds) by the
Kungurian. Finally, the body mass of the largest sphenacodont increased
to 120 kilograms (265 pounds) by the Artinskian and to 250 kilograms (551
pounds) by the Kungurian (McKinney 1990).
Thus during the 26-million-year span of the Early Permian, the body
masses of edaphosaurs increased by a factor of 22, of ophiacodonts by a
factor of 9.2, and of sphenacodonts by a factor of 7.1. These are incredible
size increases—an edaphosaur at the end of the Early Permian was fully 22
times larger than one at the beginning of the Early Permian! And, in the
Rock-Abundance model (table 4.5), oxygen levels in the Earth’s
atmosphere were well above 30 percent for this entire span of time. Note
again the mismatch between the Geocarbsulf model predictions and the
actual distribution of giant vertebrates in time: in the Geocarbsulf model,
oxygen did not even reach the 30 percent level in the atmosphere until
the middle of the Early Permian, in the Artinskin Age (table 4.5), long after
the dramatic size increases in the synapsid vertebrates were well under
way.
Giant batrachomorph amphibians, such as the two-meter-long (6.6-
foot-long) Eryops megacephalus ( g. 4.4) also existed during the Early
Permian. It is possible that these giant salamander-like amphibians
augmented their lung breathing by also breathing directly across their
skin surface area, much as the present-day lungless plethodontid
salamanders do. The plethodontids are small, only 100 to 200 millimeters
(four to eight inches) long, but in our present-day 21 percent oxygen
atmosphere they have totally lost their lungs and live solely by gas
exchange across the surface area of their naked skin. In an atmosphere
containing over 34 percent oxygen (Rock-Abundance model, table 4.5),
such an augmentative breathing mechanism may have enabled the Early
Permian amphibians to attain their giant size despite their very large
body volume relative to skin surface area.

Marine Invertebrates and Oxygen

In contrast to the temporal pattern of gigantism seen on land in the

vertebrates (table 4.5), the temporal distribution of giant marine
invertebrates (table 4.6) matches more closely the pattern seen in the
terrestrial invertebrates, the arthropods (table 4.4). Giant invertebrates
rst appear in the oceans in the Visean Age of the Early Carboniferous, as
do the giant scorpions on land ( g. 4.1), and giant marine invertebrates
are generally not present in the Late Permian, just as giant arthropods are
not. The terrestrial arthropods are all protostomous animals, however,
whereas in the oceans gigantism occurred in the clades of both the
protostomes (brachiopods, g. 4.9; bryozoans) and the deuterostomes
(echinoids, crinoids), in animals that are not bilaterians (corals), and even
in organisms that are not animals—the single-celled foraminifera
(fusulinids, g. 4.10).

TABLE 4.6 Known fossil occurrences of giant marine invertebrates in the Carboniferous and
Permian compared with the oxygen content in the Earth’s atmosphere.

Geologic Age Atmospheric O2 Giant Invertebrate Fossil Occurrences

(1) (2)
Permian
Changhsingian 24.1% 23.6%
Wuchiapingian 26.5% 29.0%
Capitanian 28.3% 30.0% ← Fusulinids, alatoconchid bivalves
Wordian 30.0% 31.0% ← Brachiopods, fusulinids
Roadian 30.6% 30.9% ← Brachiopods, fusulinids, bryozoans
Kungarian 33.0% 30.5% ← Brachiopods, fusulinids, bryozoans
Artinskian 35.0% 30.0% ← Brachiopods, fusulinids, bryozoans, crinoids
Sakmarian 34.8% 29.3% ← Brachiopods, fusulinids, bryozoans
Asselian 34.5% 28.5% ← Brachiopods, fusulinids, bryozoans
Carboniferous
Gzhelian 34.2% 27.9% ← Brachiopods, fusulinids, bryozoans
Kasimovian 33.5% 26.5% ← Brachiopods, fusulinids, bryozoans
Moscovian 32.8% 25.8% ← Brachiopods, fusulinids
Bashkirian 32.0% 25.0% ← Brachiopods, fusulinids
Serpukhovian 30.0% 23.0% ← Brachiopods, fusulinids, corals
Visean 25.8% 18.8% ← Brachiopods, echinoids, corals
Tournaisian 23.0% 17.0%

Source: Column (1) is the Rock-Abundance model (Berner et al. 2003); column (2) is the Geocarbsulf
model (Berner 2006). Oxygen data are extrapolated from gure 3.8 using the timescale of Gradstein
et al. (2012).
Note: Atmospheric oxygen contents of 25% or higher are underlined; those of 30% or higher are
also in bold.

Most marine invertebrates depend upon di usion-mediated

respiration; thus, seawater containing high concentrations of oxygen
should facilitate the same size-increase e ects in marine organisms as in
terrestrial ones (Graham et al. 1995). Seas that existed under a hyperoxic
atmosphere would have become oxygen rich via di usion and might also
have experienced little in the way of seasonal periods of anoxia that occur
in many oceanic regions today. However, it can be demonstrated that at
least some of the size increases seen in marine organisms that occurred in
the Carboniferous and Permian were also the result of increased nutrient
availability made possible by the evolution of photosynthetic symbioses, a
topic that will be explored further in the next section of the chapter.

ECOSYSTEM EVOLUTION AND ANIMAL GIGANTISM

Not all of the animal gigantism that appeared in the Carboniferous and
Permian may be directly attributable to the hyperoxic atmosphere that
existed on the Earth during that period of time. Other factors that have
been proposed to explain this animal gigantism include the evolution of
interspecies symbioses, of predator-prey interactions, and of competitive-
exclusion interactions.
 Many modern species of corals harbor photosynthetic algal species,
collectively called zooxanthellae,7 in their tissues in a complex
interspecies symbiosis. In the symbiosis, the corals metabolize the
hydrocarbons, or food, synthesized by the algae and the oxygen produced
by the algae as a waste product in photosynthesis. In turn, in their
metabolism the corals produce phosphate and nitrate wastes and carbon
dioxide that they recycle back to the algae, which use them to synthesize
more hydrocarbons via photosynthesis. In addition, the corals provide the
algae with protection from predation by grazing species of organisms by
holding them within their tissues.
Corals are sessile predators and generally survive by eating the prey
they capture. However, symbiotic coral species can use the additional
energy they receive from their symbionts to produce skeletal calcium
carbonate at accelerated rates and hence to grow the giant reef tracts
found in shallow water regions of the modern world, such as the Great
Barrier Reef in Australia. In addition, some modern species of bivalve
molluscs are also able to grow to gigantic sizes by utilizing the extra
energy provided by photosynthetic symbionts living within their tissues.
Individuals of the giant clam Tridacna gigas can have shells 1300
millimeters (4.3 feet) long, compared to an average large clam with a shell
about the size of your hand.
Given the occurrence of gigantism in some living coral cnidarians and
bivalve molluscs that are symbiotic, it is reasonable to propose that the
Carboniferous and Permian giant marine invertebrates may have been
able to achieve their large sizes because they possessed symbionts,
particularly the rugose corals (tables 4.3, 4.6). It is known that the giant
fusulinid foraminifera (tables 4.3, 4.6) that existed in the Carboniferous
and Permian contained symbiotic photosynthetic algae in their tissues:
many of these giant species evolved specialized skeletal morphologies to
contain and protect their symbionts.8 Similar to the shells of the living
giant lophotrochozoan mollusc (table 4.3) Tridacna gigas, the giant shells
of the Permian bivalves of the family Alatoconchidae (Aljinovic et al. 2008)
and the Carboniferous lophotrochozoan brachiopod Gigantoproductus
giganteus (table 4.3, g. 4.9) may indicate that algal symbioses were
widespread in the clade of the ancient lophotrochozoan animals.
However, no photosymbiotic species are known to have evolved in the
deuterostome echinoderm lineages of the crinoids and echinoids (table
4.3), the protostome lophotrochozoan lineage of the colonial bryozoans
(table 4.3), or the lophotrochozoan molluscan lineage of the ammonoid
predators (table 4.3), yet these invertebrate groups also evolved gigantism
in the Carboniferous (table 4.6). Thus, while some of the gigantism
exhibited by the Late Paleozoic marine invertebrates may be linked to the
evolution of interspecies symbioses, it is highly unlikely that mobile
predators like ammonites would ever have evolved photosymbiosis. In
addition, no terrestrial arthropod or vertebrate species has evolved
photosymbiosis, so gigantism in land animals cannot be attributed to that
mechanism.
On land, defense against predation is often invoked as a selective
mechanism that would favor the evolution of large body size. A modern
example is the combination of large body size and herd behavior in
African elephants, both of which are powerful deterrents to lion
predators. An adult elephant is simply too big, relative to the size of a lion,
and there are too many elephants together in the herd—lions know that
attacking them could lead to serious injury or death to the lion, not the
elephants. Thus lions concentrate on attacking small, juvenile elephants,
if they can be separated from the herd, or on attacking older, less agile and
formidable adults. The same ecological strategy apparently evolved in
ancient Mesozoic ecosystems with the evolution of giant body sizes in the
sauropod dinosaurs and the evolution of herd behavior as well—traits that
likely would have been deterrents to theropod dinosaur predators.9
Thus it could be argued that the evolution of giant body sizes in the
sap-sucking palaeodictyopteran winged insects in the Carboniferous was a
defensive response to the evolution of the giant meganisopteran gri en y
predators (table 4.1). How, then, can we ecologically explain the evolution
of gigantism in the gri en ies? The gri en ies had no ying predators to
defend themselves against—the rst aerial, gliding vertebrate predators
evolved in the Late Permian after the demise of the giant gri en ies (Nel
et al. 2009). In addition, modern-day aerial bird and bat predators
preferentially select large ying insects, not small ones, as prey (Clapham
and Kerr 2012). If the giant gri en ies were ecological predatory
equivalents of birds and bats, then it could be argued that predation was a
selective mechanism favoring the evolution of small body size, not large,
in the palaeodictyopteran prey species.
It has also been suggested that the absence of competition in a stable,
open niche was a selective mechanism favoring the evolution of
gigantism. In this scenario, the evolution of large body sizes is seen as the
consequence of animals’ evolving in optimal environmental conditions
with little or no competition or predation (Briggs 1985; Harrison et al.
2010; Dorrington 2015). That is, the evolution of gigantism in stable, open
niches could be argued to be the result of a form of ecological release, in
contrast to the circumstances of species evolving in habitats with
numerous competitors and/or predators (Blackburn and Gaston 1994).
Clearly competition for limited resources such as food, could be a selective
mechanism that would limit size. And, as discussed above, preferential
predation on large prey species could be a size-limiting selective
mechanism as well. Thus it could be argued that the giant gri en ies
evolved as an ecological consequence of the absence of competitors and
predators in the Carboniferous skies—the reverse of the defense-against-
predation argument for the evolution of gigantism. How, then, can we
ecologically explain the evolution of gigantism in the palaeodictyopteran
sap-suckers, species that did not evolve in an environment free of
predators?
In summary, the evolution of photosymbioses in marine invertebrate
organisms clearly provides an alternative to a hyperoxic atmosphere as a
physiological pathway to gigantism, although the two phenomena could
also operate in concert. It is more di cult to argue that the simultaneous
evolution of gigantism in numerous species of both terrestrial arthropods
and vertebrates in the Carboniferous (tables 4.4 and 4.5) was the result of
predator-prey and/or competitive-exclusion ecological interactions
within the separate evolutionary lineages of these two very di erent
groups of organisms. It is easier to argue that a common pervasive factor
—the presence of a hyperoxic atmosphere in the Carboniferous—was the
trigger for the evolution of gigantism in terrestrial arthropods and
vertebrates. Further support for this argument comes from the fact that in
another period of geologic time in which the Earth had a very oxygen-rich
atmosphere—the early Cenozoic—gigantism evolved in mammalian
vertebrates (Falkowski et al. 2005).
In this section of the chapter, we have considered physiological and
ecological hypotheses for the evolution of animal gigantism as
alternatives to the hypothesis that gigantism was triggered by hyperoxia.
In the next section, that argumentation will be reversed: we will consider
arguments that changes in the oxygen content of the Earth’s atmosphere
actually were the trigger for major physiological and ecological events in
animal evolution.
LATE PALEOZOIC OXYGEN AND ANIMAL EVOLUTIONARY EVENTS
The hyperoxic atmosphere of the Earth during the Carboniferous and
Permian may have been the trigger for far more evolutionary events than
just the appearance of animal gigantism. Je rey Graham, at the University
of California in San Diego, and his colleagues have proposed that “global
atmospheric hyperoxia possibly aided the vertebrate invasion of land”;
that “the Carboniferous diversi cation of both the insects and the
vertebrates correlates with the rise of atmospheric oxygen”; that
“hyperoxic air may have also been critically important in the evolution of
the cleidoic egg” in the rst amniotes; and that the Permian
“diversi cation of synapsids seems to be partially attributable to the
e ects of hyperoxia and a denser atmosphere on activity enhancing
specializations such as metabolic heat retention,” hence leading to the
evolution of endothermy (Graham et al. 1995, 119–120).
A comparison of some of the major events in the evolution of
vertebrates in the Carboniferous and Permian and the modeled oxygen
content in the Earth’s atmosphere during this span of time is given in
table 4.7. The invasion of land by vertebrates began in the Devonian with
the evolution of the rst tetrapods from the sarcopterygian shes, but
these early invaders were “aquatic tetrapods,” spending most of their time
in the rivers and lakes of the terrestrial realm. Only in the Visean Age of
the Early Carboniferous, following the Tournaisian Gap (table 1.7), did the
invasion of dry land by the vertebrates begin in earnest (McGhee 2013). It
is therefore of interest that the Rock-Abundance model predicts a rise in
oxygen in the Earth’s atmosphere to 25.8 percent in the Visean, 4.8
percentage points higher than in the present-day atmosphere, and that
giant scorpions appeared on land and giant invertebrates appeared in the
oceans at this same time (tables 4.4 and 4.6).
TABLE 4.7 Major events in the evolution of terrestrial vertebrates in the Carboniferous, Permian,
and Early Triassic compared with the oxygen content in the Earth’s atmosphere.

Geologic Age Atmospheric O2 Vertebrate Evolutionary Events

(1) (2)
Triassic
Olenekian 22.5% 20.0% ← Endothermic therapsids?
Induan 23.3% 21.8% ← Endothermic therapsids?
Permian
Changhsingian 24.1% 23.6% ← End-Permian mass extinction
Wuchiapingian 26.5% 29.0% ← Therapsid and reptilian diversi cations
Capitanian 28.3% 30.0% ← End-Capitanian extinction
Wordian 30.0% 31.0%
Roadian 30.6% 30.9% ← Therapsid diversi cation
Kungarian 33.0% 30.5% ← Olson’s extinction
Artinskian 35.0% 30.0%
Sakmarian 34.8% 29.3%
Asselian 34.5% 28.5%
Carboniferous
Gzhelian 34.2% 27.9%
Kasimovian 33.5% 26.5% ← Tetrapod diversi cation, ecological innovation
Moscovian 32.8% 25.8%
Bashkirian 32.0% 25.0% ← Synapsid and reptilian amniotes present
Serpukhovian 30.0% 23.0% ← Origin of amniotes?
Visean 25.8% 18.8% ← Invasion of land; origin of amniotes?
Tournaisian 23.0% 17.0%

Source: Column (1) is the Rock-Abundance model (Berner et al. 2003); column (2) is the Geocarbsulf
model (Berner 2006). Oxygen data are extrapolated from gure 3.8 using the timescale of Gradstein
et al. (2012).
Note: Atmospheric oxygen contents of 25% or higher are underlined; those of 30% or higher are
also in bold.
 Graham and colleagues argue that the elevated oxygen content in the
Earth’s atmosphere during the Visean helped the vertebrates to
successfully invade dry land in three ways. First, the hyperoxic
atmosphere helped elevate primitive lung e ectiveness in oxygen uptake.
Second, it helped the early vertebrates to lower their rate of desiccation in
breathing dry air by allowing them to take fewer breaths yet still obtain
su cient oxygen. Third, the hyperoxic atmosphere helped boost the
metabolic rates of the vertebrates, thus assisting them in overcoming the
force of gravity in walking on dry land. Dehydration is a problem that
land-dwelling animals still face today, as are the metabolic demands of
locomotion while enduring the constant pull of gravity—they point out
that, as water is 1,000 times more dense than air, an early tetrapod that
was essentially weightless in water would be 1,000 times heavier on dry
land (Graham et al. 1997). But note that this invasion-assistance
hypothesis works only for the atmospheric-oxygen predictions (table 4.7)
of the Rock-Abundance model for the Visean—the Geocarbsulf model
predicts a Visean atmosphere containing less oxygen than that of the
present-day Earth!
We know from the fossil record that the amniote vertebrates had
evolved by the Bashkirian Age at the beginning of the Late Carboniferous
and that both clades of the amniotes were present—the reptilian amniote
clade is represented by body fossils of the species Hylonomus lyelli and the
synapsid amniote clade by body fossils of the species Protoclepsydrops
haplous, both found in the famous Joggins strata with their preserved
fossil tree stumps and vertebrate skeletons in Nova Scotia, Canada (Benton
2015). However, as discussed in chapter 2, trace fossil evidence suggests
that more derived species of synapsid amniotes than Protoclepsydrops
haplous were also present in the Bashkirian, not just the basal one known
from body fossils. This evidence comes from the trackway ichnospecies
Dimetropus, found in Bashkirian-aged strata in Germany (Voigt and
Ganzelewski 2010). This trackway was made by quite a large animal—its
hind feet were around 140 millimeters (5.5 inches) long, and its forefeet
around 70 millimeters (2.8 inches) long. The trackway is most similar to
those produced by more-derived ophiacodontid, edaphosaurid, or
sphenacodontid synapsids, all of which were large animals. In contrast to
the large Dimetropus footprints, both the known basal synapsid and
reptilian amniotes were quite small—the animals were only 200
millimeters (eight inches) or so in length. If the ichnospecies Dimetropus
was produced by a highly derived synapsid amniote in the Bashkirian,
then the split between the synapsid and saurposid amniotes had to have
occurred even earlier than the Bashkirian—sometime in the Early, not the
Late, Carboniferous.
As discussed in chapter 2, an intriguing partial body fossil suggests that
the amniotes may have evolved in the Visean Age, a possibility that we
will explore in more detail here. This fossil is of the enigmatic species
Casineria kiddi, found in a sedimentary nodule that contains most of the
body skeleton—but unfortunately the head and tail are missing. The
animal was tiny: its back, from the base of its neck to its pelvic girdle, was
only 80 millimeters (three inches) long, making it the smallest known
Early Carboniferous tetrapod (Carroll 2009).
The vertebrae of Casineria kiddi are solidly ossi ed, and the animal had
long, slender, curved ribs. The well-preserved forelimb and forefoot of the
animal held the greatest surprise, as described by the Cambridge
University paleontologist Jennifer Clack: “The humerus is much more
slender than that of any other Early Carboniferous tetrapod, with an
obvious shaft and with the two ends set at di erent angles to each other
(known as torsion). The radius and ulna are also slender, with an
olecranon process on the ulna…. These features alone strongly suggest a
fully terrestrial animal.” In the forefoot, “the manus has ve slender
digits…. The last phalanx on each digit (the ungual) is noticeably curved,
and the whole arrangement suggests a hand capable of grasping. No other
Early Carboniferous tetrapod shows digits like this, but they are similar to
those found in Late Carboniferous early amniotes” (Clack 2002, 199). The
McGill University paleontologist Robert Carroll also observes: “Uniquely,
Casineria had curved terminal phalanges (unguals) forming claws, as in
many early amniotes” (Carroll 2009, 74). In summary, Clack notes: “It has
been suggested that the origin of amniotes is connected with an
evolutionary step involving small size … and here is a specimen that
accords well with that theory” (Clack 2012, 277). Because the head of the
animal is missing, it is impossible to prove that Casineria kiddi was indeed
the rst known amniote; clearly, however, it is the most derived species of
tetrapod yet known from the Visean Age.
Graham and colleagues argue that a hyperoxic atmosphere—whether
in the Visean, Serpukhovian, or Bashkirian—also assisted in the evolution
of the rst amniote vertebrates. A key step in the evolution of the
amniotes was the evolution of the amniote egg: “With specialized
membranes to prevent water loss (amnion), enhance respiratory gas
exchange (chorion), and collect waste products (allantois), the amniotic
egg (also termed the cleidoic egg), which appeared in the Carboniferous,
eliminated amphibian reliance upon aquatic egg laying and larval
development…. Accordingly, natural selection leading to the development
of the three amniote egg membranes enabled a further increase in egg
size. Moreover, the hyperoxic Carboniferous atmosphere would have
allowed the development of large amniote eggs by minimizing the ratio of
water loss to oxygen uptake” (Graham et al. 1997, 153). Note again that
this amniote-evolution-assistance hypothesis works only for the
atmospheric-oxygen predictions of the Rock-Abundance model for the
Visean Age (25.8 percent oxygen) and not the Geocarbsulf model (table
4.7), which predicts a Visean atmosphere containing less oxygen (18.8
percent) than in our present world (21 percent).
Following the evolution of amniotes (whether in the Visean or
Serpukhovian), a major diversi cation of amniotes occurred in the
Moscovian-Kasimovian interval of time. The rst fossils of the advanced
ophiacodontid synapsids are found in Moscovian strata, and fossils of the
rst advanced varanopid, edaphosaurid, and sphenacodontid ( g. 4.6)
synapsids are found in Kasimovian strata. In the reptilian amniote clade,
the rst fossils of the advanced diapsid, Petrolacosaurus kansensis, is found
in Kasimovian strata (McGhee 2013). Of particular note is the evolution of
herbivory in both the batrachomorph amphibian and the synapsid
amniote clades in the Late Carboniferous. The evolution of the ability to
e ciently use living plants as a food source was a key element in the
construction of the terrestrial ecosystem familiar to us today, with its
energy ow from living plants to herbivores to carnivores. Within the
batrachomorph clade the rst fully herbivorous vertebrates, the
diadectomorphs, are represented by the Moscovian species Limnostygis
relictus, and the herbivorous edaphosaurid synapsid species Ianthasaurus
hardestii and Xyrospondylus ecordi appeared in the following Kasimovian
Age (McGhee 2013).
Graham and colleagues further argue that “the Carboniferous
diversi cation of both the insects and vertebrates correlates with the rise
in atmospheric oxygen” (Graham et al. 1995, 119, g. 2). However,
establishing a causal link between increased speciation and evolutionary
innovation and the presence of a hyperoxic atmosphere is not as
straightforward as it is for the evolution of large body sizes or the
evolutionary transition from aquatic habitats to terrestrial habitats. Why
should higher oxygen levels trigger evolutionary innovation, such as the
evolution of herbivory, or increased speciation rate? Graham and
colleagues simply speculate that “increased oxygen availability may have
also fuelled the diversi cation and ecological radiation of late Palaeozoic
groups by acting as a substrate for the evolution of behavioural,
physiological and ecological adaptations, permitting greater exploitation
of aquatic habitats and the newly evolving terrestrial biosphere” (Graham
et al. 1995, 120). It is known that both clades of the amniotes were present
in the Bashkirian (table 4.7), when oxygen levels reached 32 percent in the
atmosphere (in the Rock-Abundance model). Was the increase in oxygen
in the atmosphere to 33.5 percent in the Kasimovian really the trigger for
amniote diversi cation and ecological innovation?
In contrast, the University of Bristol paleontologist Sarda Sahney and
colleagues argue that the Carboniferous diversi cation of tetrapods was
triggered by the ecological e ects of the Kasimovian crisis in the great
rainforests that we considered in chapter 3. They argue that the
Kasimovian crisis resulted in the fragmentation of the tropical rainforests
into numerous small habitat islands, and demonstrate that a major
increase in the number of endemic species of tetrapods occurred from the
Moscovian to the Kasimovian. Not only did the evolution of numerous
isolated pockets of endemic species occur, but that “rainforest collapse
was also accompanied by acquisition of new feeding strategies (predators,
herbivores), consistent with tetrapod adaptation to the e ects of habitat
fragmentation and resource restriction” (Sahney et al. 2010, 1079), and
“our data, which show elevated extinction rates, increased endemism, and
ecological diversi cation, apparently represent a classic community-
response to habitat fragmentation” (Sahney et al. 2010, 1081). They also
point out that the batrachomorph amphibians experienced high
extinction rates—losing nine families—and major ecological turnover
from the Moscovian to the Kasimovian, whereas the amniotes lost no
families to extinction, probably because they had an “ecologic advantage
in the widespread drylands that developed” in the Kasimovian (Sahney et
al. 2010, 1081).
Another animal evolutionary event during the Carboniferous and
Permian that may have had no relationship to the presence of a hyperoxic
atmosphere on Earth is Olson’s extinction10 in synapsid amniotes, which
occurred in the Early to Middle Permian transition (table 4.7). In this
vertebrate extinction event, the long-lived and dominant “pelycosaurs”
(table 4.2), the non-therapsid synapsids, were eliminated. Of the existing
six families of pelycosaurs present before the extinction, three went
extinct in the late Artinskian, a fourth perished in the early Kungurian,
and the nal two families succumbed in the Captanian (Kemp 2006). The
exact cause of the demise of the previously highly successful clades of
non-therapsid synapsids remains unknown. Oxygen levels in the
atmosphere are predicted to have fallen by 2 percent from the Artinskian
to the Kungurian in the Rock-Abundance model (table 4.7)—but could that
have triggered the extinction of six entire families of pelycosaurs? Olson’s
extinction will be considered in more detail in chapters 5 and 6, where the
possibility that this vertebrate extinction event was triggered by
paleoclimatic changes at the end of the Late Paleozoic Ice Age will be
explored.
Following a gap in the fossil record, named Olson’s Gap (Lucas and
Heckert 2001; Lucas 2004), a major diversi cation in therapsid synapsids
occurred in the Roadian (table 4.7). The basal therapsid group, the
Biarmosuchia, had evolved by the Early Permian and is represented by the
species Tetraceratops insignis. However, it was only in the Middle Permian
that the dinocephalian therapsids evolved—a diverse group both in terms
of numbers, over 40 genera, and in terms of ecology, as both carnivorous
and herbivorous dinocephalians existed (Benton 2015). The
dinocephalians also included some of the largest land animals to exist in
the Permian, such as the ve-meter-long (16.4-foot-long) dinocephalian
herbivore Moschops capenis ( g. 4.7). The dinocephalians appear in the
fossil record after the demise of the “pelycosaurs” (table 4.2), the non-
therapsid synapsids; thus their replacement of these earlier synapsid
herbivores and carnivores in the Permian appears to be a case of passive
ecological replacement rather than active, competitive ecological
replacement.
The dinocephalians themselves were driven to extinction at the end of
the Capitanian (table 4.7) and were replaced by the dicynodont and
gorgonopsian therapsids (table 4.2). The end-Capitanian extinction will be
considered in more detail in chapters 5 and 6, but like Olson’s extinction,
it appears not to have had any relationship to the oxygen content of the
Earth’s atmosphere. Oxygen levels fell only 1.7 percent from the Wordian
to the Capitanian (in the Rock-Abundance model, table 4.7), even less than
the 2 percent drop that preceded Olson’s extinction. Instead, the end-
Capitanian extinction is thought to have been triggered by the onset of
mantle-plume volcanism in China during this interval of time.11
The Late Permian dicynodonts were a diverse group of herbivores, over
60 genera, and the gorgonopsians comprised about 35 genera of
carnivores (Benton 2015). The dicynodonts and gorgonopsians were the
dominant herbivores and carnivores of the Late Permian, and together
they constituted about 80 percent of the terrestrial vertebrate diversity
(Erwin 1993). Finally, the therocephalian and cynodont therapsids (table
4.2) also appeared in the Late Permian—the cynodonts are particularly of
note because the rst mammals would evolve in the cynodont clade in the
Late Triassic, represented by the species Adelobasileus cromptoni.
Although the therapsids dominated Late Permian terrestrial
ecosystems, signi cant evolutionary diversi cation and innovation also
occurred in the clade of the reptilian amniotes. The pareiasaurs and
captorhinids appeared, some species of which were large animals (tables
4.2, 4.5). Also, the very rst gliding vertebrates evolved in the Early
Permian: the diapsid reptile Coelurosauravus jaekeli.12 The rst reptiles
capable of true powered ight, the pterosaurs, would evolve some 50
million years later in the Triassic (Steyer 2012).
Do any of these vertebrate evolutionary events have any relationship
with the amount of oxygen present in the atmosphere in the Permian?
Graham and colleagues argue that in the Permian the “synapsids
underwent a pronounced diversi cation, proceeding from the pelycosaurs
to the therapsids, a diverse assemblage of herbivores and carnivores….
The diversi cation of synapsids seems to be partially attributable to the
e ects of hyperoxia and a denser atmosphere on activity enhancing
specializations such as metabolic heat retention. The large ‘sails’ of
pelycosaurs such as Dimetrodon [ g. 4.6] are thought to have functioned in
heat transfer. The capacity to regulate heat gain and loss was an
important precursor to endothermy” (Graham et al. 1995, 120). The
distinctive “sails,” or crests, on the backs of both the carnivorous
sphenacodonts and herbivorous edaphosaurs are indeed thought to have
had a thermoregulatory function, allowing these large animals to shed
heat to the surrounding air when they became too hot or to absorb heat
from sunlight when the animals were cold. Thus, although the animals
were still ectotherms, they had a sophisticated anatomical and behavioral
mechanism for regulating their body temperature. If, in addition, the
sphenacodonts ( g. 4.6) and edaphosaurs had elevated metabolic rates—
triggered by the hyperoxic atmosphere of the Early Permian—the need for
such a thermoregulatory mechanism would have been more acute.
The non-therapsid synapsids were not the only vertebrates evidencing
elevated metabolic rates during the Permian. In the Early Permian, two
separate lineages of reptiles appear to have convergently evolved the
capability to stand up and locomote on their hind limbs only—they
evolved bipedalism.13 The case for Aphelosaurus lutevensis, a basal diapsid
reptile, is less clear—it may have been mostly arboreal, spending most of
its time climbing around in trees and only occasionally descending to the
forest oor to locomote on its hind limbs. The case for the anapsid
bolosaur Eudibamus cursoris is unequivocal—this animal was clearly
bipedal, capable not only of walking or running on its hind limbs only but
also of jumping, giving it the nickname “kangaroo reptile” among
vertebrate paleontologists (Steyer 2012, 134). Bipedal locomotion is
metabolically costly, and its appearance in the reptiles gives evidence for
the presence of elevated metabolic rates in both the synapsid and
reptilian amniote clades during the Early and Middle Permian.
The problem is, the dominant sphenacodonts and edaphosaurs went
extinct in the Kungurian and the therapsid diversi cation in the Roadian
occurred only after the disappearance of the non-therapsid synapsids—
that is, following Olson’s Gap. Thus there was not a continuous
diversi cation of synapsids “proceeding from the pelycosaurs to the
therapsids,” as argued by Graham and colleagues, and the Roadian
therapsid diversi cation appears to have been a separate evolutionary
event. It is not clear how this diversi cation and ecological innovation, in
the evolution of both herbivores and carnivores, in the dinocephalian
therapsids could have been driven by atmospheric oxygen. Only the giant
size of some of the dinocephalians, like the Moschops herbivores ( g. 4.7),
can clearly be attributed to the e ect of hyperoxia.
Still, in a later study, Graham and colleagues expanded their argument
that major evolutionary changes in the synapsid lineage were driven by
changes in the oxygen content of the Permian atmosphere, now adding
both hyperoxia and hypoxia to the equation:
 A signi cant event in tetrapod ecological physiology was the evolution
of endothermy … The large sails of Dimetrodon (from the Lower
Permian) and other sphenacodontids suggest the presence of a
complex behavioral repertoire revolving around the capacity to
regulate heat transfer…. The discovery of turbinate bones in the nasal
passages of therapsids indicates the presence of a water-conserving
mechanism linked to frequent ventilation and endothermy and
correspondingly suggests that the evolution of a “mammalian”
metabolic rate has occurred by the Late Permian (Hillenius, 1992;
1994). We suggest a two-part scenario for the evolution of a
mammalian-level of metabolism in the hyperoxic Carboniferous-
Permian biosphere. First … synapsids may have undergone natural
selection for a relatively high metabolic rate and also increased their
body size (thermal inertia)…. Increased metabolic expenditures …
would have been favored by an abundance of environmental oxygen.
Second, the presence of these metabolically specialized and hyperoxia
adapted organisms in a Permian environment characterized by
progressive atmospheric hypoxia could have intensi ed natural
selection on certain lineages for an increased ventilation frequency
(hence the appearance of turbinal bones in therapsids) and improved
cardiac e ciency for oxygen delivery to the tissues (i.e., separation of
systemic and pulmonary circulation) … this nding suggests a
mechanism through which the Permian oxygen decline could have
in uenced the evolution of a four chambered heart. (Graham et al.
1997, 158–160)

Graham and colleagues note that “only the crocodiles, among the reptiles,
and the birds and mammals have a four-chambered (i.e., completely
separated pulmonary and systemic circulations) heart” (Graham et al.
1997, 155), and of these animal groups both the birds and mammals are
endothermic. Thus Graham and colleagues argue that the onset of
hypoxia in the latest Permian triggered the evolution of the four-
chambered heart, and perhaps full endothermy, in the latest Permian
therapsids (presumably the cynodonts, the ancestors of the mammals,
although this is not explicitly stated in their argument).
The Rock-Abundance model does indeed predict progressive drops in
the oxygen content of the Earth’s atmosphere during the span of the Late
Permian into the Early Triassic—falling to 26.5 percent in the
Wuchiapingian (a 1.8 percentage point drop from the Capitanian), to 24.1
percent in the Changhsingian (a 2.4 percentage point drop from the
Wuchiapingian), to 23.3 percent in the Early Triassic Induan Age (a 0.8
percentage point drop from the Changhsingian), and nally to 22.5
percent in the Early Triassic Olenekian Age (another 0.8 percentage point
drop from the Induan) (table 4.7). A predicted Early Triassic atmosphere
containing 22.5 percent oxygen is still richer in oxygen than our present
world, and is certainly not hypoxic.
In the Geocarbsulf model, however, atmospheric oxygen is predicted to
have fallen precipitously from 29 percent in the early Wuchiapingian to
23.6 percent in the early Changhsingian—a 5.4 percentage point drop in
only 5.6 million years—with an additional 3.6 percentage point drop to a
hypoxic 20.0 percent by the beginning the Early Triassic Olenekian Age
(table 4.7). Yet Graham and colleagues do not attribute the end-Permian
mass extinction to the onset of hypoxia in either model, stating that the
“rate of oxygen decline was too gradual, however, to have been the
primary cause of the end-Permian extinction” (Graham et al. 1995, 117).
The problem with the hypoxia-onset model for triggering the
evolution of a four-chambered heart and the evolution of an endothermic
metabolic rate is the temporal pattern of evolution seen in the geologic
record for both the synapsid and reptilian lineages. As noted by Graham
and colleagues, only the crocodilian archosaurs, the avian dinosaurs
(birds), and the mammals possess four-chambered hearts today. However,
the rst known mammal, Adelobasileus cromptoni, appeared only in the
Late Triassic—long after the Late Permian–Early Triassic hypoxic interval.
The rst known bird, Archaeopteryx lithographica, appeared only in the Late
Jurassic, even further removed in time from the Late Permian–Early
Triassic hypoxic interval.
If the cynodont therapsids (table 4.2) had evolved a four-chambered
heart and an endothermic metabolism in the Late Permian or an Early
Triassic hypoxic interval, why was the evolution of the rst mammals—
animals that unequivocally possess both of these traits—delayed over 15
million years until the Late Triassic? Rather than the synapsids, a better
temporal case for the hypoxia-onset model could be made for the reptiles:
the crocodilian archosaurs (table 4.2) possess a four-chambered heart (but
not an endothermic metabolic rate), and the crocodilian archosaurs did
indeed evolve in the Early Triassic.
In summary, we have examined thus far many of the proposed
consequences of the evolution of the great Carboniferous rainforests, both
for the composition of the Earth’s atmosphere and for the evolution of
both trees and animals during the Late Paleozoic Ice Age. In the Permian,
the Late Paleozoic Ice Age would come to an end—with catastrophic
consequences for both plant and animal life. In the next chapter, we will
examine the prelude to disaster—the beginning of the end.
5 The End of the Late Paleozoic Ice Age

A massive expansion of ice occurred at the Pennsylvanian-Permian boundary,

and glaciation became bipolar at that time. Ice sheets are inferred to have
been at their maximum extent during the Asselian and early Sakmarian, after
which they decayed rapidly over much of Gondwana.
—Fielding, Frank, and Isbell (2008, 343)

THE EARLY PERMIAN DEEP FREEZE

Nine million years had passed since the end of the C4 continental
glaciation in eastern Gondwana (present-day Australia) and the massive
extinction in the great rainforests in the Kasimovian Age of the Late
Carboniferous (table 3.2). But the Late Paleozoic Ice Age was not nished
yet, and the dawn of the Permian saw the return of massive ice sheets
across the supercontinent Gondwana in the Southern Hemisphere, both in
the west and the east, and by the early Sakmarian Age in the Northern
Hemisphere as well (table 5.1). For the second time, the Late Paleozoic Ice
Age had become bipolar, as stated in the epigraph of this chapter. Or had
it? Once again, the bipolar controversy concerning the Late Paleozoic Ice
Age—which we rst encountered in the Carboniferous in chapter 2—
returns when we enter the Permian.

TABLE 5.1 Climatic events, area of rainforest coverage, and biological events during the Permian
phase of the Late Paleozoic Ice Age.
Age Geologic Time Climatic Events Extent of Extinction Events
(Ma) Rainforests
(1) (2) (3)
(Triassic) 252
CHANGHSINGIAN 253 ← End-Permian Mass
Extinction
254 140,000 km2
WUCHIAPINGIAN 255 183,000 km2
256 ICE 236,000 km2
257 ICE 289,000 km2
258 ICE 342,000 km2
259 ICE 395,000 km2
CAPITANIAN 260 ICE- 373,000 km2 ← Capitanian
P4 Extinction
261 350,000 km2
262 328,000 km2
263 ICE 305,000 km2
264 ICE 283,000 km2
265 ICE 261,000 km2
WORDIAN 266 ICE 239,000 km2
267 ICE 227,000 km2
268 ICE ICE? 216,000 km2
ROADIAN 269 ICE ICE? 194,000 km2
270 ICE ICE? 179,000 km2
271 ICE- ICE? 164,000 km2
P3
272 ICE? 150,000 km2
KUNGURIAN 273 127,000 km2
274 105,000 km2 ← Olson’s Extinction
275 105,000 km2
276 105,000 km2
277 105,000 km2
 278 105,000 km2
279 105,000 km2
ARTINSKIAN 280 332,000 km2
281 482,000 km2
282 633,000 km2
283 784,000 km2
284 ICE 1,011,000
km2
285 ICE 1,115,000
km2
286 ICE 1,220,000
km2
287 ICE 1,325,000
km2
288 ICE 1,430,000
km2
289 ICE 1,690,000
km2
290 ICE 1,640,000
km2
SAKMARIAN 291 ICE ICE? 1,590,000
km2
292 ICE- ICE? 1,590,000
P2 km2
293 ICE? 1,590,000
km2
294 ICE ICE ICE? 1,590,000
km2
295 ICE ICE ICE? 1,590,000
km2
ASSELIAN 296 ICE ICE 1,590,000
km2
297 ICE ICE 1,422,000
km2
298 ICE ICE- 1,255,000
 P1 km2
(Carboniferous) 299 1,087,000
km2
300 1,095,000
km2

Source: Gondwana data from Isbell et al. (2003), Frank et al. (2008), Fielding, Frank, Birgenheier et
al. (2008), and Metcalfe et al. (2015); Siberian data from Ustritsky (1973), Epshteyn (1981b),
Chumakov (1994), and Raymond and Metz (2004). Rainforest data from Cleal and Thomas (2005),
timescale modi ed from Gradstein et al. (2012).
Note: Climatic events column (1) is West and Central Gondwana and column (2) is East Gondwana
(Australia), both in the Southern Hemisphere; column (3) is Siberia (northeastern Russia) in the
Northern Hemisphere. Bold type indicates the existence of continental glaciers; normal type
indicates the presence of alpine glaciers, with glaciations in eastern Gondwana designated P1
through P4, from oldest to youngest.

The Early Permian chill is astounding: the Earth went from the balmy
Kasimovian-Gzhelian interpulse greenhouse interval in the Late
Carboniferous (table 3.2) to a globally frozen state with new ice caps in the
South in the Asselian—and perhaps in the North Pole in the Sakmarian
(table 5.1). At the beginning of the Permian, the ice sheets in western and
central Gondwana formed their third and nal phase of continental
glaciation in the Late Paleozoic Ice Age, and in eastern Gondwana the ice
sheets formed the rst glaciation phase, P1, of the Permian.
The climatic condition of the North Pole is much more uncertain. Early
stratigraphic data from the northeastern tip of Russia have been use to
argue for the presence of continental ice in this region, which at that time
was very close to the North Pole,1 during the Sakmarian, but other
workers have argued that these data were misdated and that they were in
fact of Bashkirian-Muscovian age in the Late Carboniferous (Ustritsky
1973; Epshteyn 1981a; Isbell et al. 2016). For this reason, I have placed
question marks on the Sakmarian northern hemisphere data in table 5.1.
 A brief warming phase then occurred in the middle Sakmarian Age in
the Southern Hemisphere, and the continental glaciers in Gondwana
retreated (table 5.2). Cooling resumed in the late Sakmarian, triggering
the P2 continental glacial phase in eastern Gondwana—but only in eastern
Gondwana, as western and central Gondwana remained free of continental
glaciers. Still, in eastern Gondwana the Early Permian P1 and P2
continental glacial pulses spanned a total of some 14 million years of ice
cover (table 5.1).

TABLE 5.2 Climatic events, area of rainforest coverage, and biological events during the transition
from the latest Early Permian (Kungurian Age) to the earliest Middle Permian (Roadian Age).

Age Geologic Time Climatic Events Extent of Rainforests Evolutionary Events

(Ma) rainforests:
(2) (3)
WORDIAN 266 ICE 239,000 km2
267 ICE 227,000 km2
268 ICE ICE? 216,000 km2
ROADIAN 269 ICE ICE? 194,000 km2 ← Therapsid
Radiation
270 ICE ICE? 179,000 km2 ← Olson’s Gap
271 ICE- ICE? 164,000 km2 ← Olson’s Gap
P3
272 ICE? 150,000 km2 ← Olson’s Gap
KUNGURIAN 273 127,000 km2 ← Olson’s Gap
274 105,000 km2 ← Olson’s
Extinction
275 105,000 km2
276 105,000 km2
277 105,000 km2
278 105,000 km2
279 105,000 km2
ARTINSKIAN 280 332,000 km2
 281 482,000 km2
282 633,000 km2
283 784,000 km2
284 ICE 1,011,000 km2
285 ICE 1,115,000 km2
286 ICE 1,220,000 km2
287 ICE 1,325,000 km2
288 ICE 1,430,000 km2
289 ICE 1,690,000 km2
290 ICE 1,640,000 km2
SAKMARIAN 291 ICE ICE? 1,590,000 km2
(pars)
292 ICE- ICE? 1,590,000 km2
P2

Note: Climatic events column (2) is East Gondwana (Australia) in the Southern Hemisphere, and
column (3) is Siberia (northeastern Russia) in the Northern Hemisphere; see table 5.1 for data
sources.

But change was in the air. By the end of the Sakmarian, the ice cap in
the Northern Hemisphere had melted away—if it existed!—and by the
middle to late Artinskian, the P2 continental glaciation phase in eastern
Gondwana had come to an end as well (table 5.1). The twilight of the Late
Paleozoic Ice Age had arrived.

THE MELTING OF THE ICE AGE

The Earth experienced a 12-million-year interpulse greenhouse period
spanning the period of time from the middle of the Artinskian Age to the
early Roadian Age.2 In the early Roadian Age, the planet once again cooled
—but to a lesser extent than in any previous cooling event in the Late
Paleozoic Ice Age. Two more phases of ice sheets formed in eastern
Gondwana, but only in the highlands and mountains (table 5.1). The
earlier phase of alpine glaciation, P3, was colder, and it has been argued
that the last glaciation in the Northern Hemisphere also occurred at this
time (Ustritsky 1973; Epshteyn 1981b; Chumakov 1994; Raymond and Metz
2004), but both events were brief on geologic timescales.
The last claim to continental glaciation in the Northern Hemisphere
(table 5.1), like all of the others, has been challenged by John Isbell and
colleagues. In this instance, they have actually shown that reportedly
glacial, continental sedimentary deposits in northeastern Russia were in
fact deepwater marine in nature and were formed by submarine gravity-
ow slumps and turbidity currents—at least in three stratigraphic sections
in the Okhotsk Basin (Isbell et al. 2016). Still, the very high latitude
position of the northeastern tip of Russia in the Late Permian—almost
positioned on the North Pole itself 3—makes it di cult to understand how
it could have remained ice free while the Southern Hemisphere was
glaciated (table 5.1). Isbell and colleagues note this climatic anomaly and
state that “our ndings constrain boundary conditions such that future
climate modeling can better determine factors that allowed Gondwana
glaciation to occur while inhibiting the development of land-based ice in
Northeastern Asia” (Isbell et al. 2016, 297).
In the Southern Hemisphere, the last ice sheets in the Late Paleozoic
Ice Age, the alpine glaciation phase P4 in eastern Gondwana, lasted for
some ve million years before they also melted away (table 5.1). The Late
Paleozoic Ice Age had ended.
Why did the Late Paleozoic Ice Age end? One factor that contributed to
its end was paleogeographic: the tectonic plate holding the giant
continent of Gondwana was nally moving o of the South Pole ( g. 1.4).
For over 110 million years, the South Pole—the coldest spot in the
Southern Hemisphere—had been located on Gondwana, rst in the west
and then moving slowly and erratically to the east as the tectonic plate
holding Gondwana shifted with time over the South Pole. By the end of
the Permian, the South Pole had come to be located on the edge of eastern
Gondwana—the southeast edge of present-day eastern Australia ( g. 1.4).
The open oceanic waters of Panthalassa4—the “all ocean” or “world
ocean” surrounding the single world continent Pangaea5—were located
just o shore to the south, and they would have ameliorated temperature
extremes on the neighboring continental margin just as oceanic waters do
on continental margins today. As anyone knows who lives on a coastline
near an ocean, summers are usually cooler and winters are usually
warmer than they are in the continental interior. Cities like Chicago,
located near large bodies of freshwater like Lake Michigan, also usually
experience a smaller range of temperatures from summer to winter than
cities like Tucson, surrounded only by land and rock. This e ect occurs
because water takes a long time to heat up and, once warm, takes a long
time to cool down. The opposite is true of bare soil and rock—they heat up
rapidly and become very hot in the summer, then just as rapidly lose heat
and become very cold in the winter.
In gure 1.4 we saw that the position of the South Pole moved from the
continental interior of Gondwana, present-day Antarctica, to the coastline
about 265 million years ago. During the last phase of the Late Paleozoic Ice
Age—the entire P4 glacial interval (table 5.1)—the South Pole was located
near the southeastern margin of Gondwana and hence near the oceanic
waters of Panthalassa. The entire P4 interval was also the least cold of the
Permian phases of the Late Paleozoic Ice Age, consisting of alpine glaciers
in the highlands of Gondwana, and the continental ice sheets were
unipolar, occurring only in the Southern Hemisphere.
 Another factor that contributed to the end of the Late Paleozoic Ice
Age was global climate change. The Earth became hotter and drier with
the passage of time in the Late Permian, and this global climatic change
also contributed to the death of the great lycophyte rainforests—a topic
that will be examined in detail in the next section of the chapter. The P4
glacial pulse spanned the period from the late Capitanian to the mid
Wuchiapingian (Metcalfe et al. 2015), and thus the beginning of the last
glacial pulse of the Late Paleozoic Ice Age coincided in time with the
Capitanian extinctions (table 5.1) (see Metcalfe et al. 2015, 75, g. 14). The
environmental trigger for the Capitanian biodiversity crisis will be
explored in detail in the next chapter, but here it will be noted that
massive volcanic eruptions also occurred in South China during the late
Capitanian. The geographic area of these eruptions is known today as the
Emeishan Large Igneous Province, and we have geochemical evidence that
enormous amounts of carbon dioxide and methane—both powerful
greenhouse gases—were vented into the Earth’s atmosphere in the late
Capitanian. With an atmosphere laden with heat-trapping greenhouse
gases, the entire Earth had to have become hotter, and not just in eastern
Gondwana where the last of the P4 glaciers eventually melted.

THE END OF THE GREAT RAINFORESTS

Not only did massive glaciers return on the Earth at the dawn of the
Permian, but the great lycophyte rainforests began to return as well. The
extent of the rainforests increased by 168,000 square kilometers (64,848
square miles), or about 15 percent, at the beginning of the Asselian Age—
from a low of 1,087,000 square kilometers (419,582 square miles) at the
close of the Carboniferous to 1,255,000 square kilometers (484,430 square
miles) when the ice sheets returned to Gondwana (table 5.1). The
rainforests continued to expand in the Early Permian, reaching a
maximum area of 1,690,000 square kilometers (652,340 square miles) in
the early Artinskian Age (table 5.1). Thus the maximum extent of the
Permian rainforests was comparable to the size of the rainforests present
at the Bashkirian/Moscovian boundary in the Late Carboniferous (table
3.2), some 23 million years earlier. However, the maximum size of the
Permian rainforests never quite reached the maximum of the
Carboniferous rainforests, being 70 percent of the rainforest cover present
on the Earth during the Moscovian Age (table 3.2).
As noted in chapter 3, the Permian rainforests were also more
geographically restricted than those of the Carboniferous. Only in the far
east, on the large islands of the Sino-Korean continental block (present-
day Korea and North China) and the Yangtze continental block (present-
day South China), did the lycophyte-dominated tropical mires continue to
exist ( g. 3.7). The equatorial regions of Pangaea to the west, once covered
by the lycophyte-dominated rainforests in the Carboniferous ( g. 3.7),
were now populated by plants that had previously been found mostly in
highland regions or in savannah environments in the lowlands (Cleal and
Thomas 2005).
The Permian lycophyte rainforests began to decline in the Artinskian,
coincident with the melting of the P2 continental glaciations (table 5.1).
From that point onward, the rainforests were never larger in area than
one million square kilometers, and they shrank in size by almost an entire
order of magnitude—from 1,011,000 to 105,000 square kilometers (390,346
to 40,530 square miles)—in just ve million years following the end of the
P2 continental glaciation (table 5.1). Eventually the lycophyte rainforests
vanished entirely on the island of the Yangtze continental block ( g. 3.7),
and only to the north, on the Sino-Korean block, did the rainforests
survive into the Kungurian. By the late Kungurian, the Sino-Korean
rainforests had shrunk to just over 100,000 square kilometers (38,600
square miles)—about the size of the modern state of South Korea. The
drastic reduction in the expanse of the Permian rainforests in the mid-
Artinksian to late Kungurian had a devastating e ect on the pelycosaurs,
the non-therapsid synapsids; we will examine the demise of these
tetrapods in detail in the next section of the chapter.
With the onset of the smaller P3 alpine glaciations in Gondwana, and
perhaps the return of ice in the Northern Hemisphere, the rainforests
recovered somewhat and expanded in size by almost a factor of four,
reaching an extent of 395,000 square kilometers (152,470 square miles) by
the dawn of the Wuchiapingian Age (table 5.1). This brief recovery can be
attributed largely to the reestablishment and spread of the rainforests on
the Yangtze continental block, as the Sino-Korean rainforests to the north
remained the same size as they had been in the Kungurian (Cleal and
Thomas 2005).
The rainforests began their nal decline following the end of the P4
alpine glaciation in eastern Gondwana (table 5.1). By the end of the
Wuchiapingian, the rainforests on the Sino-Korean continental block had
vanished—only those on the island of the Yangtze block survived (Cleal
and Thomas 2005). The paleobotanists Christopher Cleal and Barry
Thomas argue that the lycophyte-dominated rainforests shrank during
the Permian “as water-stress and increasing wild res made the habitat
unsuitable for the dominant lycophytes (Wang and Chen, 2001.) …
Towards the end of the Permian, even drier conditions caused the forests
to contract further, and increasing numbers are found of Mesophytic
[drier adapted] plants…. This trend towards drier and hotter conditions
may have been in part a consequence of the rain-shadow e ect from the
rising Northern Border Highlands to the north. However, Enos (1995)
favoured the progressive drift north of [North] China, so that by the
Changhsingian Age the area was outside of tropical latitudes. This is
compatible with the persistence through the end of the Permian of some
coal forests in South China, which were still in tropical latitudes” (Cleal
and Thomas 2005, 21). That is, Cleal and Thomas attribute the
paleoclimatic trend toward “drier and hotter conditions” in rainforest
regions toward the end of the Permian to tectonic e ects—both
mountainous uplift on the Sino-Korean continental block and movement
of the block to the north out of the equatorial region of the Earth.
But what if the trend toward hotter and drier conditions in the late
Permian was global, occurring across the entire planet, and not just in the
Sino-Korean region? In the next chapter we will examine the geological
evidence that massive volcanic eruptions occurred on the Yangtze
continental block during the late Capitanian, along with the geochemical
evidence that enormous amounts of powerful greenhouse gases were
vented into the Earth’s atmosphere beginning in the late Capitanian.
Regardless of the ultimate cause of the late Permian paleoclimatic
trend toward hotter and drier conditions, by the mid-Changhsingian Age
the Yangtze rainforests had also vanished. The great Late Paleozoic
rainforests of the Earth were gone, and their 70-million-year history was
at an end (tables 3.2 and 5.1).

THE PRELUDE TO DISASTER FOR PALEOZOIC LIFE

The close of the Early Permian Epoch saw a signi cant change in the
synapsid amniote faunas of the world. Most of the long-lived and
numerically dominant pelycosaurs, the non-therapsid synapsids (table
4.2), died out about 274 million years ago in the late Kungurian Age (table
5.1). This event has been named Olson’s extinction, and it does not appear
to have been triggered by changes in the oxygen content of the Earth’s
atmosphere, as discussed in chapter 4.
The non-therapsids in the synapsid clade of amniote animals were
ecologically replaced by the more derived therapsids—particularly the
dinocephalian therapsids (table 4.2)—in the following late Roadian and
Wordian Ages of the Middle Permian. However, as discussed in chapter 4,
there exists a gap in geologic time spanning most of the Roadian Age—
Olson’s Gap (table 5.2)—between the last of the sphenacodonts in the late
Kungurian and the diversi cation of the dinocephalians in the late
Roadian–Wordian (Lucas 2004; Blieck 2011). Thus, the more derived
dinocephalians do not appear to have competitively displaced their older
synapsid relatives in a case of active ecological replacement. Rather, the
ecological replacement that did occur appears to have been passive—the
dinocephalians simply moved into vacant ecospace created by the demise
of the non-therapsid synapsids. The dinocephalians ourished in the
Middle Permian, producing over 40 genera of both carnivorous and
herbivorous animals, and some of the largest land animals to exist in the
entire Permian, as discussed in chapter 4.
This standard passive-ecological-replacement model for the non-
therapsid versus therapsid synapsids turnover has been questioned by
Tom Kemp, a paleontologist at the Oxford University Museum of Natural
History. He has proposed a more nuanced model of ecological replacement
for these two evolutionary faunas that contains both passive and active
phases. First, he argues that the non-therapsid pelycosaurs (table 4.2)
were adapted for life in the ever-warm, ever-wet, equatorial zone created
by Late Paleozoic Ice Age climatic conditions ( g. 5.1) and that they
experienced no selective pressures for lifestyle changes in these stable
environments. The areal extent of the great rainforests—and the zone
they inhabited—progressively shrank in the late Artinskian–early Roadian
interval between the end of the P2 glaciation and the start of the P3
glaciation (table 5.2), and it is in this same interval of time that four of the
six families of non-therapsids perished.6 Thus, the extinction of these non-
therapsid families was driven by changing climatic conditions and was
independent of any competition by the therapsids, which were to appear
in numbers only later in the late Roadian of the Middle Permian (table
5.2).

FIGURE 5.1 Kemp’s paleoecological model for the evolution of the therapsid synapsids from the
non-therapsid pelycosaurs during the Early to Middle Permian time interval; see text for
discussion.

Source: From Kemp (2006), reprinted with permission.

Second, Kemp argues that one family of the Early Permian pelycosaurs,
the predaceous sphenacodontids ( gure 4.6), gave rise to the earliest
therapsids—but not in the ever-warm, ever-wet equatorial zone. Instead,
this evolutionary event took place in cooler and seasonally arid regions of
the Earth—geographic regions that expanded in area during the P2–P3
interpulse greenhouse period. This seasonally arid, savanna-like
environment Kemp labeled “tropical summer wet” ( g. 5.1), noting that
not only did the rst therapsids evolve there but the two remaining
families of the non-therapsids, the Caseidae and Varanopidae, also
migrated there as the equatorial ever-warm, ever-wet region shrank.
Third, Kemp argues that the early therapsids invaded even harsher
environments—the cool-temperate/winter-wet zone—in the Middle
Permian ( g. 5.1). Previously these higher-latitude regions were isolated
from equatorial regions by the temperate-zone desert bands that formed
in the Late Paleozoic Ice Age. As sea level fell with the return of expanding
ice sheets of the P3 glaciation (table 5.2), the early therapsids were able to
migrate north along the exposed east coastline of Laurussia and enter the
higher-latitude regions. Kemp argued that the key to their spectacular
success in this harsh region was the suite of morphological traits they had
evolved in dealing with the less-harsh conditions of the cool and
seasonally arid regions in which they had evolved. These traits included
much higher metabolic rates and sustained activity levels, faster growth,
and—most important—the ability to regulate their body temperatures to a
greater degree than any previous synapsid group (as discussed in chapter
4). The therapsids rapidly diversi ed in this new environment, producing
some nine new clades of both carnivores and herbivores (Kemp 2006).
Finally, Kemp notes that the fossil record shows that the last of the
non-therapsids, the caseids and the varanopids, also managed to invade
the cool temperate region and coexisted with the newly evolved
therapsids for a few more million years ( g. 5.1). He argues that the last
caseids and varanopids were eventually competitively displaced by the
more-advanced therapsids; thus, the nal extinction of the last of the
non-therapsid pelycosaurs was the result of active ecological replacement,
not passive like the initial extinction of four entire families of pelycosaurs
in the Early Permian.
Whatever its ultimate cause, Olson’s extinction triggered the loss of
some two-thirds of the biodiversity of terrestrial vertebrates and thus was
not a trivial event (Sahney and Benton 2008; Blieck 2011). Other clades of
land animals—not just the synapsid amniotes—were also a ected by
contraction of the ever-warm, ever-wet, equatorial zone that occurred in
the late Artinskian–early Roadian interval of the Early Permian. Ecological
diversity was also lost: post-Olson’s-extinction vertebrate communities
possessed a lower number of guilds than in any other period of Permian
history, being comprised almost entirely of piscivorous carnivores and
browsing herbivores (Sahney and Benton 2008).
But what about Olson’s Gap—a hiatus in the fossil record that the
University of Lille paleontologist Alain Blieck describes as being of global
extent and having a duration equivalent to most of Roadian time (Blieck
2011, 207)? This gap in the fossil record occurs in phase 2 of Kemp’s
ecological replacement model ( g. 5.1)—namely, the period of time in
which the rst therapsids are modeled as having evolved from— some of
the last sphenacodontids, an event that took place in seasonally arid,
savanna-like environments only marginal to the usual geographic range of
the sphenacodontids. The oldest known therapsid species is the basal
biarmosuchian Tetraceratops insignis (table 4.2) found in Early Permian
strata in Texas in North America (Benton 2005, 2015; McGhee 2013), yet
the fossil appearance of the oldest Middle Permian therapsid species is far
away to the east in Russia (Kemp 2006). Olson’s Gap may be a
preservational artifact; that is, the late Kungurian and early Roadian
newly evolved therapsid species may have existed in such small
population sizes that they had a very low probability of preservation in
the fossil record.
 On the other hand, Olson’s Gap might be a real ecological-evolutionary
phenomenon and not a preservational artifact. The University of Bristol
paleontologists Sarda Sahney and Michael Benton argue that “Olson’s
extinction was a dramatic extinction ‘trough’ that is a prolonged period of
very low diversity after a long and sustained diversity rise and probably
the result of prolonged environmental stress” (Sahney and Benton 2008,
761). That is, the low diversity of fossils found in the strata during the
period of Olson’s Gap did not result from a failure of representation of the
actual diversity of species because of small population sizes of those
species, and hence a low probability of their begin preserved in the fossil
record; rather, the low diversity of fossils actually records the low
diversity of species present on the Earth during the Olson’s Gap period of
time.
In contrast to extinction and biodiversity loss on land at the end of the
Early Permian, life in the oceans was rebounding from the long period of
evolutionary stagnation triggered by the Serpukhovian crisis that we
examined in detail in chapter 2. Starting in the early Sakmarian Age
(Stanley 2007), species diversity in marine ecosystems steadily increased
through the remainder of the Early Permian and through the Middle
Permian up to the middle of the Capitanian Age. That is, while on land the
glaciers slowly melted away and the great equatorial rainforests
progressively shrank to critical minimum levels during the late Early and
early Middle Permian (table 5.2), in the oceans normal evolutionary
turnover rates returned, diverse ecological specialist species re-evolved,
and rising sea levels created new shallow-water habitats on the continents
that were invaded by marine life around the Earth.
At the beginning of the Capitanian Age, life was good not only in the
oceans but also on land as therapsid species continued to diversify in the
cool-temperate zones of the Earth. Then the Earth began to heat up more
rapidly than ever seen in any of the interpulse greenhouse periods of the
Late Paleozoic Ice Age. In the oceans, dead regions of oxygen-depleted
waters formed and marine species began to go extinct around the world.
On land, arid desert regions began to expand, driving plant and animal
species to extinction. Even in the wet regions of the Earth, plant and
animal species mysteriously began to go extinct. The Capitanian crisis had
begun.
My colleagues Matthew Clapham, Peter Sheehan, Dave Bottjer, and
Mary Droser and I have demonstrated that global ecosystems of the world
began to collapse during the Capitanian Age, triggering the fth-most-
severe ecological disruption in the Phanerozoic Eon (table 1.3) (McGhee et
al. 2013). In the oceans, the ecological impact of the Capitanian
biodiversity crisis at the end of the Late Paleozoic Ice Age was greater
than that of the Serpukhovian crisis at the beginning of the “Big Chill”
phase of the Late Paleozoic Ice Age (table 2.2), but not as great as that of
the Late Devonian biodiversity crisis at the onset of the Late Paleozoic Ice
Age (table 1.3)—that is, if the Late Devonian crisis marked the onset of the
Late Paleozoic Ice Age, a controversy discussed in chapter 1. On land,
Sahney and Benton note that “the ecological impact of the Guadalupian
[Middle Permian] events is catastrophic; 8 (out of a possible 12) guilds are
lost from the Artinskian high of 10 guilds. These are recovered in the last
stages of the Permian before being devastated again by the end-Permian
event…. A dramatic change in diet type also occurs: proportions of
piscivores, insectivores, predators and browsers are thrown out of balance
during each extinction pulse” (Sahney and Benton 2008, 761).
Then, as if the Capitanian crisis were not enough, Peter Sheehan, Dave
Bottjer, Mary Droser, and I have determined that the ecological impact of
the end-Permian mass extinction—in both marine and terrestrial
ecosystems—was the worst ecological catastrophe in Earth history
(McGhee et al. 2004). The magnitude of the ecological disruption triggered
by the end-Permian mass extinction was so great that recovery was
impossible—the world of the Paleozoic Era came to an end.
What possible catastrophes could have happened on the Earth in the
Permian that would have had such a devastating e ect that they triggered
the fth-most-severe and rst-most-severe ecological crises in Earth
history (table 1.3) in the short time span of only seven million years (table
5.1)? To readers who subscribe to the Gaia hypothesis—that the
geochemical and biochemical cycles of the Earth are bu ered to protect
and nurture life—the answer to that question will be shocking, and that
horri c answer will be examined in detail in chapter 6.
6 The End of the Paleozoic World

The end of the Permian period is marked by global warming and the biggest
known mass extinction on Earth. The crisis is commonly attributed to the
formation of the Siberian Traps Large Igneous Province…. Heating of organic-
rich shale and petroleum bearing evaporites around sill intrusions led to
greenhouse gas and halocarbon generation in su cient volumes to cause
global warming and atmospheric ozone depletion…. The gases were released to
the end-Permian atmosphere partly through spectacular pipe structures with
kilometre-sized craters.
—Svensen et al. (2009, 490)

THE CAPITANIAN CRISIS

The Capitanian Age witnessed the opening salvo in the volcanic assault
that would bring an end to the Paleozoic world. Deep beneath South
China, a huge plume of hot magma was slowly rising to the Earth’s
surface. South China in the Capitanian was a large island, separate from
present-day North China, and was located in the tropics at the Earth’s
equator ( g. 6.1). The mantle plume would eventually intersect the
bottom of South China, burn its way up through the continental crust, and
pour an incredible amount of hot liquid lava across a huge expanse of the
Earth’s surface. In the process, it would create what we now call the
Emeishan “large igneous province,” or LIP in the vernacular of the
volcanologists.1

FIGURE 6.1 Paleogeography of the Late Permian world showing the locations of the Emeishan and
Siberian large igneous provinces, both of which erupted during the last nine million years of the
Permian.

Source: From Lithos, volume 79, pp. 475-489, by J. R. Ali, G. M. Thompson, M.-F. Zhou, and X. Song,
“Emeishan large igneous province, SW China,” copyright © 2005 Elsevier. Reprinted with
permission.

What is a LIP? A typical LIP is the product of mantle-plume volcanism

in which a plume of magma rises from deep in the Earth’s mantle and,
when it reaches the Earth’s surface, erupts almost unimaginable volumes
of basaltic lava and injects a huge amount of hot gases into the
atmosphere. A LIP eruption is thus characterized by (1) its large volume,
hundreds of thousands to millions of cubic kilometers of molten magma;
(2) its large areal extent, with lava covering hundreds of thousands to
millions of square kilometers of land or sea oor; (3) its rapid eruption,
consisting of numerous years- to decade-long volcanic pulses spread over
only one to ve million years—a very short period of time on geologic
timescales; and (4) the type of lava usually produced in the eruption—
basaltic.2 Basaltic lava is quite liquid and ows for considerable distances
before slowly freezing into solid rock, which partially explains why such
large areas of land are covered by lava in a “ ood basalt” LIP volcanic
event.3 The other explanation for the large size of LIPs is the fantastic
amount of lava that pours out onto the Earth’s surface through numerous
cracks and ssures produced in the Earth’s crust by the huge head of the
hot mantle plume located beneath it.
Vincent Courtillot, a geophysicist at the University of Paris, argues that
the initial head of a mantle plume could exceed 200 kilometers (124 miles)
in diameter when it initially forms at the boundary between the molten
outer core of the Earth and the lower mantle, located some 2,900
kilometers (1,800 miles) below the surface ( g. 6.2). As it slowly rises
through the mantle, he argues, the head of the plume may expand to
some 500 kilometers (310 miles) in diameter as it melts through mantle
rock and then further expand to some 2,000 kilometers (1,240 miles) in
diameter as it impacts and spreads out under the continental crust located
above it at the Earth’s surface (Courtillot 1999, 108).
FIGURE 6.2 Cross-section of the interior of the Earth showing a hot-spot LIP region (top center of
gure) produced by mantle plumes rising to the Earth’s surface from a depth of 2,900 kilometers
(1,800 miles).

Source: From Evolutionary Catastrophes, by Vincent Courtillot, copyright © 1999 Cambridge

University Press. Reprinted with permission.

A mantle plume can exist for a signi cant period of geologic time,
producing a relatively stable “hot spot” ( g. 6.2) near the surface of the
Earth with the gigantic plates of the Earth’s lithosphere slowly moving
across it. The Hawaiian Islands were produced in this fashion, with each
volcanic island forming sequentially as the mantle-plume hot spot burned
through the lithospheric plate moving across it. By dating the ages of the
Hawaiian Islands and associated submerged seamounts, it can be shown
that the mantle plume producing the island chain has been stable in its
position for at least 74 million years. The hot spot is presently located
beneath the largest island, Hawaii, which is the largest volcano on Earth—
although this is not apparent because most of the volcano is submerged
with only its tip protruding above the waters of the Paci c Ocean.
Measuring from its tip down to the sea oor, the volcano is a ten-
kilometer-high (6.2-mile-high) mass of basaltic lava produced by mantle-
plume volcanism (Courtillot 1999). The largest known volcanoes produced
by mantle-plume hot-spot eruptions are not on Earth, however, but on our
sister planet, Mars. The largest known volcano in the solar system is the
Martian Olympus Mons, which is a towering 21.3-kilometer-high (13.2-
mile-high) pile of basaltic lava ows ( g. 6.3). The Martian volcanoes
Ascraeus Mons, Pavonis Mons, Arsia Mons, and Elysium Mons are all taller
than ten kilometers (hence taller than Earth’s Hawaii volcano) (Croswell
2003), and they were all formed by mantle-plume volcanism. Mars is a
smaller planet with a smaller mantle and core, and the movement of the
plates of its crust were not as dynamic as those seen on Earth before the
core of Mars eventually cooled and the Martian magnetic eld collapsed.
As a result, most of the volcanoes on Mars were created by mantle plumes,
not, as on Earth, by the melting of plate boundaries in subduction zones
(for example, the “ring of re” distribution of volcanoes surrounding the
Paci c Ocean is created by subduction of Paci c Ocean plates). Because
plate motion ceased early in Martian history, its plates did not move
progressively across mantle-plume hot-spot locations. Almost all of the
lava that erupted at a hot spot on Mars accumulated into a single volcano
—rather than a chain of volcanoes like the Hawaiian Islands—resulting in
the even more gigantic size of the Martian LIP volcanoes.
FIGURE 6.3 The Martian volcano Olympus Mons, the largest known volcano in the solar system, a
towering 21.3-kilometer-high (13.2-mile-high) pile of basaltic lava ows produced by mantle-plume
hot-spot eruptions.

Source: Photograph courtesy of NASA.

Eleven LIP volcanic eruptions are known to have occurred on Earth in

the past 260 million years (Courtillot and Renne 2003; Saunders 2005;
Courtillot et al. 2010). No less than two of these massive eruptions
occurred during the short time span of the last nine million years of the
Permian—producing the Emeishan and the Siberian LIPs ( gure 6.1). Our
best radiometric data indicate that the Emeishan LIP began to erupt
between 257.6 ± 0.5 and 259.6 ± 0.5 million years ago, near the
Capitanian/Wuchiapingian boundary (Shellnutt 2013), whereas
biostratigraphic data indicate that the lava eruptions may have begun
earlier in the mid-Capitanian (Bond, Wignal, et al. 2010). The Emeishan LIP
may have continued to erupt sporadically for about 20 million years (Racki
and Wignall 2005; Shellnut 2013), up into the Triassic, but the majority of
the volcanic material in the LIP was emplaced in less than 1.5 million
years, essentially at the Capitanian/Wuchiapingian boundary (Shellnutt
2013). The Emeishan eruption is estimated to have produced 300,000 to
600,000 cubic kilometers (71,700 to 143,400 cubic miles) of lava, and its
main lava outcrop covers about 250,000 square kilometers (96,500 square
miles) of land today (Ali et al. 2005; Shellnut 2013). It is di cult to
estimate the original size of the Emeishan LIP because of extensive erosion
of the original lava ows and major tectonic fragmentation of the South
China continental block as it collided with and sutured together with the
North China and Indochina blocks in the early Mesozoic, plus the Indo-
Eurasian block collisions in the early Cenozoic (Shellnutt 2013). In our
modern world, outcrops of the original Emeishan LIP volcanics can be
scattered as much as 300 kilometers (186 miles) away from one another
across southwest China ( g. 6.4) (Wignall et al. 2009). Thus some estimates
put the original land area covered by the Emeishan ood basalts at over
two million square kilometers (772,000 square miles) (Racki and Wignall
2005).
FIGURE 6.4 Geologic map showing the exposed basalt outcrops (shaded regions) of the Emeishan
large igneous province in southwest China (inset map in lower right corner of the gure).

Source: From Lithos, volume 79, pp. 475–489, by J. R. Ali, G. M. Thompson, M.-F. Zhou, and X. Song,
“Emeishan Large Igneous Province, SW China,” copyright © 2005 Elsevier. Reprinted with
permission.

A volume of 300,000 to 600,000 cubic kilometers of molten lava is a

staggering amount—nothing like the Emeishan LIP eruption has ever
occurred within human history. The largest ood-basalt eruption in
recorded history is the Eldgjá eruption in Iceland, which occurred from
AD 934 to AD 940 and produced 19.6 cubic kilometers (4.7 cubic miles) of
basaltic lava (Thordarson and Self 2003). However, the best-documented
LIP eruption in history is the Laki eruption, also in Iceland, which started
on June 8, 1783, and continued for eight months until February 7, 1784
(Thordarson et al. 1996; Thordarson and Self 2003). The Laki LIP eruption
is the second largest in human history, producing 14.7 cubic kilometers
(3.5 cubic miles) of basaltic lava. This lava poured from 140 volcanic vents
in ssures that extended in a row some 27 kilometers (17 miles) long,
owing away from the ssures to covered some 580 square kilometers (224
square miles) of land in southeast Iceland (Thordarson and Self 2003;
Stone 2004).
The Laki LIP eruption produced a plume of gases that extended some
15 kilometers (9.3 miles) up into the atmosphere. The total eruption
ejected approximately 122 million tonnes (134 million tons) of sulfur
dioxide (SO2) into the atmosphere, which reacted with water vapor to
produce about 200 million tonnes (220 million tons) of sulfuric acid
(H2SO4) droplets. About 175 million tonnes (193 million tons) of sulfuric
acid precipitated out of the atmosphere as acid rain all across Europe, and
the remaining 25 million tonnes (28 million tons) remained in the upper
tropopause to lower stratosphere for over a year (Thordarson and Self
2003). The volcanic rifts also emitted huge clouds of chlorine and uorine
gas, which reacted with water vapor to produce about seven million
tonnes (7.7 tons) of hydrochloric acid (HCl) and 15 million tonnes (16.5
tons) of hydro uoric acid (HF) in the atmosphere, of which one million
tonnes (1.1 million tons) of hydro uoric acid fell as acid rain on the island
of Iceland alone (Thordarson and Self 2003; Stone 2004).
The e ect of the Laki LIP eruption on the human population of Iceland
was catastrophic: 10,000 people were killed—about 20 percent of the total
population of Iceland at the time (Stone 2004). Over 75 percent of the
grazing livestock animals were killed (Schmidt et al. 2011). Rain on the
island was so acid that it burned holes in the leaves of trees—most of the
trees and shrubs died and did not return for some three to ten years.
Cultivated grasses—food for livestock—withered. Many people died of
starvation in a famine that lasted for three years (Jackson 1982). Others
died from sulfur dioxide and sulfuric acid inhalation (Schmidt et al. 2011).
But many people and livestock animals died from chronic uorine
poisoning. The hydro uoric acid rain produced uoride salts on grasses
eaten by farm animals, poisoning the animals. The animals were eaten by
humans, and they became poisoned. Even the water was poisoned by
uorine—drinking water contained as much as 30 times the level of
uorine compounds permitted in modern drinking water. Merely drinking
a glass of water would make you sick, yet the islanders had no choice—
there was no other source of drinking water (Stone 2004).
The e ects of the Laki LIP eruption extended much farther away than
just Iceland. All across Europe, sunlight dimmed as the sulfuric aerosol
cloud spread east from the Icelandic ood basalts. This volcanic air
pollution was called the “dry fog” in English-speaking areas, as it looked
like fog but was not damp like fog; in German-speaking areas it was
described as the Höhenrauch , or smoke in the skies. Acid rain withered
summer wheat crops across Europe, and the green leaves on trees in the
middle of summer turned yellow and brown and fell to the ground as if it
were late autumn. The dry fog persisted for some ve to six months, and
winter came unusually early.4 The winter of 1783–1784 was the harshest
recorded in 250 years, and lemon crops far to the south in Italy were
destroyed by frost. In England alone, some 20,000 people died from
weather-related illnesses; in France the death rate was 25 percent higher
than normal (Stone 2004). If an equivalent eruption were to occur today,
computer simulations predict that some 142,000 people would die in
Europe from air-pollution e ects alone (Schmidt et al. 2011).
In North America, the winter of 1783–1784 was the most severe in
recorded history in the New World. In the edgling United States of
America, Benjamin Franklin proposed that the harsh winter was probably
caused by a volcanic eruption that he had heard about in Europe. In
Europe itself, the French naturalist Mourgue de Montredon attributed the
intense cold directly to the Laki eruption in Iceland (Thordarson and Self
2003),5 and Neale (2010) considers the environmental e ects of the Laki
eruption to have helped spark the French Revolution. The volcanic cloud
of sulfuric aerosols produced in the Laki LIP eruption eventually covered
the entire Northern Hemisphere from a latitude of about 35°N up to the
North Pole. The mean surface temperature of the Earth in this region
dropped by about −1.3°C (−2.3°F), and the colder-than-normal period
lasted from two to three years in the Northern Hemisphere (Thordarson
and Self 2003; Schmidt et al. 2012).
The mantle plume beneath the Icelandic hot spot is predicted to
produce a Laki-style ood-basalt eruption every 200 to 500 years
(Thordarson and Larsen 2007; Schmidt et al. 2012). It is a historical fact
that four such large-volume basaltic eruptions have occurred in the past
1,200 years, including the largest in recorded history (Eldgjá) and the
second largest (Laki), giving an average eruption frequency of one per 300
years (Thordarson and Self 2003). For a comparison with the Emeishan LIP
eruption in the Middle Permian, take the volume of lava produced in the
largest known eruption in history, 19.6 cubic kilometers, and for simplicity
in arithmetic round it up to 20 cubic kilometers (4.8 cubic miles). Now take
the shortest predicted Icelandic LIP-eruption frequency, one per 200
years, and divide 100,000 years by that. In this worst-case scenario for the
Icelandic mantle plume—the largest eruption occurring in the shortest
predicted period of time—some 500 eruptions, each producing 20 cubic
kilometers of lava, would occur in 100,000 years. Thus the total volume of
Icelandic hot-spot volcanism would be 10,000 cubic kilometers (2,400
cubic miles) of lava per 100,000 years, or 150,000 cubic kilometers (36,000
cubic miles) of lava per 1.5 million years.6 In contrast, the Emeishan LIP
produced a minimum of 300,000, and perhaps as much as 600,000, cubic
kilometers of lava in less than 1.5 million years. Thus the Emeishan
mantle-plume lava production was at least twice as large, and perhaps as
much as four times as large, as that of the Icelandic hot spot.
The Emeishan LIP eruption was coincident in time with the Capitanian
extinction (Zhou et al. 2002; Wignall et al. 2009; Bond, Wignall, et al. 2010).
It is thus logical to explore a causal relationship between the two events—
that is, to seek the cause of the biodiversity crisis in the environmental
e ects of this mantle-plume volcanic eruption. But rst let us examine
some of the biological e ects of the extinction. The loss in marine
biodiversity that occurred in the Capitanian event was once thought to be
the third largest in Phanerozoic history (Sepkoski 1996; Bambach et al.
2004), but it is now known that many of the extinctions that were once
thought to have occurred in the Capitanian in fact occurred later, in the
Changhsingian mass extinction (Clapham et al. 2009). Instead of a 36 to 47
percent loss of standing biodiversity, the true magnitude of the
biodiversity loss that occurred in the Capitanian extinction was more like
25 percent (McGhee et al. 2013). Ecologically, however, the Capitanian
extinction was the fth-most-severe event in the Phanerozoic (table 1.3)
(McGhee et al. 2013). Thus the Capitanian extinction is particularly
interesting because it is one of the bioevents in Earth history in which the
ecological impact of the event—relatively large—was markedly di erent
from the magnitude of the biodiversity loss—relatively small. Why did the
Capitanian extinction have such a large ecological impact? The answer
may perhaps be found in the cause of the event.
The kill scenario (Retallack and Krull 2006; Retallack et al. 2006;
Retallack and Jahren 2008; Bond, Hilton, et al. 2010; Clapham and Payne
2011; Payne and Clapham 2012) goes something like this: Back in the
Capitanian Age, the super-plume rising beneath South China rst caused
the land above to push up into a 1,000-kilometer-wide (620-mile-wide)
dome7 and then began to form huge cracks and ssures in the surface of
the Earth, accompanied with numerous earthquakes. Fluid basaltic lava
began to pour from the ssures, and explosive craters formed along the
ssures where hot gases vented into the atmosphere. The hot gases were
primarily sulfur dioxide, carbon dioxide (CO2), and methane (CH4). The
sulfur dioxide and carbon dioxide gases reacted quickly with water vapor
to produce droplets of sulfuric acid and of carbonic acid (H2CO3), and
oxidation of the methane produced even more carbon dioxide.8
In the lower atmosphere of the Earth, clouds of sulfuric and carbonic
acid precipitated out as acid rain on the land and in the neighboring ocean
waters. In the upper atmosphere, clouds of sulfuric acid droplets injected
by volcanic plumes re ected some of the incoming light from the sun and
caused the temperature of the Earth below to drop. Each volcanic pulse
triggered a cooling pulse in which the Earth remained cold for a period of
a decade or more. However, the venting of carbon dioxide and methane
from the eruption eventually triggered the reverse—global warming
caused by the greenhouse e ect of those gases. As the temperature of the
atmosphere and the oceans increased, methane frozen in ice in high-
latitude permafrost regions on land and buried in submarine sediments in
the oceans became increasingly unstable and began to dissolve, bubbling
even more methane up into the water and into the atmosphere. Methane
is a more potent greenhouse gas than carbon dioxide, and methane
releases triggered even steeper increases in the Earth’s temperature.
Ocean waters that had become acidi ed from sulfuric and carbonic
acid rain now became hot and stagnant, producing large “dead zone”
areas of anoxic water. With each passing year, the global climate became
warmer and the dead regions of oxygen-depleted ocean waters grew
larger and larger. Marine organisms began to die from acid poisoning
(acidosis), carbon-dioxide poisoning (hypercapnia), and asphyxiation
(hypoxia). On land, plants began to die from acid-rain poisoning.
Terrestrial areas became more arid with the increased heat of the
atmosphere; deserts began to expand and spread. Oxidation of methane in
the atmosphere caused oxygen levels to fall and carbon-dioxide levels to
rise.9 Land animals died in increasing numbers as a result of climatic stress
—from hyperthermia during the summer heat and from hypercapnia in
breathing the carbon-dioxide-laced air. Lower atmospheric oxygen levels
caused animals to retreat from the highlands of the Earth, and they began
to su er from uid buildup in and swelling of their lung tissues
(pulmonary edema). Herbivorous animals died from starvation as their
food, the land plants, died from either acid-rain poisoning or lack of water
in newly formed desert regions. Carnivorous animals died from starvation
as their food, the herbivores, died. The global ecosystem of the Capitanian
world began to collapse, and the fth-most-severe ecological disruption in
the Phanerozoic Eon was the result (McGhee et al. 2013).
As in our modern oceans, reef organisms were particularly sensitive to
the change to higher water temperatures and acid-water chemistry in the
Capitanian Age. The tropics of the Capitanian world contained many reefs
constructed by large demosponges, ancient rugose and tabulate corals,
and reef-building microbes (Weidlich 2002; Kiessling and Simpson 2011).
These reefs were destroyed in the Capitanian crisis, and, with the
exception of some small biostromes and bryozoan reefs, no signi cant
reef-building occurred until the recovery of reef ecosystems some ve to
seven million years later in the late Wuchiapingian and Changhsingian
(Fan et al. 1990; Flügel and Kiessling 2002; Weidlich 2002). Ecologically, the
Capitanian extinction triggered a global restructuring of reef ecosystems
from complex, multicellular-animal-dominated reef structures to
unicellular-microbe-dominated reefs (Flügel and Kiessling 2002).
The reef-building demosponges with their hypercalci ed mode of
growth (Kiessling and Simpson 2011) were particularly hard hit by the
acidi cation of the world’s oceans by the Emeishan eruption. Another
group of large, highly calci ed organisms that perished in the extinction
were the giant unicells of the photosymbiotic fusulinid foraminifera that
we considered in chapter 4. Acid waters impeded the growth of the
calcium-carbonate-rich skeletons of the large-bodied and morphologically
complex foraminiferal species that had specialized skeletal structures for
nurturing and protecting their photosymbiots.10 All of the large-bodied
foraminifera were driven to extinction, leaving as survivors only the
smaller and morphologically simpler species.11 Large-bodied bivalves with
thick calcareous shells that also hosted photosymbionts in their tissues
also went extinct.12 The simultaneous extinction of the hypercalci ed,
large-bodied, photosymbiont-hosting members of what the University of
Tokyo Yukio Isozaki and University of Zagreb Dunja Aljinović
paleontologists describe as the “Tropical Trio”—the “waagenophyllid
corals, verbeekinid foraminifers, and alatoconchid bivalves” (Isozaki and
Aljinović 2009)—is a diagnostic feature of the e ect of the Capitanian
crisis.
The Capitanian extinction also had a major ecological impact on the
ammonoids, which were actively swimming predators in the Capitanian
seas. The typical Paleozoic ammonoid faunas were completely replaced by
ammonoid faunas that were to become the typical Mesozoic forms—some
eight million years before the start of the Mesozoic (Leonova 2009)!
Ammonoids of the order Ceratitida rst appeared at the Early/Middle
Permian boundary, but by the Capitanian only ve genera of the order
Ceratitida existed, representing 18 percent of the ammonoid diversity.
Following the Capitanian crisis, the diversity of ceratite genera jumped to
55 percent in Wuchiapingian and further increased to 74 percent in the
Changhsingian ( g. 6.5) (Leonova 2009). The shift in abundance was just as
or even more abrupt than the shift in diversity that occurred between the
Capitanian and Wuchiapingian, coincident with the Capitanian extinction.
These ammonoid faunal changes are re ected in an ecological shift in
marine ecosystem structure from bottom-swimming (nektobenthic)
ammonoid faunas to open-ocean (pelagic) ammonoid faunas, an ecological
shift to a more Mesozoic-style ecosystem structure that we will consider
in more detail later in this chapter.
FIGURE 6.5 Occurrence counts for the three late Paleozoic ammonoid orders from data in the
Paleobiology Database (www.paleodb.org). An occurrence is the record of a species from a single
fossil collection, and is a proxy for abundance. The Capitanian biodiversity crisis is marked by the
vertical dotted line. Geologic timescale abbreviations: Capitan, Capitanian; Wuchiaping,
Wuchiapingian; Changhsing, Changhsingian.

Source: From McGhee et al. (2013).

The Capitanian extinction may have been more severe ecologically

than it was taxonomically because of the environmentally selective nature
of the event. For most animal groups, the physiological stresses induced
by the Emeishan LIP eruption may not have been su ciently intense to
drive major taxonomic extinctions. In contrast, taxa such as
hypercalci ed sponges, corals, and larger foraminifera su ered because
they had poor physiological bu ering and would have been particularly
susceptible to the e ects of ocean warming and acidi cation (Clapham
and Payne 2011; McGhee et al. 2013). Their elimination triggered major
ecological restructuring, particularly in reef ecosystems. Other typical
Paleozoic benthic organisms with less developed respiratory systems and
low metabolic rates—such as the bryozoan and brachiopod lophophorates
—su ered di erentially from hypercapnia and hypoxia, as opposed to the
more energetic gastropod and bivalve molluscs that were to become
dominant in the Mesozoic (Knoll et al. 1996; Retallack and Krull 2006).
The cause of the shift in dominance structure in ammonoid pelagic
ecosystems, from the typical Paleozoic mollusc orders of the Goniatitida
and Prolecanitida to the more typical Mesozoic order of the Ceratitida ( g.
6.5), is less clear because the biology of these extinct groups is uncertain.
It is of note, however, that the ceratite ammonoids also survived the end-
Permian mass extinction—also associated with a LIP eruption, as we will
see in the next section of the chapter—whereas the goniatites and
prolecanitids did not (McGhee et al. 2013).
Finally, the selective extinction of the large-bodied photosymbiont-
bearing foraminifera and bivalves during the Capitanian may not have
been driven entirely by acidosis. Another possibility is that all
photosynthetic organisms—including plants on land—su ered radiation
poisoning during the Capitanian extinction (Ross 1972; Bond, Wignall, et
al. 2010). In this alternative kill scenario, chlorine and uorine gases
emitted from the Emeishan LIP eruption seriously damaged the ozone
layer of the Earth’s atmosphere, permitting the ux of high-energy
ultraviolet radiation to greatly increase at the surface of the Earth. As yet,
it has not been demonstrated that enough chlorine or uorine gas was
released during the Emeishan eruption to disrupt the ozone layer, but we
will see below that the radiation-poisoning kill scenario does indeed
appear to apply to the end-Permian mass extinction.
The reader may have noticed a major di erence in the scenario
outlined above for the environmental e ects of the Emeishan LIP eruption
and those environmental e ects actually recorded in the 1783–1784 Laki
LIP eruption that we considered previously. The opening of huge ssures
in the Earth, the outpouring of ood basalts, the clouds of hot sulfur-
dioxide gas, the formation of sulfuric acid in the atmosphere, the e ect of
acid rain on land plants and animals, and short-term global cooling all
occurred in the historical Laki eruption and are also proposed to have
occurred in the Emeishan eruption. What is di erent is the major
presence of two additional gases in the Emeishan eruption—carbon dioxide
and methane—and the environmental e ects of the long-term global
warming produced by these greenhouse gases in the atmosphere of the
Capitanian world. The Laki LIP eruption produced in insigni cant amount
of carbon dioxide compared to the amount of sulfur dioxide it released
into the atmosphere (Self et al. 2005). And the Laki LIP is not unusual in
this respect—volcanic-generated carbon-dioxide release from the typical
basaltic magmas that produce LIPs usually is not large in comparison to
the amount of carbon dioxide already present in the atmosphere (Self et
al. 2005; Ganino and Arndt 2009).
Why was the Emeishan eruption so di erent from Laki—what evidence
do we have for the additional production of huge amounts of carbon
dioxide or methane in the Emeishan eruption some 360 million years ago?
There are two lines of evidence: (1) the composition of the sedimentary
rocks encountered by the molten magma of the Emeishan eruption on its
rise to the Earth’s surface and (2) the composition of the types of carbon
released into the Earth’s atmosphere and ocean waters during the time of
the eruption.
First, when the rising Emeishan super-plume eventually intersected
the bottom of South China and began to melt its way up through the
continental crust, it encountered thick layers of sedimentary rock that
were present above it in the Sichuan Basin—thick layers of reef limestone
(rocks rich in calcium carbonate; CaCO3), dolomite (rocks rich in calcium
magnesium carbonate; CaMg(CO3)2), and, in some areas, shales containing
petroleum deposits (Ganino and Arndt 2009). Clément Ganino and
Nicholas Arndt, geologists at the Université Joseph Fourier de Grenoble,
estimate that the Emeishan mantle-plume melting and heating—cooking
—of the Sichuan Basin limestone and dolomite strata released between
41.3 and 97.6 trillion tonnes (45.5 to 107.6 trillion tons) of carbon dioxide
into the Earth’s atmosphere. This is in addition to the much smaller 11.3
trillion tonnes (12.5 trillion tons) of carbon dioxide emitted directly from
the LIP magma itself; thus the total amount of carbon dioxide released
into the Earth’s atmosphere during the Emeishan eruption was between
52.6 and 108.9 trillion tonnes (58 to 120 trillion tons).13
In contrast, an Emeishan-sized ood-basalt eruption that occurs where
the magma does not encounter any carbonate strata in the surrounding
country rock is estimated to inject some 4.2 trillion tonnes (4.6 trillion
tons) of sulfur dioxide into the Earth’s atmosphere in the eruption plume
and an additional 1.2 trillion tonnes (1.3 trillion tons) from the surface
area of the lava ows, bringing the total amount of sulfur-dioxide gas
emitted to some 5.4 trillion tonnes (6 trillion tons).14 Thus, in the actual
Emeishan eruption, the estimated amount of sulfur dioxide released into
the atmosphere—5.4 trillion tonnes—is 9.7 to 20.2 times smaller than the
amount of carbon dioxide emitted into the atmosphere—52.6 to 108.9
trillion tonnes. The emission of millions of tonnes of sulfur dioxide in the
1783–1784 Laki LIP eruption chilled the Northern Hemisphere of the Earth
for two to three years (Thordarson and Self 2003; Schmidt et al. 2012), and
the emission of billions of tonnes of sulfur dioxide with each major ood-
basalt ow in the Emeishan eruption is predicted to have triggered global
cooling pulses that persisted for a decade or longer (Self et al. 2005). Still,
the climatic e ect of injecting 52.6 to 108.9 trillion tonnes of the
greenhouse gas carbon dioxide into the atmosphere would eventually
produce global warming of a magnitude that would swamp the cooling
e ects of periodic sulfur-dioxide emissions. The planet would heat up and
remain hot for hundreds of thousands of years, or for as long as the
carbon-dioxide content of the atmosphere remained high (Wignall 2001;
Bond and Wignall 2014). In the end analysis, the production of sulfuric
acid from the 5.4 trillion tonnes of sulfur dioxide produced in the
Emeishan LIP eruption would have had a much larger killing e ect by
acidifying the world’s oceans and blighting the land areas with acid rain
than by periodic global cooling.
Second, the analysis of carbon-isotope ratios not only con rms that an
enormous amount of carbon was added to the environment during the
time of the Emeishan eruptions, but also raises intriguing questions as to
the source of all of the carbon. In short, too much of the lighter isotope of
carbon, carbon-12, was injected into the Capitanian atmosphere than
could possibly have been produced by a basaltic volcanic eruption alone.
How was this conclusion reached? To explain, we know that the carbon
atom has two stable isotopes, carbon-12 and carbon-13. Carbon-12 has six
neutrons and six protons in the nucleus of the atom, and the heavier
isotope carbon-13 has seven neutrons and six protons in its nucleus
(carbon has a still heavier isotope, carbon-14, but it is unstable and
radioactively decays to nitrogen-14). We also know that living organisms
prefer the lighter isotope of carbon in their growth—plants preferentially
extract carbon-12 from the atmosphere in their photosynthetic
construction of hydrocarbons like sugar,15 and marine animals
preferentially extract carbon-12 from seawater in the construction of
their calcium-carbonate skeletons.16
Ratio proportions of the two isotopes in geologic strata are customarily
reported by an index, δ13C.17 Positive values of the index δ13C indicate an
enrichment of the strata sample in the heavier isotope carbon-13
resulting from the depletion of the lighter isotope, carbon-12. Positive
values of δ13C are commonly produced by the activity of organisms in
marine environments, such as algal blooms, as those organisms
preferentially remove carbon-12 from the environment and leave the
heavier isotope behind. Negative values of the index δ13C indicate the
opposite—an enrichment of the strata sample in the lighter isotope
carbon-12. Negative values of δ13C are commonly produced by the
weathering of organic-rich sediments—that is, sediments enriched in
carbon-12.
A major negative carbon isotope anomaly has been discovered to be
associated with the onset of Emeishan LIP volcanism and with the
Capitanian extinction (Retallack and Krull 2006; Retallack et al. 2006;
Retallack and Jahren 2008; Wignall et al. 2009; Bond, Wignall, et al. 2010).
The magnitude of the Capitanian negative carbon isotope anomaly is so
great that it could not possibly have been produced by volcanic outgassing
of carbon dioxide by the magma intrusions and ood-basalt lavas of the
Emeishan LIP eruption. Gregory Retallack and Hope Jahren, geologists at
the University of Oregon, further argue that the anomaly is too large even
to have been produced by the metamorphism and melting of the Sichuan
Basin carbonate reef strata that the Emeishan magmas burned through,
producing the massive release of carbon dioxide into the atmosphere
described by Ganino and Arndt that we examined earlier. A source of even
lower carbon-isotope ratios is needed, and Retallack and Jahren argue that
that source could only have been methane (Retallack and Jahren 2008; see
also Berner 2002).
Where could the methane have come from? Retallack and Jahren point
out three possible scenarios for the origin of the methane: (1) deep-ocean
water overturn, (2) melting of methane-rich ice deposits,18 and (3) magma
intrusion into carbonaceous sediments—sediments containing coal,
petroleum, or natural gas (Retallack and Jahren 2008). The rst scenario
requires the strati cation of large areas of the Panthalasan ocean of the
Capitanian world ( g. 6.1). Enormous amounts of methane could
accumulate and be trapped in stagnant, oxygen-poor bottom waters
contained below an upper-ocean oxygenated water-layer cap.
Catastrophic overturn of these ocean waters—perhaps triggered by global
warming—would release the deepwater methane into the Earth’s
atmosphere. Retallack and Jahren argue it is highly unlikely that the huge,
globally continuous Panthalasan ocean could have been strati ed to such
an extent and, moreover, that deepwater overturn would oxygenate the
bottom waters, whereas the actual strata and the fossil record indicate
that marine anoxia was widespread during the Capitanian extinction.
The second scenario is based on the fact that in our modern world
enormous reservoirs of methane exist frozen in the ice of permafrost
regions on land and within the pores of sediment located in cold-water,
high-pressure regions of the sea oor. If the global climate were to warm
substantially, the methane-rich ice located in the high-latitude permafrost
regions of the Earth would begin to melt and release the methane gas into
the atmosphere. Likewise, with warming ocean waters, the gas hydrates in
cold-water submarine sediments would become unstable, dissociate, and
release methane into the ocean and subsequently into the Earth’s
atmosphere. This scenario is very real and of major concern at present, as
the warming of our modern world is already triggering the release of
methane from permafrost regions of the planet. Why is this of concern?
Carbon dioxide is a potent greenhouse gas, but methane is much worse!
The Earth could become much hotter much faster with higher levels of
methane in the atmosphere.
However, several lines of evidence exist that the negative carbon
isotope anomaly seen in the Capitanian could not have been generated by
methane outbursts alone. First, the overall anomaly actually consists of
several negative isotopic excursions within the span of time of 10,000
years. Methane release from ice-deposit destabilization and melting
should occur as a single catastrophic outburst, as it apparently did in the
end-Paleocene hothouse world (Dickens et al. 1995). Following such a
catastrophic pulse, it is argued, it would take a “recharge time” of at least
200,000 years for the Earth to accumulate enough methane for another
catastrophic outburst (Dickens 2001). In contrast, several such bursts
would have had to have happened in less than 10,000 years if the negative
carbon isotopes excursions seen in the Captanian were due to methane-
rich ice destabilization alone (Retallack and Jahren 2008). Another line of
argument is that the overall negative carbon isotope anomaly was
produced slowly and gradually, not suddenly and catastrophically. In
South China, values of the index δ13C made a major shift in the negative
direction at the onset of Emeishan volcanism and the onset of Capitanian
extinctions. However, δ13C values continued to shift in the negative
direction in strata deposited after the main pulse of the extinction, and
only began to return to pre-extinction levels some two conodont zones
above the extinction interval. The University of Leeds geologist David
Bond and his colleagues argue that the negative shift in δ13C values seen in
the Chinese sections was gradual, not abrupt, and that it occurred at a
decline rate of one to two per mille per twenty meters of limestone (Bond,
Wignall, et al. 2010). Converting this stratigraphic-thickness decline rate
into one of measured time is problematic. If—and that is a big if—one
assumes that the carbonate strata accumulated at a rate of about ten
centimeters (four inches) per thousand years, then the Capitanian
negative δ13C shift occurred at a rate of 0.01 to 0.02 per mille per thousand
years. In contrast, negative carbon isotope anomalies attributed to the
catastrophic release of methane from methane-rich ice destabilization in
Jurassic strata show negative δ13C shifts that occurred at a rate of 1.5 per
mille per thousand years—roughly two orders of magnitude faster than
the negative shift seen in the Chinese Capitanian strata (Kemp et al. 2005).
Given the much slower release of carbon-12 into the atmosphere that
occurred in the Capitanian, Bond and colleagues argue against a methane-
rich ice source for the carbon-12 (Bond, Wignall, et al. 2010). If, on the
other hand, one assumes an equal duration of time per conodont zone—
another big if, as that implies a constant, clocklike, rate of evolution in the
conodont species—the negative δ13C shift occurred rapidly in the time
interval from the latest Jinogondolella altudaensis to the earliest
Jinogondolella prexuanhanensis conodont zones in South China.
Given these complications with the methane-rich ice-melting scenario,
Retallack and Jahren argue that the third scenario—magma intrusion into
carbonaceous sediments—was the source of the methane necessary to
have produced such large negative carbon isotope anomalies in the
Captanian ( g. 6.6). They point out that large diabase dikes (subterranean
injection features containing basaltic magma; see g. 6.6) from the
Emeishan LIP, some as much as 200 meters (656 feet) wide, penetrated
some 100 meters (328 feet) of Carboniferous coal deposits to the
southwest of the main magma mass, and that some of the coal seams in
these deposits are up to 12 meters (39 feet) thick. To the northwest of the
Emeishan basalts, the Sichuan Basin also contains shale and limestone
strata that contain natural gas and petroleum deposits; Ganino and Arndt
(2009) argue that these strata may have been metamorphosed by hot
magma at depth and may also have been a source of methane release into
the atmosphere. Methane release by contact metamorphism of
surrounding carbon-rich sediments by the subterranean injection of
Emeishan hot magmas would be a longer-term, more gradual
phenomenon that the sudden explosive release of methane into the
Earth’s atmosphere resulting from methane-rich ice destabilization.
FIGURE 6.6 Model showing the carbon-emission results of intruding hot basaltic lava through
strata containing layers of coal: ve separate sources of methane (CH4) generation are shown as
well as their characteristic negative carbon-isotope signatures (δ13C).

Source: From Journal of Geology, volume 116, pp. 1–20, by G. J. Retallack and A. H. Jahren, “Methane
Release from Igneous Intrusion of Coal During Late Permian Extinction Events,” copyright © 2008
University of Chicago Press. Reprinted with permission.

In summary, the debate concerning the exact sources of all of the

carbon-12 injected into the atmosphere during the Emeishan LIP eruption
continues to be lively, but there is no dispute that massive amounts of
carbon dioxide and methane were released—unlike in the historic Laki LIP
eruption, in which no carbonate or carbonaceous strata were
metamorphosed by the mantle-plume magmas. The Laki LIP eruption was
deadly enough, but to produce a really horrendous degradation of the
environment of the entire Earth, a LIP also needs to encounter additional
sources of carbon during its eruption.
The size and the environmental e ects of the Emeishan LIP eruption
were outside anything we have encountered in human history, and the
eruption triggered the fth most ecologically severe biodiversity crisis
(table 1.3) in the past 600 million years of Earth history, since the
evolution of animals in the Neoproterozoic (table 1.1). Yet the 300,000 to
600,000 cubic kilometers of lava produced by the Emeishan mantle-plume
eruption was tiny, miniscule in comparison to what was to come at the
end of the Permian. And to make matters much worse, the end-Permian
mantle super-plume encountered a huge additional source of carbon, a
source that was to ignite like a petrochemical bomb.

THE END-PERMIAN MASS EXTINCTION

We have now arrived at the catastrophic end-Permian mass extinction,
which nearly ended animal life on the planet Earth about 252 million
years ago (Benton 2003; Shen et al. 2011; Erwin 2015). All forms of life,
both terrestrial and marine, su ered in that global catastrophe. If it had
been only a little more severe, it would have erased the previous 350
million years of animal evolution, leaving only the simplest animals like
jelly sh and sponges as survivors. As it was, it triggered a restructuring of
the ecosystems of the entire planet, both on the land and in the seas
(McGhee et al. 2004), and is recorded in the fossil record as the largest loss
of biodiversity ever seen in geologic time.
The series of events that led to the greatest catastrophe in Earth
history began innocently enough: organic-rich shales began to be
deposited in the marine waters occupying the Tunguska Basin in the
eastern Siberian region of what is now Russia. Perhaps this region of the
Earth has a propensity for catastrophe, for it is the same region in which
the Tunguska asteroid exploded in the Earth’s atmosphere on June 30,
1908, with a force of some ten to 15 megatons of TNT, attening trees in
the surrounding forest over an area of 2,150 square kilometers (830 square
miles) (Farinella et al. 2001). Fortunately, no large loss of life is known to
have occurred in that event—but that was not to be case with the end-
Permian mass extinction.
The strata of the Tunguska Basin contain the oldest petroleum-bearing
deposits in the world, formed by the maturation of organic material in
Tonian- and Cryogenian-aged source rocks, Neoproterozoic strata that are
1,000 to 635 million years old (see table 1.1 for the geologic timescale)
(Svensen et al. 2009; Polozov et al. 2016). These strata range in thickness
from one to eight kilometers (0.6 to ve miles) and are overlain by thick
deposits of carbonate and evaporite rocks (rocks formed by chemical
precipitation in the evaporation of seawater) deposited in the Ediacaran
Period of the latest Neoproterozoic and the Cambrian Period of the
earliest Phanerozoic. During this period of time, the marine waters in the
Tunguska basin shallowed and began to evaporate, leaving behind vast
deposits of sea salt (NaCl)and anhydrite (CaSO4). At least ve di erent
episodes of shallowing and evaporation occurred in the Tunguska basin
during the Cambrian, leading to the deposition of up to 2.5-kilometer-
thick (1.5-mile-thick) sequences of salt- and anhydrite-rich strata. The
largest of these is the Early Cambrian Usolye salt basin, which has an
average thickness of 200 meters (656 feet) of salt spread over an area of
some two million square kilometers (772,000 square miles) (Svensen et al.
2009; Polozov et al. 2016).
The petroleum-, salt-, and anhydrite-rich strata deposited in the late
Neoproterozoic and early Cambrian in the Tunguska basin were then
buried successively by carbonates, sandstones, and eventually terrestrial
coal deposits as the basin in lled from the Ordovician to the Permian.
Under this cover of younger sediments, the oldest petroleum-rich deposits
in the world sat quietly like a ticking time bomb. In the Late Permian, that
petrochemical bomb would be triggered.
Deep in the Earth beneath the Tunguska basin in Siberia, the second
salvo in the volcanic assault that would bring an end to the Paleozoic
world was red—a mantle plume so huge that it has been designated as
the “Siberian super-plume”—was rising toward the Earth’s surface. The
Siberian super-plume volcanic eruptions would produce the largest
known continental LIP in Earth history (Reichow et al. 2009). The total
volume of magma produced in the Siberian eruption—in extrusive lava
ows and intrusive dikes and sills—is estimated to be at least ve million
cubic kilometers (1,195,000 cubic miles), with some estimates ranging as
high as 16 million cubic kilometers (3,824,000 cubic miles) (Dobretsov and
Vernikovsky 2001; Racki and Wignall 2005; Payne and Clapham 2012). Even
the lower estimate of ve million cubic kilometers is an almost
unimaginable volume: try to visualize a black cube of basaltic rock that
towers 171 kilometers (106 miles) into the sky, is 171 kilometers wide, and
is 171 kilometers long. The Earth’s oceans are only about six kilometers
(four miles) deep, on average, so if you were to place such a cube in the
middle of the ocean, it would still tower 165 kilometers (102 miles) into
the sky, reaching all the way up into the ionosphere of the Earth’s
atmosphere. If you were to drive a car at a constant 100 kilometers per
hour (62 miles per hour) alongside the base of this cube, it would take you
almost two hours to get from one corner of the cube to the next. The
minimum estimate for the Siberian LIP volume— ve million cubic
kilometers—is 8.3 to 16.7 times larger than the 300,000- to 600,000-cubic-
kilometer volume estimate for the Emeishan LIP; that is, at the very least,
the eruption of the Siberian LIP was equivalent to the eruption of 8.3, and
possibly as many as 16.7, Emeishan LIP events taking place all at once.
The Siberian LIP eruption covered ve million square kilometers
(1,930,000 square miles) of land with ood-basalt lavas. That is a land area
larger than one-half of the 48 contiguous states of the United States of
America—imagine 62 percent of the U.S. map covered with molten lava—
not once, but multiple times: the ood-basalt pile of the Siberian LIP is
composed of layer after layer of lava ows. The present-day outcrop of the
Emeishan LIP lavas covers some 250,000 square kilometers—the Siberian
LIP area is 20 times larger.
What, then, are the expected environmental consequences, in
comparison to the Emeishan LIP, of the vastly larger Siberian LIP
eruption? Given the huge magma volume and ood-basalt coverage of the
eruption alone, one might extrapolate and expect that the environmental
e ects of the Siberian eruption might be 8.3 to 20 times more severe than
those of the smaller Emeishan LIP eruption. But that calculation omits the
consequences that were to follow when the rising Siberian super-plume
ignited the petrochemical bomb buried in the strata of the Tunguska
basin.
Using thermomechanical modeling, the German Georesearch Center
(GFZ) petrologist Stephen Sobolev and colleagues argue that the Siberian
super-plume arrived below the lithosphere in the northern border of the
Siberian Shield about 253 million years ago (Sobolev et al. 2011). It was
extremely hot, with a temperature of around 1,600°C (2,900°F), and had a
plume-head diameter of around 800 kilometers (497 miles). It began to
melt the lower part of the lithosphere at a depth of between 130 and 180
kilometers (81 and 112 miles), and the plume head spread out to a width of
1,200 kilometers (745 miles) just below the lithosphere. Massive intrusion
of molten magma via sills and dykes into the lithosphere then began to
occur. The super-plume continued to rise and began to break through the
lithosphere and into crustal rocks in only 100,000 to 200,000 years. The top
of the super-plume was now located at a depth of about 50 kilometers (31
miles), and it began to melt the continental crust. This model predicts that
between six and eight million cubic kilometers (1.4 to 1.9 million cubic
miles) of molten magma was then intruded into crustal rocks (Sobolev et
al. 2011), a prediction that is in accord with the empirical estimates of the
volume of the Siberian LIP (Dobretsov and Vernikovsky 2001; Racki and
Wignall 2005; Payne and Clapham 2012).
The huge volume of molten magma intruded into crustal rocks through
subterranean sills and dykes now encountered the petroleum-, salt-, and
anhydrite-rich strata buried between one and three kilometers (0.6 to ve
miles) below the surface of the Earth in the Tunguska Basin. Beginning
around 252 million years ago, the contact of hot magma with the organic-
and evaporite-rich strata buried in the basin resulted in the generation of
huge amounts of gas at high pressure, subterranean gas that exploded and
produced gigantic blast pipes that reached all the way to the surface of the
Earth.
About 250 blast pipes lled with magnetite-rich breccia are located in
the southern part of the Tunguska Basin in the region characterized by
thick deposits of Cambrian salts at a depth of some 2,000 meters (6,560
feet) below the surface of the Earth. Several of these pipes contain iron-
ore concentrations of commercial value and are currently being mined for
magnetite (Fe3O4). In the northwest region of the Tunguska Basin, outside
the region underlain by the Cambrian evaporites, are more than 500
additional blast pipes lled with basalt (Svensen et al. 2009; Polozov et al.
2016).
A geologic cross section of one of these explosive pipes, the
Scholokhovskoie pipe, is shown in gure 6.7. Because of the commercial
value of the magnetite ores contained in the pipe, the strata both within
the pipe and surrounding it have been extensively drilled. A massive, 200-
meter-thick (656-foot-thick) intrusive-magma sill of dolerite (labeled
“Upper Sill” in g. 6.7)19 is present in the 400 meters (1,310 feet) of Late
Cambrian sandstones and siltstones of the Verkholensk Suite strata
(topographic elevations of 100 to 400 meters in g. 6.7). At greater depths,
some 800 to 900 meters (2,620 to 2,950 feet) below the surface of the Earth,
thinner dolerite sills are located in the thick Early Cambrian salt and
anhydrite evaporite strata of the Angara and Bulay Suite strata (lower sills
at topographic depths of −400 to −550 meters in g. 6.7). The blast pipe
itself contains large breccia blocks of the Cambrian evaporite strata that
have been explosively ejected up the pipe and, back at the end of the
Permian, out the mouth of the surface crater onto the surrounding
countryside. Also contained within the pipe are breccia blocks from the
dolerite sills, proving that the explosions that produced the blast pipe
were contemporaneous with the intrusion of the magma intrusions that
produced the sills. These magmatic breccia blocks are rich in glass,
produced by rapid cooling of the hot intruded magma as it was exposed to
air within the pipe.
FIGURE 6.7 Geologic cross section of the upper 1,000-meter (3281-foot) thickness of the strata
containing the Scholokhovskoie blast pipe, obtained by subsurface data from numerous drill holes
into the strata (inset map, lower left corner of the gure). Strata abbreviations in the left margin of
the gure: O, Ordovician; L-S, Litvinstsev Suite; B-S, Angara and Bulay Suites.

Source: From Earth and Planetary Science Letters, volume 277, pp. 490–500, by H. Svensen, S. Planke, A.
G. Polozov, N. Schmidbauer, F. Corfu, Y. Y. Podladchikov, and B. Jamtveit, “Siberian Gas Venting and
the End-Permian Environmental Crisis,” copyright © 2009 Elsevier. Reprinted with permission.

Note that the Scholokhovskoie pipe at the surface of the Earth ( g. 6.7)
has a diameter of some 430 meters (1,410 feet) and narrows to a diameter
of 300 meters (984 feet) at a depth of 1,000 meters (3,280 feet) below the
surface—a geometry produced by the upward and outward force of the
explosions that formed the pipe. Other pipes in the region have surface
craters that are some 1.6 kilometers wide (one mile wide), an exit-blast
diameter that is a mute geologic witness to the spectacular magnitude of
the explosions that occurred in this region of the Earth 253 million years
ago.
All in all, over 6,400 explosive pipes exist in the Tunguska Basin, and
drill-hole geologic samples from many of them have been extensively
studied by the University of Oslo geologist Henrik Svensen and his
colleagues (2009). The geophysical model they propose for the formation
of the Siberian LIP blast pipes is shown in gure 6.8. In the rst frame of
the sequence (from left to right), the intrusion of a horizontal sill of hot
volcanic magma, marked with “V” symbols in gure 6.8, occurs at depth in
the Proterozoic strata underlying the petroleum-rich Cambrian evaporite
strata (petroleum pools are black ovals marked with white “P” letters in
g. 6.8). In the second frame, a larger sill of hot magma is shown intruding
into the Cambrian strata, the thermal volatilization of the petroleum
deposits within the strata begins to occur, and the resultant gas expands,
shown by the stippled-bounded region and the upward-pointing arrows.
In the third frame, the gas explodes at depth and produces the upward
blast force (upward-pointing arrows) that creates the pipe, carrying large
blocks of the sill rock and surrounding Cambrian strata upward and out of
the blast crater at the surface of the Earth. Additional contact-
metamorphism and gas production is shown surrounding the degassing
sill at shallower depths within the Ordovician strata (upper sill marked
with “V” symbols in the upper part of the gure). Finally, in the fourth
frame, continued degassing (wavy upward-pointing arrows) is shown from
the petroleum-rich Cambrian evaporite strata and the magma layers, and
is shown continuing to occur in the pipe itself—illustrated by the bubbles
in the waters of the lake now shown occupying the blast crater. In
addition, intrusive sills are now shown to come into contact with shallow-
buried coal deposits (the uppermost black layer in the top of the gure),
and contact metamorphism of the coal produces even more methane
degassing, shown by the wavy upward-pointing arrow (Svensen et al.
2009).

FIGURE 6.8 Model of the sequential events (from left to right) that led to the formation of the
Siberian LIP blast pipes; see text for discussion.

Source: From Earth and Planetary Science Letters, volume 277, pp. 490–500, by H. Svensen, S. Planke, A.
G. Polozov, N. Schmidbauer, F. Corfu, Y. Y. Podladchikov, and B. Jamtveit, “Siberian Gas Venting and
the End-Permian Environmental Crisis,” copyright © 2009 Elsevier. Reprinted with permission.

Simply trying to imagine the eruption of one of these pipes is

staggering. Suddenly a vast expanse of land—more than 1.6 kilometers
(one mile) across, the size of the downtown area of a small city—exploded
into the air with a tremendous blast followed by the ground-shaking roar
and hiss of a gigantic column of superheated gases and steam jetting
upward into the sky. Huge chunks of country rock and glowing-hot magma
blobs were sent rocketing across the heavens, arcing away to fall back to
the Earth at incredible distances. Thick black plumes of smoke began to
billow up from the blast crater, produced by the subterranean burning of
coal beds by the hot magma sills. Coal y ash from these billowing smoke
clouds was transported as far as 20,000 kilometers (12,400 miles) from
Siberia, around the top of the world, to settle in Canadian High Arctic
sediments (Grasby et al. 2011).
If that scenario is di cult to envision, imagine it happening 6,400
times! Svensen and colleagues point out that the eruption pattern of the
Siberian LIP blast pipes is unknown. Obviously, if one pipe exploded per
year, then the entire process took 6,400 years, but it is much more likely
that pipe formation was clustered in time—probably starting out with the
occasional blast of a single or a few pipes erupting as the initial magma
intrusions reached the buried Cambrian petroleum-rich evaporite strata,
followed by a period of time with numerous pipes exploding
simultaneously across the landscape during the phase of maximum
heating and magma intrusion from the head of the mantle super-plume
located at depth, and ending with a tapering o of pipe formation as the
subterranean magma masses cooled. In that scenario, the formation of the
Siberian LIP blast pipes took place in a time interval much shorter than
6,400 years.
Svensen and colleagues have conducted extensive heating experiments
with geologic samples of the Siberian LIP strata, obtained through
numerous drill holes produced by prospectors searching for petroleum,
magnetite, and potassium salt deposits. They have concluded that the
Siberian LIP blast pipes vented between 39 and 114 trillion tonnes (43 to
125 trillion tons) of carbon dioxide, plus an additional 20 trillion tonnes
(22 trillion tons) of carbon dioxide degassed from the lava. In addition, the
contact metamorphism of coal and other organic-rich strata by the
Siberian mantle super-plume magmas is estimated to have released
between 14.3 and 41.9 trillion tonnes (16 to 46 trillion tons) of methane.
The coal y ash produced by the burning of the Siberian LIP coal beds also
contained high levels of the toxic metals chromium, arsenic, mercury, and
lead (Grasby et al. 2011, 2017).
The carbon dioxide and methane vented from the burning of
petroleum- and organic-rich rocks produced a huge injection of the light
carbon-12 isotope into the atmosphere (Svensen et al. 2009). Analyses of
the resultant negative δ13C excursion found in end-Permian strata have led
the Stanford University geologist Jonathan Payne and colleagues to
estimate an even higher volume of vented carbon-dioxide gas from the
Siberian LIP—some 100 to 160 trillion tonnes (110 to 176 tons)—than that
estimated by Svensen and colleagues (Payne et al. 2010).
In addition to organic-rich rocks, the Siberian LIP magmas encountered
a type of rock not present in the Emeishan LIP area—the Cambrian
evaporite strata, thick deposits of rock salt and anhydrite. The burning of
rock salt, or halite, released sodium metal and chlorine gas, both of which
are poisonous. The Siberian LIP chlorine gas interacted with water vapor
to produce rock-salt-derived hydrochloric acid, in addition to the amount
produced by the magma itself. Even more serious, however, is that the
vented chlorine gas combined with the huge amounts of methane also
released by the Siberian LIP to produce a very noxious gas—methyl
chloride (CH3Cl). Svensen and colleagues estimate that between 5.2 and
15.3 trillion tonnes (5.7 to 16.9 trillion tons) of methyl-chloride gas were
released into the Earth’s atmosphere by the Siberian LIP. Finally, the
presence of bromine in the Cambrian evaporite strata is estimated to have
produced another noxious halocarbon gas in bromine’s combination with
methane—methyl bromide (CH3Br). Although smaller than the huge
volume of methyl chloride produced by the Siberian LIP, Svensen and
colleagues estimate that between 87 and 255 billion tonnes (96 to 281
billion tons) of methyl bromide were emitted into the Earth’s atmosphere
at the end of the Permian (Svensen et al. 2009). The consequences of the
formation of these two halocarbon gases by the Siberian LIP were dire
indeed, as we will see shortly.
Finally, it is estimated that the Emeishan LIP eruption produced about
5.4 trillion tonnes of sulfur dioxide, as discussed previously. Since the
Siberian LIP ood-basalt eruption was 8.3 to 20 times larger than the
Emeishan, one might conclude that the Siberian LIP lavas would have
produced 8.3 to 20 times more sulfur dioxide. But that calculation does not
consider the e ect of the thermal metamorphism of the other evaporite
rock type present in the Cambrian evaporite strata—thick deposits of
anhydrite. Anhydrite is a calcium sulfate, and burning anhydrite produces
sulfur dioxide gas in addition to the amount produced by the magma
itself; thus, the potential amount of sulfuric acid produced by the Siberian
LIP eruption was even greater than would be anticipated based on the size
of the ood basalts alone.
Can the Siberian LIP scenario get any worse? The answer is yes. The
thermomechanical modeling of the Siberian super-plume by Sobolev and
colleagues suggests that about 10 to 20 percent of the magma volume
produced did not come from the deep mantle alone but rather from the
melting and recycling of oceanic crustal rock located beneath the
continental crust of Siberia—oceanic crust that had been subducted
beneath the continental crust of Asia in earlier plate-tectonic events. If so,
the magma produced by the super-plume would be much more gaseous
and carbon rich than the typical basaltic-type lavas seen in LIP eruptions
like those of Laki, Iceland. Sobolev and colleagues estimate that the
degassing of the modeled super-plume magmas could have produced
about 175 trillion tonnes (193 trillion tons) of carbon dioxide—an amount
greater than the estimated 39 to 114 trillion tonnes of carbon dioxide
vented through the Siberian blast pipes from the subsurface thermal
metamorphism of the Tunguska Basin petroleum-rich strata! In addition,
a surprising amount of hydrochloric acid is also modeled to have occurred
solely from the degassing super-plume itself—some 18 trillion tonnes (20
trillion tons) (Sobolev et al. 2011).
In this scenario, Sobolev and colleagues suggest that “CO2 from the
plume alone may have triggered the main extinction event … degassing of
the plume, rather than thermogenic gases from sediments, triggered the
biotic crises” (Sobolev et al. 2011, 314–315). However, we know that the
thick deposits of petroleum-, coal-, and evaporite-rich strata in the
Tunguska Basin were thermally metamorphosed, and the explosive result
of that metamorphism is evidenced by the 6,400 blast pipes present in the
basin today. In addition, we know that the burning of the Tunguska Basin
coal deposits produced huge clouds of carbon-rich ash, ash in such
quantities that it was transported aloft some 20,000 kilometers away to
Canada (Grasby et al. 2011). Thus, rather than an alternative source of
carbon dioxide, the Siberian super-plume model of Sobolev and colleagues
may provide an additional source of trillions of tonnes of carbon dioxide to
add to the trillions of tonnes produced by thermogenic degassing from the
Tunguska Basin petroleum-rich strata—making an originally catastrophic
environmental scenario much worse. In this combination scenario,
catastrophic degassing from the deeper super-plume itself began shortly
before the main ood-basalt eruptions, followed by additional
catastrophic degassing from the shallower Tunguska Basin petroleum-rich
evaporite strata during the ood-basalt eruptions.

THE END-PERMIAN KILL MECHANISMS

The predicted environmental consequences of the injection of
teratonnages of both methane and carbon dioxide into the Earth’s
atmosphere by the huge Siberian LIP eruption are the same as they were
for the earlier and smaller Emeishan eruption, just very much worse (table
6.1). Methane and carbon dioxide are both greenhouse gases—methane
much more so than carbon dioxide—and global heating of the planet is
expected with high concentrations of these two gases in the atmosphere.
Methane is particularly bad in that it retains heat itself and, when
oxidized, also produces carbon dioxide, a gas that continues to retain heat
in the atmosphere.

TABLE 6.1 Pollutants produced by the Siberian LIP eruption and their global environmental
impacts.

Pollutant/Pollutant Product Environmental Impact

Sulfur dioxide/sulfuric acid Acidi cation of oceans, acid rain on land
Carbon dioxide/carbonic acid Acidi cation of oceans, acid rain on land
Chlorine/hydrochloric acid Acidi cation of oceans, acid rain on land
Atmospheric sulfur dioxide Short-term global cooling
Atmospheric carbon dioxide Long-term global warming
Atmospheric methane Long-term global warming, global oxygen depletion
Chlorine/methyl chloride Atmospheric ozone destruction
Bromine/methyl bromide Atmospheric ozone destruction
Coal y ash Oceanic euxinia and anoxia, metal toxicity

Kill Mechanism 1: Heat Death

The nearly unbelievable seriousness of the magnitude of the heating of

the Earth that occurred at the end of the Permian and into the Early
Triassic can be sensed in the wording of the titles (usually pretty staid and
descriptive) of numerous scienti c papers in research journals that have
revealed the enormity of the event—titles such as “Hot Acidic Late
Permian Seas Sti e Life in Record Time” (Georgiev et al. 2011), “Lethally
Hot Temperatures During the Early Triassic Greenhouse” (Sun et al. 2012),
and “Post-Apocalyptic Greenhouse Climate Revealed by Earliest Triassic
Paleosols” (Retallack 1999). The language within many of these scienti c
papers is just as startling—for example, from a study using the ratios of
rhenium (Re) and osmium (Os) isotopes in calculating seawater variation
in temperature and acidity, “If temperature was indeed the controlling
factor, the extreme 187Re /188Os ratios in the Upper Permian time require
global warming at levels unparalleled in the geologic record…. Such ratios
are unknown from any other period in Earth history” (Georgiev et al.
2011, 397, 399).
The standard technique for calculating past temperatures is to measure
the ratios of oxygen isotopes (δ18O) preserved in the skeletal elements of
organisms alive during the study interval, usually the shells of
brachiopods or the dentary elements of conodonts (McGhee 2013, 196–
199). Using the δ18O index values found in the dentary elements of
conodonts, the China University of Geosciences geobiologist Yadong Sun
and colleagues have reconstructed seawater temperatures in the late
Changhsingian to early Middle Triassic. Their research has revealed that
seawater temperatures rose rapidly from 21°C (70°F) to 36°C (97°F) across
the Permian/Triassic boundary. This rapid rise in temperature was
interrupted by a cool pulse, when seawater temperatures dropped back to
a still-quite-warm 32°C (90°F)—because of renewed volcanic eruption of
sulfur dioxide?—before rising to the highest temperatures seen in their
study: seawater temperatures of 38°C (100°F) and sea-surface
temperatures of 40°C (104°F) in the mid-Olenekian Age of the Early
Triassic.20
Seawater temperatures above 35°C (95°F) are lethal for most marine
organisms, triggering hyperthermia (table 6.2). Yet the study by Sun and
colleagues show that seawater temperatures remained above 35°C for
almost the entire Early Triassic, a span of 5.1 million years, before nally
falling back to the very warm range of 32°C to 34°C (90°F to 93°F) at the
onset of the Middle Triassic Epoch. At present, sea-surface temperatures
in the equatorial zone of the Earth—the hottest part of the Earth—range
between 25°C and 30°C (77°F to 86°F). High seawater temperatures also
lead to lower oxygen solubility and thus facilitate the development of
marine anoxia, a fact we will return to when we consider the hypoxic kill
mechanism in detail (table 6.2).

TABLE 6.2 Kill mechanisms triggered by the Siberian-LIP-induced global environmental changes
listed in table 6.1.

Kill Mechanism Organisms Adversely A ected

Hyperthermia Large animals (both marine and land), energetic animals with higher metabolic
(heat death) rates, marine sessile benthic organisms, land plants
Acidosis (acid Marine uni- and mullticellular organisms with calci ed skeletons, marine sessile
poisoning) benthic organisms, land plants
Hypercapnia Large animals (both marine and land), marine uni- and multicellular organisms
(CO2 with calci ed skeletons, marine animals with poorly bu ered respiratory
poisoning) physiologies, marine sessile benthic organisms
Hypoxia Large animals (both marine and land), energetic animals with higher metabolic
(su ocation) rates, marine sessile benthic organisms
Radiation Land plants, land animals that cannot burrow to escape surface radiation
poisoning

On land, photorespiration starts to seriously interfere with plant

photosynthesis at temperatures above 35°C (95°F), and few terrestrial
plants can survive temperatures higher than 40°C. At temperatures of
45°C (113°F), land animals begin to succumb to hyperthermia (Sun et al.
2012). The Université Claude Bernard geochemist Kévin Rey and
colleagues have analyzed the δ18O values found in the bones and teeth of
terrestrial vertebrates in South Africa and have reported that at the end of
the Permian “an intense and fast warming of +16°C [+29°F] occurred and
kept increasing during the Olenekian” Age of the Early Triassic (Rey et al.
2016, 384). All in all, the end-Permian hot pulse in air temperatures on
land “lasted 6 Ma [million years] before temperatures decreased” in the
Anisian Age of the Middle Triassic (Rey et al. 2016, 384).
In the “Hot Earth” of the latest Permian and Early Triassic, the
equatorial tropics were lethal zones both on land and in the sea—in stark
contrast to our modern world, where the equatorial tropics harbor the
highest diversity of life on the planet. The equatorial marine zone of the
Hot Earth was full of “gaps”—the “marine sh gap,” the “marine reptile
gap”—marked by the absence of life-forms driven out of the tropics to
higher latitudes of the planet where temperatures were survivable (Sun et
al. 2012). Individuals of marine species that managed to survive in the Hot
Earth equatorial seas were all abnormally small and stunted—a
phenomenon known as the Early Triassic “Lilliput e ect” (Twitchett
2007), which Sun and colleagues (2012) argue to be the result of the low
thermal tolerance range of organisms with large body sizes. The
equatorial land areas of the Hot Earth saw the “tetrapod gap,” the “coal
gap,” and the “conifer gap”—the absence of tetrapods, the absence of peat
swamps, and the absence of conifers (Retallack et al. 1996). Instead of the
highly derived conifer seed plants, the ora of the equatorial Hot Earth
were dominated by peculiar primitive spore-reproducing lycophytes,
relatives of the ancient lycophytes that dominated the coal swamps of the
Carboniferous world that we considered in chapter 3. Only in the
marginally cooler Middle Triassic Earth would the tetrapods and conifers
once again return to the equatorial regions, and peat swamps once again
form (Sun et al. 2012).
Kill Mechanism 2: Carbon Dioxide Poisoning

Carbonic acid formed from both carbon dioxide and methane acidi ed the
oceans and produced acid rain on land. To the acid e ect of trillions of
tonnes of both gases was added the trillions of tonnes of sulfuric acid and
hydrochloric acid produced both from the mantle super-plume itself and
from the thermal metamorphism of the Tunguska Basin petroleum-rich
evaporite strata (table 6.1). To make matters even worse, the ocean
chemistry of the Paleozoic world was not like that of our present world.
Jonathan Payne and his colleagues point out that the Earth’s oceans today
are “bu ered against acidi cation by extensive, ne-grained, unlithi ed
carbonate sediments on the deep-sea oor, which could relatively rapidly
dissolve to counter acidi cation. By contrast, the Late Permian deep sea
contained no such carbonate bu er because Permian oceans lacked
abundant pelagic carbonate producers such as coccolithophorids and
planktonic foraminifers. Consequently, any bu ering against acidi cation
via dissolution of carbonate sediments could only have occurred more
slowly in the less extensive, coarse-grained, mostly lithi ed, shallow-
marine carbonate platform sediments or via chemical weathering of
silicate and carbonate rocks on land” (Payne et al. 2010, 8546; see also
Ridgwell 2005) and “In the absence of a large reservoir of ne-grained,
unlithi ed deep-sea carbonate sediment, whole-ocean acidi cation could
likely last for tens of thousands of years (Archer et al., 1997), and few
refugia would exist—survival would depend primarily on long-term
physiological tolerance of altered conditions” (Payne and Clapham 2012;
Clarkson et al. 2015).
Biological evidence corroborates the hypothesis that the acidi cation
of the world’s oceans during the end-Permian mass extinction was much
worse than it was during the earlier Capitanian crisis. During the eruption
of the Emeishan LIP, highly calci ed organisms like massive corals, large-
bodied fusulinid foraminifera, and giant alatoconchid bivalves su ered
high extinction losses relative to organisms with lightly calci ed
skeletons. However, during the eruption of the Siberian LIP, all organisms
with calcareous skeletons su ered high extinction losses. In fact, having a
skeleton composed of anything other than calcium carbonate was highly
bene cial in surviving the end-Permian mass extinction (Clapham and
Payne 2011). This di erential survival pattern is particularly striking
within clades of closely related organisms: many species of calcareous
foraminifera went extinct, whereas those foraminifera that formed their
skeletons by agglutinating siliceous sedimentary grains survived; many
species of calcareous brachiopods went extinct, whereas the inarticulated
brachiopods with phosphatic shells survived; many species of calcareous
sponges went extinct, whereas siliceous hexactinellid sponges survived;
species of calci ed corals and algae went extinct, whereas unskeletonized
corals (naked cnidarians, like modern sea anemones) and unskeletonized
algae survived; and so on (Knoll et al. 2007; Clapham and Payne 2011;
Kiessling and Simpson 2011; Garbelli et al. 2017).
A surprising contrast to the deadly e ect of acid-rich seawater on
calcareous organisms is the discovery of peculiar primary deposits of
calcium carbonate on the sea oor in some areas, deposits of odd fan-
shaped crystal structures that can be some ten centimeters (four inches)
in diameter. These calcium-carbonate fans and crusts are thought to have
been produced in the aftermath of an oceanic acidi cation event,
following the death of most benthic organisms with calcareous skeletons,
and the upwelling of alkaline deep waters supersaturated with calcium
carbonate—but no organisms to take up the calcium carbonate for their
skeletal constructions—resulted in these primary sedimentary carbonate
deposits (Payne and Clapham 2012). Finally, on land, the deadly e ects of
acid rain, such as those that were seen in Europe with the historic Laki LIP
eruption, were added to the already deadly e ects of hyperthermia in
killing o land plants.
Added to the deadly e ects of sulfuric, hydrochloric, and carbonic
acids in marine waters is the related e ect of high concentrations of
simple carbon dioxide in ocean waters—hypercapnia, or carbon-dioxide
poisoning (table 6.2). Hypercapnia preferentially kills many of the same
organisms as acidosis, but with a few di erences. Animals with poorly
bu ered respiratory physiologies are least able to deal with the deadly
e ects of carbon-dioxide poisoning; these animals include the sponges,
corals, brachiopods, bryozoans, and most echinoderms. In contrast,
energetic animals with well-bu ered physiologies are better able to deal
with carbon-dioxide stress; these animals include the molluscs,
arthropods, chordates, and infaunal organisms in general, which are
adapted to living within submarine sediments where oxygen partial
pressures are low and carbon-dioxide partial pressures are high (Knoll et
al. 1996, 2007; Clapham and Payne 2011). The di erential survival rates of
these two groups of organisms provide an important key to understanding
the radical ecological reorganization that ended the Paleozoic world—an
ending we will consider in detail later in this chapter.
Terrestrial “infaunal” animals—that is, burrowers that live
underground—also are better able to survive high carbon-dioxide levels in
atmospheric gases, for the same reason as with the infaunal marine
animals. Animals with more advanced respiratory structures, such as
muscular diaphragms for energetic breathing, lungs with large surface
areas contained in barrel-chested rib cages, and nasal turbinals are also
better able to survive high-carbon-dioxide, low-oxygen conditions
(Retallack and Krull 2006; Knoll et al. 2007). Thus the worldwide survival
and proliferation of species of the small-bodied, barrel-chested, burrowing
tetrapod Lystrosaurus in the Early Triassic—at high latitudes, far away
from the lethally hot tropics (Sun et al. 2012)—may be attributed to the
combined selective e ects of high temperatures, high carbon-dioxide
levels, and low oxygen levels in terrestrial environments following the
end-Permian mass extinction ( g. 6.9).

FIGURE 6.9 Summary reconstructions of biological conditions before and after the end-Permian
mass extinction: (a) the replacement of Permian large land animals (Dicynodon lacerticeps and
Rubidgea majora) by Triassic small land animals (Galesaurus planiceps) and burrowers (Lystrosaurus
declivis) in dry environments; (b) the replacement of Permian wetland thick peat deposits (black
layer) and trees with aquatically adapted deep roots (Glossopteris browniana) by Triassic dry
conditions; (c) the replacement of Permian large marine animals by Triassic small “Lilliput” marine
animals.
Source: From Geological Society of America Special Paper 399, pp. 249–268, by G. J. Retallack and E. S.
Krull, “Carbon Isotopic Evidence for Terminal-Permian Methane Outbursts and Their Role in
Extinctions of Animals, Plants, Coral Reefs, and Peat Swamps,” copyright © 2006 Geological Society
of America. Reprinted with permission.

Kill Mechanism 3: Su ocation

Just as the kill mechanisms of acidosis and hypercapnia adversely a ect

many of the same organisms, the e ects of the kill mechanism of hypoxia
—oxygen-deprivation stress all the way to su ocation death—are similar
as well (table 6.2). The Siberian LIP eruptions triggered a major drop in the
oxygen content of the Earth’s atmosphere in a cascade of linked
environmental e ects. First, the burning—the oxidation—of the vast
deposits of organic material contained in the Tunguska petroleum-rich
strata and coal beds would necessarily consume a tremendous amount of
atmospheric oxygen.21 As we previously considered, one of the very
negative consequences of the burning of those subterranean
hydrocarbon-rich strata was the production of trillions of tonnes of
methane. Second, the oxidation of that vast amount of methane also
sucked oxygen out of the air—two oxygen molecules per single methane
molecule—leading to a further global decline in the oxygen content of the
oceans and of the atmosphere.22
The atmospheric modeling of Robert Berner ( g. 3.8) indicates that a
decline in oxygen content of the Earth’s atmosphere began in the Middle
Permian, continued through the Late Permian and Early Triassic, and
reached a minimum level of 15.5 percent at the beginning of the Middle
Triassic, in the Geocarbsulf model (in the Anisian Age; see table 6.3), or a
minimum level of 16.0 percent in the late Middle Triassic (the Ladinian
Age; see table 6.3) in the Rock-Abundance model. In the models, the
predicted atmospheric-oxygen values are calculated at ten-million-year
intervals and thus correspond only roughly to the expected oxygen-
depleting e ects of the Emeishan and Siberian LIP eruptions, whose
catastrophic phases were of much shorter duration than ten million years
(table 6.3). Note, however, that the minimum values predicted for the
Triassic—an atmospheric oxygen content of 15.5–16.0 percent—are still
above the 13 percent level below which wild res do not take place and
charcoal is not formed.

TABLE 6.3 Major LIP volcanic events in the Permian, Triassic, and earliest Jurassic compared with
the oxygen content in the Earth’s atmosphere.

Geologic Age Atmospheric O2

(1) (2)
Hettangian 17.0% 14.5%
Rhaetian 17.0% 18.0% ← Central Atlantic LIP eruptions
Norian 16.0% 16.0% ← Central Atlantic LIP eruptions
Carnian 17.0% 15.5%
Ladinian 17.6% 16.0%
Anisian 20.9% 18.7%
Olenekian 22.5% 20.0%
Induan 23.3% 21.8% ← Smaller Siberian LIP eruptions continue
Changhsingian 24.1% 23.6% ← Catastrophic Siberian LIP eruptions
Wuchiapingian 26.5% 29.0%
Capitanian 28.3% 30.0% ← Emeishan LIP eruptions
Wordian 30.0% 31.0%
Roadian 30.6% 30.9%
Kungarian 33.0% 30.5%
Artinskian 35.0% 30.0%
Sakmarian 34.8% 29.3%
Asselian 34.5% 28.5%

Source: Column (1) is the Rock-Abundance model (Berner et al., 2003) and column (2) is the
Geocarbsulf model (Berner, 2006). Oxygen data extrapolated from gure 3.8 using the timescale of
Gradstein et al. (2012).
Note: Atmospheric oxygen contents of 25% or higher have been underlined; those of 30% or higher
are also in bold.

Abundant geologic evidence exists for widespread marine anoxia

during and after the end-Permian mass extinction; indeed, this may have
been the most geographically widespread anoxic event in the world’s
oceans for the entire Phanerozoic—a “superanoxic event” (Wignall and
Twitchett 2002). Black-shale deposits are typical sedimentary signals of
oxygen depletion in marine waters, and black-shale deposits are
widespread in shallow marine environments during the end-Permian
mass extinction. Thinly laminated sediments rich in pyrite, an iron sul de
(FeS2), are also widespread and suggest not only anoxic conditions but the
presence of deadly hydrogen sul de (H2S). Other geochemical signatures
of anoxia and of the presence of hydrogen sul de, such as shifts in the
relative isotopic abundances of nitrogen and uranium and the presence of
isorenieratane, have also been widely discovered in Permian/Triassic
boundary strata (Payne and Clapham 2012).
In the oceans, anoxia would have been triggered not only by Siberian-
LIP-induced low atmospheric pressure of oxygen and the oxidation of
Siberian-LIP-produced methane but also by the high seawater
temperatures of the Hot Earth, which substantially lowered oxygen
solubility. In addition, nutrients (chie y iron) from the coal y ash
produced by the volcanic burning of subterranean coal deposits in the
Tunguska Basin may have triggered vast algal blooms and eutrophication
of shallow marine waters (table 6.1), leading to further oxygen depletion
in bottom waters (Grasby et al. 2011).
Still, Jonathan Payne and the University of California paleontologist
Matthew Clapham argue that the hypoxic kill mechanism played a lesser
role in the total biological devastation of the end-Permian mass
extinction, stressing the primary kill mechanisms of hyperthermia,
acidosis, and hypercapnia (table 6.2). They point out that sur cial and
shallow-water anoxic and hydrogen-sul de-rich conditions “could not
have been both temporally persistent and geographically widespread
unless atmospheric oxygen levels were far lower than the values indicated
by model reconstructions (e.g., Berner 2006)” (Payne and Clapham 2012,
102) (see table 6.3).
For land animals, the e ects of hypoxia and hypercapnia are essentially
the same; thus, the survival and proliferation of species of the small-
bodied, barrel-chested, burrowing dicynodont Lystrosaurus—
approximately 90 percent of all land vertebrates alive in the Early Triassic
were lystrosaurs (Sahney and Benton 2008)—could just as well be due to
selection for low-oxygen tolerance as for high-carbon-dioxide tolerance,
and is probably due to both selective e ects acting in concert ( g. 6.9).
Land plants, however, present a more complex problem. The dominant
and more recently evolved pinophyte spermatophytes of the Permian
were replaced by more ancient, spore-reproducing isoetalean lycophytes
in the terrestrial ora of the earliest Triassic (see table 3.1 for the
phylogeny of Paleozoic land plants). Only in the late Olenekian Age of the
Early Triassic did the pinophytes regain their dominance in the terrestrial
ora, and then only in the higher-latitude regions of the Earth, whereas
the lycophytes remained dominant in the equatorial regions until the
beginning of the Middle Triassic (Sun et al. 2012). This latitudinal pattern
of retreat from the tropics by the pinophytes, then gradual return to
lower and lower latitudes until the tropics were once again reached by the
Middle Triassic, is exactly the pattern one would expect to be produced by
the onset of hot temperatures at the Permian/Triassic boundary,
persistent high temperatures in the Early Triassic, and a nal cooling
down of the Earth by the Middle Triassic. This latitudinal-shift pattern in
land plants is not a predicted result of the onset of low-oxygen, high-
carbon-dioxide conditions in the Earth’s atmosphere.
In fact, one might predict that the land plants would ourish in such an
atmosphere—if temperatures were within livable ranges. To land plants,
oxygen is a waste product. Unlike terrestrial arthropods or vertebrates,
which need oxygen to survive, plants need carbon dioxide to synthesize
their food. In the fossil record we repeatedly see that land plants ourish
during times of low oxygen content, and high carbon-dioxide content, in
the atmosphere. But did land plants ourish in the Early Triassic? No, they
did the exact opposite. Rather than ourishing, we see a six-million-year
global “coal gap” in which there was insu cient plant growth in peat
swamps to form any coal deposits (Retallack et al. 1996). In the hills and
highlands, there was insu cient plant growth to stabilize and bind the
upland soils in the Early Triassic, and downstream, low-energy
meandering rivers were replaced by high-energy braided rivers on a
worldwide basis (Ward et al. 2000; Sahney and Benton 2008). Still, Gregory
Retallack argues that the coal gap could also have been produced—at least
in part—by “soil hypoxia,” because low oxygen levels within soils will
interfere with plant-root respiration and nutrient uptake (Retallack and
Krull 2006).

Kill Mechanism 4: Radiation Poisoning

Land plants do show evidence of an additional kill mechanism not

demonstrated for the Emeishan LIP eruption: radiation poisoning (table
6.2). The predicted consequences of the injection into the atmosphere of
trillions of tonnes of methyl chloride and billions of tonnes of methyl
bromide were new to the Siberian LIP eruption (table 6.1), as the
Emeishan LIP magmas did not encounter buried evaporite strata. Chlorine
and bromine catalytically destroy ozone,23 as we know from modern times
with our e orts to stop the destruction of the Earth’s ozone layer at the
poles of the planet caused by pollutant chloro uorocarbons from our
older coolant systems. The Siberian LIP eruption produced trillions of
tonnes of hydrochloric acid, but atmospheric-photochemical computer
simulations by the University of She eld biologist David Beerling and his
colleagues suggest that that amount of hydrochloric acid, although huge,
would only destroy 33–55 percent of the stratospheric ozone shield in the
high-latitude Northern Hemisphere, where the Siberian LIP is located.
However, when additional trillions of tonnes of methyl chloride are added
to the mix, the simulations predict that 70–85 percent of the Northern
Hemisphere ozone shield would be destroyed, and that 55–80 percent of
the Southern Hemisphere ozone shield would be destroyed as well. And, to
make matters worse, the addition of methyl bromide to the mix speeds up
the catalytic process of ozone destruction (Beerling et al. 2007; Svensen et
al. 2009).
With the destruction or serious thinning of the ozone layer, the
radiation ux of high-energy ultraviolet light in the shortwave 290 to 315
nanometer spectrum24—deadly UV-B radiation—increases at the Earth’s
surface. The computer simulations of Beerling and colleagues predict that
the UV-B radiation ux could increase by as much as 50 to 100 kilojoules
(12 to 24 kilocalories) per square meter (square yard) per day (Beerling et
al. 2007). That increased radiation ux would have mutagenic e ects on
land plants, as plants cannot ee the Earth’s surface and hide
underground as burrowing animals can. Thus, to the selective e ects of
hypercapnia and hypoxia, we can add the selective e ect of radiation
poisoning in favoring the survival of the burrowing lystrosaurs (table 6.2).
 The Utrecht University paleobotanist Henk Visscher and his colleagues
argue that that bleak radiation-poisoning scenario did indeed happen in
the end-Permian mass extinction. They report nding mutation-induced
changes in land-plant reproductive structures in Late Permian strata—
anomalous variation in shape, size, wall thickness, and arrangement of
pollen and spores. Numerous spores are found still bound together in
groups of four, indicating that the normal process of spore development in
reproduction was never completed. Lycophytes in particular are able to
continue to reproduce asexually if their sexual reproductive structures—
spores—are damaged or rendered sterile.
It has been proposed that radiation-induced damage to their pollen-
and seed-producing structures contributed to the decline of the large
woody conifers and the vanishing of closed-canopied forests in the Late
Permian. The conifers were replaced by isoetalean lycophytes, the
quillworts (table 3.1), and the great majority of the lycophyte species
belonged to a single genus of isoetaleans, Pleuromeia . Thus the diverse Late
Permian forests of large, woody conifers was replaced by a monotonous
terrestrial plant cover of Pleuromeia, polelike lycophytes covered by a
sleeve of tiny leaves and topped by sporangia, standing only 1.5 to two
meters ( ve to 6.6 feet) high (Pfe erkorn 2004). The latitudinal pattern of
global deforestation in the Late Permian ts well with a kill mechanism of
radiation poisoning and hypothermia, perhaps assisted by acidosis, as
discussed in the previous sections of the chapter. The subsequent invasion
of herbaceous lycophytes ts well with an ecological scenario of forest
destruction and replacement by fast-growing, opportunistic weedy plants
that reproduced asexually. This scenario also explains why many of these
lycophytes are found to have colonized and grown in drought-prone
habitats that are usually o limits to the sexually reproducing lycophytes
that require water in the reproductive process (Visscher et al. 2004).
 The question then becomes: what sterilized the Late Permian land
plants and caused their reproductive structures to mutate? Visscher and
colleagues argue that anomalous changes in land-plant reproductive
structures were produced by exposure to high-energy UV-B radiation, and
that that exposure could only have been produced by the vanishing or
serious thinning of the Earth’s ozone shield in the Late Permian. And it did
not happen just once: fossil layers containing mutant spores and pollen
are also found in the Early Triassic, indicating a repeated cycle of collapse
and recovery of the ozone layer. Such a cyclic pattern is consistent with
the continued Siberian LIP eruptions and lethal gas emissions during the
Early Triassic (Visscher et al. 2004; Sun et al. 2012; Black et al. 2014).
In summary, the worst mass extinction in Earth history appears to have
been the result of a series of very bad events happening concurrently:
rst, the eruption in Siberia of the largest continental LIP in the
Phanerozoic Eon; second, the penetration of the rising Siberian super-
plume through the vast deposits of buried hydrocarbons (petroleum, coal)
and evaporites containing chlorine, bromine, and sulfur (halite,
anhydrite) in the Tunguska Basin (table 6.1); and third, the explosive
nature of the catastrophic phase of the eruptions, in which volcanic gases
were vented into the Earth’s atmosphere through the formation of over
6,400 explosive pipes in the Tunguska Basin ( gs. 6.7–6.8). As summarized
by the Russian Academy of Sciences geologist Alexander Polozov and his
colleagues, the “direct relationship between the Siberian Trap LIP and the
abundance and style of pipes is unlike any other LIP encountered through
the Earth’s history, and points to the thick sequences of salt and carbonate
rich sediments beneath the province playing a vital role in making the
end-Permian event so extreme compared to many of the other
LIP/extinction relationships” (Polozov et al. 2016).
 The catastrophic phase of the Siberian LIP eruption took place in only
200,000 years, from 252.3 million years ago (the very latest Permian25) to
252.1 million years ago (the earliest Triassic) (Shen et al. 2011; Payne and
Clapham 2012; Wang et al. 2014). On a geologic timescale, a mere 200,000
years is a blink of the eye, yet decades of research have progressively
con rmed the extremely short time span during which the largest loss of
animal biodiversity in Earth history occurred—and future research may
narrow that time interval even further.26 The mass extinction itself is now
estimated to have occurred in only 60,000 ± 48,000 years, based on the
radiometric dating of volcanic ash beds in South China that lie above and
below the extinction strata (Erwin 2015, xi, xiv). That is, the greatest die-
o of animal life in Earth history may have taken place in as little as
12,000 years, and no more than 108,000 years, during the 200,000-year
span of the catastrophic phase of the Siberian LIP eruption.
The Siberian LIP continued to erupt episodically all the way through
the Early Triassic (Bryan et al. 2010), and the southern part of the LIP in
Chelyabinsk experienced eruptive activity up to 243 million years ago, in
the Anisian Age of the Middle Triassic (Reichow et al. 2009). Thus there
was no respite from the end-Permian catastrophe—hot global climates
and low levels of oxygen in the atmosphere persisted throughout the ve
million years of the Early Triassic (Retallack 1999; Sun et al. 2012). The
end-Permian ecological catastrophe and the long duration of the Hot
Earth environmental conditions of the Early Triassic proved to be too
much for many of the typical life-forms of the Paleozoic to survive, and by
the dawn of the Middle Triassic the 200-million-year old ecological
structure of the Paleozoic world was gone forever.

THE EARTH POISONS ITS CHILDREN

We now know that in the end-Permian mass extinction the ultimate
horror in geologic history occurred: the Earth poisoned its own children.
Back in the 1970s, the once-popular “Gaia hypothesis” maintained that the
Earth was similar to a living organism in that its geochemical and
biochemical cycles seemed to act together to bu er and nurture the
presence of life on the planet (Ruse 2013). For example, for the past 4,560
million years, major uctuations have occurred in the temperature of the
Earth, in the carbon dioxide and oxygen content of its atmosphere, in the
salinity and other chemical components of its seas—yet all these
uctuations have occurred within upper and lower limits that have never
exceeded the ability of life to continue to survive on the planet. This
planetary physical-biological-homeostasis hypothesis was given the name
Gaia by its formulator, the British scientist James Lovelock, after the
ancient Greek goddess of the Earth.27 In the decades that followed the Gaia
proposal, some argued that the hypothesis was redundant in that
everyone already knew that the evolution of life has had major e ects on
the physical characteristics of the Earth and that these changes in the
physical characteristics of the Earth have had major e ects on the
subsequent evolution of life. Others have branded the hypothesis as
pseudoscience, particularly the more extreme proposal by some that the
planet Earth itself was a living organism (Ruse 2013).
Others have pointed out that life has had some extremely close calls to
annihilation in the past 4,560 million years—and the end-Permian mass
extinction was one of them. Rather than Gaia, the nurturing Greek
goddess, the planet Earth in the Late Permian acted more like the Hindu
god Shiva—a god that both creates and destroys entire worlds. Shiva is
usually depicted as a male humanoid gure with two legs but with six
arms, dancing barefoot on the back of a prostrate and squalling human
baby. In the case of the Late Permian, the world that was destroyed was
the Paleozoic world, the very world that had been created earlier in the
Cambrian Explosion and the Great Ordovician Biodiversi cation Event. In
the next two sections of the chapter, we will examine in detail—an
obituary?—the destruction of the Paleozoic world: rst the end of the
world in the oceans, and then the end of the world on land.

The End of the Paleozoic Marine World

What was the Paleozoic world like in the oceans? In a series of classic
papers in the 1980s and early 1990s, the University of Chicago28
paleontologist J. John (Jack) Sepkoski Jr. argued that the Phanerozoic
marine biosphere can be divided into three evolutionary faunas: the
Cambrian evolutionary fauna, which dominated the Cambrian Period; the
Paleozoic evolutionary fauna, which was dominant from the Ordovician
through the Permian; and the modern evolutionary fauna, which was
dominant in the Mesozoic and Cenozoic ( g. 6.10) (Sepkoski 1981, 1984,
1990; Sepkoski and Miller 1985). His research demonstrated that each
evolutionary fauna possessed distinctive evolutionary traits; for example,
each successive evolutionary fauna had a slower rate of diversi cation and
species turnover but a higher level of maximum diversity. Jack Sepkoski
and his student Arnie Miller demonstrated that these same three
evolutionary faunas can be reconstructed from ecological community data
(rather than just the taxonomic diversity data that Sepkoski had used in
his initial works), and also that the three evolutionary faunas have
characteristic ecological distributions along an onshore-o shore
environmental gradient—thus the evolutionary faunas are recurrent
associations in space as well as in time (Sepkoski and Miller 1985).
Subsequent research by other workers has demonstrated that each of the
three evolutionary faunas has a unique and characteristic guild structure,
with each successive evolutionary fauna occupying more “ecospace” and
having more guilds in total (Bambach 1983, 1985); that each evolutionary
fauna has its own tiering structure, both above and below the sediment-
water interface (Bottjer and Ausich 1986); and that each of the
evolutionary faunas is characterized by a unique series of periods of
ecological structural stability (Sheehan 1996).

FIGURE 6.10 Sepkoski’s classic plot of the familial diversity of marine animals in geologic time,
showing the temporal diversities of the Cambrian Evolutionary Fauna (Cm), the Paleozoic
Evolutionary Fauna (Pz), and the Modern Evolutionary Fauna (Md); see text for discussion. Geologic
timescale abbreviations: V, Ediacaran; barred-C, Cambrian; Ө, Ordovician; S, Silurian; D, Devonian;
C, Carboniferous; P, Permian; TR, Triassic; J, Jurassic; K, Cretaceous; T, Tertiary (Paleogene and
Neogene).

Source: From Paleobiology, volume 10, pp. 246–267, by J. J. Sepkoski Jr., “A Kinematic Model of
Phanerozoic Taxonomic Diversity: III. Post-Paleozoic Families and Mass Extinctions,” copyright ©
1984 The Paleontological Society. Reprinted with permission of Cambridge University Press.

Sepkoski’s three evolutionary faunas raised a few eyebrows at the time

they were published because their time designations were not standard.
First, the Cambrian Period is a part of the Paleozoic Era. Sepkoski’s
mathematical analyses showed that the Cambrian organisms are
evolutionarily and ecologically demonstrably di erent from the
Ordovician-Permian organisms, and subsequent research has shown that
the Ordovician-Permian evolutionary fauna originated in the Great
Ordovician Biodiversi cation Event (GOBE) (Webby et al. 2004; Servais et
al. 2010), hence postdating the Cambrian fauna, but how can one speak of
a “Paleozoic” evolutionary fauna without including the Cambrian
organisms (see the top and middle graphs in gure 6.11)? Second, he
lumped the Mesozoic marine fauna and the Cenozoic marine fauna
together as one “modern” evolutionary fauna, and most people do not
think of the extinct marine reptile Ichthyosaurus as a modern animal (see
animal-illustration number 23 in the bottom graph in gure 6.11). Yet the
point that Sepkoski argued was that the ecological structures of the
Mesozoic and Cenozoic organisms were the same; for example, a Mesozoic
ichthyosaur (a reptile) is ecologically equivalent to the convergently29
evolved Cenozoic porpoise (a mammal; see animal-illustration number 24
in the bottom graph in gure 6.11).
FIGURE 6.11 Sepkoski’s illustrations of some of the characteristic animals belonging to the
Cambrian, Paleozoic, and Modern Evolutionary Faunas. For geologic timescale abbreviations, see
gure 6.10.

Source: From Paleobiology, volume 10, pp. 246–267, by J. J. Sepkoski Jr., “A Kinematic Model of
Phanerozoic Taxonomic Diversity: III. Post-Paleozoic Families and Mass Extinctions,” copyright ©
1984 The Paleontological Society. Reprinted with permission of Cambridge University Press.
 In this book I consider the Paleozoic marine world to be just that—the
ecological world of the marine organisms characteristic of the Paleozoic
Era—and that that world consisted of two evolutionary phases: the initial
Cambrian Explosion diversi cation that produced the Cambrian
evolutionary fauna and the subsequent Great Ordovician
Biodiversi cation Event that produced the Ordovician-Permian
evolutionary fauna (top and middle graphs in gure 6.11). Thus, in table
6.4, I have combined Sepkoski’s two evolutionary faunas of Paleozoic age
into a single list of marine organisms characteristic of the Paleozoic
marine world, and contrast those with the marine organisms
characteristic of Sepkoski’s modern evolutionary fauna. Note that many of
the characteristic organisms of the Paleozoic world are extinct—but that
even the modern fauna have extinct components, such as the Mesozoic
marine reptiles like Ichthyosaurus.

TABLE 6.4 Characteristic marine organisms of the Paleozoic world and of the modern evolutionary
fauna.

Paleozoic World Fauna Modern Evolutionary Fauna

1. Rugose† corals 1. Rhizarian unicells
2. Tabulate† corals 2. Demosponges
3. Stenolaemate bryozoans 3. Scleractinian corals
4. Brachiopods 4. Gymnolaemate bryozoans
5. Monoplacophoran molluscs 5. Gastropod molluscs
6. Hyolithan† lophophorates 6. Bivalve molluscs
7. Ammonoid† cephalopods 7. Malacostracan crustaceans
8. Trilobite† arthropods 8. Echinoid echinoderms
9. Ostracod crustaceans 9. Chondrichthyan shes
10. Eocrinoid† echinoderms 10. Osteichthyan shes
11. Crinoid echinoderms 11. Marine reptiles† (Mesozoic)
12. Asteroid echinoderms 12. Marine mammals (Cenozoic)
13. Graptolite† hemichordates

Source: Modi ed from the classic evolutionary fauna analysis of Sepkoski (1984, 1990).
Note: Extinct taxa are marked with a dagger (†).

At rst glance (table 6.4) it might seem that the end-Permian mass
extinction triggered the extinction of older, less well adapted, clades of
organisms in favor in newly evolved, better adapted, clades. The actual
pattern of survival is more subtle than a simplistic older-clade-versus-
younger-clade scenario. If we take the list of characteristic organisms of
the Paleozoic world and of the modern evolutionary fauna (table 6.4) and
arrange those organisms in terms of their phylogenetic relationships
(table 6.5), then we quickly see that in many cases the Paleozoic world and
modern fauna organisms belong to the same clades. That is, the global
ecological restructuring that the end-Permian mass extinction triggered
was more of a within-clade than a between-clade phenomenon.

TABLE 6.5 Phylogenetic relationships of the marine organisms of the Paleozoic world (italicized)
and of the modern evolutionary fauna (underlined).

Eukarya (eukaryote cells)

– Bikonta
– – Rhizaria
– – – Actinopoda
– – – – Radiolaria
– – – Foraminifera
– – – – calcareous benthic foraminifera
– – – – agglutinated benthic foraminifera
– – – – planktic calcareous foraminifera
– Unikonta
– – Opisthokonta
– – – Choanozoa
– – – – Metazoa (animals)
– – – – – Demospongiae
– – – – – – Stromatoporoidea†
– – – – – – Sclerospongea
– – – – – Hexactinellida (glass sponges)
– – – – – Eumetazoa
– – – – – – Cnidaria (jelly sh, corals, and kin)
– – – – – – – Zoantharia (corals)
– – – – – – – – Tabulata† (tabulate corals)
– – – – – – – – Rugosa† (horn corals)
– – – – – – – – Heterocorallia†
– – – – – – – – Scleractinia
– – – – – – Bilateria
– – – – – – – Protostomia (animals with protostome development)
– – – – – – – – Lophotrochozoa
– – – – – – – – – Lophophorata
– – – – – – – – – – Bryozoa (moss animals)
– – – – – – – – – – – Stenolaemata
– – – – – – – – – – – Gymnolaemata
– – – – – – – – – – – – Ctenostomata
– – – – – – – – – – Phoronozoa
– – – – – – – – – – – Phoronida
– – – – – – – – – – – Hyolitha†
– – – – – – – – – – – Brachiopoda (lampshells)
– – – – – – – – – – – – Linguliformea
– – – – – – – – – – – – Rhynchonelliformea
– – – – – – – – – – – – – Strophomenata†
– – – – – – – – – – – – – Rhynchonellata
– – – – – – – – – – – – – – Orthida†
– – – – – – – – – – – – – – Pentamerida†
– – – – – – – – – – – – – – Atrypida†
– – – – – – – – – – – – – – Athyridida†
– – – – – – – – – – – – – – Spiriferida†
– – – – – – – – – – – – – – Spiriferinida†
– – – – – – – – – – – – – – Rhynchonellida
– – – – – – – – – – – – – – Terebratulida
– – – – – – – – Eutrochozoa
– – – – – – – – – Spiralia
– – – – – – – – – – Annelida (segmented worms)
– – – – – – – – – – – Polychaeta
– – – – – – – – – – Mollusca
– – – – – – – – – – – Eumollusca
– – – – – – – – – – – – Polyplacophora (chitons)
– – – – – – – – – – – – Conchifera
– – – – – – – – – – – – – Monoplacophora
– – – – – – – – – – – – – Ganglioneura
– – – – – – – – – – – – – – Visceroconcha
– – – – – – – – – – – – – – – ?Tentaculitoidea†
– – – – – – – – – – – – – – – Cephalopoda (octopus, squid, and kin)
– – – – – – – – – – – – – – – – Ammonoidea† (shelled cephalopods)
– – – – – – – – – – – – – – – – Nautiloidea (shelled cephalopods)
– – – – – – – – – – – – – – – – Coleoidea (octopus, squid, and kin)
– – – – – – – – – – – – – – – Gastropoda (snails)
– – – – – – – – – – – – – – Diasoma
– – – – – – – – – – – – – – – Rostroconcha†
– – – – – – – – – – – – – – – Bivalvia (clams, mussels, and kin)
– – – – – – – – Cuticulata
– – – – – – – – – Ecdysozoa
– – – – – – – – – – Panarthropoda
– – – – – – – – – – – Arthropoda
– – – – – – – – – – – – Cheliceriformes
– – – – – – – – – – – – – Eurypterida† (sea scorpions)
– – – – – – – – – – – – – Trilobita†
– – – – – – – – – – – – – Merostomata (horseshoe crabs)
– – – – – – – – – – – – – Pycnogonida (sea spiders)
– – – – – – – – – – – – Mandibulata
– – – – – – – – – – – – – Myriapoda
– – – – – – – – – – – – – Pancrustacea (crustaceans)
– – – – – – – – – – – – – – Maxillopoda
– – – – – – – – – – – – – – – Ostracoda
– – – – – – – – – – – – – – – Copepoda
– – – – – – – – – – – – – – – Cirripedia (barnacles)
– – – – – – – – – – – – – – Malacostraca (crabs, lobsters, krill, and kin)
– – – – – – – Deuterostomia (animals with deuterostome development)
– – – – – – – – Echinodermata
– – – – – – – – – Eocrinoidea†
– – – – – – – – – Crinoidea (sea lilies)
– – – – – – – – – Blastoidea†
– – – – – – – – – Asteroidea (star sh)
– – – – – – – – – Echinoidea (sea urchins)
– – – – – – – – – Holothuroidea (sea cucumbers)
– – – – – – – – Pharyngotremata
– – – – – – – – – Hemichordata
– – – – – – – – – – Graptolithina†
– – – – – – – – – Chordata
– – – – – – – – – – Myomerozoa
– – – – – – – – – – – Craniata
– – – – – – – – – – – – Conodonta†
– – – – – – – – – – – – Vertebrata
– – – – – – – – – – – – – Gnathostomata (jawed vertebrates)
– – – – – – – – – – – – – – Placodermi† (armored shes)
– – – – – – – – – – – – – – Chondrichthyes (sharks, rays, and kin)
– – – – – – – – – – – – – – Acanthodii† (spine- n shes)
– – – – – – – – – – – – – – Osteichthyes (bony shes)
– – – – – – – – – – – – – – – Actinopterygii (ray- n shes)
– – – – – – – – – – – – – – – Sarcopterygii (lobe- n shes)
– – – – – – – – – – – – – – – – Actinistia (coelacanths)
– – – – – – – – – – – – – – – – Dipnoi (lung shes)
– – – – – – – – – – – – – – – – Tetrapodomorpha
– – – – – – – – – – – – – – – – – Tetrapoda (limbed vertebrates; see table 6.8)

Source: Phylogenetic classi cation modi ed from Lecointre and Le Guyader (2006) and Benton
(2015).
Note: Extinct taxa are marked with a dagger (†).

Let us examine the pattern of the collapse of the Paleozoic world in its
evolutionary context (table 6.5). To begin, the clade of rhizarian
unicellular organisms contains organisms that perished in the end-
Permian mass extinction—namely, the calcareous bottom-dwelling
foraminifera that died in the acidi cation of the late Permian oceans.
However, this clade also contains elements of the modern evolutionary
fauna—the highly successful siliceous radiolarians, which oat up in the
water column, and the bottom-dwelling agglutinated foraminifera, both of
which groups were largely immune to the e ects of oceanic acidi cation
because neither has shells composed of calcium carbonate—as well as the
calcareous oating foraminifera that evolved after the end-Permian mass
extinction was over. Skipping down the list, the clade of the zoantharian
corals contains both the Paleozoic world heterocorals, tabulates, and
rugosans that did not survive into the Mesozoic and the scleractinian
corals that evolved in the Triassic, which are the main components of
coral reefs today. The clade of the bryozoans, the tiny colonial moss
animals, contains both the highly calci ed stenolaemates that perished in
the acidic late Permian seas and the uncalci ed ctenostomate
gymnolaemates that survived, along with the calcareous cheilostome
gymnolaemates that would evolve later in the Jurassic. Skipping further
down the list, the clade of the cephalopod molluscs contains both the
numerous ammonoid shelled cephalopods that died from hypercapnia in
the carbon-dioxide-poisoned terminal Permian oceans and the nautiloid
shelled cephalopods that, with their better-bu ered respiratory
physiologies, survived (Knoll et al. 2007). Likewise, in the clade of the
deuterostome echinoderms, the burrowing echinoids and holothurians,
with their high metabolic activity levels and their tolerance of high
carbon-dioxide levels within the sediment, survived di erentially over the
immobile or slow-moving eocrinoids, blastoids, and numerous crinoids
with poorly-bu ered respiratory physiologies.
Thus, in many cases, the e ects of the end-Permian environmental
catastrophe triggered within-clade ecological replacements, not wholesale
replacement of one evolutionary clade by another. However, some clades,
such as the brachiopod lampshells, were almost universal losers in the
collapse of the Paleozoic world fauna (table 6.5). Each biodiversity crisis in
the Paleozoic triggered diversity losses in the brachiopods, and even some
of the rhynchonelliform groups that managed to survive the end-Permian
mass extinction—the athyridids and spiriferinids—later succumbed to
extinction in the Mesozoic and are not alive today. Only the
rhynchonellids and terebratulids survive today, a small remnant of the
once highly diverse brachiopod shell sh fauna of the Paleozoic world.30
Another way of understanding the magnitude of the collapse of the
Paleozoic world is to examine the pattern of ecological replacements
triggered by the end-Permian mass extinction. Table 6.6 shows nine major
ecological roles of life, or megaguilds, for marine life (Droser et al. 2000).
Within each megaguild are listed the organisms of the Paleozoic world
that lived those ecological roles and the organisms of the modern
evolutionary fauna that replaced the Paleozoic world fauna in those
ecological roles following the end-Permian mass extinction.
TABLE 6.6 Ecological replacement of the fauna of the Paleozoic world by the modern evolutionary
fauna in the marine realm following the end-Permian mass extinction.

I. Pelagic Mode of Life (living within oceanic waters)

Detritivores Paleozoic World: chitinozoans†; tentaculitoid† molluscs; trilobite† arthropods;
conodont† craniates; graptolite† hemichordates
Modern Fauna: radiolarian and foraminiferan rhizarians; pteropod gastropods; krill
and copepod crustaceans
Carnivores Paleozoic World: ammonoid† cephalopods; eurypterid† arthropods; armored†, spine-
n†, and lobe- n shes
Modern Fauna: nautiloid and squid cephalopods; sharks and rays; ray- n shes;
marine reptiles†
II. Epibenthic Mode of Life (living on the surface of the sea bottom)
Sessile Paleozoic World: stromatoporoid† demosponges; stenolaemate bryozoans;
Detritivores acrotretid†, strophomenate†, orthid†, pentamerid†, atrypid†, athyridid†,
spiriferid†, and spiriferinid† brachiopods; hyolithan† lophophorates; bivalve
molluscs (low diversity); eocrinoid†, blastoid† and crinoid echinoderms
Modern Fauna: sclerosponge demosponges; glass sponges; ctenostomate bryozoans;
bivalve molluscs (high diversity); barnacle crustaceans
Sessile Paleozoic World: heterocoral†, rugose†, and tabulate† corals
Carnivores
Modern Fauna: scleractinian corals
Mobile Paleozoic World: monoplacophoran and gastropod molluscs, trilobite† arthropods;
Detritivores ostracod crustaceans; armored† shes
Modern Fauna: gastropod and bivalve molluscs; malacostracan crustaceans; echinoid
echinoderms; ray- n shes
Mobile Paleozoic World: monoplacophoran and gastropod molluscs; merostomate
Herbivores arthropods; ostracod and malacostracan crustaceans
Modern Fauna: polychaete annelids; polyplacophoran and gastropod molluscs;
malacostracan crustaceans; echinoid echinoderms; ray- n shes
Mobile Paleozoic World: ammonoid† cephalopods; malacostracan crustaceans; asteroid
Carnivores echinoderms
Modern Fauna: polychaete annelids; gastropod molluscs; octopus cephalopods;
pycnogonid arthropods; malacostracan crustaceans; asteroid echinoderms
III. Endobenthic Mode of Life (living within the sediment on the sea bottom)
Mobile Paleozoic World: linguliformean brachiopods; rostroconch† and bivalve molluscs
Detritivores (low diversity); trilobite† arthropods; conodont† craniates
Modern Fauna: polychaete annelids; bivalve molluscs (high diversity); echinoid and
holothurian echinoderms
Mobile Paleozoic World: polychaete annelids; merostomate arthropods
Carnivores
Modern Fauna: polychaete annelids; gastropod molluscs; malacostracan crustaceans

Source: Data from Levinton (1982), Bambach (1983), and McGhee (2011).
Note: Extinct taxa are marked with a dagger (†).

In the oceans of the Earth, the pelagic ecological structure of the

Paleozoic world’s oceanic waters was totally destroyed—essentially all of
the characteristic organisms of that world are extinct (table 6.6). The
collapse of the Paleozoic world ecological trophic pyramid in the oceans
was total: both the zooplankton—the tiny detritivores—and the larger
swimming predatory fauna were essentially eliminated. Only one
extremely rare remnant of the lobe- n shes survives in our oceans today
—the species Latimeria chalumnae (the famous “living fossil” coelacanth,
which is an actinistian sarcopterygian; see table 6.5). In contrast to the
Paleozoic, our modern oceans are lled with a zooplankton of rhizarians,
gastropods, and crustaceans and a predatory fauna of squids, sharks, and
ray- n shes. Only the marine reptiles of the Mesozoic are today extinct—
but their ecological niche has been convergently lled by the evolution of
modern-day marine mammals like killer whales and porpoises.31
On the sea bottom, the ecological structure of immobile, sessile benthic
organisms of the Paleozoic world was totally destroyed—essentially all of
the characteristic organisms of that world are also extinct (table 6.6).
Again, both the detritivore and carnivore levels of the trophic structure of
the Paleozoic ecosystem were destroyed. Instead of a bottom-dwelling
fauna almost totally comprised of vast expanses of sessile brachipod
lampshells (some of them giants; see gure 4.9) and stalked echinoderms,
and huge reefs comprised of massive calcareous stromatoporoid
demosponges, tabulate corals, and horn corals, our modern ocean bottoms
are populated by survivors of the latest Permian acid seawaters—
sclerosponges, glass sponges, and ctenostome bryozoans—and active
organisms with well-bu ered respiratory physiologies—bivalve molluscs
and crustaceans. Our modern reefs are dominantly comprised of
scleractinian corals, corals that evolved from soft-bodied, skeletonless
survivors of the Permian acid seas.
The ecological structure of the mobile bottom-dwelling organisms of
the Paleozoic world was not as severely a ected as that of the immobile
organisms, probably because these organisms, being mobile, could at least
try to ee late Permian toxic water conditions as they developed—an
option the sessile organisms did not have (table 6.6). Thus, more of the
ancient Paleozoic mobile bottom-dwelling fauna have survived to the
present day, including the monoplacophoran molluscs, horseshoe crabs,
ostracod and Paleozoic malacostracan crustaceans, and the star sh. Still,
these surviving elements of the Paleozoic world are today vastly
outnumbered by modern mollusc faunas of chitons, snails, bivalves, and
octopuses, sea spider and modern crustacean arthropods, sea urchins, and
ray- n shes (compare tables 6.5 and 6.6).
Living within the sediment on the sea bottom is a more restrictive
mode of life, so fewer organisms have evolved adaptations to actively
burrow in bottom muds and silts. Yet even here the ecological structure of
burrowing organisms of the Paleozoic world was almost totally destroyed
(table 6.6). The burrowing rostroconch molluscs, trilobites, and conodonts
of the Paleozoic are all extinct, and the living linguliform lampshells are
rare. Instead, our modern seas are dominated by burrowing polychaete
worms, numerous bivalve molluscs and snails, and echinoderm sea
urchins and sea cucumbers. All of these animals are descendants of active
ancestors with a high tolerance for high carbon-dioxide levels, and low
oxygen levels, within the marine muds of the latest Permian seas.
Although the end-Permian eruption of the Siberian LIP and the
environmental catastrophe that followed sealed the fate of the Paleozoic
marine world, the ecological structure of the Paleozoic world had already
su ered several major shocks prior to its nal collapse. The rst
biodiversity crisis to strike the Paleozoic world was the end-Ordovician
(Hirnantian Age, table 1.3) mass extinction, which was triggered by the
Late Ordovician glaciation (table 1.2). Sepkoski’s data revealed that the
Paleozoic world lost 22 percent—almost a quarter—of the animal families
alive in the world’s oceans in the Late Ordovician, a loss that is clearly
visible in the sharp drop in the standing diversity curves shown in gures
6.10 and 6.11 at the end of the Ordovician. However, the Ordovician-
Permian evolutionary fauna not only survived the extinction, it also
rebounded in diversity in the Silurian back to a level similar to that
present before the end-Ordovician mass extinction. This pattern of
survival and recovery was not seen in the Cambrian evolutionary fauna,
where this older Paleozoic fauna never recovered its previous diversity
level (top graph in gure 6.11); thus the Paleozoic world permanently lost
a small amount of its marine diversity at the end of the Ordovician.
The next biodiversity crisis in the Paleozoic world occurred in the Late
Silurian (Ludfordian Age, table 1.3) and triggered a much smaller drop in
marine diversity ( gs. 6.10, 6.11). As in the end-Ordovician crisis, the
Ordovician-Permian evolutionary fauna survived and recovered its
diversity levels postcrisis in the Early Devonian. Although both the end-
Ordovician and Late Silurian crises triggered extinctions and diversity
losses, both had minimal ecological impact in the Paleozoic marine world
(table 1.3) (McGhee et al. 2013).
 The same is not true of the Late Devonian (Frasnian Age, table 1.3)
biodiversity crisis. This next crisis struck the Paleozoic world hard, and
although Sepkoski’s data show that it triggered a loss of marine animal
families very similar to the end-Ordovician crisis ( gs. 6.10, 6.11), the
ecological impact of the crisis was devastating (McGhee 1996, 2013;
McGhee et al. 2004, 2013). The Paleozoic marine world never fully
recovered from the marine diversity losses it su ered at the end of the
Frasnian Age. A rediversi cation phase began in the following Famennian
Age of the Late Devonian, but it was terminated by the end-Devonian
crisis before previous diversity levels could be attained (middle graph in
gure 6.11). All in all, the e ects of the twin Late Devonian crises
eliminated the chitinozoans32 and tentaculitoids in the zooplankton, the
atrypid brachiopods on the sea oor, the stromatoporoid sponges in the
reef ecosystems, and the armored shes. A smaller and slower
rediversi cation followed the end-Devonian crisis, but it also was
terminated by the Early Carboniferous biodiversity crisis (Serpukhovian
Age, table 1.3). It is known that both the Famennian and Serpukhovian
crises were triggered by the stepwise onset of the Late Paleozoic Ice Age
and, as I argued in chapter 1, it is likely that the Frasnian biodiversity
crisis was also.
In summary, the Paleozoic marine world su ered a series of ecological
shocks largely triggered by the various glaciation phases of the Late
Paleozoic Ice Age. Still, Sepkoski’s mathematical analyses showed that the
Ordovician-Permian evolutionary fauna of the Paleozoic world constituted
a stable ecological structure that had persisted in the oceanic realm for
over 200 million years—until the end-Permian mass extinction (middle
graph in gure 6.11). The ancient tabulate and rugose reef coral faunas
had su ered major setbacks in the Frasnian and Serpukhovian crises, but
they survived. These hardy corals now nally encountered an
environmental crisis too severe even for them to survive, and they
perished. The trilobites are found in marine rocks from the Cambrian
through the Permian, and are virtually synonymous with the word
“Paleozoic.” They, and their close cousins the sea scorpions, were driven to
extinction at the close of the Permian and would evolve no further. The
diverse strophomenate, orthid, and spiriferid brachiopod lampshell
groups were wiped out, along with their conical-shell lophophorate
cousins, the hyoliths (Moysiuk et al. 2017), eliminating most of the typical
shell sh of the Paleozoic seas. Our modern seas are dominated by
molluscan shell sh, but even the molluscs su ered the extinction of the
long-lived rostroconchs, and the highly active ammonoid predators just
barely survived. Among the deuterostome animals, the benthic gardens of
blastoid echinoderms perished while their close cousins the crinoids, the
sea lilies, came very close to extinction. In the chordates, the pelagic
graptolites and spine- n shes died, but the tiny conodonts managed to
survive the Changhsing, only to succumb to extinction in the Triassic.
The magnitude of the end-Permian crisis was so great that it reset the
global pattern of both extinction rates (Van Valen 1984) and speciation
rates (Sepkoski 1998) in the marine ecosystems of the Earth. The hot, acid,
and poisonous seawaters of the latest Permian oceans were too much for
the Paleozoic fauna, and the 200-million-year-old ecological structure of
the Ordovician-Permian evolutionary fauna of the Paleozoic marine world
collapsed—and this time it never recovered (middle graph in gure 6.11;
table 6.6).

The End of the Paleozoic Terrestrial World

At present, there exists no mathematical division of the terrestrial animal

biosphere into evolutionary faunas equivalent to those produced for the
marine animal biosphere by Sepkoski (Sepkoski 1984, 1990). However, Jack
Sepkoski did write in 199033 that he considered the University of Bristol
vertebrate paleontologist Mike Benton’s 1985 “terrestrial tetrapod
assemblages” (Benton 1985) to be similar in ecological nature to his
marine evolutionary faunas. Benton argued that three distinct ecological-
evolutionary assemblages of terrestrial tetrapod vertebrates could be
recognized in geologic time: rst, a Paleozoic labyrinthodont-synapsid-
parareptile assemblage that was terminated in the end-Permian
extinction; second, a Mesozoic diapsid-pterosaur-dinosaur assemblage
that was terminated in the end-Cretaceous extinction; and third, the
Cenozoic assemblage of modern lissamphibians, turtles, lizards,
crocodiles, birds, and mammals.
Who were these animals? Similar to tables 6.4 and 6.5 for marine
animals, in tables 6.7 and 6.8 I have listed the characteristic land-dwelling
vertebrates of the Paleozoic and Mesozoic worlds. The labyrinthodont
component of Benton’s 1985 Paleozoic tetrapod assemblage referred to
the ancient amphibian-like animals that existed in the Carboniferous and
Permian, which consisted of older paraphyletic or “partial” evolutionary
groupings that we now recognize as basal batrachomorphs (then called
temnospondyls, and ancestral to our modern amphibians), basal
lepospondyls (microsaurs), and basal reptiliomorphs (anthracosaurs;
compare tables 6.7 and 6.8). These ancient animals were joined by the
early amniote animals—basal synapsids (pelycosaurs), advanced therapsid
synapsids, and the diapsid reptiles—to constitute the typical fauna that
one would have encountered on the continents of the Paleozoic world.

TABLE 6.7 Characteristic land-dwelling vertebrate faunas of the Paleozoic world and of the
Mesozoic world.

Paleozoic World Fauna Mesozoic World Fauna

1. “temnospondyls” (batrachomorphs) 1. cynodont/mammal synapsids
2. “microsaurs” (lepospondyls) 2. marine reptiles
3. “anthracosaurs” (reptiliomorphs) 3. archosaurian reptiles
4. “pelycosaurs” (synapsids) 4. pterosaurian reptiles
5. therapsids (synapsids) 5. dinosaurian reptiles
6. parareptiles (reptiles)

Note: Older paraphyletic tetrapod group names are in quotation marks; see table 6.8 for a
phylogenetic classi cation.

TABLE 6.8 Phylogenetic relationships of the land-dwelling vertebrates of the Paleozoic world
(italicized) and of the Mesozoic world (underlined).

TETRAPODA (limbed vertebrates)

– basal tetrapods
– Neotetrapoda
– – BATRACHOMORPHA (ancestors of amphibians)
– – – basal batrachomorphs (“temnospondyls”)
– – – Capitosauria†
– – – Trematosauria†
– – – Eryopidae† (giants)
– – – Lissamphibia (living amphibians)
– – Lepospondyli†
– – – basal lepospondyls (“microsaurs”)
– – – Nectridea†
– – – Aistopoda†
– – REPTILIOMORPHA (ancestors of amniote tetrapods)
– – – basal reptiliomorphs (“anthracosaurs”)
– – – Seymouriamorpha†
– – – Diadectomorpha† (giants)
– – – AMNIOTA (amniote tetrapods, conquerors of the land)
– – – – SYNAPSIDA (ancient ancestors of mammals)
– – – – – basal synapsids (“pelycosaurs”)
– – – – – Eothyrididae†
– – – – – Caseidae† (giants)
– – – – – Varanopidae†
– – – – – Ophiacodontidae† (giants)
– – – – – Edaphosauridae† (giants)
– – – – – Sphenacodontidae† (giants)
– – – – – THERAPSIDA (closer ancestors of mammals)
– – – – – – Biarmosuchia† (basal therapsids)
– – – – – – Dinocephalia† (giants)
– – – – – – Dicynodontia†
– – – – – – Gorgonopsia†
– – – – – – Therocephalia
– – – – – – – Cynodontia
– – – – – – – – MAMMALIA (mammals)
– – – – REPTILIA (ancestors of reptiles)
– – – – – basal reptiles
– – – – – – PARAREPTILIA
– – – – – – – Mesosauridae†
– – – – – – – Procolophonidae†
– – – – – – – Pareiasauridae† (giants)
– – – – – – EUREPTILIA
– – – – – – – Captorhinidae† (giants)
– – – – – – – DIAPSIDA
– – – – – – – – Weigeltisauridae†
– – – – – – – – Testudinata (turtles)
– – – – – – – – LEPIDOSAUROMORPHA
– – – – – – – – – Ichthyosauria† (marine reptiles; ichthyosaurs)
– – – – – – – – – Sauropterygia† (marine reptiles; plesiosaurs and placodonts)
– – – – – – – – – Lepidosauria (living lizards and snakes)
– – – – – – – – ARCHOSAUROMORPHA
– – – – – – – – – Rhynchosauria†
– – – – – – – – – Archosauria
– – – – – – – – – – Proterosuchidae†
– – – – – – – – – – Crurotarsi (ancestors of living crocodiles)
– – – – – – – – – – – Phytosauridae†
– – – – – – – – – – – Stagonolepididae† (aetosaurs)
– – – – – – – – – – – Rauisuchia†
– – – – – – – – – – – Crocodylomorpha (crocodiles)
– – – – – – – – – – Ornithodira
– – – – – – – – – – – Pterosauria† ( ying reptiles; pterodactyls and rhamphorhynchids)
– – – – – – – – – – – DINOSAURIA (dinosaurs, including living birds)
– – – – – – – – – – – – Saurischia
– – – – – – – – – – – – – Sauropodomorpha†
– – – – – – – – – – – – – Theropoda
– – – – – – – – – – – – Ornithischia†

Source: Phylogenetic classi cation modi ed from Benton (2015).

Note: Extinct taxa are marked with a dagger (†). Older paraphyletic tetrapod group names are in
quotation marks; major clades are in capitals.

In contrast to the Paleozoic, the Mesozoic world land-dwelling

vertebrate fauna is comprised almost entirely of reptiles (tables 6.7 and
6.8). Only the cynodonts and their descendants, the mammals, remained
from the once ecologically diverse and numerous land-dwelling synapsid
faunas of the Paleozoic world (Abdala and Ribeiro 2010). What happened?
The synapsid vertebrates (a group that includes us, the modern humans)
were the dominant and most advanced land animals of the world in the
Paleozoic. In the Mesozoic world, the synapsids had been ecologically
replaced by reptiles—and very advanced reptiles indeed. Not only did the
reptiles take over the continents of the Earth, they also convergently re-
evolved ns and other aquatic adaptations34 and invaded the marine realm
(the ichthyosaurs, plesiosaurs, and kin; table 6.8) and even modi ed their
forelimbs into wings and invaded the aerial realm as well (the pterosaurs;
table 6.8). The rst ying vertebrate animals in Earth history were
reptiles, not synapsids (which only managed to evolve wings much later in
the Cretaceous35).
The evolutionary and ecological innovativeness of the clade of the
reptiles (table 6.8) in the Mesozoic terrestrial world is better revealed
when we examine the ecological structure of that world in table 6.9. We
have previously considered the ecological replacements that took place in
the pelagic realm when the Paleozoic marine world ended (table 6.6). Note
at the top of table 6.9 that the “marine reptiles” listed in table 6.6 are now
identi ed as two groups of reptiles: the diapsid ichthyoperytians and the
lepidosaurian sauroptergyians (compare tables 6.8 and 6.9). At the bottom
of table 6.9, note that the reptiles had already produced some
evolutionary experiments in ight even within the Paleozoic terrestrial
world—namely, the gliding weigeltisaurid diapsids. These interesting
animals had bodies that were dorsoventrally attened and laterally
widened. Within their bodies their ribs were attened and greatly
elongated laterally away from the anteroposterior axis of the vertebral
column. In midair these animals looked like a circular discus or frisbee in
ight—but with a pair of forelimbs and a head attached on one side of the
discus and a pair of hindlimbs and a tail on the other ( g. 6.12). This
identical morphology has been repeatedly, convergently, evolved within
the diapsid reptiles and can be seen today in the modern gliding lizard
Draco melanopogon.36Although innovative, the weigeltisaurid gliders
apparently were not innovative enough because they still perished at the
end of the Changhsingian in the end-Permian mass extinction.

TABLE 6.9 Ecological replacement of the vertebrate fauna of the Paleozoic world by the Mesozoic
world fauna in the terrestrial realm in the Late Triassic.

I. Vertebrates in the Oceans

Carnivores Paleozoic World: armored†, spine- n†, and lobe- n shes
 Mesozoic World: sharks and rays; ray- n shes; ichthyosaurian† and
sauropterygian† lepidosauromorphs
II. Vertebrates in Freshwater Rivers and Lakes
Detritivores Paleozoic World: armored† shes
Mesozoic World: ray- n shes
Carnivores Paleozoic World: armored† and lobe- n shes, capitosaurian† and trematosaurian†
batrachomorphs; nectridean lepospondyls†; seymouriamorph† reptiliomorphs;
mesosaurian† parareptiles
Mesozoic World: ray- n shes; phytosaurid† archosaurs; spinosaurid† theropod
dinosaurs
III. Vertebrates on Land
Herbivores Paleozoic World: diadectomorph† reptiliomorphs; caseid†, edaphosaur†,
dinocephalian†, dicynodont† and cynodont synapsids; procolophonid† and
pareiasaurid† parareptiles; captorhinid† reptiles
Mesozoic World: rhynchosaur† archosauromorphs; aetosaur† archosaurs;
sauropodomorph† and ornithischian† dinosaurs
Carnivores Paleozoic World: eryopid† batrachomorphs; varanopid†, ophiacodont†,
sphenacodont†, and gorgonopsid† synapsids
Mesozoic World: cynodont and mammalian synapsids (low diversity); proterosuchid†
and rauisuchian† archosaurs; theropod dinosaurs (high diversity)
IV. Vertebrates in the Air
Paleozoic World: gliding weigeltisaurid† diapsids
Mesozoic World: ying pterosaurs†; avian dinosaurs (birds)

Note: Extinct taxa are marked with a dagger (†).

FIGURE 6.12 A Paleozoic world gliding reptile, the weigeltisaurid Coelurosauravus.

Source: Illustration by Mary Persis Williams. Modi ed and redrawn from Steyer (2012).

The great ying reptiles of the Mesozoic world—the pterosaurs (table

6.9)—were not descendants of the early weigeltisaurid gliders of the
Permian. These animals were ornithodirans, highly derived archosaurian
reptiles closely related to the famous dinosaurs (table 6.8). Unlike their
distant cousins the weigeltisaurid gliders, the pterosaurs had true wings
and could y. Their wings were modi ed forelimbs, originally used solely
for walking but now used also as lift-generating structures, that consisted
of a single nger—the fourth, or ring, nger—that was vastly elongated
and to which a membrane of skin was attached and stretched all the way
to the side of the animal’s body ( g. 6.13). Their membranous wings were
thus somewhat similar to those of modern bats, although a bat’s wing is
made by elongating all ve ngers, not just the fourth. Later in the
Mesozoic world, the theropod dinosaur predators would produce a ying
competitor to their pterosaur cousins—the animals we know today as
birds (table 6.8). The wing of a bird, an avian dinosaur, di ers even more
from the membranous wings of pterosaurs and bats. It consists of
numerous elongated ight feathers that are attached to its arm, but still
the basic structure is a modi cation of the forelimb of the animal that was
originally used for walking.

FIGURE 6.13 A Mesozoic world ying reptile, the pterosaur Eudimorphodon.

Source: Illustration by Mary Persis Williams. Modi ed and redrawn from Palmer et al. (2012).

In freshwater ecosystems on land, the rivers and lakes of the Paleozoic

world contained a diverse fauna of ancient shes, ancient ancestors of the
amphibians (batrachomorphs), amphibian-like lepospondyls, and
advanced seymouriamorph reptiliomorphs that were very closely related
to the amniotes (table 6.8). The only major group of reptiles in freshwater
ecosystems in the Paleozoic world were the mesosaur parareptiles (tables
6.8 and 6.9). In contrast, the rivers and lakes of the Mesozoic world were
lled with ray- n shes, which exploded in diversity in the Triassic, the
 sh-eating phytosaurs, a group of advanced archosaurs related to
modern-day crocodiles, and the sh-eating theropod dinosaurs—the
spinosaurs (tables 6.8 and 6.9).
The megaguild of herbivores in the Paleozoic terrestrial world
comprised a diverse assemblage of diadectomorph reptiliomorphs (the
giant Diadectes maximus we considered in chapter 4; see g. 4.5) and, in the
synapsid clade, the caseids, edaphosaurs, dinocephalians (the giant
Moschops dinocephalians were the largest land vertebrates in the
Paleozoic; g. 4.7), and the advanced dicynodonts and cynodont
herbivores. The reptile clade was also diverse in the herbivore megaguild
of the Paleozoic terrestrial world, with procolophonid and pareiasaur
parareptiles (relatives of the mesosaurs in freshwater ecosystems) and
captorhinids (table 6.9). In contrast, the herbivore megaguild of the early
Mesozoic terrestrial world was much less diverse, dominated by
rhynchosaur archosauromorphs, more advanced aetosaur archosaurs, and
the early dinosaurs—both sauropodomorphs (later to become the largest
land animals in Earth history) and ornithischians (table 6.9). The diverse
synapsid herbivores of the Paleozoic world were gone.
The Paleozoic carnivore megaguild was composed of eryopid early
amphibians (including the giant Eryops megacephalus; g. 4.4),
pelycosaurian synapsids (varanopids, ophiacondonts, and sphenacodonts,
some of which were giants; see g. 4.6), and advanced gorgonopsid
synapsids ( g. 6.14). The gorgonopsid predators in particular were lithe,
catlike animals with elongated fang teeth very similar to the saber teeth
to be evolved much later by the true cats (table 6.9). In the Mesozoic
world, the dominant land-dwelling carnivores were proterosuchid and
rauisuchian archosaurs (the latter a relative of the phytosaurian
carnivores in freshwater ecosystems; tables 6.8 and 6.9) and the theropod
dinosaurs (later to produce the famous Tyrannosaurus rex and other very
large carnivores). The once-dominant synapsid carnivores were rare—only
the cynodonts and, near the end of the Triassic, the rst of the true
mammals.

FIGURE 6.14 A Paleozoic world synapsid carnivore, the gorgonopsid Lycaenops.

Source: Illustration by Mary Persis Williams. Modi ed and redrawn from Steyer (2012).

Just as with the end of the Paleozoic marine world, the end of the
Paleozoic terrestrial world was not a single instantaneous event con ned
to the end of the Changhsingian Age in the Permian. The Paleozoic marine
world experienced a series of biodiversity crises before the coup de grâce
of the end-Permian mass extinction. The Paleozoic marine ecological
structure was older than the Late Paleozoic Ice Age, and the various glacial
phases of the ice age usually triggered extinctions and biodiversity losses
in the marine fauna. In essence, the Late Paleozoic Ice Age presented a
series of environmental and ecological challenges that the Paleozoic world
marine fauna encountered and survived, even while losing overall
diversity with time (middle graph in gure 6.11). Thus we can hypothesize
that the Paleozoic marine world might have continued to survive past the
end of the Late Paleozoic Ice Age if it had not been for the global
environmental catastrophe produced by the Emeishan and Siberian LIP
eruptions.
In contrast, the Paleozoic terrestrial world fauna was largely the
product of climatic conditions produced by the Late Paleozoic Ice Age, and
land vertebrates ourished in the equatorial rainforests of the Earth
during the Carboniferous and Permian. However, the Late Paleozoic Ice
Age began to wane in the Early Permian, as discussed in the last chapter,
and the P2-to-P3 interpulse greenhouse period triggered a major
contraction of the ever-warm, ever-wet, equatorial zone during the late
Artinskian–early Roadian interval of the Permian. By the beginning of the
Middle Permian Epoch, some elements of the typical Paleozoic terrestrial
world fauna had died out, including the eryopid basal amphibians, the
nectridean and aistopod groups of the amphibian-like lepospondyls, the
seymouriamorph and diadectomorph basal reptiliomorphs, the
eothyridid, ophiacodont, edaphosaur, and sphenacodont basal synapsids,
and the mesosaurs in the reptile clade (tables 6.8 and 6.9)—all in “Olson’s
extinction” that we considered in chapter 5. Then came the Emeishan LIP
eruption, and the surviving non-therapsid synapsids, the caseids and
varanopids, were eliminated, along with the biarmosuchian basal
therapsids (table 6.8) in the Capitanian. The catastrophe of the Siberian
LIP eruption killed the great dinocephalian herbivores and lithe
gorgonopsid predators in the synapsid clade, and in the reptile clade the
pareiasaurs, captorhinids, and gliding weigeltisaurids also perished (tables
6.8 and 6.9) in the Changhsingian. At the dawn of the Early Triassic, in the
synapsid clade, only the dicynodonts, therocephalians, and cynodonts had
survived (Sahney and Benton 2008). In the reptile clade, only the
procolophonid parareptiles, marine lepidosauromorphs (the
ichthyosaurs), and basal archosauromorphs lived on (Sahney and Benton
2008).
 The fate of the Paleozoic terrestrial world ora presents a somewhat
more complex picture than that of the fauna. As mentioned at the
beginning of this section, the terrestrial paleozoologists have not
produced a formal division of the terrestrial animal biosphere into
evolutionary faunas using the analytical techniques that Sepkoski used for
the marine biosphere. However, the terrestrial paleobotanists have:
Christopher Cleal and Borja Cascales-Miñana explicitly “adopted exactly
the same mathematical approach as used by Sepkoski (1981) and applied it
to the most recent compilation of families and classes of tracheophytes
(vascular plants) as revealed in the fossil record” (Cleal and Cascales-
Miñana 2014, 469). Their analyses reveal the existence of ve evolutionary
oras in the terrestrial realm during the Phanerozoic, rather than the
three evolutionary faunas that existed in the oceans.
Of particular interest here is that Cleal and Cascales-Miñana recognize
a Paleophytic ora in the interval of time from the Devonian through the
Permian, a ora dominated by lycophytes, licophytes, and
equisetophytes (lycopsids, pteropsids, and equisetopsids in their
terminology; see Cleal and Cascales-Miñana 2014, table 2). This oral
ecological assemblage collapsed following the end-Permian catastrophe in
a pattern remarkably similar to that shown by the Paleozoic marine fauna
(middle graph in gure 6.11) (Cleal and Cascales-Miñana 2014, g. 4,
factor-loading graph for factor 3). The Paleophytic ora was later replaced
in the Triassic by the Mesophytic ora, a ora dominated by pinophytes
and ginkgophytes (pinopsids and ginkgoopsids in their terminology; see
Cleal and Cascales-Miñana 2014, table 2), and Cascales-Miñana and
colleagues conclude that the end-Permian climatic catastrophe had a
“profound e ect on plant life; vegetation changed fundamentally, in e ect
resetting plant evolutionary history and marking the appearance of what
has become today’s vegetation” (Cascales-Miñana et al. 2015, 1072)—a fate
similar to that of the terrestrial faunas and the marine faunas of the Earth
in this interval of time. Thus the end-Permian mass extinction was a
completely global event, of maximum ecological severity in both the
oceans and on the land, and for both animal and plant life of the Earth
(McGhee et al. 2004; Cascales-Miñana et al. 2015).
We now enter the “post-apocalyptic greenhouse” (Retallack 1999)
world of the Early Triassic—and that world was very weird indeed. As
discussed previously, Yadong Sun and colleagues have presented evidence
that the equatorial tropical zones of the Earth were lethally hot for the
entire ve-million-year duration of the Early Triassic Epoch. The
surviving land vertebrates were con ned to latitudes higher than 30° in
the Northern Hemisphere and 40° in the Southern Hemisphere—the
tropics were barren. The same was true of the hot oceans, with marine
shes and marine reptiles, the ichthyosaurs, occurring only in the cooler
waters of the high latitudes of the Earth (Sun et al. 2012).
On the land, the tetrapod faunas of the entire Earth were dominated by
species of a single genus: the barrel-chested, burrowing dicynodont
therapsid Lystrosaurus. All of the continents of the Earth were joined
together in the single supercontinent of Pangaea, and in the Northern
Hemisphere you could have traveled from western North America to
Europe to eastern Asia without ever leaving dry land—and just about all of
the land animals you would have seen would have been lystrosaurs. Flying
over the barren lands of the super-hot tropics to the cooler high latitudes
of the Southern Hemisphere, you would have encountered scenes eerily
similar to those of the Northern Hemisphere—lumbering groups of
lystrosaurs wandering across the landscapes, from South America in the
west to Africa to India to Australia in the east.
Lystrosaurs were smaller herbivores ( g. 6.15) that also were
burrowers, and this behavioral trait probably contributed to their
surviving the end-Permian mass extinction, as discussed above (Retallack
et al. 2006; Retallack and Krull 2006). They had large, blocky heads with
massive jaw muscles for chewing (Benton 2015) the tough vegetation
present on land in the hothouse Early Triassic world—for land plants had
also su ered major extinction in the end-Permian catastrophe. The Late
Permian expanses of conifer trees were gone, replaced by low-growing
lycophyte club mosses, equisetophyte horsetail rushes, and ferns (Payne
and Clapham 2012; Sun et al. 2012; Irmis and Whiteside 2012). All of these
plants are of low nutrient content and are tough to chew—the
equisetophytes in particular, as these plants concentrate silica in their
tissues in order to discourage herbivores from eating them—so many of
the lystrosaurs, as burrowing animals, may have concentrated on digging
up the roots of the sparse Early Permian land-plant vegetation and eating
this more palatable plant material instead.

FIGURE 6.15 A synapsid disaster species that brie y ourished in the hothouse Early Triassic
world, the dicynodont herbivore Lystrosaurus.

Source: Illustration by Mary Persis Williams.

Other tetrapod groups were present on land in the Early Triassic, but in
very low diversity, as 90 percent of the land vertebrates alive in the Early
Triassic were lystrosaurs (Sahney and Benton 2008). The synapsid lithe,
catlike gorgonopsid predators of the Permian were gone, replaced by the
smaller, short-legged, and more crocodile-like archosaur Proterosuchus, a
reptile. The synapsids (our ancestors) had lost their ecological position as
the top carnivores to the archosaurs in the Early Triassic hothouse Earth,
and their fate would become even bleaker in the Late Triassic. The
diversifying archosaurs produced the freshwater phytosaur piscivores and
the land-dwelling rauisuchian carnivores (table 6.9), which further
competed with the few predatory cynodont synapsids, such as Galesaurus
and Thrinaxodon, that lived in the Early Triassic. But even these competing
archosaurian groups would become extinct by the end of the Triassic
following the evolution and spread of an entirely new type of
archosaurian predator—the theropod dinosaurs (tables 6.8 and 6.9). The
oldest known dinosaurs are Carnian in age—predators like Herrarasaurus
and Coelophysis—and they rapidly diversi ed in the Norian Age of the Late
Triassic (Benton 2015) until theropods became the top carnivores in
terrestrial ecosystems around the planet and the Mesozoic world
ecological structure was rmly established. For the next 150 million years,
the theropod dinosaurs would remain the lords of the predator megaguild,
and the cynodonts and their new descendants—the mammals—would
remain small and cryptic, hiding underground and foraging for food in the
dark of night.
At rst glance, it might seem that the synapsids had retained their
dominance as the major herbivores in terrestrial ecosystems, given the
ubiquitous planetwide distribution of the lystrosaurs in the Early Triassic.
However, the lystrosaurs were opportunistic “disaster species”—lucky
survivor species of an ecological catastrophe that rapidly spread into the
ecological space vacated by the unlucky victims of the end-Permian mass
extinction (Sun et al. 2012; Irmis and Whiteside 2012). The proliferation of
a few disaster species, both animal and plant, following a major
environmental catastrophe results in post-catastrophe ecosystems that
are very low in species diversity but, of the species that are present, high
population abundances and widespread geographic distribution. Such was
the Early Triassic world, in which 90 percent of the individuals in the
tetrapod populations of the world were lystrosaurs ( g. 6.15) (Sahney and
Benton 2008).
The synapsid disaster species of the Early Triassic world were rapidly
replaced by more diverse tetrapod faunas as the global terrestrial
ecosystem began to recover in the Middle Triassic (Sahney and Benton
2008; Irmis and Whiteside 2012). In particular, the reptiles progressively
moved into the terrestrial herbivore niche; by the Late Triassic, they
would displace the last of the herbivorous synapsids. First it was the
surviving procolonophorid parareptiles, themselves a remnant of the
Paleozoic world fauna (tables 6.8 and 6.9), that managed to achieve large
population densities in South Africa in the hothouse Early Triassic. Then,
in the Middle and Late Triassic, the more advanced rhynchosaur
archosauromorphs began to ecologically displace the remaining
dicynodont and cynodont synapsid herbivores. From small beginnings, by
the Carnian Age of the Late Triassic, the rhynchosaurs had evolved body
sizes up to two meters (6.6 feet) long and were the dominant herbivores in
terrestrial ecosystems, constituting up to 60 percent of the herbivore
populations (Benton 2015). These animals had massive heads—the back of
the skull was wider than the skull itself was long, and the front part of the
skull was laterally compressed and had a parrot-like beak ( g. 6.16). They
possessed massive muscles for chewing in the back part of the skull, and
their jaws moved from side to side with no sliding backwards or forwards
(Benton 2015). Even though they were well adapted to eating tough
vegetation, they still were driven to extinction—along with the last of the
herbivorous synapsids—at the end of the Carnian Age, and the herbivore
megaguild in terrestrial ecosystems was rmly taken over by the lords of
the Mesozoic world—the plant-eating dinosaurs (tables 6.8 and 6.9).

FIGURE 6.16 A Mesozoic world reptile that ourished and displaced the synapsid disaster species
in the Late Triassic, the rhynchosaur herbivore Hyperodapedon.

Source: Illustration by Mary Persis Williams.

The Paleozoic terrestrial world came to a nal end at the end of the
Carnian Age of the Late Triassic. It was replaced by the ecological
structure of the Mesozoic terrestrial world, with the land oras
dominated by the Mesophytic ora and the land faunas by the ruling
dinosaurs. The dinosaurian ecosystem would persist for some 150 million
years—until it too came to an end, when the Chixulub asteroid impacted
the Earth and the end-Cretaceous mass extinction began. Very late in the
Carnian Age37 of the Late Triassic, a small, furry, shrew-like animal rustled
in the underbrush, hunting for insects to eat. This tiny animal was a new
descendant of the cynodont synapsid carnivores that had been the
dominant predators on Earth in the Permian, during the time of the
Paleozoic terrestrial world. It was Adelobasileus cromptoni, the rst true
mammal (Benton 2005, 2015) and a basal member of the synapsid lineage
that would eventually include the evolution of Homo sapiens , modern
humans. Mammals would eventually come to rule the terrestrial realm—
but not for 150 million years. Back in the Late Triassic, the long night for
the mammals—the Age of Dinosaurs—had just begun.
7 The Legacy of the Late Paleozoic Ice
Age

Palaeozoic tropical rainforests and their e ect on global climates: is the past
the key to the present? … The Palaeozoic evidence clearly con rms that there
is a correlation between levels of atmospheric CO2 and global climates.
However, care must be taken in extrapolating this evidence to the present-day
tropical forests, which do not act as a comparable unsaturated carbon sink.
—Cleal and Thomas (2005, 13)

THE EFFECT OF THE LATE PALEOZOIC ICE AGE IN EARTH HISTORY

What were the e ects of the Late Paleozoic Ice Age in Earth history? First,
the climatic conditions created by the ice age—a planet with cold poles,
arid temperate zones, and an ever-wet, ever-warm tropical zone—were
perfect for forming what Stephen Greb and his colleagues, as noted in
chapter 3, called the “largest tropical peat mires in Earth history” (Greb et
al. 2003, 127), perfect habitats in which the peculiar lycophyte trees
ourished. Those peat mires and lycophyte tropical rainforests later
fossilized to produce the worldwide distribution of Carboniferous coal
deposits that would be used by humans over 250 million years later as the
principal source of energy in the process of industrialization, rst in
Europe and then around the world—continuing to the present day in
developing countries like China and India. In this chapter, we will examine
how the climatic and biological processes that created these massive coal
deposits during the Late Paleozoic Ice Age have a ected our modern
world.

Legacy 1: The Industrial Revolution?

Humans learned how to create re and burn wood in prehistory. In fact,

the use of re in the human lineage is much older than the 300,000-year-
old species Homo sapiens —we know from the fossil record that the older
hominin species Homo erectus also learned how to used re some 1.5
million years ago. We also know from historical records that our ancestors
knew about coal, the strange black rocks that they found lying on the
ground in some regions of the Earth—black rocks that could ignite and
burn. However, for the great majority of our history, our ancestors burned
wood for energy. Later humans learned how to make charcoal from wood
by heating wood in kilns, cooking out volatiles like water and partially
oxidizing the remaining carbon. Charcoal is a more concentrated form of
burnable carbon; that is, charcoal has a higher energy content per unit
mass than ordinary wood. Everyone who has ever used charcoal to grill
hamburgers or other food on a summer day is familiar with this fact,
recognizing that it would take a much larger mass of wood to accomplish
the same amount of cooking (the amount of wood required would not t
in your average backyard grill!).
In Europe, people rst began to burn coal to create hotter res in order
to melt metal ores from their rock matrix and to soften metal for working
it into tools and ornaments. People rapidly discovered that burning less
concentrated-carbon forms of coal, such as bituminous coal, created
irritating and unhealthy air pollution and thus began to use more
concentrated-carbon forms of coal, such as anthracite, for domestic
purposes like heating their homes.
Then the rst primitive steam engines were invented. It was discovered
that the pressure produced by boiling water into steam could be
controlled and used to power machines that could, for example, turn the
massive heavy millstones that were used to grind grain into our. These
steam engines were much more powerful than traditional sources of
energy for grinding grain, such as using the energy of owing water over
waterwheels or connecting the millstones to a team of horses that walked
around and around in circles. More experimentation produced more
e cient steam engines, until the Scot James Watt patented the rst
e cient, modern-type steam engine in 1769. Producing the steam power
in these engines required a very hot re—and coal was the perfect fuel for
that re. Even today, it is amazing to ride on a train powered by a now-
antique steam engine. The steam engine produces so much power that it
not only propels itself, but it also pulls along its source of energy—a
second car loaded with coal—as well as a whole series of cars connected to
the train behind the coal car. Those cars are heavy, and they can be lled
with even heavier loads of coal ore from coal mines in freight cars ( g. 7.1)
or loads of people in passenger cars. Yet the steam engine at the front of
the train pulls them all, powered by the energy released by burning the
coal in the coal car.
FIGURE 7.1 Coal-powered, steam-engine train hauling freight cars loaded with coal ore in
Australia.

Source: Photograph courtesy of Wikipedia/Christchurch. Reprinted under permission of the GNU

Free Documentation License.

Next, just as people discovered that charcoal could be made by cooking

o the volatiles in wood, people discovered that coke could be made by
cooking coal. Burning coke could then produce extremely hot res in very
large blast furnaces, and those huge blast furnaces could be used to
produce much less expensive, high-quality iron in massive amounts—
cheap iron that could be used to make more machines, particularly steam-
driven machines.
Thus was born the technology needed to drive the European Industrial
Revolution, beginning in Britain in the mid-1700s. The Industrial
Revolution probably began rst in Britain because, by geologic
coincidence, large amounts of coal-containing strata and metallic-ore-
containing rocks were both present. And those coal strata in England,
Wales, and Scotland are almost all Carboniferous in age. Then the
Industrial Revolution spread to continental Europe—particularly to
southern Belgium and the middle region of Germany, where coal-bearing
strata were also present. And those coal strata in the Ardennes of Belgium
and the Ruhrgebiet of Germany are almost all Carboniferous in age. The
Industrial Revolution spread around the world in the 1700s and 1800s,
following the geologic outcrops of Carboniferous coal. As Nick Lane
pointed out (see chapter 2), 90 percent of all the coal strata on Earth were
deposited during the height of the Late Paleozoic Ice Age—in the time
interval from the Serpukhovian Age in the Early Carboniferous to the
Wuchiapingian Age in the Late Permian (Lane 2002, 84). Coal was
originally present on the surface of the Earth, where early humans
discovered that these peculiar black rocks could burn. With the need for
coal to fuel the Industrial Revolution, intensive mining of coal spread
around the world. Coal strata are relatively shallow and could be mined by
open-pit excavations or shallow mine tunnels underground. As the
shallow deposits of coal were mined out, deeper underground mines were
dug, and entire mountain tops were removed in some regions of the Earth
to reach the coal buried at depth. The mining and burning of coal for
energy continues to the present day, but most of the coal burned today is
used to produce electricity, not to power steam engines.
The powerful steam engines of the 1700s, 1800s, and early 1900s have
been replaced today by an even more powerful, more e cient type of
engine—the internal combustion engine. The invention and near-
universal use of internal combustion engines today was made possible by
a new type of energy source—petroleum—and its re ned products such as
gasoline and jet fuel. Unlike coal, large quantities of petroleum are not
present on the surface of the Earth or at shallow depths underground. It
was only after people perfected deep-drilling machines—the early ones
powered by steam engines—that abundant petroleum became available as
an energy source.

Legacy 2: Modern Global Climate Change?

It is no accident that climate scientists divide the recent history of carbon-

dioxide levels in the Earth’s atmosphere into “preindustrial” and
“industrial” phases, in recognition of the atmospheric e ect of the tons of
coal that has been burned since the beginning of the Industrial
Revolution. The preindustrial amount of carbon dioxide in the
atmosphere was 0.03 percent—that is, in the entire recorded human
history of burning wood and charcoal, over 6,000 years, the carbon dioxide
content of the atmosphere did not exceed 0.03 percent. At the beginning
of the Industrial Revolution, humans began to burn coal in vast quantities,
and the present-day level of carbon dioxide in the atmosphere has risen to
0.04 percent in less than 200 years.
It is an empirical observation, a fact, that the Earth is heating up.
Combining that fact with the fact that the amount of the greenhouse gas
carbon dioxide in the atmosphere has vastly increased in a tiny amount of
time on geologic timescales has led climate scientists to conclude cause
and e ect—that is, that the industrially driven1 increase in the carbon-
dioxide content of the atmosphere has caused the Earth to retain more
heat in the atmosphere instead of losing it to space and, as a result, the
planet has become hotter. As 90 percent of the coal in the Earth’s strata
that has been burned since the beginning of the Industrial Revolution was
deposited during the Late Paleozoic Ice Age, by extension one could blame
the Late Paleozoic Ice Age for modern global climate change!
 However, as one might expect, the situation is more complicated than
that, and one cannot blame the Late Paleozoic peat bogs and tropical
mires for everything. The industrial burning of coal is clearly a major
culprit in the increase of carbon dioxide in our atmosphere, but since
about the mid-1900s, vast quantities of petroleum have been burned, and
the burning of petroleum has also added tonnes of carbon dioxide to the
atmosphere. It is usually impossible to determine how old that petroleum
was—was it formed from organics deposited during the Late Paleozoic Ice
Age or from some other geologic time period? Petroleum is a liquid, and a
slippery liquid at that, and it easily migrates within rocks under the
in uence of gravity and di erential pressure within the rock. Petroleum
can accumulate in vast pools deep underground that are very distant from
the original strata in which the petroleum formed.
Another complicating factor is that not only has the amount of carbon
dioxide in the atmosphere increased rapidly in less than 200 years, but the
number of human beings on the Earth has also increased rapidly in the
same period of time. In the early 1800s, early in the Industrial Revolution,
there were about one billion human beings present on the Earth. Two
hundred years later, at the beginning of the third millennium, the number
of human beings on the Earth has increased to 7.6 billion. All of those
people have depended primarily on energy from burning hydrocarbons—
rst coal, then petroleum—and that burning of hydrocarbons has added
carbon dioxide to the Earth’s atmosphere. If the Earth’s human population
had remained stable at around one billion individuals throughout those
200 years, clearly only a small fraction of the coal and petroleum that has
been burned during that time interval would have been burned—and the
Earth would be a much cooler (and less crowded) place.
Still, the fact remains that all of that Carboniferous coal (and
petroleum, whatever age it happens to be) was burned for energy—and
continues to be burned today. Where is the Earth headed in the future?
Are we headed for a hothouse world like the Earth that existed in the Early
Triassic—a world, described in detail in chapter 6, so hot that the tropical
zones were essentially lethal, a world in which complex plant and animal
life was con ned to latitudes higher than 30° in the Northern Hemisphere
and 40° in the Southern Hemisphere?

Legacy 3: The Successful Invasion of Land by Animals?

A twin result of the huge size and the peculiar growth pattern of the trees
in the ancient lycophyte rainforests was the massive removal of carbon
dioxide from the Earth’s atmosphere, the xation of the carbon from
carbon-dioxide molecules into plant tissues, and the burial of those
carbon-rich plant tissues in bogs where they would later be fossilized into
coal. Simultaneously, the oxygen released in xing carbon during plant
photosynthesis resulted in a massive injection of free oxygen into the
Earth’s atmosphere during the Late Paleozoic Ice Age. That abundance of
atmospheric oxygen led to the evolution of gigantism in both marine and
land animals and, on land, in both arthropod and vertebrate clades of
animals, as we saw in chapter 4.
In the vertebrate clade of animals, the initial boost of atmospheric
oxygen, it is argued, was a major assist in both the nal invasion of land
by the tetrapods in the Visean Age of the Carboniferous and in the
evolution of the rst amniotes (Graham et al. 1995, 1997; McGhee 2013). As
outlined in chapter 4, Je rey Graham and his colleagues have pointed out
that the hyperoxic Carboniferous atmosphere enabled the previously
aquatic tetrapods of the Devonian to obtain more oxygen with their
primitive lungs, to decrease the amount dehydration they experienced in
air breathing, and to boost their metabolic rates so they could move more
energetically and endure the constant pull of gravity experienced in
moving on dry land.
The key innovation in the successful invasion of land by vertebrates
was the evolution of the amniote egg and the anatomical antidehydration
adaptations that characterize the amniote animals. The hyperoxic
Carboniferous atmosphere would have allowed tetrapods to develop larger
eggs and, at the same time, to minimize water loss from those eggs
(Graham et al. 1997). With an atmosphere rich in oxygen, the egg could
absorb more oxygen per unit surface area of the egg. A larger egg has a
larger internal volume (hence more space for extra tissue layers and uid
in addition to the embryo) with a relatively smaller external surface area
(thus less water loss across that surface area to the outside world) than a
small egg. The amniote egg is an innovative, two-layer adaptation to
protect the developing embryo from dehydration in the harsh, dry-air
environments of the terrestrial realm. First, the embryo is enclosed in a
water- lled region contained by the surrounding amniotic membrane. In
essence, rather than oating in an actual pond of water as amphibian
embryos do, the amniote embryo oats within its own private amniotic
pond. Second, the outside layer of the egg is a tough shell rather than the
soft, gelatinous outer covering of the amphibian egg. Both of these layers
help protect the embryo from serious water loss and death by
dehydration. Thus the amniotes can lay their eggs on dry land and have
them survive. This is not true of most amphibians. Their eggs will rapidly
dehydrate if exposed to dry air, shriveling up and shrinking as moisture is
lost to the atmosphere, and the embryo will die.
Finally, a large container of food for the embryo is enclosed within the
amniote egg—the yolk sac. Thus the embryo can develop to a considerable
degree within the egg itself, feeding on nutrients from the yolk sac, rather
than hatching out of the egg at an early growth stage and foraging for
food in a free-swimming larval stage, as in many amphibians. In
consequence, the amniotes have direct development—there is no larval
stage. What emerges from an amniote egg looks like a small, scaled-down
version of the adult animal. In contrast, what emerges from a frog egg—a
tadpole—looks like a sh, not a frog. Fish, of course, need water to swim
in; a hatchling tetrapod amniote does not.
The amniotes were the victorious conquerors of land in the clade of the
vertebrate animals.2 That victory was achieved by the evolution of the
amniote egg, which freed the vertebrates from the constraint of having to
reproduce in or near a body of water such as a river or lake. Prior to this
innovation, tetrapods still laid their eggs in water, much like sh—and
many of the non-amniote tetrapods, the amphibians, still do so today. Free
from this constraint, the amniotes invaded the highlands and dry areas of
the Earth far from standing bodies of water.
Just as the amniotes were the nal conquerors of land in the vertebrate
clade, the winged insects were the victorious conquerors of land in the
clade of the arthropods (McGhee 2013). The hyperoxic atmosphere of the
Carboniferous, it is argued, was also a major assist in the evolution of the
winged insects, as we considered in chapter 4. Flying is a highly energetic
activity that requires a lot of oxygen. The arthropods invaded land back in
the Silurian, long before the vertebrates emerged from the water and the
start of the Late Paleozoic Ice Age (McGhee 2013). Why, then, did it take
them so long to take the next evolutionary step, to develop ight? The
development of a hyperoxic atmosphere in the Carboniferous and the
evolutionary diversi cation and spread of the ying insects during the
Late Carboniferous may not be a coincidence (Graham et al. 1995, 119, g.
2). With the evolution of ight, the insects could disperse over huge areas
of land in a very short period of time—the world was theirs for the taking.
Only much later, in the Late Triassic, would the vertebrates evolve
powered ight in the clade of the reptiles with the rst pterosaurs.

Legacy 4: The Evolution of Mammals and Dinosaurs?

The decline and depletion of free oxygen in the atmosphere following the
end of the Late Paleozoic Ice Age and the catastrophic Emeishan and
Siberian LIP eruptions, it is argued, was a major impetus for the evolution
of yet more e cient respiratory metabolisms—and eventually the
evolution of endothermic metabolisms—in both the synapsid and reptilian
clades of vertebrates, and the eventual evolution of both mammals and
dinosaurs (Graham et al. 1995, 1997). Geologic history—the very existence
of the Mesozoic and Cenozoic Eras of geologic time—would not have
developed the way it did if these two major groups of land animals had not
evolved. Every schoolchild knows that the Mesozoic was the Age of
Dinosaurs and the Cenozoic is the Age of Mammals. They may not be
aware, however, that both the dinosaurs and the mammals evolved in the
Late Triassic, and that the ecological conditions on land that resulted from
the end of the Late Paleozoic Ice Age may have contributed to the
evolution of both groups.
Some may argue that the evolution of dinosaurs and mammals is not
really a legacy of the Late Paleozoic Ice Age but rather a legacy of the
Emeishan and Siberian LIP eruptions, which produced the conditions that
resulted in oxygen depletion in the Earth’s atmosphere at the end of the
Permian. Still, the hyperoxic atmosphere created by the Late Paleozoic Ice
Age set the stage for the environmental selective pressures that would
result from the collapse of that atmosphere, and for the onset of the
hypoxic and poisonous atmospheric conditions in the latest Permian and
early Triassic. As we considered in detail in chapter 4, Graham and
colleagues have argued that the onset of hypoxia in the latest Permian
and earliest Triassic triggered the evolution of the four-chambered heart,
and perhaps full endothermy, in the therapsids within the synapsid clade
(Graham et al. 1997). In the reptilian clade, we know that four-chambered
hearts evolved in the Early Triassic, as the crocodilian archosaurs appear
in the fossil record at this time and they possess four-chambered hearts.
Still, as discussed in chapter 4, the oldest known (as yet) true mammals
and dinosaurs appear in the fossil record in the Late Triassic—at least 15
million years after the Late Permian–Early Triassic hypoxic interval of
time.

Legacy 5: The Destruction of the Paleozoic World?

The ecological shocks of the successive phases of glaciation during the

Late Paleozoic Ice Age triggered a series of extinctions that progressively
winnowed and depleted Paleozoic marine world animals in the Earth’s
oceans. The Frasnian, Famennian, and Serpukhovian extinctions all
preferentially eliminated marine animals that belonged to the Paleozoic
world fauna, and preferentially spared marine animals that belong to the
modern world fauna (Sepkoski 1984, 1990, 1996; Stanley 2007). We know
that the Famennian and Serpukhovian extinctions were triggered by Late
Paleozoic Ice Age glaciations, and, as discussed in chapter 1, evidence
exists that the Frasnian extinction was as well. In essence, the winnowing
e ect of the Late Paleozoic Ice Age glacially induced extinctions resulted
in a Paleozoic marine world fauna that was progressively weakened and
eventually could not withstand the catastrophic environmental changes
triggered by the Emeishan and Siberian LIP eruptions. As discussed in
chapter 6, the hot, acid, poisonous seawaters of the latest Permian oceans
were too much for the Paleozoic fauna, and the 200-million-year old
ecological structure of the Paleozoic marine world collapsed, never to
recover ( gs. 6.10, 6.11; table 6.6).
In contrast, the Paleozoic terrestrial world was largely the product of
the terrestrial climatic conditions created by Late Paleozoic Ice Age, as
discussed in chapter 6; how, then, can that world’s demise be attributed to
its creator? It is the end of the Late Paleozoic Ice Age that was the
harbinger of the end of the Paleozoic terrestrial world, and thus one may
argue that both the existence of and the end of the Paleozoic terrestrial
world were a legacy of the Late Paleozoic Ice Age. This causal relationship
can be seen most clearly in Olson’s extinction, which we considered in
detail in chapter 5. The waning of the Late Paleozoic Ice Age in the Early
Permian triggered major contractions in the ever-warm, ever-wet
equatorial zone of the Earth, and major elements of the terrestrial
vertebrate fauna of the Paleozoic terrestrial world died out as a result. One
could argue that the waning of the Late Paleozoic Ice Age produced its
own winnowing e ect in preferentially eliminating species of the
Paleozoic terrestrial world fauna, resulting in a weakened fauna that could
not withstand the catastrophic environmental changes triggered by the
Emeishan and Siberian LIP eruptions in the Late Permian.
On the other hand, it can be counterargued that the destruction of the
Paleozoic World is not really a legacy of the Late Paleozoic Ice Age but
rather a legacy of the Emeishan and Siberian LIP eruptions; those
eruptions produced the catastrophic environmental conditions that
lethally heated and poisoned vast areas of both the seas and the land of
the Earth, and the very atmosphere of the entire planet. The
counterargument would be that had it not been for the Emeishan and
Siberian LIP eruptions, the characteristic faunas of the Paleozoic world
might have survived. This hypothetical possibility is but one of many that
we will explore in the next sections of the chapter.
What If the Late Paleozoic Ice Age Had Never Happened?

From the perspective of human history, this question has major

implications. If there had been no Late Paleozoic Ice Age, then there
would have been no Earth with a tropical zone with huge expanses of
lycophyte rainforests from 326 to 254 million years ago (tables 3.2 and
5.1). If there had been no lycophyte rainforests, then there would have
been no massive coal deposits in Carboniferous strata—indeed, there
would have been no “Carboniferous” at all in the geologic timescale. If the
Carboniferous coal strata had never existed, would there ever have been
an industrial revolution?
From prehistoric times we have burned wood for heat—heat to keep
our habitats warm and heat to cook our food. Wood is readily available at
the Earth’s surface, and it is also a renewable resource, as new trees can be
grown to replace old ones that have been cut and used for fuel. However,
neither wood nor charcoal made from wood contains enough energy per
unit mass to e ciently fuel a steam engine—for that we need coal. In
human history, coal also was initially readily available at the surface of the
Earth, or only shallowly buried beneath that surface. Yet 90 percent of the
coal that was present in the Earth’s strata was formed during the Late
Paleozoic Ice Age, and if the Late Paleozoic Ice Age had never occurred,
then that coal would not have existed. How long would a nascent
industrial revolution have lasted that was fueled by only 10 percent of the
coal that was used to fuel the historical Industrial Revolution? In an
alternative world in which the Late Paleozoic Ice Age had never occurred,
could humankind have quickly made the jump—while those meager 10
percent of coal supplies still lasted—from steam engines to the use of
petroleum and the invention of the internal combustion engine?
The problem with petroleum is that it is usually not readily available at
the Earth’s surface—or even shallowly below that surface. It takes
powerful drilling machines to reach the petroleum pools located deep
beneath the surface of the Earth. In an alternative world in which the Late
Paleozoic Ice Age had never occurred, would humankind have used its
coal- red steam engines to drill for oil before the coal ran out? Or would
that alternative humankind have remained in the preindustrial stage of
our own world, forever existing in small cities and predominantly
agrarian communities that used wood for fuel, unaware of the existence of
the vast pools of oil buried deep beneath the surface of the Earth?
Clearly our modern dilemma of carbon-dioxide-induced global
warming would not have occurred if the Late Paleozoic Ice Age had not
occurred, as the vast deposits of coal that formed in that ice age would
never have existed to be burned by humans. If the Industrial Revolution
had never occurred, the human population itself might not have exploded
as it did; in an alternative world in which the Late Paleozoic Ice Age had
never occurred the human population might have grown at a much slower
rate, fueled solely by our intellectual advances in more e cient
agricultural methods to produce more food and our advances in more
e cient medical methods to preserve and prolong life.
The question “What if the Late Paleozoic Ice Age had never happened?”
also has major implications for Earth history and evolutionary ecology. If
the great lycophyte tropical rainforests had never existed, the hyperoxic
atmosphere of the Earth during much of the Carboniferous–Permian
interval of geologic time also would never have existed, as those peculiar
rainforests not only acted as massive carbon sinks but also released tons
of free molecular oxygen into the atmosphere. In the absence of a
hyperoxic atmosphere, the convergent evolution of the numerous giant
animals that we considered in detail in chapter 4 would never have
happened. Giant gri en ies would never have soared through the skies of
the Earth, alligator-size millipedes would never have crawled through the
peat bogs and rainforests below, and so on.
However, there is a more serious side to this question than the
potential absence of animal gigantism in the Carboniferous and Permian.
Without the energetic boost of that oxygen-enriched atmosphere, would
the tetrapods have nally managed to emerge from the rivers and lakes
and become fully terrestrial in the Early Carboniferous? Without that
hyperoxic atmosphere, would the amniotes have evolved? Without extra
oxygen, would the tracheal-breathing insects have nally managed to
perfect the highly energetic activity of ight? Or, in an alternative world
in which the Late Paleozoic Ice Age had never occurred, would the
tetrapods have remained in the aquatic evolutionary stage, and if the
amniotes had never evolved, would the land areas of the Earth have been
populated by amphibian vertebrates only? Within the terrestrial
arthropods, would the insects have remained grounded in the mode of life
in which arthropods had previously existed for some 100 million years
since their initial invasion of land back in the Silurian?
As discussed in the rst section of the chapter, we also know that the
Famennian and Serpukhovian extinctions were triggered by the Late
Paleozoic Ice Age glaciations, and evidence exists that the Frasnian
extinction was as well. If the Late Paleozoic Ice Age had never happened,
then the glacially induced Famennian and Serpukhovian—and possibly the
Frasnian—extinctions would never have happened. These events were the
seventh, sixth, and fourth most ecologically severe biodiversity crises in
Earth history (table 1.3). How would our Earth have been di erent if these
crises had never happened?
In the oceans, the Frasnian crisis destroyed the largest reefs in Earth
history—the massive Paleozoic-style skeletal reefs of stromatoporoid
sponges, tabuate corals, and rugose corals ( g. 7.2). The stromatoporoids
just barely survived the Frasnian extinction, only to be exterminated by
the Famennian crisis. The tabulate corals never recovered their previous
diversity, but the rugosans did begin to recover and diversify in the Early
Carboniferous, only to be decimated by the Serpukhovian crisis (McGhee
et al. 2012). One hundred thirty million years were to pass before the
corals once again became the major skeletal-building element in reefs—in
the Middle Triassic, with the evolution of the scleractinian corals (Flügel
and Stanley 1984; Fois and Gaetani 1984). If the Late Paleozoic Ice Age had
never happened, would Paleozoic-style massive skeletal reefs have
survived to the present? Would the scleractinian corals and our modern-
style reefs have never evolved?

FIGURE 7.2 The skeletal-building elements of the massive Paleozoic-style reefs were chie y
stromatoporoid sponges like Stromatopora polyostiolata (top left), tabulate corals like Halysites
catenularia (bottom left), and rugosan horn-corals like Caninia torquia (right).
Source: Stromatopora, Halysites, and Caninia modi ed and redrawn from Tasch (1973); Hoskins,
Inners, and Harper (1983); and Moore, Lalicker, and Fischer (1952), respectively.

The Devonian is also known as the “Age of Armored Fishes,” the great
placoderms (table 7.1) (Young 2010). As discussed in chapter 2, some of the
great armored shes were as big as modern-day killer whales, but unlike
killer whales, the armored shes had no teeth. Instead, they had sharp
bone blades in their mouths that functioned in a way similar to the sharp
bone beaks of modern-day snapping turtles. And even though they
possessed massive armor plates of bone around their heads, internally
these peculiar shes had skeletons made of cartilage, rather than bone,
like modern-day sharks. Imagine a huge shark with a head like a snapping
turtle and you have a sh similar to the great placoderm predators (see
g. 2.1).

TABLE 7.1 Detailed phylogeny of placoderm and sarcopterygian vertebrates showing the e ect of
the twin Late Devonian extinctions (modi ed from tables 6.5 and 6.8).

VERTEBRATA (animals with vertebrae)

– Gnathostomata (jawed vertebrates)
– – PLACODERMI (armored shes) †Famennian
– – – Acanthothoraci †Frasnian
– – – unnamed clade †Famennian
– – – – unnamed clade †Famennian
– – – – – Rhenanida †Frasnian
– – – – – Antiarchi †Famennian
– – – – unnamed clade †Famennian
– – – – – Arthrodira †Famennian
– – – – – unnamed clade †Famennian
– – – – – – Petalichthyida †Frasnian
– – – – – – Ptychtodontida †Famennian
– – CHONDRICHTHYES (sharks, rays, and kin)
– – Acanthodii (spine n shes)
– – Osteichthyes (bony shes)
– – – ACTINOPTERYGII (ray n shes)
– – – SARCOPTERYGII (lobe n shes + descendants)
– – – – CROSSOPTERYGII
– – – – – Porolepiformes †Famennian
– – – – – unnamed clade
– – – – – – Onychodontida †Famennian
– – – – – – Actinista (living coelacanths)
– – – – DIPNOI (lung shes)
– – – – TETRAPODOMORPHA (tetrapod-like shes + descendants)
– – – – – Rhizodontia
– – – – – Osteolepidiformes
– – – – – – Osteolepididae †Famennian
– – – – – – Megalichthyidae
– – – – – – Eotetrapodiformes
– – – – – – – Tristichopteridae †Famennian
– – – – – – – unnamed clade
– – – – – – – – Elpistostegalia †Frasnian
– – – – – – – – TETRAPODA (limbed vertebrates)
– – – – – – – – – Family Elginerpetontidae †Frasnian
– – – – – – – – – – Elginerpeton pancheni †Frasnian
– – – – – – – – – – Obruchevichthys gracilis †Frasnian
– – – – – – – – – Family incertae sedis †Frasnian
– – – – – – – – – – Sinostega pani †Frasnian
– – – – – – – – – unnamed clade
– – – – – – – – – – Family incertae sedis †Famennian
– – – – – – – – – – – Densignathus rowei †Famennian
– – – – – – – – – – unnamed clade
– – – – – – – – – – – Family incertae sedis †Famennian
– – – – – – – – – – – – Ventastega curonica †Famennian
– – – – – – – – – – – unnamed clade †Frasnian
– – – – – – – – – – – – Family incertae sedis †Frasnian
– – – – – – – – – – – – – Metaxygnathus denticulus †Frasnian
– – – – – – – – – – – Family incertae sedis †Famennian
– – – – – – – – – – – – Jakubsonia livnensis †Famennian
– – – – – – – – – – – unnamed clade
– – – – – – – – – – – – Family Acanthostegidae †Famennian
– – – – – – – – – – – – – Acanthostega gunnari †Famennian
– – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – Family incertae sedis †Famennian
– – – – – – – – – – – – – – Ymeria denticulata †Famennian
– – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – Family Ichthyostegidae †Famennian
– – – – – – – – – – – – – – – Ichthyostega stensioei †Famennian
– – – – – – – – – – – – – – – Ichthyostega watsoni †Famennian
– – – – – – – – – – – – – – – Ichthyostega eigili †Famennian
– – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – Family incertae sedis †Famennian
– – – – – – – – – – – – – – – – Hynerpeton bassetti †Famennian
– – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – Family Tulerpetontidae †Famennian
– – – – – – – – – – – – – – – – – Tulerpeton curtum †Famennian
– – – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – – Family Colosteidae
– – – – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – – – Family Crassigyrinidae
– – – – – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – – – – Family Whatcheeriidae
– – – – – – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – – – – – Family Baphetidae
– – – – – – – – – – – – – – – – – – – – unnamed clade
– – – – – – – – – – – – – – – – – – – – Neotetrapoda
– – – – – – – – – – – – – – – – – – – – – BATRACHOMORPHA (ancestors of amphibians)
– – – – – – – – – – – – – – – – – – – – – Lepospondyli
– – – – – – – – – – – – – – – – – – – – – REPTILIOMORPHA (ancestors of amniote tetrapods)

Source: Phylogenetic data modi ed from Lecointre and Le Guyader (2006), Ahlberg et al. (2008),
Clack et al. (2012), and Benton (2015).
Note: Lineages that went extinct in either the Frasnian or Famennian are marked with a dagger (†);
the age of their extinction is in bold; major clades are in capitals.

The Frasnian extinctions eliminated half of the world’s diversity of

armored shes (table 7.1); the Famennian extinctions eliminated them all
(table 7.1). Our modern ecologically and numerically dominant shes, the
sharks (the chondrichthyans, table 7.1) and ray- n shes (the
actinopterygians, table 7.1), diversi ed only in the Early Carboniferous,
lling the ecological void left by the vanished armored shes (Sallan and
Coates 2010; McGhee 2013). If the Late Paleozoic Ice Age had never
happened, would our rivers, lakes, and seas still be populated by giant
dinichthyids and titanichthyids—the “terrible shes” and “titanic
shes”—that were seven meters (23 feet) long, with massive bony head
armor and jaws like a snapping turtle? Would sharks and ray- n shes
never have diversi ed, but only continued to exist in the shadow of the
great armored shes if the Late Paleozoic Ice Age had never happened?
The Late Devonian extinctions were particularly severe in the clade of
the sarcopterygian vertebrates (table 7.1), where they acted as
evolutionary bottlenecks. As discussed in chapter 2, an evolutionary or
genetic bottleneck is produced when the population sizes of a given
species shrink almost to the critical minimum level from which a species
cannot recover. One of the consequences of a species’ surviving an
evolutionary bottleneck is the sharp reduction of morphological or
genetic diversity in the population of survivors. Another is a reduction in
the geographic range of a species: species that may once have had large
populations spread across a continent may be found in only a few regional
patches with small numbers of individuals after an evolutionary
bottleneck. Finally, as implied in the descriptive name “bottleneck,” only a
few lineages manage to survive past that evolutionary constriction. All of
these classic bottleneck phenomena are seen in the lineage of the
sarcopterygians in the twin Late Devonian extinctions; hence I call these
two evolutionary events the End-Frasnian Bottleneck and the End-
Famennian Bottleneck (McGhee 2013).
The sarcopterygian lobe- n shes are not familiar to us today, but they
greatly outnumbered our modern, familiar ray- n shes (the
actinopterygians, table 7.1) back in the Devonian. The sarcopterygian
clade was hit hard by the Late Devonian extinctions, and today they are
represented by only one genus of coelacanth sh, three groups of lung
sh, and, of course, ourselves—the modern living tetrapods.
The End-Famennian Bottleneck eliminated two-thirds of the lineages of
the crossopterygian lobe- n shes (table 7.1). For many years,
paleontologists thought that the crossopterygians were extinct—that is,
that they had encountered an evolutionary dead end rather than a
bottleneck. Then, in 1938, o the coast of South Africa, shermen
unexpectedly captured a very odd-looking sh. Unlike a ray- n sh,
whose ns appears to attach directly to its body, this sh had peculiar
stumpy lobes—lobes that contained bones—protruding from its body, and
its ns were attached to these lobes. To their astonishment,
paleontologists recognized the sh as a living coelacanth crossopterygian
and gave it the name Latimeria chalumnae. Since that time, other
individuals of Latimeria chalumnae have been sighted and photographed in
the waters o South Africa and Madagascar, have appeared in National
Geographic magazine and television programs, and so on. The
crossopterygians had survived the End-Famennian Bottleneck after all,
and are still alive today.
Only three families of lung shes, the dipnoian sarcopterygians (table
7.1), survive today—the Neoceratodontidae in Australia, the
Lepidosirenidae in South America, and the Protopteridae in Africa. These
widely scattered lung sh groups are rare, the only survivors of a group of
shes that were very numerous in the Devonian. What if lobe- n
crossopterygian and dipnoian shes were common today, as they were in
the Devonian, and ray- n shes were rare? The idea that we evolved from
sh might not seem so strange to people if every time they bought a sh
at the supermarket, it had four stumpy limblike ns, complete with
internal bones, as in a crossopterygian sh. What if it were not at all
unusual to go down to a river and see several di erent types of shes
crawling about on the banks, breathing air? If, like common pet turtles in
aquaria, modern-day children kept air-breathing sh as pets at home,
would their parents still be skeptical than land vertebrates are the
descendants of sh?
Next we encounter the tetrapodomorph shes—the ancestors of the
living tetrapods (table 7.1). No fewer than three groups of lobe- n,
tetrapodomorph shes were independently evolving tetrapod-like
morphologies in the Devonian—the tristichopterids like Eusthenopteron
( g. 7.3), the elpistostegalians like Tiktaalik ( g. 7.3), and, of the course,
the tetrapods themselves (table 7.1) (Ahlberg and Johanson 1998; Coates et
al. 2008; McGhee 2013). The End-Frasnian Bottleneck eliminated the
elpistostegalians, and the End-Famennian Bottleneck eliminated the
tristichopterids (table 7.1). The extinction of the elpistostegalian,
tristichopterid, and osteolepidid tetrapodomorphs created a large
phylogenetic gap between the advanced tetrapods and the basal rhizodont
and megalichthyid tetrapodomorphs (table 7.1). The rhizodonts would not
survive the Carboniferous, and the megalichthyids would not survive the
Permian. By the end of the Paleozoic, only the tetrapod clade remained of
the once diverse tetrapodomoph lineages. As a result of the Late Devonian
extinctions and bottlenecks, a large phylogenetic gap exists today
between living limbed vertebrates (the tetrapods; table 7.1) and living
nned vertebrates (the ray- n shes, the actinopterygians; table 7.1). All
of our intermediate cousins are gone, which is why a modern-day lizard
seems to us to be such a radically di erent animal from a trout.

FIGURE 7.3 From lobed ns to limbs (from top to bottom): comparison of the bones in the n of
the tristichopterid sh Eustenopteron foordi, in the n of the elpistostegalian tetrapodomorph
Tiktaalik roseae, in the forelimb of the aquatic tetrapod Acanthostega gunnari, in the forelimb of the
cynodont synapsid Thrinaxodon liorhinus, in the forelimb of the mammalian synapsid mountain lion
Felis concolor, and in the arm of the human Homo sapiens.

Source: Illustration by Kalliopi Monoyios © 2013. Reprinted with permission.

If these two bottlenecks in vertebrate evolution had not occurred,

might further parallel evolution in both the elpistostegalian and
tristichopterid lineages have produced animals with four limbs? That is,
could the condition of possessing four limbs have evolved independently
in three separate vertebrate lineages, instead of just one? This is not an
outlandish idea: it is a fact that three other groups of vertebrates have
convergently, independently, evolved wings from their forelimbs—the
pterosaurs, the birds, and the bats. The pterosaurs are ornithodiran
reptiles, and they evolved wings from their forelimbs in the Triassic. The
birds are also reptiles, but they belong to a much more derived lineage of
reptiles—the dinosaurs. They independently evolved wings from their
forelimbs in the Jurassic. The bats are very di erent from the pterosaurs
and birds as they are synapsids, not reptiles. They are also are very highly
derived synapsids—they are mammals—and yet they also independently
evolved wings from their forelimbs in the Paleocene, at the beginning of
the Cenozoic (Benton 2005). If three separate groups of vertebrates could
independently, convergently, accomplish the seemingly highly unlikely
feat of evolving wings from modi ed forelimbs, it is equally likely that
three separate groups of advanced lobe- n shes could have
independently, convergently, evolved four limbs from their four ventral
lobe ns. Today the skies of the Earth are the territory of birds and bats—
two very di erent kinds of winged animals that have no common winged
ancestor. What if the land areas of the Earth were also home to two, or
even three, di erent kinds of limbed vertebrates that also had no common
limbed ancestor but had independently evolved their four limbs? We will
never know, as the Late Devonian bottlenecks eliminated that
evolutionary possibility forever.3
The rst of the tetrapods—the true limbed vertebrates (table 7.1)—
were aquatic animals, not land dwellers. Their limbs were short, and the
manus and pes (hand and foot) morphologies on the limb, used as
swimming and crawling paddles, were very di erent from those of
modern tetrapods ( g. 7.3). Only with subsequent evolution would the
paddlelike limbs of the early aquatic tetrapods be modi ed to limbs that
would bear the weight of the animal as it walked on dry land. The
Devonian aquatic tetrapods like Acanthostega ( g. 7.3) were the classic
“ sh with feet” that are featured today on cars with Darwin bumper
stickers.
The End-Frasnian Bottleneck hit the tetrapods hard (table 7.1). Before
the bottleneck, these early aquatic tetrapods had a worldwide distribution
—from Scotland (Elginerpeton pancheni) and Latvia (Obruchevichthys gracilis)
in Europe in the west to China (Sinostega pani) and Australia
(Metaxygnathus denticulus) in the east. After the bottleneck, surviving early
Famennian tetrapods are found only in Europe. Before the bottleneck,
tetrapods had a diversity of body sizes, from large individuals up to 2.5
meters (8.2 feet) long to small individuals about a 0.5 meters (1.6 feet)
long. After the bottleneck, surviving Famennian tetrapods are about all
the same size—about a meter (3.3 feet) long. The End-Frasnian Bottleneck
triggered a sharp reduction in morphological variance and geographic
range in surviving tetrapod faunas, reductions that are classic
evolutionary bottleneck phenomena. What diverse Famennian
morphological novelties and body sizes might our cousins in the other
Frasnian tetrapod lineages have produced if they had not been eliminated
by the End-Frasnian Bottleneck? How would the future course of tetrapod
evolution have changed if these lineages had survived?4
Following the Famennian Gap cold period (table 1.7), the tetrapods
began to recover their species diversity and began to lose many of their
shlike traits as they became more adapted to life on dry land. Then they
encountered the End-Famennian Bottleneck, and it triggered an even
more severe diversity contraction than had the End-Frasnian Bottleneck
(table 1.7). Not only were almost all of the new Famennian clades
eliminated, two lineages of Frasnian survivors (the lineages of
Densignathus rowei and Ventastega curonica) were also driven to extinction.
Only three groups are thought to have survived the bottleneck—the
colosteids, the crassigyrinids, and the whatcheeriids (table 1.7). Only the
whatcheeriids are known from fossils in the Famennian (Daeschler et al.
2009; McGhee 2013), and the presumed survival of the colosteids and
crassigyrinids is based solely on phylogenetic analyses that indicate that
they are more primitive than the whatcheeriids and thus evolved before
the whatcheeriids.5
As discussed in chapter 2, following the Famennian glaciations and the
Tournaisian Gap cold period (table 1.7), the tetrapods began once again to
recover. The whatcheeriids (see g. 2.2) in particular blossomed in
diversity and achieved a worldwide distribution, with new species being
found in North America, Europe, and Australia. By the Visean Age, the
tetrapod invasion of land was in full progress, and the two major clades of
the batrachomorphs (ancestors of our modern amphibians) and the
reptiliomorphs (ancestors of the amniotes—the living reptiles, dinosaurs,
and mammals) had evolved (tables 2.4 and 7.1).
At the very least, the End-Frasnian Bottleneck set back and delayed the
successful invasion of land by tetrapods by some six million years (the
duration of the Famennian Gap, table 1.7), and the End-Famennian
Bottleneck likewise generated another delay of some ten million years
(the duration of the Tournaisian Gap). At the worst, the bottlenecks
changed the direction of tetrapod evolution forever—we will never know
how the exterminated lineages would have evolved or what ecological
diversi cations and evolutionary innovations they would have produced.
In our impoverished world, only one lineage of four-limbed vertebrates
exists—a sad legacy of the Late Paleozoic Ice Age.
 On the other hand, as discussed at the beginning of this section, the
hyperoxic atmosphere produced by the Late Paleozoic Ice Age was a major
boost to that surviving tetrapod lineage in its nal and successful invasion
of land. The other major clade of terrestrial invaders, the arthropods,
likewise received that energetic boost—a boost that sent them into the
skies. In a world where the hyperoxic atmosphere of the Late Paleozoic Ice
Age had never existed, perhaps both of those evolutionary events might
still have happened—but they might also have taken much longer than the
rapid burst of evolutionary events that occurred in the Serpukhovian–
Bashkirian interval of Carboniferous time.

WHAT IF THE SIBERIAN MANTLE PLUME HAD MISSED THE

TUNGUSKA BASIN?
The continental glaciations of the Late Paleozoic Ice Age eventually would
have ended even if the global warming produced by the Emeishan LIP
eruption and then the extreme global heating produced by the Siberian
LIP eruption had never happened. After its long 115-million-year trek, the
supercontinental mass of Gondwana was nally moving o the South Pole
in the latest Permian, as discussed in chapter 1 ( g. 1.3). By the Middle
Triassic, about 240 million years ago, the South Pole was open ocean.6 The
giant landmass of Gondwana began to break up in the Middle Jurassic: the
plate fragment holding South America and Africa began to move north,
and the plate fragment holding India, Antarctica, and Australia began to
move south, back toward the South Pole. By the Early Cretaceous, the
continental region of Antarctica was once again located over the South
Pole, and then rst India and later Australia began to split away from it.
By the Eocene Epoch of the Cenozoic Era, Antarctica was a totally separate
continent—a small fraction of the original giant continent Gondwana—
sitting directly on the South Pole, and the initial phases of the continental
glaciation that would produce the Cenozoic Ice Age had begun.7
In itself, eruption of a mantle plume the size of the Siberian one would
have a ected the climate of the entire planet, but not to the extent that it
did when combined with the vaporization and explosive venting of gases
from the huge volume of Tunguska Basin organic-rich and evaporite
strata that the Siberian mantle plume burned through on its way to the
surface of the Earth. The Siberian mantle plume alone erupted enough
lava to cover ve million square kilometers (1.93 million square miles) of
land; as discussed in chapter 6, that is enough lava to cover 62 percent of
the area of the United States between the Atlantic and Paci c coasts. The
Siberian ood-basalt eruption is estimated to have been 8.3 to 20 times
larger than the Emeishan LIP, producing some 45 to 108 trillion tonnes (50
to 119 trillion tons) of sulfur-dioxide pollutant and its byproduct, sulfuric
acid. The acidi cation of the world’s oceans and acid rain poisoning of the
world’s land areas would have proceed just as they did at the end of the
Permian in the absence of any further gaseous input from the Tunguska
Basin strata—but it can be argued that the environmental impacts would
have been measured on a smaller scale. This is because the massive
additional injection of sulfur dioxide into the atmosphere from the
Tunguska Basin anhydrite deposits—strata rich in calcium sulfate—would
not have occurred if the Siberian mantle plume had missed the Tunguska
Basin.
Beyond the e ects of sulfur dioxide, the environmental consequences
of the Siberian LIP eruption would have been radically di erent if the hot
plume had missed the Tunguska Basin. At the end of the Permian,
between 100 and 160 trillion tonnes (110 to 176 trillion tons) of carbon
dioxide and between 14 and 42 trillion tonnes (16 to 46 trillion tons) of
methane were injected into the Earth’s atmosphere from the organic-rich
strata in the Tunguska Basin (Svensen et al. 2009; Payne et al. 2010), as
discussed in chapter 6. What if this had never happened? If the Earth’s
atmosphere had never become vastly enriched and polluted with those
two greenhouse gases, the horri c Hot Earth of the latest Permian and
Early Triassic would never have occurred. The global depletion of oxygen
from the Earth’s atmosphere that occurred at the end of the Permian—
caused by the oxidation of those trillions of tonnes of methane—would
never have happened. The additional acidi cation of the world’s oceans
by carbonic acid would never have occurred; ocean acidi cation would
have been the product of the mantle plume’s own sulfur dioxide without
any extra boost from the Tunguska Basin carbon deposits. Coal y ash
from the burning Tunguska Basin coal strata would never have been
explosively injected into the atmosphere, oceanic euxinia would not have
occurred, and unknown trillions of marine organisms would never have
su ocated from the oxygen-depleting e ects of eutrophication. Thus
three of the end-Permian kill mechanisms that we considered in chapter 6
—heat death, carbon-dioxide poisoning, and su ocation—would either not
have occurred at all or would have been vastly reduced in their lethality.
Finally, we have the evaporite strata of the Tunguska Basin. At the end
of the Permian, between ve and 15 trillion tonnes (six to 17 trillion tons)
of methyl chloride and between 87 and 255 billion tonnes (96 to 281 billion
tons) of methyl bromide were injected into the Earth’s atmosphere from
the burning of the Tunguska Basin evaporite deposits (Svensen et al.
2009). If the Siberian mantle plume had missed the Tunguska Basin, those
gases never would have been injected into the atmosphere, the ozone
layer of the atmosphere never would have been destroyed, and the vast
increase in the radiation ux of ultraviolet at the Earth’s surface never
would have occurred. Yet another of the end-Permian kill mechanisms—
radiation poisoning—would either not have occurred at all or would have
been vastly reduced in its lethality.
In chapter 6 we also considered the alternative mantle-plume-
chemistry model of Sobolev and colleagues (Sobolev et al. 2011), who
argued that the Siberian LIP itself could have produced as much as 175
trillion tonnes (193 trillion tons) of carbon dioxide and as much as 18
trillion tonnes (20 trillion tons) of hydrochloric acid if the melting and
recycling of oceanic crustal rock located beneath Siberia had produced a
magma that was more gaseous and carbon rich than typical basaltic-type
lavas of other LIP eruptions like those of Laki, Iceland. Thus, in the case of
each kill mechanism discussed above, I have included the possibility that
even if the Siberian mantle plume had missed the Tunguska Basin, the kill
mechanisms of heat death, carbon-dioxide poisoning, su ocation, and
radiation poisoning still might have occurred but would have been vastly
reduced in their lethality. Even if the magma-chemistry model of Sobolov
and colleagues is eventually shown to be correct, it is a known fact that
the Siberian mantle plume did indeed burn through the organic-rich and
evaporitic strata of the Tunguska Basin. Thus, whatever the lethality of
the kill mechanisms produced by the Sobolev and colleagues model for
the Siberian mantle plume alone, all of the intensi cation of those kill
mechanisms produced by the vast tonnages of carbon dioxide, methane,
methyl chloride, and methyl bromide that were injected into the Earth’s
atmosphere at the end of the Permian would have to be subtracted if the
mantle plume had missed the Tunguska Basin.
Would the end-Permian mass extinction have happened at all if the kill
mechanisms of heat death, carbon-dioxide poisoning, su ocation, and
radiation poisoning had never occurred, or occurred with much reduced
intensities? If the Siberian mantle plume had missed the Tunguska Basin,
would the only major kill mechanism triggered have been acidi cation
resulting from the plume’s venting of sulfur dioxide? The Siberian
volcanic eruptions were the largest continental LIP in Earth history
(Reichow et al. 2009), but would sulfur-dioxide-induced acidi cation of
the world’s oceans and land areas have been lethal enough to cause the
largest loss of biodiversity and the severest ecological catastrophe in Earth
history, the end-Permian mass extinction (table 1.3)?
If an extinction event of much reduced severity had occurred at the
end of the Permian—in essence, if there had been no end-Permian mass
extinction—would the Paleozoic world have never come to an end (table
6.6)? Jack Sepkoski would have argued that the Paleozoic marine world
would still have ended, even in the absence of the end-Permian mass
extinction—it just would have taken longer and not been as abrupt in an
alternative world in which the Siberian mantle plume missed the
Tunguska Basin. Sepkoski’s analyses of the interactive evolutionary
dynamics (Sepkoski 1981, 1984, 1990; Sepkoski and Miller 1985) of his
Paleozoic evolutionary fauna and modern evolutionary fauna supported
the conclusion that the modern evolutionary fauna would have actively,
competitively, ecologically replaced the Paleozoic evolutionary fauna with
time, just as his Paleozoic evolutionary fauna competitively replaced the
Cambrian evolutionary fauna ( gs. 6.10, 6.11). Sepkoski’s analyses showed
that every major crisis in the Paleozoic had di erentially eliminated
Paleozoic marine world fauna and favored the modern evolutionary fauna
—in essence, that the Paleozoic marine world was doomed even in the
absence of the catastrophic e ects of the end-Permian mass extinction.
The winnowing e ect of the Frasnian, Famennian, Serpukhovian, and
Captanian extinctions on the Paleozoic marine world fauna would still
have occurred in a world where there had been no end-Permian mass
extinction, and Sepkoski would have argued that each future biodiversity
crisis in the marine realm would have continued to eliminate Paleozoic
marine world fauna in favor of the modern evolutionary fauna in an
inexorable process of natural selection.
However, the story might have been very di erent for the Paleozoic
terrestrial world fauna in a world in which the end-Permian mass
extinction had never happened. In such an alternative world, in the
absence of the catastrophic e ects of the Siberian LIP, would the diverse
land-dwelling vertebrate faunas of the Paleozoic world have been replaced
almost entirely by reptiles (tables 6.7 and 6.8)? And if the synapsids had
not been weakened by the end-Permian mass extinction, might they not
have been replaced by dinosaurs in the Late Triassic (table 6.9)—might the
synapsids have continued to be the numerically and ecologically
dominant land vertebrates? Might the synapsids have evolved
endothermic mammals before the dinosaurs, and might the dominant
mammals have kept the dinosaurs in check for the next 150 million years
—rather than the dinosaurs keeping the mammals in check, as happened
in our world in the Mesozoic?
The dinosaurs evolved a robust ecosystem structure that persisted for
150 million years before being destroyed in the environmental
catastrophe triggered by the Chicxulub asteroid strike at the end of the
Cretaceous. After the collapse of the dinosaur ecosystem, the same
ecosystem structure was convergently evolved by the mammals in the
Cenozoic (table 7.2). It is important to note that the structure of the 19
ecological roles, or niches, listed in table 7.2 rst appeared in the
evolution of the dinosaurian-dominated ecosystems of the Mesozoic
world, not in the mammalian-dominated ecosystem structure of the
Cenozoic world. Mesozoic dinosaurs did not independently evolve the
ecological roles of lions, wild cats, wolves, wolverines, anteaters,
elephants, deer, bison, rhinoceroses, glyptodonts, goats, and ground sloths
—it is the Cenozoic mammal fauna that has convergently re lled the
ecological niches of allosaurs, coelophysises, velociraptors, troodonts,
alvarezsaurids, sauropods, hypsilophodonts, hadrosaurs, triceratopses,
ankylosaurs, pachycephalosaurs, and therizinosaurs! Some of these two
groups of animals resemble each other morphologically (triceratopses and
rhinoceroses, ankylosaurs and glyptodonts, therizinosaurs and ground
sloths), whereas the others appear radically di erent—the mammalian
predators are all quadrupeds, whereas the dinosaurian predators walked
on their hind legs only. Yet they all were ecological analogs of each other,
making their living in the roughly same way. The goats, for example, have
a ruminant stomach system to process plant material, whereas the
pachycephalosaurs had a gizzard-like gastric mill to accomplish the same
purpose. The similarity in the herbivorous ecological roles of many of the
ornithopod dinosaurs, such as the hypsilophodonts and hadrosaurs, to
ungulate mammals prompted paleobiologists Matthew Carrano, Christine
Janis, and Jack Sepkoski to conclude: “Although late Mesozoic and late
Cenozoic terrestrial ecosystems were profoundly di erent in terms of
both animal and plant taxa, there may be universal constraints on the
ecological roles played by large herbivores, resulting in convergence in
morphology (and, by implication, behavioral ecology) between groups as
taxonomically distinct as dinosaurs and mammals” (Carrano et al. 1999,
256). Thus Carrano and colleagues suggest that Mesozoic hadrosaurs and
Cenozoic ungulates may provide an example of both ecological and
behavioral convergent evolution.8

TABLE 7.2 Convergent evolution of ecological-analog compositions of Mesozoic dinosaurian-

dominated ecosystems and Cenozoic mammalian-dominated ecosystems.

Convergent Mesozoic Ecosystem Cenozoic Ecosystem

Ecological Analog
1. Large ambush Allosaur (Dinosauria: Allosauridae; Lion (Mammalia: Felidae; Panthera
predator Allosaurus fragilis †Jurassic) leo)
2. Small ambush Coelophysis (Dinosauria: Coelophysidae; Wild cat (Mammalia: Felidae; Felis
predator Coelophysis rhodesiensis †Triassic) sylvestris)
3. Pursuit pack Velociraptor (Dinosauria: Wolf (Mammalia: Canidae; Canis
predator Dromaeosauridae; Velociraptor lupus)
mongoliensis †Cretaceous)
4. Nocturnal Troodont (Dinosauria: Troodontidae; Wolverine (Mammalia:
foraging predator Saurornithoides mongoliensis †Cretaceous) Mustelidae; Gulo gulo)
5. Ant eater “Desert bird” alvarezsaurid (Dinosauria: Anteater (Mammalia:
Alvarezsauridae; Shuvuuia deserti Myrmecophagidae; Myrmecophaga
†Cretaceous) tridactyla)
6. Large, herding, Sauropod (Dinosauria: Titanosauridae; Elephant (Mammalia:
browsing Titanosaurus madagascariensis Elephantidae; Loxodonta africana)
herbivore †Cretaceous)
7. Midsize, fast- Hypsilophodont (Dinosauria: Deer (Mammalia: Cervidae; Cervus
running, Hypsilophodontidae; Hypsilophodon foxii elaphus)
browsing †Cretaceous)
herbivore
8. Large, herding, Hadrosaur (Dinosauria: Hadrosauridae; Bison (Eutheria: Bovidae; Bison
mixed-feeding Parasaurolophus walkeri †Cretaceous) bison)
herbivore
9. Large, horned, Triceratops (Dinosauria: Ceratopsidae; Rhinoceros (Mammalia:
grazing herbivore Triceratops albertensis †Cretaceous) Rhinocerotidae; Rhinoceros
unicornus)
10. Large, Ankylosaur (Dinosauria: Glyptodont (Mammalia:
armored, grazing Ankylosauridae; Ankylosaurus Glyptodontidae; Doedicurus
herbivore magniventris †Cretaceous) clavicaudatus †Pleistocene)
11. Midsize Pachycephalosaur (Dinosauria: Goat (Mammalia: Bovidae; Capra
ruminant grazer Pachycephalosauridae; hircus)
Pachycephalosaurus wyomingensis
†Cretaceous)
12. Giant “Sloth-claw” therizinosaur (Dinosauria: Giant ground sloth (Mammalia:
frugivore (fruit Therizinosauridae; Nothronychus Megatheriidae; Nothrotheriops
eater) mckinleyi †Cretaceous) shastensis †Pleistocene)
13. Fluvial Spinosaur (Dinosauria: Spinosauridae; Gavialid crocodile (Archosauria:
piscivore ( sh Spinosaurus maroccanus †Cretaceous) Crocodilidae; Gavialis gangeticus)
eater)
14. Shallow- Plesiosaur (Diapsida: Sauropterygia: Sea lion (Mammalia: Otariidae;
marine piscivore Cryptocleididae; Cryptocleidus oxoniensis Zalophus californianus)
†Jurassic)
15. Small open- Ichthyosaur (Diapsida: Ichthyosauria: Porpoise (Mammalia:
ocean carnivore Ichthyosauridae; Ichthyosaurus platyodon Phocaenidae; Phocaena phocaena)
†Jurassic)
16. Large open- Pliosaur (Diapsida: Sauropterygia: Killer whale (Mammalia:
ocean carnivore Rhomaleosauridae; Rhomaleosaurus Delphinidae; Orcinus orca)
megacephalus †Jurassic)
17. Flying “Frog-jaw” pterosaur (Ornithodira: Bat (Mammalia: Vespertilionidae;
insectivore Pterosauria: Aneurognathidae; Myotis myotis)
(insect eater) Batrachognathus volans †Jurassic)
18. Flying “Bakony dragon” pterosaur Fruit bat (Mammalia:
frugivore (fruit (Ornithodira: Pterosauria: Pteropodidae; Rousettus
eater) Azhdarchidae; Bakonydraco galaczi aegyptiacus)
†Cretaceous)
19. Flying Pteranodon (Ornithodira: Pterosauria: Pelican (Aves: Pelecanidae;
piscivore Pteranodontidae; Pteranodon longiceps Pelicanus occidentalis)
†Cretaceous)

Source: Modi ed from McGhee (2011).

Note: The age of extinct species is marked with a dagger and a geologic date (e.g., †Jurassic).

Not all dinosaurian ecological roles are lled by Cenozoic mammals,

and not all mammalian roles were lled by dinosaurs. Although the otter
hunts sh in rivers, the gavialid crocodile is a much closer modern
ecological analog than the otter to the toothy spinosaur (table 7.2). The
dinosaurs were exclusively terrestrial animals, yet if we examine marine
and oceanic habitats, we nd that other Mesozoic animals independently
evolved ecological roles that are now played by Cenozoic mammals: the
plesiosaur, ichthyosaur, and pliosaur ecological niches have been
convergently re lled by Cenozoic sea lions, porpoises, and killer whales.
In the air, the insect-eating pterosaur and fruit-eating pterosaur niches
have been independently re-evolved by the insect-eating bats and the
fruit-eating bats. And the modern pelican, itself an avian dinosaur,
convergently re lled the niche of the Mesozoic pteranodon and even
resembles a small pteranodon in appearance. Mammals also have
convergently evolved ying sh-eaters, such as the greater bulldog bat,
Noctilio leporinus, although it looks nothing like a pteranodon (excepting
the wings).
In these 19 ecological roles (table 7.2), the Cenozoic ecosystem play is
scripted in the same way as the Mesozoic ecosystem play. The roles of the
ecological play have not changed, although the cast of actors changed
dramatically following the end-Cretaceous mass extinction. Would this
same ecosystem structure have evolved if the end-Permian mass
extinction had never happened, if the synapsids have evolved
endothermic mammals before the dinosaurs, and if the dinosaur-
dominated ecosystem structure of our Mesozoic world had never evolved?
That is, in an alternative world in which the end-Permian mass extinction
had never happened, would the mammals have evolved the same
ecosystem structure (table 7.2) back in the Late Triassic? Would the
dinosaurs have existed only as minor elements in that ecosystem
structure, coexisting with the mammals but being held in check by the
mammalian ecological dominants—the reverse of what happened in our
world in the Mesozoic?
If the mammals had “got there rst,” so to speak, and established a
mammalian-dominated ecosystem structure that persisted throughout
the Mesozoic, what would have happened when the Chicxulub asteroid
impacted the Earth at the end of the Cretaceous? That cosmic event would
still have happened, whether there had ever been an end-Permian mass
extinction or not, as that event had an extraterrestrial cause. In our world,
that impact-generated global environmental catastrophe triggered the
collapse of the dinosaurian ecosystem and the extinction of all of the
nonavian dinosaurs. In an alternative world in which the end-Permian
mass extinction had never happened, would the Chicxulub asteroid
impact have destroyed the mammalian ecosystem instead, and would the
dinosaurs have then convergently re lled the ecological niches vacated by
the extinct mammals? Would that alternative world have had a Mesozoic
fauna dominated by mammals, and a Cenozoic fauna dominated by
dinosaurs?
The answer to that hypothetical question is … probably not. One of the
traits that enabled the mammals to di erentially survive the Chicxulub
asteroid strike was their ability to burrow underground and hibernate.
That trait is an ancient synapsid trait, one that predates the evolution of
the mammals—and one that also enabled the synapsids to di erentially
survive the end-Permian mass extinction, as discussed in chapter 6. It was
no accident that the small-bodied, barrel-chested, burrowing dicynodont
synapsid Lystrosaurus survived the catastrophic environmental conditions
triggered by the Siberian LIP, and that approximately 90 percent of all
land vertebrates alive in the Early Triassic were lystrosaur synapsids
(Sahney and Benton 2008). (However, remember from chapter 6 that that
synapsid population resurgence was short and not sweet, as the reptilian
rhynchosaur and rauisuchian competitors were soon to evolve in the
Middle Triassic, and the deadly dinosaurs proliferated in the Late Triassic;
table 6.9). It is a curious quirk of evolution that this trait is found in both
highly derived, high-metabolic-rate endothermic mammals and ancient,
primitive, ectothermic animals like amphibians—animals that are not
even amniotes. The phylogenetic distribution of this trait explains some
of the great mystery of why almost all of the highly derived, ecologically
advanced, high-metabolic-rate dinosaurs were exterminated by the end-
Cretaceous mass extinction, whereas very ancient ectothermic animals
like frogs and snakes survived: the dinosaurs did not burrow underground
and they did not hibernate. The dinosaurs, to our knowledge, never
evolved the ecological equivalents of mammalian moles, woodchucks,
rabbits, and the like.
Large body size is always a survival liability in times of ecological crisis
—large-bodied animals need too much food—and in a hypothetical world
in which the mammals were ecologically dominant when the Chicxulub
asteroid struck the Earth, all of the large-bodied mammals would no doubt
have perished, just as all of the large-bodied dinosaurs did in our world.
But—just as actually happened in our world—diverse lineages of mammals
that could burrow and hibernate would have survived to rediversify in the
postextinction hypothetical world. The small-bodied dinosaurs did not
have this capability, and they also perished, along with their large-bodied
kin, in the end-Cretaceous mass extinction. Only the avian dinosaurs
survived, probably as a result of both small body size and their ability to
y: they could cover large areas in search of food when food was in very
short supply in the post-asteroid-impact world. No doubt vast,
uncountable numbers of birds starved to death at the end of the Mesozoic
terrestrial world, but enough passed through the evolutionary bottleneck
to continue the survival of the dinosaurian lineage to this day.
In summary, rather than the end-Permian mass extinction, it was the
end-Cretaceous mass extinction that created our modern Cenozoic
terrestrial world with its mammal-dominated ecosystems (McGhee et al.
2004). Even in a hypothetical world in which the end-Permian mass
extinction never happened, in which the dinosaurian ecosystem never
evolved, and in which the mammals became dominant in the Late Triassic,
the selective e ect of the end-Cretaceous mass extinction would still have
resulted in a postapocalyptic terrestrial world that continued to be
dominated by mammals, not by dinosaurs.
 However, if the end-Permian mass extinction had never happened and
the dinosaurian ecosystem had never evolved, then the mammals would
have escaped being ecologically dominated and preyed upon by the
dinosaurs. Mammals never would have been forced to remain small,
mouse- or chipmunk-size creatures during the Jurassic and Cretaceous,
digging underground to hide from the terrible small coelurosaurian
dinosaur predators, forced to remain nocturnal creatures scurrying
furtively for food in the dark of the night in hopes of not being noticed by
nocturnally hunting troodont dinosaurs. The 150-million-year-long
nightmare of our mammalian ancestors might never have happened.
 Notes

1. HARBINGERS OF THE LATE PALEOZOIC ICE AGE

1. The percentage of carbon dioxide in the atmosphere is now 0.04 and steadily rising because of
the burning of fossil hydrocarbons—many of Carboniferous age.
2. Anaerobic photosynthesizing bacteria: purple sulfur, purple nonsulfur, green sulfur, green
nonsulfur, and heliobacteria. Aerobic photosynthesizing bacteria: cyanobacteria.
3. CO2 + H2O → CH2O + O2↑.
4. CH4 + 2O2 → CO2 + 2H2O.
5. Five to 18 percent of the atmosphere’s present 21 percent level of oxygen (Lane 2002).
6. Κρυος + γενής.
7. Given its short duration, some question whether the Gaskiers was a snowball-Earth glaciation
at all—it may instead have been the rst of the Phanerozoic-style glaciations (Pu et al. 2016).
8. Φανερός + ζωον.
9. The discovery of soft-tissue fossils of the enigmatic hyoliths has now demonstrated that the
hyoliths were lophophorates (Moysiuk et al. 2017).
10. In the Oligocene Epoch.
11. Aerobic: CO2 + H2O → CH2O + O2; anaerobic: CO2 + 2H2S → CH2O + H2O + 2S.
12. CO2 + CaSiO3 → CaCO3 + SiO2.
13. CH4 + 2O2 → CO2 + 2H2O.
14. A mineral formed at very high pressures and temperatures deep in the Earth.
15. Measuring the total duration of the Late Paleozoic Ice Age from an onset in the late Frasnian
Age of the Late Devonian, 375 million years ago, to an ending in the Wuchiapingian Age of the
Late Permian, about 260 million years ago.
16. Another signi cant accumulation of coal-rich strata also occurred in the Paleogene (Nelsen et
al. 2016).
17. Zachos et al. (2001) estimate the size of the Oi-1 ice sheet to have been 50 percent of the
present-day ice sheet, or 7.0 × 106 km2, which is here taken as a minimum estimate. Pusz et al.
 (2011) give an average estimate of the Oi-1 ice sheet that is 85 percent of the present-day ice
sheet, or 11.9 × 106 km2, which is here taken as a maximum estimate.
18. By 12 million years ago in the Miocene, sea level had fallen to the level present in our modern-
day state of glaciation (Westerhold et al. 2005); thus the initial estimate of the size of the
Miocene ice sheet in Antarctica is 14 × 106 km2, the same as today, and this value is here taken
as a minimum estimate. However, Wilson and Luyendyk (2009) have argued that the
Transantarctic Mountains in Antarctica are a small remnant of a previous highland and that
West Antarctica once was exposed above sea level, increasing the total land area of Antarctica
by 10 to 20 percent in the earlier Cenozoic. The Miocene ice sheet may thus have been as
large as 15.4 to 16.8 × 106 km2, and the upper value is here taken as a maximum estimate.
19. 250 μm and larger.
20. See note 18.
21. The Famennian Gap spanned ten conodont zones and was estimated to have been seven
million years in duration based on the Gradstein et al. (2004) geologic timescale. However, the
duration of the Famennian Age has been shortened from 15.3 to 13.3 million years in the new
Gradstein et al. (2012) timescale, leading to a new estimate of 0.6 million years’ duration for
each of the 22 Famennian conodont zones. As the Famennian Gap spanned ten of those zones,
ten zones times 0.6 million-years-per-zone yields the new duration estimate of six million
years for the Famennian Gap.
22. The duration of the Tournaisian Gap was once thought to have been some 14 million years—
the duration of the entire Tournaisian Age in the Gradstein et al. (2004) timescale—and twice
the duration of the original seven-million-year estimate of the duration of the Famennian
Gap (see tables 6.2 and 6.7 in McGhee 2013). However, while McGhee (2013) was in press,
Smithson et al. (2012) argued that the Tournaisian Gap (which they called Romer’s Gap)
lasted for ten million years, not 14 million, before land vertebrate faunas began to recover
from the Lilliput-e ect small-body-size constraint imposed by “adverse conditions, such as
aridity of other climatic conditions” (Smithson et al. 2012, 4535). It should be noted that this
new ten-million-year estimate of the duration of the Tournaisian Gap is based solely on the
land vertebrate fossil record and has yet to be corroborated by data from the marine fossil
record and the land-plant fossil record, both of which were also a ected by the climatic
conditions of the Tournaisian Gap (see McGhee 2013, 184–188).
23. Additional geochemical similarities between the onset of the Cenozoic and Late Paleozoic Ice
Ages also exist; for example, both the Miocene and Famennian glaciations were characterized
by positive carbon-isotope anomalies of +0.8‰ δ13C and +1.2‰ δ13C, respectively (Zachos et
al. 2001; Kaiser et al. 2006, 2016); oxygen-isotope increases of 0.3–1.0 ‰ δ18O and 0.8–1.2 ‰
δ18O, respectively (Flower and Kennett 1995; Kaiser et al. 2006); and estimated drops in sea-
surface temperature of 6–7°C in the Miocene and at least 2–4°C in the Famennian, based on
partial data (Shevenell et al. 2004; Kaiser et al. 2006). Likewise, both the Oligocene and
proposed Frasnian glaciations were characterized by positive carbon-isotope anomalies of
+0.8‰ δ13C and +3.0‰ δ13C, respectively (Zachos et al. 2001; Joachimski and Buggisch 2002);
 oxygen-isotope increases of 0.5–1.0‰ δ18O and 1.0–1.5‰ δ18O, respectively (Pusz et al. 2011;
Joachimski and Buggisch 2002); and estimated drops in sea-surface temperature of 3–4°C in
the Oligocene, based on partial data, and 5–7°C in the Frasnian (Wade et al. 2012; Joachimski
and Buggisch 2002).

2. THE BIG CHILL

1. For a detailed examination of the aftermath of the Devonian extinctions, see McGhee (2013).
2. A lyginopterid spermatophyte, an extinct group of seed plants.
3. Lophophorate lophotrochozoans. Brachiopods still exist today, but in very low diversity in
modern oceans.
4. Eutrochozoan lophotrochozoans.
5. It was originally argued that the Tournaisian Gap lasted 14 million years (McGhee 2013, 184–
188), but a more recent estimate is ten million years (Smithson et al. 2012, 4535). For a more
detailed discussion of the data, see note 22 in chapter 1.
6. For a detailed examination of the e ect of the Devonian extinctions on the tetrapod faunas,
see McGhee (2013).
7. For a detailed examination of the post-Tournaisian-Gap fauna, see McGhee (2013).
8. Tentative evidence from bone fragments (Daeschler et al. 2009) suggest that a whatcheeriid-
like tetrapod may have been present in Pennsylvania, USA, in the latest Famennian. If true,
this makes the enigma of the lack of further evolution in the family during the ten million
years of the Tournaisian Gap even more perplexing.
9. Dated from the Lower crenulata through the isosticha conodont zones; see discussion in
McGhee (2013, 196–199).
10. Possible further geochemical corroboration of this assessment is the documented presence of
a large-magnitude positive anomaly in carbon-isotope ratios in strata dating from the middle
Tournaisian in both North America and Europe. We saw in chapter 1 that the onset of
glaciations in both the Late Paleozoic and Cenozoic ice ages was accompanied by positive
carbon-isotope anomalies (see note 23 in chapter 1), and the mid-Tournaisian anomaly is
“one of the largest known Phanerozoic δ13C events” (Saltzman et al. 2000).
11. Dated from the Upper commutata through the Middle bilineatus conodont zones; see
discussion in McGhee (2013, 196–199).
12. The midpoint of the Visean, which spans in geologic time from 330.9 million to 346.7 million
years ago, is 338.8 million years.
13. δ18O = 1.8‰ (Mii et al. 2001).
14. See Epshteyn (1981a), who argues that Ustritsky’s (1973) Early Permian (Sakmarian) glacial
strata are Bashkirian-Muscovian in age.
15. However, the marine fauna of high-latitude regions did continue to experience uctuations in
extinction rates during the waxing and waning of glaciers during the Late Carboniferous; see
the analysis of Balseiro (2016).
16. A United States quarter-dollar coin is 24 millimeters in diameter.
17. Benton (1989) documents the extinction of only four families of tetrapods in the
Serpukhovian, and all four were of low species diversity.
18. See the extensive discussion of the evolution of amniotes, and of the evolution of ight in
insects, in McGhee (2013).
19. Both are found in the famous Joggins tree-stump Lagerstätte in Nova Scotia, Canada; see
discussion in McGhee (2013).
20. The question of the evolution of the rst winged insects and their subsequent diversi cation
becomes even more enigmatic if molecular analyses are included. Molecular phylogenies
with divergence-date estimates have predicted the evolution of wings, hence the rst
Pterygota, to have occurred 406 million years ago in the Emsian Age of the Early Devonian
(Misof et al. 2014)! If true, this would mean that insects possessing the key adaptation of
wings still did not diversify and did not achieve large population sizes until the Bashkirian
Age of the Late Carboniferous, some 83 million years later. Ecologically, that scenario does not
make sense.

3. THE LATE CARBONIFEROUS ICE WORLD

1. Based on molecular phylogenies; see discussion in Bell et al. (2010).

2. Not all calamitean trees grew from rhizomes; see Rößler et al. (2012).
3. However, some cordaitean trees grew as tall as 30 meters; see Stewart (1983).
4. CO2 + H2O → CH2O + O2↑.
5. More precisely, the strata are dated to the Middle expansa conodont zone, the VH spore zone,
and the Fa2c chronozone of older timescales; see McGhee (2013).
6. The late Robert Berner (1935–2015) did modify the Geocarbsulf model in a brief, three-page
paper in 2009, bringing its predicted oxygen values closer to the older Rock-Abundance
model. Still, the revised Geocarbsulf model predicts that oxygen levels did not increase to 30
percent or higher until the latest Carboniferous and, more signi cantly, actually predicts a
lower oxygen content in the atmosphere in the Early Carboniferous than the original
Geocarbsulf model! The Early Carboniferous is the same time interval in which animal
gigantism evolved in numerous independent species lineages, a biological phenomenon that
has been used to argue for a hyperoxic atmosphere on the Earth, as will be discussed in
chapter 4.
7. Some peat res do occur today in the tropics, during the dry season.
 8. To help the reader, here is a rough correlation of some of the major older Carboniferous
geologic time divisions with the modern time scale: Namurian A (pars) = Serpukhovian;
Namurian A (pars), B, C, and Westphalian A, B = Bashkirian; Westphalian C, D, and Stephanian
(Cantabrian pars) = Moscovian; Stephanian (Cantabrian pars) A, B = Kasimovian; and
Stephanian C, Autunian (pars) = Gzhelian.
9. Παν + γαια.
10. See also DiMichele et al. (2009, 210), in which di erences in interpretation of the wettest part
of cyclothemic rhythms are discussed—one interpretation holding that the wettest periods
occurred during periods of sea-level lowstand, another holding that the wettest period
occurred during sea-level highstand.

4. GIANTS IN THE EARTH …

1. Some have argued that Megarachne servinei was a giant land-dwelling eurypterid water
scorpion, not a spider. If true, this would simply mean that this giant predatory arthropod
was a merostome chelicerate rather than an arachnid chelicerate (table 4.1).
2. Giant arthropleuran fossils are reported from strata ranging in age from the Carboniferous
Westphalian C [= Moscovian] to the Permian Lower Rotliegend [= Asselian].
3. Data from Clapham and Kerr (2012; supplemental tables S1, wing length, and S2, body width).
Total wingspan is calculated by the formula: Wingspan = 2(wing length) + (body width).
4. The basal batrachomorphs are often called “temnospondyls” in the older literature, and the
basal reptiliomorphs are called “anthracosaurs.” Both of these older taxonomic groupings are
paraphyletic.
5. A United States quarter-dollar coin is 24 millimeters in diameter.
6. Speci cally, the Cryogenian Period of the Neoproterozoic Era; see Erwin et al. (2011).
7. The most common symbionts are species of the dino agellate genus Symbiodinium; see table
4.3 for dino agellate evolutionary relationships.
8. Speci cally, species in the fusulinacean families Neoschwagerinidae and Verbeekinidae; see
discussion in McGhee et al. (2013).
9. For a discussion of the convergent evolution of ecological niches in Cenozoic mammals and
Mesozoic dinosaurs, see McGhee (2011).
10. Named for the vertebrate paleontologist Everett C. Olson, who spent his life studying these
vertebrates; Sahney and Benton (2008).
11. Speci cally, eruption of the Emeishan Large Igneous Province in China, which we will examine
in detail in chapter 6; see also McGhee et al. (2013).
12. These animals were the rst of many vertebrate lineages to convergently evolve the ability to
glide; see McGhee (2011).
13. These animals were the rst of many vertebrate lineages to convergently evolve bipedalism;
see McGhee (2011).

5. THE END OF THE LATE PALEOZOIC ICE AGE

1. See the paleogeographic map given in gure 5 of the paper by Isbell et al. (2016). However,
these same authors argue that the region was ice free in the Permian.
2. See the new dating of the P3 and P4 glacial pulses given in Metcalfe et al. (2015).
3. See note 1.
4. Παν + θάλασσα.
5. Παν + γαια; see discussion of the tectonic e ects of the assembly of Pangaea in chapter 3.
6. The Ophiacodontidae, Edaphosauridae, Eothyrididae, and Sphenacodontidae; see Kemp (2006).

6. THE END OF THE PALEOZOIC WORLD

1. LIP is pronounced like the word “lip,” and not spelled out L-I-P like an acronym.
2. The term “large igneous province” was introduced by Co n and Eldholm (1991, 1994); see
discussion in Saunders (2005). For comprehensive discussions of the LIPs that have occurred
in geologic time, see Bryan and Ernst (2008) and Bryan et al. (2010).
3. A LIP geographic region sometimes also is called the “Traps,” as in the “Emeishan Traps.” The
designation “trap” comes from “traprock,” an informal name commonly used by miners for
basaltic rocks.
4. For extensive eyewitness accounts of this period of crisis in Europe, see tables A1 and A2 in
Thordarson and Self (2003).
5. Some researchers have questioned the causal link between the Laki eruption and the
abnormally cold winter of 1783–1784; see, for example, D’Arrigo et al. (2011); Lanciki et al.
(2012). For a counterargument to these studies, see Schmidt et al. (2012).
6. (500 eruptions/100,000 years) × (20 km3 lava/eruption) = 10,000 km3 lava/100,000 years;
1,500,000 years/100,000 years = 15; thus 15 × (10,000 km3 lava)/ 15 × (100,000 years) = 150,000
km3 lava/1,500,000 years.
7. The degree of doming produced by the Emeishan super plume continues to be debated;
Ukstins-Peate and Bryan (2008) argue that no doming at all occurred. For a detailed
discussion of the debate, see Shellnutt (2013).
8. CH4 + 2O2 → CO2 + 2H2O.
9. For every molecule of CH4 oxidized, two molecules of O2 are removed from the atmosphere
and one molecule of CO2 is released into the atmosphere; see note 8.
10. Families Neoschsagerinidae and Verbeedinidae; see Vachard et al. (2010).
11. Sta elids and schubertellids; see Vachard et al. (2010).
12. The Alatoconchidae; see Aljinović et al. (2008); Bond, Wignall, et al. (2010).
13. The model of Ganino and Arndt (2009) estimates a total emission of 78.4 to 162.4 trillion
tonnes of CO2 into the atmosphere, based on the assumption of a 1 × 106 km3 LIP volume. Our
current maximum estimates of the original Emeishan LIP volume are about a third smaller, at
0.6 × 106 km3. Thus I have reduced the original Ganino and Arndt (2009) CO2 injection-mass
estimates by one-third.
14. The model of Self et al. (2005) estimates a total emission of 11.7 billion tonnes of SO2 into the
atmosphere from a small LIP magma volume of 1.3 × 103 km3. Scaling this estimate up to an
Emeishan LIP magma volume of 0.6 × 106 km3 results in a SO2 injection mass of 5.4 trillion
tonnes.
15. 12CO + H O → 12CH O + O .
2 2 2 2

16. Ca12CO3.
17. The index is calculated as follows: δ13C = [(13C/12C)sample / (13C/12C)standard − 1] × 103, where
values are reported as per mille (‰) relative to a standard. For many biostratigraphic studies,
the standard taken is the isotopic ratio of a fossil belemnite from the Cretaceous Pee Dee
Formation of the Carolinas, the “Pee Dee Belemnite” standard or simply δ13C(pdb).
18. Methane clathrates.
19. Dolerite is a coarse-crystalline form of basalt formed at depth within the Earth.
20. The Olenekian Age is subdivided, from oldest to youngest, into the Dienerian, Smithian, and
Spathian sub-ages; the highest temperatures occur in the Smithian (Sun et al. 2012).
21. Just as the photosynthetic process produces oxygen in the production of hydrocarbons, CO2 +
H2O → CH2O + O2, the burning of those hydrocarbons consumes oxygen, CH2O + O2 → CO2 +
H2O.
22. CH4 + 2O2 → CO2 + 2H2O.
23. The most general depletion cycle is Cl + O3 → ClO + O2, which destroys one ozone molecule
(O3), and then ClO + O → Cl + O2, which simultaneously prevents atomic oxygen (O) from
producing a new ozone molecule (O + O2 → O3) and produces a free chlorine atom to start the
cycle all over again; see Beerling et al. (2007).
24. For contrast, humans see light in the 400 (violet) to 700 (red) nanometer spectrum.
25. In the timescale of Gradstein et al. (2012).
26. For example, in 1998 Bowring et al. narrowed the time interval down to 900,000 years, from
252.3 to 251.4 Ma; in 2004 Mundil et al. narrowed the time interval down to 400,000 years,
from 252.8 to 252.4 Ma; and in 2011 Shen et al. narrowed the time interval down to 200,000
years, from 252.3 to 252.1 Ma.
27. Γαια.
28. However, it was at the University of Rochester in upstate New York that Jack Sepkoski
collected most of the massive amount of data that he used in his analyses. For a brief glimpse
into graduate student life and paleontological research conducted at the University of
Rochester in those years, see McGhee (1996, 168–171).
29. For a detailed discussion of convergent ecological evolution, see McGhee (2011).
30. For an analysis of the di erential fates of the rhynchonelliform brachiopods in the acidic
oceans of the end-Permian, see Garbelli et al. (2017).
31. For an extensive discussion of the phenomenon of ecological convergence, see McGhee (2011).
32. The chitinozoans were the egglike reproductive phase of as yet unknown small invertebrate
animals; see Grahn and Paris (2011).
33. Jack Sepkoski tragically died at age 50 in the year 1999.
34. For a discussion of the convergent evolution of swimming morphologies, see McGhee (2011).
35. The volaticotherians. For a discussion of the convergent evolution of ight morphologies, see
McGhee (2011).
36. For a discussion of the convergent evolution of gliding morphologies in animals, see McGhee
(2011).
37. Unfortunately, too late, as the predatory dinosaurs had already evolved at the beginning of
the Carnian and diversi ed rapidly.

7. THE LEGACY OF THE LATE PALEOZOIC ICE AGE

1. The excess carbon dioxide is overwhelmingly the lighter isotope of carbon, 12CO , which
2
would be released by the burning of coal and other hydrocarbons. As we are not living at the
end of the Permian, when coal strata and petroleum-rich strata were burned by mantle-
plume magma (chapter 6), the only other source of large excesses of light-isotope carbon
dioxide in the atmosphere is anthropogenic.
2. For a detailed examination of the process of the invasion of land by vertebrates, see McGhee
(2013).
3. See the more extensive discussion of this evolutionary possibility in McGhee (2013).
4. For a more extensive discussion of the e ects of the End-Frasnian Bottleneck, see McGhee
(2013).
5. For a more extensive discussion of the e ects of the End-Famennian Bottleneck, see McGhee
(2013).
6. See the detailed paleogeographic map sequences in Blakey (2008).
7. See the detailed paleogeographic map sequences in Blakey (2008).
8. For an extensive discussion of the phenomenon of behavioral convergent evolution, see
McGhee (2011).
 References

Abdala, F., and A. M. Ribeiro. 2010. Distribution and diversity patterns of Triassic cynodonts
(Therapsida, Cynodontia) in Gondwana. Palaeogeography, Palaeoclimatology, Palaeoecology 286:202–
217.
Ahlberg, P. E., J. A. Clack, E. Lukševičs, H. Blom, and I. Zupinš. 2008. Ventastega curonica and the
origin of tetrapod morphology. Nature 453:1199–1204.
Ahlbert, P. E. and Z. Johanson. 1998. Osteolepiforms and the ancestry of tetrapods. Nature 395:792–
794.
Algeo, T. J., R. A. Berner, J. B. Maynard, and S. E. Scheckler. 1995. Late Devonian oceanic anoxic
events and biotic crises: “Rooted” in the evolution of vascular land plants? GSA Today 5:63–66.
Algeo, T. J., S. E. Scheckler, and J. B. Maynard. 2001. E ects of the Middle to Late Devonian spread of
vascular land plants on weathering regimes, marine biotas, and global climate. In Plants invade
the land, ed. P. G. Gensel and D. Edwards, 213–236. New York: Columbia University Press.
Ali, J. R., G. M. Thompson, M.-F. Zhou, and X. Song. 2005. Emeishan large igneous province, SW
China. Lithos 79:475–489.
Aljinovic, D., Y. Isozaki, and J. Sremac. 2008. The occurrence of giant bivalve Alatoconchidae from
the Yabeina zone (Upper Guadalupian, Permian) in European Tethys. Gondwana Research 13:275–
287.
Archer, D., H. Kheshgi, and E. Maier-Reimer. 1997. Multiple timescales for neutralization of fossil
fuel CO2. Geophysical Research Letters 24:405–408.
Averbuch, O., N. Tribovillard, X. Devleeschouwer, L. Riquier, B. Mistiaen, and B. van Vliet-Lanoe.
2005. Mountain building-enhanced continental weathering and organic carbon burial as major
causes for climatic cooling at the Frasnian-Famennian boundary (c. 376 Ma)? Terra Nova 17:25–
34.
Babarro, J. M. F., and A. De Zwaan. 2008. Anaerobic survival potential of four bivalves from di erent
habitats: A comparative survey. Comparative Biochemistry and Physiology, Part A 151:108–113.
Baker, R. A., and W. A. DiMichele. 1997. Biomass allocation in Late Pennsylvanian coal-swamp
plants. Palaios 12:127–132.
Balseiro, D. 2016. Compositional turnover and ecological changes related to the waxing and waning
of glaciers during the late Paleozoic ice age in ice-proximal regions (Pennsylvanian, western
Argentina). Paleobiology 42:335–357.
Bambach, R. K. 1983. Ecospace utilization and guilds in marine communities through the
Phanerozoic. In Biotic interactions in recent and fossil benthic communities, ed. M. J. S. Tevesz and P.
L. McCall, 719–746. New York: Plenum.
Bambach, R. K. 1985. Classes and adaptive variety: The ecology of diversi cation in marine faunas
through the Phanerozoic. In Phanerozoic diversity patterns, ed. J. W. Valentine, 191–253.
Princeton: Princeton University Press.
Bambach, R. K., A. H. Knoll, and S. C. Wang. 2004. Origination, extinction, and mass depletions of
marine diversity. Paleobiology 30:522–542.
Barham, M., J. Murray, M. M. Joachimski, and D. M. Williams. 2012. The onset of the Permo-
Carboniferous glaciation: Reconciling global stratigraphic evidence with biogenic apatite δ18O
records in the late Visean. Journal of the Geological Society, London 169:119–122.
Becker, R. T., M. Piecha, M. Gereke, and K. Spellbrink. 2016. The Frasnian/Famennian boundary in
shelf basin facies north of Diemelsee-Adorf. Münster Forschungen zur Geologie und Paläontologie
108:220–231.
Beerling, D. J., and R. A. Berner. 2000. Impact of a Permo-Carboniferous high O2 event on the
terrestrial carbon cycle. Proceedings of the National Academy of Sciences USA 97:12428–12432.
Beerling, D. J., M. Harfoot, B. Lomax, and J. A. Pyle. 2007. The stability of the stratospheric ozone
layer during the end-Permian eruption of the Siberian Traps. Philosophical Transactions of the
Royal Society (London) A365:1843–1866.
Beerling, D. J., F. I. Woodward, M. R. Lomas, M. A. Wills, W. P. Quick, and P. J. Valdes. 1998. The
in uence of Carboniferous palaeo-atmospheres on plant function: An experimental and
modelling assessment. Philosophical Transactions of the Royal Society (London) B353:131–140.
Bell, C. D., D. E. Soltis, and P. S. Soltis. 2010. The age and diversi cation of the angiosperms re-
revisited. American Journal of Botany 97:1296–1303.
Benton, M. J. 1985. Patterns in the diversi cation of Mesozoic non-marine tetrapods and problems
in historical diversity analysis. Special Papers in Palaeontology 33:185–202.
Benton, M. J. 1989. Mass extinctions among tetrapods and the quality of the fossil record.
Philosophical Transactions of the Royal Society (London) B325:369–386.
Benton, M. J. 2003. When life nearly died: The greatest mass extinction of all time. London: Thames and
Hudson.
Benton, M. J. 2005. Vertebrate palaeontology. 3d ed. Oxford: Blackwell.
Benton, M. J. 2015. Vertebrate palaeontology. 4th ed. Chichester: Wiley Blackwell.
Berner, R. A. 2002. Examination of hypotheses for the Permo-Triassic boundary extinction by
carbon cycle modeling. Proceedings of the National Academy of Sciences USA 99:4172–4177.
Berner, R. A. 2006. GEOCARBSULF: A combined model for Phanerozoic atmosphere O2 and CO2.
Geochimica et Cosmochimica Acta 70:5653–5664.
Berner, R. A. 2009. Phanerozoic atmospheric oxygen: New results using the GEOCARBSULF model.
American Journal of Science 309:603–606.
Berner, R. A., D. J. Beerling, R. Dudley, J. M. Robinson, and R. A. Wildman. 2003. Phanerozoic
atmospheric oxygen. Annual Review of Earth and Planetary Sciences 31:105–134.
Bishop, J. W., I. P. Montañez, E. L. Gulbranson, and P. L. Brenckle. 2009. The onset of mid-
Carboniferous glacio-eustacy: Sedimentologic and diagenetic constraints, Arrow Canyon,
Nevada. Palaeogeography, Palaeoclimatology, Palaeoecology 276:217–243.
Black, B. A., J.-F. Lamarque, C. A. Shields, L. T. Elkins-Tanton, and J. T. Kiehl. 2014. Acid rain and
ozone depletion from pulsed Siberian Traps magmatism. 2014. Geology 42:67–70.
Blackburn, T. M., and K. J. Gaston. 1994. Animal body size distributions: Patterns, mechanisms and
implications. Trends in Ecology and Evolution 9:471–474.
Blakey, R. C. 2008. Gondwana paleogeography from assembly to breakup—A 500 m.y. odyssey. In
Resolving the Late Paleozoic Ice Age in time and space (Special Paper 441), ed. C. R. Fielding, T. D.
Frank, and J. L. Isbell, 1–28. Boulder, Colo.: Geological Society of America.
Blieck, A. 2011. From adaptive radiations to biotic crises in Palaeozoic vertebrates: A geobiological
approach. Geologica Belgica 14:203–227.
Bond, D. P. G., J. Hilton, P. B. Wignall, J. R. Ali, L. G. Stevens, Y. Sun, and X. Lai. 2010. The Middle
Permian (Capitanian) mass extinction on land and in the oceans. Earth-Science Reviews 102:100–
116.
Bond, D. P. G., and P. B. Wignall. 2014. Large igneous provinces and mass extinctions: An update. In
Volcanism, impacts, and mass extinctions: Causes and e ects, ed. G. Keller and A. C. Kerr, 29–55.
Boulder, Colo.: Geological Society of America Special Paper 505.
Bond, D. P. G., P. B. Wignall, W. Wang, G. Izon, H.-S. Jiang, X.-L. Lai, Y.-D. Sun, R. J. Newton, L.-Y. Shao,
S. Védrine, and H. Cope. 2010. The mid-Capitanian (Middle Permian) mass extinction and
carbon isotope record of South China. Palaeogeography, Palaeoclimatology, Palaeoecology 292:282–
294.
Bonelli, J. R., Jr., and M. E. Patzkowsky. 2011. Taxonomic and ecologic persistence across the onset of
the late Paleozoic ice age: Evidence from the upper Mississippian (Chesterian Series), Illinois
Basin, United States. Palaios 26:5–17.
Bottjer, D. J., and W. I. Ausich 1986. Phanerozoic development of tiering in soft substrata
suspension-feeding communities. Paleobiology 12:400–420.
Bowring, S. A., D. H. Erwin, Y. G. Jin, M. W. Martin, D. Davidek, and W. Wang. 1998. U/Pb zircon
geochronology and tempo of the end-Permian mass extinction. Science 280:1039–1045.
Boyce, C. K., and W. A. DiMichele. 2016. Arborescent lycopod productivity and lifespan:
Constraining the possibilities. Review of Palaeobotany and Palynology 227:97–110.
Braddy, S. J., M. Poschmann, and O. E. Tetlie. 2007. Giant claw reveals the largest ever arthropod.
Biology Letters 4:106–109.
Brezinski, D. K., C. B. Cecil, and V. W. Skema. 2010. Late Devonian glacigenic and associated facies
from the central Appalachian Basin, eastern United States. Geological Society of America Bulletin
122:265–281.
Briggs, D. E. G. 1985. Gigantism in Palaeozoic arthropods. Special Papers in Palaeontology 33:1571.
Brunton, C. H. C., S. S. Lazarev, and R. E. Grant. 2000. Productida. In Treatise on invertebrate
paleontology, part H, ed. R. L Kaesler, 350–643. Boulder, Colo: Geological Society of America and
Lawrence, Kan.: University of Kansas.
Bryan, S. E., and R. E. Ernst. 2008. Revised de nition of Large Igneous Provinces (LIPs). Earth-Science
Reviews 86:175–202.
Bryan, S. E., I. Ukstins-Peate, D. W. Peate, S. Self, D. A. Jerram, M. R. Mawby, J. S. Marsh, and J. A.
Miller. 2010. The largest volcanic eruptions on Earth. Earth-Science Reviews 102:207–229.
Caputo, M. V., J. H. G. Melo, M. Streel, and J. L. Isbell. 2008. Late Devonian and Early Carboniferous
glacial records of South America. In Resolving the Late Paleozoic Ice Age in time and space (Special
Paper 441), ed. C. R. Fielding, T. D. Frank, and J. L. Isbell, 161–173. Boulder, Colo.: Geological
Society of America.
Carmichael, S. K., J. A. Waters, C. J. Batchelor, D. M. Coleman, T. J. Suttner, E. Kido, L. M. Moore, and
L. Chadimová. 2016. Climate instability and tipping points in the Late Devonian: Detection of
the Hangenberg Event in an open oceanic island arc in the Central Asian Orogenic Belt.
Gondwana Research 32:213–231.
Carrano, M. T., C. M. Janis, and J. J. Sepkoski. 1999. Hadrosaurs as ungulate parallels: Lost lifestyles
and de cient data. Acta Palaeontological Polonica 44:237–261.
Carroll, R. 2009. The rise of amphibians: 365 million years of evolution. Baltimore: Johns Hopkins
University Press.
Cascales-Miñana, B., J. B. Diez, P. Gerrienne, and C. J. Cleal. 2015. A palaeobotanical perspective on
the great end-Permian biotic crisis. Historical Biology 28:1066–1074.
Chumakov, N. M. 1994. Evidence of Late Permian glaciation in the Kolyma River Basin: A
repercussion of the Gondwana glaciations in northeast Asia? Stratigraphy and Geological
Correlation 2:426–444.
Clack, J. A. 2002. Gaining ground: The origin and evolution of tetrapods. Bloomington: Indiana University
Press.
Clack, J. A. 2012. Gaining ground: The origin and evolution of tetrapods. 2d ed. Bloomington: Indiana
University Press.
Clack, J. A., P. E. Ahlberg, H. Blom, and S. M. Finney. 2012. A new genus of Devonian tetrapod from
north-east Greenland, with new information on the lower jaw of Ichthyostega. Palaeontology
55:73–86.
Clapham, M. E., and J. A. Kerr. 2012. Environmental and biotic controls on the evolutionary history
of insect body size. Proceedings of the National Academy of Sciences USA 109:10927–10930.
Clapham, M. E., and J. L. Payne. 2011. Acidi cation, anoxia, and extinction: A multiple logistic
regression analysis of extinction selectivity during the Middle and Late Permian. Geology
39:1059–1062.
Clapham, M. E., S. Z. Shen, and D. J. Bottjer. 2009. The double mass extinction revisited: Reassessing
the severity, selectivity, and causes of the end-Guadalupian biotic crisis (Late Permian).
Paleobiology 35:32–50.
Clarkson, E. N. K. 1998. Invertebrate palaeontology and evolution. Oxford: Blackwell Science.
Clarkson, M. O., S. A. Kasemann, R. A. Wood, T. M. Lenton, S. J. Daines, S. Richoz, F. Ohnemueller, A.
Meixner, S. W. Poulton, and E. T. Tipper. 2015. Ocean acidi cation and the Permo-Triassic mass
extinction. Science 348:229–232.
Cleal, C. J. 2010. Tiering on land—trees and forests (Late Palaeozoic). In Encyclopedia of life sciences
(ELS). Chichester: Wiley. doi: 10.1002/9780470015902.a0001642.pub2.
Cleal, C. J., and B. Cascales-Miñana. 2014. Composition and dynamics of the great Phanerozoic
Evolutionary Floras. Lethaia 47:469–484.
Cleal, C. J., R. M. James, and E. L. Zodrow. 1999. Variation in stomatal density in the Late
Carboniferous gymnosperm frond Neuropteris ovata. Palaios 14:180–185.
Cleal, C. J., and B. A. Thomas. 2005. Palaeozoic tropical rainforests and their e ect on global
climates: Is the past the key to the present? Geobiology 3:13–31.
Coates, M. I., M. Ruta, and M. Friedman. 2008. Ever since Owen: Changing perspectives on the early
evolution of tetrapods. Annual Review of Ecology, Evolution, and Systematics 39:571–592.
Co n, M. F., and O. Eldholm. 1991. Large igneous provinces: JOI/USSAC workshop report (Technical
Report 114). Austin: University of Texas at Austin Institute for Geophysics.
Co n, M. F., and O. Eldholm. 1994. Large igneous provinces: Crustal structure, dimensions, and
external consequences. Review of Geophysics 32:1–36.
Copper, P. 1977. Paleolatitudes in the Devonian of Brazil and the Frasnian-Famennian mass
extinction. Palaeogeography, Palaeoclimatology, Palaeoecology 21:165–207.
Copper, P. 1986. Frasnian-Famennian mass extinctions and cold water oceans. Geology 14:835–838.
Copper, P. 1998. Evaluating the Frasnian-Famennian mass extinction: Comparing brachiopod
faunas. Acta Palaeontologia Polonica 43:137–154.
Copper, P. 2002. Silurian and Devonian reefs: 80 million years of global greenhouse between two ice
ages. SEPM Special Publications 72:181–238.
Courtillot, V. E. 1999. Evolutionary catastrophes: The science of mass extinction. Cambridge: Cambridge
University Press.
Courtillot, V. E., V. A. Kravchinsky, X. Quidelleur, P. R. Renne, and D. P. Gladkochub. 2010.
Preliminary dating of the Viluy traps (Eastern Siberia): Eruption at the time of Late Devonian
extinction events? Earth and Planetary Science Letters 300:239–245.
Courtillot, V. E., and P. R. Renne. 2003. On the ages of ood basalt events. Comptes Rendus Geoscience
335:113–140.
Croswell, K. 2003. Magni cent Mars. New York: Free Press.
Daeschler, E. B., J. A. Clack, and N. H. Shubin. 2009. Late Devonian tetrapod remains from Red Hill,
Pennsylvania, USA: How much diversity? Acta Zoologica 90(Supplement 1):306–317.
D’Arrigo, R., R. Seager, J. E. Smerdon, A. N. LeGrande, and E. R. Cook. 2011. The anomalous winter of
1783–1784: Was the Laki eruption or an analog of the 2009–2010 winter to blame? Geophysical
Research Letters 38. doi:10.1029/2011GL046696.
DeConto, R. M., and D. Pollard. 2003. Rapid Cenozoic glaciation of Antarctica induced by declining
atmospheric CO2. Nature 421:245–249.
Demko, T. M., R. F. Dubiel, and J. T. Parrish. 1998. Plant taphonomy in incised valleys: Implications
for interpreting paleoclimate from fossil plants. Geology 26:1119–1122.
Dennison, R. 1978. Placodermi. In Handbook of paleoichthyology, ed. H. P. Schultze, 2:1–128. Stuttgart:
Gustav Fischer Verlag.
Dennison, R. 1979. Acanthodii. In Handbook of paleoichthyology, ed. H. P. Schultze, 5:1–62. Stuttgart:
Gustav Fischer Verlag.
Dickens, G. R. 2001. Modeling the global carbon cycle with a gas hydrate capacitor across the latest
Paleocene thermal maximum. In Natural gas hydrates: Occurrence, distribution and detection, ed. C.
K. Paull and W. P. Dillon. American Geophysical Union Geophysical Monographs 124:19–38. New
York: Wiley.
Dickens, G. R., J. R. O’Neil, D. K. Rea, and R. M. Owen. 1995. Dissociation of oceanic methane hydrate
as a cause of the carbon isotope excursion at the end of the Paleocene. Paleoceanography 10:965–
971.
DiMichele, W. A., and R. W. Hook. 1992. Paleozoic terrestrial ecosystems. In Terrestrial ecosystems
through time, ed. A. K. Behrensmeyer, J. D. Damuth, W. A. DiMichele, R. Potts, H.-D. Sues, and S. L.
Wing, pp. 205–325. Chicago: University of Chicago Press.
DiMichele, W. A., I. P. Montañez, C. J. Poulsen, and N. J. Tabor. 2009. Climate and vegetational regime
shifts in the late Paleozoic ice age earth. Geobiology 7:200–226.
DiMichele, W. A., H. W. Pfe erkorn, and R. A. Gastaldo. 2001. Response of Late Carboniferous and
Early Permian plant communities to climatic change. Annual Review of Earth and Planetary
Sciences 29:461–487.
DiMichele, W. A., and T. L. Phillips. 1996. Climate change, plant extinctions and vegetational
recovery during the middle-late Pennsylvanian transition: The case of tropical peat-forming
environments in North America. In Biotic recoveries from mass extinctions (Special Publication
102), ed. M. B. Hart, 210–221. London: Geological Society of London.
DiMichele, W. A., T. L. Phillips, and R. A. Peppers. 1985. The in uence of climate and depositional
environment on the distribution and evolution of Pennsylvanian coal swamp plants. In
 Geological factors in the evolution of plants, ed. B. Ti ney, 223–256. New Haven: Yale University
Press.
DiMichele, W. A., N. J. Tabor, D. S. Chaney, and W. J. Nelson. 2006. From wetlands to wet spots:
Environmental tracking and the fate of Carboniferous elements in Early Permian tropical oras.
In Wetlands through time (Special Paper 399), eds. S. F. Greb and W. A. DiMichele, 223–248.
Boulder, Colo: Geological Society of America.
Dobretsov, N. L., and V. A. Vernikovsky. 2001. Mantle plumes and their geologic manifestations.
International Geological Review 43:771–787.
Donoghue, M. J. 2005. Key innovations, convergence, and success: Macroevolutionary lessons from
plant phylogeny. Paleobiology 31(Supplement):77–93.
Dorrington, G. E. 2015. Heavily loaded ight and limits to the maximum size of dragon ies
(Anisoptera) and gri en ies (Meganisoptera). Lethaia 49:261–274.
Droser, M. L., D. J. Bottjer, P. M. Sheehan, and G. R. McGhee. 2000. Decoupling of taxonomic and
ecologic severity of Phanerozoic marine mass extinctions. Geology 28:675–678.
Dudley, R. 1998. Atmospheric oxygen, giant Paleozoic insects and the evolution of aerial locomotor
performance. Journal of Experimental Biology 201:1043–1050.
Ehrmann, W. U., and A. Mackensen. 1992. Sedimentological evidence for the formation of an East
Antarctic ice sheet in Eocene/Oligocene time. Palaeogeography, Palaeoclimatology, Palaeoecology
93:85–112.
Elrick, M., and B. Witzke. 2016. Orbital-scale glacio-eustacy in the Middle Devonian detected using
oxygen isotopes of condont apatite: Implications for long-term greenhouse-icehouse climatic
transitions. Palaeogeography, Palaeoclimatology, Palaeoecology 445:50–59.
Enos, P. 1995. The Permian of China. In The Permian of northern Pangea: Vol. 2. Sedimentary basins and
economic resources, ed. P. A. Scholle, T. M. Peryt, and D. S. Ulmer-Scholle, 225–256. Berlin:
Springer Verlag.
Epshteyn, O. G. 1981a. Middle Carboniferous ice-marine deposits of northeastern U.S.S.R. In Earth’s
pre-Pleistocene glacial record, ed. M. J. Hambrey and W. B. Harland, 268–269. Cambridge:
Cambridge University Press.
Epshteyn, O. G. 1981b. Late Permian ice-marine deposits of northeastern U.S.S.R. In Earth’s pre-
Pleistocene glacial record, ed. M. J. Hambrey and W. B. Harland, 270–273. Cambridge: Cambridge
University Press.
Erwin, D. H. 1993. The great Paleozoic crisis: Life and death in the Permian. New York: Columbia
University Press.
Erwin, D. H. 2015. Extinction: How life on earth nearly ended 250 million years ago. Princeton: Princeton
University Press.
Erwin, D. H., M. La amme, S. M. Tweedt, E. A. Sperling, D. Pisani, and K. J. Peterson. 2011. The
Cambrian conundrum: Early divergence and later ecological success in the early history of
animals. Science 334:1091–1097.
Falcon-Lang, H. J. 2004. Pennsylvanian tropical rain forests responded to glacial-interglacial
rhythms. Geology 32:689–692.
Falcon-Lang, H. J., and W. A. DiMichele. 2010. What happened to the coal forests during
Pennsylvanian glacial periods? Palaios 25:611–617.
Falkowski, P., M. Katz, A. Milligan, K. Fennel, B. Cramer, M. P. Aubry, R. A. Berner, and W. M. Zapol.
2005. The rise of atmospheric oxygen levels over the past 205 million years and the evolution of
large placental mammals. Science 309:2202–2204.
Fan, J. S., J. K. Rigby, and J. W. Qi. 1990. The Permian reefs of south China and comparisons with the
Permian reef complex of the Guadalupe mountains, west Texas and New Mexico. Brigham Young
University Geology Studies 36:15–55.
Farinella, P., L. Foschini, C. Froeschlé, R. Gonczi, T. J. Jopek, G. Longo, and P. Michel. 2001. Probable
asteroidal origin of the Tunguska Cosmic Body. Astronomy and Astrophysics 377:1081–1097.
Fielding, C. R., T. D. Frank, L. P. Birgenheier, M. C. Rygel, A. T. Jones, and J. Roberts. 2008.
Stratigraphic record and facies associations of the Late Paleozoic Ice Age in eastern Australia
(New South Wales and Queensland). In Resolving the Late Paleozoic Ice Age in time and space
(Special Paper 441), ed. C. R. Fielding, T. D. Frank, and J. L. Isbell, 41–57. Boulder, Colo.:
Geological Society of America.
Fielding, C. R., T. D. Frank, and J. L. Isbell. 2008. The Late Paleozoic Ice Age—a review of current
understanding and synthesis of global climate patterns. In Resolving the Late Paleozoic Ice Age in
time and space (Special Paper 441), ed. C. R. Fielding, T. D. Frank, and J. L. Isbell, 343–354.
Boulder, Colo.: Geological Society of America.
Filer, J. K. 2002. Late Frasnian sedimentation cycles in the Appalachian Basin: Possible evidence for
high frequency eustatic sea-level changes. Sedimentary Geology 154:31–52.
Finnegan, S., K. Bergmann, J. Eiler, D. Jones, D. Fike, I. Eisenman, N. Hughes, A. Tripati, and W.
Fischer. 2011. The magnitude and duration of Late Ordovician–Early Silurian glaciation. Science
331:903–906.
Flanagan, R. 1995. Mass murderers of the Devonian. Earth 4(August): 22.
Flower, B. P., and J. P. Kennett. 1995. Middle Miocene deepwater paleoceanography in the southwest
Paci c: Relations with the East Antarctica ice sheet development. Paleoceanography 10:1095–
1112.
Flügel, E., and W. Kiessling. 2002. Patterns of Phanerozoic reef crises. In Phanerozoic reef patterns, ed.
W. Kiessling, E. Flügel, and J. Golonka. SEPM Special Publications 72:691–733.
Flügel, E., and G. D. Stanley. 1984. Re-organization, development and evolution of post-Permian
reefs and reef-organisms. Paleontolographica Americana 54:177–186.
Fois, E., and M. Gaetani. 1984. The recovery of reef-building communities and the role of cnidarians
in carbonate sequences of the Middle Triassic (Anisian) in the Italian Dolomites.
Paleontolographica Americana 54:191–200.
Frakes, L. A., J. E. Francis, and J. I. Syktus. 1992. Climate modes of the Phanerozoic. Cambridge:
Cambridge University Press.
Frank, T. D., L. P. Birgenheier, I. P. Montañez, C. R. Fielding, and M. C. Rygel. 2008. Late Paleozoic
climate dynamics revealed by comparison of ice-proximal stratigraphic and ice-distal isotopic
records. In Resolving the Late Paleozoic Ice Age in time and space (Special Paper 441), ed. C. R.
Fielding, T. D. Frank, and J. L. Isbell, 331–342. Boulder, Colo.: Geological Society of America.
Ganino, C., and N. T. Arndt. 2009. Climate changes caused by degassing of sediments during the
emplacement of large igneous provinces. Geology 37:323–326.
Garbelli, C., L. Angiolini, and S.-Z. Shen. 2017. Biomineralization and global change: A new
perspective for understanding the end-Permian extinction. Geology 45:19–22.
Gastaldo, R. A., E. Purkyňová, and Z. Šimůnek. 2009. Mega oral perturbation across the Enna
marine zone in the Upper Silesian basin attests to Late Mississippian (Serpukhovian)
deglaciation and climate change. Palaios 24:351–366.
Gastaldo, R. A., E. Purkyňová, Z. Šimůnek, and M. D. Schmitz. 2009. Ecological persistence in the
Late Mississippian (Serpukhovian, Namurian A) mega oral record of the Upper Silesian basin,
Czech Republic. Palaios 24:336–350.
Gastaldo, R. A., I. M. Stevanović-Walls, W. N. Ware, and S. F. Greb. 2004. Community heterogeneity of
Early Pennsylvanian peat mires. Geology 32:693–696.
Georgiev, S., H. J. Stein, J. L. Hannah, B. Bingen, H. M. Weiss, and S. Piasecki. 2011. Hot acidic Late
Permian seas sti e life in record time. Earth and Planetary Science Letters 310: 389–400.
Giles, P. S. 2009. Orbital forcing and Mississippian sea level change: Time series analysis of marine
ooding events in the Visean Windsor Group of eastern Canada and implications for Gondwana
glaciation. Bulletin of Canadian Petroleum Geology 57:449–471.
González-Bonorino, G., and N. Eyles. 1995. Inverse relation between ice extent and the late
Paleozoic glacial record of Gondwana. Geology 23:1015–1018.
Gradstein, F. M., J. G. Ogg, M. Schmitz, and G. Ogg. 2012. The geologic time scale 2012. Amsterdam:
Elsevier B. V.
Gradstein, F. M., J. G. Ogg, and A. G. Smith. 2004. A geologic time scale 2004. Cambridge: Cambridge
University Press.
Graham, J. B., N. M. Aguliar, R. Dudley, and C. Gans. 1997. The Late Paleozoic atmosphere and the
ecological and evolutionary physiology of tetrapods. In Amniote origins: Completing the transition
to land, ed. S. S. Sumida and K. L. M. Martin, 141–167. San Diego: Academic Press.
Graham, J. B., R. Dudley, N. M. Aguliar, and C. Gans. 1995. Implications of the late Palaeozoic oxygen
pulse for physiology and evolution. Nature 375:117–120.
Grahn, Y., and F. Paris. 2011. Emergence, biodiversi cation and extinction of the chitinozoan group.
Geological Magazine 148:226–236.
Grasby, S. E., H. Sanei, and B. Beauchamp. 2011. Catastrophic dispersion of coal y ash into oceans
during the latest Permian extinction. Nature Geoscience 4:104–107.
Grasby, S. E., W. Shen, R. Yin, J. D. Gleason, J. B. Blum, R. F. Lepak, J. P. Hurley, and B. Beauchamp.
2017. Isotope signatures of mercury contamination in latest Permian oceans. Geology 45:55–58.
Greb, S. F., W. M. Andrews, C. F. Eble, W. DiMichele, C. B. Cecil, and J. C. Hower. 2003. Desmoinesian
coal beds of the Eastern Interior and surrounding basins: The largest tropical peat mires in
Earth history. In Extreme depositional environments: Mega end members in geologic time (Special
Paper 370), ed. M. A. Chan and A. W. Archer, 127–150. Boulder, Colo.: Geological Society of
America.
Grimaldi, D., and M. S. Engel. 2005. Evolution of the insects. Cambridge: Cambridge University Press.
Gulbranson, E. L., I. P. Montañez, M. D. Schmitz, C. O. Limarino, J. L. Isbel, S.A. Marenssi, and J. L.
Crowley. 2010. High-precision U-Pb calibration of Carboniferous glaciation and climate history,
Paganzo Group, NW Argentina. Geological Society of America Bulletin 122:1480–1498.
Hallam, A., and P. G. Wignall. 1997. Mass extinctions and their aftermath. Oxford: Oxford University
Press.
Harris, M. T., and P. M. Sheehan. 1998. Early Silurian stratigraphic sequences of the eastern Great
Basin (Utah and Nevada). In Silurian cycles: Linking dynamic stratigraphy with atmospheric and
oceanic changes (New York State Museum Bulletin 491), ed. E. Landing and M. E. Johnson, 51–61.
Albany: New York State Museum.
Harrison, J. F., A. Kaiser, and J. M. VandenBrooks. 2010. Atmospheric oxygen level and the evolution
of insect body size. Proceedings of the Royal Society B277:1937–1946.
Hayward, B. W. 2002. Late Pliocene to Middle Pleistocene extinctions in deep-sea benthic
foraminifera (Stilostomella extinction) in the southwest Paci c. Journal of Foraminiferal Research
32:274–307.
Heckel, P. H. 1991. Lost Branch Formation and revision of upper Desmoinesian stratigraphy along
midcontinent Pennsylvanian outcrop belt. Kansas Geological Survey Geology Series 4:1–68.
Heckel, P. H. 2008. Pennsylvanian cyclothems in Midcontinent North America as far- eld e ects of
waxing and waning of Gondwana ice sheets. In Resolving the Late Paleozoic Ice Age in time and
space (Special Paper 441), ed. C. R. Fielding, T. D. Frank, and J. L. Isbell, 275–289. Boulder, Colo.:
Geological Society of America.
Hillenius, W. J. 1992. The evolution of nasal turbinates and mammalian endothermy. Paleobiology
18:17–29.
Hillenius, W. J. 1994. Turbinates in therapsids: Evidence for late Permian origins of mammalian
endothermy. Evolution 48:207–229.
Horton, D. E., C. J. Poulsen, I. P. Montañez, and W. A DiMichele. 2012. Eccentricity-paced late
Paleozoic climate change. Palaeogeography, Palaeoclimatology, Palaeoecology 331–332:150–161.
Hoskins, D. M., J. D. Inners, and J. A. Harper. 1983. Fossil collecting in Pennsylvania (General Geology
Report G40). 3d ed. Harrisburg: Pennsylvania Geological Survey.
House, M. R. 1988. Extinction and survival in the Cephalopoda. Systematics Association Special
Volumes 34:139–154.
Irmis, R. B., and J. H. Whiteside 2012. Delayed recovery of non-marine tetrapods after the end-
Permian mass extinction tracks global carbon cycle. Proceedings of the Royal Society B279:1310–
1318.
Irving, E. 2008. Why Earth became so hot 50 million years ago and why it then cooled. Proceedings of
the National Academy of Sciences USA 105:16061–16062.
Isaacson, P. E., E. Diaz-Martinez, G. W. Grader, J. Kalvoda, O. Babek, and F. X. Devuyst. 2008. Late
Devonian–earliest Mississippian glaciation in Gondwanaland and its biogeographic
consequences. Palaeogeography, Palaeoclimatology, Palaeoecology 268:126–142.
Isbell, J. L., A. S. Biakov, I. L. Vedernikov, V. I. Davydov, E. L. Gulbranson, and N. D. Fedorchuk. 2016.
Permian diamictites in northeastern Asia: Their signi cance concerning the bipolarity of the
late Paleozoic ice age. Earth-Science Reviews 154:279–300.
Isbell, J. L., M. F. Miller, K. L. Wolfe, and P. A. Lenaker. 2003. Timing of late Paleozoic glaciation in
Gondwana: Was glaciation responsible for the development of northern hemisphere
cyclothems? In Extreme depositional environments: Mega end members in geologic time (Special Paper
370), ed. M. A. Chan and A. W. Archer, 5–24. Boulder, Colo.: Geological Society of America.
Isozaki, Y., and D. Aljinovic. 2009. End-Guadalupian extinction of the Permian gigantic bivalve
Alatoconchidae: End of gigantism in tropical seas by cooling. Palaeogeography, Palaeoclimatology,
Palaeoecology 284:11–21.
Ivany, L. D., S. Van Simaeys, E. W. Domack, and S. D. Samson. 2006. Evidence for an earliest
Oligocene ice sheet on the Antarctic Peninsula. Geology 34:377–380.
Jablonski, D. 1986. Background and mass extinctions: The alternation of macroevolutionary
regimes. Science 231:129–133.
Jackson, E. L. 1982. The Laki eruption of 1783: Impacts on population and settlement in Iceland.
Geography 67:42–50.
Joachimski, M. M., and W. Buggisch. 2002. Conodont apatite δ18O signatures indicate climate cooling
as a trigger of the Late Devonian mass extinction. Geology 30:711–714.
Kaiser, S. I., M. Aretz, and R. T. Becker. 2016. The global Hangenberg Crisis (Devonian-Carboniferous
transition): Review of a rst-order mass extinction. In Devonian climate, sea level and evolutionary
events (Special Publications 423), ed. R. T. Becker, P. Königshof, and C. E. Brett, 387–437. London:
Geological Society.
Kaiser, S. I., T. Steuber, R. T. Becker, and M. M. Joachimski. 2006. Geochemical evidence for major
environmental change at the Devonian-Carboniferous boundary in the Carnic Alps and the
Rhenish Massif. Palaeogeography, Palaeoclimatology, Palaeoecology 240:146–160.
Kalvoda, J. 1990. Late Devonian-Early Carboniferous paleobiogeography of benthic foraminifera and
climatic oscillations. In Extinction events in Earth history, ed. E. G. Kau man and O. H. Walliser,
183–187. Berlin: Springer Verlag.
Kammer, T. W., and D. L. Matchen. 2008. Evidence for eustacy at the Kinderhookian–Osagean
(Mississippian) boundary in the United States: Response to late Tournaisian glaciation? In
 Resolving the Late Paleozoic Ice Age in time and space (Special Paper 441), ed. C. R. Fielding, T. D.
Frank, and J. L. Isbell, 261–274. Boulder, Colo.: Geological Society of America.
Katz, M. E., K. G. Miller, J. D. Wright, B. S. Wade, J. V. Browning, B. S. Cramer, and Y. Rosenthal. 2008.
Stepwise transition from the Eocene greenhouse to the Oligocene icehouse. Nature Geoscience
1:329–334.
Kemp, D. B., A. L. Coe, A. S. Cohen, and L. Schwark. 2005. Astronomical pacing of methane release in
the Early Jurassic period. Nature 437:396–399.
Kemp, T. S. 2006. The origin and early radiation of the therapsid mammal-like reptiles: A
palaeobiological hypothesis. Journal of Evolutionary Biology 19:1231–1247.
Kenrick, P., and P. R. Crane. 1997a. The Origin and Early Diversi cation of Land Plants: A Cladistic Study.
Washington, D.C.: Smithsonian Institution Press.
Kenrick, P., and P. R. Crane. 1997b. The origin and early evolution of plants on land. Nature 389:33–
39.
Kent, D. V., and G. Muttoni. 2008. Equatorial convergence of India and early Cenozoic climate
trends. Proceedings of the National Academy of Sciences USA 108:21146–21151.
Kiessling, W., and C. Simpson. 2011. On the potential for ocean acidi cation to be a general cause of
ancient reef crises. Global Change Biology 17:56–67.
Knoll, A. H., R. K. Bambach, D. E. Can eld, and J. P. Grotzinger. 1996. Comparative Earth history and
the Late Permian mass extinction. Science 273:452–457.
Knoll, A. H., R. K. Bambach, J. L. Payne, S. Pruss, and W. W. Fischer. 2007. Paleophysiology and end-
Permian mass extinction. Earth and Planetary Science Letters 256:295–313.
Kopp, R. E., J. L. Kirschvink, I. A. Hilburn, and C. Z. Nash. 2005. The Paleoproterozoic snowball Earth:
A climate disaster triggered by the evolution of oxygenic photosynthesis. Proceedings of the
National Academy of Sciences USA 102:11131–11136.
Kraus, O., and C. Brauckmann. 2003. Fossil giants and surviving dwarfs: Arthropleurida and
Pselaphognatha (Atelocerata, Diplopoda); characters, phylogenetic relationships and
construction. Verhandlungen des naturwissenschaftlichen Vereins in Hamburg 40:5–50.
Krings, M., H. Kerp, T. N. Taylor, and E. L. Taylor. 2003. How Paleozoic vines and lianas got o the
ground: On scrambling and climbing Carboniferous–Early Permian pteridosperms. Botanical
Review 69:204–224.
Kukalová-Peck, J. 1985. Ephemeroid wing venation based upon new gigantic Carboniferous may ies
and basic morphology, phylogeny, and metamorphosis of pterygote insects (Insecta,
Ephemerida). Canadian Journal of Zoology 63:933–955.
Kukalová-Peck, J. 1987. New Carboniferous Diplura, Monura, and Thysanura, the hexapod ground
plan, and the role of thoracic side lobes in the origin of wings (Insecta). Canadian Journal of
Zoology 65:2327–2345.
Lanciki, A., J. Cole-Dai, M. H. Thiemens, and J. Savarino. 2012. Sulfur isotope evidence of little or no
stratospheric impact by the 1783 Laki volcanic eruption. Geophysical Research Letters 39, L01806,
 doi:10.1029/2011GL050075.
Lane, N. 2002. Oxygen: The molecule that made the world. Oxford: Oxford University Press.
Lecointre, G., and H. Le Guyader. 2006. The tree of life: A phylogenetic classi cation. Cambridge, Mass.:
Belknap Press of Harvard University Press.
Lenton, T. M., M. Crouch, M. Johnson, N. Pires, and L. Dolan. 2012. First plants cooled the
Ordovician. Nature Geoscience 5:86–89.
Leonova, T. 2009. Ammonoid evolution in marine ecosystems prior to the Permian-Triassic crisis.
Paleontological Journal 43:858–865.
Levinton, J. S. 1982. Marine ecology. Englewood Cli s, NJ: Prentice-Hall.
Lewis, A. R., D. R. Marchant, A. C. Ashworth, L. Hedenäs, S. R Hemming, J. V. Johnson, M. J. Leng, M.
L. Machlus, A. E. Newton, J. I. Raine, J. K. Willenbring, M. Williams, and P. A. Wolfe. 2008. Mid-
Miocene cooling and the extinction of tundra in continental Antarctica. Proceedings of the
National Academy of Sciences USA 105:10676–10680.
Little, E. L. 1980. The Audubon Society eld guide to North American trees. New York: Knopf.
Long, J. A. 1993. Early-Middle Palaeozoic vertebrate extinction events. In Palaeozoic vertebrate
biostratigraphy and biogeography, ed. J. A. Long, 54–63. London: Belhaven Press.
Lucas, S. G. 2004. A global hiatus in the Middle Permian tetrapod fossil record. Stratigraphy 1:47–64.
Lucas, S. G., and A. B. Heckert. 2001. Olson’s gap: A global hiatus in the record of Middle Permian
tetrapods. Journal of Vertebrate Paleontology 21:75A.
Lutz, R. A., and D. C. Rhoads. 1977. Anaerobosis and theory of growth line formation. Science
198:1222–1227.
Marynowski, L., and P. Filipiak. 2007. Water column euxinia and wild re evidence during the
deposition of the Upper Famennian Hangenberg event horizon from the Holy Cross Mountains
(central Poland). Geological Magazine 144:569–595.
Marynowski, L., P. Filipiak, and M. Zatoń. 2010. Geochemical and palynological study of the Upper
Famennian Dasberg event horizon from the Holy Cross Mountains (central Poland). Geological
Magazine 147:527–550.
McClung, W. S., C. A. Cu ey, K. A. Ericksson, and D. O. Terry Jr. 2016. An incised valley ll and
lowland wedges in the Upper Devonian Foreknobs Formation, central Appalachian Basin:
Implications for Famennian glacioeustacy. Palaeogeography, Palaeoclimatology, Palaeoecology 446:
125–143.
McClung, W. S., K. A. Eriksson, D. O. Terry Jr., and C. A. Cu ey. 2013. Sequence stratigraphic
hierarchy of the Upper Devonian Foreknobs Formation, central Appalachian Basin, USA:
Evidence for transitional greenhouse to icehouse conditions. Palaeogeography, Palaeoclimatology,
Palaeoecology 387:104–125.
McGhee, G. R., Jr. 1988. The Late Devonian extinction event: Evidence for abrupt ecosystem
collapse. Paleobiology 14:250–257.
McGhee, G. R., Jr. 1996. The Late Devonian mass extinction. New York: Columbia University Press.
McGhee, G. R., Jr. 2001. The ‘multiple impacts hypothesis’ for mass extinction: A comparison of the
Late Devonian and the late Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology 176:47–58.
McGhee, G. R., Jr. 2011. Convergent evolution: Limited forms most beautiful. Cambridge, Mass.: MIT
Press.
McGhee, G. R., Jr. 2013. When the invasion of land failed: The legacy of the Devonian extinctions. New York:
Columbia University Press.
McGhee, G. R., Jr. 2014a. The Late Devonian (Frasnian/Famennian) mass extinction: A proposed test
of the glaciation hypothesis. Geological Quarterly 58:263–268.
McGhee, G. R., Jr. 2014b. The search for sedimentary evidence of glaciation during the
Frasnian/Famennian (Late Devonian) biodiversity crisis. SEPM The Sedimentary Record
12(June):4–8.
McGhee, G. R., Jr., M. E. Clapham, P. M. Sheehan, D. J. Bottjer, and M. L. Droser. 2013. A new
ecological-severity ranking of major Phanerozoic biodiversity crises. Palaeogeography,
Palaeoclimatology, Palaeoecology 370:260–270.
McGhee, G. R., Jr., P. M. Sheehan, D. J. Bottjer, and M. L. Droser. 2004. Ecological ranking of
Phanerozoic biodiversity crises: Ecological and taxonomic severities are decoupled.
Palaeogeography, Palaeoclimatology, Palaeoecology 211:289–297.
McGhee, G. R., Jr., P. M. Sheehan, D. J. Bottjer, and M. L. Droser. 2012. Ecological ranking of
Phanerozoic biodiversity crises: The Serpukhovian (Early Carboniferous) crisis had a greater
ecological impact than the end-Ordovician. Geology 40:147–150.
McKinney, M. L. 1990. Trends in body-size evolution. In Evolutionary trends, ed. K. J. McNamara, 75–
118. Tucson: University of Arizona Press.
Meave, J., and M. Kellman. 1994. Maintenance of rain forest diversity in riparian forests of tropical
savannas: Implications for species conservation during Pleistocene drought. Journal of
Biogeography 21:121–135.
Melezhik, V. A. 2006. Multiple causes of Earth’s earliest global glaciation. Terra Nova 18:130–137.
Metcalfe, I., J. L. Crowley, R. S. Nicoll, and M. Schmitz. 2015. High-precision U-Pb CA-TIMS
calibration of Middle Permian to Lower Triassic sequences, mass extinction and extreme
climate-change in eastern Australian Gondwana. Gondwana Research 28:61–81.
Mii, H.-S., E. L. Grossman, T. E. Yancey, B. Chuvashov, and A. Egorov. 2001. Isotopic records of
brachiopod shells from the Russian Platform—evidence for the onset of mid-Carboniferous
glaciation. Chemical Geology 175:133–147.
Milne, L., and M. Milne. 1980. The Audubon Society eld guide to North American insects and spiders.
New York: Knopf.
Misof, B., L. Shanlin, K. Meusemann, R. S. Peters, A. Donath, C. Mayer, P. B. Frandsen, J. Ware, T.
Flouri, R. G. Beutel, and 91 other authors. 2014. Phylogenomics resolves the timing and pattern
of insect evolution. Science 346:763–767.
Montañez, I. P., and P. E. Isaacson. 2013. A ‘sedimentary record’ of opportunities. SEPM The
Sedimentary Record 11(January):4–9.
Montañez, I. P., and C. J. Poulsen. 2013. The Late Paleozoic Ice Age: An evolving paradigm. Annual
Review of Earth and Planetary Sciences 41:629–656.
Moore, R. C., C. G. Lalicker, and A. G. Fischer. 1952. Invertebrate fossils. New York: McGraw-Hill.
Morzadec, P. 1992. Evolution des Asteropyginae (Trilobita) et variations eustatiques au Dévonien.
Lethaia 25:85–96.
Moy-Thomas, J. A., and R. S. Miles. 1971. Palaeozoic shes. 2d ed. London: Chapman and Hall.
Moysiuk, J., M. R. Smith, and J.-B. Caron. 2017. Hyoliths are Palaeozoic lophophorates. Nature.
doi:10.1038/nature20804.
Mundil, R., K. R. Ludwig, I. Metcalfe, and P. R. Renne. 2004. Age and timing of the Permian mass
extinction: U/Pb dating of closed-system zircons. Science 305:1760–1763.
Neale, G. 2010. How an Icelandic volcano helped spark the French Revolution. Guardian, 15 April
2010.
Nel, A., G. Fleck, R. Garrouste, G. Gand, J. Lapeyrie, S. M. Bybee, and J. Prokop. 2009. Revision of
Permo-Carboniferous gri en ies (Insecta: Odonatoptera: Meganisoptera) based upon new
species and redescription of selected poorly known taxa from Eurasia. Palaeontographica
Abteilung A 289:89–121.
Nelsen, M. P., W. A. DiMichele, S. E. Peters, and C. K. Boyce. 2016. Delayed fungal evolution did not
cause the Paleozoic peak in coal production. Proceedings of the National Academy of Sciences USA
113:2442–2447.
Newell, N. D. 1949. Phyletic size increase, an important trend illustrated by fossil invertebrates.
Evolution 3:103–124.
Niklas, K. J. 1997. The evolutionary biology of plants. Chicago: University of Chicago Press.
Nutman, A. P., V. C. Bennett, C. R. L. Friend, M. J. Van Kranendonk, and A. R. Chivas. 2016. Rapid
emergence of life shown by discovery of 3,700-million-year-old microbial structures. Nature
537:535–538.
Oliver, W. A., Jr., and A. E. H. Pedder. 1994. Crises in the Devonian history of the rugose corals.
Paleobiology 20:178–190.
Palmer, D., M. Brasier, D. Burnie, C. Cleal, P. Crane, B. A. Thomas, C. Buttler, J. W. C. Cope, R. M.
Owens, J. Anderson, R. Benson, S. Brusatte, J. Clack, K. Bennis-Bryan, C. Du n, D. Hone, Z.
Johanson, A. Milner, D. Naish, K. Parsons, D. Prothero, and X. Xing. 2012. Prehistoric Life. New
York: Dorling Kindersley.
Paris, F., C. Girard, C. Feist, and T. Winchester-Seeto. 1996. Chitinozoan bio-event in the
Frasnian/Famennian boundary beds at La Serre (Montagne Noire, Southern France).
Palaeogeography, Palaeoclimatology, Palaeoecology 121:131–145.
Payne, J. L., and M. E. Clapham. 2012. End-Permian mass extinction in the oceans: An ancient
analog for the twenty- rst century? Annual Review of Earth and Planetary Sciences 40:89–111.
Payne, J. L., A. V. Turchyn, A. Paytan, D. J. DePaola, D. J. Lehrmann, M. Yu, and J. Wei. 2010. Calcium
isotope constraints on the end-Permian mass extinction. Proceedings of the National Academy of
Sciences USA 107:8543–8548.
Pfe erkorn, H. W. 2004. The complexity of mass extinction. Proceedings of the National Academy of
Sciences USA 101:12779–12780.
Pfe erkorn, H. W., R. A. Gastaldo, W. A. DiMichele, and T. L. Phillips. 2008. Pennsylvanian tropical
oras from the United States as a record of changing climate. In Resolving the Late Paleozoic Ice
Age in Time and Space (Special Paper 441), ed. C. R. Fielding, T. D. Frank, and J. L. Isbell, 305–316.
Boulder, Colo.: Geological Society of America.
Phillips, T. L., and W. A. DiMichele. 1992. Comparative ecology and life-history biology of
arborescent lycopsids in Late Carboniferous swamps of North America. Annals of the Missouri
Botanical Garden 79:560–588.
Phillips, T. L., and R. A. Peppers. 1984. Changing patterns of Pennsylvanian coal-swamp vegetation
and implications of climatic control on coal occurrence. International Journal of Coal Geology
3:205–255.
Polozov, A. G., H. H. Svensen, S. Planke, S. N. Grishina, K. E. Fristad, and D. A. Jerram. 2016. The
basalt pipes of the Tunguska Basin (Siberia, Russia): High temperature processes and volatile
degassing into the end-Permian atmosphere. Palaeogeography, Palaeoclimatology, Palaeoecology
441:51–64.
Powell, M. G. 2005. Climatic basis for sluggish macroevolution during the late Paleozoic ice age.
Geology 33:381–384.
Powell, M. G. 2008. Timing and selectivity of the Late Mississippian mass extinction of brachiopod
genera from the central Appalachian basin. Palaios 23:525–534.
Prokop, J., A. Nel, and I. Hoch. 2005. Discovery of the oldest known Pterygota in the Lower
Carboniferous of the Upper Silesian Basin in the Czech Republic (Insecta: Archaeorthoptera).
Geobios 38:383–387.
Prothero, D. R. 1994. The Eocene-Oligocene Transition: Paradise Lost. New York: Columbia University
Press.
Prothero, D. R., L. C. Ivany, and E. A. Nesbitt. 2003. From Greenhouse to Icehouse: The Marine Eocene-
Oligocene Transition. New York: Columbia University Press.
Pu, J. P., S. A. Bowring, J. Ramezani, P. Myrow, T. D. Raub, E. Landing, A. Mills, E. Hodgin, and F. A.
Macdonald. 2016. Dodging snowballs: Geochronology of the Gaskiers glaciation and the rst
appearance of the Ediacaran biota. Geology 44:955–958.
Pusz, A. E., R. C. Thunell, and K. G. Miller. 2011. Deep water temperature, carbonate ion, and ice
volume changes across the Eocene-Oligocene climatic transition. Paleoceanography 26, PA2205,
doi:10.1029/2010PA001950.
Qiao, L., and S.-Z. Shen. 2015. A global review of the Late Mississippian (Carboniferous)
Gigantoproductus (Brachiopoda) faunas and their paleogeographical, paleoecological, and
 paleoclimatic implications. Palaeogeography, Palaeoclimatology, Palaeoecology 420:128–137.
Racki, G. 1998. Frasnian-Famennian biotic crisis: Undervalued tectonic control? Palaeogeography,
Palaeoclimatology, Palaeoecology 141:177–198.
Racki, G., and P. W. Wignall. 2005. Late Permian double-phased mass extinction and volcanism: An
oceanographic perspective. In Understanding Late Devonian and Permian-Triassic Biotic and Climatic
Events, ed. D. J. Over, J. R. Morrow, and P. B. Wignall, 263–297. Amsterdam: Elsevier B. V.
Raup, D. M., and J. J. Sepkoski Jr. 1982. Mass extinctions in the marine fossil record. Science
215:1501–1503.
Raymo, M. E., and W. F. Ruddiman. 1992. Tectonic forcing of late Cenozoic climate. Nature 359:117–
122.
Raymond, A., and C. Metz. 1995. Laurussian land-plant diversity during the Silurian and Devonian:
mass extinction, sampling bias, or both? Paleobiology 21:74–91.
Raymond, A., and C. Metz. 2004. Ice and its consequences: Glaciation in the Late Ordovician, Late
Devonian, Pennsylvanian-Permian, and Cenozoic compared. Journal of Geology 112:655–670.
Reichow, M. K., M. S. Pringle, A. I. Al’Mukhamedov, M. B. Allen, V. L. Andreichev, M. M. Buslov, C. E.
Davies, G. S. Fedoseev, J. G. Fitton, S. Inger, A. Ya. Medvedev, C. Mitchell, V. N. Puchkov, I. Yu.
Safonova, R. A. Scott, and A. D. Saunders. 2009. The timing and extent of the eruption of the
Siberian Traps large igneous province: Implications for the end-Permian environmental crisis.
Earth and Planetary Science Letters 277:9–20.
Retallack, G. J. 1999. Postapocalyptic greenhouse paleoclimate revealed by earliest Triassic
paleosols in the Sydney Basin, Australia. Geological Society of America Bulletin 111:52–70.
Retallack, G. J., and A. H. Jahren. 2008. Methane release from igneous intrusion of coal during Late
Permian extinction events. Journal of Geology 116:1–20.
Retallack, G. J., and E. S. Krull. 2006. Carbon isotope evidence for terminal-Permian methane
outbursts and their role in extinctions of animals, plants, coral reefs, and peat swamps. In
Wetlands through Time (Special Paper 399), ed. S. F. Greb and W. A. DiMichele, 249–268. Boulder,
Colo: Geological Society of America.
Retallack, G. J., C. A. Metzger, T. Greaver, A. H. Jahren, R. M. H. Smith, and N. D. Sheldon. 2006.
Middle-Late Permian mass extinction on land. Geological Society of America Bulletin 118:1398–
1411.
Retallack, G. J., J. J. Veevers, and R. Morante. 1996. Global coal gap between Premian-Triassic
extinction and Middle Triassic recovery of peat-forming plants. Geological Society of America
Bulletin 108:195–207.
Rey, K., R. Amiot, F. Fourel, T. Rigaudier, F. Abdala, M. O. Day, V. Fernandez, F. Fluteau, C. France-
Lanord, B. S. Rubidge, R. M. Smith, P. A. Vigliette, B. Zipfel, and C. Lécuyer. 2016. Global climate
perturbations during the Permo-Triassic mass extinctions recorded by continental tetrapods
from South Africa. Gondwana Research 37:384–396.
Ridgwell, A. 2005. A Mid-Mesozoic Revolution in the regulation of ocean chemistry. Marine Geology
217:339–357.
Robinson, J. M. 1990. Lignin, land plants and fungi: Biological evolution a ecting Phanerozoic
oxygen balance. Geology 15:607–610.
Ross, C. A. 1972. Paleobiological analysis of fusulinacean (Foraminiferida) shell morphology. Journal
of Paleontology 46:719–728.
Rößler, R., Z. Feng, and R. Noll. 2012. The largest calamite and its growth architecture—Arthropitys
bistriata from the Early Permian petri ed forest of Chemnitz. Review of Palaeobotany and
Palynology 185:64–78.
Rubinstein, C. V., P. Gerriene, G. S. de la Puente, R. A. Astini, and P. Steemans. 2010. Early Middle
Ordovician evidence for land plants in Argentina (eastern Gondwana). New Phytologist 188:365–
369.
Ruse, M. 2013. The Gaia Hypothesis: Science on a Pagan Planet. Chicago: University of Chicago Press.
Sage, R. F. 1999. Why C4 photosynthesis? In C4 Plant Biology. ed. R. F. Sage and R. K. Monson. San
Diego, Cal.: Academic Press, pp. 3–16.
Sahney, S. and M. J. Benton. 2008. Recovery from the most profound mass extinction of all time.
Proceedings of the Royal Society B275:759–765.
Sahney, S., M. J. Benton, and H. J. Falcon-Lang. 2010. Rainforest collapse triggered Carboniferous
tetrapod diversi cation in Euramerica. Geology 38:1079–1082.
Sallan, L. C., and M. I. Coates. 2010. End-Devonian extinction and a bottleneck in the early evolution
of modern jawed vertebrates. Proceedings of the National Academy of Sciences USA 107:10131–
10135.
Saltzman, M. R., L. A. González, and K. C. Lohman. 2000. Earliest Carboniferous cooling triggered by
the Antler orogeny? Geology 28:347–350.
Sandberg, C. A., F. G. Poole, and J. G. Johnson. 1988. Upper Devonian of Western United States. In
Devonian of the World. eds. N. J. McMillan, A. F. Embry, and D. J. Glass. Canadian Society of Petroleum
Geology Memoir 14, volume I:184–220.
Saunders, A. D. 2005. Large igneous provinces: Origin and environmental consequences. Elements
1:259–263.
Saunders, W. B., and W. H. C. Ramsbottom. 1986. The mid-Carboniferous eustatic event. Geology
14:208–212.
Schmidt, A., B. Ostro, K. S. Carslaw, M. Wilson, T. Thordarson, G. W. Mann, and A. J. Simmons. 2011.
Excess mortality in Europe following a future Laki-style Icelandic eruption. Proceedings of the
National Academy of Sciences USA 108:15710–15715.
Schmidt, A., T. Thordarson, L. D. Oman, A. Robock, and S. Self. 2012. Climatic impact of the long-
lasting 1783 Laki eruption: Inapplicability of mass-independent sulfur isotopic composition
measurements. Journal of Geophysical Research 117, D23116, doi:10.1029/2012JD018414.
Schwarzacker, W. 1989. Milankovitch type cycles in the Lower Carboniferous of NW Ireland. Terra
Nova 1:468–473.
Scott, A. C. and I. J. Glasspool. 2006. The diversi cation of Paleozoic re systems and uctuations in
atmospheric oxygen concentration. Proceedings of the National Academy of Sciences USA
103:10861–10865.
Self, S., T. Thordarson, and M. Widdowson. 2005. Gas uxes from ood basalt eruptions. Elements
1:283–287.
Sepkoski, J. J., Jr. 1981. A factor analytic description of the Phanerozoic marine fossil record.
Paleobiology 7:36–53.
Sepkoski, J. J., Jr. 1984. A kinematic model of Phanerozoic taxonomic diversity: III. Post-Paleozoic
families and mass extinctions. Paleobiology 10:246–267.
Sepkoski, J. J., Jr. 1990. Evolutionary faunas. In Palaeobiology: A Synthesis. ed. D. E. G. Briggs and P. R.
Crowther. Oxford, England: Blackwell Scienti c, pp. 37–41.
Sepkoski, J. J., Jr. 1996. Patterns of Phanerozoic extinction: A perspective from global data bases. In
Global Events and Event Stratigraphy. ed. O. H. Walliser. Berlin, Germany: Springer-Verlag, pp. 35–
51.
Sepkoski, J. J., Jr. 1998. Rates of speciation in the fossil record. Philosophical Transactions of the Royal
Society of London B353:315–326.
Sepkoski, J. J., Jr., and A. I. Miller. 1985. Evolutionary faunas and the distribution of Paleozoic
marine communities in space and time. In Phanerozoic Diversity Patterns. ed. J. W. Valentine.
Princeton, New Jersey: Princeton University Press, pp. 153–190.
Servais, T., A. W. Owen, D. A. T. Harper, B. Kröger, and A. Munnecke. 2010. The Great Ordovician
Biodiversi cation Event (GOBE): The palaeoecological dimension. Palaeogeography,
Palaeoclimatology, Palaeoecology 294:99–119.
Shaw, S. R. 2014. Planet of the Bugs: Evolution and the Rise of Insects. Chicago, Illinois: niversity of
Chicago Press.
Shear, W. A., and J. Kukalová-Peck. 1990. The ecology of Paleozoic terrestrial arthropods: The fossil
evidence. Canadian Journal of Zoology 68:1807–1834.
Sheehan, P. M. 1996. A new look at Ecologic Evolutionary Units (EEUs). Palaeogeography,
Palaeoclimatology, Palaeoecology 127:21–32.
Sheehan, P. M. 2001. The Late Ordovician mass extinction. Annual Review of Earth and Planetary
Sciences 29:331–364.
Shellnutt, J. G. 2013. The Emeishan large igneous province: A synthesis. Geoscience Frontiers (2013),
http://dx.doi.org/10.1016/j.gsf.2013.07.003.
Shen, S.-Z., J. L. Crowley, Y. Wang, S. A. Bowring, and D. H. Erwin. 2011. Calibrating the end-Permian
mass extinction. Science 334:1367–1372.
Shevenell, A. E., J. P. Kennett, and D. W. Lea. 2004. Middle Miocene Southern Ocean cooling and
Antarctic cryosphere expansion. Science 305:1766–1770.
Smith, L. B. and J. F. Read. 2000. Rapid onset of late Paleozoic glaciation on Gondwana: Evidence
from Upper Mississippian strata of the Midcontinent, United States. Geology 28:279–282.
Smithson, T. R., S. P. Wood, J. E. A. Marshall, and J. A. Clack. 2012. Earliest Carboniferous tetrapod
and arthropod faunas from Scotland populate Romer’s Gap. Proceedings of the National Academy of
Sciences USA 109:4532–4537.
Sobolev, S. V., A. V. Sobolev, D. V. Kuzmin, N. A. Krivolutskaya, A. G. Petrunin, N. T. Arndt, V. A.
Radko, and Y. R. Vasiliev. 2011. Linking mantle plumes, large igneous provinces and
environmental catastrophes. Nature 477:312–316.
Stanley, S. M. 2007. An analysis of the history of marine animal diversity. Paleobiology Memoirs 4:1–
55.
Stanley, S. M., and M. G. Powell. 2003. Depressed rates of origination and extinction during the late
Paleozoic ice age: A new state for the global marine ecosystem. Geology 31:877–880.
Stearn, C. W. 1987. E ect of the Frasnian-Famennian extinction event on the stromatoporoids.
Geology 15:677–679.
Stein, W. E., C. M. Berry, L. VanA. Hernick, and F. Mannolini. 2012. Surprisingly complex community
discovered in the mid-Devonian fossil forest at Gilboa. Nature 483:78–81.
Stein, W. E., F. Mannolini, L. V. Hernick, E. Landing, and C. M. Berry. 2007. Giant cladoxylopsid trees
resolve the enigma of the Earth’s earliest forest stumps at Gilboa. Nature 446:904–907.
Stewart, W. N. 1983. Paleobotany and the evolution of plants. Cambridge: Cambridge University Press.
Steyer, S. 2012. Earth Before the Dinosaurs. Bloomington: Indiana University Press.
Stone, R. 2004. Iceland’s doomsday scenario? Science 306:1278–1281.
Strand, K., S. Passchier, and J. Näsi. 2003. Implications of quartz grain microtextures for onset
Eocene/Oligocene glaciation in Prydz Bay, ODP Site 1166, Antarctica. Palaeogeography,
Palaeoclimatology, Palaeoecology 198:101–111.
Streel, M., M. V. Caputo, S. Loboziak, and J. H. G. Melo. 2000. Late Frasnian-Famennian climates
based on palynomorph analyses and the question of the Late Devonian glaciations. Earth-Science
Reviews 52:121–173.
Streel, M., M. Vanguestaine, A. Pardo-Trujillo, and E. Thomalla. 2000. The Frasnian-Famennian
boundary sections at Hony and Sinsin (Ardenne, Belgium): New interpretation based on
quantitative analysis of palynomorphs, sequence stratigraphy and climatic interpretations.
Geologica Belgica 3:271–283.
Sun, Y., M. M. Joachimski, P. B. Wignall, C. Yan, Y. Chen, H. Jiang, L. Wang, and X. Lai. 2012. Lethally
hot temperatures during the Early Triassic greenhouse. Science 338:366–370.
Svensen, H., S. Planke, A. G. Polozov, N. Schmidbauer, F. Corfu, Y. Y. Podladchikov, and B. Jamtveit.
2009. Siberian gas venting and the end-Permian environmental crisis. Earth and Planetary Science
Letters 277:490–500.
Tasch, P. 1973. Paleobiology of the Invertebrates. New York: John Wiley and Sons, Inc.
Taylor, T. N., and E. L. Taylor. 1993. The Biology and Evolution of Fossil Plants. Englewood Cli s, N.J.:
Prentice Hall.
Thordarson, T., and G. Larsen. 2007. Volcanism in Iceland in historical time: Volcano types,
eruption styles and eruptive history. Journal of Geodynamics 43:118–152.
Thordarson, T., and S. Self. 2003. Atmospheric and environmental e ects of the 1783–1784 Laki
eruption: A review and reassessment. Journal of Geophysical Research 108, D14011,
doi:10.1029/2001JD002042.
Thordarson, T., S. Self, N. Óskarsson, and T. Hulsebosch. 1996. Sulfur, chlorine, and ourine
degassing and atmospheric loading by the 1783–1784 AD Laki (Skaftár Fires) eruption in
Iceland. Bulletin of Volcanology 58:205–225.
Twitchett, R. J. 2007. The Lilliput e ect in the aftermath of the end-Permian extinction event.
Palaeogeography, Palaeoclimatology, Palaeoecology 252:132–144.
Ukstins-Peate, I., and S. E. Bryan. 2008. Re-evaluating plume-induced uplift in the Emeishan large
igneous province. Nature Geoscience 1:625–629.
Ustritsky, V. I. 1973. Permian climate. In The Permian and Triassic Systems and Their Mutual Boundary
(Memoir 2), ed. A. Logan and L. V. Hill, 733–744. Calgary: Canadian Society of Petroleum
Geologists.
Vachard, D., L. Pille, and J. Gaillot. 2010. Palaeozoic foraminifera: Systematics, Palaeoecology, and
responses to global changes. Revue de Micropaleontologie 53:209–254.
Van Valen, L. M. 1984. A resetting of Phanerozoic community evolution. Nature 307:50–52.
Visscher, H., C. V. Looy, M. E. Collinson, H. Brinkhuis, J. H. A. van Konijnenburg-van Cittert, W. M.
Kürschner, and M. A. Sephton. 2004. Environmental mutagenesis during the end-Permian
ecological crisis. Proceedings of the National Academy of Sciences USA 101:12952–12956.
Voigt, S., and M. Ganzelewski. 2010. Toward the origin of amniotes: Diadectomorph and synapsid
footprints from the early Late Carboniferous of Germany. Acta Palaeontologia Polonica 55:57–72.
Wade, B. S., A. J. P. Houben, W. Qualijtaal, S. Schouter, Y. Rosenthal, K. G. Miller, M. E. Katz, J. D.
Wright, and H. Brinkhuis. 2012. Multiproxy record of abrupt sea-surface cooling across the
Eocene-Oligocene transition in the Gulf of Mexico. Geology 40:159–162.
Wade, N. 2006. Before the Dawn: Recovering the Lost History of Our Ancestors. New York: Penguin Press.
Walker, J. D., and J. W. Geissman. 2009. Geologic Time Scale. Geological Society of America, doi:
10.1130/2009.CTS004R2C.
Wang, Z., and A. Chen. 2001. Traces of arborescent lycopsids and dieback of the forest vegetation in
relation to the terminal Permian mass extinction in North China. Review of Palaeobotany and
Palynology 117:217–243.
Wang, X.-D., X.-J. Wang, F. Zhang, and H. Zhang. 2006. Diversity patterns of Carboniferous and
Permian rugose corals in South China. Geological Journal 41:329–343.
Wang, Y., P. M. Sadler, S.-Z. Shen, D. H. Erwin, Y.-C. Zhang, X.-D. Wang, W. Wang, J. L. Crowley, and C.
M. Henderson. 2014. Quantifying the process and abruptness of the end-Permian mass
 extinction. Paleobiology 40:113–129
Ward, P. D. 2006. Out of Thin Air: Dinosaurs, Birds, and Earth’s Ancient Atmosphere. Washington, D.C.:
Joseph Henry.
Ward, P. D., D. R. Montgomery, and R. Smith. 2000. Altered river morphology in South Africa related
to the Permian–Triassic extinction. Science 289:226–229.
Webby, B. D., F. Paris, M. L. Droser, and I. G. Percival. 2004. The Great Ordovician Biodiversi cation
Event. New York: Columbia University Press.
Weems, R. E. 1992. The ‘terminal Triassic catastrophic event’ in perspective: A review of
Carboniferous through Early Jurassic vertebrate extinction patterns. Palaeogeography,
Palaeoclimatology, Palaeoecology 94:1–29.
Wei, F., Y. Gong, and H. Yang. 2012. Biogeography, ecology and extinction of Silurian and Devonian
tentaculitoids. Palaeogeography, Palaeoclimatology, Palaeoecology 358–360:40–50.
Weidlich, O., 2002. Permian reefs re-examined: Extrinsic control mechanisms of gradual and abrupt
changes during 40 my of reef evolution. Geobios 35(Supplement 1):287–294.
Wellman, C. H. 2010. The invasion of the land by plants: When and where? New Phytologist 188:306–
309.
Wellman, C. H., P. L. Osterlo , and U. Mohiuddin. 2003. Fragments of the earliest land plants. Nature
425:282–285.
Westerhold, T., T. Bickert, and U. Röhl. 2005. Middle to late Miocene oxygen isotope stratigraphy of
ODP site 1085 (SE Atlantic): New constraints on Miocene climate variability and sea-level
uctuations. Palaeogeography, Palaeoclimatology, Palaeoecology 217:205–222.
Wignall, P. B. 2001. Large igneous provinces and mass extinctions. Earth-Science Reviews 53:1–33.
Wignall, P., Y. Sun, D. Bond, G. Izon, R. Newton, S. Védrine, M. Widdowson, J. Ali, X. Lai, H. Jiang, H.
Cope, and S. Bottrell. 2009. Volcanism, mass extinction, and carbon isotope uctuations in the
Middle Permian of China. Science 324:1179–1182.
Wignall, P. B., and R. J. Twitchett. 2002. Extent, duration, and nature of the Permian–Triassic
superanoxic event. Geological Society of America Special Paper 356:395–413.
Wilson, D. S., and B. P. Luyendyk. 2009. West Antarctic paleotopography estimated at the Eocene-
Oligocene climate transition. Geophysical Research Letters 36, L16302, doi:10.1029/2009GL039297.
Wooton, R. J., J. Kukalová-Peck, D. J. S. Newman, and J. Muzón. 1998. Smart engineering in the Mid-
Carboniferous: How well could Palaeozoic dragon ies y? Science 282:749–751.
Wright, V. P., and S. D. Vanstone. 2001. Onset of late Palaeozoic glacio-eustacy and the evolving
climates of low latitude areas: A synthesis of current understanding. Journal of the Geological
Society, London 158:579–582.
Young, G. C. 2010. Placoderms (armored sh): Dominant vertebrates of the Devonian Period. Annual
Review of Earth and Planetary Sciences 38:523–550.
Zachos, J. C., J. R. Breza, and S. W. Wise. 1992. Early Oligocene ice-sheet expansion on Antarctica:
Stable isotope and sedimentological evidence from Kerguelen Plateau, southern Indian Ocean.
 Geology 20:569–573.
Zachos, J. C., M. Pagani, L. Sloan, E. Thomas, and K. Billups. 2001. Trends, rhythms, and aberrations
across the Eocene-Oligocene transition in central North America. Nature 445:639–642.
Zhou, M.-F., J. Malpas, X.-Y. Song, P. T. Robinson, M. Sun, A. K. Kennedy, C. M. Lesher, and R. R.
Keays. 2002. A temporal link between the Emeishan large igneous province (SW China) and the
end-Guadalupian mass extinction. Earth and Planetary Science Letters 196:113–122.
Ziegler, W., and H. R. Lane. 1987. Cycles in conodont evolution from Devonian to mid-Carboniferous.
In Palaeobiology of Conodonts, ed. R. J. Aldridge, 147–163. Chichester: Horwood Press.
 Index

Page numbers refer to the print edition but are hyperlinked to the appropriate location in
the e-book.

acidosis kill model, 177, 180–181, 202, 263

Algeo, Thomas, 21–23
Aljinović, Dunja, 181
amniotes, evolution of, 60–62, 146–149, 247–248, 252
anoxia. See hypoxia kill model
Arndt, Nicholas, 184
Averbuch, Olivier, 19

Barham, Milo, 47
Beerling, David, 101, 113–114, 210–211
Benton, Mike, 62, 169–170, 227
Berner, Robert, 93–95, 101, 114, 207
biodiversity crises. See Capitanian (Middle Permian); end-Devonian (Famennian); end-Ordovician
(Hirnantian); end-Permian (Changhsingian); end-Triassic (Rhaetian); Kasimovian; Late
Devonian (Frasnian); Olsen’s; Serpukhovian (Early Carboniferous)
biodiversity crises, ecological ranking of, 8, 10–11, 52, 170, 179, 253
Blieck, Alain, 168
Bond, David, 188
Bottjer, Dave, 10–11, 52, 170
Boyce, Kevin, 100
Brauckmann, Carsten, 134–135

Capitanian (Middle Permian) biodiversity crisis, 8–9, 11, 147, 152, 158, 170, 171–190, 265
Caputo, Mário, 47
carbon-depletion kill model, 105, 111–114
carbon dioxide, drawdown, 3–4, 6, 15–23, 91, 101, 111–114, 241, 252; injection, 180, 184–185, 197–
199, 200, 244–246, 252, 263–264; poisoning. See hypercapnia kill model
carbonic acid, 180, 200, 204, 263
carboniferous trees, cordaitean pines, 69, 84, 203, 209, 237; equisetophyte horsetails, 68, 79–80, 86,
237; licophyte ferns, 68, 81–82, 86, 237; lycophyte scale trees, 57–60, 68, 76–79, 86–87, 102–103,
105–106, 110, 113, 114–115, 203, 209, 237; medullosan seed ferns, 69, 82–83
Carrano, Matthew, 268
Carroll, Robert, 149
Cascales-Miñana, Borja, 236–237
Cenozoic Ice Age, beginning of, 24–28, 262
central Atlantic volcanism, 208
Clack, Jennifer, 148–149
Clapham, Matthew, 10–11, 170, 208–209
Cleal, Christopher, 90, 106, 108, 114, 164, 236–237
coal, x, 7, 24, 34, 86–87, 93, 96–102, 110–111, 191, 196–197, 199, 200, 203, 210, 241–246, 251–252, 263
continental-positioning kill model, 105, 107–108
continental-uplift kill model, 105, 108–109
convergent evolution, of behaviors, 268; of bipedalism, 153; of C4 photosynthesis, 112; of ecological
niches, 216, 222–224, 230–231, 266–269; of gigantism, 123, 133, 252; of gliding forms, 232; of
swimming forms, 216, 228; of trees, 72–73, 82; of wings, 228, 232–233, 248, 259–260
Courtillot, Vincent, 172–173

de Montredon, Mourgue, 178

DiMichele, William, 100, 109–111
dinosaurs, evolution of, 124, 144, 155, 228, 230–231, 233–235, 238–240, 248–249, 265–271
Droser, Mary, 10–11, 52, 170

ecological ranking of biodiversity crises, 8, 10–11, 52, 170, 179, 253

Ehrmann, Werner, 27
Emeishan volcanism, 171–190
end-Cretaceous (Maastrichtian) biodiversity crisis, 8–10, 52
end-Devonian (Famennian) biodiversity crisis, 8–9, 11, 41–42, 95, 249, 253–261, 265
end-Famennian bottleneck, 43–44, 73, 257–261
end-Frasnian bottleneck, 43-44, 257–261
end-Permian (Changhsingian) biodiversity crisis, 8–10, 52, 147, 158, 170, 190–240, 262–271; duration
of, 212–213, 279n26
end-Permian kill models. See acidosis (acid poisoning); hypercapnia (carbon-dioxide poisoning);
hyperthermia (heat death); hypoxia (su ocation); radiation poisoning
end-Ordovician (Hirnantian) biodiversity crisis, 8–11, 52, 55–56, 225
end-Triassic (Rhaetian) biodiversity crisis, 8, 52
endothermy, evolution of, 146–147, 153–156, 248–249
eruption magma volumes, Eldgjá eruption, 176; Emeishan eruption, 175–176, 179, 190, 262–263; Laki
eruption, 176; Siberian eruption, 192, 262–263
evolutionary bottlenecks, cheetahs, 43; humans, 43; end-Famennian, 43–44, 73, 257–261; end-
Frasnian, 43-44, 257–261; end-Permian, 214–240
explosive blast pipes, Siberian, 193–197, 212
extinctions. See biodiversity crises
Eyles, Nicholas, 50

Falcon-Lang, Howard, 109–111

Famennian Gap, 30, 35, 37, 95, 261, 274n21
fern trees, 21, 68, 81–82, 86, 237
Fielding, Christopher, 23, 48
Filer, Jonathan, 32
Frank, Tracy, 47
Franklin, Benjamin, 178
Frasnian glaciation hypothesis, 28–39, 226, 250, 253, 274n23; predicted glacier size, 34–35

Gaia, 170, 213–214

Ganino, Clément, 184
Gaps, Famennian, 30, 35, 37, 95, 261, 274n21; Olsen’s, 151, 154, 165–166, 168–169; Tournaisian, 30, 35,
41–44, 46, 73, 261, 274n22
Gastaldo, Robert, 57, 86–87
giants, marine invertebrates, 128–133, 141–142, 246; terrestrial arthropods, 117–123, 133–138, 246,
252; terrestrial vertebrates, 123–128, 138–141, 246
gigantism, 24, 117–133; causes of, 133–146
glaciations, causes of, 3–5, 12–23
Glasspool, Ian, 92–96, 137
global warming, x–xi, 162, 164, 169–170, 200–203, 208, 213, 237–238, 244–246, 263
González-Bororino, Gustavo, 50
Graham, Je rey, 146–147, 149–150, 153–155, 246–249
Greb, Stephen, 91, 241
greenhouse gases. See carbon dioxide; methane

Hallam, Tony, 22–23, 51–52, 55

heat death. See hyperthermia kill model
hibernation, 270
horsetail trees, 68, 79–80, 86, 237
“Hot Earth”, x–xi, 200–203, 208, 213, 237–238, 244–246, 263
hydrochloric acid, 177, 198–199, 200, 210, 264
hydro uoric acid, 177
hydrogen sul de, 207
hypercapnia kill model, 180, 202, 203–205, 263
hyperthermia kill model, 200–203, 263–264
hypoxia kill model, 180, 202, 205–210, 263–264

Ice Ages, 1–12; causes of, 12–23

industrial revolution, 242–244, 251–252
insect ight, evolution of, 62–64, 248, 252
interpulse-drying kill model, 105–107
Isaacson, Peter, 32, 34
Isbell, John, 45, 51, 160–161
isorenieratane, 207
Isozaki, Yukio, 181

Jahren, Hope, 186–189

Janis, Christine, 268

Kasimovian biodiversity crisis, 102–116, 150–151

Kasimovian kill models. See carbon depletion; continental positioning; continental uplift;
interpulse drying; refugia constriction; tree blight
Kemp, Tom, 165–168
Kraus, Otto, 134–135
Krings, Michael, 104–105

Lane, Nick, 96–99, 244

large igneous provinces (LIPS), Central Atlantic, 208; Emeishan, 171–190; Siberian, 172, 190–213
Late Devonian (Frasnian) biodiversity crisis, 8–9, 11, 41–42, 52, 95, 170, 226, 250, 253–261, 265
Late Paleozoic Ice Age (LPIA), beginning of, 24, 28–39, 274n23; duration of, 23–24
Lovelock, James, 213
LPIA glaciations, Early Carboniferous, 20, 45–51, 253; Late Carboniferous, 20, 48–49, 51, 88–89; Late
Devonian, 20, 28–39, 250, 253; Permian, 20, 157–162

Mackensen, Andreas, 27
mammals, evolution of, 61, 124, 152, 155–156, 218, 224, 229, 235, 239–240, 248–249, 265–271
mantle plume volcanism, Emeishan, 171–190; Laki, 176–179; Hawaiian, 173–174; Martian, 174–175;
Siberian, 190–213
mass extinctions. See biodiversity crises
McClung, Wilson, 32–35
McGhee, George, 10–11, 28, 33–35, 43, 52, 114–115, 170, 226, 257
McGhee, Marae, v, xi
McKinney, Michael, 140
methane, 15–16, 180, 184, 186–190, 197, 207, 263
methyl bromide, 198, 200, 210, 263–264
Methyl chloride, 198, 200, 210, 263–264
Metz, Cheryl, 51, 56–57, 95
Milanković, Milutin, 14
Milankovitch cycles, 14–15, 32–33, 47, 56, 110
Miller, Arnie, 214
Monoyios, Kalliopi, 45, 259

Nel, André, 138

Newell, Norman, 130–132
Niklas, Karl, 82–83

Olsen’s biodiversity crisis, 147, 151–152, 158, 166, 168–169, 250

Olsen’s Gap, 151, 154, 165–166, 168–169
oxygen, 3–6, 91–96, 114, 205–210, 263; animal evolution, 6, 146–156, 246–249; gigantism and, 133–
146, 246, 252; wild res and, 92–93, 96–98
ozone-layer destruction, 210–213, 264

paleozoic marine world, 214–227

paleozoic terrestrial world, 227–240
Payne, Jonathan, 197, 203, 208–209
Pfe erkorn, Hermann, 102–109
pine trees, 69, 84, 203, 209, 237
Polozov, Alexander, 212
Powell, Matthew, 50, 53–55
Prokop, Jakub, 63

radiation-poisoning kill model, 202, 210–213, 263–264

rainforests, Carboniferous, 57–60, 67–69, 74–92, 100–101, 241; modern, 67, 97–98, 100–101, 241;
Permian, 158–159, 162–164
Raup, Dave, 12, 52
Raymond, Anne, 51, 56–57, 95
refugia-constriction kill model, 105, 109–111
Retallack, Gregory, 186–189, 210
Rey, Kévin, 202

Sahney, Sarda, 150–151, 169–170

scale trees, 57–60, 68, 76–79, 86–87, 102–103, 105–106, 110, 113, 114–115, 203, 209, 237
Scott, Andrew, 92–96, 137
seed-fern trees, 69, 82–83, 86
Sepkoski, Jack, 12, 52, 53, 214–216, 226–227, 236, 265, 268
Serpukhovian (Early Carboniferous) biodiversity crisis, 8–9, 51–65, 169–170, 226, 249, 253, 265
Shaw, Scott, 99
Sheehan, Peter, 10–11, 52, 170
Shiva, 214
Siberian volcanism, 190–213
Snowball Earths, 3–5, 16
Sobolev, Stephen, 193, 199, 264
solar radiation, 12–15
Stanley, Steve, 12, 50, 53–54
steam engines, 242–244, 251
strontium, 19
su ocation. See hypoxia kill model
sulfur dioxide, 177, 185, 198, 200, 263
sulfuric acid, 177, 180, 185, 200, 204
Sun, Yadong, 201–203, 237
Svensen, Henrik, 195, 197–198

therapsid-evolution ecological model, 165–168

Thomas, Barry, 106, 108, 164
Tournaisian Gap, 30, 35, 41–44, 46, 73, 261, 274n22
tree-blight kill model, 105, 114–115
Tunguska Basin, 191–197, 212, 262–272

Visscher, Henk, 211–212

volcanic winter, 177–178

Ward, Peter, 111, 113

Wegener, Alfred, 27
Wignall, Paul, 22–23, 52, 55
wild res, 92–93, 96–98
Williams, Mary, 75, 77, 80, 81, 103, 119, 120, 122, 125–128, 232–234, 238, 240

Zachos, James, 27