Professional Documents
Culture Documents
com
Convergent Evolution
www.ebook3000.com
Vienna Series in Theoretical Biology
Gerd B. Müller, Günter P. Wagner, and Werner Callebaut, editors
www.ebook3000.com
Convergent Evolution
www.ebook3000.com
© 2011 Massachusetts Institute of Technology
All rights reserved. No part of this book may be reproduced in any form by any electronic
or mechanical means (including photocopying, recording, or information storage and
retrieval) without permission in writing from the publisher.
This book was set in Syntax and Times Roman by Toppan Best-set Premedia Limited.
Printed and bound in the United States of America.
McGhee, George R.
Convergent evolution : limited forms most beautiful / George R. McGhee Jr.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-262-01642-1 (hardcover : alk. paper)
1. Convergence (Biology) I. Title.
QH373.M34 2011
576.8—dc22
2011007840
10 9 8 7 6 5 4 3 2 1
www.ebook3000.com
For Marae
’S òg a thug mi gràdh dhuit.
www.ebook3000.com
www.ebook3000.com
Contents
Series Foreword ix
Preface: Limited Forms Most Beautiful xi
2 Convergent Animals 13
Locomotion 13
Carnivores: Prey Detection 30
Carnivores: Prey Capture 42
Herbivores: Plant Processing 50
Defense: Antipredator Adaptations 60
Organ Systems 66
Reproduction 81
3 Convergent Plants 93
Arborescence 93
Photoautotrophs: Light Capture 97
Photoautotrophs: Carbon Processing 101
Defense: Antiherbivore Adaptations 105
Defense: Antidehydration Adaptations 112
Reproduction 115
www.ebook3000.com
viii Contents
www.ebook3000.com
Series Foreword
astonished to see that the eyes of the deadly pit viper, gazing coldly
at me through the glass enclosure of its pen, were identical to the eyes
of my beloved black cat back at home. Both animals had eyes with very
odd-looking vertical-slit pupils and irises of a beautiful gold color. How
could it be that this deadly snake, a cold-blooded scaly reptile so vastly
different from a warm, furry mammal, had eyes like a cat? In my later
university studies I began to realize that the number of evolutionary
possibilities was not endless, and that there was a very good reason
why the eyes of a predatory snake and a predatory mammal are so
similar. By the end of this book, I hope the reader will also share my
fascination with an evolutionary process that produces limited forms
most beautiful.
I thank the Konrad Lorenz Institute for Evolution and Cognition
Research, Altenberg, Austria, for the fellowship that enabled me to work
at the institute in 2010 and to bring this book to completion. I thank
Gerd Müller, director of the KLI, for all of his help in preparing this
book for the Vienna Series in Theoretical Biology. I thank all my col-
leagues (in particular Karl Niklas and Doug Erwin) who have discussed
the scientific implications of convergent evolution with me over the
years. I thank Simon Conway Morris, Celia Deane-Drummond, and
Michael Ruse for our discussions of the philosophical implications of
convergent evolution while we attended a conference on the subject at
the Vatican Observatory, Castel Gandolfo, Italy (and I thank Simon for
inviting me to the conference). Needless to say, philosophical opinions
expressed in this book are my own. Finally, I thank my wife, Marae, for
her patient love.
1 What Is Convergent Evolution?
A porpoise looks like a fish. It has a fusiform, streamlined body like that
of a swordfish or a tuna. It has four fins on the ventral side of its body,
instead of four legs. It has a large fin at its posterior end, instead of a tail.
And it even has a vertical fin centered on its back, so it looks very much
like a shark when it is swimming through the water toward you.
Astonishingly, all appearances to the contrary, a porpoise is not a fish;
it is a mammal. It possesses all the distinctive combination of mammalian
traits: a porpoise is placental, gives live birth, nurses its young with spe-
cialized mammary glands, has an endothermic metabolism, has three
bones in its inner ears, has mammalian milk teeth, reproduces via inter-
nal fertilization, and so on. But it has lost other traits that are found in
most mammals: it has no legs, no tail, no fur, and has instead evolved fins
like those of a fish in place of legs and a tail.
The porpoise inherits its mammalian traits from its mammalian
ancestors, and the possession of these traits indicates to us that the por-
poise belongs to the evolutionary lineage, or clade, of the Mammalia.
These traits are synapomorphies, derived traits that are inherited from
a common ancestor and that define membership within a particular
clade.
The fins of the porpoise are not directly inherited from fish ancestors;
they are independently derived convergent traits. That is, porpoises
have evolved fins that have converged on the morphology of the fins that
2 Chapter 1
are seen in the fish, and, even though the fins of the porpoise look very
much like those of a fish, they are in fact not fish fins but rather are
mammal fins.
Distinguishing synapomorphic traits from convergent traits is critical
in the recognition of convergent evolution. In the literature (particularly
the older literature), synapomorphic traits are sometimes called second-
ary homologous traits, and convergent traits are called homoplastic traits
(Lecointre and Le Guyader 2006). Thus, the critical determination, for
the purpose of recognizing convergent evolution, is sometimes referred
to as the determination of secondary homology versus homoplasy (a
term that I have always found to be most uneuphonious—it sounds like
some type of disease).
There are actually three ways in which homoplastic traits—traits that
look similar but that are not inherited from a common ancestor—might
arise in evolution: convergence, parallelism, and reversion. Let us con-
sider a hypothetical example in which we know the evolutionary rela-
tionships between six species, illustrated by the cladogram given in figure
1.1. Species 1, 2, and 3 all possess synapomorphy S, and thus belong to
the monophyletic clade S. Species 4, 5, and 6 all possess synapomorphy
T, and thus belong to the monophyletic clade T. Although clade S and
clade T have diverged in their evolution, they nevertheless evolved from
a common ancestor that evolved the derived trait R, a trait that all six
species still possess by inheritance; thus, trait R is a synapomorphy
for the larger monophyletic clade R, which contains both clade S and
clade T.
So far, so good. Now let us suppose that in clade S, the new trait Z
arises by evolutionary modification of the preexisting trait A, as seen in
species 3 (figure 1.2). However, in clade T a new trait very similar to trait
Figure 1.1
Cladogram of evolutionary relationships between six hypothetical species.
What Is Convergent Evolution? 3
Figure 1.2
Convergent evolution of trait Z in species 3 and 6.
Figure 1.3
Parallel evolution of trait Z in species 3 and 6.
Figure 1.4
Reverse evolution of trait R in species 6 within clade T, whereas all the species of clade S
simply inherit trait R from a common ancestor.
and the exact same molecular construction: DNA codes for RNA, and
RNA codes for the same 20 amino acids that are assembled to produce
proteins, the building blocks of life. This fact leads us to the concept of
“deep homology” and parallelism in evolution: “One of the most impor-
tant, and entirely unanticipated, insights in the past 15 years was the
recognition of an ancient similarity in patterning mechanisms in diverse
organisms, often among structures not thought to be homologous on
morphological or phylogenetic grounds. In 1997, prompted by the
remarkable extent of similarities in genetic regulation between organs
as different as fly wings and tetrapod limbs, we suggested the term
‘deep homology’ to describe the sharing of the genetic regulatory appa-
ratus that is used to build morphologically and phylogenetically dispa-
rate animal features” (Shubin et al. 2009, 818). However, rather than
viewing deep homology as a disproof of convergent evolution, Shubin
et al. (2009) recognize that it is proof of parallel convergent evolution:
“The deep homology of generative processes and cell-type specification
mechanisms in animal development has provided the foundation for
the independent evolution of a great variety of structures . . . common
genetic mechanisms are used to generate diverse adaptations and can
lead to the parallel evolution of novelties” (Shubin et al. 2009, 818; italics
mine).
In actual practice it is sometimes quite difficult to differentiate con-
vergent and parallel evolution. Rokas and Carroll (2008) give unambigu-
ous examples of both types of evolution at the molecular level: both
dolphins and rhinos have independently evolved the amino acid valine
at the same amino acid site. The dolphins, however, modified alanine to
valine at that site, whereas the rhinos modified tyrosine to alanine. The
same amino acid, valine, was independently evolved from two different
amino acid precursors in a process of convergent evolution (the process
illustrated in figure 1.2). In contrast, the filamentous ascomycetes Asper-
gillus clavatus and Aspergillus oryzae have independently evolved glu-
tamic acid at the same amino acid site in parallel, in that both modified
their ancestral amino acid asparagine to glutamic acid at that site (the
process illustrated in figure 1.3).
In other cases the determination is not so easy. The wings of birds and
bats are overwhelmingly used in the literature as examples of convergent
evolution. And indeed these are two very distantly related animals—the
bats belong to the clade Mammalia and the birds to the clade Dinosauria.
We have to go back in time to the Carboniferous, some 340 million years
ago, to find a common ancestor for these two animals. Both have wings,
What Is Convergent Evolution? 7
Table 1.1
The major lineages of life on Earth
Eubacteria
Archaea
Eukarya
– Bikonta
– – Chlorobiota
– – – Embryophyta
– – – – Tracheophyta
– – – – – Lycophyta
– – – – – Euphyllophyta
– – – – – – Moniliformopses
– – – – – – Lignophyta
– – – – – – – Spermatophyta
– – – – – – – – Angiospermae
– – – – – – – – – Euangiosperms
– Unikonta
– – Metazoa
– – – Cnidaria
– – – Bilateria
– – – – Protostomia
– – – – – Lophotrochozoa
– – – – – Ecdysozoa
– – – – Deuterostomia
– – – – – Echinodermata
– – – – – Chordata
– – – – – – Vertebrata
– – – – – – – Chondrichthyes
– – – – – – – Osteichthyes
– – – – – – – – Actinopterygii
– – – – – – – – Sarcopterygii
– – – – – – – – – Tetrapoda
– – – – – – – – – – Batrachomorpha
– – – – – – – – – – Reptiliomorpha
– – – – – – – – – – – Amniota
– – – – – – – – – – – – Sauropsida
– – – – – – – – – – – – – Diapsida
– – – – – – – – – – – – – – Lepidosauromorpha
– – – – – – – – – – – – – – Archosauromorpha
– – – – – – – – – – – – Synapsida
– – – – – – – – – – – – – Therapsida
– – – – – – – – – – – – – – Mammalia
Note: The complete phylogenetic classification of life used in this book is given in the
appendix.
What Is Convergent Evolution? 11
occur at the top of table 1.1 first and arrange subsequent lineages in their
sequence of evolution as outlined in table 1.1. Thus, in the discussion of
the convergent evolution of eyespots in animals given in chapter 2, the
nonbilaterian jellyfish species Leuckartiara octona (lineage Metazoa:
Cnidaria in table 1.1) is listed before the more derived bilaterian rotifer
species Asplanchna brightwelli (lineage Metazoa: Bilateria: Protostomia:
Lophotrochozoa in table 1.1), which itself is listed before the deuteros-
tomous sea cucumber species Synaptula lamperti (lineage Metazoa: Bila-
teria: Deuterostomia: Echinodermata in table 1.1).
I will take the list of nodes for each pair of convergent lineages back
to the point where the lineages diverged in evolution so the reader can
see how distantly related species are that have independently acquired
a convergent trait. Thus, in the previous example, the living sea cucumber
Synaptula lamperti and the rotifer Asplanchna brightwelli both have
eyespots, yet you have to go all the way back to the evolutionary split
between the protostomous and deuterostomous lineages of bilaterian
animals to find a common ancestor for the two species (Bilateria: Proto-
stomia versus Bilateria: Deuterostomia). In order to find a common
ancestor for these bilaterian animals and the cnidarian jellyfish Leuck-
artiara octona, you have to go all the way back to the evolution of
the metazoans themselves, over 600 million years ago (i.e., Metazoa:
Cnidaria versus Metazoa: Bilateria).
Published sources for all the convergent lineages discussed in this
book are given in the references. These have partially come from my own
lists of convergences, amassed over many years, and all the many text-
books and encyclopedia articles on evolution I have read, all of which
contain examples of convergent evolution (in many cases, the same
ones). In addition, three compendia that are exclusively devoted to the
subject of convergent evolution, and that have been particularly helpful
in my analyses, are Conway Morris’s Life’s Solution (2003), Barlow’s Let
There Be Sight! A Celebration of Convergent Evolution (2005), and
Berg’s Nomogenesis, or Evolution Determined by Law (1922), and I
recommend these to the reader as they also contain many additional
examples not included in this volume.
The reader can quickly grasp the structure of my analysis of the
phenomenon of convergent evolution by reading the section headings
within each of the individual chapters given in the table of contents of
the book. Throughout the book I will discuss the functional and devel-
opmental constraints that have resulted in convergent evolution at every
level of life, from the external forms of living organisms down to the very
12 Chapter 1
molecules from which they are constructed, from their ecological roles
in nature to the way in which their minds function. I will also use the
analytical techniques of theoretical morphology in that discussion
(McGhee 1999, 2001, 2007). In theoretical morphology, the consideration
of nonexistent biological form is just as important as that of existent
biological form, and can give us valuable insights into the phenomenon
of convergent evolution.
2 Convergent Animals
In all cases of two very distinct species furnished with apparently the same
anomalous organ, it should be observed that, although the general appearance
and function of the organ may be the same, yet some fundamental difference can
generally be detected. I am inclined to believe that in nearly the same way as
two men have sometimes independently hit on the very same invention, so
natural selection, working for the good of each being and taking advantage of
analogous variations, has sometimes modified in very nearly the same manner
two parts in two organic beings, which owe but little of their structure in common
to inheritance from the same ancestor.
—Darwin (1859, 193–194)
Locomotion
Table 2.1
Convergent evolution of animal swimming morphologies
1 Convergent structure and function: FUSIFORM BODY (missile-shaped form for drag
reduction)
Convergent lineages:
1.1 Great white shark (Vertebrata: Chondrichthyes: Elasmobranchii: Lamnidae;
Carcharodon carcharias)
1.2 Swordfish (Vertebrata: Osteichthyes: Actinopterygii: Teleostei: Xiphiidae; Xiphias
gladius)
1.3 Ichthyosaur (Osteichthyes: Saurcopterygii: Reptiliomorpha: Amniota: Sauropsida:
Diapsida: Ichthyosauria: Ichthyosauridae; Icthyosaurus platyodon †Jurassic)
1.4 Harbor porpoise (Amniota: Synapsida: Therapsida: Mammalia: Eutheria:
Laurasiatheria: Cetartiodactyla: Cetacea: Odontoceti: Phocaenidae; Phocaena
phocaena)
2 Convergent structure and function: EEL-SHAPED BODY (elongated, cylindrical form
for drag reduction)
Convergent lineages:
2.1 Reed fish (Osteichthyes: Actinopterygii: Cladista: Polypteridae; Erpetoichthyes
calabaricus)
2.2 European eel (Actinopterygii: Teleostei: Elopomorpha: Anguillidae; Anguilla
anguilla)
2.3 South American lungfish (Osteichthyes: Sarcopterygii: Dipnoi: Lepidosirenidae;
Lepidosiren paradoxa)
2.4 Mosasaur (Sarcopterygii: Reptiliomopha: Amniota: Sauropsida: Diapsida:
Lepidosauromorpha: Squamata: Scleroglossa: Autarchoglossa: Anguimorpha:
Mosasauridae; Platecarpus ictericus †Cretaceous)
2.5 Geosaur (Diapsida: Archosauromorpha: Crurotarsi: Crocodilia: Metriorhynchidae;
Geosaurus giganteus †Jurassic)
3 Convergent structure and function: PADDLE-FORM APPENDAGES (walking
appendages modified to paddle shapes for paddling/rowing through water)
Convergent lineages:
3.1 Sea scorpions (Bilateria: Protostomia: Ecdysozoa: Arthropoda: Cheliceriformes:
Merostomata: Eurypterida: Eurypteridae; Eurypterus remipes †Silurian)
3.2 Marbled diving beetle (Arthropoda: Mandibulata: Hexapoda: Dytiscidae;
Thermonectes marmoratus)
3.3 Green sea turtle (Bilateria: Deuterostomia: Chordata: Osteichthyes: Sarcopterygii:
Reptiliomorpha: Amniota: Sauropsida: Anapsida: Testudines: Chelonioidae; Chelonia
mydas)
3.4 Plesiosaur (Sauropsida: Diapsida: Lepidosauromorpha: Sauropterygia:
Elasmosauridae; Muraenosaurus leedsii †Jurassic)
3.5 King penguin (Diapsida: Archosauromorpha: Ornithodira: Dinosauria: Saurischia:
Theropoda: Maniraptora: Aves: Sphenisciformes: Spheniscidae; Aptenodytes patagonica)
3.6 African manatee (Amniota: Synapsida: Therapsida: Mammalia: Eutheria: Afrotheria:
Sirenia: Trichechidae; Trichechus senegalensis)
3.7 California sea lion (Eutheria: Laurasiatheria: Carnivora: Caniformia: Otariidae;
Zalophus californianus)
manatees and their kin, and the sea lions and their kin (table 2.1).
Although a manatee and a walrus look very similar, they are very differ-
ent types of mammals. The afrotherian manatee is more closely related
to an elephant, and the laurasiatherian walrus is more closely related to
a wolf, than they are to each other. Their apparent similarity in form is
entirely due to convergent evolution.
The second most commonly cited example of amazing feats in conver-
gent evolution is the evolution of flying in land animals. No fewer than
three separate groups of tetrapods have modified their forelimbs into
wings: the extinct pterosaurs and the living birds and bats (table 2.2). At
first glance, it would seem to be extremely unlikely that an originally
quadrupedal animal, using all four legs for walking, would somehow be
able to evolve into an animal in which the forelimb is now used as a
flying structure. Yet not only has this occurred in the evolution of life, it
has occurred three times independently! The wings of birds, bats, and
pterosaurs have an amazingly similar appearance, which is clearly due to
the functional constraints of locomotion via flying in the thin gaseous
medium of the Earth’s atmosphere. This similarity is deceptive, however,
for closer examination reveals that the wings of the three groups of
animals are constructed differently. The wing of a bat largely consists of
its hand, in which the fingers are vastly elongated and between which is
stretched a webbed membrane of skin. The wing of a pterosaur is similar,
but consists of a single finger (the fourth digit) from which a membrane
of skin is stretched all the way to the side of the animal. The wing of a
bird is even more different, in that the surface of the wing is not con-
structed of skin membranes, but consists of elongated flight feathers
attached to its arm. Still, wings in all three groups are convergent modi-
fications of an original walking structure, the forelimb.
The wings of insects, the fourth major group of animals to evolve
powered flight (table 2.2), are radically different from those of the tet-
rapods. The insects did not modify their forelimbs in the evolution of
wings. Rather, insect wings consist of modified gill branches originally
present only in the larval stages. Originally present on multiple segments
in the abdomens of larvae, these gill structures were expanded in some
body segments while being developmentally suppressed in others, leading
to the four-wing condition seen in primitive adult insects such as mayflies
(Carroll 2006).
In examining the convergent evolution of wings (table 2.2), we begin
to see the effect of a different constraint on the evolution of form in
nature, in addition to that of functional constraint. All three tetrapod
18 Chapter 2
Table 2.2
Convergent evolution of animal flying morphologies
1 Convergent structure and function: WINGS (flat to curved planar structures for
generating lift in active flying)
Convergent lineages:
1.1 Green darner dragonfly (Bilateria: Protostomia: Ecdysozoa: Panarthropoda:
Arthropoda: Mandibulata: Hexapoda: Odonata: Aeschnidae; Anax junius)
1.2 Hairy devil pterosaur (Bilateria: Deuterostomia: Chordata: Osteichthyes:
Sarcopterygii: Reptiliomorpha: Amniota: Sauropsida: Diapsida: Archosauromorpha:
Ornithodira: Pterosauria: Rhamphorhynchidae; Sordes pilosus †Jurassic)
1.3 Great blue heron (Archosauromorpha: Ornithodira: Dinosauria: Saurischia:
Theropoda: Maniraptora: Aves: Neognathae: Neoaves: Ciconiiformes: Ardeidae; Ardea
herodias)
1.4 Mouse-eared bat (Amniota: Synapsida: Therapsida: Mammalia: Eutheria:
Laurasiatheria: Chiroptera: Microchiroptera: Vespertilionidae; Myotis myotis)
2 Convergent structure and function: GLIDER MEMBRANES (expanded skin
membranes stretched between fore and hind limbs for retarding falling speed in passive
flying)
Convergent lineages:
2.1 Flying Mesozoic mammal (Synapsida: Therapsida: Mammalia: Volaticotheria:
Volaticotheridae; Volaticotherium antiquum †Cretaceous)
2.2 Flying possum or sugar glider (Mammalia: Marsupialia: Diprotodontia: Petauridae;
Petaurus breviceps)
2.3 Feathertail glider (Marsupialia: Diprotodontia: Acrobatidae; Acrobatus pygmaeus)
2.4 Greater glider (Marsupialia: Diprotodontia: Pseudocheiridae; Petauroides volans)
2.5 True flying squirrel (Mammalia: Eutheria: Euarchontoglires: Rodentia: Sciuridae;
Glaucomys volans)
2.6 Scaly-tailed flying squirrel (Eutheria: Euarchontoglires: Rodentia: Anomaluridae;
Anomalurus derbianus)
2.7 Flying lemur or colugo (Eutheria: Euarchontoglires: Dermoptera: Cynocephalidae;
Cynocephalus volans)
3 Convergent structure and function: GLIDER BODIES (dorsoventrally flattened,
laterally expanded body for retarding falling speed in passive flying)
Convergent lineages:
3.1 Flying diapsid reptile (Sauropsida: Diapsida: Weigeltisauridae; Coelurosauravus
jaekeli †Permian)
3.2 European flying Mesozoic lizard (Diapsida: Lepidosauromorpha: Kuehneosauridae;
Kuehneosaurus latus †Triassic)
3.3 North American flying Mesozoic lizard (Lepidosauromorpha: Kuehneosauridae;
Icarosaurus siefkeri †Triassic)
3.4 Black-bearded flying lizard (Lepidosauromorpha: Lepidosauria: Squamata: Iguania:
Agamidae; Draco melanopogon)
3.5 Flying paradise snake (Lepidosauria: Squamata: Scleroglossa: Autarchoglossa:
Anguimorpha: Serpentes: Colubridae; Chrysopelea paradisi)
4 Convergent structure and function: GLIDER FEET (expanded skin membranes
stretched between elongated toes in feet for retarding falling speed in passive flying)
Convergent lineages:
4.1 Wallace’s flying tree frog (Tetrapoda: Batrachomorpha: Lissamphibia: Batrachia:
Anura: Rhacophoridae; Rhacophorus nigropalmatus)
4.2 Costa Rican flying tree frog (Lissamphibia: Batrachia: Anura: Hylidae; Agalychnis
spurrelli)
Convergent Animals 19
Table 2.2
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Table 2.3
Convergent evolution of animal walking morphologies
Note: The geological age of extinct species is marked with a †. For data sources, see text.
are much more diverse than nontetrapod. However, one particular form
of tetrapod walking locomotion is curiously quite restricted: that of
bipedalism. The first tetrapods evolved in the Late Devonian, and were
quadrupedal amphibians (nonamniote tetrapods). The plesiomorphic
quadrupedal condition is the norm for almost all living tetrapods; very
few have made the transition to standing up and walking solely on their
hind limbs.
The adaptive significance of bipedalism is still open to debate. Stand-
ing erect on the hind limbs clearly gives an animal the ability to see much
farther in the distance, thus giving it an early-warning capability in preda-
tor detection. Walking only on the hind limbs also gives an animal the
capability to use the forelimbs for functions other than locomotion: both
theropod maniraptoran dinosaurs and primate mammals convergently
24 Chapter 2
Table 2.4
Convergent evolution of animal burrowing morphologies
Table 2.4
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
28 Chapter 2
and stiffened into spadelike digging structures. Last, excavator forms are
also found in the frogs, as in the Guinea shovel-snout frog.
In modern placental mammals, the mole-form has independently
evolved in three very distantly related groups: the laurasiatherian true
talpid moles, so common in Europe and North America; the afrotherian
golden chrysochlorid moles; and the xenarthran armadillos (table 2.4).
In the Oligocene, some 30 million years ago, another group of placental
mammals independently evolved the mole-form, the ancient epoicother-
ids (Rose and Emry 2005). An even more distantly related mole-form
has been convergently evolved by the marsupial mammals, as is seen in
the modern southern marsupial mole.
Last, multiple convergent evolutions of excavator-form morphologies
have taken place in another group of placental mammals, the rodents, as
documented by Nevo (1999). Mole-forms have been convergently
evolved in four groups of myomorph rodents around the world: the voles,
the lesser mole rats, the Siberian zokors, and the Asian bamboo rats. In
South America, two groups of hystricomorph rodents have evolved
mole-forms: the coruros, endemic to Chile, and the tuco-tucos, found far
to the southeast in Uruguay. In Africa, another group of hystricomorph
rodents have evolved mole-forms: the African mole rats. And last, in
North America, the sciuromorph rodents have produced the pocket
gophers.
From the perspective of theoretical morphology, it is interesting to
note that burrowing mammals have never produced penetrator-forms—
there exist no mammalian snake-forms. Yet it is easy to envision
an elongated, furry, snakelike mammal, one that has secondarily lost its
legs. Of all the mammals, the weasels and ferrets—with their elongated
bodies and small legs—are perhaps the best candidates for producing
a hypothetical snake-form mammal. In contrast, the reptiles have never
produced excavator mole-forms; many reptiles excavate (turtles come
to mind) but usually only sporadically and shallowly, as when they
are preparing to hibernate. The burrowing reptiles’ preference for
penetrator-forms and the burrowing mammals’ preference for excavator-
forms again raise the question of the role of functional versus
developmental constraint in explaining the absence of theoretically
possible morphologies in evolution.
In the animals, mode of locomotion is usually independent of how the
animal feeds; that is, whether it is a carnivore or an herbivore. Some
burrowing mammals are carnivores, like talpid moles, while others are
herbivores, like tuco-tucos (Nevo 1999). Some flying birds are carnivores,
30 Chapter 2
like hawks; others are herbivores, like parrots. Similary, some flying bats
are carnivores, like the insectivorous microbats, while others are herbi-
vores, like the frugivorous megabats. In addition to locomotion, the
particular mode of feeding of an animal imposes additional functional
constraints, which again are reflected in subsequent convergent evolu-
tion, as we shall see in the next section of the chapter.
Table 2.5
Convergent evolution of predator eye structures and vision systems
1 Convergent structure and function: VERTICAL-SLIT PUPILS IN EYES (allows [1] full
usage of the diameter of the lens of the eye in bright light as well as in very low light
intensities, with well-focused images in all light intensities; [2] particular detection of
motion in the horizontal plane)
Convergent lineages:
1.1 Pacific angel shark (Vertebrata: Chondrichthyes: Elasmobranchii: Squatinidae;
Squatina californica)
1.2 Whitetip shark (Chondrichthyes: Elasmobranchii: Carcharhinidae; Charcharhinus
longimanus)
1.3 Spotted eagle ray (Chondrichthyes: Elasmobranchii: Myliobatidae; Aetobatus
narinari)
1.4 Brownbelly leaf frog (Vertebrata: Osteichthyes: Sarcopterygii: Tetrapoda:
Batrachomorpha: Lissamphibia: Batrachia: Anura: Hylidae; Phyllomedusa tarsius)
1.5 Tailed frog (Batrachomorpha: Lissamphibia: Batrachia: Anura: Ascaphidae;
Ascaphus truei)
1.6 Mexican burrowing toad (Lissamphibia: Batrachia: Anura: Rhinophrynidae;
Rhinophrynus dorsalis)
1.7 Couch’s spadefoot toad (Lissamphibia: Batrachia: Anura: Pelobatidae; Scaphiopus
couchii)
1.8 Helmeted gecko (Tetrapoda: Reptiliomorpha: Amniota: Sauropsida: Diapsida:
Lepidosauromorpha: Lepidosauria: Squamata: Scleroglossa: Gekkonta: Gekkonidae;
Tarentola chazaliae)
1.9 Burton’s legless lizard (Lepidosauria: Squamata: Gekkonta: Pygopodidae; Lialis
burtonis)
1.10 Granite night lizard (Lepidosauria: Squamata: Scleroglossa: Autarchoglossa:
Scincomorpha: Xantusidae; Xantusia henshawi)
1.11 Indian python (Squamata: Scleroglossa: Autarchoglossa: Anguimorpha: Serpentes:
Boidae; Python molurus)
1.12 Timber rattlesnake (Squamata: Scleroglossa: Autarchoglossa: Anguimorpha:
Serpentes: Viperidae: Crotalinae; Crotalus horridus)
1.13 Nile crocodile (Diapsida: Archosauromorpha: Archosauria: Crurotarsi: Crocodylia:
Crocodylidae; Crocodylus niloticus)
1.14 Black skimmer (Archosauria: Ornithodira: Dinosauria: Saurischia: Theropoda:
Maniraptora: Aves: Charadriiformes: Rhynchopidae; Rhynchops niger)
1.15 Small cats (Amniota: Synapsida: Therapsida: Mammalia: Eutheria: Laurasiatheria:
Carnivora: Feliformia: Felidae; Felis sylvestris catus)
1.16 Red fox (Mammalia: Eutheria: Laurasiatheria: Carnivora: Caniformia: Canidae;
Vulpes vulpes)
1.17 Harp seal (Eutheria: Laurasiatheria: Carnivora: Caniformia: Phocidae; Pagophilus
groenlandicus)
1.18 Slow loris (Eutheria: Euarchontoglires: Primates: Lorisiformes; Nycticebus
coucang)
2 Convergent structure and function: HORIZONTAL-SLIT PUPILS IN EYES (allows
[1] full usage of the diameter of the lens of the eye in bright light as well as in very low
light intensities, with well-focused images in all light intensities; [2] particular detection of
motion in the vertical plane)
Convergent lineages:
2.1 Common octopus (Bilateria: Protostomia: Lophotrochozoa: Mollusca: Cephalopoda:
Coleoidea: Octopodidae; Octopus vulgaris)
Convergent Animals 33
Table 2.5
(continued)
Table 2.5
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
toads. These toads burrow into the ground during the day, and emerge
to hunt at night.
In the lepidosaurian reptiles, the great majority of both lizards and
snakes have eyes with round pupils. But among the lizards, we find
vertical-slit pupils in the helmeted geckos (whose bizarre eyes are unique
in many ways) and in all the nocturnal species of the gecko family. On
opposite sides of the planet, the xantusid night lizards of North America
and many species of the pygopodid legless lizards of Australia, like Bur-
ton’s legless lizard, have eyes with vertical-slit pupils. Among the snakes,
the two deadliest snake groups have convergently evolved eyes with
vertical-slit pupils: the boid constrictors and the venomous pit vipers.
Both of these predator groups have also evolved infrared-sensitive pit
organs, and are able to hunt warm-blooded animals in total darkness (a
trait that we shall consider in more detail in a moment).
The advanced archosaurian reptiles, such as the crocodiles and
alligators, are aquatic carnivores that have independently evolved eyes
with vertical-slit pupils. Since the great majority of the archosaurian
dinosaurs are extinct, we may never know what type of eyes they pos-
sessed. This fact does not prevent the makers of science fiction movies
from usually portraying extinct theropod predators, such as the veloci-
raptors, with eyes with vertical-slit pupils. In this context, it is interesting
to note that only one species of the modern avian dinosaurs has eyes
with vertical-slit pupils: the black skimmer Rhynchops niger (Zusi and
Bridge 1981). The many species of modern owls, which are avian noctur-
nal predators, all have eyes with round pupils. Zusi and Bridge (1981)
Convergent Animals 35
suggest that the wide range of light intensities encountered by the black
skimmer, which seeks prey during both the dark of night and brilliant
daylight, led it to evolve vertical-slit pupils to protect its eyes during
daylight fishing.
In the placental mammals, eyes with vertical-slit pupils are most char-
acteristic of the small cats, such as the European wildcat Felis sylvestris
or the North American bay lynx Lynx rufus. All the large cats that hunt
during daylight, such as the African lion Panthera leo, have eyes with
round pupils. However, eyes with vertical-slit pupils have been conver-
gently evolved within the canid predators as well, even though these
animals are so unlike the cats in many ways. One example is the modern
red fox, Vulpes vulpes. The small foxes are in essence honorary cats,
converging on a felid way of life. Many of the marine carnivores, such as
the harp seal, have also evolved eyes with pupils that are round when
the animal is hunting in the water, but that contract to vertical slits
when in the air. Last, eyes with vertical-slit pupils have independently
evolved within the primates themselves, and are present in the lorisiform
nocturnal carnivore the slow loris, Nycticebus coucang (Malmström and
Kröger 2006).
In summary, the vertical-slit pupil has repeatedly evolved in carnivores
that are nocturnal, hunting at night, or crepuscular, hunting in the dim
light of twilight or just before sunrise, as well as in those that hunt in
environments that normally have low light intensities, as in many aquatic
habitats. Malmström and Kröger (2006) have argued that in animals with
multifocal lenses, the slit-pupil system allows the full usage of the diam-
eter of the lens of the eye in bright light as well as in very low light
intensities, with well-focused images at all light intensities. In contrast,
the round-pupil system shades the peripheral zones of the lens of the
eye as it is constricted, leading to the loss of well-focused images at
wavelengths of light that are normally focused in those regions of the
lens. Zusi and Bridge (1981) report that, in the eyes of small cats, the
slit-pupil system allows a greater reduction in pupil size than is possible
with a round-pupil sphincter system; this is particularly important to
animals that have very large pupils when the pupillary opening is at its
maximum dilation. Thus, for animals with large pupils, the slit-pupil
system offers much better protection to the retina of the eye in bright
light conditions than a round-pupil system.
The catch in the reasoning of these two studies is this: it does not
matter how the slit in the slit-pupil system is oriented in order to obtain
both of these superior-focus and retinal-protection advantages. That is, a
36 Chapter 2
rather than to the visual region. These ancient fishes therefore may have
evolved the capability to “hear” the static of an electric field, rather than
“seeing” the electric field.
Modern electric fishes are predators, and use their electroreceptive
organs to locate moving prey by detecting the electric-field activity asso-
ciated with muscle contractions in the prey organisms. Many are able to
“see” the electric field of their prey even in total darkness, similar to the
ability of the pit viper to “see” the infrared radiation emitted by endo-
thermic prey species in total darkness. For example, the African electric
catfish is a voracious nocturnal predator, but in contrast to most noctur-
nal predators, it has very small eyes (even smaller than those of diurnal
fishes). Instead of using visual detection to hunt its prey, it uses electro-
sensory detection.
Water is an excellent conductor of electricity, and seven groups of
fishes have independently evolved the capability to detect an electric
field in water. But these fishes are not alone: two separate groups of
primitive monotreme mammals—the duckbill platypus and the Austra-
lian echidna—have also evolved the capability of detecting electric fields
(table 2.5). The duckbill platypus is a semiaquatic predator capable of
hunting in total darkness under water (Gregory et al. 1987; Proske and
Gregory 2003); it has thus converged on the same hunting strategy used
by the electric fishes. The Australian echidna is a land-dwelling animal,
however, and electric field intensities in air are very weak compared to
those that can be developed in water. The echidna uses the electrorecep-
tors on its beak to detect an electric field in the moist soil of its rainforest
habitat, produced by moving earthworms (Manger et al. 1997).
Charles Darwin found the convergent evolution of electroreception in
the electric fishes to be so unusual and so improbable that he included
it in On the Origin of Species in his list of difficulties for the theory of
natural selection to explain: “The electric organs offer another and more
serious difficulty; for they occur in only about a dozen fishes, of which
several are widely remote in their affinities. Generally when the same
organ appears in several members of the same class, especially if in
members having very different habits of life, we may attribute its pres-
ence to inheritance from a common ancestor; and its absence in some of
the members to its loss through disuse or natural selection. But if the
electric organs had been inherited from one ancient progenitor thus
provided, we might have expected that all electric fishes would have been
specially related to each other” (Darwin 1859, 193). Today we know that
the convergent distribution of electrosensory organs in fish is much more
40 Chapter 2
Table 2.6
Convergent evolution of predator ear structures and auditory systems
potential prey animals—and have ears that can hear the ultrasonic
echoes, or return waves. Echolocation thus allows these animals not only
to hunt prey animals by ultrasound but to safely locomote without sight.
The most spectacular example of echolocation is that seen in bats flying
after insects in the dim light of an early summer evening (while we can
still see them). Two separate groups of birds have also convergently
evolved ultrasonic hearing and echolocation capabilities: the apodid
swiftlets of the tropical Indo-Pacific and the steatornithid oilbirds of
South America. The edible-nest swiftlets are crepuscular-to-nocturnal
flying insectivores that inhabit dark caves, and thus are ecological equiva-
lents to bats. Yet the swiftlet is an avian dinosaur, whereas the bat is a
placental mammal. The closest common ancestor for the two is found
back in the Carboniferous, over 340 million years ago, yet they have
42 Chapter 2
great cat, Smilodon fatalis. As such, it would appear that the evolution
of saber-tooth predators in the nimravids and felids is an interesting, but
hardly remarkable, case of parallel evolution in the Carnivora. Not so.
During the same interval of time in which felid saber-tooth cats roamed
North America, Europe, and Asia, the marsupial mammals of South
America evolved a saber-tooth predator, Thylacosmilus atrox, that is
almost identical in morphology to Smilodon fatalis (Turner 1997, 136).
Although we have to go back in time 100 million years or so, to the mid-
Cretaceous, to find a common ancestor between the placental and mar-
supial mammals, the two groups converged on the same predator
morphology in the Pliocene and Pleistocene.
Elongated, saber-shaped teeth are an adaptation for producing shear-
ing bites, bites that can tear off whole chunks of prey flesh and that
produce gaping wounds, leading to massive blood loss and shock in the
prey animal. Shearing bites to the neck can rip out the entire neck of the
prey animal below the cervical vertebrae, or a saber-tooth bite can crush
and collapse the windpipe of large prey animals, quickly causing uncon-
sciousness due to oxygen deprivation (Turner 1997, 125). The extinction
of the saber-tooth predators in the Pleistocene appears to be ecologically
linked to the extinction of the large-prey species that formed their chief
source of food.
The modern cats are well known for another highly derived trait:
retractable claws. Most predators, like bears and wolves, have claws that
can cause serious damage to a prey animal, but that are very dull com-
pared to the razor-sharp claws of the cat. The cat’s claws are usually
retracted, and thus do not come into contact with the ground when they
are walking; hence they are not dulled by constant wear and abrasion,
like the claws of the dog. When the cat deploys its claws, its entire paw
expands as the toe digits are extended, and the sickle-shaped claws may
be used for either slashing the prey animal or clinging to it while the cat’s
fang teeth are in use.
Interestingly, retractable claws were first evolved by the dinosaurs, not
by the cats (table 2.7). Both the dromaeosaurids and the troodontids
evolved a retractable claw on the second digits of their hind feet. Like
the cat’s claws, these claws were sickle-shaped and were retracted
when not in use. The dromaeosaurids and troodontids were related mani-
raptoran theropods, but their retractable claws were independently
evolved in parallel and differ in several aspects of their anatomy (Var-
ricchio 1997). A third example of convergent evolution of this trait is
seen in the very plesiomorphic Late Cretaceous bird Rahonavis ostromi,
44 Chapter 2
Table 2.7
Convergent evolution of predator killing structures
1 Convergent structure and function: SABER TEETH (fang teeth elongated and laterally
flattened for shearing bites to produce gaping wounds)
Convergent lineages:
1.1 Marsupial saber-tooth “cat” (Mammalia: Marsupialia: Sparassodonta:
Thylacosmilidae; Thylacosmilus atrox †Pliocene)
1.2 False saber-tooth “cat” (Mammalia: Eutheria: Laurasiatheria: Carnivora: Feliformia:
Nimravidae; Barbourofelis fricki †Miocene)
1.3 True saber-tooth cat (Eutheria: Laurasiatheria: Carnivora: Feliformia: Felidae;
Smilodon fatalis †Pleistocene)
2 Convergent structure and function: RETRACTABLE CLAWS (protection of claws
when not in use, enhances sharpness of sickle claws for ripping prey)
Convergent lineages:
2.1 Dromaeosaurs (Amniota: Sauropsida: Diapsida: Archosauromorpha: Archosauria:
Ornithodira: Dinosauria: Saurischia: Theropoda: Maniraptora: Dromaeosauridae;
Velociraptor mongoliensis †Cretaceous)
2.2 Troodontids (Dinosauria: Saurischia: Theropoda: Maniraptora: Troodontidae;
Saurornithoides mongoliensis †Cretaceous)
2.3 Rahonaves (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves: incertae sedis;
Rahonavis ostromi †Cretaceous)
2.4 Red-legged seriema (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Gruiformes: Cariamae: Cariamidae; Cariama cristata)
2.5 Nimravid “cats” (Amniota: Synapsida: Therapsida: Mammalia: Eutheria:
Laurasiatheria: Carnivora: Feliformia: Nimravidae; Hoplophoneus mentalis †Eocene)
2.6 True cats (Mammalia: Eutheria: Laurasiatheria: Carnivora: Feliformia: Felidae;
Proailurus lemanensis †Oligocene)
2.7 Banded palm civet (Eutheria: Laurasiatheria: Carnivora: Feliformia: Viverridae;
Hemigalus derbyanus)
3 Convergent structure and function: RAPTORIAL BEAK (laterally compressed, hooked
beak for piercing and tearing flesh)
Convergent lineages:
3.1 Common octopus (Bilateria: Protostomia: Lophotrochozoa: Mollusca: Cephalopoda:
Coleoidea: Octopodidae; Octopus vulgaris)
3.2 Snapping turtle (Bilateria: Deuterostomia: Chordata: Osteichthyes: Sarcopterygii:
Reptiliomorpha: Amniota: Sauropsida: Anapsida: Testudines: Chelydridae; Macroclemys
temmincki)
3.3 Peregrine falcon (Sauropsida: Diapsida: Archosauromorpha: Archosauria:
Ornithodira: Dinosauria: Saurischia: Theropoda: Maniraptora: Aves: Falconiformes:
Falconidae; Falco peregrinus)
3.4 Red-tailed hawk (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Accipitriformes: Accipitridae; Buteo jamaicensis)
3.5 Secretary bird (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Accipitriformes: Sagittaridae; Sagittarius serpentarius)
3.6 Osprey (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves: Accipitriformes:
Pandionidae; Pandion haliaetus)
3.7 Turkey vulture (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Ciconiiformes: Cathartidae; Cathartes aura)
3.8 Great horned owl (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Strigiformes: Strigidae; Bubo virginianus)
3.9 Red-legged seriema (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves:
Ralliformes: Cariamae: Cariamidae; Cariama cristata)
Convergent Animals 45
Table 2.7
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
46 Chapter 2
strikingly different from the pattern seen in the troodontids and the
cariamaen birds. On the other hand, the pattern of convergent evolution
in the true cats and the nimravid false cats is identical—both groups
developed fully retractable claws on all of their digits.
Not all predators have teeth like the sharks in the sea and the cats on
the land. The cephalopods, such as the modern octopus, are ancient mol-
luscan predators that grasp their prey with multiple tentacles, and bite
with a beak that is surprisingly similar in form to the raptorial beak seen
in a hawk (table 2.7). But because the cephalopods evolved in the Late
Cambrian, some 340 million years before the evolution of the first avian
dinosaurs in the Late Jurassic, it would be more correct to state that the
hawks have beaks like the octopus rather than vice versa. Raptorial
beaks appear to be a synapomorphy for the entire Cephalopoda, where
the beaks seen in modern coleoid cephalopods are more similar to the
extinct ammonites than to the modern living ectocochleate cephalopod
Nautilus pompilius.
Raptorial beaks have been convergently evolved by gnathostome ver-
tebrates that have secondarily lost an ancient characteristic trait of most
jawed animals—their teeth. The raptorial beak is well developed in many
types of turtles, themselves very ancient sauropsids, as seen in the modern
snapping turtle. But the raptorial beak is most characteristically, and
convergently, developed in a more derived group of sauropsids, the the-
ropod dinosaurs, as seen in the modern hawks (table 2.7).
The accipitriform raptors include a very diverse group of bird species:
hawks, eagles, kites, and Old World vultures (Accipitridae), the secretary
birds (Sagittaridae), and the ospreys (Pandionidae). The hawks, eagles,
and kites hunt on the wing, whereas the secretary birds are largely ter-
restrial hunters. The ospreys are specialized piscivores that catch fish with
the talons of their feet (the single species Pandion haliaetus is considered
to be so different from the other accipitriform birds that it constitutes
the entire family Pandionidae). And the Old World vultures are mostly
scavengers instead of hunters. Although these birds differ greatly in their
morphologies and ecologies, the raptorial beak may be a synapomorphy
for the entire Accipitriformes, rather than three separate convergences
in the three separate families of accipitriforms (table 2.7).
The falconiform raptors, the falcons and caracaras, are more distantly
related to the accipitriforms, and a more substantial argument may be
made for the convergent development of the raptorial beak in this group.
The New World vultures, such as the familiar turkey vulture in the skies
of North America, are ciconiiform birds even more distantly related to
48 Chapter 2
Land plants cover extensive areas of the surface of the Earth; thus,
life would at first glance seem to be much easier for a plant-eating her-
bivore than for a carnivore. That is, plant food is generally abundant,
and plants cannot run away, as hunted prey can. The major problem
for herbivores is that plant food is difficult to digest, and essential nutri-
ents are difficult to obtain in sufficient quantities from plants. For
example, consider many plant seeds and grains. They are rich in nutrients,
but they are also very hard, particularly when dried, and can crack your
teeth if you try to chew them. Yet think of the common domestic chicken,
Gallus gallus, an animal that survives on a diet of seeds and grains but
has no teeth. The chicken has evolved an alternative method of grinding
up hard plant material—it has a gastric mill, or gizzard. The chicken
actually deliberately swallows sharp stone fragments, which are then
held in the muscular walls of its gizzard and are used to grind up seeds
and grains.
Gastric mills have been convergently evolved by an astonishing variety
of animals, from earthworms to mammals (table 2.8). The gizzard of the
domestic chicken is most familiar to us, and some consider it to be a
delicacy. But gizzards are not just found in the galliform birds or just in
herbivorous or granivorous birds. The gaviiform red-throated loon,
Gavia stellata, is more closely related to a penguin than a chicken, yet it
also possesses a gastric mill that it uses to grind up the bones of frogs
and the exoskeletons of crustaceans that form part of its diet. The plesio-
morphic paleognathaean ostrich, Struthio camelus, a modern large flight-
less bird, also possesses a gastric mill that it uses to grind the grasses,
roots, and leaves that it normally eats, plus the occasional insect, lizard,
or small mammal.
Curiously, we even have a fossil record for gastric mills. The sharp
stones located within the gizzard are termed “gastroliths,” or stomach
stones, and they become worn and rounded in a characteristic fashion
with usage. Gastroliths are occasionally found within the skeletal remains
of fossil birds and other animals, indicating that these animals possessed
gastric mills in life. Thus, we know that the Cretaceous maniraptoran
dinosaur Caudipteryx zoui had a gizzard—and it was not a bird, although
closely related (note, however, that Maryańska et al. [2002] consider the
oviraptorosaurians to be secondarily flightless birds, and so the gastric
mill of Caudipteryx zoui may be an avian synapomorphy rather than
convergent within the maniraptorans).
Convergent Animals 51
Table 2.8
Convergent evolution of gastric mills in herbivores and other animals that require
mechanical assistance to digestion
Convergent structure and function: GASTRIC MILLS (muscular gizzards with embedded
stones or teeth for grinding/processing plant material)
Convergent lineages:
1 Common European earthworm (Bilateria: Protostomia: Lophotrochozoa: Annelida:
Oligochaeta: Lumbricidae; Lumbricus terrestris)
2 Great pond snail (Lophotrochozoa: Mollusca: Gastropoda: Pulmonata: Lymnaeidae;
Lymnaea stagnalis)
3 Priapulid worm (Protostomia: Ecdysozoa: Introverta: Priapulida: Priapulidae;
Priapulus caudatus)
4 Hermit crab (Arthropoda: Mandibulata: Malacostraca: Decapoda: Paguridae;
Eupagurus bernhardus)
5 Madagascar hissing cockroach (Arthropoda: Mandibulata: Hexapoda: Blattodea:
Blattellidae; Gromphadorhina portentosa)
6 Striped mullet (Bilateria: Deuterostomia: Chordata: Osteichthyes: Actinopterygii:
Perciformes: Mugilidae; Mugil cephalus)
7 American gizzard shad (Osteichthyes: Actinopterygii: Clupeiformes: Clupeidae;
Dorosoma cepedianum)
8 Nile crocodile (Osteichtyes: Sarcopterygii: Reptiliomorpha: Amniota: Sauropsida:
Diapsida: Archosauromorpha: Crurotarsi: Crocodylia: Crocodylidae; Crocodylus
niloticus)
9 Sauropodomorph dinosaur (Archosauromorpha: Ornithodira: Dinosauria: Saurischia:
Sauropodomorpha: Plateosauridae; Plateosaurus engelhardi †Triassic)
10 Caudipteryx oviraptor (Dinosauria: Saurischia: Theropoda: Maniraptora:
Oviraptorosauria: Caudipterygidae; Caudipteryx zoui †Cretaceous)
11 Ostrich (Dinosauria: Saurischia: Theropoda: Maniraptora: Aves: Paleognathae:
Struthionidae; Struthio camelus)
12 Psittacosaurid dinosaur (Dinosauria: Ornithischia: Cerapoda: Marginocephalia:
Psittacosauridae; Psittacosaurus mongoliensis †Cretaceous)
13 Giant scaly anteater (Amniota: Synapsida: Therapsida: Mammalia: Eutheria:
Laurasiatheria: Pholidota: Manidae; Manis gigantea)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
From the fossil record, we know that gastric mills were convergently
evolved by the second major branch of the saurischian dinosaurs, the
herbivorous sauropodomorphs. Characteristic gastroliths have been
found with the skeletal remains of plesiomorphic prosauropods, like
Plateosaurus engelhardti, of the Triassic and at the very base of the sau-
ropodomorph clade. Gastroliths are later found in the rib cages of giant
herbivorous sauropod dinosaurs, like Seismosaurus, that apparently
swallowed all of their plant food whole and ground it up internally in
their gastric mill (McIntosh 1997). As many as 50 polished gastroliths
have been found in several psittacosaur skeletons, indicating that gastric
mills were convergently evolved in the herbivorous ornithischian dino-
saurs as well (Sereno 1997).
52 Chapter 2
and the gizzard shad is a clupeiform fish, and their gizzards are indepen-
dently evolved.
The alternative to grinding up plant material in an internal gizzard is
to grind it up in the mouth itself. In the modern world, mammals are the
undisputed masters of processing food in the mouth. The characteristic
heterodont dentition of the mammals is a veritable toolbox, in that
mammals have specialized teeth for slicing, piercing, shearing, crushing,
and grinding. Moreover, these specialized teeth meet in a precise arrange-
ment when the jaws of the mammal are closed, an arrangement known
as dental occlusion. Mammals represent an extreme in the development
of dental occlusion, where the complex crowns of the molar teeth in the
upper and lower jaw fit together in a precise mortar-and-pestle fashion,
even to the extent of facilitating tooth wear on the crowns to produce
wear facets resulting from tooth-to-tooth contact (DeMar and Bolt 1981;
Benton 2005, 292). In particular, the mammals possess tribosphenic
molars in which a cusp in the upper-jaw molar, the protocone, fits into
an opposing basin in the lower-jaw molar, the talonid (Benton 2005,
307). These complex molars are capable of both shearing and grinding
occlusal functions, and it was long thought that they were a unique
autapomorphy of the therian clade of extinct and extant marsupial and
placental mammals. However, we now know that an extinct group of
australosphenid mammals, the Ausktribosphenida, convergently evolved
the tribosphenic molar (table 2.9). Luo et al. (2001) argue that the
tribosphenic molar evolved vicariantly in the southern continent of
Gondwana (as seen in Ausktribosphenos nyktos), and in the northern
continent of Laurasia (as seen in marsupials and placentals) during the
Cretaceous. In addition, yet another group of even more distantly related
Jurassic mammals, the Shuotheriidae, evolved a reversed, or “pseudo-
tribosphenic,” molar in which the position of a pseudo-talonid basin is
shifted from the posterior part of the molar to the anterior in order to
receive the pseudo-protocone cusp (Luo et al. 2007). The functional end
result is the same as that for the tribosphenic molar, and Luo et al. (2007)
argue that early mammalian dental evolution may be much more itera-
tive and parallel than currently recognized.
Hunter and Jernvall (1995) point out that many of the more derived
tribosphenic mammals have added an additional cusp, a hypocone, to the
original three cusps found in the upper molar. They argue that the acqui-
sition of the hypocone is a key evolutionary adaptation for herbivory,
and that it has been convergently evolved over twenty separate times
within the tribosphenic mammals (see table 1 in Hunter and Jernvall
54 Chapter 2
Table 2.9
Convergent evolution of teeth and dental systems in herbivores and other animals that
require mechanical assistance to digestion
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Convergent Animals 55
1995). Jernvall (2000) argues that only small developmental changes are
needed to produce major changes in cusp numbers and sizes in mam-
malian teeth, and that this explains the frequent convergent evolution of
new cusps.
Just as the tribosphenic molar in mammals is convergent, dental occlu-
sion itself is not a unique trait of the mammals. In fact, mammals are
only one of four separate groups of cynodont therapsids that conver-
gently evolved complex dental occlusion in the Triassic (table 2.9). The
tritylodontids, such as Kayentatherium wellesi, are the closest relatives to
the most ancient mammal known to us, Adelobasileus cromptoni (Benton
2005, 290), but they independently evolved dental occlusion. The traver-
sodontids are more distantly related, and the diademondontids even
more so, yet they too independently evolved dental occlusion (Rowe
1993; Martinez et al. 1996). Interestingly, all three of these cynodont
groups were secondarily herbivorous, in that most of the cynodonts were
carnivores or omnivores. Thus, in the Triassic, evolution produced four
separate and different cynodont experiments in complex dental occlu-
sion and extensive food processing in the mouth, only one of which
survives to the present day (the mammals).
Dental occlusion is also not a unique trait of the synapsid amniotes.
The other branch of the amniote animals, the reptilian sauropsids, also
evolved independent lineages of animals with at least partial dental
occlusion. This is particularly interesting in that the sauropsids usually
replace their teeth continually throughout life, whereas the mammals
have only two sets of teeth—the early deciduous teeth, and the later set
of permanent teeth. Occlusion is particularly difficult to maintain in a
jaw that contains teeth of differing ages, and hence different shapes and
positions, as is commonly the case in sauropsids.
Even so, the convergent evolution of dental occlusion has been
reported in some extinct archosaurs and primitive sauropsids. Wu et al.
(1995, 678) describe a Cretaceous notosuchian crocodylomorph from
China, Chimaerasuchus paradoxus, that has teeth “very similar to that of
the postcanine teeth of tritylodontid synapsids and represents a particu-
larly striking example of convergent evolution.” This animal was second-
arily herbivorous, as most of the crocodile-like animals were carnivores,
and thus also represents an ecological convergence to the tritylodontid
cynodonts, which were also secondary herbivores. A second, indepen-
dent, evolution of molariform teeth and occlusal precision in Cretaceous
notosuchians is reported from Tanzania for Pakasuchus kapilimai (table
2.9). This species belongs to a separate notosuchian clade than the
56 Chapter 2
Table 2.10
Convergent evolution of chemical digestive structures and systems in herbivores
tylopod camels, guanocos, and their kin, and yet more different from the
stomachs of the hippopotamuses. The colobus monkeys are primates,
very different animals from the cetartiodactyls, yet they too have con-
vergently evolved a stomachal-fermentation system to deal with their
exclusively leaf-eating herbivorous diet. And last, the peculiar hoatzin or
“stink bird” is an avian dinosaur, very far removed from the clade of the
mammals, yet it also has independently evolved a stomachal-fermenting
system (table 2.10).
The cecal-fermentation system also has been convergently evolved by
a phylogenetically diverse group of mammals (table 2.10): euarchonto-
glirean lagomorphs (rabbits, hares, and pikas), afrotherian hyracoids
Convergent Animals 59
Table 2.11
Convergent evolution of parrot and duck beaks
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Many of the same structures that predators use to kill prey can also be
used to defend themselves from other predators. In this section of the
chapter, I will concentrate on the convergent evolution of structures and
Convergent Animals 61
systems that are primarily for defense. Although a few predators have
evolved eyes with horizontal-slit pupils (as we have previously discussed),
eyes of this type are very widespread in herbivores. Most familiar,
perhaps, are the odd-looking eyes of domestic goats, although the same
type of eye is present in sheep, cows, deer, elks, hippopotamuses, and so
on. It is a near universal trait of the cetartiodactyl mammals. However,
horses and their kin also have eyes with horizontal-slit pupils; thus, this
characteristic has clearly evolved independently in the perissodactyl
mammals. A third group of mammals, the afrotherian manatees, have
convergently evolved the horizontal-slit pupil in the eye as well (table
2.12). In all cases, this type of eye is used by herbivores for predator
detection at a distance, in all directions, and thus for predator
avoidance.
A typical herbivore strategy of predator avoidance is simply to run
away—as fast as possible. The evolution of fast running has anatomical
consequences, a functional constraint that leads directly to convergent
evolution. Bakker (1983) has demonstrated that six morphological
Table 2.12
Convergent evolution of herbivore predator-avoidance systems
Note: The geological age of extinct species is marked with a †. For data sources, see text.
62 Chapter 2
transitions occur in the evolution of fast runners from slow, and that
these morphological changes have repeatedly occurred in distantly
related animal groups that convergently evolved fast runners. These
evolutionary changes involve lengthening and attenuation of the distal
bone shafts in the limbs while the proximal bones of the limbs become
shorter, lengthening and attenuation of the metapodial bones in the feet
as the animals locomote first on their toes and then on the tips of their
toes, shortening of the phalangeal bones in the foot, and transformation
of claws into hooves to protect the ends of the toes from shock and
damage while running. Last, the number of hoofed toes in the foot are
reduced, leading in some cases to the animal standing upon a single
hoofed toe. The spindly legs of red deer are a familiar example of this
morphological transition: the upper bone of the hind limb (femur) is
short and surrounded by a compact mass of muscle (which controls the
limb) held next to the body of the deer, the lower limb bone shafts (tibia
and fibula) are elongated away from the deer’s body and contain much
less muscle mass, the foot is enormously elongated into a thin shaft of
very low mass, and the animal stands on the tips of only two hoofed toes.
This form of limb, so common in many ruminant cetartiodactyls, has been
independently evolved in the tylopod cetartiodactyls as well, even to the
extent of reduction of the number of digits in the foot to two (table 2.12).
The more distantly related perissodactyls have not only independently
evolved the same type of running limb but have further reduced the
number of hoofed digits in the foot to one, as is seen in the modern horse.
In the Miocene, a group of distantly related South American meridiun-
gulates convergently evolved limbs like those seen in modern horses. The
ancient one-toed litopterns that ran across the plains of South America
looked like horses, but were not (table 2.12).
Another antipredator adaptation is to acquire armor. An armored
body can be of the passive, tanklike form or the actively offensive, spine
armor form. Very distantly related animals have repeatedly evolved
tanklike body armor: the saurposid turtles, the dinosaurian ankylosaurs,
and the mammalian glyptodonts, armadillos, and pangolins (table 2.13).
When confronted by a predator, these animals retreat within their armor
and passively wait until the predator gives up trying to attack them and
moves on to more vulnerable prey. The morphological convergence
between the Cretaceous ankylosaurid dinosaurs and the Pleistocene
glyptodontid mammals is astonishing—not just in the similarity of the
body armor, but even down to the offensive club located at the ends of
their tails, which these animals used to strike out at a predator while
Convergent Animals 63
Table 2.13
Convergent evolution of body armor in animals
Note: The geological age of extinct species is marked with a †. For data sources, see text.
64 Chapter 2
crouched down under their armor. Both creatures had to live with deadly
predators—the ankylosaurids were preyed upon by the tyrannosaurs,
and the glyptodontids had to deal with saber-tooth cats—and they
evolved the same morphological solution, even though they are vastly
different animals.
Besides passively protecting the bearer, spinose body armor also
presents the danger of injury to the predator. Animals bristling with
sharp spines have repeated evolved in widely separated lineages—
monotreme echidnas, laurasiatherian hedgehogs, and afrotherian
tenrec “hedgehogs”—and twice independently have evolved among the
rodents—the North American porcupines and the African porcupines
(table 2.13).
A third type of body armor is not mechanical but chemical. Animals
with chemical defenses are also some of the prettiest, most colorful, and
most deadly in nature (table 2.13). Their brightly colored bodies are a
visible warning to predators that their body tissues are loaded with
poison. In the oceans, the wildly colorful nudibranch snails not only
manufacture their own poison, such as sesquiterpene toxin (Cimino et
al. 1985), but can store and concentrate poisons found in their environ-
ment. The bizarre puffer fish have deadly tetrodotoxin in their skin and
organs; if eaten, the predator swiftly dies.
On land, diverse arthropods such as pretty orange-and-black monarch
butterflies store and accumulate cardenolide poison from the milkweeds
on which they feed (we shall consider poisonous plants in detail in the
next chapter); blister beetles can burn with cantharidin toxin if eaten;
and brilliant flashing fireflies contain lucibufagin poison—and insecti-
vores avoid them all (Agosta 1996). In the amphibians, the beautifully
colored poison-arrow frogs are deadly, and toxin from their skin has been
used by humans to poison the tips of arrows (hence their name). Fire
salamanders have independently evolved poison glands along the tops
of their backs. Even a few birds have convergently evolved chemical
body armor; the skin and feathers of the hooded pitohui are poisonous
(Agosta 1996), and the crested auklet secretes antiectoparasite toxins
(Douglas et al. 2005), creating its own mosquito repellant.
A last line of antipredator defense is to fight back. In the seas, five
different groups of bony fishes have evolved poisonous stinger spines for
defense (table 2.14). The scorpion fishes are diverse and deadly: some
are very pretty, like the red lionfish that is popular in aquariums even
though it is dangerous; others are nondescript and cryptic, like the stone-
fish Synanceia horrida, which can kill a human if inadvertently stepped
Convergent Animals 65
Table 2.14
Convergent evolution of specialized defensive structures in animals
on. The weever fishes, rabbitfishes, toadfishes, and catfishes have all con-
vergently evolved this line of defense—as has the duckbill platypus on
land, with stinger spurs on its hind feet.
We have previously considered predators that sting; three groups of
arthropods have convergently evolved stingers for defense. These are the
bees, some ants, and some long-horned beetles (table 2.14). The social
bees in particular will defend their nests by stinging, which is an act of
self-sacrifice as the bee dies afterward. Most ants bite to defend their
nests, but a few have independently modified their ovipositors to produce
stingers, in parallel evolution with their relatives the bees. The long-
horned beetle Onychocerus albitarsis has evolved hypodermic antennae;
that is, it uses its antennae as stingers with “a delivery system almost
identical to that found in the stinger of a deadly buthid scorpion” (Berkov
et al. 2008, 257).
Rather than use a stinger, which requires close contact with the preda-
tor, other animals have convergently evolved systems to spray poisonous
chemicals at the predator from a distance (table 2.14). In the mammals,
the New World skunks and Old World polecats are legendary as well as
interesting, in that most mammals do not use chemical defenses. In the
arthropods, the bombardier beetle is most famous because it not only
sprays noxious liquids with quinone (Agosta 1996) but heats the liquid
to boiling before spraying. The common millipede also will spray quinone
if disturbed, whereas the polydesmid millipedes spray deadly hydrogen
cyanide.
Why is chemical defense so rare in terrestrial mammals and birds,
but so common in terrestrial panarthropods and amphibians? From the
perspective of theoretical morphology, one can easily visualize a non-
existent poison-skinned mouse or sparrow, convergent on a poison-
arrow frog, that has evolved a chemical defense to deter a cat or hawk
predator. Yet the poison-arrow frog exists, whereas the poison-skinned
mouse or sparrow does not. Does a developmental constraint exist
for the generation and maintenance of skin poisons in endothermic
animals? Or is the constraint functional, in that other forms of defense,
perhaps less metabolically costly and complicated, are adequate in endo-
thermic animals?
Organ Systems
about the eye itself? In the classic study of the convergent evolution of
the eye, Salvini-Plawen and Mayr (1977) presented evidence for the
definite independent evolution of light-detection organs in 40 separate
animal lineages, and the probable independent origination of light-
detection organs in 20 additional lineages, bringing the total to 60 sepa-
rate convergences. They also document the convergent evolution of the
process of eye evolution itself: from single photoreceptor cells scattered
across the body of an animal, to distinct aggregations of photoreceptor
cells into eyespots, to ocellar pits, to ocellar cups, to simple eyes, and to
complex compound and camera eyes. This same evolutionary sequence
can be observed in multiple independent lineages of animals.
In table 2.15 I have conservatively taken the 40 definite cases of con-
vergent eye evolution given by Salvini-Plawen and Mayr (1977), reana-
lyzed each case in terms of a phylogenetic classification of life (see
appendix), and added additional convergent cases that modern phyloge-
netic analyses have revealed. These additional factors have altered the
reasoning of Salvini-Plawen and Mayr (1977) somewhat; for example,
because Salvini-Plawen and Mayr considered the arthropods to be
descendants of the annelids, they considered the possible homology of
light-detection structures found in both groups of animals. We now know
that the arthropods are ecdysozoan protostomes more closely related to
introvert worms like the priapulids and nematodes, whereas the annelids
are lophotrochozoan protostomes more closely related to molluscs
(Lecointre and Le Guyader 2006), and thus the two are not closely
related. These considerations have resulted in bringing the number of
definite convergences to 49 (table 2.15).
It is clear that eyes and other light-detection organs are ancient inven-
tions of the animals. Simple eyespots, more complex ocellar cups, and
even complex camera eyes can be found in some of the most plesiomor-
phic animals still alive, the most simple of the eumetazoa, the cnidarians.
Complex camera eyes in our own lineage, the Chordata, were developed
over 500 million years ago in the craniate conodonts of the Cambrian
(Purnell 1995).
Light-detection organs have been repeatedly evolved, lost, and
reevolved to varying degrees of complexity throughout the past 600
million years. Thus, we find simple, isolated photoreceptor cells scattered
across the tissues of bilaterian animals as widely divergent as lophotro-
chozoan protostome rock chitons, ecdysozoan protostome earthworms,
and even in the tails of the ammocoete larvae of deuterostome sea
lampreys (table 2.15). These are all highly derived animals compared
68 Chapter 2
Table 2.15
Convergent evolution of eyes and light-detection systems in animals
Table 2.15
(continued)
Table 2.15
(continued)
7 Convergent structure and function: CAMERA EYE (complex eye with a single lens for
image resolving)
Convergent lineages:
7.1 Carybdeid box jellyfish (Metazoa: Cnidaria: Cubozoa: Carybdeidae; Carybdea
marsupialis)
7.2 Alciopod annelid worm (Metazoa: Bilateria: Protostomia: Lophotrochozoa:
Annelida: Polychaeta: Alciopidae; Torrea candida)
7.3 Common octopus (Lophotrochozoa: Mollusca: Cephalopoda: Coleoidea:
Octopodidae: Octopus vulgaris)
7.4 Jumping spider (Protostomia: Ecdysozoa: Arthropoda: Cheliceriformes: Arachnida:
Salticidae; Portia fimbriata)
7.5 Ogre-faced spider (Arthropoda: Cheliceriformes: Arachnida: Deinopidae; Dinopsis
subrufus)
7.6 Dawn conodont (Bilateria: Deuterostomia: Chordata: Myomerozoa: Craniata:
Conodonta: Cordylodontidae; Eoconodontus notchpeakensis †Cambrian)
bristle worms, not just once. Molecular analyses by Oakley and Cun-
ningham (2002) have revealed that the compound eyes of mydocopid
ostracodes have an origin independent from those in other arthropods.
Thus, Salvini-Plawen and Mayr’s (1977) upper estimate of 60 indepen-
dent origins of light-detection organs may ultimately prove to be closer
to the actual mark. Modern analyses do not always increase the number
of convergences discovered, however. Fitzhugh’s (1991) phylogenetic
analysis revealed that the pygidial eyes of the sabellid bristle worms
Fabriciola and Pseudofabriciola were a synapomophy, and not indepen-
dently derived.
It is interesting to note that the complex camera eye has been inde-
pendently evolved in deuterostome chordates, protostome molluscs and
arthropods, and nonbilaterian cnidarians (table 2.15), but that complex
compound eyes have appeared only in the protostomes. Compound eyes
have independently evolved eight different times within the three major
types of protostomes (lophotrochozoans, ecdysozoans, and chaeto-
gnaths), but not outside of the protostome clade, to my knowledge. Is
there a developmental constraint limiting compound eyes to the proto-
stome lineage? This question leads to the consideration of the develop-
ment of the eye itself.
Modern developmental studies have revealed that all animal eyes
contain highly conserved transcription factor Pax-6 gene homologs,
and the ubiquitous presence of the photopigment opsin proteins (Salvini-
Plawen 2008). This discovery has led to the proposal of two different
models for eye evolution: (1) prototype photoreceptor structures and
their developmental genes evolved only once, and further evolution
has elaborated them along independent lines, or (2) photoreceptor
structures evolved independently multiple times, each time co-opting
homologous genes for use in developing those structures (Oakley and
Cunningham 2002, 1429). The first model is, in essence, a model of paral-
lel evolution of the eye, and the second model is a model of convergent
evolution of the eye. Both models thus can be used to argue for func-
tional and developmental constraints in the evolution of the eye—eyes
function only if they are constructed in a limited number of ways, and
there are a limited number of genes available that can be used for their
development.
Both Oakley and Cunningham (2002) and Salvini-Plawen (2008) argue
for the second model. Oakley and Cunningham (2002) point out that the
first model requires a photoreceptor structure to be present in all
common ancestors, whereas their analyses indicate that compound eyes
72 Chapter 2
were not present in ostracode ancestors. Conway Morris (2003) has also
argued for the second model, and points out that in mammals alone the
Pax-6 gene is redeployed multiple times for other functions, as it is
associated with the development of the brain, nose, pituitary gland, pan-
creas, and the gut. The Pax-6 gene is, in essence, equivalent to the homol-
ogous tetrapod limb that has convergently evolved into wings for flying
in three separate groups of vertebrates (table 2.2), and into fins and
paddle-form swimming appendages in seven other groups of vertebrates
(table 2.1).
We have previously considered the convergent evolution of asymmet-
ric positioning and ultrasonic hearing in the ears of carnivores (table 2.6).
But what about the ear itself? As an organ for detecting pressure waves
in air, or for hearing, the ear is an adaptation of land animals. The ances-
tors of land animals, the fish, possess lateral line systems for detecting
pressure waves in water, which is much denser than air, and these lateral
line systems are still present in many aquatic amphibians, the tetrapod
descendants of fish. Experimentation has shown that lateral line systems
are extremely sensitive to minute differences in water pressure (Dijkgraaf
1967), and that fish and aquatic amphibians have well-developed direc-
tional “hearing” under water (Kuroki 1967).
Lateral line systems are not a unique synapomorphy of the verte-
brates, however. Both molluscs and arthropods have evolved species that
have convergently evolved lateral line systems (table 2.16). In the mol-
luscs, many cephalopods have evolved lines of ciliated tissue on their
heads and arms, and these epidermal lines have been shown to function
like the lateral line systems of fish, allowing the cephalopods to have
directional hearing underwater. In addition to the convergent evolution
of the complex camera eye (table 2.15), these epidermal lateral line
systems are “another example of convergent evolution between a sophis-
ticated cephalopod and vertebrate sensory system” (Budelmann and
Bleckmann 1988, 1). In the marine arthropods, some pelagic decapod
crustaceans have evolved very long antennae that are held parallel to
the long axis of the body of the animal on either side, and these structures
also function as lateral line systems for pressure-wave detection (Denton
and Gray 1985).
The thin gaseous mixture of the Earth’s atmosphere has very low pres-
sure compared to underwater habitats; consequently, land tetrapods have
evolved tympanal ear systems in order to hear in air, and have lost the
lateral line system. However, some tetrapods that have secondarily
returned to aquatic habitats have convergently reevolved a lateral line
Convergent Animals 73
Table 2.16
Convergent evolution of ears and sound-detection systems in animals
1 Convergent structure and function: LATERAL LINE SYSTEMS (lateral line row of
pores with nerve endings for detecting minute changes in water pressure, used for
directional hearing in aqueous habitats)
Convergent lineages:
1.1 Shallow-water brief squid (Bilateria: Protostomia: Lophotrochozoa: Mollusca:
Cephalopoda: Coleoidea: Loliginidae; Lolliguncula brevis)
1.2 Penaeid shrimp (Protostomia: Ecdysozoa: Arthropoda: Mandibulata: Malacostraca:
Decapoda: Penaeidae; Acetes sibogae)
1.3 Astrapid jawless fish (Bilateria: Deuterostomia: Chordata: Vertebrata:
Pteraspidomorphi: Astrapida: Astrapidae; Astrapis desiderata †Ordovician)
1.4 Florida manatee (Vertebrata: Osteichthyes: Sarcopterygii: Reptiliomorpha:
Amniota: Synapsida: Therapsida: Mammalia: Eutheria: Afrotheria: Sirenia: Trichechidae;
Trichechus manatus)
2 Convergent structure and function: TYMPANAL EAR SYSTEMS (tympanum-based
hearing system for detecting minute changes in air pressure, used for directional hearing
in terrestrial habitats)
Convergent lineages:
2.1 Field cricket (Bilateria: Protostomia: Ecdysozoa: Arthropoda: Mandibulata:
Hexapoda: Orthoptera: Gryllidae; Gryllus pennsylvanicus)
2.2 Cricket-parasite tachinid fly (Arthropoda: Mandibulata: Hexapoda: Diptera:
Tachinidae; Ormia ochracea)
2.3 Green lacewing (Arthropoda: Mandibulata: Hexapoda: Neuroptera: Chrysopidae;
Chrysopa carnea)
2.4 Grand western cicada (Arthropoda: Mandibulata: Hexapoda: Hemiptera: Cicadidae;
Tibicen dorsata)
2.5 Praying mantis (Arthropoda: Mandibulata: Hexapoda: Mantodea: Mantidae;
Mantis religiosa)
2.6 Tiger beetle (Arthropoda: Mandibulata: Hexapoda: Coleptera: Cincindela marutha)
2.7 Ornate tiger moth (Arthropoda: Mandibulata: Hexapoda: Lepidoptera: Noctuoidea:
Arctiidae; Apantesis ornata)
2.8 Mallow moth (Arthropoda: Mandibulata: Hexapoda: Lepidoptera: Geometroidea:
Geometridae; Larentia clavaria)
2.9 Red postman butterfly (Arthropoda: Mandibulata: Hexapoda: Lepidoptera:
Papilionoidea: Nymphalidae; Heliconius erato)
2.10 Butterfly moth (Arthropoda: Mandibulata: Hexapoda: Lepidoptera: Hedyloidea;
Hedylidae; Macrosoma nigrimacula)
2.11 Balanerpeton amphibian (Bilateria: Deuterostomia: Chordata: Osteichthyes:
Sarcopterygii: Tetrapoda: Batrachomorpha: Dendrerpetonidae; Balanerpeton woodi
†Carboniferous)
2.12 Eastern box turtle (Tetrapoda: Reptiliomorpha: Amniota: Sauropsida: Anapsida:
Testudines: Emydidae; Terrapene carolina)
2.13 Granite night lizard (Diapsida: Lepidosauromorpha: Squamata: Scleroglossa:
Autarchoglossa: Scincomorpha: Xantusidae; Xantusia henshawi)
2.14 Nile crocodile (Diapsida:Archosauromorpha: Crurotarsi: Crocodylia: Crocodylidae;
Crocodylus niloticus)
2.15 Duckbill platypus (Amniota: Synapsida: Therapsida: Mammalia: Monotremata:
Ornithorhynchidae; Ornithorhynchus anatinus)
2.16 Mouse-eared bat (Mammalia: Eutheria: Laurasiatheria: Chiroptera:
Microchiroptera: Vespertilionidae; Myotis myotis)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
74 Chapter 2
ropod Majungatholus atopus reveals that the animal had the avian pul-
monary air-sac system, and that the avian style of breathing evolved very
early in the evolution of the theropod clade of dinosaurs (O’Connor and
Claessens 2005). Now, a more careful examination of other skeletons of
the ornithodires in general reveals that the pterosaurs also had a pulmo-
nary air-sac system and uniflow respiration (Claessens et al. 2009). In
essence, the pterosaurs evolved the uniflow system of breathing 70
million years before the birds did. As the pulmonary air-sac system is
demonstrably absent in dinosauromorphs and in the herrerasaurids (the
basal dinosaurs), the breathing system clearly evolved convergently
twice within the archosaurs—once in the pterosaurs, and once in the
theropod dinosaurs.
It is interesting that the panarthropods have never evolved lungs.
This may well be another example of developmental constraint on the
number of evolutionary possibilities open to a group of animals, and a
limit to convergent evolution. The tracheal system of breathing limits the
size of an animal, and only in the Carboniferous—when oxygen partial
pressure in the atmosphere was higher than at present—did very large
insects evolve, such as meganeurid dragonflies with wingspans of 75
centimeters.
The small size of insects has allowed these animals to converge on a
thermoregulatory system that is usually thought to be possessed only by
the birds and mammals—endothermy. Like all ectotherms, insects use a
variety of behavioral mechanisms to warm up, such as basking in the sun
or shivering. Due to the insects’ small size, the heat produced by their
muscular actions alone is large relative to the volume of their bodies,
and the insects have evolved a series of organ systems to retain their
heat once it is acquired: insulating air sacs around their thoracic muscles,
an insulating air sac between their thorax and their abdomen, a circula-
tory system with countercurrent heat exchange, and external body
insulation, like the scales of moths and the setae fuzz of bees (Heinrich
1993, 1996).
Large ectotherms, like amphibians and reptiles, have a much larger
volume of body tissue to warm up, and a much smaller body surface to
absorb heat by basking. The larger the animal, the more serious the
problem, because volume increases as a cubic function of dimension,
whereas surface area increases only as a square function. One way to
solve this problem would be to somehow increase the surface area of the
body, in order to absorb more heat while basking. At least six different
groups of tetrapods have increased their body surface areas by vastly
78 Chapter 2
elongating the neural spines of their vertebrae, thus creating a large “sail
fin” structure on their backs (table 2.17). Interestingly, both the derived
synapsid pelycosaurs and the ancient batrachomorph amphibians con-
vergently evolved this surface-area-increase structure in the Permian.
Sail fins were convergently evolved twice, at the same time, by the pely-
cosaurs, as the sail fins of the carnivorous sphenacodontids are structur-
ally very different from those of the herbivorous edaphosaurids (Bennett
1996). The fact that the large pelycosaurs evolved sail fins is taken as
evidence that these synapsids were still ectotherms, and that endothermy
had not yet evolved in the synapsid lineage.
Even more interesting, three separate groups of dinosaurs indepen-
dently evolved sail fins (table 2.17). As I shall discuss in a moment, the
dinosaurs have been argued to have been endotherms, so why would they
evolve sail fins like an ectotherm? If the dinosaurs were endotherms, it
may well be that the sail fin structure was primarily used to shed excess
heat and to cool the body down, rather than to absorb heat.
Two modern groups of highly active, energetic animals are known for
their endothermic metabolisms—the birds and the mammals. But when
did endothermy evolve in these two separate lineages? The first bird,
Archaeopteryx lithographica, dates from the Late Jurassic. However, the
discovery of fossil non-avian coelurosaurian theropods with feathers but
no wings indicates that feathers evolved first as body insulation to retain
heat, and that the coelurosaurs were endothermic. Is endothermy thus a
theropod trait? The fact that many small ornithopod ornithischians, such
as the hypsilophodonts, have fibrolamellar bone (an endothermic char-
acteristic) has been used to argue that they also were endothermic. Did
endothermy convergently arise twice in the dinosaurs, once in the the-
ropod saurischians and once in the ornithopod ornithischians, or is endo-
thermy a synapomorphy of the entire clade of dinosaurs? Bakker (1986)
has been the most vocal advocate of the latter position, though others
still dispute the uniform distribution of endothermy within all of the
dinosaurs (see the discussion in Benton 2005, 219–223).
The ornithodiran pterosaurs have also been argued to have been endo-
thermic (see Benton 2005, 226–228). As discussed above, the uniflow
system of breathing using lungs and pulmonary air sacs was indepen-
dently evolved by the pterosaurs and the birds (Claessens et al. 2009),
and the evidence is strong that endothermy also independently evolved
in the pterosaurs and the dinosaurs (table 2.17). Endothermy clearly
arose independently in the synapsid amniote lineage, far removed from
the sauropsid dinosaur and pterosaur lineages. Most date the evolution
Convergent Animals 79
Table 2.17
Convergent evolution of animal thermoregulation systems
Note: The geological age of extinct species is marked with a †. For data sources, see text.
80 Chapter 2
Table 2.18
The alternative classification of metazoans proposed by Schierwater et al. (2009)
Metazoa Metazoa
– Placozoa – Diploblasta
– Demospongiae – – Placozoa
– Hexactinellida – – – Porifera
– Calcarea – – – – Calcarea
– Eumetazoa – – – – Demospongiae
– – Ctenophora – – – – Hexactinellida
– – Cnidaria* – – – Ctenophora
– – Bilateria – – – Cnidaria*
– – – Protostomia – Bilateria*
– – – Deuterostomia – – Protostomia
– – Deuterostomia
Note: The alternative classification of Schierwater et al. (2009) would require the indepen-
dent evolution of nervous systems in the cnidarians and the bilaterians (marked by
asterisks in the table).
Convergent Animals 81
Reproduction
Table 2.19
Convergent evolution of male internal fertilization organs
at least six separate times by amniote animals: four separate times within
the sauropsids, and once within the synapsids (table 2.19). A single penis
is present in anapsid turtles and archosaurian crocodiles but, curiously,
two penises, or “hemipenes,” were evolved by the lepidosaurian lizards
and passed down to their snake descendants. Although the lizards and
snakes have two penises, they only use one at a time in intercourse.
Equally curious, the penis was lost in highly derived archosaurs like the
avian dinosaurs, which reverted back to the use of the cloaca-versus-
cloaca method of fertilization. Only a very few birds, such as the ostrich
and the duck, have secondarily reevolved a penis organ (King 1981; Kelly
2002).
In the nonamniote tetrapods, males of the caecilian amphibians have
an evertable rear part of the cloaca, or phallodeum, that they use to
transfer sperm into the cloaca of the female (Wake 1993). Among the
frogs, species members of only one genus, the tailed frog Ascaphus, have
evolved a similar penislike extension of the cloaca. Of the invertebrate
animals, the snails are unique in that both dioecious and hermaphroditic
species of gastropods have evolved a penis. Some other marine inverte-
brates, such as the male octopus, secondarily use one of their tentacles
84 Chapter 2
Table 2.20
Convergent evolution of animal reproductive systems: viviparity in nonamniote animals
Table 2.20
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
88 Chapter 2
Table 2.21
Convergent evolution of animal reproductive systems: viviparity in amniote animals
Table 2.21
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
seen in the fossil record (Carter 2008) and in particular in the marine
reptiles that convergently returned to the marine realm from the ter-
restrial in the Mesozoic—the ichthyosaurs, plesiosaurs, and mosasaurs
(Lee and Shine 1998; Cheng et al. 2004; Carter 2008).
The numerous incidences of convergent evolution of viviparity within
the squamates are evidence of the adaptive value of this reproductive
mode, or the role of functional constraint in convergent evolution, but
evidence of developmental constraint may be present as well. Thompson
and Speake (2006) argue that the unique development of an intravitel-
line mesoderm and the formation of an isolated yolk sac makes it par-
ticularly easy for squamates to make the developmental shift to viviparity,
which they have done repeatedly (table 2.21). On the other hand, the
highly speciose clades of both the plesiomorphic turtles and highly
derived archosaurs have remained oviparous. Lee and Shine (1998)
suggest that this also may be due to developmental constraint, in that
90 Chapter 2
Table 2.22
Convergent evolution of animal reproductive systems: complex placentae
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Convergent Animals 91
Arborescence
The reverse of the old adage “not seeing the forest for the trees”—the
condition in which people overlook a larger overall pattern because they
are too narrowly focused on individual particulars—is “not seeing the
trees for the forest.” The trees in a forest do vary—an oak tree is differ-
ent from a maple tree—but the apparent similarity of tree form that can
be seen throughout thousands of trees in a forest may give one the
impression that all trees are variants of a single common ancestor. This
is not so. Nine separate groups of plants have independently and con-
vergently evolved the tree form (Niklas 1997; Donoghue 2005; Stein
et al. 2007).
Trees are an adaptation to life on land. Plant growth in a two-
dimensional plane soon leads to crowding and overgrowth, with one
plant shading out another plant in the competition for light from the sun,
the energy source for the survival of the plant. Just as in human cities,
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 construct towering skyscraper buildings or to evolve trees.
In order to grow into the third dimension of height, a major force must
be overcome—gravity. Tree forms all have a central support structure,
the trunk, that rises vertically from the land surface. At some distance
94 Chapter 3
above the ground, branches extend out from the tree trunk in order to
capture as much sunlight as possible for the tree’s survival and, in order
to ensure the survival of the species, to facilitate fertilization and disper-
sal of the tree’s offspring.
The first trees appeared in the Middle Devonian, over 390 million
years ago. Plants began their full-scale invasion of the terrestrial realm
in the Late Devonian and Carboniferous, and huge forests spread across
the landscapes of the Earth. The trees in those forests were very different
from those that constitute the forests of the Earth today. The famous
Gilboa fossil forest in New York State, from the Middle Devonian, gives
us a glimpse of the earliest forests on Earth. Numerous upright trunks
of trees are preserved as sandstone casts, each one broken off about a
meter above the ancient ground surface. The trunks range in size from
about a half meter to a meter in diameter with a wider, bulbous base
with numerous small anchoring roots, revealing that these earliest trees
had no taproot system for anchoring. The morphology of the upper por-
tions of the trees was completely unknown until the spectacular discov-
ery in 2007 of fossils of the complete tree, logs that had fallen over and
were preserved horizontally, in rock outcrops only 13 kilometers east of
the Gilboa forest (Stein et al. 2007). Some of these trees were over 8
meters tall and, most surprisingly, had no laminar leaves or horizontal
branches. Instead, they were topped with a crown resembling a spiky
pipe brush or shaving brush, consisting of numerous digitately branched
offshoots held at an acute angle to the trunk stem. The complete tree
Wattieza (Eospermatopteris) erianus belongs to an extinct group, the
Cladoxylopsida (table 3.1), that is related to modern surviving horsetail
rushes (see the appendix). Stein et al. (2007, 906) note that the cladoxy-
lopsid tree’s “body plan now stands unequivocally as the oldest known
arborescent terrestrial plant form” and that it is “therefore interesting to
see how instantly recognizable and, in a significant sense, ‘modern’ the
tree-like architecture of Wattieza seems to be.”
Both the more plesiomorphic lycophyte club mosses and the
cladoxylopsid-related equisetophyte horsetail rushes independently
evolved towering trees (table 3.1), although the tallest members of these
groups today is around 10 centimeters. In the lycophytes, the ancient
lepidodendrid scale trees constructed their trunks with two outer layers
of cortex beneath an outermost layer of leaf cushions, whereas in the
equisetophytes, the ancient calamite horsetail trees had trunks supported
by internal wedges of wood (Niklas 1997).
Convergent Plants 95
Table 3.1
Convergent evolution of arborescent plant morphologies
Convergent structure and function: TREE TRUNK (vertical support structure to resist
gravity for the elevation and spatial dispersal of photosynthetic and reproductive
substructures)
Convergent lineages:
1 Lepidodendrid scale tree (Tracheophyta: Lycophyta: Lepidodendracea;
Lepidodendron rhodumnense †Carboniferous)
2 Gilboa-forest cladoxylopsid tree (Tracheophyta: Euphyllophyta: Moniliformopses:
Cladoxylopsida: Pseudosporochnaceae; Wattieza (Eospermatopteris) erianus †Middle
Devonian)
3 Calamite horsetail tree (Euphyllophyta: Moniliformopses: Equisetophyta:
Calamitaceae; Calamites cistiiformes †Carboniferous)
4 Psaronis-type tree ferns (Euphyllophyta: Moniliformopses: Filicophyta: Marattiaceae;
Psaronius schopfii †Carboniferous)
5 Tempskya-type tree ferns (Euphyllophyta: Moniliformopses: Filicophyta:
Tempskyaceae; Tempskya dernbachii †Cretaceous)
6 Australian tree fern (Euphyllophyta: Moniliformopses: Filicophyta: Cyatheaceae;
Cyathea cooperi)
7 Archaeopterid-type woody trees (Euphyllophyta: Lignophyta: Archaeopteridales:
Archaeopteridaceae; Archaeopteris hibernica †Late Devonian)
8 Cycad palm (Euphyllophyta: Lignophyta: Spermatophyta: Cycadophyta: Cycadaceae;
Cycas media)
9 Date palm (Euphyllophyta: Lignophyta: Spermatophyta: Angiospermae:
Euangiosperms: Monocotyledons: Aracaceae; Phoenix dactylifera)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Three separate groups of the ferns have convergently evolved the tree
form through time: the psaronis-type tree ferns in the Carboniferous, the
tempskya-type tree ferns in the Cretaceous, and the tree ferns that
survive today (table 3.1). Each group invented a different 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 interweaved 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).
The first of the lignophyte trees that are so abundant today appeared
in the Devonian, particularly the archaeopterid-type woody trees of the
Late Devonian, which still reproduced with spores but had trunks sup-
ported by innermost concentric layers of heartwood (Niklas 1997). These
trees gave rise to the first seed plants in the Late Devonian, a reproduc-
tive event in plant evolution equivalent to the evolution of the amniote
egg in the tetrapods that we considered in the previous chapter, thus a
96 Chapter 3
vital event in the plant conquest of the terrestrial realm. The spermato-
phyte seed plants are incredibly diverse. Although they descend from the
original lignophyte trees, at least two groups of spermatophytes have lost
and secondarily reevolved the tree form: the cycad palms and the date
palms (table 3.1). The cycad palms have trunks that are supported by an
outermost layer of persistent leaf bases, whereas the date palms have
trunks supported by a fibrous complex of primary vascular bundles,
increasing in density of occurrence from the center to the outer margin
of the trunk (Niklas 1997), very different from the heartwood trunks of
most lignophyte trees today.
Niklas (1997, 316) argues strongly for the role of functional constraint
in the evolution of the tree form: “convergence among so many different
plant lineages for many biomechanical traits rather than only one or a
few important ones strongly suggests that these traits are the result of
adaptive evolution and not fortuitous expressions of diffusive evolution.”
Would not these same functional constraints produce the same evolu-
tionary result in alien worlds, as they have nine independent times here
on Earth? If sessile multicellular photoautotrophic organisms evolve on
distant Earth-like worlds, and if those organisms emerge from the seas
to invade the land, then they should necessarily—and convergently—
evolve the geometry of the tree form. Or, in the spirit of theoretical
morphology, can we conceive of an alternative geometry that these
organisms might evolve to solve the problems of overcrowding, gravity,
and maximizing surface area for photosynthesis? We shall return to this
question in the next chapter, when we consider the possible evolution of
balloon-like plants, plants that could float in the air, and why such hypo-
thetical plants do not exist on the Earth.
Niklas (1997) points out that both the ancient lepidodendrid scale
trees and the calamite horsetail trees not only convergently evolved tree-
trunk support structures, but also independently evolved traits such as
wood tissue, leaves, cones for sporangia, and heterosporous reproduc-
tion. These ancient trees depended upon a readily available and depend-
able supply of water for their free-spore mode of reproduction, and thus
were confined to swampy regions of the terrestrial realm—a develop-
mentally imposed constraint similar to that seen in early tetrapod invad-
ers of the terrestrial realm, the amphibians (as discussed in the previous
chapter). And just as the tetrapods escaped that constraint in the evolu-
tion of the amniotes, the plants escaped it in the evolution of the seed
plants, an event in reproductive evolution that we shall consider in detail
later in this chapter.
Convergent Plants 97
Because land plants are surrounded by air, not water, they require a
system to transport water from the soil below to the plant tissues above,
where it is needed. The simplest of the embryophytes, land plants such
as liverworts, have no roots and no vascular system. The derived and
highly successful vascular land plants are classified as tracheophytes, as
they possess tracheids, specialized cells for the transport of water up the
stem of the plant against the force of gravity (Lecointre and Le Guyader
2006). Advanced embryophytes, such as Cooksonia hemisphaerica from
the Late Silurian and Aglaophyton major from the Devonian, are now
considered to be transitional to the earliest true tracheophytes (Niklas
1997). These horneophytes had no roots or leaves, but did have small
stems that stood up against the force of gravity and bifurcated to produce
multiple tips that contained the sporangia. The tracheids of the earliest
tracheophytes were arranged in a series of circular rings, stacked one
atop another up the stem of the plant, or in a continuous climbing
helix. The walls of the tracheids were lignified, and thus also provided
some support to hold the stem of the plant erect against the force of
gravity. The question is still open as to whether tracheids are a synapo-
morphy, arising only once in plant evolution, and whether the Tra-
cheophyta are monophyletic. A possible alternative is that tracheids are
convergent, independently evolved in the lycophytes and the euphyllo-
phytes (Niklas 1997).
The next evolutionary step beyond tracheids for water transport within
the plant is the evolution of vessels for rapid and higher-volume water
transport. The evolution of vessels is clearly convergent: at least nine
separate phylogenetic lineages of tracheophytes have independently
evolved vessels (table 3.2). The angiosperms alone have convergently
evolved vessels several times (Niklas 1997); for example, the basal angio-
sperm nymphaealeans have independently evolved vessels at least twice
(Schneider and Carlquist 1995). Rates of water and food transport within
a plant are size limiting; with the convergent evolution of vessels, plants
began to achieve large sizes, and with the convergent evolution of trunks
(table 3.1) the invasion of land was in full force.
Plants are photoautotrophs: they produce their own food via photosyn-
thesis. Early land plants, such as Cooksonia hemisphaerica some 425
million years ago, had small green stems and no leaves. A stem is a small
cylinder, with a fairly large surface area where photosynthesis can take
98 Chapter 3
Table 3.2
Convergent evolution of water-conducting structures in plants
Note: The geological age of extinct species is marked with a †. For data sources, see text.
place, and a small volume of internal tissue in need of food. The key word
in the last sentence is “small”: the larger the cylinder, the smaller its
surface area relative to its volume, as surface area increases as a square
function of linear dimension whereas volume increases as a cubic func-
tion. Thus, the larger the stem of the plant, the smaller its surface area,
where photosynthesis occurs, relative to its volume of internal tissue,
where food is needed.
The solution to this classic area-volume problem is to somehow
increase the surface area of the plant that is capable of photosynthesizing
food—that is, to evolve leaves. The evolution of the leaf has occurred
independently in at least nine different phylogenetic lineages of land
plants (table 3.3). Interestingly, eight of these lineages convergently
developed leaves in the Devonian, from the simple liverworts to the first
of the derived lignophyte trees. Thus, the many leaves we see around us
each day, from ferns to maple tree leaves, can be traced back in time to
that explosion of convergent evolution in the Devonian.
Botanists divide leaves into two types: microphylls, which are small
with only one vein and no leaf gaps, and macrophylls, which are large
with complex venation and leaf gaps (Tomescu 2008). Not surprisingly,
microphylls were convergently evolved by Devonian liverworts, mosses,
Convergent Plants 99
Table 3.3
Convergent evolution of leaves in plants
Note: The geological age of extinct species is marked with a †. For data sources, see text.
and lycopods (Gensel and Kasper 2005); that is, in land plants that are
small and thus less in need of increased surface area for photosynthesis.
The analyses of Boyce and Knoll (2002) demonstrate that macrophylls
evolved convergently at least four separate times in the equisetophytes,
filicophytes, “progymnosperms” (nonspermatophyte lignophytes), and
spermatophytes. In table 3.3 I have listed a leaf species that is one of the
most ancient for each of these groups, taken from the analyses of Boyce
and Knoll (2002). Tomescu (2008) agrees that macrophylls originated
independently a minimum of four times, but presents phylogenetic analy-
ses that suggest the actual number of convergences may be nine. He
demonstrates that leaves were independently developed at least twice in
the ferns (separately in the Psilotaceae and the Zygopteridaceae), and
perhaps as many as five times. Likewise, he argues that leaves arose
independently in both the aneurophyte and archaeopterid nonspermato-
phyte lignophytes (table 3.3).
Before leaving this section on photosynthetic light-capture structures
in plants, I would like to touch briefly on the convergent evolution of a
different type of light-capture structure: eyes. Eyes are usually con-
sidered to be unique to animals, and we examined their convergent
100 Chapter 3
evolution in detail in the previous chapter (see table 2.15). However, not
only have eyes evolved in the unikontan branch of the eukaryote phy-
logenetic tree—the branch leading to animals—but they have conver-
gently evolved in the bikontan branch as well, the branch that leads to
the plants (see the appendix). And not just once—simple eyespots have
independently evolved in a wide variety of phototactic algae and algal
phytoflagellates (table 3.4). Eyespots have independently evolved no less
than 11 separate times in groups as diverse as green algae, brown algae,
yellow-green algae, and many algal flagellates (Foster and Smyth 1980;
Kivic and Walne 1983). Kivic and Walne (1983) document the convergent
evolution of different types of eyespots in the algae (table 3.4), and
Foster and Smyth (1980, 619) also demonstrate that the phototaxis
Table 3.4
Convergent evolution of eyes in algae and algal phytoflagellates
control system for the motile species often has “dramatically different
machinery” and has been independently derived in each group.
While present in some of the most plesiomorphic Chlorobionta, like
the green algae, eyespots are lost in higher plants, particularly sessile land
plants. Land plants have evolved a series of phototropic mechanisms
instead, as seen in plant species that slowly turn in the light, tracking the
path of the sun across the sky, or opening and closing their flowers in
response to the rising and setting of the sun. Charles Darwin was fasci-
nated with heliotropic movement in plants, and discovered in his experi-
ments that his plants stopped tracking the sun during the day if a filter
removing blue wavelength light was placed between them and the sun.
We now know that the genes encoding for heliotropic movement in
plants have been convergently evolved in animals, where they are used
in maintaining circadian rhythms (this example of molecular conver-
gence will be considered further in chapter 5).
Table 3.5
Convergent evolution of the C4 photosynthetic pathway in plants in the past 30 million
years
Table 3.5
(continued)
Table 3.6
Convergent evolution of physical defenses in plants
1 Convergent structure and function: LEAF SPINES (spines on apices of leaf lobes or
around leaf margins to discourage herbivore biting)
Convergent lineages:
1.1 Cycad (Euphyllophyta: Lignophyta: Spermatophyta: Cycadophyta: Cycadales:
Zamiaceae; Encephalartos ferox)
1.2 Aloe yucca (Spermatophyta: Angiospermae: Euangiosperms: Monocotyledons:
Eumonocotyledons: Liliales: Liliaceae; Yucca aloifolia)
1.3 Cogon grass (Euangiosperms: Monocotyledons: Eumonocotyledons: Commelinidae:
Poaceae; Imperata cylindrical)
1.4 Black oak (Euangiosperms: Eudicots: Rosidae: Eurosids I: Fagaceae; Quercus
velutina)
1.5 Scorpion weed (Euangiosperms: Eudicots: Asteridae: Euasterids I: Boraginaceae;
Phacelia purshii)
1.6 European holly (Euangiosperms: Eudicots: Asteridae: Euasterids II: Aquifoliaceae;
Ilex aquifolium)
1.7 Bull thistle (Euangiosperms: Eudicots: Asteridae: Euasterids II: Asteraceae; Cirsium
vulgare)
2 Convergent structure and function: TWIGS/BARK WITH THORNS/SPINES (thorns
or spines on twigs and/or trunk bark to discourage herbivores)
Convergent lineages:
2.1 Candle cactus (Euangiosperms: Eudicots: Core eudicots: Caryophyllales: Cactaceae;
Pilocereus lanuginosus)
2.2 Knotweed (Euangiosperms: Eudicots: Core eudicots: Caryophyllales: Polygonaceae;
Polygonum pennsylvanicum)
2.3 Madagascar spiny plant (Euangiosperms: Eudicots: Core eudicots: Caryophyllales:
Didiereaceae; Alluaudia montagnacii)
2.4 Canary Island spurge (Euangiosperms: Eudicots: Rosidae: Eurosids I: Euphorbiaceae;
Euphorbia canariensis)
2.5 French rose (Euangiosperms: Eudicots: Rosidae: Eurosids I: Rosaceae; Rosa
gallica)
2.6 European buckthorn (Euangiosperms: Eudicots: Rosidae: Eurosids I: Rhamnaceae;
Rhamnus cathartica)
2.7 Honey locust (Euangiosperms: Eudicots: Rosidae: Eurosids I: Fabaceae; Gleditsia
triacanthos)
2.8 Common prickly-ash (Euangiosperms: Eudicots: Rosidae: Eurosids II: Rutaceae;
Zanthoxylum americanum)
2.9 Bottle tree (Euangiosperms: Eudicots: Asteridae: Euasterids I: Apocynaceae;
Pachypodium lealii)
3 Convergent structure and function: STINGERS (stinging poison hairs and needles to
discourage herbivores)
Convergent lineages:
3.1 Stinging nettle (Euangiosperms: Eudicots: Rosidae: Eurosids I: Urticaceae; Urtica
dioica)
3.2 Spurge stinging nettle (Euangiosperms: Eudicots: Rosidae: Eurosids I:
Euphorbiaceae; Cnidoscollus stimulosus)
3.3 Cowhage vine (Euangiosperms: Eudicots: Rosidae: Eurosids I: Fabaceae; Mucuna
pruriens)
Convergent Plants 107
Table 3.6
(continued)
of this tree is like that of a cnidarian jelly fish and can actually be fatal
(Stewart 2009). The Euphorbiaceae family is famous for its eerie conver-
gence on numerous cactus forms, which we shall consider later in this
chapter, but it has also produced a convergent stinging nettle that is
virtually identical to a true nettle (table 3.6).
Last, some plants have recruited animals to defend them (table 3.6).
The defender species are sometimes ants, like those hosted by the
whistling-thorn acacia, or wasps, hosted by the valley oak. This conver-
gent mutualism is not common; much more abundant is the convergent
assistance of animals in plant reproduction, a topic that will be taken up
later in the chapter.
Convergent with animals (see table 2.13), plants have evolved chemi-
cal defenses as well. Unlike animals, however, plants have independently
acquired a staggering number of chemical defense systems (table 3.7).
This is not surprising, as the number of defensive strategies available to
organisms that cannot flee or actively fight is more limited. In table 3.7
I have listed numerous examples of the convergent evolution of poison
in different plant structures, from leaves to seeds to roots, taken largely
from the works of Stewart (2009) and Agosta (1996) and phylogeneti-
cally cross-checked using the APG II (2003). The most poisonous plant
families are the nightshades (Solanaceae), cashews (Anacardiaceae),
spurges (Euphorbiaceae), and parsleys (Apiaceae), groups that Stewart
(2009, 71) collectively terms “botanical crime families.” But this designa-
tion is only from our perspective; from the plant’s perspective, it is simply
defending itself. Although these families contain the most numerous, and
most infamous, species of poisonous plants, poison tissues are found scat-
tered across the plant phylogenetic spectrum, from nonflowering pines
and cycads, to more basal core angiosperms like magnoliids and mono-
cotyledons, to highly derived eudicots like rosids and asterids. For each
separate clade of plants, I have listed a single species example in table
3.7, regardless of whether the clade contains many poisonous species or
only a few, in order to illustrate the extent of convergent evolution of
chemical defenses across all clades.
The poison produced in convergent chemical defense is not always the
same, but even here biochemical convergence is widespread (Stewart
2009). For example, cyanide has been independently developed in mono-
cotyledons (Johnsongrass leaf), eurosids (rubber tree seeds), and asterids
(black elderberry), to name only a few examples. Poisonous alkaloid
neurotoxins like nicotine and cocaine have been independently evolved
by monocotyledons (betel nut), basal eudicots (monkshood plant), rosids
Convergent Plants 109
Table 3.7
Convergent evolution of chemical defenses in plants
Table 3.7
(continued)
Table 3.7
(continued)
Table 3.8
Convergent evolution of water-retention structures in plants
Table 3.9
Convergent evolution of cactus forms in arid-adapted desert plants of the Western
Hemisphere (Cactaceae) and the Eastern Hemisphere (Euphorbiaceae)
Reproduction
In the last chapter we saw that animals have repeatedly and indepen-
dently evolved the same reproductive systems, particularly viviparity. The
same is true of plants, except the convergent reproductive system for
these organisms is heterospory. The repeated evolution of heterospory
in independent phylogenetic lineages has been called “the most iterative
key innovation in the evolutionary history of the plant kingdom”
(Bateman and DiMichele 1994, 345).
The original Devonian tracheophytes had sporophytes that shed their
spores freely into the air; the spores, which were all the same size, then
produced free-living gametophytes that produced the next generation of
sporophytes. The free-living gametophytes were bisexual, and required
moist conditions not only for their vegetative growth but also for sperm
transfer and fertilization of their eggs, as well as the survival of the devel-
oping new sporophyte embryo (Niklas 1997). The production of bisexual
gametophytes is an ancient reproductive condition known as homospory,
which still exists in many living species of club mosses (lycophytes),
horsetails (equisetophytes), and ferns (filicophytes). In many ways,
homosporous plants “may be thought of as the amphibians of the plant
kingdom in the sense that the completion of their sexual life cycle
requires a ‘return to water’” (Niklas 1997, 190).
A major evolutionary shift in plants’ mode of reproduction occurred
independently in many separate plant lineages during the Late Devonian
(table 3.10). This was the convergent evolution of heterospory, where
plants now produced two distinctly different types of spores: numerous,
very small microspores and relatively few but much larger megaspores.
Specialized plants producing the new microspores developed into sperm-
producing gametophytes, and plants producing the larger megaspores
developed into egg-bearing gametophytes; thus, heterosporous plants
had now evolved gametophytes that were unisexual, not bisexual like the
homosporous plants (Niklas 1997). The evolution of heterospory and
unisexual gametophytes occurred independently at least eleven separate
times in disparate groups of plants (Bateman and DiMichele 1994). Two
separate groups of lycopods, one major group of equisetophytes, and
three separate groups of filicophytes convergently became heterospo-
rous (table 3.10), although homosporous species of these types of plants
exist to the present day. The filicophytes are particularly interesting in
that they have convergently evolved heterospory repeatedly throughout
geologic time: at least once in the Late Devonian, at least once in the
116 Chapter 3
Table 3.10
Convergent evolution of heterosporous reproduction, and ultimately of the seed
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Late Cretaceous (the ancestors of living giant water ferns), and sepa-
rately in the modern European water-clover ferns and Australian
platyzoma ferns (table 3.10). Four separate groups of lignophytes inde-
pendently evolved heterospory in the span of time from the Middle
Devonian to the Late Carboniferous: the major plant lineages of the
aneurophytales and archaeopteridales, and the more enigmatic noeg-
gerathiales and cecropsidales, all of which are now extinct. In the midst
of this rampant convergent evolution of heterospory in plant lineages,
one plant group took the next major evolutionary step in plant reproduc-
tion: the evolution of the seed (table 3.10).
The evolution of heterospory, the partitioning of resources between
numerous small microspores and a few large megaspores, set the stage
for the evolution of the seed (Niklas 1997). Production of microspores
required minimal resources, and because they were so very small they
Convergent Plants 117
the seed plants, are the most successful of all the land plants, the angio-
sperms, the flowering plants, are the most successful of all the
spermatophytes.
Why flowers? The plant seed, like the animal amniote egg, is a repro-
ductive structure clearly linked to antidehydration adaptations in the
hostile dry air of terrestrial habitats. Much of the success of flowering
plants is linked to solving not dehydration problems but a different
reproductive problem: the problem of fertilization in organisms that
cannot move about in search for a mate.
The problem of fertilization in sessile organisms is an ancient one that
extends back to life in the sea. Many sessile marine animals simply
release their gametes directly into the sea water and depend upon water
currents to transport them to other animals for fertilization. These
animals are often astonishingly convergent to plants in their morpholo-
gies: crinoids are deuterostomous bilaterians, closely related to chordate
animals such as ourselves, yet many ancient crinoids had root systems
that looked like those of land plants, stems like plant trunks, and branch-
ing arms like plant branches. Plant similarities are often seen in both the
colloquial and scientific names of sessile marine animals: stalked echino-
derms are known as “sea lilies” because their long stems and calyxes look
like the stems and flowers of lilies, the bryozoa are named “moss animals”
because their colonies look like moss, the gorgonian cnidarian “sea fans”
whose central colonial stem and lateral branches look like palm leaves,
the colorful polyps of an anthozoan cnidarian colony appear similar to
a garden of “sea flowers,” and so on.
The apparent convergence in form between many sessile marine
animals and land plants is in fact reversed: it is the land plants that have
converged on the sessile marine animals, not the other way around, as
the marine animals evolved these forms first and the plants converged
on them secondarily. Rather than currents in the dense water of marine
habitats, land plants must depend upon wind in the thin atmosphere of
terrestrial habitats to disperse their gametes. Just as many sessile marine
animals have elevated themselves above the sea floor in order to catch
stronger marine currents, plants have evolved stems and trunks to
elevate them up in the air in order to catch the wind with their spores
and pollen.
Many land plants are thus not only dependent upon the vagaries of
terrestrial weather patterns in bringing rain; they are also dependent
upon these same weather patterns in bringing wind for successful fertil-
ization. The spermatophytes largely depended upon wind for fertilization
Convergent Plants 119
for over 240 million years, from the Late Devonian until the evolution
of the first angiosperms some 115 to 125 million years ago (Thien et al.
2009). At that time, the angiosperm spermatophytes evolved an interest-
ing new fertilization strategy: they began to involve a sexual partner who
was mobile. Although a few nonangiosperm spermatophytes use pollen-
eating insects as pollination assistants, the really elaborate plant-animal
reproductive pas de deux began with the evolution of the first flowering
plants in the Early Cretaceous (and with the evolution of the dioecious
angiosperms, which have separate male and female plants, the plant-
animal interaction becomes a ménage à trois).
The basal angiosperms consist of three sequential cladogram branches,
in what has been termed the ANITA grade of angiosperm evolution
(Thien et al. 2009). These branches are: first, the Amborellales, second,
the Nymphaeales, and third, the Austrobaileyales (APG II 2003; Lecoin-
tre and Le Guyader 2006), based upon molecular phylogenies. Although
molecular evidence places the Amborellales at the very base of the
angiosperm clade, the oldest fossil flowers found thus far belong to the
Nymphaeales, one branch up from the base of the clade.
The oldest flower yet discovered is quite small, only about 1 centimeter
in diameter. Somewhat younger nymphaealean flowers, around 90 million
years old, have fossil morphologies very similar to modern-day flowers
that are pollinated by beetles, reinforcing long-held views that beetle
pollination was a major form of plant-animal reproductive interaction
in the earliest of the flowering plants (Bernhardt 2000; Thien et al.
2009). Examination of living members of the basal angiosperms reveal
that flies are the primary pollinators of six families, beetles the primary
pollinators of five families, and bees are major pollinators in only one
family (Thien et al. 2009). Plant-animal interactions surely must have
evolved in the 115 to 125 million years that have passed since the evolu-
tion of the first flowering plant, and so the reproductive pattern seen in
living basal angiosperms cannot be taken as proof of identical relation-
ships in the Early Cretaceous. Still, it is interesting that living basal
angiosperms are pollinated preferentially by flies and beetles, and only
rarely by bees. More derived angiosperms not only added bees to the list
of sexual partners but added butterflies, moths, wasps, birds, and mammals
as well.
Four different types of flower morphologies have been evolved by
angiosperms that are pollinated exclusively by beetles: bilobate, brush,
chamber-blossom, and painted-bowl flowers (Bernhardt 2000). These
cantharophilous flower types have been convergently evolved over and
120 Chapter 3
over again in even highly derived angiosperms (table 3.11); that is, beetle
pollination is not a reproductive mode confined to the basal angiosperms
or plesiomorphic magnoliids. Nine independent groups of monocotyle-
dons, eight of them derived eumonocotyledons, have independently
evolved cantharophilous flowers (table 3.11). In the derived eudicot
angiosperms, cantharophilous flowers have convergently appeared 12
separate times, ranging from basal eudicots like the Ranunculalaes to the
highly derived Rosidae and Asteridae (Bernhardt 2000).
Giant flowers are rarer and more specialized than cantharophilous
flowers, yet they too have convergently evolved multiple times in the
angiosperms (table 3.12). Floral gigantism is defined as the possession of
blossoms—either flowers or flowerlike inflorescences—that exceed 30
centimeters in diameter (Davis et al. 2008). Very large flowers can be up
to a meter in diameter, such as those found in rafflesia, or up to three
meters in height, like as those found in the corpse lilies (table 3.12). These
giant flowers attract a wide range of pollinators: beetles, flies, moths,
birds, fruit bats, and even lemur primates (Davis et al. 2008). However,
the great majority of pollinators of giant flowers are flies and small-
bodied beetles. It appears that the large flower size facilitates the tem-
porary trapping of these smaller animals by the flower, and the animals
subsequently pollinate the flower as they seek an escape route. The phy-
logenetic distribution of the convergent evolution of floral gigantism is
across the entire spectrum of the angiosperm clade: from the most basal
angiosperms, like the Nymphaeales, to the highly derived Eurasterids
(table 3.12).
Even more specialized than giant flowers are flowers that smell like
rotting meat. One would think that such a specialized type of flower,
specifically adapted to the restricted role of attracting carrion-eating
animals, would be unique in the evolution of flowering plants, but this is
not the case: sapromyiophilous, or carrion-mimic, flowers have conver-
gently evolved in the angiosperms no less than seven separate times
(table 3.13). Some carrion-mimic flowers are also giant, like the corpse
lilies and rafflesia (table 3.12), but others are normal sized or even small
(Davis et al. 2008; Stewart 2009). Sapromyiophilous flowers have inde-
pendently evolved in plesiomorphic angiosperms like the magnoliid
pawpaws; in monocotyledons like the corpse lilies and carrion flowers;
in basal eudicots like the stinking hellebore; and in highly derived euaste-
rids like the Zulu giant carrion plant (table 3.13).
Derived flowers have evolved shapes, colors, and odors to attract the
attention of potential pollinators. Flowers also offer food rewards to
Convergent Plants 121
Table 3.11
Convergent evolution of cantharophilous flowers for plant pollination by beetles in
angiosperms
Table 3.12
Convergent evolution of floral gigantism in plants
Table 3.13
Convergent evolution of sapromyiophilous flowers that smell like rotting meat
animal pollinators: the ancient flowers offered only excess pollen for
pollen-eating animals, a trait that is actually older than the angiosperms,
as it is found in some nonangiosperm spermatophytes like cycads and
conifers as well (Labandeira et al. 2007). Derived flowers offer not only
excess pollen but also sweet nectar, starchy food bodies, stigmatic secre-
tions, perianth segments, and edible stamenodia (Bernhardt 2000; Thien
et al. 2009).
One particularly specialized type of food reward system, known as
ornithophily, is adapted specifically to attract nectar-feeding birds. Orni-
thophilous, or hummingbird, flowers have no smell; they are of a brick-
red monocolor, are trumpet-shaped, and produce copious amounts of
nectar. The evolution of hummingbird flowers in western North America
has been extensively studied by Grant and Grant (1968), who document
the independent evolution of ornithophilous flowers in 18 separate fami-
lies of angiosperms in this region of the world alone (table 3.14). For
each family, I have listed a single species representative, but some fami-
lies have more hummingbird-flowered species than others. For example,
the Scrophulariaceae contain eight genera and 74 species with humming-
bird flowers, and a more detailed phylogenetic analysis of the family
would be needed to determine how many of these hummingbird-
flowered species are independently derived and how many are synapo-
morphic. Other families, like the Cactaceae, Nyctaginaceae, Saxifragaceae,
Fouquieriaceae, Convolvulaceae, and Rubiaceae contain a single species
with hummingbird flowers, and it is clear that each of these is an inde-
pendent convergence on ornithophily (Grant and Grant 1968). Last, of
the 18 convergently evolved hummingbird flowers listed in table 3.14,
seven belong to species that are the sole red-flowered representative of
all of the other variously hued flowers belonging to other species within
each particular genus: Silene laciniata, Ribes speciosum, Astragalus coc-
cineus, Gilia aggregata, Ipomoea coccinea, Monardella macrantha,
Mimulus cardinalis; and only two red-flowered species are found in the
generally blue- and white-flowered species of the genus Delphinium, D.
cardinale and D. nudicaule (Went 1971).
Worldwide, some 65 families of angiosperms have evolved ornithophi-
lous flowers (Cronk and Ojeda 2008). It will take a detailed analysis of
the phylogenetic distribution of these flowers to determine how many
are independent originations of ornithophily. For example, species in all
eight families of the monocotyledon order Zingiberales are pollinated
by birds (Cronk and Ojeda 2008). Do these eight families represent eight
independent convergencies on ornithophily within the Zingiberales, or
124 Chapter 3
Table 3.14
Convergent evolution of ornithophilous hummingbird flowers of western North America
Table 3.15
Convergent evolution of physical seed-dispersal systems in plants
Table 3.15
(continued)
Table 3.16
Convergent evolution of fleshy fruit in monocotyledon angiosperms
the Arecaceae), while others have not (such as the clade of the
Ruscaceae).
Interestingly, Givnish et al. (2005) also have demonstrated that former
fruit-bearing species have secondarily lost the fruit habit 11 separate
times within the clade of the monocotyledons. Their careful ecological
analyses of the phylogenetic distribution of fruit occurrence within the
monocotyledons reveals that species that have invaded shaded habitats
have often subsequently evolved fleshy fruit, whereas species possessing
fruit have often lost the fruit habit if those species returned to open
habitats. Apparently animal dispersal of seeds is needed for plants
growing under closed canopies, whereas in open habitats less costly
mechanisms of wind dispersal seem to suffice. Up to 95 percent of woody
understory species in neotropical rain forests have evolved fleshy fruits,
a phenomenon that Givnish et al. (2005, 1481) have termed “concerted
convergence”; they point out that this pattern is among “the strongest
ever demonstrated for evolutionary convergence in individual traits and
the predictability of evolution.”
Even more specialized than the evolution of fruit and subsequent seed
dispersal by vertebrate animals is the evolution of seed dispersal by ants,
or myrmecochory (Dunn et al. 2007). Myrmecochorous plants have
evolved elaiosome, a specialized lipid-rich structure that ants use as a
handle in carrying the seed from the plant back to their nest. At the nest,
the ants remove the elaiosome (which is the plant’s reward to the ants),
and the seed is discarded in a midden or outside of the nest itself. The
elaiosome has clearly been convergently evolved by numerous different
plants, as the structure has multiple developmental origins (Dunn et al.
2007).
As phylogenetic relationships within the clade of the monocotyledons
have now been resolved (Givnish et al. 2005), Dunn et al. (2007) have
used this phylogeny to study the distribution of elaiosome origins. Myr-
mecochory has independently evolved 19 separate times within the clade
of the monocotyledons (table 3.17). In table 3.17 I have listed basal or
near-basal species for each independent evolution of elaiosomes in the
cladogram of Dunn et al. (2007). Unlike monocotyledons that have con-
vergently evolved fruit (table 3.16), Dunn et al. (2007) have demon-
strated that the evolution of myrmecochory in monocotyledons is
independent of whether habitats are shaded or open, and hence inde-
pendent of the evolution of fleshy fruits themselves. Rather, the evolu-
tion of myrmecochory in plants appears to be directly linked to the
evolutionary diversification and increase in population sizes of ants
132 Chapter 3
Table 3.17
Convergent evolution of myrmecochory in monocotyledon flowering plants
beginning in the Late Eocene, some 37 million years ago. The peak in
the evolution of fleshy fruit in the monocotyledons began earlier, some
85 million years ago, and was triggered by the spread of Late Cretaceous
forests and closed-canopy habitats (Givnish et al. 2005).
Although myrmecochorous plants are equally likely to be found in
either open habitats or closed-canopy habitats, Dunn et al. (2007) do
demonstrate that monocotyledons in open habitats that are fire-prone,
such as in Mediterranean ecosystems, have preferentially evolved myr-
mecochory. They suggest that transport of seeds to ant nests confers the
selective advantage of reducing the risk of seed mortality due to fire, in
addition to the selective advantage of geographic dispersal of the seeds.
This conclusion is supported by the documented convergent evolution
of myrmecochory in the fire-prone shrub lands of both South African
and Australian ecosystems (Milewski and Bond 1982), which provides
an example of ecosystem convergence as well, the subject of the next
chapter of this book.
In summary, the incidences of reproductive convergence in plants, just
as in the animals we considered in the previous chapter, vastly over-
shadow the classic examples of convergent evolution that are usually
offered for both groups. The spectacular convergent evolution of wings
in animals pales in comparison with the astonishing number of indepen-
dent origins of viviparity, and the convergent evolution of towering trees
in plants pales in comparison with the astonishing number of indepen-
dent origins of the fruit habit. I predict the 22 separate convergences on
the fruit habit discovered in the monocotyledons alone (table 3.16) are
going to be a small number indeed when rigorous phylogenetic analyses
are eventually applied to mapping the number of independent origins of
fleshy fruit in the huge clade of the eudicots.
In many ways, plant evolution is more complex and more interesting
than animal evolution because of its overtly multispecies aspect. Plants
coevolve with the numerous animal partners that they have involved in
their reproduction and defense, and these beetles, flies, bees, wasps, but-
terflies, moths, ants, birds, and mammals have themselves evolved in
response to plant evolution. In the next chapter we shall focus on the
phenomenon of convergent evolution at the multispecies level: that is,
the convergent evolution of entire ecosystems.
4 Convergent Ecosystems
Table 4.1
Convergent evolution of carnivory in plants
Table 4.2
Convergent evolution of heterotrophy in plants
Table 4.3
Convergent evolution of woodpecker ecological equivalents in animals
lower beak, and then extracts insects from the bark with the hooked
upper beak. As their name implies, the sickle-billed vangas have sickle-
shaped beaks, which they insert into bark cracks and openings; they then
use a lever action to pull out the insect.
The woodpecker finches of the Galapagos have evolved a totally
different approach to extracting wood-boring insects—they use tools.
These birds use their beaks, not to drill with, but to hold a long, sharp
tool, usually a cactus spine. They insert the cactus spine into bark
cracks and openings, and also use a lever action to pull out insects, but
using the tool rather than a specialized beak morphology. We shall con-
sider the convergent evolution of tool usage by animals in detail in
chapter 6.
Not only did tool usage and three distinctly different types of beak
morphologies evolve in independent bird lineages to fill the same eco-
logical niche, but no less than three separate lineages of mammals have
also converged on the same restricted ecological role of eating wood-
boring insects (table 4.3). All three of these mammal groups use very
elongated fingers, tipped with claws, to insert into tree bark and to pull
out enclosed insects. The particular fingers used, however, are different
in each group. The striped possum of New Guinea, a marsupial mammal,
has evolved a very elongated fourth digit (the ring finger in humans) for
this function, whereas the aye-aye of Madagascar, a placental mammal,
142 Chapter 4
has evolved an elongated third digit (the middle finger in humans) for
the same function. These two groups of mammals are alive today, but
there existed a third group, the extinct apatemyids, that also converged
on this same ecological role (table 4.3). The apatemyids evolved two
elongated fingers rather than one, the second digit (the pointing finger
in humans) and the third digit, for use in extracting insects from tree
bark (Koenigswald and Schierning 1987).
Again, what is convergent in all of these animals is the ecological role
of eating wood-boring insects, not the morphologies of the animals as
previously considered in detail in chapter 2. Birds, bats, pterosaurs, and
butterflies have all convergently evolved a morphologically similar struc-
ture, namely wings, to accomplish the same function: powered flight (see
chapter 2). Here, greatly different morphological structures and behav-
iors have evolved in independent phylogenetic lineages in order to con-
vergently fill the same ecological niche: eating wood-boring insects. A
bat superficially looks like a bird, but a red-headed woodpecker looks
very different from a striped possum. Morphologically, these animals are
very different, but ecologically they are equivalents.
Consider yet another highly specialized form of insectivory: attacking
and eating ants in their nests. Huge numbers of ants can inhabit an ant
colony, and they are highly aggressive in defending that colony. Thus,
attacking an ant nest is not a trivial task, and one might conclude that
only one type of highly specialized animal would fill the niche of such a
restrictive mode of food acquisition. But within the mammals alone, five
separate groups have converged on myrmecophagy, three times indepen-
dently in the placental mammals, once in the marsupial mammals, and
once in the monotreme mammals (table 4.4). Interestingly, in the placen-
tal mammals both morphological convergence and ecological conver-
gence have occurred, in that the South American giant anteater, the
Asian pangolin, and the African aardvark have all evolved long claws
and strong forelimbs for tearing into ant nests. All three groups of animals
have evolved tubular-shaped mouths that contain a very long, sticky
tongue for capturing ants. The anteaters and pangolins have totally lost
their teeth, and the aardvark retains only a few teeth. All three groups
have highly developed salivary glands, which assist in swallowing copious
amounts of ants, and all three have convergently evolved strongly mus-
cular pyloric regions in their stomachs to assist in digesting ants (Lecoin-
tre and Le Guyader 2006). Yet these animal groups are only very distantly
related: the anteaters are xenarthrans, basal eutherians; the aardvarks
are afrotherians; and the pangolins are laurasiatherians (table 4.4).
Convergent Ecosystems 143
Table 4.4
Convergent evolution of myrmecophagous ecological equivalents in animals
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Table 4.5
Convergent evolution of necrophagous ecological equivalents in animals
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Table 4.6
Convergent evolution of flying nectarivorous ecological equivalents in animals
nectarivore ecological niche (Cronk and Ojeda 2008). But they are not
alone, for no less than five other families of passeriform birds and one
family of psittaciform birds have convergently evolved nectarivory (table
4.6). In the Western Hemisphere the red-legged American honeycreep-
ers (Thraupidae) and Tennessee warblers (Parulidae) have convergently
evolved nectarivory, and coexist with hummingbirds. In the Eastern
Hemisphere, the South African sugarbirds (Promeropidae) and Japanese
white-eyes (Zosteropidae) have convergently evolved nectarivory, and
coexist with the sunbirds. In Australia the swift parrots (Psittacidae) and
in Hawaii the ‘I’iwi honeycreepers (Drepanididae) have convergently
evolved nectarivory, and coexist with honeyeaters. All of these birds have
independently evolved many of the nectarivorous adaptations seen in
the hummingbirds: lowered metabolic rates and protein requirements in
honeyeaters; glossy feathers and slender, thin beaks in swift parrots;
extensible brush tongues independently in both honeyeaters and swift
parrots, and so on (Gartrell 2000).
Convergence on this extremely specialized way of surviving is not
limited to the birds. In North America, the hummingbird clearwing moth
is an insect that has not merely converged on the ecological role of nec-
tarivory, it is also a hummingbird morphological mimic. This moth has
converged so closely on the shape and color of a hummingbird that it is
often mistaken for one—only when one notices a pair of antennae on
the head of the animal does one suddenly realizes that it is not a bird at
all. And last, even the mammals have evolved nectarivory (table 4.6). The
Mexican long-tongued bat, aptly named as its tongue can be as long as
one-quarter of its body length, has convergently filled the nectarivore
ecological niche even though it looks nothing like a hummingbird.
I will use one last—and major—example of ecological niche conver-
gence to conclude this section of the chapter, and to set the stage for the
next section, in which we will consider the ecological convergence of
entire ecosystems. This is the convergent evolution of ground-dwelling,
flightless birds. Many different phylogenetic lineages of birds have inde-
pendently devolved the ability to fly, losing the characteristic set of traits
that sets them apart from the other theropod dinosaurs (Roff 1994;
Roots 2006). Their wings become reduced and vestigial, with the shorten-
ing and weakening of the humerus, ulna, and radius within the wing.
Their sterna become flat, losing the prominent keel for anchoring power-
ful flight muscles. Flight feathers in the wing and tail are shortened until
they become useless for flight. Finally, the feathers lose their aerody-
namic asymmetry and their smooth aerodynamic contours, which were
148 Chapter 4
Table 4.7
Convergent evolution of mammalian ecological equivalents in birds
Table 4.7
(continued)
Table 4.7
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
In the previous section of the chapter, we have seen that the number of
ecological roles or niches available for Earth organisms is demonstrably
limited, in that species from many different phylogenetic lineages have
156 Chapter 4
Table 4.8
Convergent evolution of ecological-analog compositions of marsupial-dominated ecosys-
tems in isolated Australian and Tasmanian regions and placental-dominated ecosystems in
the rest of the world
Table 4.8
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
another, yet they existed at the same time and were remarkably similar
in appearance.
Many of the South American ecological convergences were made by
meridiungulate mammals, a group of mammals that are now extinct (see
appendix). Completely in isolation, the meridiungulates convergently
filled the ecological niches of wild pigs, woodchucks, mastodons, horses,
rabbits, camels, hippopotamuses, beavers, and tapirs with their own
analogs (Prothero and Schoch 2002).
It is interesting that 13 ecological niches in Australia and 12 ecological
niches in South America, each evolving in isolation, converged with
niches in other regions of the world. However, by comparing the lists of
ecological analogs given in tables 4.8 and 4.9, we can see that the two
sets of niches listed are not all the same. In fact, only 5 of the 12 (or 13)
in the two sets are the same: the great cat, anteater, woodchuck, deer,
and rabbit niches. The South American mammals evolved a series of
ecological roles that are analogs of those independently evolved in North
America, but that appear not to have been discovered by the Australian
marsupials. These include the elephant, camel, hippopotamus, tapir,
beaver, and wild pig niches. Amphibious herbivores and large frugivores,
in particular, appear to be missing in Australian ecosystem evolution.
These absent ecological roles may well be due to the aridity of Australian
ecosystems and its effect on the Australian flora. Yet if that is the reason,
why did the Australian marsupials not evolve the camel niche, an ecologi-
cal role that independently appeared in both South and North American
ecosystems?
In the reverse comparison, several predator niches appear to be
missing in South American ecosystem evolution. Why did the South
American marsupials not evolve the wild cat, wolf, and wolverine
niches? Why were there no South American marsupial mole, rat, or
flying squirrel analogs? Some of these absences may be artificial, and
simply due to the vagaries of fossil preservation. All of the Australian
animals listed in table 4.8 either are alive today or have only relatively
recently become extinct (that is, going back in time at most about two
million years, to the beginning of the Pleistocene). In contrast, the pat-
terns of South American ecosystem evolution examined in table 4.9
extend back in time to the Eocene, some 55 million years into the past
(Prothero and Schoch 2002). Thus, some of these missing niches may
in fact have been filled in the past, but their fossil species either have
not been found yet or were never preserved in the fossil record in the
first place.
162 Chapter 4
Table 4.9
Convergent evolution of ecological-analog compositions of terrestrial ecosystems in iso-
lated South American regions and terrestrial ecosystems in the rest of the world in the
Cenozoic
Table 4.9
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Table 4.10
Convergent evolution of ecological-analog compositions of Mesozoic dinosaur-dominated
ecosystems and Cenozoic mammal-dominated ecosystems
Table 4.10
(continued)
Note: The geological age of extinct species is marked with a †. For data sources, see text.
168 Chapter 4
the identity of the players, whether they are dinosaurs or mammals (table
4.10), placentals or marsupials (table 4.8), and that these players have
convergently filled those roles independently in different positions in
both space and time. While stressing the importance of the phenomenon
of ecological convergence, Pianka (1978: 300) also cautions that “evolu-
tionary convergence can easily be read into a situation by placing undue
emphasis upon superficial similarities but failing to appreciate fully
the inevitable dissimilarities between pairs of supposed ecological
equivalents.”
Thus, an alternative approach to the comparison of existent ecosys-
tems would be to attempt to compare existent ecosystems with nonexis-
tent ecosystems; then we could seek to understand what it is about
existent ecosystems that has led organisms to construct them repeatedly
rather than constructing theoretically possible but nonexistent ecosys-
tems. That is, we can apply the analytical techniques of theoretical mor-
phology (McGhee 2007) to ecology. As Maclaurin and Sterelny (2008,
128–129) argue, “one way of thinking about ecological diversity is in
terms of a phenomenological ecospace. The dimensions of that space
. . . define a space of possibilities,” analogous to a theoretical morpho-
space; but they also caution that “[a]s with morphospace, though, a total
ecospace is of high and somewhat arbitrary dimensionality” (Maclaurin
and Sterelny 2008, 111).
Pianka (1978) in fact used this approach in his attempt not only to
create an empty niche—to specify an ecological role a priori without
reference to any living organism—but also to arrange those empty niches
into a “periodic table of niches,” a space of potential ecological roles,
again analogous to a theoretical morphospace. He used two dimensions
to construct that space: a trophic-niche dimension and a life-history-
niche dimension (table 4.11). The trophic-niche dimension spans the
possible trophic spectrum of autotrophy (primary producers) to heter-
otrophy (herbivores and carnivores). The life-history-niche dimension
spans the possible spectrum from r-selected organisms (those with high
reproductive rates, low parental care for offspring, and short life spans)
to K-selected organisms (those with low reproductive rates, high parental
care for offspring, and long life spans).
Within this space, Pianka (1978) thus created nine permutation posi-
tions, empty niches, or potential ecological roles, in the absence of any
actual living organisms. The next step in the analysis is to ask: are these
potential niches actually empty in nature? The answer is no—Pianka
(1978) discovered that actual living organisms have evolved each of the
nine potential ecological roles (table 4.11). Pianka (1978) then added an
170 Chapter 4
Table 4.11
The ecological periodic table of niches
Trophic niche
Ecological life-history
niche Primary producers Herbivores Carnivores
Table 4.12
Pianka’s inclusion of organisms that exploit space in three dimensions (3D), as opposed
to just two dimensions (2D), in his periodic table of niches
Trophic niche
Ecological
life-history niche Primary producers Herbivores Carnivores
Note: Modified from Pianka (1978). Note that the three possible ecological niches of
three-dimensional space exploitation by primary producers are not filled.
Table 4.13
Occupation of marine-niche ecospace in the Cambrian
Trophic niche
Habitat-mobility
niche Detritivores Herbivores Carnivores
Pelagic Trilobites — —
Epibenthic
1. Sessile Brachiopods, — —
Eocrinoids
2. Mobile Trilobites, Ostracodes Monoplacophorans, —
Ostracodes
Endobenthic
1. Mobile, shallow Trilobites, Bristle — Bristle
worms, Brachiopods worms
2. Mobile, deep — — —
Table 4.14
Occupation of marine-niche ecospace in the Middle Paleozoic
Trophic niche
Habitat-mobility
niche Detritivores Herbivores Carnivores
Table 4.15
Occupation of marine-niche ecospace in the Mesozoic
Trophic niche
Habitat-
mobility niche Detritivores Herbivores Carnivores
mobile plants? Might this odd ecological role be possible after all, if only
theoretically? In chapter 7 we will explore in more detail the implications
of theoretically possible, but nevertheless nonexistent, ecological roles
on Earth for the concept of convergent evolution.
The phenomenon of ecosystem convergence supports the idea that
there may be “universal constraints on the ecological roles” played in
those ecosystems (Carrano et al. 1999, 256), just as the phenomenon of
morphological convergence argues for evolutionary constraint on the
spectrum of possible forms available to animals and plants, as discussed
in chapters 2 and 3. In the case of ecosystems, the nature of these poten-
tial universal constraints is far from clear, and is part of the larger debate
concerning the degree of integration and structure of ecosystems them-
selves (for an overview of this larger question, see Maclaurin and Sterelny
2008, 106–131). The creation of theoretical ecospaces could add much
more rigor to the debate concerning ecosystem convergence by revealing
the total spectrum of what is ecologically possible, and what is not, in
potential ecosystems. The works of Pianka (1978), Bambach (1983), and
Bush et al. (2007) are first steps in this direction, an analytic approach
to theoretical morphospace that needs much more attention from
ecologists.
Thus far in this book, we have considered the phenomena of conver-
gence in individual animal and plant forms and, on a larger scale, in the
ecological roles of entire assemblages of animal and plant species. In the
next chapter we shall consider convergence on much smaller scales; that
is, in the very molecules that make up living animals and plants.
5 Convergent Molecules
Recent advances in protein chemistry suggest that at least one set of biological
forms—the basic protein folds—is determined by physical laws similar to those
giving rise to crystals and atoms. . . . If it does turn out that a substantial amount
of higher biological form is natural, then the implications will be radical and
far-reaching. It will mean that physical laws must have had a far greater role in
the evolution of biological form than is generally assumed . . . underlying all the
diversity of life is a finite set of natural forms that will recur over and over again
anywhere in the cosmos where there is carbon-based life.
—Denton and Marshall (2001, 417)
Convergent Molecules?
The source code for all life on Earth is contained in the deoxyribonu-
cleic-acid molecule, DNA. DNA codes for RNA molecules, ribonucleic
acids, and RNA codes for the assembly of amino acids into proteins,
the building blocks of life. Each of the essential molecules of life has a
finite number of possible states. Since DNA and RNA molecules contain
only four different nucleotides, the code for Earth life is a base-four
system. As such, the probability, p, of convergent molecular evolution of
the same nucleotide at the same site in two different DNA or RNA
molecules, via random mutation, is one in four ( p = 0.25 per site). Protein
molecules contain only 20 different amino acids; thus, the probability of
the convergent molecular evolution of the same amino acid at the same
site in two different protein molecules is one in twenty ( p = 0.05 per
site). Proteins are complex molecules, and the function of the protein is
often determined not only by its amino acid composition but also by the
complex manner in which the molecule is folded. Basic protein folds also
exist in a finite number of geometries, and the number of fold geometries
that are potentially functional are only a tiny subset of all possible folds
(Axe 2004). Denton and Marshall (2001) estimate that the number of
178 Chapter 5
DNA
the “age of complete genomes,” Castoe et al. (2010) point out that only
in the past five to seven years have substantial numbers of mitochondrial
genome data sets become available for the vertebrates, and that com-
plete nuclear genome data sets number only in the tens for these highly
derived animals (bacterial genomes are densely sampled, but these
organisms have ancient phylogenies and thus the number of changes
separating their genomes are large, making the detection of convergent
evolution difficult; Castoe et al. 2010).
The few data that do exist are startling. Vision in vertebrate animals
is due to the activation of the visual pigments in the eye by impacting
photons. The visual pigments are composed of opsin proteins and a
chromophore. Most vertebrates have a single type of chromophore, but
five groups of opsin proteins exist: rhodopsin (RH1) and RH1-like
(RH2), which are maximally activated by light in the blue-green part of
the spectrum (about 480 nm to 530 nm); short-wavelength-sensitive
opsin-1 (SWS1), which is maximally activated by violet-blue wavelengths
(about 410 nm to 490 nm); short-wavelength-sensitive opsin-2 (SWS2),
which is maximally activated by ultraviolet-violet wavelengths (about
355 nm to 440 nm); and a long-wavelength- to medium-wavelength-
sensitive opsin (LWS/MWS), which is maximally activated by green-red
wavelengths (about 490 nm to 570 nm; Zhang 2003; Bowmaker and Hunt
2006).
The DNA coding for the opsin proteins has been intensively investi-
gated, and early studies suggested that the evolution of red-sensitive
opsins in humans and some fish independently evolved from green-
sensitive opsins by the same nucleotide substitutions in the encoding
DNA (Yokoyama and Yokoyama 1990). Subsequent studies of the evolu-
tion of trichromatic color vision in primates revealed that the precise
same three substitutions in DNA nucleotides that coded for LWS/MWS
opsins independently evolved in Old World primates (Catarrhini) and
New World primates (Platyrrhini; table 5.1). Parallel evolution in the
LWS/MWS opsin amino acid sequences in both primate groups (Hunt
et al. 1998) resulted in the replacement of serine with alanine at site 180
(substituting G for T in the first position in the DNA codon for serine),
the replacement of tyrosine with phenylalanine at site 277 (substituting
T for A in the second position in the DNA codon for tyrosine), and the
replacement of threonine with alanine at site 285 (substituting G for A
in the first position in the DNA codon for threonine).
These DNA coding changes resulted in a 30-nanometer spectral shift
in the maximum activation of the LWS/MWS opsin, from 530 to 560
Table 5.1
Convergent evolution of identical nuclear DNA molecules coding for photosensitive opsin
proteins
nanometers. Hunt et al. (1998, 3299) suggest that these changes were
adaptive, the functional result of visual foraging for yellow and orange
fruits against a green foliage background, and that “the separate origin
of trichromacy in New and Old World primates would indicate that the
selection of these three [amino acid] sites is the result of convergent
evolution.”
The catarrhine and platyrrhine primates are closely related—both are
members of the simiiform clade (table 5.1)—and the two lineages
diverged only some 35 million years ago in the Eocene (Benton 2005).
One could thus argue that these two lineages demonstrate parallel
molecular evolution; that is, independent parallel nucleotide substitu-
tions in similar DNA molecules. In contrast, the marsupial mammals and
placental mammals diverged some 125 million years ago, in the Early
Cretaceous (Benton 2005). Yet even in these two ancient, independent
lineages of mammals, convergent evolution in opsin-encoding DNA has
been demonstrated (table 5.1).
Deeb et al. (2003) report that the identical nucleotide substitution in
DNA coding for the SWS1 opsin amino acid sequence occurred inde-
pendently in the marsupial tammar wallaby and in the placental cow:
replacement of phenylalanine by tyrosine at site 86 (substituting A for
T in the second position in the DNA codon for phenylalanine). This
DNA coding change resulted in a 66-nanometer spectral shift in the
maximum activation of the SWS1 opsin, from 358 to 424 nanometers.
The tammar wallaby grazes on grass and low tree branches in the late
afternoon, and Deeb et al. (2003) suggest that the convergent evolution
in SWS1-encoding DNA observed in it and in the placental grazing
ungulates enables the animals to visually distinguish features in grasses
with differing shades of green.
Primates are unique among mammals in having trichromatic color
vision, in that most mammals have only dichromatic vision. In contrast,
some fish, reptiles, and birds have tetrachromatic vision, and may be
able to see twice the number of colors that even trichromatic primates
can see (Ödeen and Håstad 2003). Many of these same animals can
see ultraviolet light, which has much shorter wavelengths than the
400-nanometer violet light that we are able to see. Ödeen and Håstad
(2003) have demonstrated that ultraviolet vision evolved four times inde-
pendently in the birds (table 5.1) by the identical nucleotide substitution
in DNA coding for the SWS1 opsin amino acid sequence: replacement
of serine by cysteine at site 90 (substituting G for C in the second posi-
tion in the DNA codon for serine). This DNA coding change results in
182 Chapter 5
Table 5.2
Convergent evolution of identical nuclear DNA and mitochondrial DNA molecules coding
for digestive proteins
Table 5.3
Convergent evolution of identical nuclear DNA molecules coding for aldehyde oxidore-
ductase enzymes
Table 5.4
Convergent evolution of identical antifreeze protein molecule in independent lineages, and
of different molecules that serve the same antifreeze molecular function
are nearly identical in amino acid composition and are comprised mainly
of threonine-alanine-alanine repeats, usually encoded by ACA-GCT/G-
GCA codons in the Antarctic notothenioids, whereas the Arctic cod use
DNA codons rarely observed in the notothenioid fish (Logsdon and
Doolittle 1997). The convergent amino acid triplet-repeating structure of
AFGP appears to be required for antifreeze-ice interaction to occur, and
probably evolved in the notothenioids about 10 to 14 million years ago,
as the Antarctic Ocean began to freeze, and about two to three million
years ago in the cod, during the Arctic glaciation (Chen et al. 1997;
Logsden and Doolittle 1997).
Subsequent work has revealed the convergent evolution of yet another
antifreeze molecule: AFP Type II. This molecule evolved independently
in cold-water clupeomorph fish (Atlantic herring) and in two clades of
euteleostian fishes, the protocanthopterygian American smelts and the
acanthopterygian sea ravens (table 5.4). Unlike in the case of AFGP, the
convergent evolution of AFP Type II in these three fish groups appears
to have been a case of parallel independent changes in the same C-type
lectin genes ancestrally found in all three groups (Fletcher et al. 2001).
Convergence in antifreeze protein function does not always produce
convergent evolution, or parallel evolution, of the same antifreeze mol-
ecules, as in the case AFGP or AFP Type II. Consider the convergent
evolution of wings in insects, pterosaurs, birds, and bats (see table 2.2),
all of which serve the common function of generating lift for powered
flight. The structure of the wing is different in each animal group, but the
function is the same. Fletcher et al. (2001) point out that three additional
antifreeze proteins have evolved in the cold-water fishes—AFP Types I,
III, and IIV—and Davies et al. (2002) discuss the evolution of two more
antifreeze proteins that have evolved in the arthropods, AFP Cf and Tm
(table 5.4).
The amino acid sequences in these additional antifreeze proteins are
different, but aspects of their geometries are very similar (Davies et al.
2002), and their functions are the same. In particular, Davies et al. (2002,
932) point out that the helical geometries of AFP Cf and AFP Tm are
such that “their ice-binding sites are virtually superimposable” even
though AFP Cf is a left-coiled β-helix and AFP Tm is a right-coiled β-
helix. Davies et al. (2002, 932) describe this geometric result as a “remark-
able example of convergent evolution” in that the “two ranks of threonine
residues line up perfectly because the threonines are in the same rota-
meric configuration” in the helices.
192 Chapter 5
Table 5.5
Convergent evolution of similar oxygen-binding protein molecules in independent lin-
eages, and of different molecules that serve the same oxygen-binding molecular function
forms of anaerobic life (Bailly et al. 2008). With the evolution of aerobic
life, transport of oxygen within the organism for metabolic purposes is a
critical function, and van Holde et al. (2001, 15566) hypothesize that
oxygen-transport proteins developed “in several independent ways,
hemoglobins from myoglobins, hemerythrins from myohemerythrins,
and the two kinds of hemocyanins from two different classes of phenol
oxidases.”
Hemocyanins evolved twice, independently, in the clade of the mol-
luscs (lophotrochozoans) and the clade of the arthropods (ecdysozoans;
table 5.5). In the molluscs, Class 3a hemocyanins are used for oxygen
transport in the blood of cephalopods, such as the octopus, and of gas-
tropods, including both marine and terrestrial snails. In the arthropods,
Class 3b hemocyanins are present in the blood of cheliceriformes, such
as the horseshoe crab, and mandibulate crustaceans, such as the true
crabs (van Holde et al. 2001).
The hemocyanins bind oxygen with copper (thus, the blood of an
octopus is bluish-green when it is oxygenated, not red like ours), but
copper used as an oxygen-binding metal is not as efficient as iron. Both
hemoglobin and hemerythrin use iron, but the latter is a nonheme protein
(despite its deceptive name) unlike hemocyanin and hemoglobin, and is
less efficient in oxygen transport than hemoglobin. The hemerythrins are
patchily present throughout life, from bacteria to archaea to the eukarya,
and are used in a variety of functions, the most ancient of which is oxygen
sensing (Bailly et al. 2008). Only late in the evolution of life have three
groups of protostome invertebrates—the brachiopods, serpulids, and pri-
apulids (table 5.5)—independently co-opted the hemerythrin protein for
the primary function of oxygen transport. Blood sufficiently enriched in
hemerythrin is pale violet when oxygenated, not blue-green as for hemo-
cyanin or red as for hemoglobin.
Last, mammals possess a specialized plasma protein, haptoglobin, that
can be used to inhibit hemoglobin’s oxidative activity, if needed. A
variant of this protein, two-CCP-domain haptoglobin, independently
evolved in the ruminant mammals and, curiously, in humans (table 5.5).
Because the two-CCP-domain haptoglobin is found throughout the
clade of the ruminants, the mutation producing this protein occurred
early in the Cenozoic for these mammals (Wicher and Fries 2007). The
same protein evolved much more recently in humans, for it is not present
in our close primate cousins. In both cows and humans, the convergent
duplication of the gene segment coding for the CCP domain of the α-
chain in the haptoglobin molecule resulted in a much longer α-chain in
194 Chapter 5
the new protein, which contains two CCP domains in tandem. Wicher
and Fries (2007) point out that the new protein can form much higher-
molecular-weight oligomers than the conventional protein, and suggest
that the different binding properties of this molecule is useful in prevent-
ing infection in the ruminant mammals. Its function in humans is less
clear, but data do exist that indicate humans with the mutated haptoglo-
bin molecule are more resistant to life-threatening streptococcus infec-
tions, and possibly to malarial infections (Wicher and Fries 2007).
Preventing tissue death by freezing or oxygen starvation is a matter
of life or death, yet convergent molecular evolution also occurs in situ-
ations that are not so dire. Mundy (2005) reviews the recent discovery
that melanic polymorphism in three groups of distantly related wild birds
has evolved in parallel, in that the plumage color variations seen in these
birds are produced by point mutations in amino acid substitution in the
same gene coding for the MC1R protein (table 5.6). Many birds are well
known for the dazzling colors of their plumage, and so it was unexpected
to find parallel changes in the same gene to be producing the color varia-
tions in such distantly related species. Mundy (2005, 1638) concludes by
noting that the “examples discussed here clearly only represent the tip
of the iceberg as far as avian plumage colouration genetics in the wild is
concerned. Over 300 avian species exhibit plumage polymorphisms”;
thus, he implies that more examples of parallel evolution of the MC1R
protein may soon be discovered.
Likewise, Protas et al. (2005) have shown that mutations in the
protein-encoding sequences of the same gene, the one coding for
Table 5.6
Convergent evolution of similar melanin-controlling protein molecules
1 Convergent protein and function: MC1R (protein producing plumage color variations
in birds)
Convergent lineages:
1.1 Snow goose (Aves: Neognathae: Galloanserae: Anseriformes: Anatidae; Anser
caerulescens)
1.2 Bananaquit (Aves: Neognathae: Neoaves: Passeriformes: Coerebidae; Coereba
flaveola)
1.3 Arctic skua (Aves: Neognathae: Neoaves: Charadriiformes: Stercorariidae;
Stercorarius parasiticus)
2 Convergent protein and function: OCA2 (protein producing albinism in fish)
Convergent lineages:
2.1 Molino cave populations of the Mexican tetra, Astyanax mexicanus
2.2 Pachón cave populations of the Mexican tetra, Astyanax mexicanus
2.3 Japonés cave populations of the Mexican tetra, Astyanax mexicanus
Table 5.7
Convergent evolution of the EDA protein in independent freshwater populations of the
three-spine stickleback fish species Gasterosteus aculeatus
Convergent protein and function: EDA (protein used in constructing rows of bony armor
plates on the bodies of stickleback fish)
Convergent populations:
A European populations
1 Schwale (Germany) freshwater G. aculeatus
2 Loch Fada (Scotland) freshwater G. aculeatus
3 Norway stream (Norway) freshwater G. aculeatus
4 Blautaver (Iceland) freshwater G. aculeatus
B Eastern North American population
5 Olmstead Park (Massachusetts) freshwater G. aculeatus
C Western North American populations
6 Santa Clara River (California) freshwater G. aculeatus
7 Friant (California) freshwater G. aculeatus
8 San Francisco Creek (California) freshwater G. aculeatus
9 Conner Creek (Washington) freshwater G. aculeatus
10 Paxton Lake (Canada) freshwater G. aculeatus
11 Salmon River (Canada) freshwater G. aculeatus
12 Wallace Lake (Alaska) freshwater G. aculeatus
13 Alaska stream (Alaska) freshwater G. aculeatus
D Japanese population
14 Nakagawa Creek (Japan) freshwater G. aculeatus
Table 5.8
Convergent evolution of similar structural protein molecules
whereas the insect protein is secreted in the midge’s saliva, and is used
in constructing external tube walls within which the insect lives. Rey et
al. (1998, 6215) note that the two genes coding for the two proteins, one
fungal and one insect, had to have originated independently of each
other, and that the “ancestor of each new gene created de novo has to
be a noncoding DNA sequence that abounds in every eukaryotic
genome”; furthermore, the observed protein convergence “was not
between the genes but between the similar noncoding repeats whence
the two [new] genes sprung quite independently of each other.”
More enigmatic is the convergent evolution of the protein
apolipoprotein(a), or apo(a), in the hedgehog and in humans (table 5.8).
Apo(a) is part of the low-density lipoprotein complex (LDL) that is a
known risk factor in the development of atherosclerosis in humans,
hence its research interest in medicine. It is known that apo(a) is found
in Old World primates (Catarrhini), but it was a surprise to find the same
protein in the very distantly related European hedgehogs: Lawn et al.
(1997, 11992) conclude that it is “a remarkable example of ‘parallel gene
evolution.’ . . . By apparent remodeling of a plasminogen-like gene,
hedgehog and human ancestors independently evolved an apo(a)
protein” that functions identically in both organisms. Unfortunately,
some of those functions increase the risk of cerebral stroke in humans,
which is one of the reasons the protein was under study in the first place.
Last, we have the convergent evolution of protein function in which
multiple independent proteins with their own independent functions are
co-opted to serve the convergent protein function, a process known as
gene sharing. Convergent evolution of protein function by gene sharing
was first discovered in the crystallin proteins of the eye lens in animals:
“The term ‘gene sharing’ was used initially to generalize the finding that
crystallins with a structural, optical function in the lens may also be
expressed in other tissues [in the body], where they have a metabolic,
nonrefractive function. Thus, the protein encoded in the identical gene
may perform entirely different functions depending upon its expression
pattern . . . gene regulation is a critical factor for gene sharing, leading
to the use of a metabolic [enzyme] or stress protein as a structural
crystallin. It is ironic that the abundantly expressed crystallins, long
considered inert bricks as it were, squirreled away in a tiny transparent
tissue . . . have become landmarks for multifunctional uses of widely
expressed proteins that illustrate the dynamism of evolutionary change!”
(Piatigorsky 2007, 55–56; author’s emphasis).
Convergent Molecules 199
Table 5.9
Convergent evolution of crystallin protein molecules by gene sharing
Convergent molecular function: Construction of a transparent eye lens that optimizes the
refractive index for focusing and image formation
Convergent molecules:
1 J1-, J2-, J3-CRYSTALLINS
Lineage: Box jellyfish (Metazoa: Cnidaria: Cubozoa: Carybdeidae; Tripedalia
cystophora)
2 Ω/L-CRYSTALLIN
Lineage: Sea scallop (Metazoa: Bilateria: Protostomia: Lophotrochozoa: Mollusca:
Bivalvia: Pectinidae; Placopecten magellanicus)
3 SL11/Lops4-, S-CRYSTALLINS
Lineage: Octopus (Protostomia: Lophotrochozoa: Mollusca: Cephalopoda: Octopodidae;
Octopus vulgaris)
4 DROSOCRYSTALLIN
Lineage: Fruit fly (Protostomia: Ecdysozoa: Arthropoda: Hexapoda: Drosophilidae;
Drosophila melanogaster)
5 Ci-βγ-CRYSTALLIN
Lineage: Sea squirt (Bilateria: Deuterostomia: Urochordata: Ascidiacea: Cionidae;
Ciona intestinalis)
6 τ-CRYSTALLIN
Lineage: Sea lamprey (Deuterostomia: Chordata: Vertebrata: Petromyzontiformes:
Petromyzontidae; Petromyzon marinus)
7 ρ-CRYSTALLIN
Lineage: Leopard frog (Chordata: Osteichthyes: Sarcopterygii: Tetrapoda:
Batrachomorpha: Lissamphibia: Anura: Ranidae; Rana pipiens)
8 π-, ι-, ρB-CRYSTALLINS
Lineage: Striped day gecko (Tetrapoda: Reptiliomorpha: Amniota: Sauropsida:
Lepidosauromorpha: Scleroglossa: Gekkonidae; Phelsuma lineata)
9 ε-CRYSTALLIN
Lineage: Crocodile (Sauropsida: Archosauromorpha: Crurotarsi: Crocodylidae;
Crocodylus niloticus)
10 δ1-, δ2-CRYSTALLINS
Lineage: Chicken (Sauropsida: Archosauromorpha: Ornithodira: Dinosauria: Saurischia:
Theropoda: Maniraptora: Aves: Galliformes: Phasianidae; Gallus gallus)
11 υ-CRYSTALLIN
Lineage: Platypus (Amniota: Synapsida: Therapsida: Mammalia: Monotremata:
Ornithorhynchidae; Ornithorhynchus anatinus)
12 μ-CRYSTALLIN
Lineage: Kangaroo (Mammalia: Marsupialia: Diprotodontia: Macropodidae; Macropus
rufus)
13 η-CRYSTALLIN
Lineage: Elephant shrew (Mammalia: Eutheria: Afrotheria: Macroscelidea:
Macroscelididae; Elephantulus rozeti)
14 ζ-CRYSTALLIN
Lineage: Camel (Mammalia: Eutheria: Laurasiatheria: Cetartiodactyla: Camelidae;
Camelus dromedarius)
15 λ-CRYSTALLIN
Lineage: Rabbit (Mammalia: Eutheria: Euarchontoglires: Lagomorpha: Leporidae;
Oryctolagus cuniculus)
Table 5.10
Convergent evolution of different enzyme molecules that serve the same metabolic
catalytic function
Table 5.11
Convergent evolution of different enzyme molecules that serve the same catalytic
function
present in these organisms. The plant and animal chitinase proteins are
very different in structure, and the chitin binding domains (CBDs) of the
two enzymes do not share significant amino acid sequence similarity;
nevertheless, the fact that “plant and invertebrate CBDs share a common
core structure and chitin-binding mechanism while being unrelated in
evolution, argue[s] in favor of convergent evolution” of these molecules
(Shen and Jacobs-Lorena 1999, 343).
Last, convergent evolution also exists in enzymatic functions involving
RNA molecules, as well as the proteins. Within the cell, the aminoacyl-
tRNA synthetases are responsible for the synthesis of the substrates for
204 Chapter 5
Table 5.12
Convergent evolution of different enzyme molecules that serve the same RNA catalytic
function
Table 5.13
Convergent evolution of the same ribozyme molecular structure
1.0” per minute (Salehi-Ashtiani and Szostak 2001, 82). They argue that
their “results suggest that the evolutionary process may have been chan-
nelled . . . towards repeated selection of the simplest solution to a bio-
chemical problem” (Salehi-Ashtiani and Szostak 2001, 82), and that
“purely chemical constraints (that is, the ability of only certain sequences
to carry out particular functions) can lead to the repeated evolution
of the same macromolecular structures” (Salehi-Ashtiani and Szostak
2001, 84).
In summary, we have examined examples of (1) the convergent evolu-
tion of the same enzyme function produced by the convergent evolution
of the same protein producing that function, (2) the convergent evolu-
tion of the same catalytic core structure in enzyme molecules that have
no similarity in amino acid sequence or in the three-dimensional struc-
ture of the molecules, and (3) the convergent evolution of the same
macromolecular structure in unrelated enzymes. Enzymes converge in
every possible way.
the cell (Denton et al. 2002, 339). Second is cell form itself which, when
experimentally altered by microsurgery, can “recover its proper form by
searching a conformational space for its correct conformation, just like
a folding protein” (Denton et al. 2002, 340). Therefore, like the molecules
composing them, the macromolecular structures of the cell itself
converge.
West et al. (1999) assert that biological convergence in form due to
the laws of physics extends upward from the cell to entire tissue systems,
such as capillary, intestinal, and lung geometries and surface areas in
animals and root, branch, and leaf geometries and surface areas in
plants. They note that biological form follows an allometric quarter-
power scaling, rather than the expected third-power from Euclidean
geometric scaling, and argue that this follows from the fractal-like archi-
tecture of all vascular branching systems in living organisms: “The vast
majority of organisms exhibit scaling exponents very close to 3/4 for
metabolic rate and 1/4 for internal times and distances. These are the
maximal and minimal values, respectively, for the effective surface
area and linear dimensions for a volume-filling fractal-like network . . .
it is testimony to the severe geometric and physical constraints on meta-
bolic processes, which have dictated that all of these organisms obey a
common set of quarter-power scaling laws” (West et al. 1999, 1679). They
continue: “[u]nlike the genetic code, which has evolved only once in the
history of life, fractal-like distribution networks that confer an additional
effective fourth dimension have originated many times” (West et al. 1999,
1679).
In summary, convergent evolution occurs across the entire spectrum
of molecules that make up life. We have studied examples of the conver-
gent evolution of identical nucleotide substitutions in nuclear and
mtDNA molecules of distantly related organisms, similar amino acid
sequences in unrelated protein molecules, similar structural geometries
in proteins with different amino acid sequences, and similar protein func-
tions by gene sharing; the convergent evolution of the same enzyme
function produced by the convergent evolution of the same protein
producing that function; the same catalytic core structure in enzyme
molecules that have no similarity in amino acid sequence or in the three-
dimensional structure of the molecules; and the convergent evolution of
the same macromolecular structure in unrelated enzymes. The effects of
functional constraint can be seen throughout molecular evolution, in that
selection for a given function repeatedly produces convergent molecular
compositions and/or molecular geometries. The number of functional
Convergent Molecules 207
Convergent Minds?
Problem-Solving Behavior
It was long thought that the tool was the unique invention of the human
mind. Only the human mind had the ability to understand the limits to
210 Chapter 6
Table 6.1
Convergent evolution of tool-using and tool-making behavior
1 Convergent behavior and function: TOOL USAGE (using an existing external object as
a tool to attain an immediate goal)
Convergent lineages:
1.1 Black-and-yellow mud wasp (Bilateria: Protostomia: Ecdysozoa: Arthropoda:
Mandibulata: Hexapoda: Hymenoptera:Vespoidea: Sphecidae; Sceliphron caementarium)
1.2 Forest ant (Arthropoda: Mandibulata: Hexapoda: Hymenoptera: Formicidae:
Myrmicinae; Aphaenogaster rudis)
1.3 Woodpecker finch (Bilateria: Deuterostomia: Chordata: Osteichthyes: Sarcopterygii:
Reptiliomorpha: Amniota: Sauropsida: Archosauromorpha: Dinosauria: Saurischia:
Theropoda: Maniraptora: Aves: Neognathae: Neoaves: Passeriformes: Emberizidae;
Cactospiza pallida)
1.4 Egyptian vulture (Aves: Neognathae: Neoaves: Falconiformes: Accipitridae;
Neophron percnopterus)
1.5 Asian elephant (Amniota: Synapsida: Therapsida: Mammalia: Eutheria: Afrotheria:
Proboscidea: Elephantidae; Elephas maximus)
1.6 Sea otter (Mammalia: Eutheria: Laurasiatheria: Carnivora: Mustellidae; Enhydra
lutris)
1.7 Polar bear (Mammalia: Eutheria: Laurasiatheria: Carnivora: Ursidae; Ursus
maritimus)
1.8 Bottlenose dolphin (Mammalia: Eutheria: Laurasiatheria: Cetartiodactyla: Cetacea:
Odontoceti: Delphinidae; Tursiops truncatus)
1.9 Black-striped capuchin monkey (Mammalia: Eutheria: Euarchontoglires: Primates:
Platyrrhini: Cebidae; Cebus libidinosus)
1.10 Yellow baboon (Mammalia: Eutheria: Euarchontoglires: Primates: Catarrhini:
Cercopithecoidea: Cercopithecidae; Papio cynocephalus)
2 Convergent behavior and function: TOOL CONSTRUCTION (modifying an existing
external object or objects to create a tool to attain an immediate goal)
Convergent lineages:
2.1 New Caledonian crow (Amniota: Sauropsida: Archosauromorpha: Dinosauria:
Saurischia: Theropoda: Maniraptora: Aves: Neognathae: Neoaves: Passeriformes:
Corvidae; Corvus moneduloides)
2.2 Orangutan (Amniota: Synapsida:Therapsida: Mammalia: Eutheria: Euarchontoglires:
Primates: Catarrhini: Hominoidea: Hominoidae: Pongidae; Pongo pygmaeus)
to protect the eggs from predators (Baber 2003). Other insects, such as
the forest ant, collect and use leaves as containers in which to carry food.
In chapter 4 we considered the case of the finch that has converged on
the ecological role of the woodpecker (see table 4.3). The woodpecker
finch searches for and carries away a tool, a cactus spine, which it uses
to extract wood-boring insects. And last, the mighty elephant has been
routinely observed to carry and use branches to scratch itself (Baber
2003), and the ocean-dwelling bottlenose dolphin collects and uses
sponges to protect its skin from abrasion while foraging on the sea
bottom (Hansell and Ruxton 2008). Animals with extremely different
brains, scattered across very distantly related phylogenetic lineages, have
independently evolved tool-using behavior (table 6.1).
Taking a different perspective, Hansell and Ruxton (2008, 77) argue
that entirely too much emphasis has been placed on tool behavior in
animals because it is always interpreted in terms of human evolution:
“Those who feel that tools are ‘special’ might be correct—but they might
be more special to researchers than to the animals that use them.” As
they point out, a great deal of attention was given to the discovery that
gorillas make tools, while at the same time researchers ignored the fact
that these animals build complicated nest structures out of modified
branches and twigs on a daily basis. Likewise, relatively few birds make
or use tools (table 6.1), but the majority of birds build nests, some of
which are structurally very complex.
Before considering the convergent evolution of architectural behavior
in animals, let us briefly review the evolution of architectural behavior
in the most diversified animal architect on the planet at present, namely
humans. Our ancestors, like our living cousins the chimpanzees and
bonobos, surely built nests in trees, habitats constructed to provide a safe
and comfortable place to sleep, high above the ground and thus out of
the reach of ground-dwelling predators. In contrast, many of our ground-
dwelling ancestors looked for preexisting natural structures, such as
caves, that could be used as habitats. Caves and other naturally occurring
habitats are rare, and with further growth in human population numbers,
humans took the next step: the construction of an artificial habitat on
the ground. Two potential construction materials are readily available:
vegetative matter, such as wood, and dirt. Humans began to build habi-
tats made of modified wood, an architectural behavior still widely in use
today, or with adobe, which many people also still use today. Only later
did we begin to build habitats with cut stone, fired bricks (which are
much more durable than adobe), and concrete (artificial stone). At
Convergent Minds 213
present, we have added metal, glass, and plastic to the construction mate-
rials used for our habitats.
Other animals have evolved the exact same sequence of architectural
behavior, and they did so long before we existed (table 6.2). Hansell
(2005, 33) argues that “[a]rchitecture results from the application of
behaviour to materials”; for this reason, I have omitted the behavioral
step where animal species simply hunt for preexisting structures that can
be used for habitats, like hermit crabs that use snail shells, and will move
directly to animals that actively gather and modify material to construct
a habitat. Like humans, other animal species construct habitats with
material that is either vegetative or earthen. Unlike humans, however,
animals have not evolved the capability of working with materials like
metal, glass, or plastic in their construction, although some do steal and
inventively use our worked materials in their habitats!
Hansell (2005) identifies five architectural behaviors found in nest-
building birds, listed here in terms of increasing behavioral complexity:
stacking, entangling, Velcro-fastening, stitching, and weaving. At present
there exists no comprehensive phylogenetic analysis of how many inde-
pendent times each of these behaviors have evolved in separate groups
of birds. Thus, I will discuss as many convergent examples that I am aware
of, knowing that future phylogenetic analyses will almost certainly
expand that list.
The eagle (table 6.2) is an example of a bird that constructs its nest by
first searching for and collecting branches of the desired length and
diameter, and then transporting those branches back to the tree it has
chosen for the nest. The eagle simply stacks the branches atop one
another until a bowl-shaped nest has been created. This same nest-
building architectural behavior is used by some ground-nesting birds as
well, such as the swan (table 6.2). In general, many of the large tree-
dwelling or ground-dwelling birds use stacking to build their nests: storks
(Cicioniidae), pelicans (Pelicanidae), and herons (Ardeidae), in addition
to eagles and swans (Hansell 2005).
Nests built by entangling architectural behavior are sturdier than those
built by simple stacking, as the ends of the branches or twigs in these
nests are actively bent and pulled back into the core of the nest by the
bird. The familiar wood pigeon (table 6.2) is an example of a bird that
constructs its nest using entangling (Hansell 2005). The same method is
used by the gorilla (table 6.2) and chimpanzee (and presumably many
of our ancestors), who search out a spot in the trees with an upright fork
and then bend branches back into this foundation while holding them in
214 Chapter 6
Table 6.2
Convergent evolution of architectural behaviors
Table 6.2
(continued)
place with their feet; finally, they tuck leafy twigs growing around the rim
of the nest back into the core of the nest (von Frisch 1974). Unlike birds,
however, apes construct their sleeping nests on a daily basis.
Many plants have convergently evolved seeds with the tiny barbs and
hooks that provided the inspiration for the modern fastening product
Velcro (see table 3.15). Some birds use this same fastening principle in
building their nests: they actively seek out materials with tiny projecting
processes and position these in the nest so that they ensnare each other,
thereby creating even sturdier nests than those built using entangling.
The long-tailed tit (table 6.2) uses small-leaved mosses for Velcro, and
the related bushtit (Psaltiparus minimus) uses lichen (Hansell 2005).
These nests are so sturdy that they can be totally enclosed into a baglike
or socklike nest, with an opening at one end, as seen in the nests built
by European penduline tits (table 6.2; von Frisch 1974). Rather than
216 Chapter 6
grams (Hansell 2005). Gould and Gould (2007, 184) comment: “The
invention of adobe cups that then develop into enclosed cavities, so
evident in the progression from barn to cliff swallows, is also seen in
some birds that build in trees.” The most striking example of a tree-
dwelling adobe builder is the rufous hornero of South America (table
6.2), nicknamed the “ovenbird” because its vertical adobe nest looks like
a baker’s oven (von Frisch 1974). The globe-shaped nest contains a large
interior room that is used as a brooding chamber and a narrow, hall-like
entrance passage that opens into a ten-centimeter-diameter hole to the
outside. Both the male and female birds cooperate in its construction,
which takes them over two weeks to complete (von Frisch 1974).
Wasps have evolved a similar behavioral pathway in their habitat
constructions, as noted by Hansell (2005, 232): “I have identified two
alterations in nest building behaviour that are associated with the transi-
tion from solitary to social behaviour of wasps. . . . The first is the transi-
tion from an excavated to a constructed nest. . . . This occurred in both
the Sphecidae and Vespidae (Eumeninae). The second change was in the
choice of the building material from mud to paper.” Both paper-building
behavior and adobe-building behavior is convergent in wasps (table 6.2).
An interesting transition between excavating tunnels for below-ground
habitats to constructing above-ground adobe habitats is seen in the
eumenin funnel wasp (Paralastor emarginatus) of Australia. This wasp
first digs a tunnel, lines it with mud, and then extends these mud walls
above ground to build an external cylinder, which it increases in diameter
to build a downward-pointing funnel opening (Gould and Gould 2007).
The funnel serves the purpose of concealing the entrance to the tunnel
from parasites while the female is hunting for caterpillars to paralyze
and place in the tunnel as food for her later-developing larvae. Solitary
sphecid wasps, such as the Neotropical mud wasp (table 6.2), do no
excavating but build adobe brood chambers entirely above ground.
While adobe building is the norm for many sphecid wasps, the eusocial
Malasian hover wasp (Stenogastrinae) also builds with adobe, as does
the Ceylonese potter wasp (Eumeninae) within a subfamily of wasps that
normally builds with paper (table 6.2; Hansell 2005).
Termites also have converged on the extension of their below-ground
excavations to above-ground adobe-building behavior (table 6.2). Com-
pared to the size of their inhabitants, termite adobe dwellings are huge:
the Australian compass termite, Amitermes meridionalis, builds above-
ground nests that are up to 5 meters high and 3 meters long, but only
around 30 centimeters thick. The species is given the name “compass”
220 Chapter 6
because its slablike buildings are always oriented with the long direction
north-south, such that the habitat always has a broad side warmed by
the sun (either to the east or the west, depending upon the time of day),
and an opposite side that is cool and shaded (von Frisch 1974). The
temperature difference between the two broad sides of the habitat pro-
duces air flow from the cool side of the habitat through to the warm
side—the termites have invented buildings with air conditioning! And
not just one type of air-conditioning system—the much-studied African
warlike termite (table 6.2) builds five-meter-high pyramidal-shaped
dwellings with hard, cement-like adobe outer walls and an interior adobe
nest that is suspended within the outer walls by a series of pillars, arches,
and buttresses. The inner dwelling is surrounded by an air space, through
which warm air rises from the bottom of the building, creating a lower-
pressure region that then draws cooler, denser air into the interior of the
building from the outside. The source of heat at the bottom of the dwell-
ing is the fermentation in fungus gardens arranged in special adobe
chambers—the termites have thus also invented agriculture (a point that
we shall consider in more detail below). A large habitat of the warlike
termite can contain as many as two million termites. The various air-
conditioning systems that termites build also serve the purpose of flush-
ing the carbon-dioxide-rich air created by their breathing out the top of
the building; oxygen-enriched air is drawn in through the ventilation
holes in the side of the building (von Frisch 1974).
Australian compass termites and African warlike termites inhabit gen-
erally arid regions; in wet tropical regions, termites change the architec-
ture of their dwellings to deal with the problem of frequent rainfall. The
termite species Cubitermes fungifaber builds tall, cylindrical-shaped
adobe dwellings with multiple roofs that are slanted downward and
extend out beyond the diameter of the main habitat (Gould and Gould
2007). These multiroofed towers are reminiscent of the pagodas build by
humans in Asia.
Last, beavers have evolved diversified architectural behaviors that
converge on those of humans: they build with logs, adobe, and stone. The
beavers are the largest rodents alive today, and may weigh up to 30
kilograms (von Frisch 1974). The North American beaver habitat, or
lodge, is constructed of both vegetative and earthen material; it is archi-
tecturally very similar to the log cabins constructed by early European
settlers in North America (table 6.2). The beavers first build a domelike
habitat with cut tree branches and logs, and then carefully plaster and
caulk the openings between the branches in the walls of the lodge with
Convergent Minds 221
mud and clay. Unlike a log cabin, however, the door to the lodge is under
water, and the lodge itself is located on an artificially created island in a
pond or lake. Both of these features serve to protect the beavers’ dwell-
ing from invasion by predators. These round dome-shaped dwellings,
isolated from the shore out in a lake, are reminiscent of the crannogs
built by prehistoric humans in Scotland.
The pond or lake that surrounds the beavers’ lodge is often the cre-
ation of the beavers themselves: “beavers are experts not only in the
building of dwellings but also in hydro-engineering, and have performed
tremendous feats in this line long before man attempted anything of the
kind” (von Frisch 1974, 268). Blocking a flowing stream or river with a
dam is not an easy task, yet beavers do so by first ramming cut branches
and logs into the river bed, supporting these against the flow of the river
with forked branches pointing upstream, and adding further anchoring
by bringing heavy stones to hold the dam superstructure in place. They
further excavate the river bed on the upstream side of the dam, which
not only reduces the speed of water flow but also provides the beavers
with the earthen material they need to finish sealing the dam. The fin-
ished dam is higher in the center than at the termini, thus forcing the
dammed river to flow over the dam near the two river banks. The beavers
then actively control the level of the lake by either lowering or raising
these two overflow sluices—they frequently lower the level of the lake
during winter to created a breathing space under the ice of the frozen
surface of the lake, an air space that extends the entire area of the lake,
providing them with a swimming region protected from the winter cold.
Von Frisch (1974) reports that the largest beaver dam on record was 700
meters long, and was strong enough to support the weight of human
riders on horseback.
Do convergent architectures reveal convergent minds? Gould and
Gould (2007, 271–272) argue that “[m]ental activity is, by its nature,
private; what goes on in the brain has to be inferred. In tracing the evolu-
tion of cognitive strategies, the most tangible evidence is found among
animals that build. . . . These abilities seem to have evolved independently
in several different groups, but always apparently in about the same
order, and to serve analogous ends.” That is, given the same problem—
such as building a protective habitat—animals with radically different
brains have evolved the same architectural behaviors. Those analogous
behaviors thus reflect analogous mental activities and cognitive strate-
gies taking place in independent lineages, which is convergent mental
evolution.
222 Chapter 6
On the other hand, do humans really think like termites when they
build an adobe habitat? Gould and Gould (2007, 5) also note that
“[w]e know that animal building, like most tasks animals accomplish,
depends on many different neural mechanisms. For most creatures,
though, instinct rather than learning seems to guide behavior.” This con-
clusion was revealed in the pioneering work with animal instinctive
behavior by Konrad Lorenz, Niko Tinbergen, and Karl von Frisch, work
for which they were awarded the 1973 Nobel prize in physiology or
medicine. Behavioral evolution by the trial-and-error process of natural
selection requires very long periods of time and generations of reproduc-
tive cycles. Behavioral evolution by learning, in contrast, is very rapid:
one generation simply teaches the next generation the new behavioral
innovations discovered by the parent generation. Cognitive strategies are
transmitted by teachers, not by genes. Yet stereotypic instinctive behav-
iors and flexible learned behaviors have both converged on the same
result in architecture (table 6.2).
The discipline of theoretical morphology (McGhee 2007) takes a dif-
ferent perspective on the question of convergent architectural behavior
by examining the geometries of the architecture itself: Are there only a
limited number of ways to build structures? If so, convergent behavior
must necessarily result because independent lineages of organisms have
a limited number of building options to discover. The convergent usage
of the cantilever and the arch by such disparate organisms as mud wasps,
termites, cliff swallows, and humans results from the fact that cantilevers
and arches are necessary in the construction of adobe architectures.
Convergent minds are here the product of the functional constraints of
geometry. The implications of this theoretical morphologic perspective
on convergent evolution will be explored in more detail in chapters 7
and 8, but I would point out here that we should not be surprised to find
cantilevers and arches on Earth-like worlds elsewhere in the universe,
produced by totally alien minds.
To conclude this section of the chapter, let us consider one last
extremely complex problem-solving behavior: agriculture. Modern
humans, Homo sapiens, are the only vertebrate species that has evolved
agriculture. Agricultural behavior is convergent, however, in that several
populations of humans made the transition from hunter-gatherers to
agriculturalists independently of each other around the world about
10,000 years ago: in the Tigris and Euphrates valleys in the Near East, in
South Asia, in Central Asia, and in Central America (Gupta 2004).
Convergent Minds 223
Was agriculture clearly an idea whose time had come, beginning about
10,000 years ago, as a unique behavioral trait of the human species? Not
so. In contrast to the single human species, the hexapod arthropods have
independently evolved agriculture nine separate times in different lin-
eages (table 6.3). The transition from hunter-gatherer to agriculturalist
was made once by the ants, once by the termites, and no less than seven
separate times by the ambrosia beetles (Farrell et al. 2001; Mueller et al.
2005). And they did so long before humans existed: the platypodine
ambrosia beetles evolved agricultural behavior 60 million years ago
Table 6.3
Convergent evolution of agricultural behaviors
(Farrell et al. 2001), the ants did so 50 million years ago (Schultz and
Brady 2008), and the termites evolved agricultural behavior between 34
to 24 million years ago (Mueller et al. 2005).
Agricultural behavior is incredibly complex. It includes (1) preparing
the substrate to be used for growing the crops, (2) planting the crops, (3)
monitoring the growth and potential disease status of the crops, (4) pro-
tecting the crops from disease, (5) protecting the crops from crop-eating
species other than the farmers, (6) weeding invasive species out of the
crops, (7) using chemical herbicides for weed control, (8) using microbes
for biological pest control, (9) using microbial symbionts to procure
nutrients for the crops, and (10) the sustainable harvesting of the crops
for food (Mueller et al. 2005). This list is not a list of agricultural behav-
iors found in humans—it is a list of behaviors found in ants! In fact,
Mueller et al. (2005, 564) argue that “[b]ecause of the universality of crop
diseases in both human and insect agriculture, it may be fruitful to
examine the . . . solutions that have evolved convergently in insect agri-
culture for possible application to human agriculture.” Here we should
note that ant agriculturalists use microbes for biological pest control,
whereas most human agriculturalists (thus far) use chemical pesticides.
There is one major difference in human and hexapod agriculture:
hexapods cultivate fungus crops, whereas humans usually cultivate plant
crops (although we do also farm mushrooms). The most advanced
hexapod agriculturalists have in fact evolved a new way to digest cellu-
lose: rather than digesting it internally with the usage of anaerobic bac-
teria and specialized stomachs like many vertebrate animals (itself a
convergent trait; see table 2.10), advanced hexapods use fungus gardens.
Indeed, Schultz and Brady (2008, 5435) argue that a colony of leaf-cutter
ants is “the ecological equivalent of a large mammalian herbivore in
terms of collective biomass, lifespan, and quantity of plant material
consumed.”
Fungi are the only major group of organisms that are able to digest
cellulose in the presence of free oxygen (Gould and Gould 2007). They
also do not need light, as they are not photoautotrophs like plants, but
they do need heat and humidity for efficient digestion. Both ants and
termites have independently evolved the behaviors needed to create
warm, moist agricultural chambers for fungus gardens, the ants under-
ground and the termites in their above-ground adobe buildings. The ants
are particularly important in analyzing the evolution of agricultural
behavior, in that living ants still use five distinctly different agricultural
systems that can be demonstrated to have evolved in a sequence of
Convergent Minds 225
some species of which they have domesticated (Farrell et al. 2001). Dif-
ferent beetle lineages also use different tree hosts, in that some infest
species of conifers, and others attack species of angiosperms.
Animal husbandry in agriculture, where animals are domesticated
instead of crops, is also convergent, but rare (table 6.3). Humans
domesticated sheep, goats, cattle, and pigs for food some 7,000 years ago
(Gupta 2004). In the hexapods, the formicine and dolichoderine ants in
particular are known for their domestication of aphids (Aphididae), soft
scale insects (Coccidae), tree hoppers (Membracidae), and mealy bugs
(Pseudococcidae; Delabie 2001; Stadler and Dixon 2005).
Aphids are herbivores that feed on the phloem sap of plants, which is
rich in sugar but poor in nitrogen. To obtain enough nitrogen, aphids
have to consume large amounts of sap, and they excrete the excess sugar
as honeydew. This honeydew waste product is a food source for ants, who
collect it from the aphids. Thus begins the evolution of the trophobiotic
relationship between the aphid herbivore and the ant farmer. When
tended by the ant farmers, aphids produce much more honeydew, and
higher-quality honeydew, than they do when solitary. To offset this meta-
bolic cost, the aphids benefit by being protected from predators by the
ant farmers. The ants are so good at protecting their aphid herbivores
that farmed aphids can become very numerous and gregarious. The cost
of protecting the aphids is therefore offset by the production of aphid
“herds,” which provide a food source for the ants that is concentrated in
a small farmed area—rather than having a diffuse source of food scat-
tered over a wide geographic region. Some aphids have become so
domesticated that they depend upon the ant farmer for existence; for
example, the aphid Stomaphis quercus is cultivated by the ant Lasius
fuliginosus, and is only found on oak trees farmed by this ant (Stadler
and Dixon 2005).
Ants are known to construct protective shelters out of plant debris for
some of their farmed insects, both to shelter the insects from the weather
and to conceal them from predators. The ant Formica obscuripes goes
further and shelters its herd of aphids in galleries within its own under-
ground dwelling, leading the herd of aphids out to “pasture” during the
day, and herding them back underground for the night (Delabie 2001).
Other ants will pick up and transport their farmed insects, particularly
soft scale insects, to more favorable plant “pastures” for them to feed.
When an ant farmer feeds, it will either drum the aphid’s abdomen with
its antennae or stroke the aphid’s sides to encourage it to excrete hon-
eydew droplets, which the ant collects (Delabie 2001).
Convergent Minds 227
At first glance, it would appear that ants have converged on the animal
husbandry behavior of the human dairy farmer (table 6.3). The dairy
farmer benefits from a herd of cows, which are a concentrated source of
milk for food. The cows benefit from being protected from wolves and
other predators by the farmer, and farmed cows produce much more
milk than wild ones. The farmer leads the herd of cows out of a barn
shelter to the open pasture on a daily basis, and later strokes the mammary
glands of the cows to encourage them to excrete milk, which the farmer
collects. The problem with this analogy is the fact that ants evolved aphid
farming in the Early Oligocene, some 30 million years before the exis-
tence of humans (Stadler and Dixon 2005). It is the human dairy farmer
that has converged on the animal husbandry behavior of ants, not the
other way around.
In summary, it took ant agriculturalists 30,000,000 years to go from
farming fungus to domesticating fungus: from the Early Eocene paleoat-
tine ants to the Early Miocene neoattine ants (Schultz and Brady 2008).
In contrast, human farmers have accomplished the same feat with plants
in less than 10,000 years. Yet, as we also saw with the evolution of archi-
tectural behaviors, the evolution of stereotypic instinctive behaviors in
ants and flexible learned behaviors in humans have both converged on
the same result: farming of domesticated crops (table 6.3). Ants and
humans also convergently evolved animal husbandry, but in this case the
fossil record does not reveal how long it took the ants to accomplish the
feat of domesticating aphids.
The complex of agricultural behaviors independently evolved by
hexapod arthropods and vertebrate humans is astonishingly convergent.
However, as with architecture, using the analytical technique of theoreti-
cal morphology to examine the question of convergent agricultural
behavior provides a different perspective by focusing on the cultivation
process instead: Is there only a limited number of ways to successfully
cultivate crops? If so, convergent behavior must necessarily result, for
independent lineages of organisms have a limited number of farming
options to discover. The convergent usage of chemical herbicides by such
disparate organisms as ants and humans results from the fact that such
pesticides are necessary for the cultivation of crops. As with architecture,
convergent agricultural minds are here the product of functional con-
straint on evolution.
It is difficult for a single individual to run a really efficient farm.
Farming is labor intensive, and is usually the sum product of many indi-
viduals who have taken on different tasks. Task partitioning and the
228 Chapter 6
Group Behavior
Table 6.4
Convergent evolution of eusocial behavior
Table 6.5
Convergent evolution of collective animal behavior
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Convergent Minds 233
range’ of perception much larger than their actual sensory range.” In the
European starling (table 6.5), this behavioral coupling is a function of
topological distance: each bird interacts with a fixed number of neighbor-
ing birds, typically six or seven, instead of with all birds contained in a
fixed volume of metric space (Ballerini et al. 2008). This topological
interaction allows the size of the flock of birds to expand and contract
without altering the behavioral coupling between the birds.
Field studies with the desert locust (table 6.5) have revealed a critical
threshold effect that triggers herd behavior. Buhl et al. (2006, 1402)
observed that each locust behaves individually until the number of
locusts present per square meter of land area reaches the number eight,
at which point “a rapid transition occurs from [the] disordered move-
ment of individuals within the group to highly aligned collective motion.”
Buhl et al. (2006) further argue that this critical density effect is a key
triggering factor in all herding behavior, although the threshold number
is different for different animal groups.
In table 6.5 I have listed as many disparate convergent lineages that
have evolved collective minds, or herd behavior, that I am aware of. I
have made no effort to try to determine how many times herd behavior
may have convergently arisen within these major lineages; future phylo-
genetic studies of within-lineage convergence in herd behavior will cer-
tainly expand the list given in table 6.5. Within the mammalian clade, I
have listed examples of convergent lineages in the three major branches
of existent mammals: the afrotherians, laurasiatherians, and euarchonto-
glires. Some ruminant ungulates form enormous self-organized groups,
as seen in the behavior coupling that produces the wavelike front of
100,000 moving individuals in a wildebeest herd in Africa (Couzin and
Krause 2003). Elephants form smaller herds, and even we humans exhibit
markedly different behavior when in groups than when alone, and
unconsciously adopt coordinated behavior when we are moving as
pedestrians along the sidewalks of our cities (Raafat et al. 2009). At the
opposite extreme, swarming ants and marching locust juveniles can form
even larger herds that cover many kilometers in area even though the
individual animals themselves are tiny (Buhl et al. 2006; Conradt and
Roper 2005).
The independent evolution of herding behavior by the Mesozoic dino-
saurs is yet another example of the ecosystem convergence that exists
between the dinosaurian and mammalian ecosystems (see table 4.10). In
table 6.5 I have listed examples of convergent lineages in the two major
branches of the dinosaur clade, the saurischians and the ornithischians.
234 Chapter 6
Table 6.6
Convergent evolution of social hunting behavior
Note: The geological age of extinct species is marked with a †. For data sources, see text.
236 Chapter 6
and will run down, attack en masse, rip to shreds, and entirely eat a
gazelle in ten minutes (Wilson 1980). As such, they approach the deflesh-
ing abilities of the army ants.
The top carnivores of the mammalian ecosystem, the felids, are mostly
solitary predators. But one lineage, the African lion, has convergently
evolved pack-hunting behavior (table 6.6). These cats form social prides,
or herds, of numerous females and one to two males. Pack-hunting
behavior in the lions is also a gendered behavior, in that the females hunt
as a cohesive social unit but the males do not. Last, modern humans are
pack hunters but the reverse of the lions, in that the males were tradition-
ally the hunters and the females the gatherers. Pack-hunting behavior
enabled our small ancestors (1.5 to 1.8 meters tall, 50 to 100 kilograms
in weight) to bring down large prey like the woolly mammoths (Mam-
muthus primigenius) in Europe, which were 3 to 4 meters tall and weighed
8,000 kilograms.
Another odd form of social hunting has convergently evolved in the
carrion-eating birds (table 6.6). As discussed in chapter 4, Old World and
New World carrion-eating birds have independently evolved many mor-
phological features associated with their feasting on carcasses. Curiously,
they also have converged in aspects of their behavior: Eurasian black
vultures and North American turkey vultures are gregarious, roosting
together in large numbers at night, and often assembling in large groups
during the day as well, perched in trees or on top of buildings. Yet Eur-
asian vultures are related to hawks and eagles, which are usually not
gregarious birds, unlike the storks and flamingos that are the close rela-
tives of the turkey vultures. In both Eurasia and North America, these
birds use a soaring-in-circular-formation flight pattern in searching for
rotting corpses and, when a carcass is discovered, they will often continue
to soar in circles until other vultures arrive. Once a group has been
assembled, the flock descends to feed.
The function of this convergent behavior is not well understood. Some
have proposed that the vultures continue to circle in order to make sure
the target animal is indeed dead, or that there are no ground-dwelling
predators in the area that might pose a danger to the bird. However, the
frequency of this behavior seems to lend more weight to the suggestion
that the vultures are actually waiting for other birds to arrive, or even
summoning them. They certainly do not need a larger number of vultures
to complete the kill, because the prey animal is already dead. Carrion
is an unpredictable food resource, and vultures must often wait long
periods between feedings. When a carcass is discovered, a vulture will
238 Chapter 6
often gorge itself to the point that it has difficulty flying away. It might
seem advantageous to the bird to selfishly keep the discovery of a carcass
to itself, but at the same time, a large animal carcass is usually too much
food for a single bird to consume by itself. Thus, it might be advantageous
to share the information with other hungry vultures, in anticipation that
they will return the favor in the future when they discover a carrion
source, in a case of reciprocal altruism. Whatever the real function
of vulture group behavior, it is clear the minds of these birds have
converged.
Convergent Mentalities
Table 6.7
Convergent evolution of self-awareness and metacognition
measuring the weight of the brain relative to the weight of the body
(Prior et al. 2008), the magpie has a brain-body weight index of 31, which
is greater than the index of 21 for a human. The bottlenose dolphin has
an index of only 9.0, and the Asian elephant an index of 1.6. Clearly the
brain-body weight index favors animals of small size over those of large
(Prior et al. 2008).
In chapter 5 we considered the effect of size and allometric scaling in
producing convergent tissue structures (West et al. 1999). If we plot the
logarithm of brain mass versus the logarithm of body mass, most mammals
and birds fall along a single linear function of allometric scaling (Gould
and Gould 2007). Primates, cetaceans, corvids, and elephants plot above
this line, however, indicating that they have larger brains than are strictly
necessary for neural control of their body masses. Humans and porpoises
have brains that plot the farthest away from the general allometric
scaling, but even here problems of brain-size comparison arise. The por-
poise lives in a state of weightlessness, of neutral buoyancy in water, and
does not need the large muscle masses (which require neural control)
that land animals must have in order to exist in the gravitational field of
the Earth. Much of their body mass is actually blubber fat, which they
240 Chapter 6
use for thermal insulation in the cold waters of the ocean (Gould and
Gould 2007). Thus, their large brains may be even more unusual than
that of humans.
Although self-aware animals all have unusually large brains (in the
allometric sense), their brain structures are radically different: “primate-
cetacean cognitive and behavioral convergence is a dramatic example of
functional convergence in the face of profound structural or mechanistic
divergence” (Marino 2002, 30). Even though the elaboration of the corti-
cal circuitry in primates and cetaceans is very different, they are both
members of the clade of the eutherian mammals and share much of their
subcortical neuroanatomy (Marino 2002). The corvid birds, however, do
not even have a prefrontal cortex! Although these animals are avian
dinosaurs, with vastly different brain structures from those of mammals,
“intelligence in both corvids and primates has evolved through a process
of divergent brain evolution yet convergent mental evolution . . . [showing
that] intelligence can evolve in the absence of a prefrontal cortex”
(Clayton and Emery 2008, 138–139).
Another mental state associated with high degrees of intelligence is
metacognition, the ability of an animal not only to be aware of its own
existence, but to think about its own thinking. Humans know when they
are uncertain, are aware of that cognitive condition, and are able to think
about how to deal with their uncertainty—to seek further information,
or to defer a response and to wait (Smith 2009). Thus far in this new area
of research, experimentation has revealed that the bottlenose dolphin
and the Barbary macaque monkey are aware of their own cognitive
states (table 6.7), that capuchin monkeys (Cebus apella) show equivocal
evidence of metacognition, and that pigeons (Columba livia) show no
evidence of being aware of their cognitive state (Smith 2009). Smith
(2009, 395) notes that the corvid birds, “having proven themselves cog-
nitively sophisticated, would be the usual suspects for testing in this
area,” as are elephants and great apes other than humans. It is known
that both New World and Old World monkeys have convergently evolved
tool use (see table 6.1), and the evidence that macaque monkeys have
metacognition suggests that these monkeys should be carefully exam-
ined for evidence of possible mirror self-recognition as well.
Finally, another mental state associated with convergent intelligence
is the awareness of death, as well as the experience of grief and sorrow
at the loss of another individual (table 6.8). Both gorillas and dolphins
have been observed to mourn the death of their parent or their offspring.
Convergent Minds 241
Table 6.8
Convergent evolution of mourning
Note: The geological age of extinct species is marked with a †. For data sources, see text.
Both in the wild and in zoos, gorilla and chimpanzee mothers carefully
carry around their dead infants for days, caressing them, or placing them
on the ground and pacing in circles around them, checking over and over
again to see if they have resumed breathing. Dolphin mothers repeatedly
lift their infants up in the water for days in apparent attempts to help
the calf breathe. The mothers do not eat during this period, and make
repeated distress cries.
Other animals are known to extend their mourning behavior to other
members of their own kind that are not members of their immediate
family (table 6.8). Elephants repeatedly touch the skulls and tusks of
long-dead elephants, gently lifting the bones up and down or moving
them around. They also repeatedly pace away and then return to the
remains of the dead elephant. In a stunning photograph, Monica Szczu-
pider (2009, 12–13) captured the facial expressions and body language
of 16 chimpanzees who were mourning the death of an elder female at
the Sanaga-Yong Chimpanzee Rescue Center.
Two human species evolved a mourning behavior not seen in our
great ape cousins: the burial of our dead. The oldest fossils of anatomi-
cally modern humans found to date are 200,000 years old, and are from
northeast Africa (modern Ethiopia). The Neanderthal humans have long
roots in Europe, going back some 400,000 years. It is thought that the
Neanderthal humans evolved from Homo heidelbergensis in Europe,
242 Chapter 6
Our results strongly support the hypothesis that the essential elements of organic
structure are highly constrained by geometric rules, growth processes, and the
properties of materials. This suggests that, given enough time and an extremely
large number of evolutionary experiments, the discovery by organisms of “good”
designs—those that are viable and that can be constructed with available mate-
rials—was inevitable and in principle predictable.
—Thomas and Reif (1993, 342)
Was mich eigentlich interessiert, ist, ob Gott die Welt hätte anders machen
können.
—Einstein (quoted in Seelig 1956, 72)
very molecules from which they are constructed, from their ecological
roles in nature to the way in which their minds function.
Since convergent evolution is so ubiquitous in nature, as we have seen
in the previous five chapters, the total extent of convergent evolution
might best be revealed by studying its opposite: unique evolution. That
is, rather than compiling lists of convergences, we might compile lists of
solitary evolutionary innovations in species that have not been indepen-
dently discovered by other species in their evolutionary pathways.
Vermeij (2006) set out to do just that, and compiled a list of evolutionary
innovations said to be unique. He discovered that “purportedly unique
innovations either arose from the union and integration of previously
independent components or belong to classes of functionally similar
innovations” and that “important ecological, functional, and directional
aspects of the history of life are replicable and predictable” (Vermeij
2006, 1804).
What are the possibilities for evolution in our universe? Can we
even think about considering the total spectrum of what is possible and
not possible in biological evolution? The answer is yes, by using the
analytical techniques of theoretical morphology, in particular by the
construction of theoretical morphospaces (McGhee 2001, 2007). The
concept of the theoretical morphospace originated in evolutionary
biology (McGhee 1999), but it has subsequently caught the attention of
philosophers (Maclaurin 2003), linguists, cultural anthropologists, and
neuroscientists (Hauser 2009) who are seeking to explore the spectrum
of both possible and impossible languages and cultures. Here we will
use the concept to analyze the phenomenon of convergent evolution
with respect to the spectrum of existent, nonexistent, and impossible
biological form.
Figure 7.1
A theoretical hyperdimensional space of possible form (modified from McGhee 2007).
Each dimension of the space represents a morphological trait that may be measured on a
given biological form, f. All possible coordinate combinations (points) within the theoreti-
cal morphospace represent the set of all possible biological forms. Although only eight
dimensions are shown in this schematic diagram, the dimensionality of an actual hyper-
space of form will be much larger.
Figure 7.2
Convergent evolution of form within a theoretical morphospace. Organisms in different
regions of the morphospace, and thus possessing different initial morphologies, have evo-
lutionary trajectories that take them to the same region within the morphospace, and their
morphologies converge.
Figure 7.3
Functional constraint in theoretical morphospace. Functional forms are found within the
rectangular region in the morphospace, and nonfunctional forms are found outside the
rectangular region. The boundaries of the rectangle thus delimit the functional constraint
boundary on possible form within the morphospace.
Functional and Developmental Constraint in Convergent Evolution 249
Figure 7.4
Combining functional and developmental constraint in theoretical morphospace. The
boundaries of the dotted-line rectangle delimit the developmental constraint boundary on
possible form within the morphospace: forms within the dotted-line rectangle are devel-
opmentally possible, while forms outside the dotted-line rectangle are developmentally
impossible. Forms f:0 are thus both nonfunctional and developmentally impossible, forms
f:1 (shaded region) are both functional and developmentally possible, forms f:2 are non-
functional but developmentally possible, and forms f:3 are functional but developmentally
impossible. See text for discussion.
250 Chapter 7
(1) Biological forms that do not function and that cannot be developed
by Earth life, abbreviated f:0. These are all of the hypothetical forms
shown in figure 7.1 that fall outside the two intersecting rectangles shown
in figure 7.4.
(2) Biological forms that do function and that can be developed by
Earth life, abbreviated f:1. These are all of the hypothetical forms shown
in figure 7.1 that fall in the region where the two rectangles shown in
figure 7.4 intersect (the shaded region).
(3) Biological forms that could be developed by Earth life, but that do
not function and thus are lethal, abbreviated f:2. These are all of the
hypothetical forms shown in figure 7.1 that are located in figure 7.4 in
the region bounded by the dotted-line rectangle but not located in the
region where it intersects with the solid-line rectangle (the shaded
region).
(4) Biological forms that do function but that cannot be developed by
Earth life, abbreviated f:3. These are all of the forms shown in figure 7.1
that are located in figure 7.4 in the region bounded by the solid-line
rectangle but are not in the region where it intersects with the dotted-line
rectangle (the shaded region).
Now let us consider these four regions of possible form from the point
of view of actual existent life on Earth. The myriad forms of life that
surround us here on Earth clearly are both functional and developable,
and we can collect all of these forms into a set of forms { f:1}. Every living
thing on Earth, from butterflies to bacteria, belong to the form set { f:1}.
Needless to say, all of the convergent forms of life that we have consid-
ered throughout this book are also members of the form set { f:1}. The
phenomenon of convergent evolution immediately reveals to us that the
size of form set { f:1} is not infinite, in that life on Earth has been con-
strained to repeatedly reevolve the same forms within this set over and
over. The hypothetical universe in which every species has its own unique
Functional and Developmental Constraint in Convergent Evolution 251
Life as We Know It
All life on Earth belongs to the form set { f:1}, biological forms that both
function and can be developed by Earth life. As we have seen in chapters
2 through 6 in this book, life on Earth is highly convergent. The spectrum
of forms in the set { f:1} have been reevolved over and over again in the
history of life on Earth.
The set of existent forms of life on Earth, { f:1}, is a function of the
functional constraint boundary and the developmental constraint bound-
ary in the theoretical morphospace (figure 7.4). We believe the functional
constraints of physics and geometry to be the same throughout the uni-
verse and to be extrinsic with respect to biology (McGhee 2007); there-
fore, this constraint boundary should also apply to life forms throughout
the universe. That is, the spatial position of the functional constraint
boundary shown in figure 7.4, relative to the spectrum of hypothetical
forms shown in figure 7.1, should remain the same throughout the uni-
verse. Alien life forms should belong to the same set of forms { f:1} that
Earth life forms belong to, given the same physical conditions on that
alien planet.
Thus, we could predict that those alien organisms with powered flight
will have evolved wings, just as animals on Earth have done (table 2.2),
and fast-swimming organisms in alien seas will evolve streamlined, fusi-
form bodies (table 2.1). Land-dwelling, sessile, photoautotrophic organ-
isms will evolve tree forms, just as plants on Earth have done (table 3.1),
and those tree forms will have leaflike structures (table 3.3). The laws of
physics and geometry also extend to activities, in that there are a limited
number of ways to build structures (table 6.2) and to successfully culti-
vate crops (table 6.3). The geometries of alien buildings, and the algo-
rithms of farming procedures, should be similar to those found on
Earth. If they live in highly social groups, natural selection should drive
evolution to produce aliens that are self-aware, just as it has on Earth
(table 6.7).
Functional and Developmental Constraint in Convergent Evolution 253
simply because life on other planets cannot develop these forms either.
Several lines of reasoning, however, suggest that this may not true, and
that the set of forms { f:3} potentially exists in the universe. NASA, the
space exploration agency of the United States, is also very interested in
the possible existence of non-Earth-type life, as discussed in Peter Ward’s
book Life as We Do Not Know It (2005). In trying to conceive of possible
forms of life that are nonexistent on Earth, the scientists at NASA are
engaging in classic theoretical morphology.
Let us also engage in a theoretical-morphology thought experiment.
Development of life on Earth starts from a single cell and, as far as we
currently understand it, depends upon the genetic coding within the cell
and the interaction of the different molecules and tissue geometries
produced as the cell grows. Fundamentally, we are dealing with mole-
cules and geometries: all Earth life is carbon-based, uses DNA to code
for amino acid assembly, and is constructed of amino acids that have
left-coiled geometries; furthermore, multicellular life develops complex
tissue geometries from simple, spherical, topologically solid cell clusters
(Newman 2010). So, let us consider each of these characteristics in
turn with regard to the question: Could hypothetical life elsewhere be
different? Can we conceive of possible life forms that are nonexistent
on Earth?
All life on Earth is carbon-based: non-carbon-based life would surely
experience different developmental constraints than those present in
Earth life, and therefore could possibly develop functional biological
forms that Earth life cannot. The most widely discussed chemical alterna-
tive to carbon is silicon-based life. From the periodic table of elements,
we see that silicon can form four bonds like carbon (for example, in the
formation of monosilane, SiH4, the analog of methane, CH4), which is so
important for building the carbon-backboned molecules of Earth life.
Unfortunately, silicon bonding simply cannot produce the huge macro-
molecules that carbon can—macromolecules that are very strong and
stable at normal Earth temperatures and pressures. Curiously, however,
it is thought that silicon-based life might be possible at either very high
temperatures or very low temperatures (Ward 2005). At very low tem-
peratures, silicon can form polymers, large macromolecules that are
analogs to those formed by carbon. Rather than using water as a solvent,
as with carbon-based life, silicon-based life could use liquid methane or
nitrogen. As a result, it has been hypothesized that silicon-based life
might exist in the methane lakes that exist on Titan, Saturn’s planet-like
moon that possesses its own nitrogen atmosphere, or in possible liquid
Functional and Developmental Constraint in Convergent Evolution 255
here, however, the nonexistent code “shows behavior very similar to that
of the natural code” (Freeland and Hurst 1998, 247); thus, it is not clear
whether life evolved using this alternative code would be that much dif-
ferent from life on Earth.
The amino acid molecules that are used as the building blocks of pro-
teins in Earth life are all left-coiled. Could life exist elsewhere that uses
right-coiled amino acids? Experiment shows that chemical selection in
protein synthesis produces homochirality, the use of only one type of
amino acid geometry. But proteins can be constructed in the laboratory
with either all left-coiled or all right-coiled amino acids, and it is not clear
why life on Earth has chosen the left-coiled chemical pathway (Castelvec-
chi 2007). Both right-coiled and left-coiled amino acids are produced by
inorganic chemical processes in nature, and so perhaps chance has pro-
duced life forms elsewhere that are constructed of proteins made with
right-coiled amino acids. Would such life produce biological forms that
are simply the chemical mirror-image of Earth life forms? Or would
those forms be different from anything seen on Earth, and might they
belong to form set { f:3}?
Just as the element carbon has unique properties and the DNA coding
system found in Earth life is highly efficient, some have argued that
molecular evolution preferentially produces left-coiled amino acids.
Experiments have suggested (1) that left-coiled amino acids are slightly
more stable than right-coiled, and thus could have accumulated more
readily in prebiotic environments than right-coiled, and (2) that left-
coiled amino acids crystallize slightly faster than right-coiled, and thus
could have become more enriched in prebiotic environments than right-
coiled (Castelvecchi 2007). Either process would result in prebiotic envi-
ronments containing more left-coiled amino acids than right-coiled
amino acids as potential building blocks for proteins. Still, a chirality-
biasing process would not rule out the possibility that life constructed
with right-coiled amino acids could exist in the universe; it would simply
mean that such life forms would be less abundant in the universe than
life forms with left-coiled amino acids.
Modern molecular studies have revealed that a great deal of conver-
gent evolution of life here on Earth is in fact due to developmental
constraint: the phenomenon of “deep homology” that we considered in
chapter 1. “Studies of deep homology are showing that new structures
need not arise from scratch, genetically speaking, but can evolve by
deploying regulatory circuits that were first established in early animals.
. . . The more that researchers look, the more they will find that the same
258 Chapter 7
tools have been used to build a great variety of structures long thought
to have independent histories” (Shubin et al. 2009, 822). Would life else-
where in the universe, even if it is also carbon-based, coded for by base-
four DNA, and composed of proteins made of left-coiled amino acids,
have evolved the same developmental regulatory systems seen in Earth
life? If so, the form set { f:3} would not exist, but this at first seems
unlikely. The deep-homology regulatory circuits discussed by Shubin
et al. (2009) can be traced back to the Cambrian diversification of animal
life some 540 to 600 million years ago. Yet life existed on Earth 3.5 billion
years ago, and animals are a late addition to the history of life. It took
roughly 3 billion years of evolution for life to evolve the deep-homology
regulatory circuits present in animal life on Earth—what is the probabil-
ity that the very same regulatory system would convergently evolve with
the convergent evolution of multicellular, heterotrophic life elsewhere?
Multicellular, heterotrophic life forms—animals—that have evolved
with a different set of deep-homology regulatory circuits could poten-
tially be able to develop forms that cannot be developed by life on Earth,
forms that could belong to the set { f:3} and that would not be convergent
on Earth life.
Because we do not know the answer to the question of how develop-
ment originated in multicellular life forms on Earth, we also do not know
how development might originate in life forms elsewhere in the universe.
Theoretical modeling, however, suggests that a significant part of devel-
opmental constraint might indeed be a function of physical and geomet-
ric constraints (Newman et al. 2006; Müller 2007; Newman 2010), and
therefore that developmental systems in alien life might be similar to
those found in Earth life. In modeling the evolution of the development
of multicellular organisms, Newman et al. (2006) start with four different
kinds of physical and chemical patterning mechanisms: diffusion gradi-
ents, sedimentation gradients, reaction-diffusion mechanisms, and chemi-
cal oscillation mechanisms. Most importantly, these four patterning
mechanisms are found in nonliving as well as living chemical systems.
Then Newman et al. (2006) add two basic cell properties: differential
adhesion and cell polarity. Differential adhesion in pairs of tissues that
differ in cohesivity produces mixing behavior like that seen in immiscible
liquids, a behavior that is found in nonliving chemical systems and thus
not unique to life. Likewise, the acquisition of polarization in the adhe-
sion of cells is a mechanism not unique to life. Newman et al. (2006) next
explored the possible interactions of the four patterning mechanisms
with the two cell properties in a manner similar to that used in theoretical
Functional and Developmental Constraint in Convergent Evolution 259
Table 7.1
A periodic table of life, based upon locomotory type and evolutionary sequence of
origination
Spectrum of
locomotion
Swimming
Sequence of Crawling Walking (fusiform Flying
evolution Sessile (legless) (legs) body) (wings)
Plants Plants — — — —
Invertebrates Barnacles Worms Arthropods Squid Insects
Amphibians — Caecilians Amphibians Tadpoles —
Reptiles — Snakes [reptiles] Ichthyosaurs Pterosaurs
Dinosaurs — — [dinosaurs] Penguins Birds
Mammals — — [mammals] Porpoises Bats
Complexity theory also suggests a new take on an old question, long a staple of
science fiction and speculative science: When we do find aliens, or they find us,
what will they look like? By revealing many forms of Earth life to be governed
by deep geometrical rules of self-organization in nature, complexity suggests a
universal geometry of life that should transcend worlds. . . . There is no way to
predict precisely what aliens will look like, but the fractal geometry of life gives
us reason to believe that when they do finally land on the White House lawn,
whatever walks or slithers down the gangplank may look strangely familiar.
—Grinspoon (2003, 272–273)
The laws of physics, it is believed, are the same everywhere in the universe. This
is unlikely to be true of biology.
—Crick (1988, 138)
could not have been deduced from any laws of nature: “Contingency is
the affirmation of control by immediate events over destiny. . . . Our own
evolution is a joy and a wonder because such a curious chain of events
would probably never happen again, but having occurred, makes eminent
sense” (Gould 1989, 284–285).
Some evolutionary biologists find Gould’s view of evolution as an
unpredictable, nonrepeatable, and historically contingent process to be
liberating and uplifting. For example, Stuart Kauffman (2008, 5) com-
ments on Gould’s view that the path of evolution cannot be deduced
from any laws of nature: “a ‘natural law’ is a compact description before-
hand of the regularities of a process. But if we cannot even prestate the
possibilities, then no compact descriptions of these processes beforehand
can exist. These phenomena, then, appear to be partially beyond natural
law itself. This means something astonishingly and powerfully liberating.
We live in a universe, biosphere, and human culture that are not only
emergent but radically creative. We live in a world whose unfoldings we
often cannot prevision, prestate, or predict—a world of explosive creativ-
ity on all sides.”
Similarly, in writing of Gould’s view that evolution could not be
repeated, Kauffman (2008, 130) expands on this concept with obvious
joy: “The vast nonrepeatability, or nonergodicity, of the universe at all
levels of complexity above atoms—molecules, species, human history—
leaves room for a creativity in the way the universe unfolds at these
levels, a creativity that we cannot predict. . . . Now I want to make my
outrageous claim: the evolution of the biosphere is radically nonpredict-
able and ceaselessly creative.”
And last, as I promised at the beginning of this chapter, the concepts
of spirituality and of God also arise from this consideration of evolution
as an unpredictably creative process: “I believe we can reinvent
the sacred. We can invent a global ethic, in a shared space, safe to all
of us, with one view of God as the natural creativity in the universe”
(Kauffman 2008, xiii).
Other biologists welcome Gould’s view of evolution as the ultimate
dismissal of the older idea of teleology, that evolution is design-like and
progressive, and thus has a direction and ultimately a destination. But
Reiss (2009, 174) goes beyond Gould to attack the very idea of adaptive
improvement in evolution as being teleological, as in his discussion of
Sewall Wright’s concept of the adaptive landscape and Darwin’s concept
of natural selection: “Wright seems never to have realized that while in
the context of domestic breeds it may be reasonable to speak of ‘improve-
268 Chapter 8
We have now reviewed the arguments for the view that evolution is
unpredictable, contingent, nonrepeating, and directionless and those for
the view that evolution is predictable, predetermined, and inevitable.
What are the implications of the phenomenon of convergent evolution,
the subject of this book, with regards to these radically different views
of evolution?
First, the view that the evolutionary process is nonrepeating (noner-
gotic) is demonstrably false. Chapters 2 through 6 of this book have
demonstrated that evolution is a highly repeatable process. In fact,
repeatability in evolution is rampant at every level of life on Earth, from
tiny organic molecules to entire ecosystems of species, and even to the
ways in which we think.
Second, the view that evolution is entirely historically contingent, and
thus unpredictable (and nonrepeating), is demonstrably false. Conver-
gent evolution—the repeated, independent, evolution of the same trait
in multiple phylogenetic lineages at different points in geological time—
272 Chapter 8
mine what that destination might be. Even our models for the physical
evolution of the universe have changed radically over the past century:
once it was thought the universe was steady-state, and would infinitely
persist. Then it was thought that the universe would eventually cease to
expand, and then collapse upon itself in a sort of reverse-Big-Bang fire-
ball. Now it is generally thought the universe will continue to expand
infinitely and that all the stars will eventually burn out, one by one, in a
cold, dark universal death. But, as we currently still do not truly under-
stand dark matter or dark energy, that model of the ultimate destination
of the physical evolution of the universe may change in the future as
well. As yet, we have no equivalent models for the ultimate destination
of biological evolution. If we cannot, at present, confidently predict the
ultimate destination of the physical evolution of the universe, how can
we hope to predict the ultimate destination of the vastly more complex
process of biological evolution?
There are hints that such a destination might exist: the increase in
complexity produced by evolution in the past 3.5 billion years of Earth
history, and the evolution of self-awareness in the past 5 to 10 million
years in multiple independent lineages (see table 6.7). The universe has
become aware of itself (Grinspoon 2003). Is such a trend intrinsic to the
process of biological evolution? The self-assembling properties of bio-
logical evolution, and its constant building upon previous evolutionary
steps via the process of natural selection, would lead us to predict that
such a process should produce increasing complexity in other life forms
as well, on other worlds.
Have life forms on other worlds become self-aware? Has matter awak-
ened, has the universe become aware of itself elsewhere? We have no
evidence that this has occurred, although we keep searching the skies for
possible electromagnetic or other signals that would point to the exis-
tence of extraterrestrial civilizations of self-aware organisms in search of
other self-aware organisms. On Earth, the convergent evolution of self-
awareness has been linked to the convergent evolution of complex social
systems, as discussed in chapter 6. The fact that, on Earth, life in the sea
and life on land—life in radically different environments—both have
evolved convergent social systems and self-awareness would lead us to
predict that self-awareness would evolve on extraterrestrial worlds as
well.
Or is self-awareness an intrinsic property of the developmental regula-
tory systems present in animal life on Earth? Would life elsewhere,
evolving under different developmental constraints, have evolved social
276 Chapter 8
a few forms or into one; and that, whilst this planet has gone cycling on
according to the fixed law of gravity, from so simple a beginning endless
forms most beautiful and most wonderful have been, and are being,
evolved.”
We now have the benefit of an additional 150 years of analysis of
biological evolution. The phenomenon of convergent evolution leads us
to reconsider Darwin’s closing sentence, at least for carbon-based life,
and to suggest that it be rewritten (with the modern alterations in the
original sentence indicated by italics): “There is grandeur in this view of
life in the universe, with its several powers of functional and developmen-
tal constraint, having been originally breathed into a few forms of life or
into just the one that is carbon based; and that, whilst this planet and
others have gone cycling on according to the fixed law of gravity, from
so simple a beginning limited forms most beautiful and most wonderful
have been, and are being, re-evolved throughout the universe.”
Appendix: A Phylogenetic Classification of Life
Table A.1
A phylogenetic classification of life forms discussed in this book
Eubacteria
Archaea
Eukarya
– Bikonta
– – Excavobionta
– – Rhizaria
– – Chromoalveolata
– – Green eukaryotes
– – – Glaucophyta
– – – Metabionta
– – – – Rhodobionta
– – – – Chlorobiota > see Table A.3
– Unikonta
– – Amoebozoa
– – Opisthokonta
– – – Fungi
– – – Choanozoa
– – – – Choanoflagellata
– – – – Metazoa
– – – – – Placozoa
– – – – – Demospongiae
– – – – – Hexactinellida
– – – – – Calcarea
– – – – – Eumetazoa
– – – – – – Ctenophora
– – – – – – Cnidaria
– – – – – – Bilateria
– – – – – – – Protostomia
280 Appendix
Table A.1
(continued)
– – – – – – – – Lophotrochozoa
– – – – – – – – – Lophophorata
– – – – – – – – – – Bryozoa
– – – – – – – – – – Phoronozoa
– – – – – – – – – – – Phoronida
– – – – – – – – – – – Brachiopoda
– – – – – – – – – Eutrochozoa
– – – – – – – – – – Syndermata
– – – – – – – – – – – Rotifera
– – – – – – – – – – – Acanthocephala
– – – – – – – – – – Spiralia
– – – – – – – – – – – Entoprocta
– – – – – – – – – – – Annelida
– – – – – – – – – – – Sipuncula
– – – – – – – – – – – Mollusca
– – – – – – – – – – – Parenchymia
– – – – – – – – – – – – Platyhelminthes
– – – – – – – – – – – – Nemertea
– – – – – – – – Cuticulata
– – – – – – – – – Gastrotricha
– – – – – – – – – Ecdysozoa
– – – – – – – – – – Introverta
– – – – – – – – – – – Nematozoa
– – – – – – – – – – – – Nematoda
– – – – – – – – – – – – Nematomorpha
– – – – – – – – – – – Cephalorhyncha
– – – – – – – – – – – – Kinorhyncha
– – – – – – – – – – – – Loricifera
– – – – – – – – – – – – Priapulida
– – – – – – – – – – Panarthropoda
– – – – – – – – – – – Onychophora
– – – – – – – – – – – Tardigrada
– – – – – – – – – – – Arthropoda
– – – – – – – – – – – – Cheliceriformes
– – – – – – – – – – – – Mandibulata
– – – – – – – – Chaetognatha
– – – – – – – Deuterostomia
– – – – – – – – Echinodermata
– – – – – – – – Pharyngotremata
– – – – – – – – – Hemichordata
A Phylogenetic Classification of Life 281
Table A.1
(continued)
– – – – – – – – – Chordata
– – – – – – – – – – Urochordata
– – – – – – – – – – Myomerozoa
– – – – – – – – – – – Cephalochordata
– – – – – – – – – – – Craniata
– – – – – – – – – – – – †Conodonta
– – – – – – – – – – – – Myxinoidea
– – – – – – – – – – – – Vertebrata > see Table A.2
Note: Major clades used frequently in the text for comparative purposes are in boldface
type. Extinct taxa are marked with a †. Classification modified from Lecointre and Le
Guyader (2006) and Benton (2005).
Table A.2
A phylogenetic classification of vertebrate taxa discussed in this book
Vertebrata
– Petromyzontiformes
– †Pteraspidomorphi
– †Osteostraci
– Gnathostomata
– – †Placodermi
– – Chondrichthyes
– – †Acanthodii
– – Osteichthyes
– – – Actinopterygii
– – – Sarcopterygii
– – – – Dipnoi
– – – – Crossopterygii
– – – – – Actinistia
– – – – Tetrapodomorpha
– – – – – Tetrapoda
– – – – – – Batrachomorpha
– – – – – – – †Dendrerpetontidae
– – – – – – – †Dissorophidae
– – – – – – – Lissamphibia
– – – – – – – – Gymnophiona
– – – – – – – – Batrachia
– – – – – – – – – Anura
– – – – – – – – – Urodela
– – – – – – †Lepospondyli
282 Appendix
Table A.2
(continued)
– – – – – – Reptiliomorpha
– – – – – – – †Seymouriamorpha
– – – – – – – †Diadectomorpha
– – – – – – – Amniota
– – – – – – – – Sauropsida
– – – – – – – – – Anapsida
– – – – – – – – – – Testudines
– – – – – – – – – Diapsida
– – – – – – – – – – †Ichthyosauria
– – – – – – – – – – Lepidosauromorpha
– – – – – – – – – – – †Sauropterygia
– – – – – – – – – – – Lepidosauria
– – – – – – – – – – – – Sphenodontia
– – – – – – – – – – – – Squamata
– – – – – – – – – – – – – Iguania
– – – – – – – – – – – – – Scleroglossa
– – – – – – – – – – – – – – Amphisbaenia
– – – – – – – – – – – – – – Gekkonta
– – – – – – – – – – – – – – Autarchoglossa
– – – – – – – – – – – – – – – Scincomorpha
– – – – – – – – – – – – – – – Anguimorpha
– – – – – – – – – – – – – – – – †Mosasauridae
– – – – – – – – – – – – – – – – Serpentes
– – – – – – – – – – Archosauromorpha
– – – – – – – – – – – Archosauria
– – – – – – – – – – – – Crurotarsi
– – – – – – – – – – – – Ornithodira
– – – – – – – – – – – – – †Pterosauria
– – – – – – – – – – – – – Dinosauria
– – – – – – – – – – – – – – Saurischia
– – – – – – – – – – – – – – – †Sauropodomorpha
– – – – – – – – – – – – – – – Theropoda
– – – – – – – – – – – – – – – – †Ceratosauria
– – – – – – – – – – – – – – – – Tetanurae
– – – – – – – – – – – – – – – – – †Carnosauria
– – – – – – – – – – – – – – – – – Coelurosauria
– – – – – – – – – – – – – – – – – – Maniraptora
– – – – – – – – – – – – – – – – – – – Aves
– – – – – – – – – – – – – – – – – – – – Paleognathae
– – – – – – – – – – – – – – – – – – – – Neognathae
– – – – – – – – – – – – – – – – – – – – – Galloanserae
– – – – – – – – – – – – – – – – – – – – – Neoaves
A Phylogenetic Classification of Life 283
Table A.2
(continued)
– – – – – – – – – – – – – – †Ornithischia
– – – – – – – – – – – – – – – †Fabrosauridae
– – – – – – – – – – – – – – – †Genasauria
– – – – – – – – – – – – – – – – †Thyreophora
– – – – – – – – – – – – – – – – †Cerapoda
– – – – – – – – – – – – – – – – – †Ornithopoda
– – – – – – – – – – – – – – – – – †Marginocephalia
– – – – – – – – Synapsida
– – – – – – – – – †Ophiacodontidae
– – – – – – – – – †Edaphosauridae
– – – – – – – – – †Sphenacodontidae
– – – – – – – – – Therapsida
– – – – – – – – – – †Dinocephalia
– – – – – – – – – – †Dicynodontia
– – – – – – – – – – †Gorgonopsia
– – – – – – – – – – Cynodontia
– – – – – – – – – – – Mammalia
– – – – – – – – – – – – †Volaticotheria
– – – – – – – – – – – – †Yinotheria
– – – – – – – – – – – – Australosphenida
– – – – – – – – – – – – – †Ausktribosphenida
– – – – – – – – – – – – – Monotremata
– – – – – – – – – – – – †Multituberculata
– – – – – – – – – – – – Theria
– – – – – – – – – – – – – Marsupialia
– – – – – – – – – – – – – Eutheria
– – – – – – – – – – – – – – Xenarthra
– – – – – – – – – – – – – – Epitheria
– – – – – – – – – – – – – – – Afrotheria
– – – – – – – – – – – – – – – Boreoeutheria
– – – – – – – – – – – – – – – – †Apatemyidae
– – – – – – – – – – – – – – – – Laurasiatheria
– – – – – – – – – – – – – – – – – Eulipotyphles
– – – – – – – – – – – – – – – – – Scrotifera
– – – – – – – – – – – – – – – – – – Chiroptera
– – – – – – – – – – – – – – – – – – Fereungulata
– – – – – – – – – – – – – – – – – – – Ferae
– – – – – – – – – – – – – – – – – – – – †Creodonta
– – – – – – – – – – – – – – – – – – – – Carnivora
– – – – – – – – – – – – – – – – – – – – Pholidota
284 Appendix
Table A.2
(continued)
– – – – – – – – – – – – – – – – – – – Perissodactyla
– – – – – – – – – – – – – – – – – – – †Meridiungulata
– – – – – – – – – – – – – – – – – – – Cetartiodactyla
– – – – – – – – – – – – – – – – Euarchontoglires
– – – – – – – – – – – – – – – – – Glires
– – – – – – – – – – – – – – – – – – Lagomorpha
– – – – – – – – – – – – – – – – – – Rodentia
– – – – – – – – – – – – – – – – – Euarchonta
– – – – – – – – – – – – – – – – – – Dermoptera
– – – – – – – – – – – – – – – – – – Primates
– – – – – – – – – – – – – – – – – – – Strepsirrhini
– – – – – – – – – – – – – – – – – – – Haplorrhini
– – – – – – – – – – – – – – – – – – – – Tarsiiformes
– – – – – – – – – – – – – – – – – – – – Simiiformes
– – – – – – – – – – – – – – – – – – – – – Platyrrhini
– – – – – – – – – – – – – – – – – – – – – Catarrhini
– – – – – – – – – – – – – – – – – – – – – – Cercopithecoidea
– – – – – – – – – – – – – – – – – – – – – – Hominoidea
– – – – – – – – – – – – – – – – – – – – – – – Hylobatoidae
– – – – – – – – – – – – – – – – – – – – – – – Hominoidae
– – – – – – – – – – – – – – – – – – – – – – – – Pongidae
– – – – – – – – – – – – – – – – – – – – – – – – Hominidae
Note: Major clades used frequently in the text for comparative purposes are in boldface
type. Extinct taxa are marked with a †. Classification modified from Lecointre and Le
Guyader (2006) and Benton (2005).
Table A.3
A phylogenetic classification of plant taxa discussed in this book
Chlorobionta
– Ulvophyta
– Prasinophyta
– Streptophyta
– – Chlorokybophyta
– – Klebsormidiophyta
– – Phragmoplastophyta
– – – Zygnematophyta
– – – Plasmodesmophyta
– – – – Chaetosphaeridiophyta
– – – – Charophyta
A Phylogenetic Classification of Life 285
Table A.3
(continued)
– – – – Parenchymophyta
– – – – – Coleochaetophyta
– – – – – Embryophyta
– – – – – – Marchantiophyta
– – – – – – Stomatophyta
– – – – – – – Anthocerophyta
– – – – – – – Hemitracheophyta
– – – – – – – – Bryophyta sensu stricto
– – – – – – – – Polysporangiophyta
– – – – – – – – – †Horneophyta
– – – – – – – – – Tracheophyta
– – – – – – – – – – Lycophyta
– – – – – – – – – – Euphyllophyta
– – – – – – – – – – – Moniliformopses
– – – – – – – – – – – – †Cladoxylopsida
– – – – – – – – – – – – Equisetophyta
– – – – – – – – – – – – Filicophyta
– – – – – – – – – – – Lignophyta
– – – – – – – – – – – – †Aneurophytales
– – – – – – – – – – – – †Archaeopteridales
– – – – – – – – – – – – Spermatophyta
– – – – – – – – – – – – – †Lyginopteridales
– – – – – – – – – – – – – †Medullosanales
– – – – – – – – – – – – – †Gigantopteridales
– – – – – – – – – – – – – Ginkgophyta
– – – – – – – – – – – – – Cycadophyta
– – – – – – – – – – – – – Pinophyta
– – – – – – – – – – – – – Gnetophyta
– – – – – – – – – – – – – Angiospermae
– – – – – – – – – – – – – – Amborellales
– – – – – – – – – – – – – – Nymphaeales
– – – – – – – – – – – – – – Austrobaileyales
– – – – – – – – – – – – – – Euangiosperms
– – – – – – – – – – – – – – – Chloranthales
– – – – – – – – – – – – – – – Ceratophyllales
– – – – – – – – – – – – – – – Magnoliidae
– – – – – – – – – – – – – – – – Magnoliales
– – – – – – – – – – – – – – – – Laurales
– – – – – – – – – – – – – – – – Canellales
– – – – – – – – – – – – – – – – Piperales
286 Appendix
Table A.3
(continued)
– – – – – – – – – – – – – – – Monocotyledons
– – – – – – – – – – – – – – – – Acorales
– – – – – – – – – – – – – – – – Alismatales
– – – – – – – – – – – – – – – – Eumonocotyledons
– – – – – – – – – – – – – – – – – Miyoshiales
– – – – – – – – – – – – – – – – – Dioscoreales
– – – – – – – – – – – – – – – – – Pandanales
– – – – – – – – – – – – – – – – – Liliales
– – – – – – – – – – – – – – – – – Asparagales
– – – – – – – – – – – – – – – – – Commelinidae
– – – – – – – – – – – – – – – Eudicots
– – – – – – – – – – – – – – – – Ranunculalaes
– – – – – – – – – – – – – – – – Sabiales
– – – – – – – – – – – – – – – – Proteales
– – – – – – – – – – – – – – – – Buxales
– – – – – – – – – – – – – – – – Trochodendrales
– – – – – – – – – – – – – – – – Core eudicots
– – – – – – – – – – – – – – – – – Berberidopsidales
– – – – – – – – – – – – – – – – – Santalales
– – – – – – – – – – – – – – – – – Dilleniales
– – – – – – – – – – – – – – – – – Caryophyllales
– – – – – – – – – – – – – – – – – Saxifragales
– – – – – – – – – – – – – – – – – Vitales
– – – – – – – – – – – – – – – – – Rosidae
– – – – – – – – – – – – – – – – – – Crossosomatales
– – – – – – – – – – – – – – – – – – Geraniales
– – – – – – – – – – – – – – – – – – Myrtales
– – – – – – – – – – – – – – – – – – Eurosids I
– – – – – – – – – – – – – – – – – – Eurosids II
– – – – – – – – – – – – – – – – – Asteridae
– – – – – – – – – – – – – – – – – – Cornales
– – – – – – – – – – – – – – – – – – Ericales
– – – – – – – – – – – – – – – – – – Euasteridae
– – – – – – – – – – – – – – – – – – – Euasterids I
– – – – – – – – – – – – – – – – – – – Euasterids II
Note: Major clades used frequently in the text for comparative purposes are in boldface
type. Extinct taxa are marked with a †. Classification modified from Lecointre and Le
Guyader (2006), Donoghue (2005), APG II (2003), and Niklas (1997).
References
Aczel, A. D. 2007. The Jesuit and the skull. New York: Riverhead Books.
Agosta, W. 1996. Bombardier beetles and fever trees: A close-up look at chemical warfare
and signals in animals and plants. Reading, Mass.: Helix Books.
Alberch, P. 1989. The logic of monsters: Evidence for internal constraint in development
and evolution. Geobios (Lyon, France): Memoire Special 12:21–57.
Albert, V. A., S. E. Williams, and M. W. Chase. 1992. Carnivorous plants: Phylogeny and
structural evolution. Science 257:1491–1495.
APG II. 2003. An update of the Angiosperm Phylogeny Group classification for the orders
and families of flowering plants: APG II. Botanical Journal of the Linnean Society
141:399–436.
Axe, D. D. 2004. Estimating the prevalence of protein sequences adopting functional
enzyme folds. Journal of Molecular Biology 341:1295–1315.
Baber, C. 2003. Cognition and tool use: Forms of engagement in human and animal use of
tools. London: Taylor and Francis.
Bailly, X., S. Vanin, C. Chabasse, K. Mizuguchi, and S. N. Vinogradov. 2008. A phylogenetic
profile of hemerythrins, the nonheme diiron binding respiratory proteins. BMC Evolution-
ary Biology 8:244, doi:10.1186/1471-2148-8-244.
Bakker, R. T. 1983. The deer flees, the wolf pursues: Incongruencies in predator-prey coevo-
lution. In Coevolution, ed. D. J. Futuyma and M. Slatkin, 350–382. Sunderland, Mass.:
Sinauer Associates, Inc.
Bakker, R. T. 1986. The dinosaur heresies. New York: William Morrow.
Ballerini, M., N. Cabibbo, R. Candelier, A. Cavagna, E. Cisbani, I. Giardina, V. Lecomte, A.
Orlandi, G. Parisi, A. Procaccini, M. Viale, and V. Zdravkovic. 2008. Interaction ruling
animal collective behavior depends on topological rather than metric distance: Evidence
from a field study. Proceedings of the National Academy of Sciences of the United States of
America 105 (4): 1232–1237.
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. M. McCall, 719–746. New York: Plenum.
Barlow, C. 2005. Let there be sight! A celebration of convergent evolution. http://www
.thegreatstory.org.
Barrett, P. M. 2005. The diet of ostrich dinosaurs (Theropoda: Ornithomimosauris). Palae-
ontology 48 (2): 347–358.
Baskin, D. G., and D. W. Goldring. 1970. Experimental studies on the endocrinology
and reproductive biology of the viviparous polychaete annelid, Nereis limnicola Johnson.
Biological Bulletin 139:461–475.
288 References
Bastian, J. 1982. Vision and electroreception: Integration of sensory information in the optic
tectum of the weakly electric fish Apteronotus albifrons. Journal of Comparative Physiol-
ogy 147A:287–297.
Bateman, R. M., and W. A. DiMichele. 1994. Heterospory: The most iterative key innovation
in the evolutionary history of the plant kingdom. Biological Reviews of the Cambridge
Philosophical Society 69:345–417.
Bennett, S. C. 1996. Aerodynamics and thermoregulatory function of the dorsal sail of
Edaphosaurus. Paleobiology 22 (4):496–506.
Benton, M. J. 2005. Vertebrate palaeontology. 3rd ed. Oxford: Blackwell Publishing.
Berg, L. S. 1922. Nomogenesis; or, evolution determined by law. Cambridge, Mass.: Massa-
chusetts Institute of Technology Press (1969 reprint).
Berkov, A., N. Rodríguez, and P. Centeno. 2008. Convergent evolution in the antennae of
a cerambycid beetle, Onychocerus albitarsis, and the sting of a scorpion. Naturwissen-
schaften 95 (3):257–261.
Berman, D. S., R. R. Reisz, D. Scott, A. C. Henrici, S. S. Sumida, and T. Martens. 2000. Early
Permian bipedal reptile. Science 290:968–972.
Bernhardt, P. 2000. Convergent evolution and adaptive radiation of beetle-pollinated
angiosperms. Plant Systematics and Evolution 222:293–320.
Blackburn, D. G. 1992. Convergent evolution of viviparity, matrotrophy, and specializations
for fetal nutrition in reptiles and other vertebrates. American Zoologist 32:313–321.
Blackburn, D. G. 1993. Chorioallantoic placentation in squamate reptiles: Structure, func-
tion, development, and evolution. Journal of Experimental Zoology 266:414–430.
Blackburn, D. G. 2000. Reptilian viviparity: Past research, future directions, and appropriate
models. Comparative Biochemistry and Physiology Part A 127A:391–409.
Blackburn, D. G., L. J. Vitt, and C. A. Beuchat. 1984. Eutherian-like reproductive specializa-
tions in a viviparous reptile. Proceedings of the National Academy of Sciences of the United
States of America 81:4860–4863.
Block, B. A., J. R. Finnerty, A. F. R. Stewart, and J. Kidd. 1993. Evolution of endothermy in
fish: Mapping physiological traits on a molecular phylogeny. Science 260:210–214.
Blumberg, M. S. 2009. Freaks of nature: What anomalies tell us about development and
evolution. Oxford: Oxford University Press.
Boomsma, J. J. 2009. Lifetime monogamy and the evolution of eusociality. Philosophical
Transactions of the Royal Society of London: Series B, Biological Sciences B364:3191–
3207.
Bork, P., C. Sander, and A. Valencia. 1993. Convergent evolution of similar enzymatic func-
tion on different protein folds: The hexokinase, ribokinase, and galactokinase families of
sugar kinases. Protein Science 2:31–40.
Bowmaker, J. K., and D. M. Hunt. 2006. Evolution of vertebrate visual pigments. Current
Biology 16 (13):R484–R489.
Boyce, C. K., and A. H. Knoll. 2002. Evolution of developmental potential and the
multiple independent origins of leaves in Paleozoic vascular plants. Paleobiology 28
(1):70–100.
Brady, S. G. 2003. Evolution of the army ant syndrome: The origin and long-term
evolutionary stasis of a complex of behavioral and reproductive adaptations. Proceedings
of the National Academy of Sciences of the United States of America 100 (11):6575–
6579.
Brady, S. G., S. Sipes, A. Pearson, and B. N. Danforth. 2006. Recent and simultaneous origins
of eusociality in halictid bees. Philosophical Transactions of the Royal Society of London:
Series B, Biological Sciences B273:1643–1649.
Brakefield, P. M. 2006. Evo-devo and constraints on selection. Trends in Ecology and Evolu-
tion 21 (7):362–368.
References 289
Budelmann, B., and H. Bleckmann. 1988. A lateral line analogue in cephalopods: Water
waves generate microphonic potentials in the epidermal head lines of Sepia and Lolligun-
cula. Journal of Comparative Physiology 164A:1–5.
Buhl, J., D. J. T. Sumpter, I. D. Couzin, J. J. Hale, E. Despland, E. R. Miller, and S. J. Simpson.
2006. From disorder to order in marching locusts. Science 312:1402–1406.
Bull, J., and J. Farrand. 1988. The Audubon Society field guide to North American birds. New
York: Alfred A. Knopf.
Bush, A. M., R. K. Bambach, and G. M. Daley. 2007. Changes in theoretical ecospace utiliza-
tion in marine fossil assemblages between the mid-Paleozoic and late Cenozoic. Paleobiol-
ogy 33 (1):76–97.
Byrne, M. 2005. Viviparity in the sea star Cryptasterina hysteria (Asterinidae): Conserved
and modified features in reproduction and development. Biological Bulletin 208:81–91.
Camazine, S., J.-L. Deneubourg, N. R. Franks, J. Sneyd, G. Theraulaz, and E. Bonabeau.
2001. Self-organization in biological systems. Princeton: Princeton University Press.
Cameron, K. M., K. J. Wurdack, and R. W. Jobson. 2002. Molecular evidence for the common
origin of snap-traps among carnivorous plants. American Journal of Botany 89 (9):
1503–1509.
Cameron, S. A., and P. Mardulyn. 2001. Multiple molecular data sets suggest independent
origins of highly eusocial behavior in bees (Hymenoptera: Apinae). Systematic Biology 50
(2):194–214.
Carrano, M. T., C. M. Janis, and J. J. Sepkoski. 1999. Hadrosaurs as ungulate parallels: Lost
lifestyles and deficient data. Acta Palaeontologica Polonica 44 (3):237–261.
Carroll, L. [C. W. Dodgson]. 1865. Alice’s adventures in Wonderland. London: Macmillan.
Carroll, R. L., and W. Lindsay. 1985. Cranial anatomy of the primitive reptile Procolophon.
Canadian Journal of Earth Sciences 22:1571–1587.
Carroll, S. B. 2006. Endless forms most beautiful: The new science of EvoDevo. New York:
W. W. Norton and Co.
Carter, A. M. 2008. What fossils can tell us about the evolution of viviparity and placenta-
tion. Placenta 29:930–931.
Cashmore, A. R., J. A. Jarillo, Y.-J. Wu, and D. Liu. 1999. Cryptochromes: Blue light recep-
tors for plants and animals. Science 284:760–765.
Castelvecchi, D. 2007. Alien pizza, anyone? Biochemistry may have taken a different turn
on other worlds. Science News 172:107–109.
Casti, J. L. 1989. Paradigms lost: Tackling the unanswered mysteries of modern science.
New York: Avon Bard.
Castoe, T. A., A. P. J. de Koning, H.-M. Kim, B. P. Noonan, G. Naylor, Z. J. Jiang, C. L. Par-
kinson, and D. D. Pollock. 2009. Evidence for an ancient adaptive episode of convergent
molecular evolution. Proceedings of the National Academy of Sciences of the United States
of America 106 (22):8986–8991.
Castoe, T. A., A. P. J. de Koning, and D. D. Pollock. 2010. Adaptive molecular convergence:
Molecular evolution versus molecular phylogenetics. Communicative and Integrative
Biology 3 (1):67–69.
Charnock, S. J., I. E. Brown, J. P. Turkenburg, G. W. Black, and G. J. Davies. 2002. Convergent
evolution sheds light on the anti-β-elimination mechanism common to family 1 and 10
polysaccharide lyases. Proceedings of the National Academy of Sciences of the United States
of America 99 (10):12067–12072.
Chen, L., A. L. DeVries, and C.-H. C. Cheng. 1997. Convergent evolution of antifreeze
glycoproteins in Antarctic notothenioid fish and Arctic cod. Proceedings of the National
Academy of Sciences of the United States of America 94:3817–3822.
Cheng, Y.-N., X.-C. Wu, and Q. Ji. 2004. Triassic marine reptiles gave birth to live young.
Nature 432:383–386.
290 References
Cimino, G., S. De Rosa, S. De Stefano, R. Morrone, and G. Sodano. 1985. The chemical
defense of nudibranch molluscs: Structure, biosynthetic origin and defensive properties of
terpenoids from the dorid nudibranch Dendrodoris grandiflora. Tetrahedron 41
(6):1093–1100.
Clack, J. A. 2002. Gaining ground: The origin and evolution of tetrapods. Bloomington:
Indiana University Press.
Claessens, L. P. A. M., P. M. O’Connor, and D. M. Unwin. 2009. Respiratory evolution
facilitated the origin of pterosaur flight and aerial gigantism. PLoS ONE 4 (2):e4497. www
.plosone.org.
Clark, J. M., L. L. Jacobs, and W. R. Downs. 1989. Mammal-like dentition in a Mesozoic
crocodylian. Science 244:1064–1066.
Clark, R. N., J. M. Curchin, J. W. Barnes, R. Jauman, L. Sonderblom, D. P. Cruikshank,
R. H. Brown, et al. 2010. Detection and mapping hydrocarbon deposits on Titan. Journal
of Geophysical Research-Planets, doi:10.1029/2009JE003369.
Clayton, N. S., and Emery, N. J. 2008. Canny corvids and political primates: A case for
convergent evolution in intelligence. In The deep structure of biology, ed. S. Conway Morris,
128–142. West Conshohocken, Pa.: Templeton Foundation Press.
Colosimo, P. F., K. E. Hosemann, S. Balabhadra, G. Villarreal, M. Dickson, J. Grimwood,
J. Schmutz, R. M. Myers, D. Schluter, and D. M. Kingsley. 2005. Widespread parallel
evolution in sticklebacks by repeated fixation of Ectodysplasin alleles. Science 307:1928–
1933.
Conradt, L., and T. J. Roper. 2005. Consensus decision making in animals. Trends in Ecology
and Evolution 20 (8):449–456.
Conway Morris, S. 2003. Life’s solution: Inevitable humans in a lonely universe. Cambridge:
Cambridge University Press.
Conway Morris, S. 2008. Evolution and convergence: Some wider considerations. In The
deep structure of biology, ed. S. Conway Morris, 46–67. West Conshohocken, Pa.: Templeton
Foundation Press.
Cooper, A., C. Lalueza-Fox, S. Anderson, A. Rambaut, J. Austin, and R. Ward. 2001. Com-
plete mitochondrial genome sequences of two extinct moas clarify ratite evolution. Nature
409:704–707.
Couzin, I. D. 2007. Collective minds. Nature 445:715.
Couzin, I. D., and J. Krause. 2003. Self-organization and collective behavior in vertebrates.
Advances in the Study of Behavior 32:1–75.
Crespi, B. J., D. A. Carmean, and T. W. Chapman. 1997. Ecology and evolution of galling
thrips and their allies. Annual Review of Entomology 42:51–71.
Crick, F. H. C. 1988. What mad pursuit: A personal view of scientific discovery. New York:
Basic Books.
Cronk, Q., and I. Ojeda. 2008. Bird-pollinated flowers in an evolutionary and molecular
context. Journal of Experimental Botany 59 (4):715–727.
Darwin, C. 1859. On the origin of species by means of natural selection, or the preservation
of favoured races in the struggle for life. London: John Murray.
Darwin, C. 1875. Insectivorous plants. London: John Murray.
Daugherty, C. H., G. W. Gibbs, and R. A. Hitchmough. 1993. Mega-island or micro-conti-
nent? New Zealand and its fauna. Trends in Ecology and Evolution 8 (2):437–442.
Davies, P. L., J. Baardsnes, M. J. Kuiper, and V. K. Walker. 2002. Structure and function of
antifreeze proteins. Philosophical Transactions of the Royal Society of London: Series B,
Biological Sciences B357: 927–935.
Davis, C. C., P. K. Endress, and D. A. Baum. 2008. The evolution of floral gigantism. Current
Opinion in Plant Biology 11:49–57.
References 291
Fitzhugh, K. 1991. Further revisions of the Sabellidae subfamilies and cladistic relationships
among the Fabriciinae (Annelida: Polychaeta). Zoological Journal of the Linnean Society
102:305–332.
Flemming, A. F., and D. G. Blackburn. 2003. Evolution of placental specializations in vivipa-
rous African and South American lizards. Journal of Experimental Zoology 299A:33–47.
Fletcher, G. L., C. L. Hew, and P. L. Davies. 2001. Antifreeze proteins of teleost fishes.
Annual Review of Physiology 63:359–390.
Foster, K. W., and R. D. Smyth. 1980. Light antennas in phototactic algae. Microbiological
Reviews 44 (4):572–630.
Freeland, S. J., and L. D. Hurst. 1998. The genetic code is one in a million. Journal of
Molecular Evolution 47:238–248.
Gartrell, B. D. 2000. The nutritional, morphologic, and physiologic bases of nectarivory in
Australian birds. Journal of Avian Medicine and Surgery 14 (2):85–94.
Gaudin, T. J., and D. G. Branham. 1998. The phylogeny of the Myrmecophagidae (Mam-
malia, Xenarthra, Vermilingua) and the relationship of Eurotamandua to the Vermilingua.
Journal of Mammalian Evolution 5 (3):237–265.
Gensel, P. G., and A. E. Kasper. 2005. A new species of the Devonian lycopod genus,
Leclercqia, from the Emsian of New Brunswick, Canada. Review of Palaeobotany and
Palynology 137:105–123.
Givnish, T. J., J. C. Pires, S. W. Graham, M. A. McPherson, L. M. Prince, T. B. Patterson, H.
S. Rai, et al. 2005. Repeated evolution of net venation and fleshy fruits among monocots
in shaded habitats confirms a priori predictions: Evidence from an ndhF phylogeny. Philo-
sophical Transactions of the Royal Society of London: Series B, Biological Sciences
B272:1481–1490.
Gould, E. 1965. Evidence for echolocation in the Tenrecidae of Madagascar. Proceedings
of the American Philosophical Society 109 (6):352–360.
Gould, J. L., and C. G. Gould. 2007. Animal architects: Building and the evolution of intel-
ligence. New York: Basic Books.
Gould, S. J. 1989. Wonderful life: The Burgess Shale and the nature of history. New York:
W. W. Norton.
Grant, K. A., and V. Grant. 1968. Hummingbirds and their flowers. New York: Columbia
University Press.
Gregory, J. E., A. Iggo, A. K. McIntyre, and U. Proske. 1987. Electroreceptors in the platypus.
Nature 326:386–387.
Grinspoon, D. 2003. Lonely planets: The natural philosophy of alien life. New York:
HarperCollins Publishers Inc.
Gupta, A. K. 2004. Origin of agriculture and domestication of plants and animals linked to
early Holocene climate amelioration. Current Science 87 (1):54–59.
Hagan, H. R. 1951. Embryology of the viviparous insects. New York: Ronald Press.
Hamilton, W. D. 1964. The genetical evolution of social behavior I, II. Journal of Theoretical
Biology 7:1–52.
Hansell, M. 2005. Animal architecture. Oxford: Oxford University Press.
Hansell, M., and G. D. Ruxton. 2008. Setting tool use within the context of animal construc-
tion behaviour. Trends in Ecology and Evolution 23 (2):73–78.
Hansen, D. M., and M. Galetti. 2009. The forgotten megafauna. Science 324:42–43.
Hardison, R. C. 1996. A brief history of hemoglobins: Plant, animal, protist, and bacteria.
Proceedings of the National Academy of Sciences of the United States of America
93:5675–5679.
Harshman, J., E. L. Braun, M. J. Braun, C. J. Huddleston, R. C. K. Bowie, J. L. Chojnowski,
S. J. Hackett, et al. 2008. Phylogenetic evidence for multiple losses of flight in ratite birds.
References 293
King, N., C. T. Hittinger, and S. B. Carroll. 2003. Evolution of key cell signaling and adhesion
protein families predates animal origins. Science 301:361–363.
Kivic, P. A., and P. L. Walne. 1983. Algal photosensory apparatus probably represent mul-
tiple parallel evolutions. Bio Systems 16:31–38.
Klok, C. J., R. D. Mercer, and S. L. Chown. 2002. Discontinuous gas-exchange in centipedes
and its convergent evolution in tracheated arthropods. Journal of Experimental Biology
205:1019–1029.
Koenigswald, W. v., K. D. Rose, L. Grande, and R. D. Martin. 2005. First apatemyid skeleton
from the Lower Eocene Fossil Butte Member, Wyoming (USA), compared to the Euro-
pean apatemyid from Messel, Germany. Palaeontographica Abteilung A272:149–169.
Koenigswald, W. v., and H.-P. Schierning. 1987. The ecological niche of an extinct group of
mammals, the early Tertiary apatemyids. Nature 326:595–597.
Kornegay, J. R., J. W. Schilling, and A. C. Wilson. 1994. Molecular adaptation of a leaf-eating
bird: Stomach lysozyme of the Hoatzin. Molecular Biology and Evolution 11 (6):921–928.
Kuhn-Schnyder, E., and H. Rieber. 1986. Handbook of paleozoology. Baltimore: Johns
Hopkins University Press.
Kunz, T. H., M. S. Fujita, A. P. Brooke, and G. R. McCracken. 1994. Convergence in tent
architecture and tent-making behavior among Neotropical and Paleotropical bats. Journal
of Mammalian Evolution 2:57–58.
Kuroki, T. 1967. Theoretical analysis of the role of the lateral line in directional hearing. In
Lateral line detectors, ed. P. H. Cahn, 217–237. Bloomington: Indiana University Press.
Labandeira, C. C., J. Kvač ek, and M. B. Mostovski. 2007. Pollination drops, pollen, and insect
pollination of Mesozoic gymnosperms. Taxon 56:663–695.
Laughlin, S., A. D. Blest, and S. Stowe. 1980. The sensitivity of receptors in the posterior
median eye of the nocturnal spider, Dinopis. Journal of Comparative Physiology
141:53–65.
Lawn, R. M., K. Schwartz, and L. Patthy. 1997. Convergent evolution of apolipoprotein(a)
in primates and hedgehog. Proceedings of the National Academy of Sciences of the United
States of America 94:11992–11997.
Lecointre, G., and H. Le Guyader. 2006. The tree of life: A phylogenetic classification. Cam-
bridge, Mass.: Belknap Press of Harvard University Press.
Lee, M. S. Y. 1998. Convergent evolution and character correlation in burrowing reptiles:
Towards a resolution of squamate relationships. Biological Journal of the Linnean Society
65:369–453.
Lee, M. S. Y., and R. Shine. 1998. Reptilian viviparity and Dollo’s law. Evolution: Interna-
tional Journal of Organic Evolution 52 (5):1441–1450.
Loewer, P. 1995. Seeds: The definitive guide to growing, history, and lore. Portland, Ore.:
Timber Press.
Logsdon, J. M., and W. R. Doolittle. 1997. Origin of antifreeze protein genes: A cool tale in
molecular evolution. Proceedings of the National Academy of Sciences of the United States
of America 94:3485–3487.
Long, J. A., K. Trinajstic, G. C. Young, and T. Senden. 2008. Live birth in the Devonian
period. Nature 453:650–653.
Longrich, N. R., and P. J. Currie. 2008. Albertonykus borealis, a new alvarezsaur (Dinosauria:
Theropoda) from the Early Maastrichtian of Alberta, Canada: Implications for the system-
atics and ecology of the Alvarezsauridae. Cretaceous Research 30 (1):239–252.
Luo, Z.-X., R. L. Cifelli, and Z. Kielan-Jaworowska. 2001. Dual origin of tribosphenic
mammals. Nature 409:53–57.
Luo, Z.-X., Q. Ji, and C.-X. Yuan. 2007. Convergent dental adaptations in pseudo-tribos-
phenic and tribosphenic mammals. Nature 450:93–97.
References 295
Maclaurin, J. 2003. The good, the bad and the impossible. Biology and Philosophy
18:463–476.
Maclaurin, J., and K. Sterelny. 2008. What is biodiversity? Chicago: University of Chicago
Press.
Malmström, T., and R. H. H. Kröger. 2006. Pupil shapes and lens optics in the eyes of ter-
restrial vertebrates. Journal of Experimental Biology 209:18–25.
Manger, P. R., R. Collins, and J. D. Pettigrew. 1997. Histological observations on presumed
electroreceptors and mechanoreceptors in the beak skin of the long-beaked echidna,
Zaglossus bruijnii. Philosophical Transactions of the Royal Society of London: Series B,
Biological Sciences B264:165–172.
Marino, L. 2002. Convergence of complex cognitive abilities in cetaceans and primates.
Brain, Behavior and Evolution 59:21–32.
Martin, A. J. 2006. Introduction to the study of dinosaurs. Oxford: Blackwell.
Martin, V. J. 2002. Photoreceptors of cnidarians. Canadian Journal of Zoology 80:
1703–1722.
Martinez, R. N., C. L. May, and C. A. Forster. 1996. A new carnivorous cynodont from the
Ischigualasto Formation (Late Triassic, Argentina), with comments on eucynodont phylog-
eny. Journal of Vertebrate Paleontology 16:271–284.
Maryańska, T., H. Osmólska, and M. Wolsan. 2002. Avialan status for Oviraptorosauria.
Acta Palaeontologica Polonica 47:97–116.
Masonjones, H. D., and S. M. Lewis. 1996. Courtship behavior in the dwarf seahorse, Hip-
pocampus zosterae. Copeia (3):634–640.
Mayr, E. 1964. Introduction. In A facsimile of the first edition of On the origin of species,
ed. E. Mayr, vii–xxvii. Cambridge, Mass.: Harvard University Press.
Mayr, G., and D. S. Peters. 2006. Response to comment on “A well-preserved Archaeopteryx
specimen with theropod features.” Science 313: 1238c, doi: 10.1126/science.1130964.
Mayr, G., B. Pohl, and D. S. Peters. 2005. A well-preserved Archaeopteryx specimen with
theropod features. Science 310: 1483–1486.
McGhee, G. R. 1999. Theoretical morphology: The concept and its applications. New York:
Columbia University Press.
McGhee, G. R. 2001. Exploring the spectrum of existent, nonexistent and impossible bio-
logical form. Trends in Ecology and Evolution 16:172–173.
McGhee, G. R. 2007. The geometry of evolution: Adaptive landscapes and theoretical
morphospaces. Cambridge: Cambridge University Press.
McGhee, G. R. 2008. Convergent evolution: A periodic table of life? In The deep structure
of biology, ed. S. Conway Morris, 17–31. West Conshohocken, Pa.: Templeton Foundation
Press.
McGhee, G. R., P. M. Sheehan, D. J. Bottjer, and M. L. Droser. 2004. Ecological ranking of
Phanerozoic biodiversity crises: Ecological and taxonomic severities are decoupled. Pal-
aeogeography, Palaeoclimatology, Palaeoecology 211:289–297.
McIntosh, J. S. 1997. Sauropoda. In Encyclopedia of dinosaurs, ed. P. J. Currie and K. Padian,
654–658. San Diego: Academic Press.
McKay, C. P., and H. D. Smith. 2005. Possibilities for methanogenic life in liquid methane
on the surface of Titan. Icarus 178:274–276.
McKinney, F. K. 2007. The Northern Adriatic ecosystem: Deep time in a shallow sea.
New York: Columbia University Press.
McKinney, F. K., S. J. Hageman, and A. Jaklin. 2007. Crossing the ecological divide:
Paleozoic to modern marine ecosystem in the Adriatic Sea. Sedimentary Record 5
(2):4–8.
296 References
Ödeen, A., and O. Håstad. 2003. Complex distribution of avian color vision systems revealed
by sequencing the SWS1 opsin from total DNA. Molecular Biology and Evolution 20
(6):855–861.
O’Reilly, J. C., D. A. Ritter, and D. R. Carrier. 1997. Hydrostatic locomotion in a limbless
tetrapod. Nature 386:269–272.
Ostrom, J. H. 1990. Dromaeosauridae. In The Dinosauria, ed. D. B. Weishampel, P. Dodson,
and H. Osmólska, 269–279. Berkeley: University of California Press.
Peet, R. K. 1978. Ecosystem convergence. American Naturalist 112:441–459.
Pianka, E. R. 1978. Evolutionary ecology. 2nd ed. New York: Harper and Row.
Pianka, E. R., and W. S. Parker. 1975. Ecology of horned lizards: A review with special refer-
ence to Phrynosoma platyrhinos. Copeia 1975:141–162.
Piatigorsky, J. 2007. Gene sharing and evolution. Cambridge, Mass.: Harvard University
Press.
Poelwijk, F. J., D. J. Kiviet, D. M. Weinreich, and S. J. Tans. 2007. Empirical fitness landscapes
reveal accessible evolutionary paths. Nature 445:383–386.
Popov, I. Y. 2002. “Periodical systems” in biology (a historical issue). Verhandlungen zur
Geschichte und Theorie der Biologie 9:55–68.
Prior, H., A. Schwarz, and O. Güntürkün. 2008. Mirror-induced behavior in the magpie
(Pica pica): Evidence of self-recognition. PLoS Biology 6 (8):1642–1650.
Pritchard, G., M. H. McKee, E. M. Pike, G. J. Scrimgeour, and J. Zloty. 1993. Did the first
insects live in water or in air? Biological Journal of the Linnean Society 49:31–44.
Proske, U., and E. Gregory. 2003. Electrolocation in the platypus: Some speculations. Com-
parative Biochemistry and Physiology: Part A, Molecular and Integrative Physiology 136
(4):821–825.
Protas, M. E., C. Hersey, D. Kochanek, Y. Zhou, H. Wilkens, W. R. Jeffery, L. I. Zon, R.
Borowsky, and C. J. Tabin. 2005. Genetic analysis of cavefish reveals molecular convergence
in the evolution of albinism. Nature Genetics 38(1): 107–111.
Prothero, D. R., and R. M. Schoch. 2002. Horns, tusks, and flippers: The evolution of hoofed
mammals. Baltimore: Johns Hopkins University Press.
Prud’homme, B., and S. B. Carroll. 2006. Monkey see, monkey do. Nature Genetics 38
(7):740–741.
Purnell, M. A. 1995. Large eyes and vision in conodonts. Lethaia 28:187–188.
Raafat, R. M., N. Chater, and C. Frith. 2009. Herding in humans. Trends in Cognitive
Sciences 13 (10):420–428.
Rasmussen, D. T., E. L. Simons, F. Hertel, and A. Judd. 2001. Hindlimb of a giant terrestrial
bird from the Upper Eocene, Fayum, Egypt. Palaeontology 44 (2):325–337.
Reardon, D., and G. K. Farber. 1995. The structure and evolution of α/β barrel proteins.
Journal of the Federation of American Societies for Experimental Biology 9:497–503.
Reep, R. L., C. D. Marshall, and M. L. Stoll. 2002. Tactile hairs on the postcranial body in
Florida manatees: A mammalian lateral line? Brain, Behavior and Evolution 59:141–154.
Reiss, J. O. 2009. Not by design: Retiring Darwin’s watchmaker. Berkeley: University of
California Press.
Rey, M., S. Ohno, J. A. Pintor-Toro, A. Llobell, and T. Benitez. 1998. Unexpected homology
between inducible cell wall protein QID74 of filamentous fungi and BR3 salivary protein
of the insect Chironomus. Proceedings of the National Academy of Sciences of the United
States of America 95:6212–6216.
Reznick, D. N., M. Mateos, and M. S. Springer. 2002. Independent origins and rapid evolu-
tion of the placenta in the fish genus Poeciliopsis. Science 298:1018–1020.
Rich, T. H., J. A. Hopson, A. M. Musser, T. F. Flannery, and P. Vickers-Rich. 2005. Indepen-
dent origins of middle ear bones in monotremes and therians. Science 307:910–914.
298 References
Robert, D., J. Amoroso, and R. R. Hoy. 1992. The evolutionary convergence of hearing in
a parasitoid fly and its cricket host. Science 258:1135–1137.
Robert, D., and U. Willi. 2000. The histological architecture of the auditory organs in the
parasitoid fly Ormia ochracea. Cell and Tissue Research 301:447–457.
Robinson, P. L. 1956. An unusual saurposid dentition. Zoological Journal of the Linnean
Society 43:283–293.
Robson, P., G. M. Wright, J. H. Youson, and F. W. Keeley. 2000. The structure and organiza-
tion of lamprin genes: Multiple-copy genes with alternative splicing and convergent evolu-
tion with insect structural proteins. Molecular Biology and Evolution 17 (11):1739–1752.
Rodríguez-Trelles, F., R. Tarrio, and F. J. Ayala. 2003. Convergent neofunctionalization by
positive Darwinian selection after ancient recurrent duplications of the xanthine dehydrog-
enase gene. Proceedings of the National Academy of Sciences of the United States of America
100 (23):13413–13417.
Roff, D. A. 1994. The evolution of flightlessness: Is history important? Evolutionary Ecology
8:639–657.
Rokas, A., and S. B. Carroll. 2008. Frequent and widespread parallel evolution of protein
sequences. Molecular Biology and Evolution 25 (9):1943–1953.
Roots, C. 2006. Flightless birds. Westport, Conn.: Greenwood Press.
Rose, R. D., and R. J. Emry. 2005. Extraordinary fossorial adaptations in the Oligocene
palaeanodonts Epoicotherium and Xenocranium (Mammalia). Journal of Morphology
175:33–56.
Rowe, T. 1993. Phylogenetic systematics and the early history of mammals. In Mammal
phylogeny: Mesozoic differentiaton, multituberculates, monotremes, early therians, and mar-
supials, ed. F. S. Szalay, M. J. Novacek, and M. C. McKenna, 129–145. Berlin: Springer-
Verlag.
Ruse, M. 2003. Darwin and design: Does evolution have a purpose? Cambridge, Mass.:
Harvard University Press.
Sage, R. F. 1999. Why C4 photosynthesis? In C4 plant biology, ed. R. F. Sage and R. K.
Monson, 3–16. San Diego: Academic Press.
Sage, R. F. 2001a. C4 plants. In Encyclopedia of biodiversity. Vol. 1, ed. S. A. Levin, 575–598.
San Diego: Academic Press.
Sage, R. F. 2001b. Environmental and evolutionary preconditions for the origin and diver-
sification of the C4 photosynthetic syndrome. Plant Biology 3:202–213.
Salehi-Ashtiani, K., and J. W. Szostak. 2001. In vitro evolution suggests multiple origins of
the hammerhead ribozyme. Nature 414:82–84.
Salvini-Plawen, L. von. 2008. Photoreception and the polyphyletic evolution of photore-
ceptors (with special reference to Mollusca). American Malacological Bulletin 26
(1/2):83–100.
Salvini-Plawen, L. von, and E. Mayr. 1977. On the evolution of photoreceptors and eyes.
Evolutionary Biology 10:207–263.
Scantlebury, M., J. R. Speakman, M. K. Oosthuizen, T. J. Roper, and N. C. Bennett. 2006.
Energetics reveal physiologically distinct castes in a eusocial mammal. Nature
440:795–797.
Schierwater, B., M. Eitel, W. Jakob, H.-J. Osigus, H. Hadrys, S. L. Dellaporta, S.-O. Koloko-
tronis, and R. DeSalle. 2009. Concatenated analysis sheds light on early metazoan evolution
and fuels a modern “urmetazoan” hypothesis. PLoS Biology 7 (1):36–44.
Schmitz, A., M. Gebhardt, and H. Schmitz. 2008. Microfluidic photomechanic infrared
receptors in a pyrophilous flat bug. Naturwissenschaften 95:455–460.
Schneider, E. L., and S. Carlquist. 1995. Vessel origins in Nymphaeaceae: Euryale and
Victoria. Botanical Journal of the Linnean Society 119:185–193.
References 299
Schultz, T. R., and S. G. Brady. 2008. Major evolutionary transitions in ant agriculture.
Proceedings of the National Academy of Sciences of the United States of America 105
(14):5435–5440.
Seelig, C. 1956. Helle Zeit—Dunkle Zeit. In Memoriam Albert Einstein. Zurich: Europa
Verlag.
Sereno, P. 1997. Psittacosauridae. In Encyclopedia of dinosaurs, ed. P. J. Currie and
K. Padian, 611–613. San Diego: Academic Press.
Sereno, P., Z. Xijin, and T. Lin. 2009. A new psittacosaur from Inner Mongolia and the
parrot-like structure and function of the psittacosaur skull. Philosophical Transactions of
the Royal Society of London: Series B, Biological Sciences, doi:10.1098/rspb.2009.0691.
Shen, Z., and M. Jacobs-Lorena. 1999. Evolution of chitin-binding proteins in invertebrates.
Journal of Molecular Evolution 48: 341–347.
Shi, Y., and S. Yokoyama. 2003. Molecular analysis of the evolutionary significance of
ultraviolet vision in vertebrates. Proceedings of the National Academy of Sciences of the
United States of America 100 (14):8308–8313.
Shingleton, A. W., and W. A. Foster. 2001. Behaviour, morphology and the division of labour
in two soldier-producing aphids. Animal Behaviour 62:671–679.
Shubin, N., C. Tabin, and S. Carroll. 2009. Deep homology and the origins of evolutionary
novelty. Nature 457:818–823.
Sinha, N. R., and E. A. Kellogg. 1996. Parallelism and diversity in multiple origins of C4
photosynthesis in the grass family. American Journal of Botany 83 (11):1458–1470.
Smith, J. D. 2009. The study of animal metacognition. Trends in Cognitive Sciences 13
(9):389–396.
Spinney, L. 2007. Back to their roots. New Scientist 194 (2608):48–51.
Stadler, B., and A. F. G. Dixon. 2005. Ecology and evolution of aphid-ant interactions.
Annual Review of Ecology Evolution and Systematics 36:345–372.
Steadman, D. W. 2006. Extinction and biogeography of tropical Pacific birds. Chicago:
University of Chicago Press.
Stein, W. E., F. Mannolini, L. V. Hernick, E. Landing, and C. M. Berry. 2007. Giant cladoxy-
lopsid trees resolve the enigma of the Earth’s earliest forest stumps at Gilboa. Nature
446:904–907.
Stensiö, E. 1963. The brain and the cranial nerves in fossil lower craniate vertebrates.
Skrifter utgitt av det Norske Videnskaps-Akademi i Oslo. I. Matematisk-Naturvidenskapelig
Klasse 13:1–120.
Stern, D. L. 2010. Evolution, development, and the predictable genome. Greenwood Village,
Colo.: Roberts and Company.
Stern, D. L., and V. Orgogozo. 2008. The loci of evolution: How predictable is genetic evolu-
tion? Evolution: International Journal of Organic Evolution 62 (9):2155–2177.
Stern, D. L., and V. Orgogozo. 2009. Is genetic evolution predictable? Nature 323:
746–751.
Stewart, A. 2009. Wicked plants. Chapel Hill: Algonquin Books.
Stewart, C. B., J. W. Schilling, and A. C. Wilson. 1987. Adaptive evolution in the stomach
lysozymes of foregut fermenters. Nature 330:401–404.
Strobel, D. F. 2010. Molecular hydrogen in Titan’s atmosphere: Implications of the mea-
sured tropospheric and thermospheric mole fractions. Icarus, doi:10.1016/j.icarus.2010.03.003.
Sumpter, D. J. T. 2006. The principles of collective animal behaviour. Philosophical Transac-
tions of the Royal Society of London: Series B, Biological Sciences B361:5–22.
Szczupider, M. 2009. Visions of Earth: Cameroon Sanaga-Yong chimpanzee rescue center.
National Geographic 216 (5):12–13.
300 References
Terada, T., O. Nureki, R. Ishitani, A. Ambrogelly, M. Ibba, D. Söll, and S. Yokoyama. 2002.
Functional convergence of two lysyl-tRNA synthetases with unrelated topologies. Nature
Structural Biology 9 (4):257–262.
Thien, L. B., P. Bernhardt, M. S. Devall, Z.-D. Chen, Y.-B. Luo, J.-J. Fan, L.-C. Yuan, and
J. H. Williams. 2009. Pollination biology of basal angiosperms (ANITA grade). American
Journal of Botany 96 (1):166–182.
Thomas, R. D. K., and W.-E. Reif. 1993. The skeleton space: A finite set of organic designs.
Evolution: International Journal of Organic Evolution 47:341–360.
Thompson, D’A. W. 1942. On growth and form. Cambridge: Cambridge University
Press.
Thompson, M. B., and B. K. Speake. 2006. A review of the evolution of viviparity in lizards:
Structure, function and physiology of the placenta. Journal of Comparative Physiology
176B:179–189.
Thorne, B. L., and J. F. A. Traniello. 2003. Comparative social biology of basal taxa of ants
and termites. Annual Review of Entomology 48:283–306.
Tomescu, A. M. F. 2008. Megaphylls, microphylls and the evolution of leaf development.
Trends in Plant Science 14 (1):5–12.
Tripp, B. C., K. Smith, and J. G. Ferry. 2001. Carbonic anhydrase: New insights for an ancient
enzyme. Journal of Biological Chemistry 276 (52):48615–48618.
Turner, A. 1997. The big cats and their fossil relatives. New York: Columbia University
Press.
van Holde, K. E., K. I. Miller, and H. Decker. 2001. Hemocyanins and invertebrate evolu-
tion. Journal of Biological Chemistry 276 (19):15563–15566.
van Schaik, C. P., E. A. Fox, and A. F. Sitompui. 1996. Manufacture and use of tools in wild
Sumatran orangutans. Naturwissenschaften 83 (4):186–188.
Van Valen, L. 1973. A new evolutionary theory. Evolutionary Theory 1:1–30.
Varricchio, D. J. 1997. Troodontidae. In Encyclopedia of dinosaurs, ed. P. J. Currie and
K. Padian, 749–754. San Diego: Academic Press.
Vermeij, G. J. 2006. Historical contingency and the purported uniqueness of evolutionary
innovations. Proceedings of the National Academy of Sciences of the United States of
America 103 (6):1804–1809.
von Frisch, K. 1974. Animal architecture. New York: Harcourt Brace Jovanovich.
Wade, N. 2006. Before the dawn: Recovering the lost history of our ancestors. New York:
Penguin Press.
Wake, M. H. 1992. Evolutionary scenarios, homology and convergence of structural spe-
cializations for vertebrate viviparity. American Zoologist 32:256–263.
Wake, M. H. 1993. Evolution of oviductal gestation in amphibians. Journal of Experimental
Zoology 266:394–413.
Wald, G., and S. Raypart. 1977. Vision in annelid worms. Science 196:1434–1439.
Ward, P. D. 2005. Life as we do not know it. New York: Viking Penguin.
Watson, T. 1993. Why some fishes are hotheads. Science 260:160–161.
Weinreich, D. M., N. F. Delaney, M. A. DePristo, and D. L. Hartl. 2006. Darwinian evolution
can follow only very few mutational paths to fitter proteins. Science 312:111–114.
Weishampel, D. B., and J. R. Horner. 1990. Hadrosauridae. In The Dinosauria, ed. D. B.
Weishampel, P. Dodson, and H. Osmólska, 534–561. Berkeley: University of California
Press.
Went, F. W. 1971. Parallel evolution. Taxon 20:197–226.
West, G. B., J. H. Brown, and B. J. Enquist. 1999. The fourth dimension of life: Fractal
geometry and allometric scaling of organisms. Science 284:1677–1679.
References 301
Whitehead, H. 2008. Social and cultural evolution in the ocean: Convergences and contrasts
with terrestrial systems. In The deep structure of biology, ed. S. Conway Morris, 143–160.
West Conshohocken, Pa.: Templeton Foundation Press.
Wicher, K. B., and E. Fries. 2007. Convergent evolution of human and bovine haptoglobin:
Partial duplication of the genes. Journal of Molecular Evolution 65:373–379.
Williams, D. S., and P. McIntyre. 1980. The principal eyes of a jumping spider have a tele-
photo component. Nature 288:578–580.
Wilson, E. O. 1980. Sociobiology: The abridged edition. Cambridge, Mass.: Harvard Univer-
sity Press.
Wilson, E. O., and B. Hölldobler. 2005. Eusociality: Origin and consequences. Proceedings
of the National Academy of Sciences of the United States of America 102 (38):
13367–13371.
Wourms, J. P., and J. Lombardi. 1992. Reflections on the evolution of piscine viviparity.
American Zoologist 32:276–293.
Wu, X.-C., H.-D. Sues, and A. Sun. 1995. A plant-eating crocodyliform reptile from the
Cretaceous of China. Nature 376:678–680.
Wyndham, J. 1951. The day of the triffids. New York: Doubleday and Company.
Yanoviak, S. P., R. Dudley, and M. Kaspart. 2005. Directed aerial descent in canopy ants.
Nature 433:624–626.
Yeager, M., and A. L. Hughes. 1999. Evolution of the mammalian MHC: Natural selection,
recombination, and convergent evolution. Immunological Reviews 167:45–58.
Yee, M. S. Y. 1999. Molecular phylogenies become functional. Trends in Ecology and Evolu-
tion 14 (5):177–178.
Yokoyama, R., and S. Yokoyama. 1990. Convergent evolution of the red- and green-like
visual pigment genes in fish, Astyanax fasciatus, and human. Proceedings of the National
Academy of Sciences of the United States of America 87:9315–9318.
Zakon, H. H., and G. A. Unguez. 1999. Development and regeneration of the electric organ.
Journal of Experimental Biology 202:1427–1434.
Zhang, J. 2003. Paleomolecular biology unravels the evolutionary mystery of vertebrate
UV vision. Proceedings of the National Academy of Sciences of the United States of America
100 (14):8045–8047.
Zhang, J. 2006. Parallel adaptive origins of digestive RNases in Asian and African leaf
monkeys. Nature Genetics 38 (7):819–823.
Zhang, J., and S. Kumar. 1997. Detection of convergent and parallel evolution at the amino
acid sequence level. Molecular Biology and Evolution 14 (5):527–536.
Zusi, L., and D. Bridge. 1981. On the slit pupil of the Black Skimmer (Rhynchops niger).
Journal of Field Ornithology 52 (4):338–340.
Index of Common Names
Wolf
European, 158, 166, 236
marsupial, 158–159
Wolverine, 158–159, 166
Wombat, 158, 160
Woodchuck, 158, 160, 161–162
Woodpecker, red-headed, 140, 142
Worms
bristle, 71, 84
velvet, 76, 84
Wren, Carolina, 216
Zokor, Siberian, 29
Index of Species
Varanus komodensis, 28
Vatica parvifolia, 121
Velociraptor mongoliensis, 44, 166
Veratrum viride, 109
Vermiuora peregrina, 146
Vespula maculifrons, 229
Vestiaria coccinea, 146
Index of Topics
nonexistent, 12, 16, 19, 21, 25, 29, 38, 66, Methane, 254–255
96, 103, 170–171, 174–175, 250–254, Mimicry, 8–9
260–264 Minds, convergent, 209–243
Freedom, 265–277 Mitochondrial DNA, 185–186
Fruit, 127, 129–133 Molecules
Functional constraint, 7–8, 13, 17, 22, 25, convergent, 177–209
29, 38, 46, 66, 96, 125, 245–264 DNA, 178–189
enzymes, 199–208
Gastric mills, 50–53 proteins, 189–199
Gliding forms, 20–22 Morphology
Ground-dwelling birds, 147–155 convergent (see Convergent evolution)
impossible, 250–252
Hammerhead ribozyme, 204–205 nonexistent, 12, 16, 19, 21, 25, 29, 38, 66,
Haptoglobin, 192–193 96, 103, 170–171, 174–175, 250–254,
Hemerythrin, 192–193 260–264
Hemocyanin, 192–193 Morphospaces. See Theoretical
Hemoglobin, 192–193 morphospaces
Herbivores, 50–60 Mourning, 240–242
Herding, 231–235 Myrmecochory, 131–133
Heterosporous reproduction, 115–118
Homology, 2 Nervous systems, 80–81
“deep homology,” 6, 38, 71, 101, 257–260 New Zealand ecosystems, 156–157
Homoplasy, 2 Niches
Horizontal-slit pupils, 32, 36, 61 carnivorous plant, 135–138
Hunting, 235–238 insectivore, 140–142
Hydrogen, 255 myrmecophage, 142–144
necrophage, 144–145, 235, 237–238
Impossible forms, 250–252 nectarivore, 145–147
Infrared detection, 37–38 parasitic plant, 139–140
Insectivore niche, 140–142 periodic table of, 168–175
Intrinsic constraints, 253 Nonexistent forms, 12, 16, 19, 21, 25, 29, 66,
103, 170–171, 174–175, 250–252, 254,
Lateral-line system, 72–73 260–264
Leaves, 97–99
Life Occlusive teeth, 53–57
carbon-based, 254–256 Opsin, 179–183
periodic table of, 260–264 Ornithophilous flowers, 123–125
silicon-based, 254–256, 276 Oxygen, 255
Titan and, 254–255, 264, 276
Locomotion Pancreatic ribonuclease, 184–185
burrowing, 26–29 Parallel evolution, 3–7, 71, 81, 178, 183,
flying, 17–20, 260–264 188
gliding, 20–22 recognition of, 3–7
swimming, 13–17, 260–264 Parasitic plants, 139–140
walking, 22–26, 260–264 Parrotlike beaks, 59–60
Lysyl-tRNA synthesis, 203–204 Penis, 82–84
Peptide-bond hydrolysis, 202–203
Madagascar ecosystems, 157 Periodic table
Marine-niche ecospace, 171–175 of life, 260–264
Marsupial ecological equivalents, 160–163 of niches, 168–175
Melanin-controlling proteins, 194–195 Photosynthesis, 101–105
Mentalities Placentae, 90–91
metacognitive, 239–240 Placental ecological equivalents, 160–163
mourning, 240–242 Plants
self-aware, 238–240 arborescent, 93–97
Mesozoic-Cenozoic ecosystems, 160–163 carnivorous, 135–138
Metacognition, 239–240 convergent, 93–133
322 Index of Topics