23 Chordate Phylogeny and Development Timothy Rowe

Back

384

Chordata, our own lineage (fig. 23.1), belongs to the successively

more inclusive clades Deuterostomata, Bilateria,

Metazoa, and so forth. The organization of chordates is distinctively

different from that of its metazoan relatives, and

much of this distinction is conferred by unique mechanisms

of development (Slack 1983, Schaeffer 1987). Throughout

chordate history, modulation and elaboration of developmental

systems are persistent themes underlying diversification.

Only by understanding how ontogeny itself evolved can

we fully apprehend chordate history, diversity, and our own

unique place in the Tree of Life. My goal here is to present a

contemporary overview of chordate history by summarizing

current views on relationships among the major chordate

clades in light of a blossoming understanding of molecular,

genetic, and developmental evolution, and a wave of exciting

new discoveries from deep in the fossil record.

Chordates comprise a clade of approximately 56,000

named living species that includes humans and other animals

with a notochord—the embryological precursor of

the vertebral column. Chordate history can now be traced

across at least a half billion years of geological time, and

twice that by some estimates (Wray et al. 1996, Ayala et al.

1998, Bromham et al. 1998, Kumar and Hedges 1998, Hedges

2001). Chordates are exceptional among multicellular animals

in diversifying across eight orders of size magnitudes

and inhabiting virtually every terrestrial and aquatic environment

(McMahon and Bonner 1985). New living chordate

species are still being discovered both through traditional

explorations and as molecular analyses discover cryptic taxa

in lineages whose diversities were thought to be thoroughly

mapped. But it is unknown whether the pace of discovery is

now keeping up with the pace of extinction, which is accelerating

across most major chordate clades in the wake of

human population growth (Dingus and Rowe 1998).

Many chordate clades have long been recognized by characteristic

adult features, for instance, birds by their feathers,

mammals by their hair, or turtles by their shells. But owing

in large part to such distinctiveness, few adult morphological

features have been discovered that decisively resolve the

relationships among the chordate clades, and even after 300

years of study broad segments of chordate phylogeny remain

terra incognita.

Much of the hypothesized hierarchy of higher level chordate

relationships has been deduced from paleontology and

developmental biology (Russell 1916). Thanks to the advent

of phylogenetic systematics, both fields are expressing resurgent

interest and progress on the question of chordate phylogeny.

And, as they are becoming integrated with molecular

systematic analyses, a fundamental new understanding of

chordate evolution and development is emerging.

In most other metazoans, the adult fate of embryonic cells

is determined very early in ontogeny. However as chordate

ontogeny unfolds, the fates of embryonic cells are plastic for a

longer duration. Chordate cells differentiate as signals pass

Chordate Phylogeny and Development 385

between adjacent cells and tissues during the integration of

developing cell lineages into functioning tissues, organs, and

organ systems. Seemingly subtle modulations in early ontogeny

by this information exchange system have occurred many

times over chordate history to yield cascades of subsequent

developmental effects that underlie chordate diversity (Hall

1992). Molecular and developmental genetic studies are now

revealing the intricate details of this unique, hierarchical system

of information transfer as genes are expressed in cells and

tissues in early ontogeny. These analyses, moreover, generate

data that possess a recoverable phylogenetic signal and are yielding

fundamental insights into the evolution of development.

An important conclusion already evident is that major

innovations in chordate design were generally derived from

preexisting genetic and developmental pathways, whose alteration

transformed ancestral structures into distinctive new

features with entirely different adult functions (Shubin et al.

1997). Increase in numbers of genes was a primary mediator

of this change, and the inductive nature of chordate development

amplified that change via epigenesis, which occurs

as familiar physical forces and dynamic processes interact

with the cells and tissues of a developing organism. These

include gravity, adhesion, diffusion, mechanical loading, electrical

potentials, phase separations, differential growth among

tissues and organs, and many others (Rowe 1996a, 1996b).

Morphogenic and patterning effects are the developmental

outcomes of these recognized physical phenomena, because

they affect interactions among virtually all developing cells,

Figure 23.1. Chordate phylogeny, showing the relationships of extant lineages and the oldest

fossils, superimposed on a geological time column. Nodal numbers are keyed to text headings.

Vendian

Cambrian

Ordovician

Silurian

Devonian

Carboniferous

Permian

Triassic

Jurassic

Cretaceous

Cenozoic

Cathaymyrus

Myxinikela

Mayomyzon

Ambulacraria

Cephalochordata

Urochordata

Chordata

Euchordata

Craniata

Vertebrata

Gnathostomata

Osteichthyes

Sarcopterygii

Choanata

Tetrapoda

Amniota

Reptilia

Myxini

Petromyzontida

Chondrichthyes

Actinopterygii

Dipnoi

Actinistia

Amphibia

Mammalia

Yunannozoon

Haikouella

Haikouichthyes

Myllokunmingia

Heterostraci

Conodonta

Anaspida

Galeaspida

Osteostraci

Acanthodii

Placodermi

Psarolepis

Achoania

Onychodontiformes

Eusthenopteron

Elpistostegalia

Ichtyhostega

Acanthostega

Diadectomorpha

Seymouriamorpha

Casineria

Cheungkongella

Arkarua

Pikaia

Andreolepis

Yongolepis

Paleothyris

Archaeothyris

Millions of

Years Ago

0

600

100

200

300

400

500

Deuterostomata

20

2

4

6

8 14

16

10

Ligulalepis

12

1

21

18

3

5

7

9

11

13

15

17

19

386 The Relationships of Animals: Deuterostomes

tissues, and organs (Newman and Comper 1990). In the

inductive environment of chordate ontogeny, epigenesis has

been especially influential, triggering its own cascades of rapid

and nonlinear developmental change. Understanding how

epigenesis mediates the genetic blueprint of ontogeny is fundamental

to understanding how such diverse chordates as

sea squirts, coelacanths, and humans emerged from their

unique common ancestor.

Recognizing that most biologists reading this volume

study living organisms, the focus below is on extant taxa.

However, extinct taxa are discussed as well, and their inclusion

helps to emphasize the timing of origins of the major

extant chordate clades and to acknowledge the diversity and

antiquity of the lineages of which they are a part. Moreover,

the framework of chordate relationships presented below

came from the simultaneous consideration of all available

evidence. In resolving several parts of the chordate tree discussed

below, evidence afforded by fossils proved more important

than that derived from living species (Gauthier et al.

1988a, 1989, Donoghue et al. 1989).

Taxonomic Names, Ancestry, and Fossils

Older views of chordate relationships make reference to

groups united on general similarity or common gestalt. In

contrast, the names used below designate lineages whose

members appear to be united by common ancestry (de

Queiroz and Gauthier 1992). To avoid ambiguity, the meanings

of these names are defined in terms of particular ancestors

of two or more living taxa (i.e., node-based or crown

clade names). I follow an arbitrary but useful narrative convention

in specifying the crown clade names used below in

terms of their most recent common ancestry with humans.

For example, the name Chordata refers to the clade stemming

from the last common ancestor that humans share

with living tunicates and lancelets; the name Vertebrata designates

the clade stemming from the last common ancestor

that humans share with lampreys; and so on (fig. 23.1). This

is arbitrary in the sense that many other possible living

specifiers among amniotes (viz., birds, turtles, crocodilians,

lizards) in place of humans would designate the

same clades.

Stem-based names are used in reference to a node or terminal

taxon, plus all extinct taxa that are more closely related

to it than to some other node or terminal taxon. In the

interests of simplifying the complex taxonomy that evolved

under the Linnaean system, I follow a convention now gaining

popularity that employs the prefix “Pan-” to designate

stem + crown lineages (Gauthier and de Queiroz 2001). For

example, Pan-Mammalia refers to the clade Mammalia, plus

all extinct species closer to Mammalia than to its extant sister

taxon Reptilia. The clade Pan-Vertebrata includes Vertebrata

plus all extinct taxa closer to Vertebrata than to hagfishes,

and so forth.

Chordate Relationships

Node 1. The Chordates (Chordata)

Chordata (fig. 23.1) comprise the lineage arising from the

last common ancestor that humans share with tunicates and

lancelets. Tunicates are widely regarded as the sister taxon

to all other chordates (Gegenbaur 1878, Schaeffer 1987,

Cameron et al. 2000), and tunicate larvae are commonly

viewed as manifesting the organization of the adult ancestral

chordate (e.g., Meinertzhagen and Okamura 2001). But

some systematists contend that lancelets are the more distant

outgroup (Lшvtrup 1977, Jeffries 1979, 1980, 1986,

Jeffries and Lewis 1978). The controversy stems in part from

the fact that living adult tunicates are small and built from a

small number of cells. Even their larvae appear highly divergent

from other living chordate larvae. It now seems likely

that they were secondarily simplified in having lost half or

more of the Hox genes from the single cluster that was probably

present in deuterostomes ancestrally (Holland and

Garcia-Fernаndez 1996), hence, too, the loss of adult structures

governed by these genes. As adults, tunicates are derived

in losing the coelom and hindgut (Holland and Chen

2001) and are speculated to be pedomorphic in having lost

segmentation (Holland and Garcia-Fernаndez 1996). One

character shared by tunicates and craniates, to the exclusion

of lancelets, is expression of the Pax 2/5/8 gene in a region

of the developing brain known as the isthmocerebellarmidbrain-

hindbrain boundary. The lack of Pax 2/5/8 expression

in lancelets implies either secondary loss, or independent

expression in tunicates and craniates (Butler 2000), or that

tunicates share closer common ancestry with other chordates

than do lancelets. Having separated from other chordates by

at least a half-billion years ago (Wray et al. 1996, Bromham

et al. 1998, Kumar and Hedges 1998, Hedges 2001), and

without a useful fossil record (below), relationships among

these chordates must be viewed as tenuous (Gauthier et al.

1988a, Donoghue et al. 1989). More for narrative convenience

than conviction, I follow current convention in treating

tunicates as sister lineage to all other chordates.

Chordate Characters

The notochord. The namesake feature of chordates is a

premiere example of embryonic induction and patterning,

in which differentiation of the embryo along a dorsoventral

axis launches a cascade of subsequent developmental events

(Slack 1983, Schaeffer 1987). “Dorsalization” is controlled

by the Hedgehog gene and signaling by bone morphogenesis

protein, or BMP (Shimeld and Holland 2000). As in other

bilaterians, chordates develop from three primary embryonic

layers. These are the outer ectoderm, the inner endoderm,

and the mesoderm, which arises from cells that migrate between

the inner and outer layers. Chordate mesoderm develops

in the upper hemisphere of the embryonic gastrula,

its identity being induced partly as its cells stream across the

Chordate Phylogeny and Development 387

dorsal lip of the primordial opening (blastopore) into the

inner cavity (archenteron) of the embryo, and partly by signaling

from endoderm at the equator of the embryo (Hall

1992). Mesoderm cells reaching the dorsal midline condense

into a strip of cells known as chordamesoderm, which later

differentiates to become the notochord. The notochord in

turn induces overlying ectoderm to form the dorsal neural

plate, triggering another morphogenic chain of events as the

chordate central nervous system (CNS) differentiates and

begins to grow. In most chordates, the mesoderm immediately

adjacent to the notochord takes on special properties, as

does the ectoderm immediately adjacent to the neural plate.

Elaboration of these dorsal structures is tied closely to evolution

of the organs of information acquisition and integration,

as well as to locomotion.

The chordate central nervous system. Induction of a dorsal

neural plate is directed by the underlying chordamesoderm

(above). This is the first step of neurulation, in which

the nervous system arises, becomes organized, and helps

direct the integration of other parts of the developing embryo.

During neurulation, longitudinal neural folds arise

along the edges of the neural plate, perhaps under the direction

of the adjacent mesoderm (Jacobson 2001), and meet

on the midline to enclose a space that initially lay entirely

outside of the embryo. This “hollow” comprises the adult

ventricular system of the brain and central canal of the spinal

cord. It is lined with ciliated ependymal cells and its lumen

fills with cerebrospinal fluid. This original “periventricular”

layer becomes the primary region from which subsequent

neural cells arise in the brain (Butler and Hodos 1996).

Molecular signaling during neurulation also produces

anteroposterior regionalization in chordate embryos. The

rostral end of the central nerve cord swells to form the brain,

which differentiates into three regions that express distinct

gene families and which have distinct adult fates. The rostralmost

(diencephalic) domain of the neural tube expresses

the Otx gene family and is connected to specialized lightsensitive

cells. Behind this is a caudal (hindbrain–spinal cord)

division, in which Hox genes are active and which receives

nonvisual sensory inputs. Between the two lies an intermediate

region marked by expression of the Pax 2/5/8 patterning

gene that is more problematically compared with a region

known as the isthmocerebellar-midbrain-hindbrain boundary

and involves the ear (Meinertzhagen and Okamura 2001,

Butler 2000, Shimeld and Holland 2000). Pax 2/5/8 is expressed

in tunicates and craniates, but not lancelets (below).

Other bilaterians have a longitudinal nerve cord and brain

but it is ventrally positioned; hence, biologists long maintained

that the chordate dorsal nerve cord arose independently.

However, both brains express orthologous homeobox

genes in similar spatial patterns. For instance, the fruit fly

has a regionalized neural tube with similarities in rostrocaudal

and mediolateral specification to chordates (Arendt and

Nьbler-Jung 1999, Nielsen 1999, Butler 2000; for alternative

view, see Gerhart 2000). Its rostral brain is specified by

the regulatory gene Orthodenticle, a homologue to the chordate

Otx family genes, and it receives input from paired eyes.

This suggests a common blueprint. Biologists long found it

difficult to accept the two nerve cords as homologous owing

to their different positions relative to the mouth, but it

now appears that the deuterostome mouth is a new structure

and not homologous to the mouth in protostomes

(Nielsen 1999).

Special sensory organs of the head. An eye and ear of unique

design were probably present in chordates ancestrally. The

master control gene Pax6 is expressed during early development

in paired neural photoreceptors—eyes—in chordates

and many other bilaterians. Paired eyes and ears, however

rudimentary, were almost certainly present in chordates

ancestrally (Gehring 1998). However, Pax6 expression in

chordates is manifested in eye morphogenesis that follows a

unique hierarchy of pathways and inductive signals, and in

which considerable diversity evolved among the different

chordates lineages. Living tunicates, lancelets, and hagfish

each appear uniquely derived, leaving equivocal exactly what

type of eye was present in chordates ancestrally. In tunicates,

the larval eye forms a small vesicle that contains a sunken,

pigmented mass. Internal to the pigment lies a layer of cells

that are directed radially toward it, and overlying the pigment

are two hemispherical refractive layers (Gegenbaur 1878).

These same relationships occur in all other chordates. However,

in tunicates an optic vesicle is present only in larvae and

is generally unpaired. Nevertheless, it is an outgrowth of the

Otx-expressing region of the forebrain and it expresses Pax6,

as do the paired eyes of vertebrates and unlike the median

pineal eye (Meinertzhagen and Okamura 2001). In lancelets

there is a single, median frontal eye, which also expresses

Pax6, and like the bilateral eyes of vertebrates it is linked with

cells in the primary motor center (Lacalli 1996a, 1996b,

Butler 2000). In the case of lancelets, the forward extension

of the notochord may be implicated in secondary fusion of

the single eye. Hagfish have paired eyes, but they are poorly

developed compared with most vertebrates.

The chordate ear or otic system eventually differentiated

into the organs of both balance and hearing in vertebrates.

Adult tunicates have sensory hair cells that support a pigmented

otolith and are grouped into gelatinous copular

organs located in the atrium of the adult. These cells express

members of the Pax 2/5/8 gene family, as do the otic placodes

in craniates (but not lancelets), and in early development they

are topographically similar to craniate otic placodes. However,

placodes themselves are not yet present. Similar gene

expression, cellular organization, and topography point to

the probable homology of the otic organ in all chordates

(Shimeld and Holland 2000, Jeffries 2001, Meinertzhagen

and Okamura 2001).

Hormonal glands. Two hormonal glands arose in chordates

ancestrally to exert novel control over growth and metabolism.

The pituitary is a compound structure that forms via

the interaction between neurectoderm, which descends from

388 The Relationships of Animals: Deuterostomes

the developing brain toward the roof of the pharynx, and oral

ectoderm that folds inward to line the inside of the mouth.

Ectoderm forms Rathke’s pouch and becomes the glandular

part of the pituitary, whereas neural tissue from the floor of

the diencephalon becomes its infundibular portion. The infundibulum

is present in lancelets and craniates, but its homologue

in tunicates is unclear. However, in tunicates the

homologue of the glandular portion, known as the neural

gland, lies in the same position with respect to both brain

and pharyngeal roof (Barrington 1963, 1968, Maisey 1986).

The second hormonal gland, the endostyle, develops in

a groove in the floor of the larval pharynx in tunicates, lancelets,

and in larval lampreys. Its cells form thyroid follicles

that secrete iodine-binding hormones. Its homologue in

gnathostomes is probably the thyroid gland, which also

develops in a median out-pocketing in the floor of the pharynx,

and also forms thyroid follicles that secrete iodinebinding

hormones (Schaeffer 1987). Thyroid hormone

production is controlled in large measure by the pituitary

gland and affects growth, maintenance of general tissue

metabolism, reproductive phenomena, and in some taxa

metamorphosis.

Tadpole-shaped larva. Unlike the ciliated egg-shaped larvae

of hemichordates and echinoderms, the chordate larva

is tadpole shaped, with a swollen rostral end and a muscular

tail. The rostral end houses the brain, beneath which lie the

rostral end of the notochord, and the pharynx and gut tube.

Behind the pharynx is a tail equipped with muscle deriving

from caudal mesoderm (Maisey 1986, Schaeffer 1987). Although

lacking tails as adults, the larvae of many species have

tails of comparatively simple construction with muscle that

form bilateral bands, in contrast to the segmental muscle

blocks found in euchordates (below). A recent study of tailed

and tailless tunicate larvae (Swalla and Jeffery 1996) found

that the Manx gene is expressed in the cells of the tailed form

but it is down-regulated in the tail-less species, and that complete

loss of the tail can be attributed to disrupted expression

of the single gene. Whether Manx was central to the

origin of the tail in chordates is unknown, but this study

highlights the potential genetic simplicity underlying complex

adult structures.

Pan-Chordata

Although an extensive fossil record is known for many clades

lying within Chordata, no fossils are known at present that

lie with any certainty on its stem.

Node 2. The Tunicates or Sea Squirts (Urochordata)

Chordate species all can be distributed between the tunicates

and euchordates, its two principal sister clades (fig. 23.1).

The tunicates comprise a diverse marine clade that includes

roughly 1300 extant species distributed among the sessile

ascidians, and the pelagic salps and larvaceans (Jamieson

1991). Tunicate monophyly is well supported (Gegenbaur

1878, Maisey 1986). As adults, the tunicate body is enclosed

within the tunic, an acellular membrane made of celluloselike

tunicin. It is derived from ectoderm, and in tunicates it

may contain both amorphous and crystalline calcium carbonate

spicules (Aizenberg et al. 2002). Echinoderms possess

crystalline calcium in ectodermal structures, raising the

question of whether biomineralization was present in deuterostomes

ancestrally (see below). The tunic presents an

outwardly simple body, but it cloaks a much more complex

and derived organism. The pharynx is perforated by two pairs

of slits and is enormously enlarged for suspension feeding.

The pharynx size obliterates the coelom, a cavity inside the

body walls that surrounds the gut in tunicate larvae and most

adult chordates. Unique incurrent and excurrent pores supply

a stream of water through the huge pharynx, which in

some species serves in locomotion. All tunicates are mobile

as larvae, but not all species have larval tails. The pelagic salps

and larvaceans are thought to be more basal and to reflect

the primitive adult lifestyle.

Pan-Urochordata

The fossil record of tunicates is sparse and tentative, but

potentially long. The oldest putative tunicate, Cheungkongella

ancestralis, from the Early Cambrian of China (Shu et al.

2001a) is known from a single specimen. It evidently preserves

a two-fold division of the body into an enlarged pharyngeal

region with pharyngeal openings, a large oral siphon

surrounded by short tentacles, and a smaller excurrent siphon.

The body appears wholly enclosed in a tuniclike outer

covering. It has short tail-like attachment structure, a derived

feature placing Cheungkongella among crown tunicates. This

fossil, if properly interpreted, marks the Early Cambrian as

the minimum age of divergence of tunicates from other chordates

and implies a Precambrian origin for Chordata.

A possible stem tunicate fossil was brought to light through

a reinterpretation of Jaekelocarpus oklahomensis, a Carboniferous

“mitrate” (Dominguez et al. 2002). High-resolution

X-ray computed tomography (e.g., Rowe et al. 1995, 1997,

1999, Digital Morphology 2003) provided new details of

internal anatomy and revealed the presence of paired tunicate-

like gill skeletons. Jaekelocarpus and a number of similar,

tiny Paleozoic fossils have a calcite exoskeleton over their

head and pharynx and are generally thought to lie as stem

members of echinoderms or various basal chordate clades

(Jeffries 1986, Dominguez et al. 2002). The mitrates may

prove to be paraphyletic, and its members assignable to different

deuterostome clades. The eventual placement of all of

these fossils will have bearing on our interpretation of basal

chordate relationships, and on the structure and history of

mineralized tissues.

Node 3. Chordates with a Brain (Euchordata)

Euchordata comprise the last common ancestor that humans

share with lancelets (but see caveats above), and all of its

Chordate Phylogeny and Development 389

descendants (fig. 23.1). Apart from the tunicates and a single

ancient fossil of uncertain affinities (below), all other chordates

are members of Euchordata. Expanding on the innovations

that arose in chordates ancestrally, euchordates

manifest more complex genetic control over development.

This was accompanied by further elaboration of the CNS and

special sense organs, and a fundamental reorganization of the

trunk musculature and locomotor system.

Euchordate Characters

Increased genetic complexity I. Euchordates express Msx,

HNF-3, and Netrin genes, whereas only Hedgehog is expressed

in tunicates. This evident increase in homeobox expression

corresponds to elaborated dorsoventral patterning in the

CNS. Additional genes are also expressed in more elaborate

anteroposterior regionalization, including BF1 and Islet genes

(Holland and Chen 2001). Tunicates express only one to five

Hox genes, whereas lancelets express 10 Hox genes in one

cluster, affecting broader regions of the brain and nerve cord.

Although poorly sampled, at least one hemichordate (Saccoglossus)

expresses nine Hox genes in its single cluster. Tunicates

therefore may have lost genes that were present in

deuterostomes ancestrally (Holland and Garcia-Fernаndez

1996).

Elaboration of the brain I. Lancelets were long thought

to have virtually no brain at all, but recent structural studies

reveal an elaborate brain and several unique resemblances

to the brain in craniates (Lacalli 1996a, 1996b, Butler 2000).

Reticulospinal neurons differentiate in the hindbrain, where

they are involved in undulatory swimming and movements

associated with the startle reflex. Also present in lancelets are

homologues of trigeminal motor neurons, which are involved

in pharyngeal movement, and possibly other cranial nerves

(Fritzsch 1996, Butler 2000). Additionally, the neural tube

is differentiated into an inner ependymal cell layer (gray

matter) and synaptic outer fibrous layer (white matter; Maisey

1986) and is innervated by intermyotomal dorsal nerve roots

that carry sensory and motor fibers (Schaeffer 1987). Several

of these features lie partly or wholly within the expression

domain of Hox genes.

Elaboration of the special senses I. An olfactory organ occurs

in lancelets, in the form of the corpuscles of de Quatrefages.

These are a specialized group of anterior ectodermal

cells that send axonal projections to the CNS via the rostral

nerves. They are marked by expression of the homeobox gene

AmphiMsx, which is also expressed in craniate ectodermal

thickenings known as placodes (below), but no true placodes

have been observed in lancelets or tunicates (Shimeld and

Holland 2000). The olfactory organ is highly developed in

nearly all other euchordates.

Segmentation. Segmentation arises when mesoderm along

either side of the notochord subdivides to form somites. These

are hollow spheres of mesoderm that mature into muscle

blocks known as myomeres, which are separated by sheets

of connective tissue (myocomata). Only the mesoderm lying

close to the notochord becomes segmented, whereas more

laterally the mesoderm produces a sheet of muscle that surrounds

the coelomic cavity. The segmented muscles enable

powerful locomotion, producing waves of contraction that

pass backward and propel the body ahead. Segmentation is

accompanied by Fringe (or its homologue) expression and

signaling by the Notch protein, features shared with other

segmented bilaterians. These regulate the timing and synchronization

of cell-to-cell communication required of segmental

patterning and the formation of tissue boundaries

(Evrard et al. 1998, Jiang et al. 2000).

Other features. Also arising from mesoderm is a blood

circulatory system of stereotyped arterial design, with a dorsal

and ventral aorta linked by branchial vessels, and a complementary

venous system (Maisey 1986). Other transformations

traceable to the ancestral euchordate yielded a larva that

is essentially a miniature, bilateral adult. As adults, a median

fin ridge increases thrust area while helping to stabilize movement

through the water (Schaeffer 1987).

Pan-Euchordata

The oldest stem euchordate fossil may be the Early Cambrian

Yunannozoon from the Chengjiang lagerstдtte of southern

China (Chen et al. 1995, Shu et al. 2001b, Holland and Chen

2001). It is known from a single specimen that shows evidence

of segmental muscle blocks, an endostyle, a notochord,

and a nonmineralized pharyngeal skeleton. Little more than

a flattened smear, the chordate affinities of this problematic

fossil are debatable.

Node 4. The Lancelets (Cephalochordata)

The lancelets, sometimes known as amphioxus, form an ancient

lineage that today consists of only 30 species (Gans and

Bell 2001). Branchiostoma consists of 23 species and Epigonichthyes

includes seven (Poss and Boschung 1996, Gans et al.

1996). Lancelets are suspension feeders distributed widely in

tropical and warm-temperate seas. The larvae are pelagic, and

one possibly pedomorphic species remains pelagic as an adult.

Adults of the other species burrow into sandy substrate, protruding

their heads into the water column to feed.

Adult lancelets lack an enlarged head. They are unique in

the extent of both the notochord and cranial somites, which

extend to the very front of the body. A single median eye also

distinguishes them, which, based on AmphiOtx expression,

may be homologous to the paired eyes of other chordates and

bilaterians (Lacalli 1996a, Butler 2000). Their feeding apparatus

involves a unique ciliated wheel organ surrounding the

mouth, and a membranous antrum that surrounds the pharynx

(Maisey 1986, Holland and Chen 2001).

Pan-Cephalochordata

A single fossil from the Early Cambrian of China, known as

Cathaymyrus (Shu et al. 1996), may be a stem cephalochordate

and the oldest representative of the clade. Pikaia gracilens

390 The Relationships of Animals: Deuterostomes

from the Middle Cambrian Burgess Shale is known from

numerous specimens and is popularly embraced as a cephalochordate

(Gould 1989), but this is now questionable (Holland

and Chen 2001). A mitrate known as Lagynocystis

pyramidalis, from the lower Ordovician of Bohemia, may

also be a stem cephalochordate (Jeffries 1986). In all cases,

more specimens and more detailed anatomical preservation

are needed to have any confidence in these assignments.

Node 5. Chordates with a Head (Craniata)

Craniata contain the last common ancestor that humans share

with hagfish, and all its descendants (fig. 23.1). Even contemporary

literature often confuses this clade name with the

designation Vertebrata. However, Vertebrata are properly

regarded as a clade lying within Craniata (Janvier 1996).

Compared with their euchordate ancestors, craniates have

increased genetic complexity, a larger brain, and more elaborate

paired sense organs. Larvae probably persisted as suspension

feeders (Mallatt 1985), but adults shifted to active

predation with higher metabolic levels, more powerful locomotion,

and a sensory system perceptive to multiple modes

of environmental signal (Jollie 1982, Northcutt and Gans

1983;, but see Mallatt 1984, 1985). A rigid skull protects and

supports the brain, special sense organs, and feeding apparatus.

Most important, the neural crest blooms in early development

as a unique population of motile cells that induce

new structures and assist the many parts of the increasingly

complex head and pharynx to integrate as a functional whole.

Craniate Characters

Increased genetic complexity II. Craniates have at least two

Hox gene clusters, and perhaps three or four clusters were

present ancestrally (Holland and Garcia-Fernаndez 1996).

This increase in number is correlated with further elaboration

of the neurosensory system over that of lancelets and

tunicates. Several additional gene families increased in number,

including those encoding transcription factors (ParaHox,

En, Otx, Msx, Pax, Dlx, HNF3, bHLH), signaling molecules

(hh, IGF, BMP), and others (Shimeld and Holland 2000). The

mechanism of duplication is uncertain.

Elaborated brain and sensory organs II. The craniate brain

includes new cell types and neuronal groups. It now integrates

input from elaborated special sensory organs that develop from

paired ectodermal thickenings known as placodes, with the

assistance of cells of the neural crest (Northcutt and Gans 1983,

Webb and Noden 1993, Butler 2000, Shimeld and Holland

2000). Placodes are typically induced by the underlying

mesoderm, and they develop into organs and structures that

contribute sensory input to the brain. Although there is evidence

for olfactory, optic, and otic organs earlier in chordate

history, the integration of placodes with neural crest cells

marks a first blossoming of acute, highly complex special

sense organs. At least two placode types can now be distinguished.

Sensory placodes are involved in the olfactory sacs,

lens, ear vesicles, and lateral line system, whereas neurogenic

placodes contribute sensory neurons to cranial ganglia. Both

categories include some rather different structures, and the

different placodes probably had separate histories (Northcutt

1992, Webb and Noden 1993).

The craniate brain is also fully segmented in early ontogeny

and differentiates into discrete adult regions associated

with special cranial nerves that have specific sensory functions,

motor components, or both. Up to 22 cranial nerves

are know in some craniates (Butler 2000). The fore- and

midbrain regions are expanded and compartmentalized to

degrees not seen in other chordates. The forebrain differentiates

from segmented prosomeres into an anterior telencephalon

that receives input from highly developed olfactory

nerves, and the diencephalon to which project the paired eyes

(Butler and Hodos 1996). The pineal eye was probably also

a part of this system ancestrally. Adult hagfish lack a pineal

eye, evidently an ontogenetic loss as the entire visual system

degenerates (Hardisty 1979, Forey 1984b). The midbrain

arises from segmental mesomeres (Butler and Hodos 1996).

The hindbrain develops from segmental rhombomeres controlled

by Hox genes via Krox-20 and Kreisler expression

(Shimeld and Holland 2000). Also elaborated is the otic system,

which functions in both vestibular and acoustic reception.

Two semicircular canals were present ancestrally (Maisey

2001, Mazan et al. 2000). A lateral line system also arises from

head and body placodes (Northcutt 1992). Its functions in

electroreception (Bodsnick and Northcutt 1981), and also in

mechanoreception by sensing water currents and turbulence,

aiding locomotion and hunting (Pohlmann et al. 2001). Also,

an autonomic nervous system helps control the endocrine

system and other internal functions, and the spinal cord is

equipped with dorsal root ganglia.

The internal skeleton. The cartilaginous precursor of an

internal skeleton was present in the head, and along the

notochord as paired neural and hemal arches. These elements

develop via induction between the mesodermal sclerotome

and the adjacent notochord and/or spinal chord (Maisey

1986, 1988), but only later in chordate history do they become

mineralized or ossified (below). Although lacking jaws

and teeth, the ancestral craniate probably had specialized

hard mouthparts built from noncollagenous enamel proteins

that formed mineralized denticles along the pharyngeal

arches at the borders of the gill clefts. These are sites where

endoderm and ectoderm interact, and neural crest may also

contribute to their mineralization (Smith and Hall 1990).

Even in hagfish, high molecular weight amelogens are associated

with pharyngeal tissues (Slavkin et al. 1983, Delgado

et al. 2001) and the calcium regulatory hormone calcitonin

is present (Schaeffer 1987, Maisey 1988).

The neural crest. Origin of the neural crest was perhaps

the most remarkable morphogenic event in deuterostome

history, owing to the diverse structures that these cells induce

or contribute to directly, and help to integrate (Northcutt

and Gans 1983, Schaeffer 1987). Neural crest cells are

Chordate Phylogeny and Development 391

themselves induced by mesoderm along the edges of the

overlying neural plate. They migrate to new locations

throughout the head, where they produce the cartilaginous

neurocranium, a unique structure housing the expanded

brain and providing a rigid armature that suspends the special

sense organs. Neural crest cells also form a cartilaginous

branchial arch system. Neural crest cells also arise from the

developing spinal cord to form spinal ganglia, the sympathetic

nervous system, pigment cells, and adrenalin glands.

Neural crest cells do not differentiate nor are the structures

that they build present in tunicates or lancelets. However,

several neural crest cell–inducing genes occur in

lancelets. These include the Msx, Slug/Snail, and Distalless

gene families, which are expressed in lateral neural plate, and

Pax-3/7, which is expressed in immediately adjacent ectoderm

(Butler 2000, Shimeld and Holland 2000). Hox regulatory

elements have also been identified in lancelets that in

craniates drive spatially localized expression of neural crest

cells in the derivatives of placodes and the branchial arches

(Manzanares et al. 2000). Thus, well before the emergence

of the ancestral craniate, the relative spatial expression patterns

of several genes involved in neural crest induction were

present.

Pharyngeal arch elaboration. In lancelets, there is a more

or less stiff framework of several pairs of collagenous arches.

Between adjacent arches are branchial clefts that function

primarily in suspension feeding (Mallatt 1984, 1985). In

contrast, craniate pharyngeal arches are major structural elements,

composed of segmented cartilage or bone that suspend

heavily vascularized gills within the clefts. The arches

are muscular, and under CNS control they power a pump

involved in both respiration and feeding. In craniates, for the

first time, the pharyngeal clefts may properly be called gill

slits (Schaeffer 1987, Maisey 1988). Each arch develops from

an outer covering of ectoderm, an inner covering of endoderm,

and a mesenchymal core derived from neural crest and

mesoderm (Graham and Smith 2001). The majority of the

neural crest cells forming the arches arise adjacent to the

hindbrain rhombomeres, each arch with a neural crest population

tied to a specific group of rhombomeres. This ensures

the faithful transfer of segmental patterning information from

the CNS to the arches, establishing a correspondence between

innervations and effector muscles. The neural crest

segregates into discrete arch populations partly through

apoptosis, or preprogrammed cell death, in a process similar

to that which sculpts the discrete digits in the tetrapod

hands and feet (below). In both instances, key components in

the cell death program are the genes encoding Msx2 and BMP4

(Graham and Smith 2001, Zhou and Niswander 1996).

Elaborated muscular system. Muscle ontogeny follows a

unique pathway in craniates. First, mesodermal somitomeres

appear in strict rostral to caudal order during gastrulation,

as segmental arrays of paraxial mesenchymal cells condense

along the length of the embryo (Jacobson 1988, 2001). Cranial

somitomeres then disperse to form the striated muscles

of the head, including extrinsic muscles of the eye (except

in hagfish, which may have lost them secondarily), and branchial

musculature. In the trunk, the somitomeres gradually

condense to form somites. Lateral to the developing somites

the mesoderm differentiates into three separate populations

of cells. These are the sclerotome, which later forms part of

the cranium and much of the vertebral column, the dermatome,

which forms the connective tissues of the dorsal

trunk, and the myotome, which forms the striated muscles

of the trunk. The adult trunk musculature consists of sequential

chevron-shaped myomeres. Finally, the unsegmented

lateral plate splits and the coelomic cavity forms between its

two layers. The gut, which is no longer ciliated internally,

becomes invested by a layer of smooth muscle that provides

peristaltic contractions for the movement of ingested food

(Schaeffer 1987, Maisey 1986).

Powerful heart and circulatory system. A powerful twochambered

heart is present in craniates along with red blood

cells, hemoglobin, and vasoreceptors that monitor pressure

and gas levels of the blood passing through the heart. Associated

with the elaborated circulatory system is a highly innervated

kidney (Schaeffer 1987, Maisey 1988).

Additional endodermal derivatives. The liver and pancreas

arise from endoderm through new inductive signals from

mesoderm. Also deriving form this source are elaborate endocrine

glands including the parathyroids, which control

calcium and phosphate metabolism with the plasma calciumregulatory

hormone (calcitonin), and the adrenal glands, all

of which are controlled to varying degrees by the autonomic

nervous system. The larval endostyle metamorphoses into the

adult thyroid gland, becoming a true endocrine gland, directing

its secretions into the circulatory rather than digestive

system (Schaeffer 1987).

Paired and median fin folds. Primordia of the paired lateral

and median appendages arise in craniates via mesodermal-

epithelial induction, whereas the dorsal fin arises via

interaction between the epidermis and trunk neural crest. A

median fin fold is present in lancelets, but it develops without

the neural crest interaction.

High metabolic capacity. Craniates possess a well-developed

capacity for anaerobic metabolism, resulting in the formation

of lactic acid. This probably evolved in association

with burst activity that is unobtainable by relying solely on

aerobic metabolism (Ruben and Bennett 1980).

Pan-Craniata

The oldest putative pancraniate is Haikouella lanceolata,

known by more than 300 specimens from the Chengjiang

lagerstдtte of southern China (Chen et al. 1995, 1999, Shu

et al. 2001a, 2001b, Holland and Chen 2001). It has a threepart

brain and paired eyes. Its mouth has 12 oral tentacles,

and the pharynx has six nonmineralized pharyngeal arches

bearing gill filaments that lie in separate visceral clefts. A pair

of grooves in its floor suggests an endostyle. There may be

several mineralized denticles on the third arch, but preser392

The Relationships of Animals: Deuterostomes

vation leaves this uncertain. About two dozen paired straight

myomeres are separated by myosepta behind the 5th visceral

arch. Stains are preserved that may represent a heart with

ventral and dorsal aorta, and anterior branchial artery. The

notochord extends about 85% the length of the body, stopping

short of the rostrum, and slight banding can be seen

resembling the immature vertebral elements of lampreys

(Holland and Chen 2001). It also has dorsal, caudal, and

ventral midline fins. Haikouella has also been hypothesized

to lie on the lamprey stem (Chen et al. 1999), but support is

weak (Janvier 1999). From the same deposits, possibly lying

on the craniate stem, are Haikouichthyes and Myllokunmingia,

each known from a single fusiform fossil (Shu et al. 1999a,

1999b). The rostral two-thirds of their bodies comprises the

pharyngeal region, with Z-shaped myomeres making up the

rest. A median dorsal fin shows faint striations that may be

fin rays. There are also paired lateral structures, but it is

doubtful whether they are homologous with the fins of

gnathostomes (below). In Haikouichthyes are nine pharyngeal

arches and a complex skull, probably built of cartilage, suggesting

the presence of neural crest cells. Neither specimen

shows evidence of mineralization (Shimeld and Holland

2000, Holland and Chen 2001).

Node 6. The Hagfish (Myxini)

Hagfish comprise a poorly known chordate lineage that includes

58 living species (Froese and Pauly 2001). Throughout

their life cycles, hagfish generally occupy deep marine

habitats in temperate seas, ranging from 25 to 5000 m in

depth (Moyle and Cech 2000). They scavenge large carcasses,

burrow into soft substrate for invertebrates, and pursue small

prey through the water column. But they are difficult to

observe and little is known of their development.

The monophyly of Myxini is well supported. They have

three pairs of unique tactile barbels around the nostril and

mouth, and a single median nostril of distinct structure. Many

other features distinguish them from other craniates, but

some may reflect secondary loss, including absence of the

epiphysis and pineal organ, reduction of the eyes, presence

of only a single adult semicircular canal, and a vestigial lateral

line system confined to the head (Hardisty 1979, Maisey

1986, 2001).

Pan-Myxini

Only three fossil species have been allied to the hagfish. The

least equivocal is Myxinikela siroka, from the Carboniferous

Mazon Creek deposits of Illinois (Bardack 1991). A second

specimen from these same beds, Pipiscus zangerli (Bardack

and Richardson 1977), is more problematically a hagfish and

has also been allied to lampreys (below). Xidazoon stephanus,

known by three specimens from the Lower Cambrian of

China, has been compared with Pipiscius (Shu et al. 1999a,

1999b). Its mouth is defined by a circlet of about 25 plates,

and it may have a dilated pharynx and segmented tail. But

other assignments are equally warranted by the vague

anatomy it preserves, and whether it is even a chordate remains

questionable.

Node 7. Chordates with a Backbone (Vertebrata)

Vertebrata comprise the last common ancestor that humans

share with lampreys, and all its descendants. The relationship

of hagfish and lampreys to other craniates is long debated.

Hagfish and lampreys were once united either as

Cyclostomata or Agnatha, jawless fishes grouped by what its

members lacked instead of by shared unique similarities, and

they were considered ancestral to gnathostomes (e.g., Romer

1966, Carroll 1988). This grouping was largely abandoned

as diverse anatomical data showed lampreys to share more

unique resemblances with gnathostomes than with hagfish

(Stensiц 1968, Lшvtrup 1977, Hardisty 1979, 1982, Forey

1984b, Janvier 1996). But controversy persists, and recent

studies of the feeding apparatus have resurrected a monophyletic

Cyclostomata (Yalden 1985, Mallatt 1997a, 1997b).

Cyclostome monophyly is also supported by ribosomal DNA

(rDNA; Turbeville et al. 1994, Lipscomb et al. 1998, Mallatt

and Sullivan 1998, Mallatt et al. 2001), vasotocin complementary

DNA (cDNA; Suzuki et al. 1995), and globin cDNA

(Lanfranchi et al. 1994). However, the results from small subunits

of rDNA were overturned when larger ribosomal sequences

were used, and morphological analyses that sample

many different systems also refute cyclostome monophyly

(Philippe et al. 1994, Donoghue et al. 2000). The question may

not be settled, but I follow current convention and treat lampreys

and hagfish as successive sister taxa to gnathostomes.

Vertebrate Characters

Increased genetic complexity III. A tandem duplication of

Hox-linked Dlx genes occurred in vertebrates ancestrally,

encoding transcription factors expressed in several developing

tissues and structures. They are expressed in an expanded

forebrain, cranial neural crest cells, placodes, pharyngeal

arches, and the dorsal fin fold. An additional duplication

evidently occurred independently in lampreys and gnathostomes

(Amores et al. 1998, Niedert et al. 2001, Holland and

Garcia-Fernаndez 1996).

Elaboration of the brain and special senses III. In vertebrates,

exchange of products between blood and cerebrospinal

fluid occurs via the choroid plexus, a highly vascularized

tissue developing in the two thinnest parts of the ventricular

roof of the brain. Vertebrate eyes are also enhanced by a

retinal macula, a small spot of most acute vision at the center

of the optic axis of the eye, and by synaptic ribbons that

improve retinal signal processing. Extrinsic musculature

originating from the rigid orbital wall provides mobility to

the bilateral eyeballs. The pineal body is also photosensory,

and in some vertebrates differentiates into a well-developed

pineal eye with retina and lens. In addition, the lateral line

system extends along the sides of the trunk (Maisey 1986).

Chordate Phylogeny and Development 393

Correspondingly, an extensive cartilaginous braincase that

includes embryonic trabecular cartilages arises beneath the

forebrain, and an elaborate semirigid armature supports the

brain and its special sensory organs.

Locomotor and circulatory systems. Vertebrates have dorsal,

anal, and caudal fins that are stiffened by fin rays, increasing

thrust and steering ability. The circulatory and muscular

systems were also bolstered. The heart comes under nervous

regulation and a stereotyped vascular architecture carries

blood to and from the gills. Myoglobin stores oxygen in the

muscles, augmenting scope and magnitude in bursts of activity.

The kidney is also elaborated for more sensitive osmoregulation

and more rapid and thorough filtration of the

blood (Maisey 1986).

Pan-Vertebrata

The oldest putative stem vertebrates are the heterostracans,

an extinct lineage extending from Late Cambrian (Anatolepis)

to the Late Devonian (Maisey 1986, 1988, Gagnier 1989,

Janvier 1996). Their skeleton consists of plates of acellular

membranous bone. Precise relationships of this clade are

controversial, but if correct the position of heterostracans as

the sister taxon to Vertebrata may suggest that lampreys may

have secondarily lost a bony external skeleton. However, in

the absence of direct evidence that lampreys ever possessed

bone, heterostracan fossils and the characteristics of bone are

treated below (see Pan-Gnathostomata, below).

Node 8. The Lampreys (Petromyzontida)

There are approximately 35 living lamprey species, all but

three of which inhabit the northern hemisphere (Froese and

Pauly 2001). In most, larvae hatch and live as suspension

feeders in freshwaters for several years, then migrate to the

oceans as metamorphosed adults, where they become predatory

and parasitic. Nonparasitic freshwater species are known

(Beamish 1985) and in some cases the metamorphosed adults

are nonpredatory and do not feed during their short adult

lives (Moyle and Cech 2000).

Lamprey monophyly is diagnosed by a unique feeding

apparatus. It consists of an annular cartilage that supports

a circular, suction-cup mouth lined with toothlike keratinized

denticles. A mobile, rasping tongue is supported by

a unique piston cartilage and covered by denticles whose

precise pattern diagnoses many of the different species.

Lampreys attach to a host, rasp a hole in its skin, and feed

on its body fluids. Lampreys also eat small invertebrates.

The structure of the branchial skeleton (Mallatt 1984,

Maisey 1986) and the single median nasohypophysial opening

(Janvier 1997) are unique. Lampreys have a distinctive

suite of olfactory receptor genes that serves in the detection

of odorants such as bile acids (Dryer 2000). There is

also evidence that lampreys are apomorphic in having undergone

duplication of a tandem pair of Dlx genes, followed

by loss of several genes, independent of a comparable duplication

and subsequent loss that occurred in gnathostomes

(Niedert et al. 2001)

Pan-Petromyzontida

Haikouichthyes ercaicunensis (Shu et al. 1999b) from the Early

Cambrian of China is the oldest fossil lamprey reported, but

the data for its placement are tenuous (Janvier 1999).

Mayomyzon pieckoensis, known by several specimens from the

Late Carboniferous Mazon Creek beds of Illinois (Bardack

and Zangerl 1968), is the oldest unequivocal lamprey, preserving

unique lamprey feeding structures, including the

annular and piston cartilages. Hardistiella montanensis (Janvier

and Lund 1983) from the Lower Carboniferous of Montana

preserves less detail, and it is not clear whether either lies

within or outside of (crown) Petromyzontida. Pipiscus zangerli

(Bardack and Richardson 1977) from the same Mazon Creek

beds as Mayomyzon is sometimes also tied to lampreys, as well

as hagfish, but it preserves little relevant evidence.

Node 9. Chordates with Jaws (Gnathostomata)

Gnathostomata comprise the last common ancestor that

humans share with Chondrichthyes, and all of its descendants

(fig. 23.1). Its origin was marked by additional increases

in complexity of the genome, which mediated several landmark

innovations, including jaws, paired appendages, several

types of bone, and the adaptive immune system.

Although the positions of certain basal fossils are debated,

there is little doubt regarding gnathostome monophyly.

Gnathostome Characters

Increased genetic complexity IV. Gnathostomes have at

least four Hox gene clusters, and some have as many as seven.

In addition to specifying the fate of cell lineages along the

anteroposterior axis, these gene clusters mediate limb development

and other outgrowths from the body wall. It is questionable

whether as many as four Hox clusters arose earlier,

either in vertebrates or craniates ancestrally (Holland and

Garcia-Fernаndez 1996), but in gnathostomes their expression

nevertheless manifests more complex morphology.

There was also duplication of Hox-linked Dlx genes and several

enhancer elements, leading to elaboration of cranial neural

crest in the pharyngeal arches, placodes, and the dorsal fin

fold (Niedert et al. 2001). Immunoglobin and recombinase

activating genes also arose in gnathostomes, marking the

origin of the adaptive immune system.

Brain and sensory receptor enhancement IV. The gnathostome

forebrain is enlarged, primarily reflecting enhancement

of the olfactory and optic systems. The extrinsic muscles of

the eyeball are rearranged and an additional muscle (the

obliquus inferior) is added to the suite present in vertebrates

ancestrally (Edgeworth 1935). In the ear, a third (horizontal)

semicircular canal arises, lying in nearly the same plane

as the synaptic ribbons of the eye, and correlates with Otx1

expression (Maisey 2001, Mazan et al. 2000). In addition,

394 The Relationships of Animals: Deuterostomes

the lateral line system is elaborated over much of the head

and trunk. On the trunk, it is developmentally linked to

the horizontal septum and becomes enclosed by mineralized

tissues that insulate and tune directional electroreception

by the lateral line system (Northcutt and Gans

1983). The gnathostome lateral line system derives from

neural crest and lateral plate mesoderm induction, heralding

a new stage in developmental complexity. Myelination

of many nerve fibers improves impulse transmission through

much of the body (Maisey 1986, 1988).

Mineralized, bony skeleton. Many bilaterians produce

mineralized tissues, and both echinoderms and tunicates generate

amorphous and crystalline calcium carbonate spicules

(Aizenberg et al. 2002). Biomineralization is thus an ancient

property, although its erratic expression outside of Craniata

affords only equivocal interpretations of its history in this part

of the tree. Certain other components required for bone

mineralization, such as calcitonin, were already present but

did not lead to bone production. However, in gnathostomes,

different types of bone form in the head and body (Maisey

1988). Bone development requires the differentiation of

specialized cell types, including fibroblasts, ameloblasts,

odontoblasts, and osteoblasts, which are derived from the

ectoderm and cephalic neural crest. In the formation of

membranous bone, fibroblasts first lay down a fibrous collagen

framework around which the other cells deposit calcium

phosphate as crystalline hydroxyapatite. Another type

of bone development typically involves preformation by cartilage,

followed by deposition of hydroxyapatite crystals

around the cartilage (perichondral ossification), or within and

completely replacing it (endochondral ossification). Chondral

ossification occurred first in the head in the oldest extinct

gnathostomes (see Pan-Ganthostomata, below), and it

later spread to the axial skeleton and shoulder girdle. Ossification

in the shoulder girdle is of interest because it is the

first such transformation of the embryonic lateral plate mesoderm

and because it signals the initiation of neural crest

activity in the trunk (Maisey 1988). In the shark lineage, the

internal skeleton consists of cartilage that is sheathed in a

layer of crystalline apatite, but fossil evidence suggests that

this is a derived condition (below).

Elaborated skull. Cartilage and/or chondral bone surround

the brain and cranial nerves, providing a semirigid

armature for the special sensory organs. At the back of the

head, the cephalicmost vertebral segment is “captured” during

ontogeny by the skull to form a back wall of the braincase.

Thereby, it confines several cranial nerves and vessels

to a new passage through the base of the skull, known in

embryos as the metotic fissure. Cellular membranous bone

was also present, covering the top and contributing to other

parts of the skull (Maisey 1986, 1988).

Jaws. The namesake characteristic of gnathostomes arises

in ontogeny from the first pharyngeal arch, known now as

the mandibular arch. Its upper half is the palatoquadrate

cartilage, which is attached to the braincase primitively by

ligaments, whereas the lower half of the arch, Meckel’s cartilage,

forms the lower jaw and hinges to the palatoquadrate

at the back of the head. Teeth and denticles develop on inner

surfaces of these cartilages through an induction of ectoderm

and endoderm. Neural crest cells populating the

mandibular arch derive from the mesomeres and from hindbrain

rhombomeres 1 and 2, whereas the second pharyngeal

arch, the hyoid arch, derives its neural crest from rhombomere

4 (Graham and Smith 2001).

Paired appendages. Other bilaterians have multiple sets

of paired appendages that serve a broad spectrum of functions.

It was long believed that their evolution was entirely

independent of the paired appendages in gnathostomes, but

this appears only partly true today. Common Hox patterning

genes were likely present in the last common ancestor of

chordates and arthropods, if not a more inclusive group. The

SonicHedgehog gene specifies patterning along anteroposterior,

dorsoventral, and proximodistal axes of the developing

limb, via BMP2 signaling proteins (Shubin et al. 1997). In

gnathostomes, independent expression of orthologous genes

occurs in the elaboration of fins, feet, hands, and wings. As

expressed in gnathostomes, the distal limb elements are the

most variable elements. In basal gnathostomes they comprise

different kinds of stiffening rays, whereas in tetrapods they

are expressed as fingers and toes (Shubin et al. 1997). Moreover,

somite development transformed to provide for muscularization

of the limbs, as certain somite cells became motile

and moved into the growing limb buds (Galis 2001). Thus,

although the Hox genes have a more ancient history of expression,

in gnathostomes they are expressed across a unique

developmental cascade.

The adaptive immune system. One of the most remarkable

gnathostome innovations is the adaptive immune system

(Litman et al. 1999, Laird et al. 2000). It responds adaptively

to foreign invaders or antigens such as microbes, parasites,

and genetically altered cells. Other animals have immune

mechanisms, but unique to gnathostomes is a system that is

specific, selective, remembered, and regulated. Its fundamental

mediators are immunoglobin and recombinase activation

genes, which are present throughout gnathostomes but absent

in lampreys and hagfish. The immune system is expressed

in a diverse assemblage of immunoreceptor-bearing

lymphocytes that circulate throughout the body in search of

antigens. Gnathostome lymphocytes present an estimated

1016 different antigen receptors, which arose seemingly instantaneously

as an “immunological big bang” (Schluter et al.

1999) in gnathostomes ancestrally.

New endodermal derivatives. In gnathostomes, the endoderm

elaborates to form the pancreas, spleen, stomach, and

a spiral intestine (Maisey 1986).

Pan-Gnathostomata

Several extinct lineages lie along the gnathostome stem. Their

relationships remain problematic, and most have been allied

with virtually every living chordate branch (Forey 1984a,

Chordate Phylogeny and Development 395

Maisey 1986, 1988, Donoghue et al. 2000). All preserve

mineralized and bony tissues of some kind, and the phylogenetic

debate revolves in large degree around interpreting

the history of tissue diversification. The most ancient, if

problematic extinct pangnathostome lineage is Conodonta.

Known to paleontologists for decades only from isolated,

enigmatic mineralized structures, conodonts range in the

fossil record from Late Cambrian to Late Triassic. The recent

discovery of several complete body-fossils demonstrated that

these objects are toothlike structures aligned along the pharyngeal

arches and bordering the gill clefts. They are built of

dentine, calcified cartilage, and possibly more than one form

of hypermineralized enamel (Sansom et al. 1992). Microwear

features indicate that they performed as teeth, occluding

directly with no intervening soft tissues. They formed along

the same zones of endoderm-ectoderm induction as the

pharyngeal teeth in more derived vertebrates. The mineralized

oropharyngeal skeleton and dentition arose at the base

of the gnathostome stem, Cambrian conodont fossils providing

its oldest known expression (Donoghue et al. 2000).

Branching from or possibly below the gnathostome stem

are the heterostracans (see Pan-Vertebrata, above), whose

skeleton consists of external plates of acellular membranous

bone. In heterostracans, bones formed around the head, and

the cranial elements seemingly grew continually throughout

life. Their bone is formed of a basal lamina, a middle layer

of spongy arrays of enameloid, and an outer covering of

enameloid and dentine. Heterostracan fossils suggest that

bone was acellular at first.

The next most problematic taxon is Anaspida, which range

from Middle Silurian to Late Devonian (Forey 1984a, Maisey

1986, 1988, Donoghue et al. 2000). Anaspids are diagnosed

by the presence of branchial and postbranchial scales, pectoral

plates, and continuous bilateral fin folds. Perichondral

ossification occurred in neural and hemal arches, and the appendicular

skeleton, whereas endochondral ossification occurred

in fin radials and dermal fin rays in the tail. The anaspid

trunk squamation pattern suggests the presence of the horizontal

septum, a critical feature in the trunk-powered locomotion

that is also tied developmentally to the lateral line

system. Anaspid lateral fin folds may prove to be precursors

of the paired appendages of crown gnathostomes.

Lying closer to the gnathostome crown clade is Galeaspida,

which range through the Silurian and Devonian. Its members

are distinguished by a large median dorsal opening that communicates

with the oral cavity and pharyngeal chamber.

Galeaspids also have 15 or more pharyngeal pouches. Their

chondral skeleton appears mineralized around the brain and

cranial nerves, however the bone is primitive in being acellular

(Maisey 1988). Lying closer to the gnathostome crown

is Osteostraci, a lineage with a similar character and temporal

range as galeaspids. Osteostracans have a dorsal head

shield with large dorsal and lateral sensory fields. They share

with crown gnathostomes cellular calcified tissues and

perichondral ossification of the headshield, which encloses the

brain and cranial nerve roots. Ossification surrounds the orbital

wall, otic capsules, and calcified parachordal cartilages,

structures developing in extant gnathostomes via inductions

between the CNS, notochord, and the ectomesenchyme. Perichondral

mineralization of the otic capsule implies interaction

between mesenchyme and the otic placode (Maisey 1988).

Also present are lobed, paired pectoral fins that are widely

viewed as homologous to the pectoral appendages in crown

Gnathostomata (Forey 1984a, Maisey 1986, 1988, Shubin

et al. 1997, Donoghue et al. 2000). Supportive of this view is

the ontogenetic sequence in most extant gnathostomes, in

which pectoral appendages arise before pelvic.

Node 10. Sharks and Rays (Chondrichthyes)

Chondrichthyes includes sharks, skates, rays, and chimaeras

(fig. 23.1). The chimaeras (Holocephali) include roughly

30 living species, and there are about 820 living species of

skates and rays (Batoidea) plus sharks (Moyle and Cech 2000).

Morphology suggests that the species commonly known as

sharks do not by themselves constitute a monophyletic lineage,

and that some are more closely related to the batoids

than to other “sharks” (Maisey 1986).

Earlier authors argued that these different groups evolved

independently from more primitive chordates, and that

Chondrichthyes was a grade that also included several cartilaginous

actinopterygians (below). Cartilage is an embryonic

tissue in all craniates, and it persists throughout life in sharks

and rays (and a few other chordates), but the perception that

“cartilaginous fishes” are primitive is mistaken. In its more

restricted reference to sharks, rays, and chimaeras, the name

Chondrichthyes designates a monophyletic lineage. Histological

examination reveals bone at the bases of the teeth,

dermal denticles, and some fin spines. This suggests that

this restricted distribution of bone is a derived condition

in chondrichthyans (Maisey 1984, 1986, 1988).

Other apomorphic characters include the presence of

micromeric prismatically calcified tissue in dermal elements and

surrounding the cartilaginous endoskeleton. Chondrichthyans

also possess a specialized labial cartilage adjacent to the mandibles,

the males possess pelvic claspers, and the gill structure

is unique. The denticles (scales) possess distinctive neck

canals (but these may not be unique to chondrichthyans),

and the teeth have specialized nutrient foramina in their bases

with a unique replacement pattern in which replacing teeth

attach to the inner surface of the jaws as dental arcades

(Maisey 1984, 1986). Fin structure also presents a number

of unique modifications (Maisey 1986). Relationships among

chondrichthyans have received a great deal of attention

(Compagno 1977, Schaeffer and Williams 1977, Maisey 1984,

1986, Shirai 1996, de Carvalho 1996).

Pan-Chondrichthyes

The extinct relatives of chondrichthyans have a long, rich

fossil record. The oldest putative fossils are scales with neck

396 The Relationships of Animals: Deuterostomes

canals from the Late Ordovician Harding Sandstone of Colorado

(Sansom et al. 1996). Although present in extant sharks

and chimeroids, most well-known Paleozoic sharks lack them.

From the Silurian onward, chondrichthyan teeth are abundantly

preserved, although in most cases their identification

rests on solely phenetic grounds, and they provide little useful

information on higher level phylogeny. The oldest anatomically

complete fossils are the Late Devonian Symmoriidae and

Cladoselache, which are known from numerous skeletons that

in some case preserve body outlines and other evidence of soft

tissues. Both are stem chondrichthyans.

Node 11. Chordates with Lungs (Osteichthyes)

Osteichthyes (fig. 23.1) comprise the lineage stemming from

the last common ancestor that humans share with actinopterygians.

The name means “bony fishes” and was coined

in pre-Darwinian times in exclusive reference to the fishlike

members of this clade. In the phylogenetic system (de

Queiroz and Gauthier 1992), the name now refers to all

members of the clade, roughly half of which are the chordate

species adapted to life on land.

Osteichthyan Characters

An extensive composite bony skeleton. All conclusions about

skeletal evolution at this node are weak, because chondrichthyans

lack an ossified internal bony skeleton that can be

compared directly with that in osteichthyans. Nonetheless,

the fossil record offers assistance and suggests that a bony

skeleton likely arose in early pangnathostomes, and that it

was further elaborated in Osteichthyes. The membranous

skeleton of the head forms laminae that descend from the

braincase and offer attachment to muscles of the jaws and

pharyngeal skeleton. The jaws themselves are invested in a

layer of membranous bone, with teeth attached to their margins

(Rosen et al. 1981, Maisey 1986). Around the pharyngeal

chamber is an extensive series of dermal gular and

opercular bones, which improve pharyngeal function as a

suction chamber in both respiration and feeding. The pectoral

girdle became ossified, primitively more through perichondral

than endochondral processes. Lastly, in the fins are

stiffening rays known as lepidotrichia, which represent rows

of slender scales that replace the primitive covering of body

scales (Maisey 1986, 1988).

Lungs. Lungs develop as ventral outgrowths from the

rostral end of the gut tube and are often associated with skeletal

structures of mesenchymal origin. Over the course of

osteichthyan history, these diverticula become modified for

radically different functions that range from respiration, to

buoyancy regulation, to communication. In most terrestrial

members of the clade, lungs completely replace gills. They

are secondarily lost in some small living amphibian species,

where cutaneous respiration takes over. Lungs develop as

branching tubular networks constructed of sheetlike cellular

epithelia. There can be hundreds to millions of branches

in the network, yet they must also have a regular patterning

and structure to ensure proper function. A signaling pathway

mediated by fibroblast growth factor (FGF) occurs in

development of the branched lungs in the mouse, as well as

in the branched respiratory tracheae in the fruit fly, raising

the question of whether their common ancestor had a

branched respiratory structure. But because the tracheal system

lungs in insects are ectodermal and the osteichthyan lung

is endodermal, this seems unlikely. Moreover, FGF is implicated

in other branched structures and has probably been

co-opted throughout metazoan history to produce different

kinds of structures. The patterning mechanism is ancient, but

its expression in the osteichthyan lung is unique (Metzger

and Krasnow 1999).

Pan-Osteichthyes

Two problematic extinct lineages, Acanthodii and Placodermi,

arguably lie along the osteichthyan stem, but the evidence

is equivocal and a wide spectrum of other possibilities

have been proposed. Although some gnathostomes went on

to lose one or both sets of limbs, acanthodians are the only

clade to exceed the primitive number of two pairs. An anterior

spine stiffens each fin. Acanthodian fossils are known

from the Late Silurian to the Late Devonian. Placoderms

comprise a much more diverse clade whose fossil record

extends from Early Devonian to Early Carboniferous. Placoderms

are heavily armored, with a distinctive pattern of

membranous bones forming a head shield that hinges to a

membranous thoracic shield in a pair of ball-in-socket joints.

Acanthodians and placoderms share with Osteichthyes the

presence of the clavicle and interclavicles and other membranous

elements in the pectoral girdle. Placoderms lie closer

to Osteichthyes based on descending laminae of membranous

bone in the neurocranium, lepidotrichia in the fins, and

other features (Gardiner 1984). Difficulties in comparing

skeletal features in these fossils with chondrichthyans, which

largely lack a bony skeleton, complicate understanding the

relationships of these extinct lineages (Maisey 1986).

Node 12. The Ray-Finned Fishes (Actinopterygii)

The ray-finned fishes (fig. 23.1) include nearly 23,000 living

species and comprise nearly half of extant chordate diversity

(Lauder and Liem 1983). The most basal divergence

among extant actinopterygians is represented by the bichirs

and reedfish (Polypteriformes), which commonly (but not

unanimously) are regarded as sister taxon to all others. Next

most basal was the divergence between the sturgeons and

paddlefishes (Acipenseriformes), followed by gars (Ginglymodi)

and bowfins (Halecomorpha). Among these basal

clades alone are nearly 300 extinct genera named for fossils.

However, this part of the actinopterygian tree remains a frontier,

in large part because the fossil morphology is known only

superficially (Grande and Bemis 1996). The rest of extant

actinopterygian diversity resides among the teleosts (De Pinna

Chordate Phylogeny and Development 397

1996). Today actinopterygians occupy virtually every freshwater

and marine environment. Their economic importance

underlies the base of a huge global market, and actinopterygian

conservation increasingly is involved in conflicts with

development and use of the world’s water resources. One

member of this clade, the zebrafish, is growing in importance

for biomedicine as an important model organism. Actinopterygian

history and diversity are reviewed by Stiassny et al.

(Stiassny et al., ch. 24 in this vol.).

Pan-Actinopterygii

The fossil record of stem actinopterygians extends tentatively

into the Late Silurian (Long 1995, Arratia and Cloutier 1996).

The Late Silurian Andreolepis and Early Devonian Ligulalepis

are the oldest purported panactinopterygians fossils. They

are known only from scales, which overlap in a seemingly

distinctive tongue-in-groove arrangement often considered

diagnostic of actinopterygians. However, an ossified Early

Devonian braincase, possibly referable to Ligulalepis (Basden

et al. 2000) closely resembles the braincase in the extinct

Early Devonian shark Pucapampella (Maisey and Anderson

2001), although it is ossified. Hence, we may expect continued

reassessment of character distributions and view as tentative

the phylogenetic assignment of extinct taxa at this deep

part of the tree. By the Early Devonian, actinopterygian fossils

are found worldwide, but diversity is low. In the Middle

and Late Devonian, only 12 species and seven genera are

recognized. The best known is Cheirolepis, whose skeleton

is known in detail (Arratia and Cloutier 1996). From Middle

and Late Devonian rocks, abundant fossils of Mimia and

Moythomasia have been recovered, representing the oldest

members of crown clade Actinopterygii (Grande and Beamis

1996).

Node 13. Chordates with Lobe Fins (Sarcopterygii)

Sarcopterygians include the last common ancestor that humans

share with coelacanths, and lungfishes and all its descendants

(fig. 23.1). Just less than half of chordate diversity

lies within this clade (Cloutier and Ahlberg 1996). Its early

members were all aquatic, but from the Carboniferous onward

most sarcopterygians have been terrestrial (Gauthier

et al. 1989). Today only eight living species retain the ancestral

life style. Two are coelacanths and the other six are

lungfish, whereas the remainder of sarcopterygian diversity

resides among the tetrapods.

Sarcopterygian monophyly is strongly supported, but

relationships within are far from settled, especially when

fossils are concerned. Leaving fossils aside for the moment,

morphological, and molecular analyses continue to provide

conflicting results (Marshall and Schultze 1992, Schultze

1994, Meyer 1995, Zhu and Schultze 1997). Older studies

placed coelacanths outside of tetrapods + actinopterygians

(von Wahlert 1968, Wiley 1979), and even with chondrichthyans

(Lшvtrup 1977, Lagios 1979). Parvalbumin sequences

also support the placement of the Latimeria outside of

Osteichthyes (Goodwin et al. 1987). Morphology consistently

places Actinistia closer to tetrapods than to actinopterygians

(Romer 1966, Rosen et al. 1981, Maisey 1986, Nelson

1989, Chang 1991), a position also supported by 28S rDNA

(Hillis and Dixon 1989). But whether lungfish or coelacanths

are closer to tetrapods, or whether lungfish and coelacanths

together form a clade independent of tetrapods is still debated.

A larger 28S sequence (Zardoya and Meyer 1996)

found coelacanths and lungfishes to be the sister lineage to

tetrapods. A genomic DNA analysis (Venkatesh et al. 1999,

2001) and morphology (Rosen et al. 1981, Maisey 1986,

Cloutier and Ahlberg 1996) favor lungfishes and coelacanths

as successive outgroups to tetrapods, the position that is

followed here.

Sarcopterygian Characters

Lobe fins. The sarcopterygian pectoral and pelvic appendages

form muscular lobes that protrude from the lateral

body wall with a distinct skeletal architecture. In

gnathostomes ancestrally there were multiple basal elements

in each limb, but in sarcopterygians there is a single proximal

element, followed distally by a pair of radial cartilages. This

arrangement enables the insertion of muscles between the

radials, giving the fin flexibility along its axis (Clack 2000).

Fundamental similarities in branching occur within the embryonic

digital arch in lungfishes and tetrapods, producing the

familiar pattern of a single proximal element (humerus or femur),

followed by a pair of elements (radius/ulna or tibia/

fibula), followed by the more complex pattern of wrist and

ankle bones. This branching sequence is known as the metapterygial

axis, and it reflects further influence by SonicHedgehog

(via BMP2 signaling proteins), which specifies patterning along

anteroposterior, dorsoventral, and proximodistal axes of the

developing limb. Expressed from the beginnings of gnathostome

history in the development of fins, modified expression

of orthologous genes lead to the elaboration of lobefins, feet,

hands, and wings in sarcopterygians (Shubin and Alberch

1986, Shubin et al. 1997, Cloutier and Ahlberg 1996).

Enamel. A thin layer of enamel covers the teeth in sarcopterygians,

and at their bases the enamel is intricately

infolded into the dentine, in a pattern known as labyrinthodonty.

Infolded enamel enhances tooth strength as well as

the strength of attachment to the jaw (Long 1995).

Pan-Sarcopterygii

The acceptance by earlier researchers of paraphyletic groups

such as the crossopterygians (e.g., Romer 1966) and the

search for direct ancestors of tetrapods in these “amphibianlike

fishes” left controversial the relationship among the extinct

Paleozoic sarcopterygians (Rosen et al. 1981, Maisey

1986). However, most of these extinct taxa are now assignable

as stem lungfish (Pan-Dipnoi) or stem tetrapods (Pan-

Tetrapoda). However, two recent fossil discoveries lie on the

sarcopterygian stem and provide the oldest evidence of the

398 The Relationships of Animals: Deuterostomes

clade. These are Psarolepis romeri, from the Late Silurian and

Early Devonian of Asia (Ahlberg 1999, Zhu et al. 1999) and

Achoania jarvikii (Zhu et al. 2001) from the Early Devonian

of China. Lying at the base of either the sarcopterygian stem

(Long 1995, Clack 2000) or the choanate stem is Onychodontiformes,

a poorly known Devonian lineage whose

members reached 2 m in length and are characterized by

daggerlike tooth whorls. It is possible that Psarolepis lies

within this clade.

Node 14. The Coelacanths (Actinistia)

Coelacanth history is at least 400 million years (Myr) long

(Forey 1998), but only two species survive today. Latimeria

chalumnae inhabits coastal waters along southeastern Africa,

and a second population was recently discovered in the waters

off Sulawesi (Erdmann et al. 1998). Divergent DNA sequences

reportedly diagnose Latimeria menadoensis (Pouyaud

et al. 1999), but it shows little morphological distinction.

However, sequences from parts of two mitochondrial genes

also diagnose the Sulawesi species, and molecular clock estimates

suggest that it diverged from its common ancestor

with the African species 5.5 Mya (Holder et al. 1999). Monophyly

of the lineage has never been seriously questioned, and

it is diagnosed by such features as the absence of the maxilla,

absence of the surangular, absence of the branchiostegal

rays, presence of a rostral electric organ, presence of numerous

supraorbital bones, and a distinctive tassle on the tail.

Pan-Actinistia

The coelacanth fossil record ranges back to the Middle Devonian

but it ends in the Late Cretaceous, or more tenuously

the Paleocene (Cloutier and Ahlberg 1996). Approximately

125 extinct coelacanth species have been named (Cloutier

1991a, 1991b, Cloutier and Ahlberg 1996, Forey 1998).

Although often described as a living fossil (Forey 1984b), a

phylogenetic analysis of Latimeria chalumnae and its extinct

relatives showed that the living species differ by many dozens

of apomorphies from their Paleozoic relatives (Cloutier

1991a). Some of these characters represent losses of elements

in the cheek and opercular region, leading to suggestions that

coelacanth history was characterized by pedomorphosis

(Lund and Lund 1985, Forey 1984b). However, there are also

elaborations in complexity of skeletal elements, which indicate

that the history of actinistians involved more than a single

developmental trend and that living coelacanths are not “living

fossils” (Cloutier 1991a).

Node 15. The Breathing Chordates (Choanata)

Choanata comprise the last common ancestor that humans

share with lungfishes (fig. 23.1), and all its descendants

(= Rhipidistia of Cloutier and Ahlberg 1996). Choanata

monophyly is supported by genomic DNA (Venkatesh et al.

2001, Hyodo et al. 1997) and morphology (Rosen et al. 1981,

Maisey 1986, Cloutier and Ahlberg 1996), although it remains

among the more controversial nodes within Chordata

(above).

Choanata Characters

The choanate nose and respiratory system. Its namesake

feature is a palatal opening called the choana that communicates

externally via paired external nostrils to the lungs and

pharynx. The interpretation of this region is controversial in

both Paleozoic fossils and Recent taxa, and whether the

choana was actually present ancestrally is in dispute (Rosen

et al. 1981, Maisey 1986, Carroll 2001). Despite debate over

this feature, other transformations of nasal architecture and

function were underway. A nasolacrimal canal is present,

connecting the orbit with the narial passageway (Maisey

1986). The snout in front of the orbits is elongated in association

with these passageways. These facial changes appear

related to modifications in the internal structure of the lung

tied to increase in efficiency of air breathing with the addition

of pulmonary circulation and augmentation of the heart

with two auricles ( Johansen 1970, Rosen et al. 1981).

Simplification of the pharyngeal skeleton. The opercular elements

that enclosed the pharynx in osteichthyans ancestrally

are reduced and the pharyngeal arches are simplified with

the loss of their dorsal (pharyngobranchial) and ventral (interhyal)

elements (Rosen et al. 1981). The upper division of

the second arch, the hyomandibula, is reduced and freed

from its primitive role as a support between the cranium and

jaws. This may signal the beginning of its function in sound

transduction.

Tetrapodous locomotion. Well-developed pectoral and

pelvic skeletons with two primary joints are present, signaling

the beginnings of stereotyped locomotor patterns (Rosen

et al. 1981). In the forelimb, the humerus articulates with the

shoulder girdle in a ball-in-socket joint. Distal to that is the

radius and ulna, which articulate to the humerus in a synovial

elbow joint. The presence of these elements represents

the unfolding of fundamental patterning at a cellular level

(Oster et al. 1988) that persists through most members of

the clade. The pelvis is also strengthened by ventral fusion

of its right and left halves to form a single girdle. In addition,

the musculature that powers the limbs is segmented,

paving the way for a blossoming of limb diversification.

Pan-Choanata

Lying along the stem of either Choanata or Sarcopterygii lies

a poorly known lineage known as Onychodontida (Cloutier

and Ahlberg 1996). If this placement proves correct, its Early

Devonian fossils would be the oldest crown sarcopterygians

yet discovered.

Node 16. The Lungfishes (Dipnoi)

The lungfishes (fig. 23.1) have a 400 Myr history but today

include only six living species. Four live in freshwaters of tropiChordate

Phylogeny and Development 399

cal Africa (Protopterus dolloi, P. annectens, P. aethiopicus, and P.

amphibious), one in South America (Lepidosiren paradoxa), and

one in Australia (Neoceratodus forsteri). The monophyly of

dipnoans has never been challenged. Their most distinctive

features involve the feeding apparatus (Schultze 1987, 1992,

Cloutier and Ahlberg 1996). Lungfish may have teeth along

the margins of their jaws as juveniles, but they are lost in adults.

The adult dentition consists of tooth plates that line the roof

and floor of the mouth. The plates grow by the continual addition

of new teeth and dentine, which consolidate into dental

plates that are not shed (Reisz and Smith 2001).

Pan-Dipnoi (= Dipnomorpha)

Approximately 280 extinct species are known, their record

extending back to the Early Devonian. The earliest dipnomorphs

retain marginal teeth but also have palatal tooth

plates. The earliest members of the lineage are from the Early

Devonian and occupied marine waters, but by the mid-

Devonian skeletal structures associated with air breathing had

appeared and soon thereafter members of the lineage had

moved to the freshwaters that all living species inhabit

(Cloutier and Ahlberg 1996). Yongolepis and Porolepiformes

are extinct lineages known from Devonian rocks that lie at

the base of the stem of the lungfish lineage (Clack 2000).

Node 17. Chordates with Hands and Feet (Tetrapoda)

Tetrapoda (fig. 23.1) comrpise the last common ancestor that

humans share with amphibians, and all its descendants. The

sister relationship between amphibians and amniotes (below)

is supported by molecular (Hedges et al. 1993) and morphological

data (Schultze 1970, 1987, Rosen et al. 1981, Cloutier

and Ahlberg 1996). Historically, the name Tetrapoda designated

all sarcopterygians possessing limbs with digits rather

than fin rays, such as the Devonian Ichthyostega and Acanthostega.

Although it is true that a wide morphological gap separates

the fingers and toes of Ichthyostega, from more basal

sarcopterygians that lack discrete digits such as Eusthenopteron,

the limbs of Ichthyostega are quite different from those inferred

to have been present in the last common ancestor of living

tetrapod species. It was once believed that some of the extant

tetrapod lineages arose independently from fishlike sarcopterygians,

(Jarvik 1996), but recent phylogenetic analyses conclude

that extant amphibians and amniotes share a more recent

common ancestor that is not also shared with Ichthyostega or

Acanthostega. The history of Tetrapoda was long considered

to extend back to the Late Devonian, but under this more restrictive

definition of the name, the oldest known tetrapods

are Carboniferous fossils (Paton et al. 1999).

Tetrapod Characters

The tetrapod limb. In crown tetrapods, the shoulder girdle

has a prominent scapular blade and a posterior corocoidal

region, and the humerus has a discrete shaft. There are fully

differentiated proximal and distal carpals in the wrist and

phalanges in the hand. The ankle also has separate proximal

and distal tarsals and phalanges (Gauthier et al. 1988b). The

evolution of fingers and toes is associated with changes in

the timing and position of expression of the more ancient

Hox genes that regulate development of the body axis and

appendages (Shubin et al. 1997, Carroll 2001). In sampled

actinopterygians, the Hoxd-9 to Hoxd-13 genes are expressed

in an overlapping sequence from the proximal to distal ends

of the posterior surface of the fin. In tetrapods the most distal

gene, Hoxd-13, is expressed over a more anterior portion

of the distal end of the limb, directing distal expansion of

the limb and the formation of fingers and toes. Key components

in the development of separate digits are cell death

(apoptosis) programs directed by the genes encoding Msx2

and BMP4 (Graham and Smith 2001). These were first expressed

in the development of separate pharyngeal arches.

In tetrapods they are co-opted to direct apoptosis in the tissues

that lie between the digits, to produce discrete fingers

and toes (Zhou and Niswander 1996). Lost from the tetrapod

limb are the ectodermal lepidotrichia, along with axial

elements tied to axial locomotion through water, including

the caudal fin rays.

Tetrapod skull. Reduction occurred in the dermal bones

tied to aquatic feeding and respiration, including loss of the

last opercular elements (subopercular, preopercular) and

anterior tectal and internasal (Gauthier et al. 1988b). The

braincase is further enclosed, as the metotic fissure becomes

floored by the basioccipital and basisphenoid, and ossified

lateral “wings” of the parasphenoid expand beneath the otic

capsules. An elongated parasphenoidal cultriform process

extends forward below much of the brain. Tetrapods also

develop an ossified occiput and craniovertebral joint, heralding

independence and mobility of the head on the neck.

Also, the lateral line system of the skull lies almost entirely

in open canals.

Vomeronasal organ. The vomeronasal organ is a paired

structure located in the floor of the nasal chamber, on either

side of the nasal septum. It is a chemoreceptor similar in

general function to the olfactory epithelium and olfactory

nerves and bulb. But unlike olfactory epithelium, its lining

is nonciliated and it has separate innervation by the vomeronasal

nerve, which projects to an accessory olfactory bulb,

rather than to the main bulb as do the olfactory nerves. Its

function is largely in reception of pheromones and other

molecular mediators of social interaction. There is great

elaboration of the vomeronasal organ in squamates, in which

it takes on more general environmental functions. The vomeronasal

organ was once thought to be absent in primates. But

it is present in early development in nearly all mammals, and

may be present in humans (Margolis and Getchell 1988,

Butler and Hodos 1996, Keverne 1999).

Pan-Tetrapoda

The fossil record of stem tetrapods extends from the Middle

Devonian through the Permian and is represented in many

400 The Relationships of Animals: Deuterostomes

parts of the world. However, the fossil record of its sister

taxon (Pan-Dipnoi) suggests that the tetrapod stem extends

to the Early Devonian or Late Silurian (Clack 2000). At the

base of Pan-Tetrapoda lies Osteolepiformes, a diverse group

that ranged from the Middle Devonian to Early Permian. One

especially well-known member is Eusthenopteron ( Jarvik

1996), long thought to be ancestral to tetrapods, now seen

as a distant cousin. Monophyly of Osteolepiformes is not

strongly defended, and some of its members may eventually

find other positions near the base of this part of the chordate

tree. Also near the base of Pan-Tetrapoda is Rhizontida,

which ranged through much of the Devonian and Carboniferous.

Its monophyly is well supported by pectoral fin morphology

and scale composition (Cloutier and Ahlberg 1996).

Some of its members were predators that grew to great size.

Still closer to the tetrapod crown is Elpistostegalia, which

include only Elpistostege and Panderichthyes, from the Late

Devonian of North America and Eastern Europe. These taxa

are similar to Tetrapoda in having a cranial roofing pattern

consisting of paired frontals that lie anterior to the parietals,

and in the flattened shape of the head. They also have a

straight tail lacking dorsal and ventral lobes, and the dorsal

and anal fins are lost. All of these may indicate a shallowwater

lifestyle (Clack 2000).

The Devonian taxa Ichthyostega (Jarvik 1996) and Acanthostega

(Clack 1998) are still closer to the tetrapod crown and

were long considered to be the basalmost tetrapods because

they have hands and feet with discrete digits. However, their

hands and feet were very different from those of extant tetrapods,

as well as from the condition that was present in their

last common ancestor (Gauthier et al. 1989). They have up

to eight toes and retain primitive features such as a welldeveloped

gill arch skeleton and lepidotrichia along the tail

(lost in Tetrapoda), suggesting that they remained primarily

aquatic (Coates and Clack 1990, Cloutier and Ahlberg 1996).

One important feature Ichthyostega and Acanthostega share

with crown tetrapods is the fenestra vestibuli, an opening

through which the stapes communicates to the inner ear,

signaling the beginnings of an airborne-impedance-matching

ear.

Node 18. The Amphibians (Amphibia)

Extant amphibians (= Lissamphibia) comprise 4700 extant

species that are all distributed among the distinctive frog,

salamander (fig. 23.1), and limbless caecilian lineages. All are

small and insectivorous and have wet skins that in many cases

convey oxygen and other exogenous materials into the body.

Hence, they are important as sensitive barometers of freshwater

and riparian environments, and many species are facing

decline. Their skeletons are pedomorphic in many respects,

for example, in the maintenance of extensive cartilage in the

adult skeleton, and in the absence of many membranous

roofing bones (Djorovi7 and Kaleziv7 2000). However, they

are also highly derived in other respects, and none of the

extant species closely resembles its Paleozoic ancestors. Both

molecular and morphological data suggest that frogs and

salamanders are more closely related than either is to caecilians

(Zardoya and Meyer 1996).

Pan-Amphibia

Relationships at the base of Pan-Amphibia are especially

problematic, and more than 100 extinct species have been

named for Permo-Carboniferous fossils alone. The problematic

aпstopods (Carroll 1998, Anderson et al. 2003), nectrideans,

and microsaurs are often regarded as basal members

of Pan-Amphibia. However, all are highly derived and their

positions uncertain. The most basal divergence among panamphibians

was that of the extinct Paleozoic loxammatids

(Beaumont and Smithson 1998, Milner and Lindsay 1998).

Temnospondyles are generally regarded to include all other

panamphibians. Temnospondyles include large extinct Edops,

Eryops, and mastodonsaurids (Damiani 2001) in addition to

extant amphibians and a host of other fossils. These basal taxa

include large and fully aquatic or amphibious carnivores,

some exceeding 2 m in total length. They are distinguished

by the opening of large fenestrae in the roof of the palate.

However, the extinct lepospondyles have also been regarded

as closer relatives of extant amphibians than temnospondyles

(Laurin 1998a, 1998b), and the debate remains active. In

either case, amphibians and amniotes had diverged from the

ancestral tetrapod by the early Carboniferous.

By the Late Triassic, frogs, salamanders, and caecilians

had diverged, and left a fairly detailed fossil record. One of

the most exciting discoveries occurred in Late Jurassic sediments

of northern China, where 500 exceptionally wellpreserved

salamander specimens were recently recovered.

The new finds implicate Asia as the place of salamander diversification

(Gao and Shubin 2001). Amphibian history is

reviewed in detail by Cannatella and Hillis (ch. 25 in this vol.).

Node 19. Terrestrial Chordates (Amniota)

Amniota (fig. 23.1) comprise the last common ancestor that

humans share with Reptilia, and all its descendants (Gauthier

et al. 1988a). Although some members became secondarily

aquatic, the origin of amniotes heralded the first fully terrestrial

chordates. Its monophyly is strongly supported, and its

membership is noncontroversial (with the exception of certain

Paleozoic fossils). However, relationships among the

major living amniote clades are debated. Of principle concern

is whether mammals are closest to birds (Gardiner 1982,

Lшvtrup 1985) or are the sister taxon to other amniotes

(Gauthier et al. 1988a, Laurin and Reisz 1995). Arguments

linking birds and mammals are based on analyses confined

to extant taxa alone, or they treat extant taxa primarily and

then secondarily fit selected fossils to that tree. However,

when all evidence is analyzed simultaneously, mammals are

the sister taxon to other amniotes (Gauthier et al. 1988a,

Laurin and Reisz 1995).

Chordate Phylogeny and Development 401

Amniote Characters

Amniote egg. The amniote egg and attendant equipment

for internal fertilization present a complex of ontogenetic

innovations affording reproductive independence from the

water. Incubation of the amniote embryo is a more protracted

process than before, because the larval stage and metamorphosis

are lost, and instead a fully formed young emerges

from the egg. Amniote eggs are larger than those of most

nonamniotes, with larger volumes of yolk. As the embryo

grows, its size produces special problems with respect to

metabolic intensity, the exchange of respiratory gases, structural

support, and the mobilization and transport of nutrients

(Packard and Seymour 1997, Stewart 1997). The outer

eggshell takes on an important role in mediating metabolism.

It is made of semipermeable collagen fibers and varying proportions

of crystalline calcite, which permits respiration while

preventing desiccation. The eggshell also provides a calcium

repository for the developing skeleton. The embryo is also

equipped with several novel extra-embryonic membranes.

The amnion encloses a fluid filled cavity in which the embryo

develops. The allantois stores nitrogenous wastes, and

the chorion is a respiratory membrane. A single penis with

erectile tissue is also apomorphic of Amniota (Gauthier et al.

1988a).

The amniote skeleton and dentition. Amniotes have a ballin-

socket craniovertebral joint, which increases the mobility

and stability of the head on the neck. They also have two

coracoid ossifications in the shoulder girdle, an ossified astragalus

in the ankle joint, and they lose fishlike bony scales

from the dorsal surface of the body. Teeth are present on the

pterygoid transverse process, but there is no infolding of

enamel anywhere in the dentition. Also present is an enlarged

caniniform maxillary tooth. These changes reflect fully terrestrial

feeding and locomotor patterns (Gauthier et al.

1988b, Laurin and Reisz 1995, Sumida 1997).

Loss of lateral line system. The lateral line placodes fail

to appear in amniotes (Northcutt 1992), and with their loss

is the complete absence of a lateral line system. This is consistent

with the view that amniote origins represent increasingly

terrestrial habits.

Pan-Amniota

The amniote stem is represented by fossils that extend to the

Early Carboniferous (Gauthier et al. 1988b), the oldest being

Casineria (Paton et al. 1999). The best-known members

of the amniote stem include the Carboniferous-Permian

anthracosauroids, seymouriamorphs, and diadectomorphs,

and a handful of other extinct taxa (Gauthier et al. 1988a,

Sumida 1997). Many of the osteological transformations

occurring among stem amniotes involved modifications of

the dentition and palate, and specialization of the atlantoaxial

joint between the head a neck. These modifications

reflect an increased role of the mouth in capturing and manipulating

terrestrial prey items. Also, there was increased

strengthening of the vertebral column via swelling of the

neural arches, the girdles were expanded, the pelvis has an

expanded attachment to the sacrum, and the limbs are elongated.

Loss of the lateral line system was marked by the disappearance

of the canals that it etches into the skull roofing

bones. Collectively these features indicate that increasingly

terrestrial patterns of locomotion, predation, and prey manipulation

preceded the origin of Amniota.

Node 20. The Turtles, Lizards, Crocodilians,

and Birds (Reptilia)

The Reptilia are the lineage stemming from the last common

ancestor of birds and turtles (fig. 23.1). Reptilians comprise

nearly 17,000 living species and enjoy a long and rich fossil

record (Gauthier et al. 1988b, Laurin and Reisz 1995, Dingus

and Rowe 1998). The name Reptilia was long used in reference

to a paraphyletic assemblage of ectothermic amniotes,

including turtles, lizards and snakes, crocodilians, and a host

of extinct forms. Although long considered to have evolved

from reptiles, mammals and birds were excluded from actual

membership within it. More recently, the name Reptilia

was brought into the phylogenetic system by defining its

meaning in reference to the last common ancestor of turtles

and birds, and by including birds within it. The name Reptilia

has also been used to encompass the extinct relatives of

mammals, once known as “mammal-like reptiles.” But in the

phylogenetic system, these taxa are now referred to under

the term Pan-Mammalia (= Synapsida), and the name is rendered

monophyletic by including mammals, plus all extinct

taxa closer to mammals than to reptiles, within it. Reptile

phylogeny is discussed elsewhere in this volume (Lee et al.,

ch. 26, and Cracraft et al., ch. 27).

Pan-Reptilia

The fossil record of Pan-Reptilia extends into the Late Carboniferous

(Gauthier et al. 1988a). Archaeothyris, from the

Joggins fauna of Nova Scotia, is the oldest panreptile that is

known in some detail. In the Early Permian a diversity of

poorly known forms are allied as Parareptilia (Gauthier et al.

1988a, Laurin and Reisz 1995, Berman et al. 2000), a tentatively

monophyletic clade of extinct taxa that all differ considerably

from one another. Their relationships to one another,

and to extant turtles and diapsids remains unstable. Included

among parareptiles are the Carboniferous-Permian mesosaurs,

which seemed highly derived and adapted to a fully aquatic

existence. Also often included are the small terrestrial bolosaurids,

milleretids, and possibly also the procolophonids and

pareiasaurs. The latter two are considered as possible extinct

relatives of turtles, and the pareiasaurs are the only members

of this basal part of the tree that grew to large adult weights

(1000 kg). Pan-Mammalia (see below) dominates the early

fossil record of amniotes, because many of its members expressed

an early trend toward size increase. Panreptiles, with

the exception of pareiasaurs, remained small. By the end of

402 The Relationships of Animals: Deuterostomes

the Triassic, however, these roles reversed, and from then on,

the panreptiles dominate the fossil record and extant reptiles

are far more numerous and diverse than mammals.

Node 21. Chordates with Hair (Mammalia)

Mammalia comprise the last common ancestor that humans

(fig. 23.1) share with living monotremes, plus all its descendants

(Rowe 1987, 1988, 1993, Gauthier et al. 1988a, Rowe

and Gauthier 1992). It includes approximately 5000 living

species and a long fossil record. The mammalian crown extends

to the Middle or Early Jurassic, whereas the base of the

mammalian stem (Pan-Mammalia or Synapsida) traces to the

Late Carboniferous. Mesozoic mammals and their closest

extinct relatives were tiny animals, and their fossils are notoriously

difficult to collect. Most Mesozoic taxa are named

from isolated dentitions or broken jaws, and the early history

of mammals was long shrouded by incompleteness. But

a host of exciting new discoveries from Asia and South

America have yielded relatively complete ancient skeletons.

Some were announced together with detailed phylogenetic

analyses that are rapidly revising and detailing the early phylogeny

of mammals (Hu et al. 1997, Luo et al. 2001a, 2001b,

2002, Rougier et al. 1998). Mammalia is apomorphic in the

brain and special senses, body covering, musculature, skeleton,

circulatory system, respiratory system, digestive system,

reproductive system, metabolism, molecular structure, and

behavior (see Rowe 1988, 1993, 1996a, 1996b, Gauthier et al.

1988a: appx. B). Only a few of these are discussed below.

Mammalian Characters

The neocortex. Compared with even their closest extinct

relatives, mammals have large brains. The additional volume

marks an episode of heterochrony (peramorphosis) in which

the brain began to grow further into ontogeny and more rapidly

than in their extinct relatives, marked by the origin of

the mammalian neocortex. Its two hemispheres each have a

columnar organization of six radial layers, generated in ontogeny

by waves of migrating cells that originate from the

ventricular zone and move radially outward to their adult

positions. This inside-out pattern of neural growth produces

a huge cortical volume in mammals. The developing mammalian

forebrain hypertrophies into inflated lobes that swell

backward over the midbrain and forward around the bases

of the olfactory bulbs, which themselves are inflated. The

cerebellum is also expanded and deeply folded. The neocortex

supports heightened olfactory and auditory senses, and

coincident, overlapping sensory and motor maps of the entire

body surface. The enlarged cerebellum is related to acquisition

and discrimination of sensory information, and

the adaptive coordination of movement through a complex

three-dimensional environment. These changes may reflect

invasion of a nocturnal and/or arboreal niche and have been

implicated in the evolution of endothermy (Rowe 1996a,

1996b).

The mammalian middle ear. In adults, the middle ear

skeleton lies suspended beneath the cranium and behind the

jaw. It is an impedance matching lever system that contains

a chain of tiny ossicles connecting an outer tympanum to the

fluid-filled neurosensory inner ear. Its parallel histories in

ontogeny and phylogeny are among the most famous in comparative

biology. The middle ear arose in premammalian

history as an integral component of the mandible. Over a 100

Myr span of premammalian history, its bones were gradually

reduced to tiny ossicles, reflecting specialization for

increasingly high-frequency hearing, whereas the dentary

undertook a greater role in the mandible. Hearing and feeding

were structurally linked in premammalian history, but

in mammals these functions became decoupled as the auditory

chain was detached from the mandible and repositioned

behind it, and a new craniomandibular joint arose between

the dentary and squamosal bones. Separation of the ossicles

from the mandible occurs in all adult mammals and was

widely regarded as the definitive mammalian character under

Linnaean taxonomy (Rowe 1987, 1988). In ontogeny the

auditory chain differentiates and begins growth attached to

the mandible. But the connective tissues joining them are torn

as the brain grows, and the entire auditory chain (stapes,

incus, malleus, ectotympanic) is carried backward during the

next few weeks to its adult position behind the jaw. Transposition

of the auditory chain is a consequence of its differential

growth with respect to the brain. The tiny ear bones

quickly reach adult size, whereas the brain continues to grow

for many weeks thereafter. As the developing brain balloons,

it loads and remodels the rear part of the skull, detaching

the ear ossicles from the developing mandible. Many other

features of the skull were altered by this dynamic epigenetic

relationship between the rapidly growing brain and the tissues

around it (Rowe 1996a, 1996b).

Enhanced olfactory system. The mammalian olfactory system

is unique in the breadth of its discriminatory power.

Approximately 1000 genes encode odorant receptors in the

mammalian nose, making this the largest family in the entire

genome (Ressler et al. 1994). Each gene encodes a different

type of odorant receptor, and the individual receptor

types are distributed in topographically distinct patterns in

the olfactory epithelium of the nose. Their discriminatory

power is multiplied by increased surface area provided by

elaborate scrolling of the bony ethmoid turbinals. This rigid

framework enhances olfactory discrimination by facilitating

the detection of spatial and temporal information as odorant

molecules disperse within the nasal cavity. Each odorant

receptor transmits signals directly to a single glomerulus

in the olfactory bulb without any intervening synapses;

hence, the topographic distribution of odorant receptors over

the ethmoid turbinals is mapped in the spatial organization

of the olfactory bulb. Ossified turbinals occur only in mammals

(and independently in a few birds), although there is

ample evidence of unossified turbinals among their extinct

relatives. Bone is fundamentally structural, and turbinal osChordate

Phylogeny and Development 403

sification may have arisen in response to tighter scrolling,

increased surface area, and an increase in the number of olfactory

odorant receptors in mammals compared with their

closest extinct relatives. The ossified ethmoid turbinal complex

may thus be viewed as the skeleton of the olfactory

system, arising as an integral component of its distinctive

forebrain.

Pan-Mammalia

The mammalian stem lineage, also known as Synapsida, contains

mammals plus all extinct species closer to mammals

than to Reptilia. Panmammalian fossils range back to the Late

Carboniferous, and an exceptionally complete sequence of

fossils links extant mammals to the base of their stem. Before

phylogenetic systematics, the focus of study was to elucidate

the reptile-to-mammal transition. The premammalian

segment of this history was believed marked by rampant

convergence in the evolution of mammal-like sensory, masticatory,

and locomotor systems, and Mammalia itself was

held to be a grade rather than a clade. The major debate involved

rationalizing which character should mark the boundary

between reptilian and mammalian grades. Few claims of

homoplasy were substantiated when the characters were

subjected to rigorous parsimony analyses, and as synapsids

were placed in a taxonomy based on common ancestry (Rowe

and Gauthier 1992).

Pan-Mammalia are diagnosed by the lower temporal

fenestra and a forward-sloping occiput. Its early history saw

enhancement of the locomotor system for fast, agile movement,

and elaboration of the feeding system for macro predation.

The primitive armature of a tympanic impedance

matching ear also appeared early on (Kemp 1983).

A major node on the mammalian stem is Therapsida, whose

fossils date back to the Late Permian. The temporal fenestra

is larger than before, and there is a deeply incised reflected

lamina of the angular (the homologue of the mammalian

ectotympanic), and a deep external auditory meatus. These

denote an ear more sensitive to a broader range of frequencies.

Limb structure indicates a somewhat more erect posture

and narrow-tracked gait, possibly facilitating breathing

while running and a higher metabolic rate (Kemp 1983).

Cynodontia comprise a node within Therapsida whose

monophyly is supported by numerous characters that where

passed on to living mammals. The overriding feature of basal

cynodonts is that their brain had expanded to completely fill

the endocranial cavity, impressing its outer surficial features

into the inner walls of the braincase (Rowe 1996a, 1996b,

Rowe et al. 1995). Osteological synapomorphies include a

broad alisphenoid (epipterygoid) forming the lateral wall to

the braincase, and a double occipital condyle that permitted

wide ranges of stable excursion of the head about the craniovertebral

joint (Kemp 1983). The dentition is differentiated

into simple incisiform teeth, a long canine, and postcanine

teeth with multiple cusps aligned into a longitudinal row. The

dentary was elongated over the postdentary elements, which

are reduced and more sensitive to higher frequencies.

Among nonmammalian cynodonts, those closest to

crown Mammalia were tiny animals. Miniaturization involved

elaborate repackaging of the brain and special sense organs,

remodeling of the masticatory system, an accelerated rate of

evolution in a complex occlusal dentition. The vertebral column

became more strongly regionalized, and the limbs and

girdles were modified for scansorial movement. Several episodes

of inflation in the size of the brain occurred before the

origin of mammals. The recent discovery of Hadrocodon (Luo

et al. 2001b), from the Early Jurassic of China, may indicate

that the neocortex and middle ear transformation originated

just outside the mammalian crown, but it is questionable

whether Hadrocodon lies outside or within the crown. In either

event, inflation of the neocortex and detachment of the

middle ear appear to coincide.

Discussion

Many of the innovations in chordates design described above

arose as unique expressive pathways or as elaborations of

preexisting genetic and developmental mechanisms. For example,

in all chordates, molecular signaling during neurulation

produces anteroposterior regionalization of the embryo,

and a brain that divides into rostral, middle, and caudal divisions,

each with its own region of unique genetic expression.

The genes themselves are more ancient, being expressed

in the same tripartite anteroposterior regionalization of the

brain in arthropods and other bilaterians. But the inductive

pathway of expression in chordates is unique, and it produces

a nervous system radically different from that in arthropods,

or in what was likely to have been present in bilaterians

ancestrally.

Another pattern of morphogenesis and diversification

corresponds to successive increases in the numbers of genes.

The first episode occurred in either Chordata or Euchordata

ancestrally, and in either case was associated with elaboration

of brain and sensory organs, as well as with the appearance

of mesodermal segmentation. The second occurred in

craniates ancestrally and was accompanied by segmentation

of the brain into prosomeres, mesomeres, and rhombomeres

in early development, as well as enhancement of the adult

brain and sensory organs. The third increase occurred in

Vertebrata, and the fourth in Gnathostomata ancestrally, each

in association with further elaboration of the brain and special

senses. Mammalian origins also coincided with an unprecedented

increase in the number of olfactory genes. Mammalian

olfaction is the most sensitive of any chordate, and

with up to 1000 genes coding for different odorant molecule

receptors, olfactory genes comprise the largest single mammalian

gene family. We can expect many similar examples

of this pattern of gene increase and structural elaboration to

be mapped in the near future.

404 The Relationships of Animals: Deuterostomes

The inductive nature of chordate ontogeny provided an

especially rich substrate for evolutionary change. The most

spectacular example is the neural crest, whose motile cells

are induced by the underlying mesoderm and in turn induce

many tissues and structures. The neural crest arose in craniates

ancestrally, building the embryonic cartilaginous cranium,

providing a rigid armature for the brain and special

senses, and the skeleton of the pharynx, and providing a

novel substrate for the tremendous range of evolutionary

variation.

Epigenesis further multiplied these agents of morphogenesis.

Origin of the mammalian middle ear may have been one

such episode, in which early changes in the timing of development

and rate of growth of the brain altered the adjacent

connective tissues and the adult structures forming within

them. In the wake of the ballooning brain, the rear of the

developing mammalian skull is remodeled, and the middle

ear ossicles and eardrum were detached and displaced backward

from their embryonic attachment to the mandible. The

differentiation of neurectoderm is one of the earliest events

in ontogeny, and virtually anything that affects its pattern of

development will set into motion a new dynamic in the surrounding

connective tissues, potentially altering the adult

structures that form within them. Just how much adult chordate

morphology is epigenetically produced remains to be

determined. These examples illustrate that mapping and

understanding the relationship between molecules and morphology,

as it unfolds in the course of ontogeny, is fundamental

to chordate systematics and comparative biology, and

understanding our place in the Tree of Life.

Literature Cited

Ahlberg, P. E. 1999. Something fishy in the family tree. Nature

397:564–565.

Aizenberg, J., G. Lambert, S. Weiner, and L. Addadi. 2002.

Factors involved in the formation of amorphous and

crystalline calcium carbonate: a study of an ascidian

skeleton. J. Am. Chem. Soc. 124:32–39.

Amores, A., A. Force, Y. L. Yan, L. Joly, C Amemiya, A. Fritz,

R. K. Ho, J. Langeland, V. Prince, Y. L. Wang, et al. 1998.

Zebrafish hox clusters and vertebrate genome evolution.

Science 282:1711–1714.

Anderson, J. S., R. L. Carroll, and T. Rowe. 2003. New information

on Lethiscus stocki (Tetrapoda, Lepospondyli, Aпstopoda)

from high-resolution x-ray computed tomography and a

phylogenetic analysis of Aпstopoda. Can. J. Earth Sci.

40:1071–1083.

Arendt, D., and K. Nьbler-Jung. 1996. Common ground plans

in early brain development in mice and flies. Bioessays

18:255–259.

Arendt, D., and K. Nьbler-Jung. 1999. Comparison of early

nerve cord development in insects and vertebrates.

Development 126:2309–2325.

Arratia, G., and M. Cloutier. 1996. Reassessment of the

morphology of Cheirolepis canadensis (Actinopterygii).

Pp. 165–197 in Devonian Fishes and Plants of Miguasha,

Quebec, Canada (H.-P. Schultze and R. Cloutier, eds.).

Verlag Dr. Friedrich Pfeil, Munich.

Ayala, F. J., A. Rzhetsky, and F. J. Ayala. 1998. Origin of the

metazoan phyla: molecular clocks confirm paleontological

estimates. Proc. Natl. Acad. Sci. 95:606–611.

Bardack, D. 1991. First fossil hagfish (Myxinoidea): a record

from the Pennsylvanian of Illinois. Science 254:701–703.

Bardack, D., and E. S. Richardson. 1977. New agnathous fishes

from the Pennsylvanian of Illinois. Fieldiana Geol. 33:489–

510.

Bardack, D., and R. Zangerl. 1968. First fossil lamprey: a record

from the Pennsylvanian of Illinois. Science 162:1265–1267.

Barrington, E. J. W. 1963. An introduction to general and

comparative endocrinology. Clarendon Press, Oxford

Barrington, E. J. W. 1968. Phylogenetic perspectives in

vertebrate endocrinology. Pp. 1–46 in Perspectives in

Endocrinology (E. J. W. Barrington and C. B. Jorgensen,

eds.). Academic Press, New York.

Basden, A. M., G. C. Young, M. I. Coates, and A. Ritchie. 2000.

The most primitive osteichthyan braincase? Nature

403:185–188.

Beamish, R. J. 1985. Freshwater parasitic lamprey on Vancouver

Island and a theory of the evolution of the freshwater parasitic

and nonparasitic life history types. Pp. 123–140 in Evolutionary

biology of primitive fishes (R. E. Foreman, A. Gorbman,

M. J. Dodd, and R. Olsson, eds.). Plenum, New York.

Beaumont, E. H., and T. R. Smithson. 1998. The cranial

morphology and relationships of the aberrant Carboniferous

amphibian Spathicephalus mirus Watson. Zool. J. Linn. Soc.

122:187–209.

Berman, D. S., R. R. Reisz, D. Scott, A. C. Henrici, S. Sumida,

and T. Martens. 2000. Early Permian bipedal reptile.

Science 290:969–972.

Bodsnick, D., and R. G. Northcutt. 1981. Elecroreception in

lampreys: evidence that the earliest vertebrates were

electroreceptive. Science 212:465–467.

Bromham, L., A. Rambaut, R. Forey, A. Cooper, and D. Penny.

1998. Testing the Cambrian explosion hypothesis by using

a molecular dating technique. Proc. Natl. Acad. Sci. USA

95:12386–12389.

Butler, A. B. 2000. Chordate evolution and the origin of

craniates: an old brain in a new head. Anat. Rec. (New

Anat.) 261:111–125.

Butler, A. B., and W. Hodos. 1996. Comparative vertebrate

neuroanatomy. Wiley-Liss, New York.

Cameron, C. B., J. R. Garey, and B. J. Swalla. 2000. Evolution of

the chordate body plan: new insights from phylogenetic

analyses of deuterostome phyla. Proc. Natl. Acad. Sci. USA

97:4469–4474.

Carroll, R. L. 1988. Vertebrate paleontology and evolution.

W. H. Freeman, New York.

Carroll, R. L. 1998. Cranial anatomy of ophiderpetontid

aпstopods: paleozoic limbless amphibians. Zool. J. Linn.

Soc. 122:143–166.

Carroll, R. L. 2001. The origin and early radiation of terrestrial

vertebrates. J. Paleontol. 75:1202–1213.

Chang, M.-M. 1991. “Rhipidistians,” dipnoans, and tetrapods.

Pp. 3–28 in Origins of the higher groups of tetrapods (H. P.

Schultze and L. Treub, eds.). Comstock, Ithaca, NY.

Chordate Phylogeny and Development 405

Chen, J.-Y., J. Dzik, G. D. Edgecombe, L. Ramskцld, and G.-Q.

Zhou. 1995. A possible Early Cambrian chordate. Nature

377:720–722.

Chen, J.-Y., D.-Y. Huang, and C.-W. Li. 1999. An early

Cambrian craniate-like chordate. Nature 402:518–522.

Clack, J. A. 1998. The neurocranium of Acanthostega gunnari

Jarvik and the evolution of the otic region in tetrapods.

Zool. J. Linn. Soc. 122:61–97.

Clack, J. A. 2000. The origin of tetrapods. Pp. 979–1029 in

Amphibia Biology, Vol. 4: Paleontology, the evolutionary

history of amphibians (H. Heatwole and R. L. Carroll, eds.).

Surrey Beatty and Sons, London.

Cloutier, R. 1991a. Diversity of extinct and living actinistian

fishes (Sarcopterygii). Environ. Biol. Fishes 32:59–74.

Cloutier, R. 1991b. Patterns, trends, and rates of evolution

within the Actinistia. Environ. Biol. Fishes 32:23–58.

Cloutier, R., and P. E. Ahlberg. 1996. Morphology, characters,

and the interrelationships of basal sarcopterygians. Pp. 445–

479 in Interrelationships of fishes (M. L. J. Stiassny, L. R.

Parenti, and G. D. Johnson, eds.). Academic Press, San

Diego.

Coates, M. I., and J. A. Clack, 1990. Polydactyly in the earliest

known tetrapod limbs. Nature 347:66–69.

Compagno, L. J. V. 1977. Phyletic relationships of living sharks

and rays. Am. Zool. 17:303–322.

Damiani, R. J. 2001. A systematic revision and phylogenetic

analysis of Triassic mastodonsaurids (Temnospondyli:

Stereospondyli). Zool. J. Linn. Soc. 133:379–482.

de Carvalho, M. R. 1996. Higher-level elasmobranch phylogeny,

basal squaleans, and paraphyly. Pp. 35–62 in Interrelationships

of fishes (M. L. J. Stiassny, L. R. Parenti, and G. D.

Johnson, eds.). Academic Press, San Diego.

Delgado, S., D. Casane, L. Bonnaud, M. Laurin, J.–Y. Sire, and

M. Girondot. 2001. Molecular evidence for Precambrian

origin of amelogenin, the major protein of vertebrate

enamel. Mol. Biol. Evol. 18:2146–2153.

De Pinna, M. C. C. 1996. Teleostean monophyly. Pp. 147–162 in

Interrelationships of fishes (M. L. J. Stiassney, L. R. Parenti,

and G. D. Johnson, eds.). Academic Press, San Diego.

de Queiroz, K., and J. A. Gauthier. 1992. Phylogenetic taxonomy.

Annu. Rev. Ecol. Syst. 23:449–480.

Digital Morphology. 2003. Available: http://www.digimorph.

org. Last accessed 25 December 2003.

Dingus, L., and T. Rowe. 1998. The mistaken extinction:

dinosaur evolution and the origin of birds. W. H. Freeman,

New York.

Djorovi7, A., and M. L. Kaleziv7. 2000. Paedogenesis in

European newts (Triturus: Salamandridae): cranial morphology

during ontogeny. J. Morphol. 243:127–139.

Dominguez, P., A. G. Jacobson, and R. P. F. Jeffries. 2002.

Paired gill slits in a fossil with a calcite skeleton. Nature

417:841–844.

Donoghue, M. J., J. Doyle, J. A. Gauthier, A. G. Kluge, and T.

Rowe 1989. Importance of fossils in phylogeny reconstruction.

Annu. Rev. Ecol. Syst. 20:431–460.

Donoghue, P. C. J., P. L. Forey, and R. J. Aldridge. 2000.

Conodont affinity and chordate phylogeny. Biol. Rev.

75:191–251.

Dryer, L. 2000. Evolution of odorant receptors. Bioessays

22:803–810.

Edgeworth, F. H. 1935. The cranial muscles of vertebrates.

Cambridge University Press, Cambridge.

Erdmann, M. V., R. L. Caldwell, S. L. Jewett, and M. K. Moosa.

1998. Indonesian “king of the sea” discovered. Nature

395:335.

Evrard, Y. A., Y. Lun, A. Aulehla, L. Gan, and R. L. Johnson.

1998. Lunatic fringe is an essential mediator of somite

segmentation and patterning. Nature 394:377–381.

Forey, P. L. 1984a. The coelacanth as a living fossil. Pp. 166–

169 in Living fossils (N. Eldredge and S. M. Stanley, eds.).

Springer Verlag, New York.

Forey, P. L. 1984b. Yet more reflections on agnathangnathostome

relationships. J. Vert. Paleontol. 4:330–343.

Forey, P. L. 1998. History of the coelacanth fishes. Chapman

and Hall, London.

Fritzsch, B. 1996. Similarities and differences in lancelet and

craniate nervous systems. Isr. J. Zool. 42:S147–S160.

Froese, R., and D. Pauly. 2001. Fishbase. Database. Available:

http://www.fishbase.org. Last accessed 25 December 2003.

Gagnier, P.–Y. 1989. The oldest vertebrate: a 470 million years

old jawless fish, Sacabambaspis janvieri, from Ordovician of

Bolivia. Natl. Geogr. Res. 5:250–253.

Galis, F. 2001. Evolutionary history of vertebrate appendicular

musculature. Bioessays 23:383–387.

Gans, C., and C. J. Bell. 2001. Vertebrates, overview. Pp. 755–

766 in Encyclopedia of biodiversity (S. A. Levin, ed.).

Academic Press, San Diego.

Gans, C., N. Kemp, and S. Poss (eds.). 1996. The lancelets

(Cephalochordata): a new look at some old beasts. Isr. J.

Zool. 42(suppl.).

Gao, K.-Q., and N. H. Shubin. 2001. Late Jurassic salamanders

from northern China. Nature 410:574–577.

Gardiner, B. 1982. Tetrapod classification. Zool. J. Linn. Soc.

74:207–232.

Gardiner, B. G. 1984. The relationships of placoderms. J. Vert.

Paleontol. 4:379–395.

Gauthier, J. A., D. Cannatella, K. de Queiroz, A. G. Kluge, and

T. Rowe. 1989. Tetrapod phylogeny. Pp. 337–353 in The

hierarchy of life. (B. Fernholm, H. Bremer, H. Jцrnvall,

eds.). Nobel Symposium 70. Excerpta Medica, Amsterdam.

Gauthier, J. A., and K. de Queiroz. 2001. Feathered dinosaurs,

flying dinosaurs, crown dinosaurs, and the name “Aves.”

Pp. 7–41 in New perspectives on the origin and early

evolution of birds (J. Gauthier and L. Gall, eds.). Yale

University Press, New Haven, CT.

Gauthier, J., A. G. Kluge, and T. Rowe. 1988a. Amniote

phylogeny and the importance of fossils. Cladistics 4:105–

209.

Gauthier, J., A. G. Kluge, and T. Rowe. 1988b. The early

evolution of the Amniota. Pp. 103–155 in The phylogeny

and classification of the tetrapods, Vol. 1: Amphibians,

reptiles and birds (M. Benton, ed.). Syst. Assoc. Spec. Vol.

35a. Clarendon Press, Oxford.

Gegenbaur, C. 1878. Elements of comparative anatomy.

Macmillan, London.

Gehring, W. J. 1998. Master control genes in development and

evolution. Yale University Press, New Haven, CT.

Gerhart, J. 2000. Inversion of the chordate body axis: are there

alternatives? Proc. Natl. Acad. Sci. USA 97:4445–4448.

Goodwin, M., M. M. Miyamoto, and J. Czelusniak. 1987.

406 The Relationships of Animals: Deuterostomes

Pattern and process in vertebrate phylogeny revealed by

coevolution of molecules and morphology. Pp. 141–176 in

Molecules and morphology in evolution: conflict or

compromise? (Colin Patterson, ed.). Cambridge University

Press, Cambridge.

Gould, S. J. 1989. Wonderful life. Norton, New York.

Graham, A., and A. Smith. 2001. Patterning the pharyngeal

arches. Bioessays 23:54–61.

Grande, L., and W. E. Bemis. 1996. Interrelationships of

Acipenseriformes, with comments on “Chondrostei.”

Pp. 85–116 in Interrelationships of fishes (M. L. J. Stiassney,

L. R. Parenti, and G. D. Johnson, eds.). Academic Press, San

Diego.

Hall, B. K. 1992. Evolutionary developmental biology. Chapman

and Hall, London.

Hardisty, M. W. 1979. Biology of the cyclostomes. Chapman

and Hall, London.

Hardisty, M. W. 1982. Lampreys and hagfishes: analysis of

cyclostome relationships. Pp. 165–260 in The biology of

lampreys (M. W. Hardisty and I. C. Potter, eds.), vol. 4B.

Academic Press, New York.

Hedges, S. B. 2001. Molecular evidence for the early history of

living vertebrates. Pp. 119–134 in Major events in early

vertebrate evolution (P. Ahlberg, ed.). Taylor and Francis,

London and New York.

Hedges, S. B., C. A. Hass, and L. R. Maxson. 1993. Relations of

fish and tetrapods. Nature 363:501–502.

Hillis, D. M., and M. T. Dixon 1989. Vertebrate phylogeny:

evidence from 28S ribosomal DNA sequences. Pp. 355–367

in The hierarchy of life (B. Fernholm, K. Bremer, and

H. Jцrnvall, eds.). Elsevier, New York.

Holder, M. T., M. V. Erdmann, T. P. Wilcox, R. L. Caldwell, and

D. M. Hillis. 1999. Two living species of coelacanths? Proc.

Natl. Acad. Sci. USA 96:12616–12620.

Holland, N. D., and J. Chen. 2001. Origin and early evolution

of the vertebrates: new insights from advances in molecular

biology, anatomy, and palaeontology. Bioessays 23:142–

151.

Holland, P. W. H., and J. Garcia-Fernаndez. 1996. Hox genes

and chordate evolution. Dev. Biol. 173:382–394.

Hu, Y., Y. Wang, Z. Luo, and C. Li. 1997. A new symmetrodont

mammal from China and its implications for mammalian

evolution. Nature 390:137–142.

Hyodo, S., S. Ishii, and J. M. P. Joss. 1997. Australian lungfish

neurohypophysial hormone genes encode vasotocin and

[Phe2] mesotocin precursors homologous to tetrapod-type

precursors. Proc. Natl. Acad. of Sci. USA 94:13339–13344.

Jacobson, A. G. 1988. Somitomeres: mesodermal segments of

vertebrate embryos. Development 104(suppl.):209–220.

Jacobson, A. G. 2001. Somites and head mesoderm arise from

somitomeres. Pp. 16–29 in The origin and fate of somites

(E. J. Sanders, ed.). IOS Press, Amsterdam.

Jamieson, B. G. M. 1991. Fish evolution and systematics:

evidence from spermatozoa. Cambridge University Press,

Cambridge.

Janvier, P. 1996. Early vertebrates. Clarendon Press, Oxford.

Janvier, P. 1997. Hyperoarteia in Tree of Life (D. Maddison,

ed.). Available: http://www.tolweb.org. Last accessed 25

December 2003.

Janvier, P. 1999. Catching the first fish. Nature 402:21–22.

Janvier, P., and R. Lund. 1983. Hardistiella montanensis n. gen.

et sp. (Petromyzontida) from the Lower Carboniferous of

Montana, with remarks on the affinities of lampreys. J. Vert.

Paleontol. 2:407–413.

Jarvik, E. 1996. The evolutionary importance of Eusthenopteron

foordi (Osteolepiformes). Pp. 285–315 in Devonian fishes

and plants of Miguasha, Quebec, Canada (H.-P. Schultze

and R. Cloutier, eds.). Verlag Dr. Friedrich Pfeil, Munich.

Jeffries, R. P. S. 1979. The origin of chordates—a methodological

essay. Pp. 443–477 in The origins of major invertebrate

groups (M. R. House, ed.). Syst. Assoc. Spec. Vol. 12.

Academic Press, London.

Jeffries, R. P. S., 1980. Zur Fossilgeschichte des Ursprungs der

Chordaten und der Echinodermaten. Zool. Jahrb. Jena Abt.

Anat. 103:285–353.

Jeffries, R. P. S. 1986. Ancestry of the vertebrates. Cambridge

University Press, New York.

Jeffries, R. P. S. 2001. The origin and early fossil record of the

acustico-lateralis system, with remarks on the reality of the

echinoderm-hemichordate clade. Pp. 40–66 in Major events

in early vertebrate evolution (P. Ahlberg, ed.). Taylor and

Francis, London.

Jeffries, R. P. S., and D. N. Lewis. 1978. The English Silurian

fossil Placocystites forbesianus and the ancestry of the

vertebrates. Philos. Trans. R. Soc. Lond. 282:205–323.

Jiang, Y. J., B. L. Aerne, L. Smithers, C. Haddon, D. Ish-

Horowicz, and J. Lewis. 2000. Notch signaling and the

synchronization of the somite segmentation clock. Nature

408:475–479.

Johansen, K. 1970. Air breathing in fishes. Pp. 361–413 in Fish

phylogeny (W. S. Hoar and D. J. Randall, eds.), vol. 4.

Academic Press, London.

Jollie, M. 1982. What are the “Calcichordata”? and the larger

question of the origin of chordates. Zool. J. Linn. Soc.

75:167–188.

Kemp, T. S. 1983. Mammal-like reptiles and the origin of

mammals. Academic Press, London.

Keverne, E. B. 1999. The vomeronasal organ. Science 286:716–720.

Kumar, S., and S. B. Hedges. 1998. A molecular time scale for

vertebrate evolution. Nature 392:917–920.

Lacalli, T. C. 1996a. Frontal eye circuitry, rostral sensory

pathways and brain organization in amphioxus larvae:

evidence from 3D reconstructions. Philos. Trans. R. Soc.

Lond. B 351:243–263.

Lacalli, T. C. 1996b. Landmarks and subdomains in the larval

brain of Branchiostoma: vertebrate homologs and invertebrate

antecedents. Isr. J. Zool. 42:S131–S146.

Lagios, M. D. 1979. The Coelacanth and Chondrichthyes as

sister groups: a review of shared apomorphous characters

and a cladistic analysis and reinterpretation. Pp. 25–44 in

The biology and physiology of the living coelacanth ( J. E.

McCosker and M. D. Lagios, eds.). Calif. Acad. Sci.

Occasional Papers 134.

Laird, D. J., A. W. De Tomaso, M. D. Cooper, and I. L.

Weissman. 2000. 50 million years of chordate evolution:

seeking the origins of adaptive immunity. Proc. Natl. Acad.

Sci. USA 97:6924–6926.

Lanfranchi, G., A. Pallavicini, P. Laveder, and G. Valle. 1994.

Ancestral hemoglobin switching in lampreys. Dev. Biol.

164:402–408.

Chordate Phylogeny and Development 407

Lauder, G. V., and K. F. Liem. 1983. The evolution and

interrelationships of the actinopterygian fishes. Bull. Mus.

Comp. Zool. 150:95–197.

Laurin, M. 1998a. The importance of global parsimony and

historical bias in understanding tetrapod evolution. Part I—

systematics, middle ear evolution, and jaw suspension. Ann.

Sci. Nat. Zool. Paris 19:1–42.

Laurin, M. 1998b. The importance of global parsimony and

historical bias in understanding tetrapod evolution. Part

II—vertebral centrum, costal ventilation, and paedomorphosis.

Ann. Sci. Nat. Zool. Paris 19:99–114.

Laurin, M., and Reisz, R. R. 1995. A reevaluation of early

amniote phylogeny. Zool. J. Linn. Soc. 113:165–223.

Lipscomb, D. L., J. S. Farris, M. Kдllersjo, and A. Tehler. 1998.

Support, ribosomal sequences and the phylogeny of

eukaryotes. Cladistics 14:303–338.

Litman, G. W., M. K. Anderson, and J. P. Rast. 1999. Evolution of

antigen binding receptors. Annu. Rev. Immunol. 17:109–147.

Long, J. A. 1995. The rise of fishes. Johns Hopkins University

Press, Baltimore, MD.

Lшvtrup, S. 1977. The phylogeny of Vertebrata. John Wiley and

Sons, London.

Lшvtrup, S. 1985. On the classification of the taxon Tetrapoda.

Syst. Zool. 34:463–470.

Lund, R., and W. L. Lund. 1985. Coelacanths from the Bear

Gulch Limestone (Namurian) of Montana and the evolution

of the Coelacanthiformes. Bull. Carnegie Mus. Nat. Hist.

25:1–74.

Luo, Z.-X., R. L. Cifelli, and Z. Kielan-Jaworowska. 2001a. Dual

origin of tribosphenic mammals. Nature 409:53–57.

Luo, Z.-X., A. W. Crompton, and A.-L. Sun. 2001b. A new

mammaliaform from the Early Jurassic and evolution of

mammalian characteristics. Science 292:1535–1540.

Luo, Z.-X., Z. Kielan-Jaworowska, and R. L. Cifelli. 2002. In

quest for a phylogeny of Mesozoic mammals. Acta Palaeontol.

Pol. 47:1–78.

Maisey, J. 1984. Chondrichthyan phylogeny: a look at the

evidence. J. Vert. Paleontol. 4:359–371.

Maisey, J. 1986. Heads and tails: a chordate phylogeny.

Cladistics 2:201–256.

Maisey, J. 1988. Phylogeny of early vertebrate skeletal induction

and ossification patterns. Pp. 1–36 in Evolutionary biology

(M. K. Hecht, B. Wallace, and G. T. Prance, eds.), vol. 22.

Plenum, New York.

Maisey, J. G. 2001. Remarks on the inner ear of elasmobranches

and its interpretation from skeletal labyrinth morphology.

J. Morphol. 250:236–264.

Maisey, J. G., and E. M. Anderson. 2001. A primitive chondrichthyan

braincase from the Early Devonian of South

Africa. J. Vert. Paleontol. 21:702–713.

Mallatt, J. 1984. Feeding ecology of the earliest vertebrates.

Zool. J. Linn. Soc. 82:261–272.

Mallatt, J. 1985. Reconstructing the life cycle and the feeding of

ancestral vertebrates. Pp. 59–68 in Evolutionary biology of

primitive fishes (R. E. Foreman, A. Gorbman, J. M. Dodd,

and R. Olsson, eds.). Plenum, New York.

Mallatt, J. 1997a. Crossing a major morphological boundary:

the origin of jaws in vertebrates. Zoology 100:128–140.

Mallatt, J. 1997b. Shark pharyngeal muscles and early vertebrate

evolution. Acta Zool. Stockh. 78:279–294.

Mallatt, J., and J. Sullivan. 1998. 28S and 18S rDNA sequences

support the monophyly of lampreys and hagfishes. Mol.

Biol. Evol. 15:1706–1718.

Mallatt, J., J. Sullivan, and C. J. Winchell. 2001. The relationship

of lampreys to hagfishes: a spectral analysis of

ribosomal DNA sequences. Pp. 106–118 in Major events in

early vertebrate evolution (P. Alberch, ed.). Taylor and

Francis, London.

Manzanares, M., H. Wada, N. Itasaki, P. R. Trainor, R. Krumlauf,

and P. W. H. Holland. 2000. Conservation and elaboration

of Hox gene regulation during evolution of the vertebrate

head. Nature 408:854–857.

Margolis, F. L., and T. V. Getchell (eds.). 1988. Molecular

neurobiology of the olfactory system. Plenum, New York.

Marshall, C., and H.-P. Schultze. 1992. Relative importance of

molecular, neontological and paleontological data in

understanding the biology of the vertebrate invasion of the

land. J. Mol. Evol. 35:93–101.

Mazan, S., D. Jaillard, B. Baratte, and P. Janvier. 2000. OTX1

gene-controlled morphogenesis of the horizontal semicircular

canal and the origin of the ganthostome characteristics.

Evol. Dev. 2:186–193.

McMahon, T., and J. T. Bonner. 1985. On size and life.

Scientific American Library, New York.

Meinertzhagen, I. A., and Y. Okamura 2001. The larval ascidian

nervous system: the chordate brain from its small beginnings.

Trends Neurosci. 24:401–410.

Metzger, R. J., and M. A. Krasnow. 1999. Genetic control of

branching morphogenesis. Science 284:1635–1639.

Meyer, A. 1995. Molecular evidence on the origin of tetrapods

and the relationships of the coelacanth. Trends Ecol. Evol.

10:111–116.

Milner, A. R., and A. C. Lindsay. 1998. Postcranial remains of

Baphetes and their bearing on the relationships of the

Baphetidae (= Loxommatidae). Zool. J. Linn. Soc. 1998:211–

235.

Moyle, P. B., and J. J. Cech, Jr. 2000. Fishes, an introduction to

ichthyology. 4th ed. Prentice Hall, New York.

Nelson, G. 1989. Phylogeny of major fish groups. Pp. 325–336

in The hierarchy of life (B. Fernholm, K. Bremer, and H.

Jцrnvall, eds.). Nobel Symposium 70. Excerpta Medica,

Amsterdam.

Newman, S. A., and W. D. Comper. 1990. “Generic” physical

mechanisms of morphogenesis and pattern formation.

Development 110:1–18.

Niedert, A. H., V. Virupannavar, G. W. Hooker, and J. A.

Langeland. 2001. Lamprey Dlx genes and early vertebrate

evolution. Proc. Natl. Acad. Sci. USA 96:1665–1670.

Nielsen, C. 1999. Origin of the chordate central nervous system

and the origin of chordates. Dev. Genes Evol. 209:198–205.

Northcutt, R. G. 1992. The phylogeny of the octavolateralis

ontogenies: a reaffirmation of Garstang’s phylogenetic

hypothesis. Pp. 21–47 in The evolutionary biology of

hearing (D. B. Webster, R. R. Fay, and A. N. Popper, eds.).

Springer Verlag, New York.

Northcutt, R. G., and C. Gans. 1983. The genesis of neural crest

and epidermal placodes: a reinterpretation of vertebrate

origins. Q. Rev. Biol. 58:1–28.

Oster, G. F., N. Shubin, J. D. Murray, and P. Alberch. 1988.

Evolution and morphogenetic rules: the shape of the

408 The Relationships of Animals: Deuterostomes

vertebrate limb in ontogeny and phylogeny. Evolution

42:862–884.

Packard, M. J., and R. S. Seymour. 1997. Evolution of the

amniote egg. Pp. 265–290 in Amniote origins (S. S. Sumida

and K. L. M. Martin, eds.). Academic Press, San Diego.

Paton, R. L., T. R. Smithson, and J. A. Clack. 1999. An amniotelike

skeleton from the Early Carboniferous of Scotland.

Nature 1999:508–513.

Philippe, H., A. Chenuil, and A. Adoutte. 1994. Can the

Cambrian explosion be inferred through molecular

phylogeny? Development (suppl.):15–25.

Pohlmann, K., F. W. Grasso, and T. Breithaupt. 2001. Tracking

wakes: the nocturnal predatory strategy of piscivorous

catfish. Proc. Natl. Acad. Sci. USA 98:7371–7374.

Poss, S. G., and H. T. Boschung. 1996. Lancelets (Cephalochordata:

branchiostomidae): How many species are valid?

Isr. J. Zool. 42:S13–S66.

Pouyaud, L., S. Wirjoatmodjo, I. Rachmatika, A. Tjakrawidjaja,

R. Hadiaty, and W. Hadie. 1999. Un nouvelle espиce de

coelacanthe: preuves gйnйtiques et morphologiques. Compt.

Rend. Acad. Sci. 322:261–267.

Reisz, R. R., and M. M. Smith. 2001. Lungfish dental pattern

conserved for 360 million years. Nature 411:548.

Ressler, K. J., S. L. Sullivan, and L. B. Buck. 1994. Information

coding in the olfactory system: evidence for a stereotyped

and highly organized epitope map in the olfactory bulb.

Cell 79:1245–1255.

Romer, A. S. 1966. Vertebrate paleontology. University of

Chicago Press, Chicago.

Rosen, D. E., P. L. Forey, B. G. Gardiner, and C. Patterson.

1981. Lungfishes, tetrapods, paleontology, and plesiomorphy.

Bull. Am. Mus. Nat. Hist. 167:162–275.

Rougier, G. W., J. R. Wible, and M. J. Novacek. 1998. Implications

of Deltatheridium specimens for early marsupial

history. Nature 396:459–463.

Rowe, T. 1987. Definition and diagnosis in the phylogenetic

system. Syst. Zool. 36:208–211.

Rowe, T. 1988. Definition, diagnosis and origin of Mammalia.

J. Vert. Paleontol. 8:241–264.

Rowe, T. 1993. Phylogenetic systematics and the early history of

mammals. Pp. 129–145 in Mammalian phylogeny (F. S.

Szalay, M. J. Novacek, and M. C. McKenna, eds.). Springer-

Verlag, New York.

Rowe, T. 1996a. Brain heterochrony and evolution of the

mammalian middle ear. Pp. 71–96 in New perspectives on

the history of life (M. Ghiselin and G. Pinna, eds.). Calif.

Acad. Sci. Memoir 20.

Rowe, T. 1996b. Coevolution of the mammalian middle ear and

neocortex. Science 273:651–654.

Rowe, T., C. A. Brochu, and K. Kishi (eds.). 1999. Cranial

morphology of Alligator and phylogeny of Alligatoroidae.

J. Vert. Paleontol. 19(2 suppl.), mem. 6.

Rowe, T., W. Carlson, and W. Bottorff. 1995. Thrinaxodon:

digital atlas of the skull (CD-ROM). 2nd ed. (MS Windows

and Macintosh). University of Texas Press, Austin.

Rowe, T., and J. Gauthier. 1992. Ancestry, paleontology, and

definition of the name Mammalia. Syst. Biol. 41:372–378.

Rowe, T., J. Kappelman, W. D. Carlson, R. A. Ketcham, and

C. Denison. 1997. High-resolution computed tomography:

a breakthrough technology for Earth scientists. Geotimes

42:23–27.

Ruben, J. A., and A. F. Bennett. 1980. Antiquity of the vertebrate

pattern of activity metabolism and its possible relation

to vertebrate origins. Nature 286:886–888.

Russell, E. S. 1916. Form and function. John Murray, London.

Sansom, I. J., M. P. Smith, H. A. Armstrong, and M. M. Smith.

1992. Presence of earliest vertebrate hard tissues in

conodonts. Science 256:1308–1311.

Sansom, I. J., M. M. Smith, and M. P. Smith. 1996. Scales of

thelodont and shark-like fishes from the Ordovician of

Colorado. Nature 379:628–630.

Schaeffer, B. 1987. Deuterostome monophyly and phylogeny.

Pp. 179–235 in Evolutionary biology (M. K. Hecht,

B. Wallace, and G. T. Prance, eds.), vol. 21. Plenum,

New York.

Schaeffer, B., and M. Williams. 1977. Relationships of fossil and

living elasmobranches. Am. Zool. 17:293–302.

Schluter, S. F., R. M. Bernstein, H. Bernstein, and J. J. Marchalonis.

1999. “Big bang” emergence of the combinatorial immune

system. Dev. Comp. Immunol. 23:107–111.

Schultze, H.-P. 1970. Folded teeth and the monophyletic origin

of tetrapods. Am. Mus. Nov. 2408:1–10.

Schultze, H.-P. 1987. Dipnoans as sarcopterygians. J. Morphol.

1(suppl.):39–74.

Schultze, H.-P. 1992. Dipnoi. Pp. 1–464 in Fossilium catalogus

I: Animalia (F. Westphal, ed.), pt. 131. Kugler Publications,

Amsterdam.

Schultze, H.-P. 1994. Comparisons of hypotheses on the

relationships of sarcopterygians. Syst. Biol. 43:155–173.

Shimeld, S. M., and P. W. H. Holland. 2000. Vertebrate

innovations. Proc. Natl. Acad. Sci. USA 97:4449–4452.

Shirai, S. 1996. Phylogenetic interrelationships of neoselachians

(Chondrichthyes: Euselachii). Pp. 9–34 in Interrelationships

of fishes (M. L. J. Stiassny, L. R. Parenti, and G. D. Johnson,

eds.). Academic Press, San Diego.

Shu, D. G., L. Chen, J. Han, and X.-L. Zhang. 2001a. An Early

Cambrian tunicate from China. Nature 411:472–473.

Shu, D. G., S. Conway Morris, J. Han, L. Chen, X. L. Zhang,

H. Q. Liu, Y. Li, and J. N. Liu. 2001b. Primitive deuterostomes

from the Chengjiang Lagerstдtte (Lower Cambrian,

China). Nature 414:419–424.

Shu, D. G., S. Conway Morris, and X. L. Zhang. 1996. A Pikaialike

chordate from the Lower Cambrian of China. Nature

384:157–158.

Shu, D. G., S. Conway Morris, X.-L. Zhang, L. Chen, Y. Li, and

J. Han. 1999a. A pipiscid-like fossil from the Lower

Cambrian of south China. Nature 400:746–749.

Shu, D. G., H. L. Luo, S. Conway Morris, X. L. Zhang, S. X. Hu,

L. Chen, J. Han, M. Zhu, Y. Li, and L. Z. Chen. 1999b.

Lower Cambrian vertebrates from south China. Nature

402:42–46.

Shubin, N. H., and P. Alberch. 1986. A morphogenetic

approach to the origin and basic organization of the

tetrapod limb. Pp. 319–387 in Evolutionary biology (M. K.

Hecht, B. Wallace, and G. T. Prance, eds.), vol. 20. Plenum,

New York.

Shubin, N. H., C. Tabin, and S. Carroll. 1997. Fossils, genes

and the evolution of animal limbs. Nature 388:639–647.

Chordate Phylogeny and Development 409

Slack, J. M. W. 1983. From egg to embryo: determinative events

in early developmental history. Cambridge University Press,

London.

Slavkin, H. C., E. E. Graham, M. Zeichner-David, and

W. Hildemann. 1983. Enamel-like antigens in hagfish:

possible evolutionary significance. Evolution 37:404–412.

Smith, M. M., and B. K. Hall. 1990. Development and evolutionary

origins of vertebrate skeletogenic and odontogenic

tissues. Biol. Rev. 65:277–373.

Stensiц, E. A. 1968. They cyclostomes with special reference

to the diphyletic origin of the Petromyzontida and

Myxinoidea. Pp. 13–71 in Current problems in lower

veterbrate phylogeny (T. Шrvig, ed.). Nobel Symposium 4.

Excerpta Medica, Amsterdam.

Stewart, J. 1997. Morphology and evolution of the egg of

oviparous amniotes. Pp. 291–326 in Amniote origins (S. S.

Sumida and K. L. M. Martin, eds.). Academic Press, San

Diego.

Sumida, S. S. 1997. Locomotor features spanning the origin of

amniotes. Pp. 353–398 in Amniote origins (S. S. Sumida

and K. L. M. Martin, eds.). Academic Press, San Diego.

Suzuki, M., K. Kubokawa, H. Nagasawa, and A. Urano. 1995.

Sequence analysis of vasotocin cDNAs of the lamprey

Lampetra japonica, and the hagfish Eptatretus burgeri:

evolution of cyclostome vasotocin precursors. J. Mol.

Endocrinol. 14:67–77.

Swalla, B. J., and W. R. Jeffery. 1996. Requirement of the Manx

gene for expression of chordate features in a tailless ascidian

larva. Science 274:1205–1208.

Turbeville, J. M., J. R. Schulz, and R. A. Raff. 1994. Deuterostome

phylogeny and the sister group of the chordates—

evidence from molecules and morphology. Mol. Biol. Evol.

11:648–655.

Venkatesh, B., M. V. Erdmann, and S. Brenner. 2001. Molecular

synapomorphies resolve evolutionary relationships of extant

jawed vertebrates. Proc. Natl. Acad. Sci. USA 98:11382–

11387.

Venkatesh, B., Y. Ning, and S. Brenner. 1999. Late changes in

spliceosomal introns define clades in vertebrate evolution.

Proc. Natl. Acad. Sci. USA 96:10267–10271.

von Wahlert, G. 1968. Latimeria und die Geschichte der

Wirbeltiere: eine evolutionsbiologische Untersuchung.

Gustav Fischer Verlag, Stuttgart.

Webb, J. F., and D. M. Noden. 1993. Ectodermal placodes:

contributions to the development of the vertebrate head.

Am. Zool. 33:434–447.

Wiley, E. O. 1979. Ventral gill arch muscles and the phylogenetic

relationships of Latimeria. Pp. 56–67 in The biology

and physiology of the living coelacanth (J. E. McCosker and

M. D. Lagios, eds.). Calif. Acad. Sci. Occasional Papers 134.

Wray, G. A., J. S. Levinton, and L. H. Shapiro. 1996. Molecular

evidence for deep Precambrian divergences among metazoan

phyla. Science 274:568–573.

Yalden, D. W. 1985. Feeding mechanisms as evidence of

cyclostome monophyly. Zool. J. Linn Soc. 84:291–300.

Zardoya, R., and A. Meyer. 1996. Evolutionary relationships of

the coelacanth, lungfishes, and tetrapods based on the 28S

ribosomal RNA gene. Proc. Natl. Acad. Sci. USA 93:5449–

5454.

Zhou, H., and L. Niswander. 1996. Requirement for BMP

signaling in interdigital apoptosis and scale formation.

Science 272:738–741.

Zhu, M., and H.–P. Schultze. 1997. The oldest sarcopterygian

fish. Lethaia 30:293–304.

Zhu, M., X. Yu, and P. E. Ahlberg. 2001. A primitive fossil

sarcopterygian fish with an eyestalk. Nature 410:81–84.

Zhu, M., X. Yu, and P. Janvier. 1999. A primitive fossil fish sheds

light on the origin of bony fishes. Nature 397:607–610.

Навигация

• Главная
• Книги
• Статьи
• Обратная связь
• Поиск