Benton M.J. Introduction to paleobiology and the fossil record

Подождите немного. Документ загружается.

428 INTRODUCTION TO PALEOBIOLOGY AND THE FOSSIL RECORD

The backbone is the key. Human beings are

vertebrates, and so are horses, sparrows, alli-

gators, turtles, frogs and trout. What they all

share is their bony internal skeleton, and, in

particular, vertebrae – the individual elements

of the backbone. The skeleton consists of a

backbone, a skull enclosing the brain and

sense organs, and bones supporting the ﬁ ns

or limbs. Vertebrates are important today

because humans are such a successful species,

and also because of the huge diversity and

abundance of species of bony ﬁ shes, birds and

mammals. Other groups, such as insects

and microbes, are even more abundant and

diverse, but vertebrates include the largest

animals on land, in the sea and in the air.

Vertebrates are a subgroup of the Phylum

Chordata, a major deuterostome clade.

Current views and debates about the nearest

relatives of vertebrates are considered in

Chapter 14. In this chapter, we look at the

origin of vertebrates, the evolution of ﬁ shes

from the Cambrian to the present day, and the

Paleozoic tetrapods. The end-Permian mass

extinction reset the clock for vertebrates on

land, so we save the dinosaurs and their allies

and the mammals for Chapter 17. If the ver-

tebrate skeleton is so signiﬁ cant, what is so

special about it?

ORIGIN OF THE VERTEBRATES

The skeleton

The skeleton of vertebrates is made from bone

and cartilage. Bone consists of a network of

collagen ﬁ bers on which needle-like crystals

of hydroxyapatite (a form of apatite, calcium

phosphate, CaPO

) accumulate. Hence bone

has a ﬂ exible component and a hard compo-

nent, which explains why bones may undergo

a great deal of strain before they break, and

also why bones do not break along simple

brittle faces. Cartilage is a ﬂ exible, gristly

tissue, usually unmineralized, and containing

collagen and elastic tissues. In humans, most

of the bones are laid down in the early embryo

in the form of cartilage, and this progressively

mineralizes by deposition of apatite. In adult

humans, cartilage can be found in ﬂ exible

parts like the ears and the nose, as well as at

the ends of the ribs and some limb bones.

The ﬁ rst vertebrates probably had a carti-

laginous skeleton. Some of the oldest ﬁ sh

fossils, such as Sacabambaspis from the Ordo-

vician of Brazil (Fig. 16.1a), had the begin-

nings of a bony skeleton, but only on the

outside of the body, and there is no trace

preserved of an internal mineralized skeleton.

The rigid armor is made up of lots of little

tooth-like structures, each equivalent to an

individual shark scale, but united by continu-

ous sheets of bone arranged like plywood.

This shows how adaptable the vertebrate

skeleton can be, and this is perhaps why ver-

tebrates became such a diverse and abundant

group. The internal skeleton of vertebrates

has a unique property – it allows them to

(a)

(b)

(c)

20 mm

pectoral fin

10 mm

dorsal spine

dorsal plate

rostral plate

orbital plate

ventral plate

branchial plate

Figure 16.1 Early jawless ﬁ shes: (a)

Sacabambaspis from the Mid Ordovician of

Brazil, the oldest well-preserved ﬁ sh; (b) the

osteostracan Hemicyclaspis from the Devonian;

and (c) the heterostracan Pteraspis, also from the

Devonian. (a, b, based on Gagnier 1993; c,

based on Moy-Thomas & Miles 1971.)

FISHES AND BASAL TETRAPODS 429

grow very large because the skeleton can grow

with the animal. An external skeleton cannot

grow so fast, and is less adaptable in support-

ing a large volume of soft tissues. Further, the

external skeleton is vulnerable to damage and

either has to be repaired by extending ﬂ eshy

parts outside the shell (mollusks, graptolites)

or by molting the skeleton (arthropods), a

wasteful process that uses up energy and

leaves the animal vulnerable until the new

exoskeleton hardens. By contrast, the verte-

brate skeleton is maintained and remodeled

constantly within the body, and can act as a

support for small, medium, large and massive

organisms.

Jawless ﬁ shes: slurping rather than biting

Two key deﬁ ning characters of vertebrates are

the head and neural crest tissues. Our head is

so essential that we rarely stop to think that

actually only vertebrates have heads – indeed

vertebrates are sometimes called craniates,

meaning “with a skull”. Mollusks, worms,

brachiopods and echinoderms do not have

heads – we might call the front end of a worm

its “head”, but it really is not any more than

its front end. The vertebrate head is unique in

providing an organized structure that con-

tains the brain, the major sense organs and

the mouth.

The vertebrate head is formed from cells

derived from the neural crest, a second key

apomorphy of vertebrates. The neural crest

appears in the early embryo as a strip of cells

lying just below the outer skin, the ectoderm,

of the embryo, above the line of where the

backbone will develop. As tissues begin to

differentiate in the early embryo, cells derived

from the neural crest spread through the

embryo and stimulate the development of

muscles, nerves and blood vessels along the

trunk and around the heart and gut, but a

major target is the head region. The cranial

neural crest cells give rise to bones, cartilage,

nerves and connective tissue in the head and

neck region, forming the face, teeth, eyes,

inner ear, the thymus, thyroid and parathy-

roid glands, and the gills and gill arches of

ﬁ shes.

The ﬁ rst vertebrates had no jaws (Fig.

16.1). Until recently, these ﬁ rst ﬁ shes were

said to be Ordovician in age, but controver-

sial new specimens from the remarkable fossil

sites at Chengjiang in China (Box 16.1) have

pushed the range back to the Early Cambrian.

In the Late Cambrian and Ordovician, the

commonest vertebrates were the conodont

animals. Fishes became common and diverse

during the Late Silurian and Devonian.

The jawless ﬁ shes are sometimes referred

to as ostracoderms (Box 16.2). Ostracoderms

were jawless, they were generally armored,

although some were not, and they had their

heyday in the Devonian. Osteostracans like

Hemicyclaspis (Fig. 16.1b) have a semicircu-

lar head shield bearing openings on top for

the eyes and nostrils, as well as porous regions

round the sides that may have served for the

passage of electrical sense organs, perhaps

used in detecting other animals by their move-

ments in the water. Heterostracans like Pter-

aspis (Fig. 16.1c), are more streamlined in

shape, and were perhaps more active swim-

mers. Both forms have their mouths under-

neath the head shield, and they probably fed

by sieving organic matter from the sediment.

These armored jawless ﬁ shes died out at the

end of the Devonian, and their place was

taken over by ﬁ shes with jaws.

Jawless ﬁ shes still exist today, the 50 or so

species of lampreys and hagﬁ shes, eel-shaped

animals. Hagﬁ shes scavenge on dead ﬂ esh,

while lampreys are often parasitic. Although

they have no jaws, their mouths are ﬁ lled with

tooth-bearing bones, and these are used to

grip prey animals and to rasp off lumps of

ﬂ esh. Salmon and trout are commonly caught

in the American Great Lakes with huge circu-

lar craters in the sides of their bodies, where

ﬂ esh has been torn out by a sea lamprey.

Conodonts: animals of mystery

The commonest early vertebrates were the

conodont animals (Sweet & Donoghue 2001).

For over 150 years conodonts had been a

mystery, known only from their jaw elements

– no one knew which animal had produced

them.

Conodonts were ﬁ rst identiﬁ ed by the

Latvian embryologist and paleontologist

Christian Pander in 1856. They occur as

phosphatic tooth-like microfossils, termed

elements. Three main conodont groups have

been established (Fig. 16.3): (i) protocon-

odonts such as Hertzina are simple cones with

deep basal cavities; (ii) paraconodonts like

430 INTRODUCTION TO PALEOBIOLOGY AND THE FOSSIL RECORD

Box 16.1 The world’s oldest vertebrates

There was a sensation in 1999 when Shu Degan and colleagues (Shu et al. 1999) announced a new

fossil vertebrate, Myllokunmingia, from the Early Cambrian locality Chengjiang in China. This site

has become celebrated for the exceptional preservation of all kinds of animal fossils, and it rivals

the Burgess Shale (see p. 249) as a window into Cambrian life. Until 1999, the oldest vertebrates

were much debated, with some tentative Middle and Late Cambrian candidates, but nothing really

certain until the Ordovician.

Myllokunmingia (Fig. 16.2) is tiny, less than 30 mm long – you could hold a hundred or so of

them, like a handful of wriggling whitebait. The head is poorly deﬁ ned, but there seems to be a

mouth at one end. Relatives seem to show detail in the head, possible eyes and a brain. If it has

such differentiated head features, it is a vertebrate. Behind the “head” are six gill pouches, a possible

heart cavity and a gut. Above these are the notochord, a key chordate character (see p. 410),

and myotomes or V-shaped muscle blocks. There is a narrow dorsal ﬁ n along the back, and possibly

a ventral ﬁ n below. Myllokunmingia presumably swam by ﬂ icking its body and ﬁ ns from side to

side and wriggling forward through the water. None of the Chinese specimens have mineralized

bone – but this does not rule them out as vertebrates. Evidently, the vertebrate skeleton began

as a cartilaginous structure in early forms, and became mineralized with apatite later in the

Cambrian.

In the same paper, Shu et al. (1999) also named Haikouichthys, a similar early vertebrate from

Chengjiang. A rival team, Hou et al. (2002), suggest that Haikouichthys was the same as Myllokun-

mingia, although Shu and colleagues disagree. The two groups, led by Shu and Hou, also disagree

over the identiﬁ cation of different organs within these fossils, and this affects where they are placed

in the vertebrate phylogeny. The Chengjiang fossils are preserved in grey or yellow sediment, and

the fossils may be grey or reddish, with the internal organs picked out in grey, brown and black

colors. Interpreting these multicolored blobs and squiggles would test the patience of a saint, and

yet it is remarkable that such details have been preserved for 500 Myr. There are now more than

500 specimens of these early vertebrates, so further intensive study may clarify their anatomy

further.

See http://www.blackwellpublishing.com/paleobiology/ for relevant web links.

(a)

5 mm

(b)

NotochordMyotomes

Gut Ventrolateral fin ?Heart cavity

Gill pouch Mouth

Dorsal fin

Figure 16.2 The basal vertebrate Myllokunmingia from the Early Cambrian of Chengjiang,

China: (a) photograph of specimen, and (b) interpretive drawing showing possible identities of the

internal organs. (Courtesy of Shu Degan.)

FISHES AND BASAL TETRAPODS 431

Furnishina are mainly simple cones; and (iii)

euconodonts or true conodonts are more

complex, with cones, bars and blades. The

protoconodonts are almost certainly unre-

lated to true conodonts; they may be chaeto-

gnaths or arrow worms, a group of basal

metazoans of uncertain afﬁ nities.

Euconodonts occur as three broad types of

element, consisting of laminae of apatite.

These are formed by outer accretion from an

initial growth locus. White matter often occurs

between, or crosscuts, lamellae; this material

compares well with the composition and

structure of vertebrate bone. The three main

morphotypes of conodont element have been

used in the past as the basis of a crude single

element or form taxonomy (Fig. 16.4). The

cones or coniform elements are the simplest,

with the base surmounted by a cone-like cusp,

tapering upward, and sometimes ornamented

with ridges or costae (Fig. 16.4a, b). Bars or

ramiform elements consist of an elongate

blade-like ridge with up to four processes

developed posteriorly, anteriorly or laterally

to the cusp (Fig. 16.4c, d, g). Platforms or

pectiniform elements have a wide range of

shapes, with denticulate processes extending

both anteriorly, posteriorly and/or laterally

from the area of the basal cavity (Fig. 16.4e,

f, h–j); some also have primary lateral pro-

cesses. The cusp is attenuated, whereas the

base may be expanded to form a platform

with denticles on its upper surface. The basal

cavity is ﬁ lled by the basal body of the element

in the form of a dentine-like material, although

this is not always preserved.

Conodonts are common in certain marine

facies from the Cambrian to the Triassic.

Box 16.2 Classiﬁ cation of ﬁ shes

“Fishes” form a paraphyletic grouping, consisting of several distinctive clades of swimming vertebrates.

Ordovician and Silurian records of placoderms, acanthodians, chondrichthyans and osteichthyans

are mainly isolated scales and teeth; these groups are best known from the Devonian onwards.

Subphylum VERTEBRATA

“Class AGNATHA”

• A paraphyletic group of jawless ﬁ shes, including armored and unarmored Paleozoic ostraco-

derms, and modern lampreys and hagﬁ shes

• Late Cambrian to Recent

Class PLACODERMI

• Heavily armored ﬁ shes with jaws and a hinged head shield

• Mid Silurian to Late Devonian

Class CHONDRICHTHYES

• Cartilaginous ﬁ shes, including modern sharks and rays

• Late Ordovician to Recent

Class ACANTHODII

• Small ﬁ shes with many spines and large eyes

• Late Ordovician to Early Permian

Class OSTEICHTHYES

• Bony ﬁ shes, with ray ﬁ ns (Subclass Actinopterygii) or lobe ﬁ ns (Subclass Sarcopterygii), the latter

including ancestors of the tetrapods

• Late Silurian to Recent

432 INTRODUCTION TO PALEOBIOLOGY AND THE FOSSIL RECORD

Paraconodonts are reported from the Mid

Cambrian; older records are doubtful. During

the Late Cambrian, simple conical eucon-

odonts appeared. In the Early Ordovician,

apparatuses with coniforms, and some with

coniform and ramiform element types,

appeared. Conodont diversity peaked during

the Mid Ordovician, with a global maximum

of over 60 genera. During this interval of

experimentation, there was a huge diversity

of apparatus patterns never again matched;

later apparatuses are relatively uniform,

perhaps indicating stabilization of feeding

modes. Pectiniform elements were common

from the Early Ordovician, together with a

wide variety of blades and platforms in the

Mid to Late Ordovician. This great diversity

of forms was wiped out by the Late Ordovi-

cian mass extinction (see p. 169). Silurian

faunas are less variable, mainly apparatuses

with ramiform and pectiniform elements. The

conodonts again radiated during the Late

Devonian, with specialized ramiform and pec-

tiniform elements; over 1000 conodont taxa

have been named from the Upper Devonian.

Carboniferous conodonts (Fig. 16.5a) were

characterized by a lack of coniform elements,

together with pectiniform elements in the P

apparatus position, whereas ramiform ele-

ments occupied the M and S positions (see

FurnishinaHertzina

Ozarkodina

Prioniodina

Polygnathus

Amorphognathus

posterior

posterioranterior

posterior

basal cavity

(a)

(c)

(b)

basal cavity

basal

cavity

basal cavity

blade

main cusp

denticles

posterior

blade

platform

secondary

platforms

lateral

process

blade

carina

platform

inner side

outer side

Figure 16.3 Descriptive morphology of the main types of conodont elements: (a) protoconodont

Herzina (×40); (b) paraconodont Furnishina (×40); and (c) euconodonts Ozarkodina (×40), Prionodina

(×20), Polygnathus (×40) and Amorphognathus (×40). (Based on Armstrong & Brasier 2004.)

FISHES AND BASAL TETRAPODS 433

next paragraph). Conodonts became rarer

during the Early Permian, and most Late

Permian and Triassic species had small appa-

ratuses. They became extinct at the end of the

Triassic.

The ﬁ rst piece of evidence about the iden-

tity of the conodont animal is that the ele-

ments sometimes occur in a cluster or

apparatus consisting typically of 15 elements,

14 of them arranged bilaterally and one sym-

metric element positioned on the midline. The

elements are arranged in a particular way in

the apparatus: pectiniform (P elements) at the

back, makelliform (M elements) at the front,

and symmetry transition series (S elements) in

between. Generally bars and platforms occupy

P positions, whereas bars and cones are found

in M and S positions. The P, M and S posi-

tions may be deﬁ ned more precisely with sub-

scripts, for example P

and P

elements.

The ﬁ rst conodont apparatus was found in

1879, and this gave some idea about the func-

tion of conodonts, perhaps as some sort of

teeth, and provided some clues about the

whole animal. Several supposed conodont

animals were identiﬁ ed in the 1960s, but most

of these turned out to be predatory critters

that had just eaten a conodont animal, and

so had lots of conodont elements inside

them!

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

Figure 16.4 Conodont elements: (a, b) coniform,

lateral view; (c, d) ramiform, lateral view;

(e) straight blade, upper view; (f) arched blade,

lateral view; (g) ramiform, posterior view; and

(h-j) platform, upper view. Magniﬁ cation ×20–35

for all. (Courtesy of Dick Aldridge.)

(a)

(b)

Figure 16.5 Homing in on the conodont animal:

(a) natural assemblage of conodonts from the

Carboniferous of Illinois (×24); and (b) the

conodont animal from the Carboniferous

Granton Shrimp Bed, Edinburgh, Scotland, with

the head at left-hand end (×1.5). (Courtesy of

Dick Aldridge.)

434 INTRODUCTION TO PALEOBIOLOGY AND THE FOSSIL RECORD

Despite the mystery of their identity, con-

odonts became key tools in biostratigraphy

(Box 16.3). In addition, because color changes

of the elements can be related to changing

temperature, conodonts are important indica-

tors of thermal maturation. Now paleontolo-

gists believe they know what conodont

animals looked like, but it took 150 years to

work this out.

The solution came in 1983, when the ﬁ rst

complete conodont animal was found in the

Granton Shrimp Bed, a dark Carboniferous

mudstone on the seacoast near Edinburgh,

Scotland (Fig. 16.5b). This was an eel-like

animal with a conodont apparatus at its front

end. Detailed examination showed that the

elements were in place and, this time, had not

been merely eaten by the animal. Ten con-

odont animals have now been found, as well

as examples from other localities (Aldridge et

al. 1993a). The Scottish conodont animal is

up to 55 mm long, and has a short, lobed

head with large goggling eyes that are fossil-

ized black, perhaps a stain produced by the

visual pigments. Below and behind the eyes is

the conodont apparatus, clearly located where

the mouth should be, showing that conodont

elements really did function as teeth. The

Box 16.3 Conodonts and biostratigraphy

Detailed biostratigraphic schemes based on conodonts have been established for many parts of the

Paleozoic and Triassic. For example, over 20 conodont zones have been determined for the Ordovi-

cian System, while the Upper Devonian is the most congested interval, with over 30 biozones, each

less than 500,000 years long. In northwest Europe the Carboniferous is routinely correlated on the

basis of conodont zones.

Remarkable precision is now available in some zonal schemes. This has permitted the development

of models for global environmental change during the Early Silurian (Fig. 16.6) tied to a tight con-

odont zonation (Aldridge et al. 1993b). Two oceanic states are recognized: those with oxygenated

cool oceans that had a good vertical circulation and adequate supplies of nutrients (termed “primo”),

and those with warm stratiﬁ ed oceans that had deep saline levels and poor nutrient supplies (termed

“secundo”). Sudden changes between ocean states altered the vertical circulation and nutrient supply

dramatically, perhaps causing extinction events.

These kinds of stratigraphic schemes may depend on geographic zonations. Cambrian conodont

faunas were divided into equatorial (low latitude) warm-water associations and polar (high latitude)

cool-water associations. During the Early Ordovician, these low- and high-latitutde assemblages

further divided into six discrete provinces. Conodonts evolved independently at high latitudes, and

there were only a few incursions from lower-latitude faunas. Towards the end of the Ordovician

high-latitude, cold-water faunas migrated into lower latitudes. Thus Late Ordovician equatorial

mid-continent assemblages originated in polar and subpolar regions and themselves formed the

foundation for the Silurian fauna. During the Mid and Late Paleozoic, conodonts were mainly

restricted to tropical latitudes. Devonian and Carboniferous faunas show some biogeographic dif-

ferentiation among shelf associations. These differences among the geographic provinces can affect

the stratigraphic schemes and the possibility of correlation from area to area.

Conodonts occur in a wide range of marine and marine-marginal environments, although the

group is most common in nearshore carbonate facies, commonly in the tropics. Distinct environ-

ment-related conodont paleocommunities have been identiﬁ ed in many parts of the Paleozoic, and

statistical analysis may discriminate, for example, deeper-water from shallow-water assemblages. It

is important to be aware of the inﬂ uence of depth and other factors on the distribution of communi-

ties before they are used in establishing biostratigraphic zones. It would clearly be a mistake to

identify distinctive depth-determined conodont assemblages and then to interpret them as indicators

of different time intervals.

Web links are available through http://www.blackwellpublishing.com/paleobiology/.

FISHES AND BASAL TETRAPODS 435

anterior comb-like ramiform elements proba-

bly grasped prey items that were sucked

towards the mouth, and the posterior pectini-

form elements may have chomped the food

before swallowing. One the greatest mysteries

in paleontology had been solved.

JAWS AND FISH EVOLUTION

The ﬁ rst jaws

The basal vertebrates, including conodonts,

lacked jaws, and jaws probably evolved during

the Ordovician. Study of the anatomy of

modern vertebrates suggests that jaws may

have evolved from the strengthening bars of

cartilage or bone between the gill slits, each

of which consists of several elements, all

linked by tiny muscles. The transition cannot

be followed in fossils because the gill skeleton

of jawless ﬁ shes was not mineralized. Molecu-

lar biologists have even suggested that the

origin of jaws was so profound that it must

have been associated with a dramatic genome

duplication event – but the fossils say no (Box

16.4).

Some of the oldest jaw-bearing ﬁ shes were

the placoderms, such as Coccosteus (Fig.

16.8a), which had an armor of large bony

plates over the head and shoulder region, as

in the ostracoderms, and a more lightly

armored posterior region. They swam by

beating this tail region from side to side. The

edges of the jaws did not carry teeth, but

instead sharp bony plates that would have

been just as effective in snapping at prey.

CONODONT ZONES

Solvik Rytteråker Vik Skinnerbukta

Ireviken Event

Snipklint Primo

Episode

Malmøykalven

Secundo

Episode

Sandvika Event

Jong Primo Episode

Spirodden

Secundo

Episode

Telychian

Aeronian

Rhuddanian

Pterospathodus

amorphognathoides

Pterospathodus

celloni

Distomodus

staurognathoides

Distomodus

kentuckyensis

Icriodella discreta

Ozarkodina hassi

O. oldhamensis

O? kentuckyensis

Pranognathos tenuis

Distomodus spp.

Pseudolonchodina fluegeli

Pterospathodus celloni

Carniodus carnulus

Aulacognathus bullatus

Pseudooneotodas tricornis

Pterospathodus amorphognathoides

STAGES

Figure 16.6 The use of conodont assemblages in stratigraphy: alternation of primo and secundo

oceanic states correlated with part of the Lower Silurian succession of the Oslo region, Norway.

In the stratigraphic column, limestone is shown by a blocky pattern and mudstone by gray.

(Courtesy of Dick Aldridge.)

436 INTRODUCTION TO PALEOBIOLOGY AND THE FOSSIL RECORD

Placoderms were fearsome predators, some of

them, like Dunkleosteus from the Late Devo-

nian of North America, reaching the impres-

sive length of 10 m. This was the largest

animal that had lived until then, and its size

and fearsome jaws may explain why so many

Devonian ﬁ shes were armored.

Other Devonian ﬁ shes were more modern

in appearance. The ﬁ rst shark-like chondrich-

thyans, or cartilaginous ﬁ shes, came on the

scene during the Early Devonian. Acanthodi-

ans were small ﬁ shes, mostly in the range

50–200 mm in length, and they bore numer-

ous spines at the front of each ﬁ n and in

Box 16.4 Genome duplications and vertebrate evolution

Vertebrates have larger genomes than other animal groups. The genome is the entire sequence of

genes contained on all the chromosomes within the nuclei of cells. Various worms and insects have

around 15,000 genes in their genomes, while the ﬁ gure is 31,000 for humans, 30,000 for the mouse

and 38,000 for the pufferﬁ sh. However, vertebrates do not just have more genes than invertebrates,

they have two, four or even eight copies of many individual invertebrate genes. At one time, molecu-

lar biologists thought that humans had as many as 100,000 genes, but the reduced ﬁ gure was

established in 2004 after the intense gene sequencing efforts of the Human Genome Project. What

does genome size mean?

Some have suggested that genome size maps on to the complexity of an organism. Surely, a single-

celled bacterium does not need many genes because it does not do much, and vertebrates, as much

more complex organisms, would need more genes. Humans ought to have the largest genomes since

we are somehow very complex and important. In fact, genome size is only loosely related to bodily

complexity: the largest genome reported so far comes from a lungﬁ sh! Much of the genome is so-

called junk DNA, or at least duplicate genes and non-coding sections, so the functional genome size

might be a better correlate of function or bodily complexity.

Whether functional or not, molecular biologists have proposed that there were at least

three genome duplication events (GDEs) in the history of vertebrates – times when evolutionary

change was dramatic and large sectors of the genome duplicated. GDEs are identiﬁ ed at the

origin of vertebrates, the origin of gnathostomes and the origin of teleosts, the hugely diverse modern

bony ﬁ shes (Furlong & Holland 2004). Could the evolutionary jump have caused the GDE, or

perhaps the GDE stimulated rapid and fundamental reorganization of the ﬁ shes at these three

points?

Donoghue and Purnell (2005) suggest that molecular biologists have been misled. By omitting

fossils, they see artiﬁ cial morphological jumps in their cladograms, and then link this to the

postulated GDE. In fact, when fossils are inserted, the “jumps” seem less clear. For the origin of

gnathostomes, biologists have compared lampreys with sharks, and there is a wide gulf

between these two groups, so suggesting quite a leap in terms of anatomic change and in terms of

genome duplication. However, when fossils are inserted (Fig. 16.7), seven major ostracoderm and

placoderm clades fall between the living groups, and the evolutionary transition is stretched. Some

of the fossil groups (especially pteraspidimorphs, conodonts and placoderms) were diverse, and it is

not clear that the GDE drove, or permitted, a single dramatic burst of speciation, as had been

proposed. Further, it is not clear that there was a single reorganization of anatomy associated with

the origin of jaws and the GDE: the fossils show step-by-step character changes over a long

interval.

This is a developing ﬁ eld of study. The claim that genome duplication can drive major bursts

of evolution is dramatic, and perhaps overstated. Paleontologists can make profound contribu-

tions in new areas of science by working hand-in-hand with molecular and developmental

biologists.

Read more through http://www.blackwellpublishing.com/paleobiology/.

FISHES AND BASAL TETRAPODS 437

spaced rows on their undersides (Fig. 16.8b).

Acanthodians are often found preserved in

vast numbers in the rock layers, so they prob-

ably swam in huge shoals in open water,

perhaps feeding on small arthropods and

plankton. They escaped predators by rapid

darting from side to side in their shoals, and

perhaps their exceptional spininess made them

difﬁ cult to swallow.

Bony ﬁ shes: ray ﬁ ns and lobeﬁ ns

The osteichthyans, or bony ﬁ shes, also ap -

peared in the Devonian. There are two groups:

(i) those with ray-like ﬁ ns, the actinopteryg-

ians, ancestors of most ﬁ shes today from carp

to salmon, and seahorse to tuna; and (ii) the

lobeﬁ ns, the sarcopterygians, that had thick,

muscular, limb-like ﬁ ns. Today, the lobeﬁ ns

are rare, being represented by only three

species of lungﬁ shes and the rare coelacanth.

The coelacanth Latimeria is a famous “living

fossil”. Until 1938, coelacanths were only

known as Devonian to Cretaceous fossils, but

in 1938 the world was astounded to hear that

a living coelacanth had been ﬁ shed out of

deep waters off East Africa, and more have

been caught since then.

The ray ﬁ ns of the Devonian include Chei-

rolepis (Fig. 16.8c), which had a ﬂ exible body

covered with small scales and a plated head.

This was an active predator that may have fed

on acanthodians. The Devonian lobeﬁ ns

include both lungﬁ shes and “rhipidistians”.

Number of families

Heterostraci

Conodonta

Tunicata

Cephalochordata

Myxinoidea

Petromyzontida

Thelodonti

Galeaspida

Osteostraci

Placodermi

Chondrichthyes

Acanthodii

Actinopterygii

Sarcopterygii

100

Anaspida

Duplication

Stem CrownGnathostomes

Figure 16.7 Phylogeny of the basal ﬁ shes. One major genome duplication event was apparently

associated with the origin of jaws. When the fossil groups (open lines) are omitted, there is a

large morphological and genomic leap from jawless lampreys and hagﬁ shes; when the fossil

groups are included, as here, the transition appear much more gradual. The timing of the genome

duplication events is uncertain, and falls within the area of the gray box. The number of families

within each living and fossil group is shown by the shaded vertical bars. (Courtesy of Phil

Donoghue.)