A Critical Look at Creationist Paleontology

"If creation is true, we'd expect each one of the created kinds . . . to appear abruptly and fully formed, with no indication that they had evolved from a common ancestor" (Duane Gish; Saladin - Gish debate, 1988)

The past 2 decades have seen an amazing number of paleontological discoveries. These discoveries allow for an ever-more-rigourous assessment of various theories about life as it has existed and diversified through time.

The theories of common descent CD and biblical special creation BSC, together with background assumptions about the nature of the geologic record, both make predictions about the patterns we should find in the fossil record. Common descent predicts, for example, that more "derived" groups should appear over time in the fossil record, and that the ratio of living to extinct organisms will increase as a function of time/stratigraphic position. BSC, on the other hand, predicts the simultaneous origination of all "major" groups in the fossil record. [note: Creationists have in fact seriously argued that the arrangement of organisms in the geologic record is an ecological/hydrodynamic artifact of Noah's flood. Here it is taken for granted that this view is false and that the geologic record was formed over a "long" period of time.]

The first prediction of BSC --the simultaneous origination of all "major" groups, is clearly falsified. It is simply not the case that all "kinds" of organisms appear simultaneously in the fossil record. In contrast, the fossil record is fully consistent with the prediction that more "derived" groups should appear over time in the fossil record. Nowhere is this better documented than within out own subphylum, the vertebrates. The first vertebrates appear about 40 million years later, at the end of the Cambrian or the beginning of the Ordovician. The first vertebrates with limbs appear about 365 million years ago, or about 145 million years after the first vertebrates. The first reptiles appear in the early Pennsylvanian, about 320 million years ago. The first mammals and the first birds appear in the late Triassic and late Jurassic, about 225 and 150 million years ago, respectively. And so on. Nearly all of the terrestrial vertebrates familiar to us today, especially the mammals such as horses, elephants, whales, etc., appear in the fossil record recently, less than 65 million years ago. As Hunt (1999) notes, the fossil record is in general characterized by a "remarkable temporal pattern of fossil morphology, with . . . animal groups appearing in a certain unmistakable order. For example, primitive fish appear first, amphibians later, then reptiles, then primitive mammals, then (for example) legged whales, then legless whales. This temporal- morphological correlation is very striking, and appears to point overwhelmingly toward an origin of all vertebrates from a common ancestor." The following graph shows this pattern of first-occurences for some of the taxa considered in this paper.

So the first and most basic prediction that SC makes about the fossil record is falsified, and the first and most basic prediction of CD is confirmed; but what about the prediction that the earliest members of "major groups" of organisms appear "fully formed," with no evidence linking them to some previously existing taxon? The obvious way to evaluate this claim is by examining the earliest known occurences of various taxa in the fossil record. Did the earliest members of these taxa (say amphibians, mammals, and birds) appear "all at once," looking much like they do today, or is there evidence linking them to some preexisting group of organisms? In the following sections, we will examine the earliest known members of various taxa, and show that once again the predictions of CD are confirmed whereas those of SC are repeatedly falsified.

Fish and the Earliest Amphibians

See: Glenn Morton's Fish to Amphibian Transition - Documented

The earliest semi-terrestrial animals, the amphibians, are not found in the fossil record until the late Devonian period, about 365 million years ago. The first 150 million years or so of the fossil record, including the Cambrian, Ordovician, Silurian, and most of the Devonian period, are lacking completely in terrestrial animals of any sort.

Paleontologists noted at the beginning of the 20th century that one particular genus of Devonian lobe-finned fish, Eusthenopteron, shared a range of anatomic features with the earliest known (at the time) tetrapod specimens. Eusthenopteron shared the same intricate pattern of skull bones as the Icthyostegids, including the presense of internal nostrils, which are found only in osteolepiform fish and later amphibians. Eusthenopteron also had robust fin bones homologous with the limbs of the earliest tetrapods. For most of this century, the gap between Eusthenopteron was not illuminated by intermediate fossils. This situation has changed drastically over the past 2 decades.

The more recently discovered Panderichthys is an even more amphibian-like fish than Eusthenopteron, so much so that it was initially confused for an amphibian. For instance, the dorsal and anal fins have disappeared in Panderichthys, leaving only the four bony fins (pectoral and pelvic) which later become legs; the fin bones themselves show a striking similarity to the tetrapod pattern of humerus, ulna and radius in the forelimb and femur, tibia and fibula in the hindlimb. Panderichthys still had gills, but had lungs and nostrils also. The braincase and body is much more flattened and amphibian-like than Eusthenopteron, essentially intermediate between Eusthenopteron and the later Acanthostega and Icthyostega. Morton points out that Panderichthys also displayed a choana, the opening that allows air to pass from the nasal passage to the mouth (Schultze, 1991, p. 58). The choana is a derived characteristic found only in tetrapods. Finally, unlike fish but like amphibians, Panderichthys had developed a fin-topped tail. Both Eusthenopteron and Panderichthys are dated to about 378 million years ago.

Suaripterus, dated to about 370 million years ago, is known from only a partial find. It is a fish with 8 bony digits buried in its fins (Daeschler and Shubin, 1998). As we will see below, this peculiar configuration of bones is found in only one other fossil species, Acanthostega, the earliest known tetrapod.

Two more excellent morphologic intermediate forms have been dated to about 363 million years ago. Although both Acanthostega and Ichthyostega have tetrapod limbs, they could just as easily be called tetrapod fins. Both Acanthostega and Ichthyostega have basically the same compliment of limb bones found in later tetrapods, yet they are covered with a fin-like covering; it is doubtful that either species could support their weight outside of the water. Both animals had a tail covered with long fish fins almost identical to that of Panderichthys. Coates & Clack (1991) recently described a well-preserved fossil of Acanthostega, which revealed the presense of internal gills, an obvious fish trait not found in any modern amphibians.

It's interesting to note that aside from Eusthenopteron and Ichthyostega, most of these intermediate forms were discovered in the past 2 decades. In 1985, for example, Duane Gish could claim that "not a single transitional form has ever been found showing an intermediate stage between the fin of the crossopterygian and the foot of the Ichthyostegid. The limb and the limb girdle of Icthyostega were already of the basic amphibian type, showing no vestige of a fin ancestry" (p 72). Clearly this gap has been filled, not only by Pandericthys and Acanthostega, but also by several other recent discoveries. Gish's claim that the limbs of Icthyostega show "no vestige of a fin ancestry" is most impressively refuted by the recent disovery of the first well-preserved Icthyostegid hindlimbs. Kathleen Hunt (1999) notes:

"Coates & Clack (1990) . . . recently found the first really well- preserved feet, from Acanthostega (front foot found) and Ichthyostega (hind foot found) . . . The feet were much more fin-like than anyone expected. It had been assumed that they had five toes on each foot, as do all modern tetrapods. This was a puzzle since the fins of lobe-finned fishes don't seem to be built on a five-toed plan. It turns out that Acanthostega's front foot had eight toes, and Ichthyostega's hind foot had seven toes, giving both feet the look of a short, stout flipper with many 'toe rays' similar to fin rays."

If a Sauripterus-like eight-digit appendage arranged in a fin ray pattern and covered with a fleshy fin does not suggest a "fin ancestry," I don't know what will! The fossil record now documents the step-wise evolution of tetrapod limbs from fish-fins quite clearly, complete with the leg-fin intermediates Gish asked for. New specimens continue to emerge, as the recent discoveries of Elginerpeton, Ventastega, and Sauripterus demonstrate. Sauripterus is especially interesting, as we said, since its well preserved fins show an even more primitive version of the eight-digit limb pattern seen in Acanthostega (Daeschler and Shubin, 1998).

Reptiles and the Earliest Mammals

From Reptiles to Mammals
The Gulf Between Reptiles and Mammals
Mammal like Reptiles from the Karoo Basin
Mesozoic Mammals

In the modern world, mammals like ourselves are easily distinguished from reptiles by a number of consistent anatomic differences. For example, whereas reptiles have uniform teeth, usually conical and pointed, mammalian teeth are multicuspate, and are differentiated into canines, incisors, and molars. Whereas reptiles replace their teeth throughout their lives, mammals like ourselves only replace our teeth once during a lifetime (ie. "milk teeth"). Whereas the reptilian jaw joint is formed by the conjunction of the articular bone of the jaw with the quadrate bone of the skull, the mammalian jaw joint is formed by the conjunction of the dentary bone of the jaw with the squamosal bone of the skull. Whereas the reptilian ear contains only one bone or ossicle, the stapes, the mammalian ear has three bones, the stapes, the incus and the malleus. Whereas reptiles have the phalangeal formula (number of phalanges in digits 1-5) of 2-3-4-5-4, the mammaliam formula is 2-3-3-3-3. Whereas reptiles have ribs on all of the thoracic and lumbar vertebrae, mammals do not have ribs on the lumbar vertebrae. There are numerous other features that distinguish modern mammals from modern reptiles, but the features we have listed are some of the most obvious, and will help us to identify forms which are genuinely intermediate between reptile and mammal.

Defined by these characteristics, "mammals" first appear in the fossil record during the late Triassic, about 225 million years ago. Again we ask, do mammals appear abruptly and fully formed at their earliest appearance in the fossil record, or is there evidence linking them to some preexisting group of organisms? Carroll notes that the "sequence from the early amniotes to the early mammals is the most fully documented of the major transitions in vertebrate evolution. The entire skeleton was modified, as was the soft anatomy, behavior, and physiology down to the level of cellular metabolism. Many of these changes are demonstrable, either directly or indirectly, through the fossil record" (p 361).

The first late Triassic mammals are linked via numerous anatomic characteristics to the early and middle Triassic cynodont therapsids, known appropriately as the "mammal-like reptiles." The cynodonts are in turn linked via numerous anatomic characteristics to the Permian therocephalian therapsids, which are in turn linked to late Pensylvannian/early Permian sphenacodont pelycosaurs, such as Dimetrodon. The sphenacodonts can in turn be linked to the earliest captorhinomorph reptiles, amphibian-like reptiles which appear in the early Pennsylvanian. While this series is illustrated by numerous successive genera, we will consider only a few of the intermediate forms which illustrate mammalian evolution.

First, we'll consider teeth. Reptilian teeth are generally uniform in size, undiffereniated and lack cusps. Mammalian teeth possess multiple cusps, and are differentiated into molars, incisors, and canines. This reflects the fact that, in mammals, the upper and lower teeth occlude with one another, allowing for the tearing and grinding of food. In the early Permian pelycosaurs such as Dimetredon, the dentition still consists of noncuspate cones, but are differentiated in terms of size as in mammals. In the early cynodonts, such as the late Permian Procynosuchus and Dvinia, teeth start to develop simple cusps. In late cynodonts such as Thrinaxodon, the teeth are not only multicuspate like mammals, but are also functionally differentiated into canines, incisors and cheek teeth.

Next, we'll consider phalanges. In most reptiles, the phalangeal formula is 2-3-4-5-4, whereas the mammaliam formula is 2-3-3-3-3. The early therapsids possessed the reptilian formula. In Thrinaxodon, the phalangeal formula is reduced to 2-3-4-4-3, close to the mammalian count, but still intermediate between the two conditions. The "extra" phalanges in digits 3 and 4 are very small, presaging their later absense. In later cynodonts, the count is reduced to the mammalian formula of 2-3-3-3-3.

Whereas reptiles generally display ribs all the way back to the pelvis, mammals lack ribs in the pelvic region. The gradual loss of these pelvic ribs can be seen in the cynodonts or mammal-like reptiles. Probelesodon, for example, displays ribs all the way back to the pelvis, but the last several ribs are greatly attenuated. In other advanced cynodonts, such as Thrinaxodon or the tritylodonts, the pelvic ribs are absent altogether, as in the mammalian configuration.

Perhaps the most remarkable anatomic change associated with the reptile-mammal transition is the development of a new jaw joint. In all mammals, the lower jaw is composed of only one bone, the dentary, which articulates with the skull at the squamosal bone. In reptiles, the lower jaw is composed of the dentary plus at least 3 additional bones, and the jaw joint is formed by the articulation of the articular bone of the jaw with the quadrate bone of the skull.

Is there any evidence for a "transition" between these two types of jaw configurations? Indeed there is dramatic evidence for such a transition, not the least of which are multiple genera of cynodonts which possessed two jaw joints -- a standard reptilian joint formed by the articular and the quadrate, and a new mammalian joint formed by the dentary and the squamosal. Throughout the therapsid lineage leading up to mammals, the dentary becomes progressively larger, and the post dentary bones become smaller. In Probainognathus, the ascending process of the dentary finally makes contact with the squamosal, forming the typically mammalian jaw joint. Carroll notes:

"In Probainognathus, the surangular and dentary extend back to the squamosal to form a second jaw articulation. As the dentary is further elaborated, we can recognize two functional jaw joints, a medial reptilian joint, which consists of the articular and the quadrate, and a lateral mammalian joint, which is formed by the dentary and squamosal" (p 395).

Not surprisingly, the same double jaw joint is also present in Morganucodont, one of the earliest known mammal, although the mammalian dentary-squamosal joint is now clearly dominant. Although the articular and dentary are now in contact with the stapes, and hence would have served to transmit vibrations, they are little modified from the advanced cynodont configuration. Kemp states that the "secondary, dentary-squamosal jaw hinge had enlarged and took a greater proportion if not all of the stresses at the jaw articulation. The articular-quadrate hinge was free to function soley in sound conduction" (p. 256).

The formation of the mammalian-type jaw and jaw joint is directly related to another remarkable development associated with the reptile-mammal transition, the addition of two new auditory bones or ossicles (malleus and incus) to the mammalian ear. Reptiles have only one ossicle, the stapes. Mammals have three ossicles. According to research by Manley (1972), auditory acuity is no greater in mammals than it is in reptiles or birds, both of which possess only the reptilian stapes. Where did the new ear bones come from, and how are they related to the formation of the mammalian jaw?

The extra bones in each of our ears are the atrophied, vestigial remains of an ancestral reptilian jaw joint. As Romer points out, "the malleus is the old articular; the incus, the reptilian quadrate" (p. 191). As the secondary dentary-squamosal joint became dominant in early mammals, the articular-quadrate bones of the original jaw joint continued to atrophy, just as they had done throughout the Triassic in the cynodont lineage leading up to them.

There are several lines of evidence which converge in support this conclusion. First, the progressive reduction in size and increase of mobility of the postdentary bones is clearly seen in the cynodont fossil record. For instance, Dimetredon, the therocephalians, Thrinaxodon, Probainognathus and Morganucodont show the postdentary bones in progressively more reduced form, and illustrate the step-wise transformation from the reptilian to the mammalian configuration. Second, "the malleus articulates with the incus in exactly the same way as the articular articulates with the quadrate in advanced therapsids and the quadrate (incus) articulates with the stapes" (Carroll, 1988, p 395). Third, the ontogeny of the incus and the malleus reflects or recapitulates their reptilian derivation. When marsupials are still in the pouch, for instance, "the malleus and the incus maintain the reptilian role of the articular and quadrate. Only when the young leave the pouch do these bones seperate from the lower jaw and enter the middle ear" (Carroll, p. 395; see also Crompton and Jenkins, 1979; McGowan, 1984). Yet again, the evidence from the fossil record, comparative anatomy and developmental biology converge in support of the same evolutionary sequence.

Dinosaurs and the Earliest Birds

Armand Hampé’s experiment
Kollar and Fisher’s experiment.
New theropod with feathers
Dinosaurs and Birds- an Update
All About Archaeopteryx
More on Archaeopteryx

Birds are [among other things] feathered, scaled, long-necked, bipedal, egg-laying vertebrates who possess hollow bones, a furcula, digitigrade stance, and a back-turned pubis. Birds do not appear in the fossil record until the end of the Jurassic period, about 150 million years ago. They remain rare until the late Cretaceous period, but diversify rapidly in the Tertiary period.

Until very recently, discussions about bird origins have centered almost exclusively upon one particular genus, Archaeopteryx, the earliest known bird. Archaeopteryx is now known from 7 specimens, the first discovered in 1855, and the most recent in 1992. Although Archaeopteryx is classified as a bird, it's anatomic details are much more similar to an earlier group of scaly, sometimes feathered, long-necked, bipedal, egg-laying vertebrates who possessed hollow bones, a furcula, digitigrade stance, and a back-turned pubis -- the therapod dinosaurs-- than they are to any modern birds. [Note: a fossil specimen of the therapod Oviraptor was recently found not only with eggs, but positioned on top of them in a typically avian brooding position!]

Specifically, Archaeopteryx displays an overwhelming similarity to one particular taxon of therapod dinosaurs, the dromaeosaurs (which includes the famous Cretaceous dinosaur, Velociraptor). The resemblance to dinosaurs is so close, in fact, that several Archaeopteryx specimens were originally misidentified as the contemporary dinosaur Compsognathus (Dingus and Rowe, p 120). This includes the earliest specimen, discovered in 1855, 4 years before Darwin published his theory of evolution. In 1964, John's Ostrom's discovery of the dromaeosaur Deinonychus provided an even closer approximation to the form of Archaeopteryx. Deinonychus, aside from its larger size, is virtually identical to Archaeopteryx.

For example, Archaeopteryx retains a long tail of unfused vertebrae, almost identical to the tail of Deinonychus and other therapod dinosaurs. In modern birds, on the other hand, the tail is absent altogether, and the distal caudal vertebrae are fused into a stubby structure called a pygostyle. Like all therapod dinosaurs, Archaeopteryx had noncuspate, pointed teeth. In modern birds, on the other hand, teeth are absent altogether, and the maxillary and premaxillary bones are instead covered with keratinous beak. Like therapod dinosaurs, Archaeopteryx retains curved raptorial claws on each of its 3 digits. No modern birds retains these 3 raptorial claws, although many birds display temporary embryonic claws (Nedin, 1999), and a few other birds, like the Hoatzin, retain 2 small claws during childhood. Like Deinonychus, Archaeopteryx and other early birds retain 3 unfused digits on the forelimb, whereas in modern birds these three digits are fused together. Like Deinonychus, Archaeopteryx possessed a peculiar, crescent-shaped wristbone called the semilunate carpal, which is formed by the fusion of distal carpals I and II. Only Archaeopteryx and dromaeosaurs like Deinonychus are known to share this bone. In modern birds, by contrast, distal carpal I is absent, and distal carpal II has been fused along with metacarpals I,II, and III into a single bone, the carpometacarpus. Like all dinosaurs, Archaeopteryx had a long neck attached at the back of the skull, whereas in birds it is attached at the bottom of the skull. Continuing on to the pelvis, we see that both Archaeopteryx and Deinonychus display an unfused ischium and ilium, and a "footed" pubis. In modern birds, the ilium and ischium are fused together, and the pubis lacks the "foot." In all of these characteristics and many others, Archaeopteryx clearly and consistently reveals its dinosaurian ancestry.

As we saw with the early tetrapods, many of the features that we tend to associate exclusively with birds had already appeared in nonavian therapod dinosaurs. Recent discoveries from China, for example, show that several species of therapod dinosaurs actually possessed feathers of varying degree of complexity (Protoarchaeopteryx robusta, Sinosauropteryx, Sinornithosaurus, and Caudipteryx zoui, for example. Qiang et al., 1998; Wu, 1999). It is important to point out that none of these feathered non-avian therapods were known when the dinosaur-bird theory was first proposed. The subsequent discovery of simple feathers in small, Archaeopteryx-like theropod dinosaurs was therefore a straightforward and compelling confirmation of that theory.

Headshot of feathered dino from China.

We also know that the furcula or wishbone, long thought to be unique to birds, was present in many mesozoic dinosaurs, including the therapods Oviraptor and Velociraptor (Dingus and Rowe, p 184). As it turns out, several of the most obvious bird-features of Archaeopteryx are not exclusively diagnostic of birds, for they also appeared in non-avian therapod dinosaurs. As the situation now stands, there is no non-arbitrary diagnostic criteria by which birds can be distinguished from their dromaeosaurid ancestors. Carroll (1997) notes that "[a]s the specific relationship of Archaeopteryx to therapod dinosaurs has become better established, only a very few skeletal features associated with flight are known to have changed between the two groups. Many more modifications occurred between Archaeopteryx and more advanced birds . . . in the early Cretaceous" (p, 388). In other words, from a standpoint of comparative anatomy, Archaeopteryx is far more similar to known therapods than it is to modern birds. How then can creationists accept the evidence that Archaeopteryx is related to modern birds, but reject the even more compelling anatomic evidence linking Archaeopteryx to therapod dinosaurs?

While Archaeopteryx is the most "primitive" known bird, by virture of its numerous dinosaurian characteristics, later bird specimens illustrate the gradual loss of these primitive characters, and the gradual accumulation of modern avian characters [note: primitive in this sense means present in the ancestral group, not "simpler" or less "advanced" in any absolute sense].

Sinornis, from the early Cretaceous, illustrates a stage in the evolution of flight anatomy intermediate between Archaeopteryx and modern birds. With Sinornis, for instance, which is about 15 million years younger than Archaeopteryx, we see the earliest appearance of the pygostyle in place of the typically dinosaurian tail, as well as the development of an enlarged sternum and strutlike corocoid, which would have greatly enhanced flight capability. Other derived characteristics, listed by Carroll (1997), include: laterally-facing glenoid socket, modified wrist joint allowing hyperflexion of the manus against the forearm during the recovery phase of the flight stroke, dominance of the second metacarpal, and a lowered hallux, which would have been better suited for perching (p. 318). According to Carroll (1997), "Sinornis differs more from Archaeopteryx than the latter genus differs from the most birdlike dinosaurs." This again highlights the arbitrarity of creationist taxonomy, which accepts the anatomic evidence for the common ancestry of Sinornis and Archaeopteryx, but not the more powerful anatomic evidence for the common ancestry of Archaeopteryx and the birdlike dinosaurs.

Despite these advanced characteristics, Sinornis still retains most of the primitive dinosaurian characteristics seen in Archaeopteryx, including the presence of teeth, clawed manus (hand), a footed pubis bone, and gastralia. Also in common with Archaeopteryx, the ilium and iscium bones of the pelvis of Sinornis remain unfused (and thus do not form the ilio-ischiadic foramen), as do the tarsal bones (no tarsometatarsus) and the bones of the manus (no carpometacarpus).

Leaving the fossils aside, there is striking corroborative evidence from avian embryology which supports the theory that birds are the evolutionary descendents of therapod dinosaurs. During development, for example, the avian hindlimb undergoes a series of transformations that closely recapitulate anatomic changes inferred from the fossil record. Dingus and Rowe, on p 211 of their excellent book, explain:

" . . . a lot of subtle evolutionary changes occurred as the distinctive feet and legs of living birds evolved from more primitive dinosaurs. The ancestral dinosaur had a foot with five toes. Each toe was supported by its own metatarsal bone (the metatarsals are the main girders of the foot), which was connected to the tarsal bones that form the hingelike anle joint. In the theropod lineage, metatarsal I (above our 'big toe') was shortened. The first toe remained functional, but it was no longer connected directly to the ankle bones. At the same time, digit V (our 'little toe') became reduced, losing all its phalanges so that only a thin metatarsal splint remained attached to the ankle. In avialians, digit I rotated to the back of the foot, affording a crudde grasping capability. The remaining large metatarsal bones (II, III, IV) began to fuse to each other and to some of the ankle bones, forming a single compound bones called the tarsometatarsus . . . still later, digit V disappeared altogether, leaving three toes directed forward (II, III, and IV), and one directed backward (I) as in modern birds.

"Comparable changes uccur early during the development or ontogeny in most birds. These changes generally mirror the same sequence documented in evolutionary history, although the transformations occur in embryonic tissues (like cartilage) rather than bone. While still inside the egg, evidence of five toes is visible as the hindlimb starts to grow. The cartilaginous beginnings of five metatarsal bones arise, all in contact with the developing ankle cartilages, in the same configuration found in dinosaurs ancestrally. But soon digit I seperates from the ankle and slides down the side of metatarsal II, later rotating around back to afford a grasping capacity. The growing cartilages for the three remainging metatarsals (II, III, IV) eventually coalesce as they turn to bone, to form the compound tarsometetarsus. The tiny remnant of metatarsal V eventually disappears so that no trace of a fifth digit is seen in adults. In all of these changes, the developing embryo repeats of recapitulates the same changes that occurred during the evolutionary history of its ancestors. In the avian foot, ontogeny recapitulates phylogeny."

A verbal description of these recapitulative changes fails to do them justice, and the reader is advised to consult the comparative diagrams offered on p 210-211. Putatively recapitulative changes are seen in other portions of the hindlimb as well. Therapod dinosaurs, including basal birds such as Archaeopteryx, possessed a full-length fibula, for example, spanning from the ankle to the knee joint, but in modern birds the fibula has shrunken into a small splint near the knee-joint. During early development in birds, there is also a full-length fibula, which only later shrinks into the vestigial splint which characterizes birds.

In one experiment conducted by Hampe, the fibula was induced to retain its full length by the insertion of a piece of mica between the fibula and tibia. Interestingly, this affected the development of the ankle bones as well, such that the two metatarsals that normally fuse to each other and to the fibula and tibia remained seperate also. Thus, as Gould states, "Hampe's simple manipulation not only produced . . . expression of an ancestral relationship in leg bones; it also evoked the ancestral pattern of ankle bones as well" (p 185).

Other "vestigial" patterns of development are seen in different areas of the avian body. For example, we noted that no modern birds have teeth, but that their putative dinosaurian ancestors did. Is their evidence from modern birds of a toothed ancestry? Indeed there is. One group of researchers was actually able to induce the formation of "tooth buds" in embryonic chicks, demonstrating that at least part of the teeth-making genetic program has been retained, even though all modern birds are toothless (Kollar and Fisher, 1980).

Also notable are changes occurring in mandibular structure. In Deinonychus, for example, the lower mandible is composed of 6 seperate bones. In modern birds, some of these bones are either absent or fused with other mandibular bones into a single solid structure. As Chatterjee notes (p 19), all six bones of the dromaeosaurid mandible are recognizable in young birds, but become indistinguishable in adult birds.

Terrestrial Tetrapods and the Earliest Whales

The Pakicetidae
Almost like a whale
When whales walked the earth
Origin of Whales and the Power of Independent Evidence
Creationists and whale evolution
Hans Thewissen's Whale Origins Pages
Gould's article: "Hooking Leviathan by its Past"

Members of the class Cetacea, which includes modern whales, dolphins and porpouises, first appear about 50 million years ago. Cetaceans are distinguished from most other aquatic vertebrates by a number of features common to terrestial mammals. For example, cetaceans have a four-chambered heart, are warm-blooded, breath exclusively through lungs, lack gills, feed their young from mammary glands, have three ear ossicles, and so on. Though lacking hindlimbs altogether, modern whales retain a small, vestigial pelvic bone.

On the basis of comparative anatomy, several paleontologists concluded that the cetaceans are more closely related to artiodactyls such as sheep, pigs and cows than they are to sharks, fish or any other aquatic vertebrates. This conclusion has since been strongly confirmed by genetic data, including the recent discovery that whales share a number of SINE insertions with hippos, cows and sheep that are not present in other mammals (Nature 388:666), and the discovery of shared retroposons in whales and hippos that are not found in any other artiodactyls (Nikaido et al, PNAS 96:10261). In terms of evolutionary history, both anatomic and genetic data have indicated that cetaceans shared a common terrestrial ancestor with hippos and the other artiodactyls.

Since all the evidence suggests that the first mammals were clearly terrestrial animals, and possessed four limbs, their ought to have existed a four legged ancestor of the whales. Until very recently, however, the early cetacean fossil record was poorly known, and a number of creationist authors (P. Davis, D.H. Kenyon, D. Gish, etc) pointed to the lack of forms connecting these aquatic animals with their putatively terrestrial ancestors as evidence that whales did not evolve from terrestrial ancestors at all, but rather appeared "all at once," fully specialized for the aquatic environment. Michael Behe for example thought it seemed like "quite a coincidence that all of the intermediate species that must have existed between Mesonychid and whale, only species that are very similar to the end species have been found" (qtd in Miller, p 315. Note that this was in 1994. Behe has since unambiguously accepted common descent). Actually, this over-representation of "end species" was not a coincidence at all, but an expected consequence of the fact that the greater part of the evolution of whales from terrestrial mammals took only about 5 million years, whereas "essentially modern" whales have existed for about 45 million years.

Yet again, the creationists seem to have spoken too soon. New discoveries from the past decade have closed the terrestrial mammal-whale gap in spectacular fashion. Several very early whale species are now known, and these convincingly illustrate the relationship of early whales to their terrestrial mammalian ancestors. While no modern cetaceans have legs, the earliest whales all apparently possessed legs, losing them gradually over a period of about 10 million years. The earliest known whale-like terrestrial mammal is Pakicetus, which dates to about 52 mya. Although resembling Mesonchynids like Hapalodectes in most respects, Pakicetus also displays some characteristics of later whales, including the elongate jaws and skull crests. Ambulocetus, Rodhocetus, and Basilosaurus, known from deposits dated to 50, 46, and 45 mya respectively, show an ongoing reduction in the hindlimbs, and an accumulation of features typical of whales. By 45 mya, the legs of early whales had all but disappeared. Basilosaurus, a 50ft long whale, retained tiny legs about the size of a small human child's, and these legs were not even attached to the vertebral column. Prozeuglodon, dated to about 40 mya, retained 6inch long legs on its 15ft body.

Although the record is still incomplete, the broad evolutionary history of the whales is nonetheless evident. The earliest whales, unlike their modern counterparts, all possessed hindlimbs and various anatomic features linking them to earlier terrestrial mammals, such as Ichthyolestes. Its hard to understand why an intelligent engineer would create whales with features suggesting a terrestrial ancestry, but if common descent is true, we would fully expect to find this pattern. It is false that all kinds of organisms "appear abruptly and fully formed, with no indication that they had evolved from a common ancestor."

Evolution of the Horse

One of the most complete and compelling examples of intermediate forms occurs in the horse lineage. Unlike the evolution of birds and cetaceans, which is documented by only a few fossil specimens, the evolution of the horse is documented by literally thousands of fossils, spanning the past 53 million years. Though Equus appears in the fossil record a scant 4 mya, it was preceded in time by numerous morphologically intermediate forms. Carroll (1988) writes that the "extensive fossil record of the family Equidae provides an excellent example of long-term, large-scale evolutionary change. Changes in body size, skull proportions, dentition, limb structure, and relative brain size have all been thoroughly documented" (p. 533).

Carroll notes that the ancestry of the Perissodactyl clade, which includes horses, rhinos, tapirs, and the extinct titanotheres and chalicotheres, can be traced "with little question" to the phenacodonts (p. 527). Though Hyracotherium is usually considered the earliest 'horse,' it should be remembered that this animal differed very little from the basal perissodactyls, such as Radinskya, or from the preexisting phenacondonts, such as Phenacodus and Tetraclaenodon. As Simpson notes, "there was no family Equidae when [Hyracotherium] lived . . . the whole process is gradual and we assign the categorical rank [of Equidae] after the result is before us" (1953, p. 345). To a hypothetical paleontologist living in the Eocene, Hyracotherium would have been seen only as a species or variety of Phencodont, not as some new "kind" of animal. The same point holds for the earliest tapirs, rhinos, and titanotheres, which are also connected via fossil intermediate to Hyracotherium-like forms which lived during the early Eocene.

Figure 1. (A) The Eocene horse (Hyracotherium) and representatives of the condylarths, (B) Phenacodus (early Eocene) and (C) Mesonyx (middle Eocene). (From Vertebrate Paleontology by Alfred Sherwood Romer published by The University of Chicago Press, copyright © 1945, 1966 by The University of Chicago. All rights reserved. This material may be used and shared with the fair-use provisions of US copyright law, and it may be archived and redistributed in electronic form, provided that this entire notice, including copyright information, is carried and provided that the University of Chicago Press is notified and no fee is charged for access. Archiving, redistribution, or republication of this text on other terms, in any medium, requires both the consent of the authors and the University of Chicago Press.)

Hyracotherium appears in the fossil record in early Eocene strata, about 53mya. Derived traits linking Hyracotherium to the later Equids include the absence of the foramen ovale (or its confluence with the medial lacerate foramen), and the posteroventral location of the optic foramen (McFadden, 1976). These derived traits are shared with all later Equids, but are not present in non-horse perissodactyls.

All of the Hyracotheres were very small compared to their largest descendants; the smallest was about the size of a housecat (H. sandrae, McFadden, 1992, p. 221), although the average size seems to have been comparable to a medium-sized dog. They possessed 4 digits on the front legs (manus), and 3 on the back (pes). Both back and front legs display reduced splints of digits that were lost (digit 1 on front foot, digit 5 on the rear).

Hyracotherium possessed 4 roughly triangular premolars, and three roughly square or quadrate molars. These teeth occude with one another, such that the cusps would have interlocked when the jaws closed. In the lineage leading from Hyracotherium to later Equids, several gradual changes are evident in the dentition, changing from a dentition suitable to browsing to one highly effective for grazing. The premolars are changed into molars, a series of ridges form on the cheeck teeth, theses ridges form increasingly higher crowns. Though each stage of the change is morphologically small, the sum of those changes are considerable. Radius and ulna are in the 'primitive,' unfused state; fibula is full-length; orbit is about midway along the length of the skull, and lacks a postorbital bar.

Orohippus, which dates to approximately 50 mya, is much like Hyracotherium, although a few subtle changes are evident. For example, although Orohippus retains the digit configuration of Hyracotherium, the vestigial splints of digits 1 and 5 are now absent. Also notable is a change in Orohippus' last premolar (P3 and p3), which develops low crests and has been reshaped into a more molariform shape, giving Orohippus an 'extra' molar. Epihippus, which appears about 47 mya, continues this trend of molarization of the premolars, with the second premolars, P2 and p2, also assuming a molariform shape.

Mesohippus appears in the late Eocene, about 40 mya. Mesohippus is larger, about 2 ft tall at the shoulders. Both front and back legs now display 3 prominent digits (is tridactyl), with digit 4 on the front limbs now reduced to a vestigial splint. The ulna, although reduced compared to Hyracotherium, is still distinct and unfused throughout its length. All of the premolars, P1-P3 and p1-p3, have now assumed a molariform shape, but still possess relatively small crests. Miohippus, which appears about 36 mya, is essentially identical to Mesohippus, except that a new crest appears in the upper premolars. This new feature is present in all subsequent species leading to Equus.

With Parahippus and Merychippus, which date to about 23 and 17 mya respectively, further anatomical changes are apparent. Body size has increased, with Merychippus about 40 inches tall at the shoulders. The orbit is further towards the rear of the skull, and the postorbital bar is complete. Some of the most obvious and interesting changes occur in the dentition. The dental crests are now joined together to form high ridges. These ridges would have allowed for efficient, side-to-side mastication of fibrous grasses, leaves and plants. These changes in dentition occur at about the same time that grasslands became widespread in North America. Some horses, Merychippus probably included, adapted to these changes by developing high-crowned teeth suited for grazing on the abrasive, silica-rich grasses, while other horses remained browsers.

Also notable in Merychippus are changes in leg anatomy. Whereas in previous Equids the radius and ulna are unfused throughout their length (McFadden, p. 248), and the fibula is full-length (both of these are part of the ancestral eutherian condition), in Merychippus the radius has now fused with the ulna, and the fibula is reduced. This and other changes suggest that Merychippus was probably the first truly adept runner of the Equidae. Digit size varies in Merychippus, with some species showing a reduction of the two side digits.

Figure 2. Stages in horse evolution showing the reduction in the number of toes and foot bones. Forefeet above, hind feet below. (A) Hyracotherium, a primitive early Eocene horse with four toes in front and three behind, (B) Miohippus, an Oligocene three-toed horse, (C) Merychippus, a late Miocene form with reduced lateral toes, and (D) Equus. (From Vertebrate Paleontology by Alfred Sherwood Romer published by The University of Chicago Press, copyright © 1945, 1966 by The University of Chicago. All rights reserved. This material may be used and shared with the fair-use provisions of US copyright law, and it may be archived and redistributed in electronic form, provided that this entire notice, including copyright information, is carried and provided that the University of Chicago Press is notified and no fee is charged for access. Archiving, redistribution, or republication of this text on other terms, in any medium, requires both the consent of the authors and the University of Chicago Press.)

Pliohippus, at its earliest occurence about 15 mya (Miocene), still possessed three digits. Hunt notes that "Gradual loss of the side toes is seen in Pliohippus through 3 successive strata of the early Pliocene" (1995). Except for curved teeth and deep facial fossae, Pliohippus is very much like Equus. Dinohippus, first appearing about 12 mya, is very close to Equus in terms of anatomy, but retains some ancestral characteristics. Dinohippus is monodactyl, like Equus, has reduced facial fossae, intermediate between Merychippus and Equus, and has only slightly curved teeth, also intermediate between Merychippus and Equus. Changes in these seatures can be traced though several successive species of the genus Dinohippus. According to Hunt, summarizing Hulbert, 1989, "Dinohippus showed a gradual decrease in the facial fossae, straightening of the teeth, and other gradual changes, as Dinohippus smoothly graded into Equus."

Fossil Horses: A Bush, Not a Ladder

The foregoing summary focuses only on the ancestors of the extant genus Equus, and therefore gives the impression that all fossil horses evolved in a straight line towards Equus. But in fact, this particular lineage represents only a portion of horse evolution. In other lineages, there were often different changes than those occurring in the Equid line. Colbert notes:

"The horses are often cited as an outstanding example of 'straight-line' evolution or 'orthogenesis,' and it is frequently maintained that these animals evolved with little deviation along a straight path from the little Eocene Hyracotherium . . . to the modern horse, Equus. It is true that most of the progressive changes listed above can be followed through time from Hyracotherium to the modern horses, but in middle and late Tertiary times there were various lateral branches of horses that were progressive in some features and conservative in others. When all fossils are taken into account the history of horses in North America is seen to be anything but a simple progression along a single line of development" (1980, p. 379).

Similar statements can be found in paleontology books by many authors. Out of either misunderstanding or deception, creationists have frequently quoted such statements as evidence which somehow refutes horse evolution. The proposition that horses evolved at all is not a matter of contention in Colbert's statement, nor in similar statements quoted by creationists.

The only view of horse evolution that has been discarded is the outdated, orthogenetic view that all fossil horses are part of a nonbranching progression from Hyracotherium towards Equus. In reality, several of the taxa considered ancestral to Equus (Hyracotherium, Mesohippus, Merychippus, and others) also gave rise to other lineages as well, many of which were short lived, and all of which eventually became extinct. In this example, as in many others, the fossil record has shown that evolutionary descent often produces a branching bush of lineages, not a straight line. We trace the fossil record from Hyracotherium to Equus not because horse evolution led inexorably to Equus, but because Equus is the only surviving member of the Equidae.

For more online information on the evolution of the horse, see Kathleen Hunt's summary, Kenneth Miller's essay on Taxonomy, Transitional Forms, and the Fossil Record, and the Equine Studies Institute's web pages on the subject.

Conclusion

The fossil record is not perfect, nor should it be, given what we know about the processes of fossil preservation. But the fossil record is more than sufficiently detailed to demonstrate the reality of descent with modification. Negatively, the fossil record is also more than sufficient to refute the notion that all "kinds" of living things appear simultaneously and are "fully formed" at their earliest occurence. As we've seen in these examples, the earliest members of different animal taxa -- amphibians, reptiles, mammals, birds, whales and horses, for example -- are consistently more similar to some preexisting animal group than they are to modern members of their own group. In many cases, the fossil record offers a whole series of temporally and morphologically sorted fossils illustrating major transitions in vertebrate evolution. As each decade passes, new fossil specimens are discovered and the evolutionary pattern of the fossil record becomes clearer still. Ken Miller notes:

"Time after time, species after species, the greater our knowledge of earth's natural history, the greater the number of examples in which the appearance of a new species can be linked directly to a similar species preceding it in time. These histories reveal a pattern of change, a pattern that Darwin aptly called . . . 'descent with modification' " (Finding Darwin's God, p. 40).

Works Cited

Carroll, Robert L. (1988). Vertebrate Paleontology and Evolution. W.H. Freeman.

------------------ (1997). Patterns and Processes of Vertebrate Evolution.

Chatterjee. Sankar. (1997). The Rise of Birds. Johns Hopkins University Press, Baltimore.

Crompton, A.W., and Jenkins, F.A. (1979). Origin of Mammals. In J.A. Lillegraven, Z. Kielan-Jaworowska, and W.A. Clemens (eds.), Mesozoic Mammals: The First Two-Thirds of Mammalian History, pp. 59-73. University of California Press, Berkeley.

Daeschler, Edward B., and Neil Shubin. (1998). Fish with Fingers? Nature, 391:133.

Dingus, Lowell and Rowe, Timothy. (1998). The Mistaken Extinction: Dionsaur Evolution and the Origin of Birds. W.H. Freeman and Co, New York.

Gould, Stephen J. (1983). Hen's Teeth and Horse's Toes. Norton, New York.

Hunt, Kathleen. (1999). Transitional Vertebrate Fossils FAQ. Available online at Talkorigins.org.

Kollar, E.J., and Fisher, C. (1980). Tooth Induction in Chick Epithelial: Expression of Quiescent Genes for Enamal Synthesis. Science 207: 993-995.

Manley, G.A. (1972). A review of some current concepts of the functional evolution of the ear in terrestrial vertebrates. Evolution, 26:608-621.

McGowan, Josh. (1984). In the Beginning. Prometheus books, New York.

Miller, Kenneth. (1999). Finding Darwin's God. Harper Collins, New York.

Muller, G. B. (1989). Ancestral Patterns in bird limb development: A new look at Hampe's experiment. Journal of Evolutionary Biology 1:31-47.

Nedin, Chris. (1999). All About Archaeopteryx. Available online at Talkorigins.org.

Shipman, Pat. (1998). Taking Wing: Archaeopteryx and the Evolution of Bird Flight. Simon and Schuster, New York.