Ambulocetus as a Fossil Transitional

by Lenny Flank

© 1995 Lenny Flank

One of the most commonly-heard creationist arguments is the supposed lack of transitionals in the fossil record. As Duane Gish puts it, "Ever since Darwin the fossil record has been an embarrassment to evolutionists. The predictions concerning what evolutionists expected to find in the fossil record have failed miserably. Not only have they failed to find the many tens of thousands of undoubted transitional forms that are demanded by evolutionary theory, but the number of arguable, let alone demonstrable, transitional forms that have been suggested are few indeed. This has placed evolutionists in a most difficult situation, made even more embarrassing by the fact that the fossil record is remarkably in accord with predictions based on special creation.": (Gish, ICR Impact, September 1983) ":This situation is strikingly true," Gish concludes, ":concerning the origin of whales, dolphins and other marine mammals." (Gish, ICR Impact, April 1994)

In reality, the evolutionary history of the whales has become increasingly more clear, thanks to several fossil discoveries made within the past few years. The evidence of these fossils lends strong support to the theory of descent with modification, and is totally at variance with the so-called "creation model."

Until fairly recently, the fossil history of the earliest whales (known as Cetaceans) was quite unknown. Edwin Colbert pointed out in 1955, "These mammals must have had an ancient origin, for no intermediate forms are apparent in the fossil record between the whales and the ancestral Cretaceous placentals. Like the bats, the whales (using the term in a general and inclusive sense) appear suddenly in early Tertiary times, fully adapted by profound modifications of the basic mammalian structure for a highly specialized mode of life." (Colbert, 1955, p. 303) The oldest whales then known, the Archaeocetes, already exhibited all of the typical whale characteristics, including lack of rear limbs, paddle-like front limbs, and a tail with a horizontal fluke for propulsion. The teeth of the Archaeocetes, however, very closely resembled an ancient group of carnivores called Mesonychids, which were wolf-sized scavengers that lived in the early Eocene period. Based on these similarities, most paleontologists hypothesized that the whales were the evolutionary descendents of the terrestrial Mesonychid carnivores.

The first hint that they were probably right came in 1983, when researcher Phil Gingerich found a 52-million year old skull in shallow deposits in Pakistan. Although fragmentary, the skull had teeth that were nearly identical with those of Mesonychids and the Archaeocetes. The configuration of the bones at the rear of the skull, however, were different from those in the Mesonychids, and were identical to that of the Archaeocetes. Gingerich thus concluded that the animal, which he named Pakicetus, was a very primitive whale. "In time and in its morphology," Gingerich reported, "Pakicetus is perfectly intermediate, a missing link between earlier land mammals and later, full-fledged whales." (Gingerich, The Whales of Tethys, Natural History, April 1994, p. 86)

Despite the whale-like characteristics of the skull, however, Pakicetus lacked two important adaptations which are present in modern whales. In living whales, the ears contain large sinuses that can be filled with blood, allowing the animal to maintain pressure while diving. Modern whales also transmit sound vibrations to the inner ear using a "fat pad," which allows them to hear directionally underwater. Pakicetus lacked both of these features, indicating that it was unable to dive deeply and that it could not hear well underwater. These anatomical clues meshed well with its habitat, since the Pakicetus bones were found in deposits that had been laid down at the mouth of a river on the shore of a shallow sea, where the opportunities for deep diving would be limited. Although no post-cranial bones of Pakicetus were found, it seemed logical to assume, from the teeth and ear structure, that the animal spent a great deal of time in shallow water looking for food, but returned to the land to rest, somewhat like a modern sea lion. It was expected, therefore, that the limb bones of the earliest Archaeocetes would be large and functional, unlike the atrophied remnants of limb bones found in modern whales. (This hypothesis was supported by the fact that several species of modern whales have well-developed rear limbs while embryos, which later atrophy and become nonfunctional).

This conclusion was bolstered in 1990, when Gingerich found another fossil in the Zeuglodon Valley in Egypt. This was a specimen of Basilosaurus, an animal that had originally been uncovered back in 1834, but misidentified as a marine reptile. It was identified as a whale in 1840. Like the earlier specimen, Gingerich's new Zeuglodon specimen included a four-foot long skull and a number of ribs, all of which bore the unmistakable characteristics of whales, and which matched the corresponding points in the old Pakicetus skull. More importantly, Gingerich also found a femur, a kneecap, both lower leg bones, a number of ankle bones, and three toes, showing beyond any doubt that Basilosaurus had complete rear limbs.

Although the limbs were fully-formed and functional, however, they did not connect to the whale's pelvic bones, and could not have been used to support the animal's weight out of the water. Basilosaurus, therefore, was a whale with a prominent rear leg, but one that was so adapted to a marine existence that it could no longer leave the water. As Stephen Gould puts it, "They are anatomically complete, and they did project from the body wall (unlike the truly vestigial hind limbs of modern whales), but they could not have made any important contribution to locomotion--the real functional test of intermediacy.

(Gould, Hooking Leviathan By Its Past, Natural History, May 1994, p. 13). This is exactly what paleontologists had predicted they would find in ancient whales.

Gingerich also pointed out, however, that Basilosaurus had a number of unique features in the spinal column which are not shared by modern whales, leading him to believe that, while it was a whale and a descendent of a Pakicetus-like animal, Basilosaurus died out without leaving any modern descendents.

However, the paleontologists predictions concerning a limbed whale-ancestor were further confirmed in 1993, when Gingerich found two new specimens of Indocetus, an early Archaeocete which was intermediate in age between that of Pakicetus and Basilosaurus. The new Indocetus specimens included a piece of pelvic bone, tibia and femur, but no foot bones. Although the find was fragmentary, it was evident from the size and structure of the bones that Indocetus possessed fully-formed functional rear limbs. "The pelvis has a large and deep acetabulum, the proximal femur is robust, the tibia is long . . . All these features, taken together, indicate that Indocetus was probably able to support its weight on land, and it was almost certainly amphibious." (Cited in Gould, Natural History, May 1994, p. 13) Indicetus had not yet, however, developed the loosely constructed sacral bones that allow the modern whales to swim by flexing their spines up and down--a mode of locomotion which is unique among marine animals.

In the same month as Gingerich's Indocetus discovery, however, Michigan University grad student Xiaoyuan Zhou found an Archaeocete skeleton in Pakistan that was slightly younger than Indocetus. This new fossil, which has not yet been named, possessed a proportionately smaller femur, small fused neck bones and also loose sacral bones, indicating that it had achieved the full up-and-down streamlined swimming motion exhibited by modern whales. Zhou's skeleton was found in deeper marine deposits than any of the older Archaeocetes.

The most recent discovery in cetacean evolution has also been the most spectacular. In January 1994, Hans Thewissen announced the discovery of several 49 million year old Archaeocete skeletons, the most complete one consisting of parts of the skull and jaw, a number of vertebrae, some ribs and nearly complete front and hind limbs. The large limb bones were fully capable of supporting the animal's weight on land, and were also capable of paddling it through the water using an up-and-down motion of the spine (although it lacked the loose sacral bones found in the Zhou skeleton). Thewissen named the animal Ambulocetus natans ("the swimming whale that walks"). In morphology and in timing, it is a perfect intermediary between the Mesochynids and the younger Archaeocetes.

Ambulocetus was about the size of an adult sea lion, and weighed something around 650 pounds. "Ambulocetus is clearly a cetacean", Thewissen concluded. In addition to the same Mesonychid type of teeth possessed by all of the Archaeocetes and a small hoof at the end of each toe like the Mesonychids, Ambulocetus also had the whalelike skull characteristics that are found in the Archaeocetes, including an ectotympanic with a large sygmoid process, a reduced zygomatic arch, a wide supraorbital process and a narrow muzzle. While these characteristics may also be present in the terrestrial Mesonychids, Ambulocetus also possessed the small protocones and large accessory cusps which distinguish the whales from the Mesochynids.

The limbs were large and strong, and were capable of carrying the body weight on land. "The skeleton of Ambulocetus indicates that it could locomote on land and in the water. As in extant cetaceans, Ambulocetus swam by means of dorsoventral undulations of its vertebral column, as evidenced by the shape of the lumbar vertebra. Unlike modern cetaceans, however, Ambulocetus had a long tail and thus probably lacked a tail fluke." (Thewissen, Science, January 1994, p. 211). Thus, the skeleton of Ambulocetus demonstrates a mixture of characteristics from the terrestrial Mesonychid carnivores and those of fully aquatic whales. "As such," Thewissen concludes, "Ambulocetus represents a critical intermediate between land mamals and marine cetaceans." (Thewissen, Science, January 1994, p. 212)

Because of this recent series of discoveries, we now know much more about the process of cetacean evolution than we did twenty years ago. The earliest known cetacean, Pakicetus, demonstrates a mixture of traits which are unique to the terrestrial Mesonychids as well as marine whales, and indicates that the cetaceans are descended from the Mesonychid carnivores. Although we have not found any post-cranial bones from Pakicetus yet, those which have been found with Ambulocetus demonstrate that the earliest members of the Mesonychid--whale transition were animals which spent large amounts of time on land and in the sea, and presumably had lifestyles somewhat similar to those of modern sea lions.

After Ambulocetus, there is a trend towards increasing specialization for a totally marine existence. The Indocetus fossils show a reduction in the size of the limbs, while the Basilosaurus skeleton shows a reduction in the size of the hind limbs to the point where the animal could no longer leave the water. The unnamed limbless Zhou fossil shows the development of a streamlined body and the loosely-constructed sacral bones found in modern whales, which allowed them to swim more efficiently and move into the deep sea. Finally, the later Archaeocetes show the rear limbs shrinking to near non-existence, along with the development of such cetacean specializations as the pressurized inner ear and the "fat pad." Taken as a whole, the Pakicetus-Ambulocetus-Indocetus series is a perfect vindication of the predictions made by paleontologists fifty years ago.

The creationists, of course, cannot accept such a transition, and have made efforts to discredit both Pakicetus and Ambulocetus. The chief critic has been Duane Gish, the creationists' "expert" on the fossil record.

Gish's first criticism of Pakicetus centers around its habitat. After noting that whales are marine creatures (there are no modern freshwater whales in existence), Gish goes on to point out: "The fossil remains associated with Pakicetus is dominated by land mammals. Nonmammalian remains include other terrestrial remains such as snails, fishes (particularly catfishes), turtles and crocodiles. This evidence indicates a fluvial and continental rather than a marine environment, as would be expected for a whale or whale-like creature." (Gish, ICR Impact, September 1983).

Leaving aside the question of how deposits that contain "snails, fishes (particularly catfishes), turtles and crocodiles" can be considered "terrestrial" and "continental" (all of these animals are largely aquatic), we can here point out that Gish is engaging in the typical creationist tactic of creating a strawman and then attacking evolutionary theory because it doesn't live up to his strawman. The argument that any cetacean ancestor "would be expected" to be found in a marine deposit is wholly Gish's own construction. There is no reason whatsoever to expect that the earliest ancestors of marine cetaceans must of necessity have been marine creatures, and plenty of reasons to suppose otherwise. Since the earliest whale-like fossils we know of do not exhibit any of the specialized adaptations possessed by modern whales for a life in the deep sea--vascularized ears to maintain pressure while diving, directional underwater hearing, horizontal flukes for efficient swimming--there is no reason to suppose that the ancestors of whales must have lived in the deep sea while they were developing these adaptations. Apparently Gish would have us believe that it is the opinion of paleontologists that a wolf-like animal decided to jump in the deep ocean one day and live there. It is more probable that the terrestrial ancestors of the whales lived in the mouths of rivers near shallow seashores, where food would be abundant and could be obtained without the necessity for any deep-sea adaptations, and where increasing specialization for an aquatic existence could be done gradually, in stages, allowing the animals the opportunity to return to the land when necessary. And this is in fact precisely where we find the remains of such animals. As the fossil record shows, the adaptations for deep-sea life came later, after the terrestrial ancestors had already become fully aquatic.

The clear implication of his criticism is that Pakicetus really isn't a whale at all--it is merely another of the known Mesonychids, who were ordinary terrestrial carnivores. Of Pakicetus, Gish says, "One should be immediately suspicious of the term 'whale' being given to such a creature, whatever it is, since whales are totally incapable of living or breeding on land." (Gish, ICR Impact, September 1993). Pakicetus, he concludes, is "a land mammal, with no relationship to marine mammals". (Gish, ICR Impact, April 1994) And of Ambulocetus, he echoes, "Since whales don't walk on land, skeptics would immediately question the basis for designating this creature a whale, whatever it may have been. . . . It certainly was not an intermediate between a land mammal and a whale, but was more likely a near-shore carnivore." (Gish, ICR Impact, April 1994)

Gish is here playing a word game, similar to the one he uses in his "Archaeopteryx is just a bird" arguments. The system of classification which biologists use, which places all organisms into categories consisting of kingdoms, classes, phyla, orders, families, genera and species, makes it difficult to acknowledge the transitional chacter of many species which possess characteristics of two quite different categories, since every organism must be placed in one pre-labelled category or another, and no categories exist for something that is "half-this and half-that." As L. Beverley Halstead puts it: "With many groups we have what we term mosaic evolution; beginning with animals that are completely reptilian one ends up with forms that are completely avian or mammalian. The whole picture gradually changes over as more birdlike or mammalian features develop. What we have to do, because the transition is so gradual, is draw an arbitrary line: if it has character X we will call it A, if not we will call it B. Hence by definition there can never be an intermediate, because we have drawn arbitrary lines in such a way that an animal is forced to be either one thing or another." (Montagu, 1984, p. 253)

Gish has taken it upon himself to assume that whales (even ancient whales) do not have legs, and therefore anything that has legs must be called something other than a whale, no matter how many other whale-like characteristics it may have. In reality, it matters not at all whether we classify Ambulocetus or Pakicetus as a Mesonychid or an Archaeocete, since they are part of a smooth series which grades from one group to the other. One could just as accurately depict Ambulocetus as an advanced Mesonychid with many Cetacean features or as a primitive Cetacean with many Mesonychid features (just as one could equally well class Archaeopteryx as a dinosaur with birdlike features or a bird with dinosaur-like features--neither classification removes its transitional character).

In terms of their skeletal structure, however, whales are distinguished from the similar Mesonychids using five basic anatomical characteristics: (1) all of the incisors are parallel with the tooth row, (2) the medial lamboidal crest is semicircular, (3) the nasals are retracted, (4) the protocones are small, and (5) the accessory cusps are large. Pakicetus, Ambulocetus and Indocetus all possess one or more of these characteristics, and are thus classified as "Archaeocetes" (whales) rather than Mesonychids. (The Ambulocetus skeletons are lacking the relevant portions of the skull to test the first three of these characteristics, and thus Thewissen also added several other skull characteristics in concluding that Ambulocetus was an Archaeocete rather than a Mesonychid, prompting charges from creationists that he "changed the definition of what is a whale.")

Gish's argument that "if it has legs, it can't be a whale" is simply irrelevant, since the skeletal structure of the limbs is not a part of the accepted definition of what constitutes a fossil whale. Basilosaurus also had legs, but they weren't functional--Gish could scarcely argue that Basilosaurus was some kind of terrestrial Mesonychid and "not a whale." Indeed, several species of modern whales have developed rear limbs as embryos; would Gish cite this as proof that they are "not really whales"? Gish also ignores all of the anatomical characteristics which Pakicetus, Ambulocetus and Indocetus share with modern whales as well as the terrestrial Mesonychids.

Taken as a whole, the Archaeocete series from Pakicetus to Indocetus is very convincing evidence of descent with modification. Beginning with terrestrial Mesonychids, we can trace the path through Ambulocetus, which was a terrestrial animal that spent much time in the water, to Basilosaurus, which had nearly lost its functional legs, to the later Archaeocetes, which possessed no external legs at all and were specialized for a deep-sea life, to the modern whales. As Stephen Gould concludes, "If you had given me a blank piece of paper and a blank check, I could not have drawn you a theoretical intermediate any better or more convincing than Ambulocetus. Those dogmatists who by verbal trickery can make white black, and black white, will never be convinced of anything, but Ambulocetus is the very animal that they proclaimed impossible in theory." (Gould, Natural History, May 1994, p. 14)