A Different Kind of Whale

Three restorations (top, left side, and bottom) of the skull of Andrewsiphius. From the Journal of Paleontology paper.

During the past 30 years the evolution of fully aquatic whales from terrestrial ancestors has gone from one of the most enigmatic evolutionary transitions to one of the best documented. Evidence from the fossil record, genetics, and embryology have been combined to document how early whales walked into the sea, but what often has gone unnoticed is the diversity of early whales. In a new paper published in the latest issue of The Journal of Paleontology, cetacean experts J.G.M. Thewissen and Sunil Bajpai describe new fossils from two very strange creatures that represent a unique part of early whale evolution.

In many popular treatments on whale evolution the emphasis is on the transformation of hooved terrestrial mammals, akin to Indohyus, into something as monstrous as a modern blue whale. The evolution of aquatic characteristics is traced in almost a straight line, yet our present understanding of whale evolution cannot be crammed into a unilinear progression of types. As the metaphor suggests the whale family tree has a branching pattern, and some of these branches terminated without leaving any living descendants.

This picture of early whale diversity has emerged thanks to recent discoveries that have made the detailed comparison of early whales possible. Andrewsiphius, one of the archaeocetes (‘archaic whales’) described in the paper, was first described in 1975 as a fully aquatic toothed whale. This identification was revised in 1998 after more material and more early whales became known. It was not fully aquatic, as initially thought, but showed some similarity to another group of recently-described archaeocetes called the remingtonocetids. This revision was followed two years later by the description of a similar animal that was dubbed Kutchicetus.

A restoration of the skull of Remingtonocetus. From “Whale Origins as a Poster Child For Macroevolution” in BioScience.

Unfortunately the first fossil material used to describe Andrewsiphius and Kutchicetus was fragmentary and of relatively poor quality. The bones were unique enough to establish the genera but many parts of their skeletons were still missing. In the past nine years, however, some of these missing bones have been found in the 42-46 million year old rock along India’s western border. When put together, these additional fossils provide a more complete look at these odd creatures.

When compared to each other, Andrewsiphius and Kutchicetus shared more similarities with each other than they do with other archaeocetes. As such the authors place them in a new group, the Andrewsiphiinae, which was closely related to the remingtonocetids but differed in some interesting ways. Both Andrewsiphius and Kutchicetus had very long snouts that were wider from top-to-bottom than from side-to-side, and the presence of foramina (small holes in the bone) near the tip suggests that they may have had whiskers. They also had eyes that were placed very high up on the skull towards the midline rather than on the side. This gave them a crocodile-like profile, and this would have allowed them to keep on eye on the shore while still submerged in the shallows.

Andrewsiphius and Kutchicetus also have very large sagittal crests on the back of their skulls. These bony ridges were areas of muscle attachment for the jaws, and in Andrewsiphius, at least, the structure was so large that it overhung the back of the skull. Why Andrewsiphius and Kutchicetus had so much space for jaw muscles is still unknown, though, and the authors suggest they might have fed on prey in a unique way that has not previously been documented.

A reconstruction of Maiacetus, from PLoS One.

Kutchicetus also differed from earlier archaeocetes like Ambulocetus in having flattened vertebrae that would have supported an otter-like tail. In fact, otters generally seem to provide a fair analog for what the bodies of Andrewsiphius and Kutchicetus looked like; just attach a really weird, long-snouted head on the front. Indeed, Andrewsiphius and Kutchicetus had hip vertebrae that were still fused together and connected tightly to the hip bones meaning that they still supported themselves on land. This differs from the arrangement seen in other kinds of contemporary archaeocetes like the protocetids. Exemplified by forms such as Rodhocetus and Maiacetus, the protocetids were more aquatic and their hip vertebrae were becoming unfused from each other and the hip bones for better propulsion in the water. The protocetids undulated their tails and bodies up and down to swim, and these changes suggest that the protocetids were spending most of their time swimming in shallow, near-shore ocean habitats.

An early sketch of the form of Kutchicetus. As described in the paper, we now know that it had a long snout like Andrewsiphius. From “Whale Origins as a Poster Child For Macroevolution” in BioScience.

These anatomical differences are important to take into account as during the middle of the Eocene, the time during which Andrewsiphius and Kutchicetus lived, archaeocetes were spread through a variety of environments. The protocetids swam in the open water of the ocean near the shore, remingtonocetids were abundant throughout nearshore environments, and Andrewsiphius and Kutchicetus lived in the marshy areas further from the waves. It is not as if early whales began to become adapted to the water and they all began to transform into creatures like living whales. Instead there was an adaptive radiation in which a range of near-shore environments, from swampy estuaries to the shallow parts of the sea, were inhabited by different forms of early whales.

A cladogram of early whale relationships. Andrewsiphius and Kutchicetus are circled in red. From the Journal of Paleontology paper.

I find that this sort of perspective is often missing from discussions about transitional fossils. We focus on the traits possessed by different organisms at different times and track the changes, but it is rare that attention is paid to the evolutionary pattern in which these changes are nested. We all abhor the use of the phrase “missing link” yet we still often portray evolving lineages as a “chain” running from a starting point to the “finish line.” We know this is not true, but vestiges of these discarded ideas still persist here and there, and I hope that more writers will use the radiation of early whales as an antidote to classic “March of Progress” storytelling.

Thewissen, J., & Bajpai, S. (2009). New Skeletal Material of Andrewsiphius and Kutchicetus, Two Eocene Cetaceans from India Journal of Paleontology, 83 (5), 635-663 DOI: 10.1666/08-045.1