The Evolution of Whales

Whales are known to be descendants of land mammals via several physical characteristics: bones in their flippers which resemble the forelimbs of land mammals, vertical (up-and-down) movement of their spines which resembles a running terrestrial animal, and the fact that they must breathe air (presumably an animal evolving solely in an aquatic environment would take the easier energetic route of adopting gills from earlier ancestors). Many ancient and modern whales carry evolutionary "baggage" marking their ancestry, such as reduced (vestigial) hind legs. Some whales still make use of their pelvic bones for reproductive purposes. The cetaceans' associations with other mammals can also be confirmed by genetic evidence.

But the question remains how whales made the transition back to the water from land, and when.

A 1959 woodcut of Basilosaurus, an early whale originally classified as a reptile (Basilosaurus means "King Lizard")

In 1693, John Ray penned his realization that whales were mammals, not fish. William Henry Flower recognized that whales have vestigial features characteristic to land mammals in 1883, confirming that whales evolved from terrestrial animals rather than being the ancestors of today's land mammals. Flower was also the first person to link whales specifically to ungulates, or hoofed mammals.

Though we now have access to a wide variety of genetic evidence for the ancestry of whales, all past hypotheses on their origins were based on physical characteristics, also known as morphology. Scientists can actually tell a great deal about an animal simply from bones, including details of muscle placement and maturity. Because morphology studies are so advanced in the present day, some of the evidence for whale ancestry still comes from studying the physical features of fossil whales.

One of the creodonts, extinct carnivores once proposed as the ancestors of whales

But scientists initially came up with a wide variety of hypotheses as to what animals were ancestral to whales. In 1859 Charles Darwin suggested that whales were descended from bears (after a furor of scientific criticism he withdrew the point from later editions of The Origin of the Species). In the early 1900s Eberhard Fraas and Charles Andrews hypothesized that whales were descendants of extinct carnivores called creodonts. A little later on, W.D. Matthew postulated that whales descended from insectivores (insect-eating animals), and even later E.J. Slijper attempted to combine both ideas, saying that whales were descendants of credonts-turned-insectivores. No animals matching this imaginative description were ever found.

The prevailing hypothesis on whale evolution until the 1990s was founded in 1966 by Leigh Van Valen, and also independently by F.S. Szalay in 1969. Both scientists noticed that the characteristics of whale teeth closely matched those of an extinct group of carnivorous ungulates called mesonychids. Specifically the earliest whale was linked to a wolf-like animal from the genus Sinonyx. In addition to matching teeth, Sinonyx had an elongated muzzle and other characteristics similar to whales in the shape of its skull.

Wadi Al-Hitan (Whale Valley), situated in the former Tethys Sea, was made a UNESCO World Heritage site due to the many recent whale fossil discoveries made there

In the 1990s additional morphological evidence was added to the understanding of whale origins, pulling away from Van Valen's mesonychid hypothesis and toward a group of mammals called artiodactyls. Artiodactyla (even-toed), as the order is called, contains a huge variety of mammals: camels, cows, deer, pigs, antelopes, giraffes, hippos, and so on. In 2000, Philip D. Gingerich discovered distinct ankle bones in fossil whales that gave them a direct morphological link to artiodactyls. Gingerich has been working in the field for many years, and his contributions also include the 1981 discovery of Pakicetus inachus, the oldest distinctly cetacean animal.

Gingerich's ankle bone discovery set up a new artiodactyl family tree: a proto-ungulate (that is, an early hoofed mammal) gave rise to both Artiodactyla and Cetacea, with hippopotami as the closest to the cetaceans (i.e. the least diverged).

Hippos are thought to be the closest living relatives of whales

However, it is important to note that in 1985, Vincent Sanrich of UC-Berkeley had already noticed a link between cetaceans and hippopotami via similarities in blood proteins. In 1997, based on morphological evidence and a collection of 20 or so DNA studies, it was suggested that Cetacea and Artiodactyla be combined into a superorder called Cetartiodactyla to reflect the new connections between the two. In 2005 the link between hippos and cetaceans specifically was confirmed by molecular phylogeny (the study of the shapes of molecules-- specifically RNA, DNA, and proteins --to determine relatedness), lending more support to the Cetartiodactyla hypothesis.

Via these analyses, hippos and whales can be linked to a common water-loving ancestor that lived 50 to 60 million years ago. One population of this ancestral species returned to the water to become cetaceans, while another population became a group of pig-like animals called anthracotheres, of which hippos are the only descendants.

Evolutionary chart showing the relationships of cetaceans, hippos, and artiodactyls

Most recently in November 2007, Hans Thewissen discovered the tiny Indohyus, a 48-million-year-old artiodactyl. Indohyus is significant because like Pakicetus, it can be linked to whales via morphological characteristics in the skull that are unique to cetaceans and it is herbivorous. However, Indohyus is likely to be an outgrouping from whales rather than an ancestor. Check the Indohyus module for more information. Funding for Dr. Thewissen's study came from the NSF (National Science Foundation).

New discoveries are being made all the time. Many ancestral cetaceans are found in the former Tethys Sea, the remnant of which is the modern Mediterranean Sea. This page will be updated as soon as any new information is unearthed.

An articulated composite skeleton of Pakicetus, the earliest known cetacean

Pakicetus was known only from a skull until 2001. It was linked to cetaceans via a component of the skull unique to whales, specifically the way the bones of the inner ear are structured. Only living and fossil cetaceans have their ears structured in this way. The known species of Pakicetus form the Family Pakicetidae. Pakicetus was about the size of a wolf and was originally postulated to have adaptations for underwater hearing and wading activity.

However, Pakicetus also had many characteristics of land animals, as shown by the additional parts found in 2001 by Thewissen et al and the fact that Pakicetus fossils were found accompanied by other terrestrial animal fossils. These discoveries suggested that Pakicetus was a transitional form that did not yet have strong underwater capabilities.

One note on "transitional" forms: though we see Pakicetus as a stepping stone toward cetaceans, it was still a fully-formed, complex animal in its own right. It is a mistake to view evolution as an endless progression upward, i.e. saying a blue whale or bottlenose dolphin is the end-all for cetaceans. Evolutionary reactions are responses to the environment and selective pressures on animals, so every animal must be reviewed as constantly adapting to its surrounds. There is no peak or "ultimate form," so avoid using "primitive" or "archaic" terminology around early whales like Pakicetus.

An artistic rendering of Pakicetus by Carl Buell

A skeleton drawing of Rodhocetus

Rodhocetidae is a genus within the family Protocetidae. The species Rodhocetus balochistanensis is the one used by Gingerich to demonstrate the connection between cetaceans and artiodactyls: the ankle bones of this species have a double-pulley mechanism unique to artiodactyls/whales. All other mammals only have a single-pulley (or "single-spooled") ankle structure.

Rodhocetus had very good hearing underwater, but it still lacked the definitive cetacean style of swimming via tail. It is unknown if protocetids had tail flukes, but in the Rodhocetid genus the animals would paddle with their legs while using a thick, muscular tail as a rudder. Some people speculated that Rodhocetus had the push-and-glide swimming style of a sea otter, but Gingerich demonstrated that in fact the limb proportions of Rodhocetus are closest to the fast-paddling Russian desman, which is closely related to moles. However, this does not mean that Rodhocetus was a close relation of moles, only that it swam like one of their cousins.

Rodhocetus was about 10 ft in length. In addition to the cetacean characteristics mentioned above, the spine of Rodhocetus can tell us a great deal about the animal's transitional status. A segment of the spine called the sacrum articulated directly with the animal's pelvis, supporting its weight on land (a characteristic of early terrestrial mammals). However, other segments of the spine were shortened to make the animal's front half more rigid (good for powering through the water), while the legs of Rodhocetus were shortened to streamline the body. These latter traits are characteristics of cetaceans, so Rodhocetus carried both elements of its terrestrial mammal ancestors and of the cetaceans that would be derived from it.

An artist's rendering of Rodhocetus

Brian Choo's rendering of Janjucetus, a 25-million-year-old offshoot of the baleen whales

Paleoartists have the challenging job of bringing to life animals they know only by skeletons that are often incomplete. Each artist presents their own take on the animal, and what you should understand as a viewer is that no depiction is concretely the "correct" depiction. Some aspects of animals (ex. where the muscles and bones attached) can be precisely determined even by a handful of bones. For example, if you have a skull you can tell where the spine was fastened and how the animal carried itself. If you have a femur you will have indentations where the muscles attached. But how bulky an animal might be, or how much fat it carried with it, generally remains a mystery. Many early whales are shown to be thin and fat-free, as they lived in the warm Eocene periods.

Rob van Assan's rendering of Squalodon, an extinct odontocete (toothed whale) that was one of the earliest to demonstrate echolocation

But without skin imprints or fur, we are still struggling to grasp what the surface of these animals might have looked like. External features change rapidly, allowing for the incredible diversity of animal skins and hides we see today.

Carl Buell's rendering of Remingtoncetidae, small cousins to Ambulocetus

The following is an excerpt from the blog of Carl Buell, the paleoartist who inspired this lens. Here he discusses his restoration of Pakicetus:

"Unless you find that once-in-a-lifetime, fully articulated skeleton with skin, scale, fur or feather impressions, the further back you go in time, the more difficult the process becomes of bringing an extinct animal to life. Hopefully, any guesswork involved is educated guesswork based on comparative anatomy, but Pakicetus, as portrayed here, is based on a skull and non-articulated post-cranial material. That means it's a composite. The bones are from a number of individuals. As more fossils are found and described, Pakicetus' look will change. My take on any fossil-based animal is just that, my take. You can only be true to the fossil evidence you have presented to you and then try to make informed decisions about surface features and color based on possible relatives (or ecological equivalents) of your subject that are alive today and the type of environment the animal lived in.

The one thing I always do, is try to look at each animal as a completely formed individual. What I mean by that in this case, is that I didn't think of Pakicetus as a 'future whale' or an early step toward whaleness - even though it certainly was from our viewpoint. Any naturalist transported into the past would see Pakicetus as a successful creature in the world of its time. Changing conditions, differential survival, and genetic mutation would do their work, but Pakicetus lived and breathed and foraged and mated and died a complete creature. This little white-sided Dolphin - a fast, acrobatic, and conspicuous resident of the North Atlantic - is not an 'end' result, but simply a glorious, present-day manifestation of Darwin's final words in The Origin of Species: 'endless forms most beautiful and most wonderful have been, and are being evolved.'"

--- Carl Buell, November 22nd 2006