Forward Into the Past: the Origin of Whales

The notion that evolution moved in a steady forward progression agreed well with the Victorian conception of history. People’s lives in Europe were better at the end of the nineteenth century than they were at the beginning, thanks to science and industry, and they would keep getting better. That steady improvement seemed to be reflected in the history of life itself.

But Victorian biologists knew that if there was some sort of imperative for progress, many animals were ignoring it. Barnacles, for example, descended from free‑swimming crustaceans, but they had given up that independent life for a lazy existence clamped to a piling or a ship hull. If evolution was a steady march, it could start moving backward at any moment. Victorian biologists didn’t give up their notion of progress; they simply turned it into a two‑way street, with progress in one direction and degeneration in the other. Ray Lankester, a British biologist, worried that degeneration could strike human society if people weren’t careful. “Perhaps we are all drifting towards the condition of intellectual barnacles,” he wrote.

But just as evolution is not a steady march of progress, it cannot run backward either. Evolution is change, nothing more or less. Tetrapods took their heroic crawl out of the water 360 million years ago, and their descendants have gone back in more than a dozen times. When they entered the water, they did not degenerate into lancelets, let alone lobe‑fins. Instead, they became things altogether new, such as whales.

Whales have been trouble for scientists ever since Linnaeus put together the first modern taxonomy in 1735. “Amidst the greatest apparent confusion, the greatest order is visible,” Linnaeus wrote about classification, and yet when he tried to classify whales, he seemed only to add more confusion. Were they fish or mammals? “These are necessarily arranged with the Mammalia,” he demanded, “though their habits and manners are like those of fish.” Whales, he pointed out, have hearts with ventricles and auricles like mammals, they are warm‑blooded, have lungs, nurse their young–just like mammals on land. They even have eyelids that move.

Linnaeus’s classification was hard for the public to accept. In 1806 the naturalist John Bigland complained that it “will never prevent the whale from being considered as a fish, rather than as a beast, by the generality of mankind.” And in Moby Dick , Ishmael probably spoke for most people in the nineteenth century when he said, “I take the good old‑fashioned ground that the whale is a fish, and call upon holy Jonah to back me.”

Darwin saw a way out of this confusion. Linnaeus was not playing some sort of meaningless game by grouping whales with mammals. The similarities that Linnaeus found were signs that whales (including porpoises and dolphins) descended from mammals that lived on land. Evolution had produced a metamorphosis that Ovid would have loved: it had transformed away their legs, given them flukes on their tail, had put their noses on top of their heads, and had made them so big that the heaviest whales can balance the scales with a town of 2,000 citizens. It had created a fishlike mammal, but it had not in the process gotten rid of the evidence of its ancestry.

How exactly evolution had accomplished all this, Darwin could not say. He could think of no living intermediates between whales and land mammals. But not knowing how did not bother him much, because he could imagine ways. He pointed out that bears sometimes swim for hours with their mouths open, catching insects. “Even in so extreme a case as this,” he wrote in Origin of Species, “if the supply of insects were constant, and if better adapted competitors did not already exist in the country, I can see no difficulty in a race of bears being rendered, by natural selection, more and more aquatic in their structure and habits, with larger and larger mouths, till a creature was produced as monstrous as a whale.”

The idea didn’t go over well. One newspaper complained that “Mr. Darwin has, in his most recent and scientific book on the subject, adopted such nonsensical ‘theories’–as that of a bear swimming about a certain time till it grew into a whale, or to that effect.” Darwin dropped the example from later editions of his book.

In the 120 years that followed, paleontologists found a steady supply of fossils of whales, but even the oldest specimens, dating back more than 40 million years, were fundamentally like whales today. They had long backbones, hands in the shape of flippers, and no back legs. Their teeth were another matter, though. Living whales have either no teeth or simple pegs. The oldest whales had teeth that had the cusps and bumps of mammals on land. They looked particularly like the teeth of an extinct line of mammals called mesonychids. These animals were hoofed mammals–relatives, in other words, of cows and horses but they had powerful teeth and strong necks adapted for a life of eating meat, which they got either by scavenging or hunting.

Finally, in 1979, Philip Gingerich, a paleontologist from the University of Michigan, discovered a whale that lived on land.

Gingerich and his team were searching for 50‑million‑year‑old mammal fossils in Pakistan. Today Pakistan is nestled within Asia, but when the fossil mammals were alive, it was little more than a collection of islands and coastlines. India at the time was a gigantic island drifting north toward the southern edge of Asia. Gingerich’s team found many bits and pieces of mammals, most of which they could immediately identify. But a few were harder to pin down. One particularly baffling fossil was the back part of a 50‑million‑year‑old skull. It was about the size of a coyote’s and had a high ridge running like a mohawk over the top of its head, where muscles could attach and give the mammal a powerful bite. When Gingerich looked underneath the skull, he saw ear bones. They were two shells shaped like a pair of grapes and were anchored to the skull by bones in the shape of an S.

For a paleontologist like Gingerich, these ear bones were a shock. Only the ear bones of whales have such a structure; no other vertebrate possesses them. Gingerich named his creature Pakicetus, meaning “whale of Pakistan,” and in the years since, he has found its teeth and bits of its jaw. They are intermediate between mesonychids and later whales, confirming that Pakicetus was in fact a 50‑million‑year‑old whale–the oldest whale known at the time. Yet the rocks where Pakicetus fossils have been found showed that this coyote‑like creature had lived and died on land, among low, shrubby plants and shallow streams only a few inches deep. Pakicetus was a terrestrial whale.

Fifteen years later, in 1994, a former student of Gingerich’s named Hans Thewissen discovered another primitive whale. Thewissen found not just bits and pieces of his creature but what would ultimately turn out to be its entire skeleton. This 45‑million‑year‑old whale had giant feet that looked as if they could fit in clown shoes and a bulky skull in the shape of an alligator’s head. He named it Ambulocetus, meaning “walking whale.” By the close of the twentieth century, Thewissen, Gingerich, and other paleontologists had found several other species of whales with legs in Pakistan, India, and the United States. What once seemed an impossibility is now commonplace.

To understand how these early whales evolved into their fishlike life, paleontologists compared their fossils to living and extinct species. They drew an evolutionary tree, which told a provisional story of how whales evolved. Darwin shouldn’t have been thinking of bears. He should have been thinking of cows and hippos. These hoofed mammals are among the closest living relatives of whales. Paleontologists generally agreed that whales were even more closely related to an extinct group of hoofed mammals called mesonychids. Mesonychids took many forms, from creatures the size of squirrels to a terrifying monster called Andrewsarchus that measured 12 feet long–the biggest known meat‑eating mammal of all time. Among their ranks, the first whales walking on land wouldn’t have drawn much notice.

Pakicetus, which lived 50 million years ago, was far from the first whale. Mesonychids and whales all share a common ancestor, and that common ancestor must have lived before both the oldest known whale and the oldest known mesonychid. While the oldest known whales are about 50 million years old, there are fossils of mesonychids dating back 64 million years. Thus, whales must have diverged from mesonychids before 64 million years ago, more than 14 million years before Pakicetus lived.

Early whales still had legs attached to their shoulders and hips, which in turn were firmly attached to their spines. Their ears still resembled the ears of land mammals, which can pick up sounds in the air. Their teeth still resembled those of mesonychids in their general outline, but they were already changing. Maureen O’Leary, a paleontologist at the State University of New York at Stony Brook, has looked closely at the teeth of early whales and has discovered long gouges running along the outward sides of the lower molars. These gouges formed as the whales scraped their molars with their upper teeth. The whales had to have been making only vertical bites, not side‑to‑side chewing, to form them. There’s fossil evidence that later whales, which also had these gouges, fed on fish. That has led O’Leary to suggest that Pakicetus and its contemporaries had already started eating fish or other aquatic animals. Even without a modern whale’s body, an early whale could have swum, if only in a dog paddle.

But soon after Pakicetus emerged, evolution began changing other parts of whale anatomy, producing animals better adapted to swimming. Ambulocetus, Hans Thewissen’s walking whale, has short legs, a long snout, big feet, and a powerful tail. That sort of anatomy would have suited it for swimming like an otter–pushing back with its hind feet and adding more thrust by moving its tail up and down. Yet Ambulocetus, like otters, had hips that were still articulated to its spine. In other words, it could still walk on land. It probably hauled itself out of the water to bask, sleep, mate, and bear young.

In the border zone between land and sea, many species of walking whales arose. Some were adapted for wading, others for diving. For the most part, these lineages became extinct, for reasons that may never be known. But one lineage of whales adapted to life farther out to sea. It produced species such as Rodhocetus , a whale Gingerich found in Pakistan that had stubby legs and hips that were barely connected to its spine. In the water it could have raised and lowered its tail and trunk together, swimming like whales do today. Modern whales vastly improve their swimming performance with the flukes at the end of their tails, made of connective tissue. Because that sort of flesh rarely fossilizes, no one knows if Rodhocetus already had flukes to help it along.

By 40 million years ago, whale evolution had produced fully marine whales. Basilosaurus measured 50 feet long, with a slender serpent body, a long snout, and arms that had become stout flippers. It lived far from shore, and dry land would have been a death sentence for it. When paleontologists look at where the stomach once was on Basilosaurus fossils, they sometimes find a preserved meal of shark bones. These animals were much closer to our picture of a whale. But Basilosaurus lived for millions of years alongside semiaquatic and seallike whales that still retained some parts of their terrestrial past.

And Basilosaurus still had a few vestiges of its own. Its nostrils had moved only halfway back its snout toward the position where blow holes are on living whales. In 1989 Gingerich found a fossil in Egypt that tied Basilosaurus even more tightly to the past. Along its giant snaky body he found hips, and attached to the hips were hind legs. They were only a few inches long, but they even had five delicate toes.

Like all evolutionary trees, this one is a hypothesis. And like all hypotheses, it may have to be refined as more evidence comes in. It’s possible, for example, that Basilosaurus is not the closest relative to living whales–that honor may go to a species called Dorudon. Meanwhile, some scientists have been uncovering fascinating information in the genes of whales. Whale DNA shows that they are hoofed mammals, just as the paleontologists had concluded years earlier. But studies of their genes suggest that whales are most closely related to one particular hoofed mammal: the hippopotamus. Paleontologists were skeptical. Hippos belong to a group of hoofed mammals called artiodactyls. All artiodactyls have a specially shaped ankle with two rounded ends. Mesonychids don’t. Since whales were believed to be closely related to mesonychids, they couldn’t be artiodactyls. So the DNA results didn’t make sense. But in 2001, Gingerich and his colleagues discovered an early whale fossil in Pakistan, with the first ankle ever found. The ankle had the rounded ends like a hippo’s.

Although these open questions are important, they don’t alter the basic lessons of this tree. A whale is no more a fish than a bat is a bird. Early whales evolved into remarkably fishlike forms through a gradual series of steps. But inside every whale’s finlike flipper there still remains a hand, complete with fingers and a wrist. And while a tuna swims by moving its tail from side to side, whales swim by moving their tails up and down. That’s because whales descend from mammals that galloped on land. Early whales adapted that galloping into an otter‑like swimming style, arching their back in order to push back their feet. Eventually new whales emerged in which evolution had adapted that back‑arching movement to raise and lower a tail.

Although whales went through an extraordinary burst of evolution, their history put limits on what they could become. And that was not the only constraint on how they could evolve. Whales and other mammals did not undergo their dramatic radiations until the dominant vertebrates of the day–dinosaurs and giant marine reptiles–were gone. The first mammals evolved more than 225 million years ago, and for the next 150 million years they remained squirrel‑sized, barely distinguishable from one another. It was not until after the Cretaceous period ended, 65 million years ago, that the oldest fossils of most of the living orders of mammals appear. It was only then that the first primates leaped through the trees, only then that whales split off from other hoofed mammals and began their return to the ocean. In a few million years mammals became flying bats; they became gigantic relatives of rhinos and elephants; they became powerful, lion‑sized predators. Mammals experienced their own explosion, one somewhere on the scale between the Cambrian explosion and the burst of cichlids in Lake Victoria. Ever since, mammals have dominated the land and thrived in the oceans. But mammals only underwent this explosion because millions of species–including marine reptiles and dinosaurs suddenly disappeared. It was not some steady, gradual perfection that was responsible for the rise of mammals. It was, instead, an asteroid from space, sweeping away the old to make way for the new.

Seven

Extinction