Text Size
June 2011 | Home
Whales of the desert
Philip Gingerich had little interest in whales until he started discovering mysterious fossils in the deserts of Pakistan and Egypt. His research since then has transformed our understanding of modern day whales—and of how evolution works. His life's work is encapsulated in a new, permanent display at the University of Michigan Museum of Natural History. (Photo: Eric Bronson, U-M Photo Services.)
June 15, 2011
More than 50 million years ago, in a shallow sea that stretched from what is now Spain to present-day Indonesia, the first whales tested the waters of a whole new way of life. These early ancestors of today's largest marine mammals bore little resemblance to Moby-Dick, Monstro or Shamu. In fact, if you'd seen one paddling around the shoreline, you might have mistaken it for an oversized Labrador retriever.
Instead of propelling themselves through the water as modern whales do, with up-and-down strokes of their tail flukes, these creatures used their legs to swim. That's right, their legs.
Was this the early evolutionary path toward modern whales?
Click image for larger size. Elomeryx, top, was a land animal related to modern-day goats, pigs and hippos. Pakicetus was clearly a water creature, but it spent some of its life on land and had the feet of a land mammal. Rodhocetus's feet worked for both walking and swimming. Dorudon is striking for its resemblance to modern whales (note the front flippers and horizontal flukes)—but it still sported tiny back feet. (Images by John Klausmeyer.)
The oldest whales known to science had front and hind limbs that were modified for swimming. The shoreline was home, where they rested and reproduced, but they went into rivers and seas on a regular basis to find food. The story of how these four-legged mammals, perhaps no larger than a large sheep, transformed over time into today's fully aquatic bowheads, humpbacks, orcas and such is the stuff of U-M paleontologist Philip Gingerich's research. Spanning more than three decades, the work is showcased in a recently updated, permanent exhibit, "Back to the Sea: The Evolution of Whales," at the U-M Exhibit Museum of Natural History.
An accidental discovery
Gingerich wasn't looking for—or even especially interested in—whales, when a member of his team made the discovery that would redefine Gingerich's research focus.
"We were in Pakistan in 1978, collecting fossils of land mammals and looking for the ancestors of dawn horses, when a colleague working with me split open a block of conglomerate with bones in it and exposed the side of a skull," Gingerich recalls. The skull was a little larger than that of a wolf, but the size of the braincase suggested the animal's brain was much smaller than a wolf's—scarcely bigger than a walnut. Soon after, Gingerich happened upon some museum specimens of archaic whales and noticed how small their braincases were, relative to their skulls. Could the small-brained creature from Pakistan—later named Pakicetus inachus—possibly be an early whale?
Gingerich displays one of the first real clues to whales' evolution: the skulls of a dog and the ancient creature Pakicetus. Though the skulls are similar, the Pakicetus brain case is much smaller—and its ears contain structures that help whales hear underwater. (Photo: Eric Bronson, U-M Photo Services.)
Back in Michigan, Gingerich discovered another link between the mystery mammal and whales, in a pair of bones at the base of the skull called auditory bullae. In land mammals, these bones are small and securely anchored to the skull; in modern whales, the bullae are larger, denser and only loosely attached. These shell-shaped structures help whales hear underwater by amplifying sound vibrations, which are then transmitted to the inner ear.
Pakicetus, Gingerich noticed, had dense-looking bullae like those of whales, suggesting it spent a lot of time in the water. However, it didn't have other adaptations that help whales determine which direction underwater sounds are coming from, so Gingerich thinks this primitive "proto-whale," which lived about 50 million years ago, was semi-aquatic, swimming to hunt for food, but frequently returning to terra firma.
What lured Pakicetus and its proto-whale predecessors into the sea? In a word, dinner.
"We think that whales made the transition from land to sea opportunistically, to take advantage of food in the sea that no one else was eating," says Gingerich. It's not so hard to envision the scenario that led to the first, tentative excursions.
"We have in Michigan today foxes and coyotes that trot along the edge of Lake Michigan scavenging the dead fish that wash up on the shore," he says. "It's easy to imagine a coyote-like or wolf-like mammal going into the water to pick up fish that are dying, and over the course of generations, beginning to chase fish that are still living and becoming more and more efficient at catching them."
It seemed like a joke
The year before the Pakicetus discovery, in 1977, Gingerich's team found a sturdy pelvic bone, with a socket for a large upper leg bone like that of a four-legged land mammal, in marine deposits, prompting jokes about "walking whales." Of course, no one—least of all Gingerich—really believed the bone was from a whale. It was possible, in theory, but very few pelvic bones were known for whales, and none had a socket for a large upper leg bone. Assuming the bone was from some yet to be identified terrestrial quadruped, Gingerich gave it little thought.
"Why is a rib bone here?" Gingerich wondered. But it wasn't a rib. It was a leg bone, with a well- formed knee. Nobody had ever seen the knee of a whale before.
Not until the 1990s did he find complete skeletons that changed the concept of walking whales from a joke to a scientific certainty. But even before then, he made another discovery that hinted at the link between whales and their land-locked ancestors. It was 1989, and he and his team had come to a fossil-rich area in the Egyptian desert that would come to be called Wadi Al-Hitan (Valley of the Whales) to unearth the skull of a Basilosaurus—a peculiar, 50-foot long whale with a massive head and a sea-serpent body—which lived 37 million years ago, 10 million years after Pakicetus.
The area was littered with Basilosaurus skeletons—far too many to ever collect—and after excavating the skull, Gingerich wanted to measure the other fossils and map their locations. As he went about the task, he noticed what looked like a partially-exposed rib bone in one skeleton. But the bone was in an odd place for a rib, near the base of the tail.
In the deserts of Pakistan and the Wadi Al-Hitan (Valley of the Whales, above) in Egypt, Gingerich made astounding finds. (Photo: Philip Gingerich.)
"I thought, 'Why is a rib here?'" Gingerich recollects. "I started to brush away the sand and realized it wasn't a rib. It was the distal end of an upper leg bone, and it had a well-formed knee. Nobody had ever seen the knee of a whale before—of any whale anywhere."
Armed with the new knowledge of exactly where on the skeleton to look and what size of bones to search for, Gingerich and co-workers revisited the other Basilosaurus skeletons mapped at the site.
"Immediately we started to find leg bones, toe bones and foot bones," says Gingerich. "When we finished we had the complete pelvis and the legs, feet and toes of Basilosaurus."
Basilosaurus was a fearsome, 50-foot-long predator with tiny feet that could never have supported its weight on land. Its discovery provided important evidence of the evolutionary steps whales took as they moved from life on land to the sea. (Photo: Eric Bronson, U-M Photo Services.)
They were tiny legs—no bigger than a human child's—that couldn't possibly have supported the leviathan's weight on land. These holdovers were possibly still functional for positioning and clasping a partner during mating, but they must also represent a link to land-dwelling ancestors, Gingerich reasoned. After all, whales are mammals that nurse their young with mammary glands; mammals evolved on land, and most mammals still live on land. The surprise was finding legs, feet and toes, on a whale living 10 million years after Pakicetus.
In finding whale legs and feet, the researchers now possessed powerful evidence to counter one of the leading arguments put forth by opponents of evolutionary theory. If whales evolved from land mammals, there must have been intermediate stages between the fully terrestrial ancestors and today's fully aquatic whales—transitional forms with adaptations for both land and sea. Why, the anti-evolutionists always asked, had no one found fossils of those intermediates? Until Gingerich's work, paleontologists had no good comeback. Now, with Pakicetus and the Basilosaurus legs, they had some of those missing fossils in hand.
Rodhocetus, discovered in Pakistan, provided even more evidence that early proto-whales could move on land as well as sea. (Image by John Klausmeyer.)
Even more evidence of the transition from land to sea came when Gingerich shifted his research from Egypt back to Pakistan where he worked through the 1990s.
"By the end of that time, we had found complete skeletons of whales that could move on land, with legs and feet and toes that are not reduced like they are in Basilosaurus," he says.
One of those whales was Rodhocetus, found in northern Pakistan in 1992. Older than Pakicetus but younger than Basilosaurus, Rodhocetus had four fully-functional legs, with both fore- and hind legs modified for foot-powered swimming. "Tail swimming evolved later in the history of whales, when the link to land was lost," Gingerich says.
Hunting for the source
As the discoveries continued, and Gingerich's scientific papers piled higher and higher, the connections between modern whales and land mammals became clearer and clearer. But exactly which group of mammals had given rise to whales was still a much-debated question.
Scientists debated which animal family had spawned the creatures that would eventually evolve into whales. The critical piece of evidence would come from a humble source: ankle bones. (Photo courtesy Philip Gingerich.)
On one side of the debate, paleontologists were convinced—mainly on the basis of anatomical similarities to extinct land mammals—that whales came from a group of hoofed carnivores, mesonychids, that lived in the Eocene epoch, around 55 million to 34 million years ago. On the other side, immunologists and molecular geneticists were swayed by evidence indicating that, among living mammals, whales were most closely related to the mammalian order Artiodactyla, which includes goats, pigs, cattle, deer, camels, hippos and other even-toed ungulates.
Gingerich was a mesonychid man, and he knew there was one elusive fossil that could prove him right—or wrong: a key whale ankle bone called the astragalus.
Artiodactyls have a distinctive "double-pulley" astragalus with grooves at the top and bottom of the bone, easily distinguished from the single-pulley astragali of other four-legged mammals, including mesonychids. Gingerich had found astragalus bones in the ankles of Basilosaurus, but these were reduced and fused with other ankle bones. However, any astragalus associated with the skeleton of an earlier and more primitive whale might show which side—the mesonychid champions or the artiodactyl adherents—had the story right.
That find came in 2000, when colleagues working with Gingerich in Pakistan found two grooved pieces of bone amid the skeletal remains of a newly discovered fossil whale. Gingerich wanted to believe the bones were two separate single-pulley astragali, but he soon realized they were both part of the same bone: a typical artiodactyl double-pulley astragalus, from a whale later named Artiocetus clavis. A week later, complete hind limbs with a double-pulley astragalus were found for Rodhocetus too.
To Gingerich's surprise, the leg of Rodhocetus proved a connection between whales and modern mammals such as goats, pigs and hippos. (Photo: Eric Bronson, U-M Photo Services.)
So whales descended from some fully terrestrial artiodactyl, but what exactly was that land-based ancestor? That's still an open question. Based on skeletal features, Gingerich favors a hoofed mammal like Elomeryx, an extinct artiodactyl that lived along muddy shorelines and had habits similar to those of hippos.
If he's right, then whale family history goes something like this: an Elomeryx-like ancestor begat Pakicetus (or something like it) begat Rodhocetus (or something like it) begat … what? Not Basilosaurus—that was a weird evolutionary experiment that led to a dead end. Something else must be the link between the walking whale Rodhocetus and today's legless leviathans.
That something else could be Dorudon, which lived at the same time and inhabited the same seas as Basilosaurus. Much smaller than Basilosaurus, with a body length of 15 to 18 feet, Dorudon had tiny legs like those of Basilosaurus, but in other respects its anatomy was more like that of today's whales.
"It had a flattened tail with a fluke, so it could push itself through the water with up and down strokes just like whales do today," Gingerich says.
Dorudon's big, sharp teeth suggest it was a predator, probably devouring the plentiful fish in its oceanic environment. But Dorudon couldn't just cruise around with impunity; it had to watch its own back. Big bad Basilosaurus also had large, sharp teeth and probably made many a meal of luckless Dorudons. Still, enough of their ilk survived to give rise to modern whales.
The "mother" of all discoveries
With many of the blanks filled in, a few questions remain. For instance: once whales became successful in the sea, why did they retain their ties to land for so long? An extraordinary find in Pakistan in 2000 provided insights into that puzzle.
Gingerich's discovery of Maiacetus provided revolutionary insight into whale evolution. The animal had big teeth for catching fish, sturdy legs to support it on land and, most revealing of all, delivered babies head-first just like a land mammal. (Image by John Klausmeyer.)
"We were lucky because then-student Iyad Zalmout found a line of chalky residue on the surface," Gingerich recalls. "I followed it down, brushing carefully, with the morning sun illuminating the surface. The chalk turned into bone, and then I found part of the specimen full of little teeth. I was excited, thinking I'd found the skeleton of a small whale I believed to be in those beds. But as I kept working my way along the skeleton, nothing else made sense: the ribs were all going the wrong direction, and the vertebrae were too big for an animal with such small teeth."
Not until Gingerich reached the end of the skeleton and found a large skull did he realize what he had found: a female whale with a near-term fetus in the womb—another first among fossil whale finds.
It took eight years of work to piece together the story of this whale, dubbed Maiacetus ("mother whale"). During that time, Gingerich and coworkers also found a virtually complete male skeleton of the same species, which showed that while Maiacetus had big teeth for catching and eating fish, it also had legs sturdy enough to support its weight on land.
"I don't think there's any doubt that they were eating fish in the sea, but why were whales still coming out on land if they'd been spending so much time in the water for a couple million years?" Gingerich wondered.
The fetus's position in the womb—head first—provided the answer.
"Modern whales are born tail first," says Gingerich. "We're not sure why—maybe so they don't drown as they're being born, before they're free of their mother, or maybe so they can keep up with the mother as she's moving through the water. Land mammals are born head first. The position of this fetus tells us Maiacetus gave birth as land mammals do. We had speculated that they came onto land to give birth, but we never thought we'd find the evidence to demonstrate it."
Now Gingerich is revisiting earlier finds, trying to answer more lingering questions, not only about the path that led to today's whales, but also about the blind alleys.
One special interest is Basilosaurus, that serpentine hunter of Dorudon. The largest whale of its time and the top predator in the sea, Basilosaurus thrived for several million years before going extinct.
"It's an experiment that failed, and we'd like to understand it better, both to have some idea why it succeeded in life for so long, but also to understand why it ultimately failed," Gingerich says. The answers may lie in a nearly-complete, 50-foot-long Basilosaurus skeleton excavated in 2005 and brought to U-M for cleaning and stabilization. Before returning the skeleton to Egypt, Museum of Paleontology staff, aided by work-study and Undergraduate Research Opportunity Program students, made casts of each bone, producing several replicas of the complete skeleton for study and exchange with other institutions.
Exhibit Museum preparators also cast a lighter-weight version of the skeleton, which now appears to swim through the air above the "Back to the Sea" exhibit, commanding visitors' attention as it once commanded its ocean environs.
Even after decades of delving into fossil finds, Gingerich is as captivated by the gargantuan skeleton as any school kid on a field trip.
"Part of its significance lies in its unusual shape and the mystery that it represents in terms of what sort of animal it was," he says. "But to me what's significant is also that this is such a symbol of paleontology in Egypt and of the international scope of paleontological research at the University of Michigan. I'm pleased that we have it here, are studying it here, and now have a permanent exhibit of it here."