News & Politics » Feature

The Descent of Dinosaurs

Fossil finds have made Paul Sereno a media darling, but what he's really looking for is a family tree.


1 comment

Dinosaurs may be extinct, but they're everywhere. They spawn anthropomorphic storybooks for the young (Bronto the Baby Dinosaur) and potboilers (Jurassic Park) for their elders. They breed museum exhibits, TV shows, movies, models, and robots; encyclopedias and dictionaries; lunch boxes, Kool-Aid flavors, cookies, and canned pasta.

Paul Sereno doesn't really mind that dinosaurs are the world's favorite dead animals. He has even taken part in the hoopla by writing a children's book, How Tough Is a Tyrannosaurus? ("Kids can relate to this subject, and then you can be an emissary of science in general.") He also let National Geographic take "five or six CAT scans and an X ray" of him to get a skeletal image of his hand holding one of his discoveries, the skull of the oldest known dinosaur, Eoraptor lunensis. (His own slide show includes a shot of him holding the skull of the distinguished 19th-century paleontologist Edward Drinker Cope.)

It's a safe bet that Sereno's finger bones would not be in National Geographic if he had specialized in the adaptive radiation of trilobites. But sometimes all the hype worries him. An assistant professor in the Department of Organismal Biology and Anatomy at the University of Chicago who at 35 can still pass for an undergraduate, he has to defend his subject against suspicions of frivolity. "It's almost with embarrassment that you admit to working with dinosaurs," he says. Lay people sometimes suppose that scholars already know all there is to know about them (in fact, more than half the known species were discovered in the last 20 years, and there are more to come). Academics tend to look down on subject matter that fascinates five-year-olds. And evolutionary biologists know that the fossil records of mammals and invertebrates are more extensive.

Are dinosaurs more than very large curiosities to be named and filed away? Can they help answer genuine scientific questions? Sereno's answer is yes, or he wouldn't be a paleontologist. Even after graduating from Northern Illinois University in biology in 1979, Sereno was still seriously considering a career in art. He chose paleontology not because of any youthful infatuation with gigantic reptiles, but because he discovered the field was in the midst of a revolution. New methods of systematically categorizing species and deducing their family trees--known as "cladistics"--were enabling scientists to "look back at dinosaur evolution with a new set of glasses and to find things it would have been unimaginable to find before. The idea is that if we are more rigorous we can reconstruct the actual tree of speciation events"--the points at which each new species developed from its ancestors. "Up until 10 or 15 years ago no one believed we could actually do something like that." Sereno and his colleagues have transformed dinosaur studies from a musty hoard of antiquities into an organized science capable of tracking even the physical migration of creatures a hundred million years dead.

To understand what Sereno's doing, you should first quit calling all large obsolescent things "dinosaurs." "People don't realize that the dinosaurs' world persisted for some 165 million years" (from 230 to 65 million years ago), says Sereno. "That's three times as long as mammals as we know them and more than 100 times as long as anatomically modern man." Focusing on their extinction can be misleading; Sereno is more interested in how they lived. Fossils record dinosaurs' existence from beginning to end, he explains, making them a perfect laboratory in which to study evolution. "With mammals we're still in mid-stride."

But a laboratory is no good if everything in it is helter-skelter. People have known about dinosaurs for 150 years, but until the last 15 years we had only a sketchy knowledge of their family tree. The name itself has even been questioned. Paleontologists earlier in this century suspected that "dinosaurs" might not be all that closely related and that therefore coining the word might have been a romantic error. There are two big subgroups of dinosaurs, saurischians (those with lizardlike hip joints, including Tyrannosaurus and Brontosaurus) and ornithischians (those with birdlike hip joints, such as Triceratops and the duckbills.). The question is, where did these two subgroups come from?

In BC (Before Cladistics), Sereno says, what passed for dinosaur family trees were extremely vague: a big balloon labeled "thecodonts" (a generic term for ancestral reptiles) sprouted lines for crocodilians, ornithischians, saurischians, and birds, with no attempt to follow the actual evolutionary sequence.

If we want to go beyond the balloon-and-sticks model and get specific--and that's Sereno's whole idea--there are two possible ways dinosaurs originated. One is that ornithischians and saurischians branched off from a common ancestor that was itself a dinosaur, already distinct from the usual run of reptiles. If so, then "dinosaur" would be a meaningful name for a gigantic extended family of related creatures. Their family tree would look like a capital Y atop the reptile line.

The other possibility--popular among paleontologists until fairly recently--is that ornithischians and saurischians might have branched off from the general run of reptiles separately, at different times. If so, then there was no common ancestor of all dinosaurs (other than the primitive reptile that was the ancestor of all later reptiles as well), and the dinosaur family tree would not be a tree at all, just two parallel lines coming off the reptile line at different points. "Dinosaur" would simply be a popular word for any big ugly extinct reptile, with no more evolutionary significance than the word "bug."

Cladistics, the 1950 brainchild of German entomologist Willi Hennig, makes it possible to ask and answer the question of where the two subgroups came from in specific terms, transforming it from a matter of opinion into a matter of science. The reasoning goes through three deceptively simple steps: (1) Both ornithischians and saurischians have many features in common with crocodiles, turtles, plesiosaurs, and other reptiles--such as chewing teeth and hard-shelled eggs. These are "primitive" features, and we can ignore them. (2) Ornithischians and saurischians also differ in some ways, in the shape of their hip joints, for one. These are "derived" characteristics, and we can ignore them too for the moment. (3) The key question is, do the two groups have any anatomical features in common that they do not share with other reptiles? The more such features dinosaurs share, the more likely it is that they form a group of their own with a single common ancestor. If dinosaurs don't have any distinctive features in common, then they're probably not a single related group.

That's the logic. The research involves examining and measuring lots of dinosaur skeletons, usually much more closely than past paleontologists did. And that research has shown that dinosaurs do have distinctive new features in common. For instance, both ornithischians and saurischians lack the "prefrontal" skull bone just above the eye that other reptiles possess. Both also have a hingelike ankle joint very different from the crocodiles' ball-and-socket joint. Sereno is particularly fond of this ankle joint, an adaptation that made it easier for dinosaurs to walk upright. "It's one of those evolutionary 'decisions' that is pretty committal. It creates a pathway you can follow for millions of years."

Cladistic reasoning does not depend on finding an actual fossil of the common ancestral dinosaur (though that would certainly help). It simply shows that such a creature must have existed, because the alternative is absurd: if ornithischians and saurischians were independent branches off the reptile line, then we would have to suppose that they both, on their own, happened to develop the same hinged ankle joint, the same skull without a prefrontal bone, and all the rest. The chances are much better that the Cubs will threepeat the World Series.

Paleontologists had pretty well concluded that dinosaurs were a legitimate group by the time Sereno joined the profession in the mid 80s. But within the group, genealogies of the 250 to 300 well-described species remained little known. Sereno's big research project has been to use careful observation and cladistic reasoning to sort out ancestors from descendants in the ornithischian half of the dinosaur clan. (Jacques Gauthier of the California Academy of Sciences has done the same for the saurischian side.) Sereno has a yard-wide family tree (cladogram) of all the dinosaurs taped up on an office door; he hopes the relationships it depicts will all be fully documented and published within two years. In this process of setting their evolutionary laboratory to rights, Sereno and his colleagues have already learned some surprising things about how evolution works.

You could spend a lifetime on dinosaur cladistics and never go outdoors. You have to study a lot of fossils close up, and the best place to find them is in the back rooms of the world's great museums, where bones have been piling up since the 19th century, often without having been described precisely enough for cladistic purposes. That's where Sereno started. But he didn't find everything he wanted to measure.

"When we saw this [dinosaur family tree] coming together, we decided we needed a solid foundation for it. The earliest dinosaurs were known only from fragments or by inference.

"Since Robert Bakker and others argued for a common dinosaur ancestor in 1974, dozens of [distinctively dinosaurian] features have been alleged to have evolved." Various paleontologists' lists overlapped, but they contained many different features, a total of 59. "We weren't sure what the common ancestor actually had. I went to Argentina [in 1988] to see if it was on the shelf. It wasn't." So he went into the field there, to the Ischigu-alasto Provincial Park, with leading South American paleontologist Jose Napoleon Bonaparte. "And we were very fortunate."

Northwestern Argentina is one of the few places on earth that fulfills two essential conditions for Sereno's search. First, sediments were accumulating there 228 million years ago, so that when a creature died its bones might be covered with silt. Under eons of accumulation and pressure, the silt became rock and the bones mineral. Second, in this part of the world these rocks have since been exposed by erosion or uplift and are now on the surface where we can get at them.

Much has been written about Sereno's luck in finding good fossil specimens, but it's to a considerable extent the luck of the prepared. Last summer the Tribune magazine called him an "Indiana Jones," but that misses the context of his famous and usually successful expeditions. His work is more like that of the person who comes to organize your attic--and finds unexpected treasure in the process.

On the 1988 visit to Argentina Sereno found a complete skull and skeleton of Herrerasaurus ischigualastensis, an early saurischian between 12 and 15 feet long. It ran on two legs and had short forearms with a clawed thumb and two fingers that could bend backward, the better to slash its prey with. It had a hinge in the middle of its lower jaw, enabling it to accommodate struggling victims. This adaptation proved useful enough that it persisted in later species of its line, including Tyrannosaurus rex.

On a second Sereno trip in 1991 Argentinean paleontology student Ricardo Martinez found the fossil of the dog-size predator Eoraptor lunensis. Eoraptor was smaller than Herrerasaurus and had not yet developed the double-jointed jaw.

These discoveries made news--more because they're very old than because they help clear up some mysteries of dinosaur descent. Sereno has had to deal with throngs of media people ever since. (Among the better results are Bill Kurtis's video "Fragments of Time" from the New Explorers series and the relevant portions of Don Lessem's book Kings of Creation.)

The new fossils did confirm many of Jacques Gauthier's 1986 predictions about what an ancestral dinosaur might look like. They also helped Sereno cut the list of confirmed common dinosaur characteristics to 11.

Neither Herrerasaurus nor Eoraptor is the common ancestor of all dinosaurs. Yet Sereno places them low down on the saurischian side of the family tree, closer to its roots than ever before--possibly as close as we will ever get. Eoraptor's ancestors were probably smaller than it was, and small skeletons are easily scattered, unlikely to survive intact long enough to be fossilized.

"How many characteristics do you look for in an average skull?" I asked Sereno. "Hundreds," he said, including aspects of its shape, the way the bones fit together, the openings in it. How does he know which features to look for? Practice, mostly. And practice required seeing the bones in person--another reason for museum visits. (His scholarly articles are frequently illustrated with "stereopairs" of important skulls, double images that produce a three-dimensional illusion when properly viewed.)

"The best way I've been able to explain it," he says, laughing, "is to show people slides of Richard Nixon and Winston Churchill along with editorial cartoon caricatures of them. People have no trouble matching each leader with his caricature. Then I tell them, 'What you've done in three seconds with these pictures is what I've taken ten years to train myself to do with dinosaur skulls.'" Sereno hasn't given up an art career. He's just working in a different medium.

Just as we tend to shorten the age of dinosaurs in our minds, emphasizing their extinction over their 150-million-year rule, we also tend to think of them as having lived at the same time. Eoraptor and Tyrannosaurus "both" lived in the Mesozoic era. But in fact we are much closer in time to Tyrannosaurus (which flourished 65 million years ago) than it was to Eoraptor (150 million years ago). An astonishing number of dinosaur species developed, adapted, died out, and were replaced by new ones in the interim. The 300 species described so far are almost certainly just the tip of the iceberg. "We're looking at real small time slices" in existing fossil records, Sereno explains. "When you go from one rock layer to another, you see different species," he says. So whenever a rock layer is missing, which is often, some new species are probably missing too. Also, he points out, "We don't have a record for all the continents."

But however incomplete, this is Sereno's "laboratory of evolution," and the things he's learned there are as striking as the badlands scenery of good fossil country. They range from the very specific (migration routes) to the quasi-philosophical (why dinosaurs succeeded)--but they all depend on knowing the dinosaurs' family trees.

Getting bigger all the time? As he's worked to reconstruct dinosaur family trees, Sereno has been struck by how, over millions of years, the descendant species gradually become physically larger than their ancestors. This is still a somewhat controversial notion. Some paleontologists argue that it could be just a statistical artifact: since most species of animals are small, if they have even a few larger descendants the average size will increase. But Sereno focuses on individual family trees. And though occasionally one dinosaur species happens to be smaller than its immediate ancestor species, he's impressed that in only one lineage has he ever seen two successive specimens become smaller. "I know dinosaurs pretty well, and I was very surprised to find this pattern when I began to look at them in detail."

Why do successive species grow larger, if indeed they do, as time passes? This question has vexed generations of evolutionary biologists. Is it some kind of inherent force? Is it a way of occupying more ecological niches? "Once we have established the pattern with hard data, it may be possible to make these into testable hypotheses," says Sereno. But first you have to have a well-established dinosaur family tree.

Dinosaurs discover America. When Herrerasaurus and Eoraptor were flourishing the earth's continents were united in a single land mass now called Pangaea. Later, during the last half of the dinosaur era, Pangaea separated, first into two supercontinents ("Laurasia" and Gondwana"), then into approximately the arrangement we know today.

A land bridge would appear between western North America and eastern Asia during low-water millennia, and Sereno has evidence that dinosaurs crossed it more than once (probably from west to east the first time). The evidence is not footprints under the Bering Strait. Again it's cladistics.

Sereno carefully analyzed the characteristics of the ceratopsian dinosaurs, the line that began with the little parrot-beaked Psittacosaurus and ended millions of years later with the large-ruffed, three-horned plant eater Triceratops. When he lined up their skulls in order of descent, a strange pattern appeared: the most primitive fossil came from Asia, the next most primitive from western North America, the next from Asia, and so on. The same alternating pattern showed up when Sereno plotted the ancestry of another group of dinosaurs, the domeheads, the familiar extreme of which is Pachycephalosaurus. Either evolution was following amazingly parallel paths on both continents or "small groups were getting across every few million years." The second hypothesis seems much simpler.

"This whole northern-highway thing demonstrates the power of tree thinking," says Sereno. Without clear family trees for the ceratopsians and the domeheads he would have been able to say only that similar species had lived on both continents.

First two legs, then wings. Even before cladistics was invented, one of the intellectual embarrassments of dinosaur paleontology was the grab-bag term "thecodonts," used to refer to a confused cluster of early dinosaurs and their cousins, including ancestral crocodiles. In a long, heavily illustrated, and extremely technical supplement to the December 31, 1991, Journal of Vertebrate Paleontology, Sereno attempted to properly categorize this messy group.

In the course of the article he cast doubt on some conventional wisdom. Evolution didn't necessarily "progress" steadily from sprawling to upright posture, for instance. Some evidence suggests that the common ancestor of crocodiles, dinosaurs, birds, and the flying pterosaurs may have been upright and that later crocodilians reverted to a sprawling stance.

Sorting through the thecodonts gave Sereno another surprise. Of the two separate groups of archosaurs that evolved powered flight--first pterosaurs, later birds--all their near ancestors had already become erect bipeds. Without a well-established family tree, he could only have guessed at this relationship. "That's a strange 'coincidence,'" he says, chuckling. "And again it shows the power of tree thinking."

Coevolution with plants? Toward the end of the dinosaur era flowering plants, aka angiosperms, first appeared on earth. Apparently, as they became prevalent, three groups of vegetarian dinosaurs independently evolved new tooth structures for chewing plants more efficiently. Citing several colleagues, Sereno writes that these happenings may not be coincidence, that they "strongly suggest plant-herbivore co-evolution." But until we have better family trees for both the flowering plants and vegetarian dinosaurs, it remains only a suggestion.

Only a few miles north of the U. of C. is the Field Museum, where angiosperm authority Peter Crane works on his end of this problem. "I find it very helpful to be in Chicago," says Sereno, even though few dinosaur-age fossil beds are nearby. "It's not a widely known fact, but there are more paleobiologists and evolutionary biologists in this area than in any other place in the world. This is a great place for fossils and the study of evolution."

The casino theory of evolution. Sereno's 1988 and 1991 trips to Argentina weren't devoted exclusively to finding Herrerasaurus or Eoraptor. His group also excavated and counted vertebrate fossils of all kinds. They found that though many kinds of dinosaurs were alive 228 million years ago, they were quite rare, as were early ancestral mammals. Rhynchosaurs held sway--stocky, sprawling beasts with a beaklike snout and a very broad, flat skull. Also common were squat, mammallike traversodonts. Both groups have since become extinct.

Ever since Darwin (though it isn't really his fault), most people have believed that the dinosaurs took over from the rhynchosaurs because the dinosaurs, being two-legged, were able to drive the sprawling reptiles out of existence. (Even little Eoraptor looks smarter and faster than his vaguely crocodilian rivals.) Likewise, millions of years later, at the end of the Mesozoic era, the dinosaurs themselves supposedly lost the "struggle for existence" to the faster, smaller, warm-blooded mammals, which either ate the dinosaurs' eggs or could more easily survive in a cooling climate.

One problem with this conventional wisdom--Oxford University's Michael Benton was the first to make a strong point of this--is that it doesn't fit the fossilized facts. As Sereno's Argentinean dig confirmed, the principal varieties of dinosaurs were all already present 228 million years ago. They coexisted (in very small numbers) with rhynchosaurs and traversodonts for a good 20 million years without making any move on them that's visible in the fossil record. The dinosaurs didn't come to dominate until after the rynchosaurs died out for unknown reasons. According to this view, dinosaurs didn't beat out rhynchosaurs any more than the 1992 Chicago Bulls beat the 1964 Boston Celtics. They simply took their place after it became vacant.

More damaging to mammalian self-esteem, our distant ancestors were on the scene 228 million years ago too. Where was their alleged competitive advantage? Why didn't they preempt the Age of Dinosaurs altogether and get on with developing us?

Maybe competitive advantage had nothing to do with it. Quite possibly dinosaurs and, later, mammals took over the earth because they hung around until luck punched their evolutionary lottery ticket--not because they outhustled their predecessors. Once rhynchosaurs were gone, dinosaurs spread out and took over their various ecological niches. Once dinosaurs died off, mammals did the same.

Even outside biology this view of evolution has unsettling implications. Generations of believers in competitive capitalism, for instance, have taken comfort in the conviction that their worldview mirrored the reality of "nature red in tooth and claw." Just like the early mammals, little warm-blooded, garage-spawned Apple computers could sneak out at night and eat up IBM's means of reproduction. But if luck is a main player in evolution, one support of that predatory philosophy is gone.

Sereno draws less provocative conclusions. Even if you don't particularly care how evolution happened, he says, it's a great source of perspective. "Dinosaurs, more than other fossils, are like an icon of a lost world. They force people out of the human realm and into thinking about the distant past. You can say, 'Here's a species about to go extinct, don't kill it.' But there is nothing like showing people how small they are in the history of life."

Art accompanying story in printed newspaper (not available in this archive): photos/Peter Barreras.


Showing 1-1 of 1


Add a comment