The Rise and Fall of the Dinosaurs Read online

  The Permian world was dominated by sprawlers. After the extinction, however, one new group of reptiles evolved from these sprawlers but developed an upright posture—the archosaurs. This was a landmark evolutionary event. Sprawling is all well and good for cold-blooded critters that don’t need to move very fast. Tucking your limbs under your body, however, opens up a new world of possibilities. You can run faster, cover greater distances, track down prey with greater ease, and do it all more efficiently, wasting less energy as your columnar limbs move back and forth in an orderly fashion rather than twisting around like those of a sprawler.

  Grzegorz Niedźwiedzki examines a life-size model of the Prorotodactylus trackmaker: a proto-dinosaur very similar to the ancestor that gave rise to dinosaurs.

  Courtesy of Grzegorz Niedźwiedzki.

  A handprint overlapping a footprint of Prorotodactylus, from Poland. For scale, the handprint is about 1 inch long.

  Photo courtesy of the author

  We may never know exactly why some of these sprawlers started walking upright, but it probably was a consequence of the end-Permian extinction. It’s easy to imagine how this new getup gave archosaurs an advantage in the postextinction chaos, when ecosystems were struggling to recover from the volcanic haze, temperatures were unbearably hot, and empty niches abounded, waiting to be filled by whatever mavericks could evolve ways to endure the hellscape. Walking upright, it seems, was one of the ways in which animals recovered—and indeed, improved—after the planet was shocked by the volcanic eruptions.

  Not only did the new upright-walking archosaurs endure, but they thrived. From their humble origins in the traumatic world of the Early Triassic, they later diversified into a staggering variety of species. Very early, they split into two major lineages, which would grapple with each other in an evolutionary arms race over the remainder of the Triassic. Remarkably, both of these lineages survive today. The first, the pseudosuchians, later gave rise to crocodiles. As shorthand, they are usually referred to as the crocodile-line archosaurs. The second, the avemetatarsalians, developed into pterosaurs (the flying reptiles often called pterodactyls), dinosaurs, and by extension the birds that, as we shall see, descended from the dinosaurs. This group is called the bird-line archosaurs. The Prorotodactylus tracks from Stryczowice are some of the first signs of archosaurs in the fossil record, traces of the great-great-great-grandmother of this whole menagerie.

  Exactly what kind of archosaur was Prorotodactylus? Some peculiarities in the footprints hold important clues. Only the toes make an impression, not the metatarsal bones that form the arch of the foot. The three central toes are bunched very close together, the two other toes are reduced to nubbins, and the back end of the print is straight and razor-sharp. These may seem like anatomical minutiae, and in many ways they are. But as a doctor is able to diagnose a disease from its symptoms, I can recognize these features as hallmarks of dinosaurs and their very closest cousins. They link to unique features of the dinosaur foot skeleton: the digitigrade setup, in which only the toes make contact with the ground when walking, the very narrow foot in which the metatarsals and toes are bunched together, the pathetically atrophied outer toes, the hinge-like joint between the toes and the metatarsals, which reflects the characteristic ankle of dinosaurs and birds, which can move only in a back-and-forth direction, without even the slightest possibility of twisting.

  The Prorotodactylus tracks were made by a bird-line archosaur very closely related to the dinosaurs. In scientific parlance, this makes Prorotodactylus a dinosauromorph, a member of that group that includes dinosaurs and the handful of their very closest cousins, those few branches just below the bloom of dinosaurs on the family tree of life. After the evolution of the upright-walking archosaurs from the sprawlers, the origin of dinosauromorphs was the next big evolutionary event. Not only did these dinosauromorphs stand proudly on their erect limbs, but also they had long tails, big leg muscles, and hips with extra bones connecting the legs to the trunk, all of which allowed them to move even faster and more efficiently than other upright-walking archosaurs.

  As one of the first dinosauromorphs, Prorotodactylus is something of a dinosaur version of Lucy, the famous fossil from Africa that belongs to a very humanlike creature but is not quite a true human, a member of our species, Homo sapiens. In the same way that Lucy looks like us, Prorotodactylus would have appeared and behaved very much like a dinosaur, but it’s simply not considered a true dinosaur by convention. That’s because scientists decided long ago that a dinosaur should be defined as any members belonging to that group including the plant-eating Iguanodon and the meat-eating Megalosaurus (two of the first dinosaurs found by scientists in the 1820s) and all descendants of their common ancestor. Because Prorotodactylus did not evolve from this common ancestor, but slightly before it, it is not a true dinosaur by definition. But that’s just semantics.

  In Prorotodactylus we’re looking at traces left behind by the type of animal that evolved into dinosaurs. It was about the size of a house cat and would have been lucky to tip the scales at ten pounds. It walked on all fours, leaving handprints and footprints. Its limbs must have been quite long, judging from the big gaps between successive prints of the same hands and feet. The legs must have been particularly long and skinny, because the footprints often are positioned in front of the handprints, a sign that its feet were overstepping its hands. The hands were small and would have been good at grabbing things, whereas the long, compressed feet were perfect for running. The Prorotodactylus animal would have been gangly looking, with the speed of a cheetah but the awkward proportions of a sloth, perhaps not the type of animal you would expect the great Tyrannosaurus and Brontosaurus to ultimately evolve from. And it wasn’t very common either: less than 5 percent of all the tracks found at Stryczowice belong to Prorotodactylus, an indication that these proto-dinosaurs were not especially abundant or successful when they first arose. Instead, they were far outnumbered by small reptiles, amphibians, and even other types of primitive archosaurs.

  These rare, weird, not-quite-true-dinosaur dinosauromorphs continued to evolve as the world healed in the Early and Middle Triassic. The Polish track sites, stacked orderly in time sequence like the pages of a novel, document it all. Sites like Wióry, Pałęgi, and Baranów yield an equally unfamiliar array of dinosauromorph tracks—Rotodactylus, Sphingopus, Parachirotherium, Atreipus—which diversify over time. More and more track types show up; they get larger; they develop a greater diversity of shape, some even losing their outer toes entirely so that the center toes are all that remain. Some of the trackways stop showing impressions of the hand—these dinosauromorphs were walking on only their hind legs. By about 246 million years ago, dinosauromorphs the size of wolves were racing around on two legs, grabbing prey with their clawed hands, acting a whole lot like a pint-size version of a T. rex. They weren’t living only in Poland; their footprints are also found in France and Germany and the southwestern United States, and their bones start showing up in eastern Africa and later Argentina and Brazil. Most of them ate meat, but some of them turned vegetarian. They moved quickly, grew fast, had high metabolisms, and were active, dynamic animals compared to the lethargic amphibians and reptiles they were cohabitating with.

  At some point, one of these primitive dinosauromorphs evolved into true dinosaurs. It was a radical change in name only. The boundary between nondinosaurs and dinosaurs is fuzzy, even artificial, a by-product of scientific convention. The same way that nothing really changes as you cross the border from Illinois into Indiana, there was no profound evolutionary leap as one of these dog-size dinosauromorphs changed into another dog-size dinosauromorph that was just over that dividing line on the family tree that denotes dinosaurs. This transition involved the development of only a few new features of the skeleton: a long scar on the upper arm that anchored muscles to move the arms in and out, some tablike flanges on the neck vertebrae that supported stronger muscles and ligaments, and an open-window-like joint where the thighbone
meets the pelvis. These were minor changes, and to be honest, we don’t really know what was driving them, but we know that the dinosauromorph-dinosaur transition wasn’t a major evolutionary jump. A far bigger evolutionary event was the origin of the swift-running, strong-legged, fast-growing dinosauromorphs themselves.

  The first true dinosaurs arose some time between 240 and 230 million years ago. The uncertainty reflects two problems that continue to cause me headaches but are ripe to be solved by the next generation of paleontologists. First, the earliest dinosaurs are so similar to their dinosauromorph cousins that it is hard to tell their skeletons apart, never mind their footprints. For instance, the puzzling Nyasasaurus, known from part of an arm and a few vertebrae from approximately 240-million-year-old rocks in Tanzania, may be the world’s oldest dinosaur. Or it may be just another dinosauromorph on the wrong side of the genealogical divide. The same is true of some of the Polish footprints, particularly the larger ones made by animals walking on their hind legs. Maybe some of these were made by real, true, honest-to-goodness dinosaurs. We just don’t have a good way of telling apart the tracks of the earliest dinosaurs and their closest nondinosaur relatives, because their foot skeletons are so similar. But maybe it doesn’t matter too much, as the origin of true dinosaurs was much less important than the origin of dinosauromorphs.

  The other, much more glaring issue is that many of the fossil-bearing rocks of the Triassic are very poorly dated, particularly those from the early to middle parts of the period. The best way to figure out the age of rocks is to use a process called radiometric dating, which compares the percentages of two different types of elements in the rock—say, potassium and argon. It works like this. When a rock cools from a liquid into a solid, minerals form. These minerals are made up of certain elements, in our case including potassium. One isotope (atomic form) of potassium (potassium-40) is not stable, but slowly undergoes a process called radioactive decay, in which it changes into argon-40 and expels a small amount of radiation, causing the beeps you’d hear on a Geiger counter. Beginning the moment a rock solidifies, its unstable potassium starts changing into argon. As this process continues, the accumulating argon gas becomes trapped inside the rock where it can be measured. We know from lab experiments the rate at which potassium-40 changes into argon-40. Knowing this rate, we can take a rock, measure the percentages of the two isotopes, and calculate how old the rock is.

  Radiometric dating revolutionized the field of geology in the middle of the twentieth century; it was pioneered by a Brit named Arthur Holmes, who once occupied an office a few doors down from mine at the University of Edinburgh. Today’s labs, like the ones run by my colleagues at New Mexico Tech and the Scottish Universities Environmental Research Centre near Glasgow, are high-tech, ultramodern facilities where scientists in white lab coats use multi million-dollar machines bigger than my old Manhattan apartment to date microscopic rock crystals. The techniques are so refined that rocks hundreds of millions of years old can be precisely dated to a small window of time, within a few tens or hundreds of thousands of years. These methods are so fine-tuned that independent labs routinely calculate the same dates for samples of the same rocks analyzed blindly. Good scientists check their work this way, to make sure their methodology is sound, and test after test has shown that radiometric dating is accurate.

  But there is one major caveat: radiometric dating works only on rocks that cool from a liquid melt, like basalts or granites that solidify from lava. The rocks that contain dinosaur fossils, like mudstone and sandstone, were not formed this way, but rather from wind and water currents that dumped sediment. Dating these types of rocks is much more difficult. Sometimes a paleontologist is lucky and finds a dinosaur bone sandwiched between two layers of datable volcanic rocks that provide a time envelope for when that dinosaur must have lived. There are other methods that can date individual crystals found in sandstones and mudstones, but these are expensive and time-consuming. This means that it’s often difficult to date dinosaurs accurately. Some parts of the dinosaur fossil record have been well dated—when there are enough interspersed volcanic rocks to give a timeline or the individual-crystal technique has been successful—but not the Triassic. There are just a handful of well-dated fossils, so we are not entirely confident of what order certain dinosauromorphs appeared in (especially when trying to compare the ages of species found in distant parts of the world) or when true dinosaurs emerged out of the dinosauromorph stock.

  ALL UNCERTAINTIES ASIDE, we do know that by 230 million years ago, true dinosaurs had entered the picture. The fossils of several species with unquestionable signature features of dinosaurs are found in well-dated rocks of that age. They’re found in a place far from where the earliest dinosauromorphs were cavorting in Poland—the mountainous canyons of Argentina.

  Ischigualasto Provincial Park, in the northeastern part of Argentina’s San Juan Province, is the type of place that just looks as though it should be bursting with dinosaurs. It’s also called Valle de la Luna—the Valley of the Moon—and you could easily imagine its being on some other planet, full of wind-sculpted hoodoos, narrow gullies, rust-covered cliffs, and dusty badlands. To the northwest are the towering peaks of the Andes, and far to the south are the dry plains that cover most of the country, where cows graze on the grass that makes Argentine beef so delicious. For centuries Ischigualasto has been an important crossing for livestock making their way from Chile to Argentina, and today many of the few people who live in the area are ranchers.

  This stunning landscape also happens to be the best place in the world for finding the oldest dinosaurs. That’s because the red, brown, and green rocks that have been carved and eroded into such magical shapes were formed in the Triassic, in an environment both full of life and perfect for preserving fossils. In many ways, this landscape was similar to the Polish lakelands that preserved the tracks of Prorotodactylus and other dinosauromorphs. The climate was hot and humid, although perhaps a little more arid and not pounded by such strong seasonal monsoons. Rivers snaked their way into a deep basin, occasionally bursting their banks during rare storms. Over a period of 6 million years, the rivers built up repeating sequences of sandstone, formed in the river channels, and mudstone, formed from the finer particles that escaped the river and settled out on the surrounding floodplains. Many dinosaurs frolicked on these plains, along with a wealth of other animals—big amphibians, piglike dicynodonts whose ancestors managed to make it through the end-Permian extinction, beaked plant-eating reptiles called rhynchosaurs (primitive cousins of the archosaurs) and furry little cynodonts that looked like a cross between a rat and an iguana. Floods would occasionally interrupt this paradise, killing the dinosaurs and their friends and burying their bones.

  The area is so heavily eroded today, and so little disturbed by buildings and roads and other human nuisances that cover up fossils, that the dinosaurs are relatively easy to find, at least compared to so many other parts of the world where we hike around for days just praying to find anything, even just a tooth. The very first discoveries here were made by cowpokes or other locals, and it wasn’t until the 1940s that scientists began to collect, study, and describe fossils from Ischigualasto, then still another few decades until intensive expeditions were launched.

  The first major collecting trips were led by one of the giants of twentieth-century paleontology, the Harvard professor Alfred Sherwood Romer, the man who wrote the textbook that I still use to teach my graduate students in Edinburgh. During his first trip, in 1958, Romer was already sixty-four years old and regarded as a living legend, yet there he was driving a rickety car through the badlands because he had a hunch that Ischigualasto would be the next big frontier. On that trip he found part of a skull and skeleton of a “moderately large” animal, as he so modestly put it in his field notebook. He brushed away as much rock as he could, coated the bones in newspaper, applied a coat of plaster that would harden and protect the bones, and chiseled them out of the ground. He sent the bone
s back to Buenos Aires, where they would be loaded on a ship to the United States, so he could carefully clean and study them in his lab. But the fossils took a detour. They were impounded for two years at the port in Buenos Aires before customs officials finally gave the go-ahead. By the time the fossils arrived at Harvard, Romer had occupied himself with other things, and it was only years later that other paleontologists recognized that the master had found the very first good dinosaur from Ischigualasto.

  Some Argentines weren’t so happy that a Norteamericano had come down to their neighborhood to collect fossils, which were being removed from Argentina and studied in the United States. That spurred a pair of up-and-coming homegrown scientists, Osvaldo Reig and José Bonaparte, to organize their own expeditions. They assembled a team and set out for Ischigualasto in 1959, and then again three times during the early 1960s. It was during the 1961 field season that Reig and Bonaparte’s crew met a local rancher and artist named Victorino Herrera, who knew the hills and crevasses of Ischigualasto the way an Inuit knows snow. He recalled seeing some bones crumbling out of the sandstone and led the young scientists to the spot.

  Herrera had found bones all right, lots of them, and clearly they were part of the back end of a dinosaur skeleton. After a few years of study, Reig described the fossils as a new species of dinosaur that he called Herrerasaurus in the rancher’s honor, a mule-size creature that could sprint on its hind legs. Later detective work showed that Romer’s impounded fossils belonged to the same animal, and future discoveries revealed that Herrerasaurus was a fierce predator with an arsenal of sharp teeth and claws, a primitive version of T. rex or Velociraptor. Herrerasaurus was one of the very first theropod dinosaurs—a founding member of that dynasty of smart, agile predators that would later ascend to the top of the food chain and ultimately evolve into birds.