Which Came First: the Dinosaur or the Bird?

The consensus is in: Birds are living dinosaurs. But how that epic evolutionary leap took place remains one of science’s greatest mysteries.

The Structure & Motion Laboratory at the Royal Veterinary College’s Hertfordshire campus, about 25 minutes north of London by train and bus, fills a cavernous rectangular space nearly half the size of a soccer field. At one end of the lab, some students are putting a galloping horse through its paces, measuring the animal’s performance as its hooves pound a treadmill. “Let’s wait until they’re done,” says evolutionary biologist Ashley Heers, “so the horse doesn’t scare the birds.”

The lab is one of the world’s true state-of-the-art facilities, a place where scientists studying biomechanics and motion come to advance their studies of all sorts of animals, from dogs to squirrels. On any given day researchers and students might be dissecting a giraffe or an ostrich, measuring the force a Barn Owl produces when it takes off to pursue prey, or using X-ray video to analyze the wobbling gait of broiler chickens. 

The students lead the horse out a large door, and it’s Heers’s turn to experiment. She climbs the metal stairs to the lab’s mezzanine, fetches five grayish guinea fowl chicks from a wooden box warmed with heat bulbs, and clutches them to her chest. Heers takes the birds to her station along one wall of the lab and pops them into an open cardboard box turned sideways. As the chicks huddle together, looking confused and a little anxious, Heers trains two high-speed video cameras on the box and boots up a computer on a table a few feet away. “Now let’s collect some data,” she says. 

Heers, 31, is a petite woman with a head full of dark curls. She has been watching birds for much of her life. But as a postdoctoral researcher studying avian evolution, she sees them for what most paleontologists think they really are: living dinosaurs, the one surviving lineage of extinct carnivorous beasts like Tyrannosaurus rex. In the fusty world of paleontology, where traditional scientists spend hours at a time on their bellies in the dirt with picks, hammers, and brushes, Heers represents the next generation of researchers, who grew up with Jurassic Park and Google. She is using the latest video and computer modeling technology to study how baby birds develop the ability to fly, looking for clues as to how dinosaurs evolved into birds—one of the most dramatic and successful evolutionary transformations in the history of life on earth.

“Every time I look at a bird, I am seeing a dinosaur,” she says. “That’s what makes it exciting for me.”

Heers came to England fresh from doing a Ph.D. with Kenneth Dial at the University of Montana’s Flight Laboratory in Missoula, Montana, probably the world’s leading research center on bird aeronautics. In the course of her research, she has piled up considerable evidence that baby birds do indeed resemble the avian dinosaur ancestors, which had feathers but little if any ability to fly. Thus at nine days old, her downy guinea fowl chicks have very small wings and shafts of primitive-looking feathers still pervious to the air. Their tufts are simple and symmetrical, lacking the complex structure of adult feathers, and the birds are too young to fly very far. 

Heers takes one of the chicks out of the cardboard box and places it on an inclined pedestal about a foot away. The chick hesitates a second, then gathers up its wings, and its courage, and jumps back into the box. The video footage, captured on the computer, shows that while the chick relied mostly on its legs to make the leap, it also flapped its wings between two and three times, adding just enough lift to successfully bridge the gap. “They can’t yet fly very far,” Heers says, “but those tiny little wings are still very useful.” This strongly suggests, she adds, that even if early dinos evolved feathers for other reasons—to keep warm or for peacock-like sexual displays—they were already on their way to becoming birds.


Until the 1990s most researchers thought that both feathers and flight did not appear until birds first evolved, probably about 150 million years ago. That assumption had held since the early 1860s, when paleontologists began unearthing fossils of Archaeopteryx, widely considered the “first bird,” from a limestone quarry at Solfhofen, Germany. The discovery of Archaeopteryx got Charles Darwin—who just a few years earlier had published On the Origin of Species, his groundbreaking treatise on natural selection—pretty excited, because it seemed to represent a clear evolutionary transition. 

But scientists of the time quickly launched into fierce debates over just what had transitioned into what. Darwin’s colleague and defender, the biologist Thomas Henry Huxley, argued presciently that birds were closely related to dinosaurs, jousting with fellow scholars who insisted they were cousins of reptiles like the crocodiles. The failure of paleontologists to discover more than a handful of other early bird specimens over the next century kept the argument raging despite the lack of evidence to fuel it. Such dramatic finds as the birdlike dinosaur Deinonychus, described by Yale University paleontologist John Ostrom in the 1960s, bolstered the case for dino ancestry but did not clinch it. 

Then, in 1996, a farmer and amateur fossil collector from China’s northeastern Liaoning Province found a one-meter-long dinosaur embedded in two separate limestone slabs, which he sold to two different Chinese museums. Paleontologists quickly put two and two together and realized that the specimen, which they named Sinosauropteryx and dated to about 125 million years ago, was the first discovery of a feathered dinosaur. 

Although a small number of skeptical researchers have since argued that the claimed feathers are really degraded collagen fibers, further discoveries of similarly plumed dinos bolstered the case. Some of them—such as the four-winged Microraptor, first reported from China in 2000—probably could fly (see “Raising the Dead”). To date more than 20 suspected flying dinosaur species have been reported, and there is now an overwhelming consensus that birds represent a lineage of flying dinosaurs that survived the mass dinosaur extinction at the end of the Mesozoic Era, about 66 million years ago. A flood of new genetic research released at presstime confirmed that birds subsequently saw an early, rapid, evolutionary “big bang” that led to the more than 10,000 species we have today.

“There was a seamless transition between dinosaurs and birds,” says Stephen Brusatte, a paleontologist at the University of Edinburgh in the United Kingdom. Brusatte, a slim, handsome rising star who recently completed groundbreaking doctoral work on dinosaur and bird family trees at Columbia University and the American Museum of Natural History in New York City, says dinos were not dim-witted, slow-moving reptilian beasts. Rather, he argues, many dinos could be seen as “overgrown birds,” with efficient, bird-like lungs, keen eyesight and other senses, and bird-like reproductive strategies. Indeed, important new discoveries made over the past year strongly suggest that many, if not all, dinosaurs had feathers. 

Nearly every feathered dinosaur discovered so far has been a meat-eating theropod, cousin to creatures like T. rex and Velicoraptor of Jurassic Park fame. But over the past dozen years the evidence has grown that ornithischians, dinosaurs only very distantly related to the theropods—including horned dinos such as Triceratops and Stegosaurus—also had feathers.

While many paleontologists accepted that evidence, there was still no smoking gun. But that changed four years ago when a team of Russian researchers led by geologist Sofia Sinitsa undertook a geological survey in Siberia’s Kulinda Valley, near the city of Chita, and discovered some dinosaur fossil fragments unlike any they’d ever seen. When they found several complete specimens in the next two field seasons, they called in the expert Pascal Godefroit, a paleontologist at the Royal Belgian Institute of Natural Sciences in Brussels, who is well known for his work with both dinosaurs and early birds. Godefroit was shocked. This new beast had clear feather-like structures over most of its body, even though it clearly was an ornithischian and not a theropod. The team named this creature, which could date to as early as 144 million years ago, Kulindadromeus zabaikalicus when they filed their report last July (adding yet another to the long line of unpronounceable dinosaur names).

The separate lineages that led to ornithischians like Kulindadromeus, and to the theropods, split at least 200 million years ago, fairly soon after the rise of dinosaurs, about 30 million years earlier. The finding therefore points to feathers as an evolutionary innovation dating all the way back to the dawn of dinos.

There are a number of theories about why dinosaurs evolved feathers in the first place, but most researchers agree that it was not so they could fly. In fact, it was probably an evolutionary accident, Brusatte says. “They found themselves with structures on their bodies that had evolved for a different reason.” Darwinian natural selection favors traits that lead to greater reproductive success; given the growing evidence that dinosaurs were warm-blooded, paleontologists theorize that feathers would have helped keep the creatures from losing body heat, and also helped to keep their eggs warm. (Indeed, finds of dinosaur skeletons, such as that of Oviraptor and Citipati, brooding on their egg nests, show that bird-like reproduction was an early evolutionary innovation.) 

Another possibility, which has been raised by the finding of signs of pigments in fossils of numerous feathered dinosaurs, is that they sported multicolored feathers perfect for sexual display, an early version of the peacock’s colorful ensemble. Lawrence Witmer, a paleontologist at Ohio University in Athens, says that such feathers might have served as “showy flags” to attract females, and that only later in the dino-bird transition did smaller, lighter species exploit the plumes’ aerodynamic properties.  

Evidence for the hypothesis that dinosaurs sprouted features for sexual display was reported in 2013 by Natural History Museum of Los Angeles County paleontologist Luis Chiappe and his colleagues. That team looked at fossils of Confuciusornis sanctus, a 125-million-year-old bird, of which hundreds of fossils have been found in northeastern China. Researchers have been puzzled that some C. sanctus specimens have long tail feathers and some don’t. Many scientists had assumed that the long-tailed birds were males, though there was no direct evidence for this until Chiappe and his colleagues found a type of bone called a medullary—which is present in the limbs of female, egg-laying birds and stores calcium that helps form eggshell—in a specimen of C. sanctus that did not have tail feathers. This provided “undisputed evidence,” they argued, that long tail feathers were a male rather than a female attribute, a conclusion that, if correct, would support the sexual display hypothesis.


Ashley Heers is focusing her work directly on how and when dinosaurs took flight. The daughter of a veterinarian and a vet technician who grew up in the central California town of Tulare, she was a high school student when she took an online course in geology at the University of California-Davis and wrote an essay about Archaeopteryx. That got her hooked on birds, past and present. She did her senior thesis at Davis with paleobiologist Ryosuke Motani. Heers studied how the center of mass, critical to flight ability, varies in living birds. “She was a star student here,” Motani recalls. “When I first met her, her eyes were shining like those of a child.”

Later, as a graduate student at the flight lab in Montana, Heers teamed up with Kenneth Dial to investigate how the transition from dinosaurs to birds might have happened. In 2003 Dial had published evidence that living birds engage in what he called “wing-assisted incline running.” Using Chukar partridges as his model, Dial demonstrated that while the chicks of this species can use their legs to climb tree branches and other steeply inclined surfaces almost immediately after hatching, they help themselves by flapping their wings and creating aerodynamic force— similar to Heers’s later findings with the leaping guinea fowl.

Heers went from there, focusing her Ph.D. work on how the developmental changes in feathers, wings, and body structure transform a baby bird that cannot fly into an adult that can easily take to the air. In one set of experiments, Heers attached Chukar wings from birds of various ages to a revolving apparatus mounted on top of a force plate, an instrument that can measure both the aerodynamic lift and the drag produced by the wings. The whole setup resembled an old airplane motor with the Chukar wings as the spinning propeller, and allowed Heers to closely track how baby birds become better flyers as they grow up.

She found that chicks and juvenile birds, whose feathers are fairly simple—they’re relatively symmetrical and permeable to the air—can’t muster anywhere near the lift that adult birds, with their stiff, asymmetrical feathers, can easily accomplish. And chick wings also produce much more drag, the enemy of soaring flight. In her doctoral thesis, Heers demonstrated that feathered dinosaurs had gone through similar stages during their evolution into birds capable of fully powered flight.

The key to these insights, Heers says, was the realization that living adult birds provided only limited clues to how flight had evolved. “They are so different from early dinos. Most of the anatomical adaptations that we think of in adult birds are conspicuously absent” in their dino ancestors. Those ancestors instead resembled baby birds both in their wing and feather structure and their skeletons, which were flexible rather than partly fused and stiff, as in modern adult birds.

But when Heers first presented her results at the 2010 meeting of the Society of Vertebrate Paleontology in Pittsburgh, she was met with a mixed reception. Some paleontologists in the audience fired skeptical questions at her, and she feared that her scientific career might be over before it began.

Dial wasn’t at that meeting, but at the following year’s conference, he gave an ardent defense of his graduate student and her work. So Heers kept at it, presenting new results at each year’s meeting and publishing papers with Dial and others, despite sometimes harsh reviews from scientists who thought that experts in living birds shouldn’t be meddling in avian evolution.

Today these ideas have won new respect and credibility, especially from leading paleontologists. University of California-Berkeley researcher Kevin Padian, long a fan of Dial’s and Heers’s work, says the initial skepticism was totally off base: “This was exactly what was needed.” 

Brusatte adds, “Their work has been groundbreaking. They are studying living birds, and don’t just speculate based on fossils.” As for Heers, he says, “She’s brilliant. She’s not a traditional paleontologist but a real ornithologist.”

The windows of Mark Norell’s office at the American Museum of Natural History provide stunning views of Central Park. But what really excites Norell, the museum’s curator of paleontology, are the fossils in the white cabinets that line his office. “Look at this beautiful specimen of Alioramus,” he says, pulling open a drawer to reveal the skull of a small theropod dinosaur that lived about 70 million years ago. It was discovered in Mongolia and is a close relative of T. rex. “You are looking at the beginnings of the modern avian brain.” 

Earlier, more primitive theropods such as Allosaurus, which lived about 150 million years ago, “have a more linear brain,” Norell says. “In Alioramus you are starting to get the more S-shaped brain typical of things that are closer to living birds.” That S-shape, Norell and other researchers have found, is due to a “hyperinflation” of the front part of the brain as well as a reorganization of the entire organ, both of which are responsible for the keen eyesight and intricate muscle coordination needed for flight. Norell opens another drawer, revealing the fossilized remains of Gallimimus, another theropod that looks something like an ostrich and is about the same age as Alioramus. “That brain is equally avian-like,” he says.

In a series of recent studies, Norell and his colleagues have argued that the evolution of wings and feathers were not the only innovations that gave rise to the birds. Compared to reptiles, birds have brains that are relatively large for their body size, which could explain why many of them are so smart. Corvids, such as crows and jays, for example, make and use tools, and may even have a rudimentary ability to know what other birds are “thinking.” But Norell and other researchers believe that the avian brain originally evolved to deal with the complexities of flight, which requires a level of neurological and muscular coordination far beyond that of most land-dwelling creatures.

“If you look at bats, or really advanced pterosaurs [prehistoric flying reptiles], they have brains that are similar to those of birds,” Norell says. “Maybe there is something about flight; when you go from a 2D world to a 3D world, you fundamentally reorganize parts of your brain.” Brusatte, a former student of Norell’s, says the transformation from dinosaurs to birds involved the evolution of a “flight-ready” brain. He compares it to the central processor of a computer, and the feathers and wings to its peripheral components. “You can evolve the monitor and the keyboard and the mouse, but if you don’t have that processor, you aren’t going to have a computer.”

Yet while being a bird meant evolving a bigger brain, it also meant getting smaller overall—a trend that started back in the early days of dinosaurs. In papers published last year, two research teams independently reached the conclusion that although many dinosaurs—like the 115-foot-long, 200,000-pound Argentinosaurus—got bigger over the course of their time on earth, other dinosaur lineages began downsizing early on. One such group was the maniraptorans, which include Velociraptor, and to which birds also belong. Of course, if an animal is smaller, it is more vulnerable to predators, and Heers thinks that’s where the evolution of feathers and wings really got momentum. “Regardless of what feathers originally evolved for, if they have even a little bit of surface area and flap their forelimbs, that’s going to be an advantage when they try to escape,” she says. And as a rule, she adds, “small animals are better flyers.”

Heers finished up her work in England last spring, perfecting her skills in biomechanical computer modeling, and then spent part of the summer at Brown University in Providence, Rhode Island, taking the X-ray videos of the Chukars in preparation for the next phase of her career: a postdoctoral research position with Norell. In this new role, she will use computer modeling to virtually turn birds into dinosaurs and back again, altering their body plans, their wing dimensions, and their feather structures to see how changes in these vital components led to the evolution of powered flight and made birds so successful. “I’m going to start out with Archaeopteryx and Confuciusornis,” Heers says, then work backward in time, tracing some of the first steps that dinos took on the road to becoming birds.

Today, instead of telling Heers she’s wasting her time, most paleontologists are cheering her on. The dramatic evolutionary transition she is unraveling, Brusatte says, led to “something incredibly successful and totally new.” And Heers’s work, he suspects, might just show that “dinosaurs were birds all along.”

Correction: A previous photo incorrectly identified a Coelophysis as an Archaeopteryx. We have replaced it with a photo of an Archaeopteryx.

Editor’s Note: Read more about the relationship between dinosaurs and birds from Audubon Magazine: Raising Dinos From the Dead and Did Dinosaurs Have Feathers?