Published: February 2011
Evolution of Feathers
The long curious extravagant evolution of feathers
By Carl Zimmer

Most of us will never get to see nature's greatest marvels in person. We won't get a glimpse of a colossal squid's eye, as big as a basketball. The closest we'll get to a narwhal's unicornlike tusk is a photograph. But there is one natural wonder that just about all of us can see, simply by stepping outside: dinosaurs using their feathers to fly.

Birds are so common, even in the most paved-over places on Earth, that it's easy to take for granted both their dinosaur heritage and the ingenious plumage that keeps them aloft. To withstand the force of the oncoming air, a flight feather is shaped asymmetrically, the leading edge thin and stiff, the trailing edge long and flexible. To generate lift, a bird has merely to tilt its wings, adjusting the flow of air below and above them.

Airplane wings exploit some of the same aerodynamic tricks. But a bird wing is vastly more sophisticated than anything composed of sheet metal and rivets. From a central feather shaft extends a series of slender barbs, each sprouting smaller barbules, like branches from a bough, lined with tiny hooks. When these grasp on to the hooklets of neighboring barbules, they create a structural network that's featherlight but remarkably strong. When a bird preens its feathers to clean them, the barbs effortlessly separate, then slip back into place.

The origin of this wonderful mechanism is one of evolution's most durable mysteries. In 1861, just two years after Darwin published Origin of Species, quarry workers in Germany unearthed spectacular fossils of a crow-size bird, dubbed Archaeopteryx, that lived about 150 million years ago. It had feathers and other traits of living birds but also vestiges of a reptilian past, such as teeth in its mouth, claws on its wings, and a long, bony tail. Like fossils of whales with legs, Archaeopteryx seemed to capture a moment in a critical evolutionary metamorphosis. "It is a grand case for me," Darwin confided to a friend.

The case would have been even grander if paleontologists could have found a more ancient creature endowed with more primitive feathers—something they searched for in vain for most of the next century and a half. In the meantime, other scientists sought to illuminate the origin of feathers by examining the scales of modern reptiles, the closest living relatives of birds. Both scales and feathers are flat. So perhaps the scales of the birds' ancestors had stretched out, generation after generation. Later their edges could have frayed and split, turning them into the first true feathers.

It made sense too that this change occurred as an adaptation for flight. Imagine the ancestors of birds as small, scaly, four-legged reptiles living in forest canopies, leaping from tree to tree. If their scales had grown longer, they would have provided more and more lift, which would have allowed the protobirds to glide a little farther, then a little farther still. Only later might their arms have evolved into wings they could push up and down, transforming them from gliders to true powered fliers. In short, the evolution of feathers would have happened along with the evolution of flight.

This feathers-led-to-flight notion began to unravel in the 1970s, when Yale University paleontologist John Ostrom noted striking similarities between the skeletons of birds and terrestrial dinosaurs called theropods, a group that includes marquee monsters like Tyrannosaurus rex and Velociraptor. Clearly, Ostrom argued, birds were the living descendants of theropods. Still, many known theropods had big legs, short arms, and stout, long tails—hardly the anatomy one would expect on a creature leaping from trees. Other paleontologists argued that birds did not evolve from dinosaurs—rather, their similarities derived from a shared common ancestor deeper in the past.

In 1996 Chinese paleontologists delivered startling support for Ostrom's hypothesis. It was the fossil of a small, short-armed 125-million-year-old theropod, Sinosauropteryx, which had one extraordinary feature: a layer of thin, hollow filaments covering its back and tail. At last there was evidence of truly primitive feathers—found on a ground-running theropod. In short, the origin of feathers may have had nothing to do with the origin of flight.

Soon paleontologists were finding hundreds of feathered theropods. With so many fossils to compare, they began piecing together a more detailed history of the feather. First came simple filaments. Later, different lineages of theropods evolved various kinds of feathers, some resembling the fluffy down on birds today, some having symmetrically arranged barbs. Other theropods sported long, stiff ribbons or broad filaments, unlike the feathers on any living birds.

The long, hollow filaments on theropods posed a puzzle. If they were early feathers, how had they evolved from flat scales? Fortunately, there are theropods with threadlike feathers alive today: baby birds. All the feathers on a developing chick begin as bristles rising up from its skin; only later do they split open into more complex shapes. In the bird embryo these bristles erupt from tiny patches of skin cells called placodes. A ring of fast-growing cells on the top of the placode builds a cylindrical wall that becomes a bristle.

Reptiles have placodes too. But in a reptile embryo each placode switches on genes that cause only the skin cells on the back edge of the placode to grow, eventually forming scales. In the late 1990s Richard Prum of Yale University and Alan Brush of the University of Connecticut developed the idea that the transition from scales to feathers might have depended on a simple switch in the wiring of the genetic commands inside placodes, causing their cells to grow vertically through the skin rather than horizontally. In other words, feathers were not merely a variation on a theme: They were using the same genetic instruments to play a whole new kind of music. Once the first filaments had evolved, only minor modifications would have been required to produce increasingly elaborate feathers.

Until recently it was thought that feathers first appeared in an early member of the lineage of theropods that leads to birds. In 2009, however, Chinese scientists announced the discovery of a bristly-backed creature, Tianyulong, on the ornithischian branch of the dinosaur family tree—about as distant a relative of theropods as a dinosaur can be. This raised the astonishing possibility that the ancestor of all dinosaurs had hairlike feathers and that some species lost them later in evolution. The origin of feathers could be pushed back further still if the "fuzz" found on some pterosaurs is confirmed to be feathers, since these flying reptiles share an even older ancestor with dinosaurs.

There's an even more astonishing possibility. The closest living relatives of birds, dinosaurs, and pterosaurs are crocodilians. Although these scaly beasts obviously do not have feathers today, the discovery of the same gene in alligators that is involved in building feathers in birds suggests that perhaps their ancestors did, 250 million years ago, before the lineages diverged. So perhaps the question to ask, say some scientists, is not how birds got their feathers, but how alligators lost theirs.

If feathers did not evolve first for flight, what other advantage could they have provided the creatures that had them? Some paleontologists have argued that feathers could have started out as insulation. Theropods have been found with their forelimbs spread over nests, and they may have been using feathers to shelter their young.

Another hypothesis has gained strength in recent years: that feathers first evolved to be seen. Feathers on birds today come in a huge range of colors and patterns, with iridescent sheens and brilliant streaks and splashes. In some cases their beauty serves to attract the opposite sex. A peacock unfolds his iridescent train, for instance, to attract a peahen. 
The possibility that theropods evolved feathers for some kind of display got a big boost in 2009, when scientists began to take a closer look at their structure. They discovered microscopic sacs inside the feathers, called melanosomes, that correspond precisely in shape to structures associated with specific colors in the feathers of living birds. The melanosomes are so well preserved that scientists can actually reconstruct the color of dinosaur feathers. Sinosauropteryx's tail, for example, appears to have had reddish and white stripes. Perhaps the males of the species flashed their handsome tails when courting females. Or perhaps both sexes used their stripes the way zebras use theirs—to recognize their own kind or confuse predators.

Whatever the original purpose of feathers, they were probably around for millions of years before a single lineage of dinosaurs began to use them for flight. Paleontologists are now carefully studying the closest theropod relatives of birds for clues to how this transition occurred. One of the most revealing is a recently discovered wonder called Anchiornis, more than 150 million years old. The size of a chicken, it had arm feathers with black-and-white portions, creating the spangled pattern you might see on a prize rooster at a county fair. On its head it wore a gaudy rufous crown. In structure, Anchiornis's plumes were nearly identical to flight feathers, except that they were symmetrical rather than asymmetrical. Without a thin, stiff leading edge, they may have been too weak for flight.

What the plumes lacked in strength, however, they made up for in number. Anchiornis had an embarrassment of feathers. They sprouted from its arms, legs, and even its toes. It's possible that sexual selection drove the evolution of this extravagant plumage, much as it drives the evolution of peacock trains today. And just as their long, heavy trains pose a burden to peacocks, the extravagant feathers of Anchiornis may have been a bit of a drag, literally.

Corwin Sullivan and his colleagues at the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing have found a way that Anchiornis could have overcome this problem. In the theropods that were closely related to living birds, a particular wristbone was wedge-shaped, allowing them to bend their hands. Anchiornis's wrist bone was so wedge-shaped that it could fold its arms to its sides, keeping its arm feathers off the ground as it walked. Modern birds use a similar bone in flight, drawing their wings toward their bodies during an upstroke. If Sullivan and his colleagues are right, this crucial flight feature evolved long before birds took wing. It's an example of what evolutionary biologists call exaptation: borrowing an old body part for a new job. It now looks like bird flight was made possible by a whole string of such exaptations stretching across millions of years, long before flight itself arose.

The way in which that final transition occurred continues to inspire lively debate. Some scientists argue that feathered dinosaurs evolved flight from the ground up, flapping their feathered arms as they ran. Others challenge this notion, pointing out that the "leg wings" on Anchiornis and other close relatives of birds would have made for very clumsy running. These researchers are reviving the old idea that protobirds used feathers to help them leap from trees, glide, and finally fly.

Ground up, trees down—why not both? Flight did not evolve in a two-dimensional world, argues Ken Dial, a flight researcher at the University of Montana-Missoula. Dial has shown that in many species a chick flaps its rudimentary wings to gain traction as it runs from predators up steep inclines, like tree trunks and cliffs. But flapping also helps steady the chick's inevitable return to lower terrain. As the young bird matures, such controlled descent gradually gives way to powered flight. Perhaps, says Dial, the path the chick takes in development retraces the one its lineage followed in evolution—winging it, so to speak, until it finally took wing.