Waiting in the wings

Feathered fossils found in China recently seem to confirm that it was dinosaurs, not reptiles, that were the ancestors of birds. Steve Connor asks if the debate over their origins is finally over

The discovery of a remarkable four-winged creature – a feathered dinosaur – in the rich fossil beds of Liaoning Province in north-eastern China is said to have ended the 140-year-old debate about the origin of birds. Not only does the gliding animal show that dinosaurs were the ancestors of birds, it supports the idea that powered flight was a "tree-down" invention rather than a phenomenon that emerged from a series of "ground-up" evolutionary adaptations. So has the mystery of bird origins been solved?

The discovery of a remarkable four-winged creature – a feathered dinosaur – in the rich fossil beds of Liaoning Province in north-eastern China is said to have ended the 140-year-old debate about the origin of birds. Not only does the gliding animal show that dinosaurs were the ancestors of birds, it supports the idea that powered flight was a "tree-down" invention rather than a phenomenon that emerged from a series of "ground-up" evolutionary adaptations. So has the mystery of bird origins been solved?

It was Thomas Henry Huxley, the Victorian scientist who famously became Darwin's "bulldog" defender, who first suggested that birds were in effect feathered reptiles with a beak instead of teeth, and with three reptilian fingers concealed in their wings. For more than a century it was assumed that birds were the descendants of an ancient stock of reptiles that had learnt powered flight with the help of aerodynamic feathers.

More recently, the argument over avian origins has shifted to a possible dinosaur ancestor. The idea is that birds are living dinosaurs – perhaps their feathers and flying helped them to survive the cataclysm that wiped out the rest of their kin.

In the 1970s, John Ostrom of Yale University carried out a meticulous comparison of the anatomical features of dinosaurs and the oldest known bird, the fossil Archaeopteryx discovered in 1861 at Solnhofen in Germany. Ostrom's work led eventually to the discovery of more than 100 anatomical features that birds shared with the therapods (meat-eating dinosaurs), including a wishbone, swivelling wrists and three forward-pointing toes. Of all the therapods, the swift-running dromaeosaurs seemed most like birds.

In the 1990s, scientists unearthed a series of stunning fossils at Sihetun in China's Liaoning Province. The frenzy started in 1994 when a peasant farmer discovered a beautifully preserved bird named Confuciusornis sanctus. This "sacred Confucius-bird" had small, feathered wings, the earliest known beak and a pygostyle – the reduced bony tail of modern birds. But it still had archaic features such as clawed fingers on each feathered wing.

In 1996, scientists at Sihetun announced the first "feathered" dinosaur, named Sinosauropteryx prima, which means the "first Chinese dinosaur wing". Inside the rib cage of a specimen of the two-legged animal was a pair of eggs ready for laying, and in the stomach of another the bony remains of a small mammal – its last meal. The feather-like bristles ran down its spine from nose to tail, as well as other parts of its body.

Those who thought that birds are living dinosaurs had at last some hard evidence to back up their case. However, critics soon pointed out that the "feathers" of Sinosauropteryx lacked the quill and other structures typical of modern feathers. Either these bristles were very rudimentary feathers, or they were not feathers at all but some other form of body covering that served as insulation.

"Finding Sinosauropteryx provoked a lot of squawking but little resolution," says Pat Shipman, an anthropologist at Penn State University. Liaoning Province, however, still had many other surprises in store. Consisting of volcanic and sedimentary rock, the area has yielded a huge variety of exceptionally well-preserved animals and plants, from fossil fish and birds to shrimp, flowers and mammals dating from between 120 million and 145 million years ago.

At that time the area was dotted with freshwater lakes and volcanoes which sporadically rained down layers of very fine ash. Animals caught in a volcanic eruption would have been preserved in a fine-grained sediment that would have protected some of the smallest and most delicate anatomical structures, such as feathers.

Angela Milner, a dinosaur expert at the Natural History Museum in London (which has a special exhibition of the Chinese fossils), says that the Liaoning specimens are exceptional. "The creatures being discovered at Liaoning Province died quickly after being covered in volcanic dust. Their bodies have been remarkably well preserved," Dr Milner says.

In 1997, a turkey-sized creature called Protarchaeopteryx emerged from the Liaoning excavations. It had long arms and legs and true feathers, which were symmetrical in shape. To be aerodynamically useful, however, feathers need to be asymmetrical, suggesting that those of Protarchaeopteryx were used perhaps for insulation rather than flight.

The real breakthrough discovery occurred in 2000 when local farmers made a remarkable discovery which many scientists described as the true "missing link" between birds and dinosaurs. This "fuzzy raptor" had the bony skeleton of a predatory dinosaur but a coat fringed with feathers. It was clearly a close relative of the meat-eating dinosaurs as well as sharing many characteristics of Archaeopteryx.

Now with the latest announcement in this week's Nature of six specimens of the four-winged "Microraptor gui", the argument appears to be settled. The find supports the idea that birds descended from small, meat-eating dinosaurs that had grown feathers and taken to the trees where they had become adapted to gliding and then powered flight.

Xing Xu of the Institute of Vertebrate Palaeontology and Paleoanthropology in Beijing, the leader of the research team, believes that Microraptor seals the dinosaur origin of birds and proves that flight evolved from animals that took to the trees and subsequently learnt to glide.

He argues that feathered hind limbs would have made it difficult for the creature to live on the ground – its feathers would have quickly become caked in mud. It is more likely that it lived in the trees and used its limbs for gliding from one branch to another. Nevertheless, it is too early to rule out the possibility that Microraptor was able to walk and perhaps to run on the ground using all four of its limbs.

Like many of the big questions in evolution, it is unlikely that everyone will agree. There are still those who believe that birds have a far older origin than suggested by the Liaoning fossils – old enough to put their first appearance firmly with the reptiles. And there are those who still support the idea that avian flight was the result of adaptations driven from the ground up.

Sceptics of the tree-down theory of flying for instance question whether feathered wings could truly have come out of the pressure to adapt to a gliding mode of transport. Modern gliding animals, such as "flying" squirrels and reptiles, have flaps of skin stretching between the fore and hind limbs. Why don't feathered dinosaurs have similar flaps of skin if they were gliders?

Many of these sceptics believe that powered flight is more likely to have resulted from powered running and jumping. This was recently adapted by Kenneth Dial of the University of Montana who saysflapping feathered "proto-wings", which would have been useless for flying, could still provide the necessary downforce on the feet to improve traction for running up steep slopes.

"It turns out that the proto-wings – precursors of wings that birds have today – actually acted more like a spoiler on the back of a race car to keep the animal sure-footed even while climbing up nearly vertical surfaces," says Professor Dial.

An even more contentious issue still surrounds the dinosaur-as-ancestor theory. Archaeopteryx is known to be about 150 million years old, about 25 million years older than the feathered dinosaurs of China. Evidently these particular dinosaurs could not be the ancestor of the oldest-known bird.

The discrepancy in dates has been seized on by those who still believe that reptiles are the true ancestors of the birds. "Just as you can't be your own grandmother, birds can't have come from therapod dinosaurs because the fossil record shows that the time line is all wrong," says Alan Feduccia, of the University of North Carolina.

Working with Russian scientists, Feduccia has identified what he claims to be the earliest fossilised feather dating back to 220 million years ago. It came from a small reptile that probably used feathers for gliding. "These are the earliest structures in the fossil record that can be called feathers. They predate the so-called 'fuzzy dinosaurs' from China by at least 100 million years," he says. These feathers were "unequivocally" used in the context of flight rather than insulation. So, according to Feduccia, birds are really feathered reptiles rather than living dinosaurs.

But Feduccia seems to be in a minority. According to Pat Shipman, the sheer number and variety of feathered dinosaurs discovered in recent years has tipped the balance in favour of a "strong and nearly irrefutable" link between dinosaurs and modern birds. "There are no other contenders for the ancestry of modern birds that are anywhere near as convincing as the smaller, fast-moving dinosaurs," Shipman says. "The key questions now are whether one, or more than one, species of dinosaur gave rise to birds and, if birds had many dinosaurian ancestors, who are they?"