Flight: a bird hand is worthy in the bush.

By examining the claws on the most famous fossil animal, an ornithologist has scratched open an old and bloody debate over how birds developed the ability to propel themselves through the skies.

Most paleontologists in recent years have backed the theory that birds evolved from leathered dinosaurs that ran along the ground jumping after insects. But the new work suggests that the earliest known bird had claws designed to perch in trees, countering the "ground-up" theory for the origin of flight, says Alan Feduccia, who reported his work in the Feb. 5 SCIENCE.

The ancient bird at the center of this debate is Archaeopteryx, a crow-size creature that lived 150 million years ago and is the oldest known bird. Paleontologists discovered the first Archaeopteryx fossil in 1861 and have since unearthed five other specimens. All come from a particular site in southern Germany, where they were preserved in exquisite detail by a fine-grained limestone prized for its use in lithography

Archaeopteryx has captivated succeeding generations of scientists because the specimens show feathers and other structures not normally preserved in fossils.

Feduccia, an ornithologist at the University of North Carolina at Chapel Hill, began his study by analyzing the claw geometry of nearly 300 modern birds from 30 different species. He found that the degree of claw curvature is a clear indicator of lifestyle: Ground-dwelling species have straighter claws, perching birds have moderately curved claws, and birds that climb tree trunks have very curved claws. Almost all modern birds have claws only on their feet.

Comparing Archaeopteryx claws to the modern examples, Feduccia found that the oldest bird did not fit into the ground-dwelling group. Instead, the foot claws of Archaeopteryx most closely resemble those of perching birds, while the hand claws of Archaeopteryx matched the foot claws of tree-climbing birds.

"1 think the evidence is overwhelming that Archaeopteryx was an arboreal [tree-dwelling], flying bird. And once you show that, there is a crack in the dam of the entire ground-up theory of arian origins," Feduccia says.

The new work will surely rekindle the smoldering conflict over the origins of birds and flight. Feduccia and many other ornithologists subscribe to the long-standing theory that birds evolved from primitive tree-dwelling reptiles that used leathered limbs to break their fall while leaping from branch to branch.

Many paleontologists, however, support a theory championed 20 years ago by paleontologist John H. Ostrom of Yale University Ostrom, who identified a previously unrecognized specimen of Archaeopteryx, proposed that the early bird had legs, feet, and claws best suited for running along the ground rather than perching. He also observed that Archaeopteryx shared substantial anatomical similarities with the small bipedal dinosaurs that lived at the same time during the late Jurassic period.

Putting the two observations together, Ostrom proposed that birds evolved from small theropod dinosaurs that had evolved feathers. Although first used for insulation and for display, the feathers could have gradually helped these running bipeds jump and trap insects.

As paleontologists have solidified the link between birds and theropods, the dinosaurian origin theory has come to dominate its competitor in the scientific journals. Meanwhile, the two opposing camps have entrenched their positions, and the argument has turned personal, with researchers attacking one another in newspaper and magazine articles.

Avoiding the vitriol, Ostrom praises Feduccia's study even though it disproves Ostrom's contention that Archaeopteryx' claws were poorly designed for perching. "I think Alan has put together a very solidly based study, I'm not set in concrete," says Ostrom. While he grants that Archaeopteryx could probably perch in trees, Ostrom maintains that the bulk of the anatomical evidence indicates Archaeopteryx could run well and had evolved from running animals.