Evolutionary biologists have argued for a century about how birds learned to fly. Did they -- or their dinosaur ancestors -- run faster and faster until their primitive wings allowed them to take off? Or did they jump and glide between trees until they learned to fly without branches?

The answer, it turns out, may be both.

In research reported today in the journal Science, University of Montana biologist Kenneth P. Dial described how some poor-flying modern birds use their wings to help them gain traction as they run up steep inclines or even straight up vertical walls.

"Like the way spoilers on the back of a race car push the tires to the pavement," he said. The birds flap their wings to hold them against the climbing surface even as their powerful hind legs propel them upward.

The same strategy is what may have enabled small, feathered dinosaurs to get into the trees and develop the flapping talents that enabled them to evolve into flying birds. The technique marries the two sides of the old debate: It is "bottom-up" in that the animal starts on the ground, but "top-down" in that the animal may have needed to be in the tree before it could learn true flight.

"It's a better compromise between the two extremes," said Luis Chiappe, curator of paleontology at the Los Angeles County Natural History Museum and a bottom-up proponent. "There are people who are always going to stick to extreme ideas, but I'm more inclined to find a middle ground."

Dial said he began the experiments that led to his insight by accident. He was studying barnyard chicks to see how much their tiny wings help them lift their relatively large hindquarters in a vertical jump. When he returned home from a trip, his 13-year-old son, who had been minding the experiments, said the chicks were "cheating." Whatever they were doing to get up the wall, he added, they weren't jumping.

It was true, but the movements were so fast they were hard to follow with the naked eye, Dial said. So he used high-speed photography -- 500 to 1,000 frames per second -- to freeze the chicks in the act: "When you're just watching them, it looks like frenetic fluttering," he said. "But in slow motion it's beautiful to see -- we could see they were running and flapping their wings at the same time."

Dial said he expanded his research to focus on various species of birds known as Galliformes -- chickens, turkeys, partridges, guinea fowl, peafowl, quail and pheasant, among others. Known as "ground birds," they could just as easily be described as "table birds." Their large hind legs make them poor fliers, but provide them with attractive dark meat for dinner.

Dial used chukar partridges for his pivotal series of experiments, testing and filming their ability to run up inclined planes. Fully feathered partridges running on a gritty surface did not use their wings until the slope reached 45 degrees, but, then, by flapping and changing their wing angle on steeper slopes, they were able to run up to a 105-degree plane -- 15 degrees past vertical.

Clipping and then plucking the partridges' most important flying feathers caused a sharp dropoff in performance, an indication that the wings play a crucial aerodynamic function in climbing. Even more important, however, Dial found that the birds could not handle slopes greater than 50 degrees when they ran on smooth surfaces.

"If the wings were giving them lift, the [running surface] shouldn't have mattered," he said. "They wanted to run," and it was clear that the flapping wings were helping the partridges get a better grip on the running surface -- not lifting them off it.

This finding had important implications for evolutionary paleontology, where the debate over the origin of flight had been obscured for years by the equally contentious question of whether birds descended from dinosaurs. The evolution and subsequent extinction of the flying -- or gliding -- reptiles known as pterosaurs occurred independently and has nothing to do with the evolution of birds.

The dinosaur-bird debate still seethes with occasional eruptions, but a consensus has emerged in recent years on the linkage connection, principally because of a rich harvest of feathered dinosaur fossils from China. One example, Caudipteryx, an 18-inch-tall omnivore that lived 125 million years ago, is usually depicted as a bantam rooster-like animal with powerful hind feet and smallish wings.

The problem with the dinosaurs, as with the partridges, said evolutionary biologist Lawrence Witmer of Ohio University-Athens, was that scientists could not figure out how to get the big-hindquartered dinosaur into the trees so it could evolve the big wings and delicate feet of the modern flying bird.

"The tree-down theory has always been appealing, except that scientists have always believed that gliding is an evolutionary dead end," Witmer said. "Flapping is what it's all about, and this study offers a whole new way to look at it. You're not getting lift-off with flapping, but you're getting to heights. These little dinosaurs would have sought trees, either as refuge or as places to hunt. This could be how they did it."

Kenneth P. Dial's research is reported in the journal Science.