The new study, led by Andrea Cau of the OPHIS Paleontological Museum and Herpetological Center in Italy, supports the second idea.

The team's analysis identified 194 anatomical changes across microraptor evolution. About 30% of those changes also appeared in the bird lineage. The shared features include:

• Fusion of bony projections onto the ribs, producing a rigid torso
• Forelimb bones relatively robust compared with the hind limbs
• An ulna — the main weight-bearing bone of a wing — longer than the humerus

Crucially, when the researchers compared N. barsboldi with other fossils, they found that the two groups acquired these shared features in a different order. That sequence difference points to separate evolutionary paths.

"This demonstrates that microraptors acquired traits convergent with the early evolutionary stages of true birds, albeit through a distinct evolutionary pathway, supporting the hypothesis that flapping flight evolved independently in non-avialan dinosaurs and birds," said Rui Pei, a paleontologist at the Institute of Vertebrate Paleontology and Paleoanthropology at the Chinese Academy of Sciences, who was not involved in the work.

"I'm very happy that they've come to the conclusion that there were probably independent origins of flight amongst various groups," Hartman said. "I strongly agree with them."

How did a four-winged dinosaur fly?

Many features of microraptors resemble those of modern flying birds. "It is reasonable to expect that these non-avialan theropods were capable of behaviors comparable to those of modern avian species," Pei said by email.

Hartman suggested these animals combined gliding with flapping. A large, well-ossified sternum, or breastbone, could have anchored sizeable pectoral muscles. "So they probably have a pretty solid downstroke, which is what generates the thrust to keep it going forward," he said.

The hind wings likely sat directly beneath the body, stabilizing flight, acting as brakes and enabling rapid, tight turns. "These are clearly little predatory buggers," Hartman said. "If you're leaping onto things or dropping onto things, that kind of control would be pretty important."

Climbing trees squirrel-style was probably out of the question, since their forelimbs lacked the necessary range of motion. Instead, microraptors may have taken off by running down slopes and leapt from hills or cliffs rather than gliding down from trees, Hartman said.

The wings might have served other purposes too, such as signaling fitness to potential mates or warming eggs on a nest.

How solid is the conclusion?

The findings rely on one exceptionally preserved specimen plus comparisons with other fossils, and the study covers a single lineage within a diverse group. Still, the independent experts contacted for the story found the convergent-flight argument convincing, and the pattern of shared-but-reordered traits offers a concrete mechanism for how two dinosaur groups could each arrive at flight on their own.