Beetle wings are often hidden. Nestled behind armored shields on the beetle's back, they unfurl in whirring sheets, whisking their clumsy owners from danger. Beetles don't have more than two sets of wings — unless they're in Yoshinori Tomoyasu's lab.

In research published in the Proceedings of the National Academy of Sciences, Tomoyasu and his co-author, David Linz, genetically engineered beetle larvae with wings on their abdomens, part of an attempt to unpack one of evolution's greatest mysteries: how insects gained the ability to fly.

Insects took to the empty skies between 300 and 360 million years ago, long before birds, bats or pterosaurs. Wings allowed them to conquer new habitats and ecological niches, and establish themselves as one of the most diverse and successful animal classes.

The vast majority of living insects either have wings or evolved from flying ancestors, said Linz, an evolutionary biologist now at Indiana University. "When the average person thinks about an insect wing, they think about a dragonfly — these two pairs of really pretty, long wings," he said. "But it's different in different lineages."

There's a frustrating lack of fossil evidence from the period when insect flight evolved, said Tomoyasu, an evolutionary biologist at Miami University.

"With the flight wing in vertebrates, there's a clear origin," he said. But insect wings evolved so long ago, "it's hard to tell what happened."

According to Floyd Shockley, an entomologist at the Smithsonian's National Museum of Natural History, there have long been two competing hypotheses. The "tergal hypothesis" suggests that wings originated on the tergum — the top of the insect body wall — perhaps as gliding membranes. The "pleural hypothesis" argues that wings were created from ancient leg segments.

The rise of evolutionary developmental biology, along with advances in genetics, has lent weight to a third possibility, Linz said. Originally proposed in 1974, the "dual origin" hypothesis suggests that insect wings began with a fusion of the two separate tissues: the dorsal body wall provided the membrane, while its articulation arose from leg segments.

Tomoyasu and Linz worked with Tribolium, or flour beetles, a common subject because of its fully sequenced genome. The team used master switches in the beetles' genome to manipulate which segments of the body had wings. To their surprise, doing so disrupted portions of anatomy that had seemed unconnected to flight.

This offered some support for the idea that wings were composite tissues. To add support for the dual origin hypothesis, Linz said, evolution would have had to fuse a structure on the dorsal region and one from the pleural tissue.

The team introduced a fluorescent protein into the beetles that marked the expression of certain wing-related genes, making it easy to tell which tissues were being affected by genetic tampering. After manipulating genes of the abdomen, they saw two green tissues: one at the dorsal gin-trap, and one down in the pleural tissue.

And by doing so, they were able to produce pupae in which both tissues fused to form pairs of tiny wings.

The debate about how insect wings evolved is far from over, Tomoyasu said. "We're still relying on one species," he said, which means the results "could be unique to this lineage."