New bat-like dinosaur had wings. Scientists don't know why.

For 160 million years, the remains of a scrappy little creature with features no one has yet seen in a dinosaur fossil remained hidden in long-buried lake sediments in China's Hebei Province.

Now, researchers there have unveiled the creature – a bird-like dinosaur that sported what appears to be bat-like wings or perhaps a parachute-like structure. They have dubbed it Yi qi, or "strange wing."

Whatever the odd airfoil represents for Jurassic air travel, it's turning heads among dinosaur specialists.

"It's been a while since I've been really surprised by the anatomy of a dinosaur," says Thomas Holtz Jr., a paleontologist at the University of Maryland at College Park. "But this one did it.... We definitely need more specimens of this thing to figure out what's going on."

Yi is not an ancestor of modern birds, says Dr. Holtz, who was not a member of the team making the discovery. But it still may say something about the evolution of flight in birds.

Once small, feathered theropods are able to climb trees, "you can start experimenting with what you're doing up in the trees. This represents one experiment that didn't go very far."

The creature, described in Thursday's issue of the journal Nature, appears as a partial skeleton, essentially from the waist up. Back in the day, it would have been a Chihuahua among theropods – about a foot long from snout to tail and would have weighed a bit less than one pound.

Based on Yi's general traits, a research team led by Xing Xu and Xiaoting Zheng, with Linyi University in Linyi City and Corwin Sullivan at the Chinese Academy of Sciences in Beijing assigned it to a group of diminutive, feathered meat-eaters known as scansoriopterygids.

Only two other species within this group have been identified, also from partial skeletons. Based on the evidence, some paleontologists hold that the the creatures lived in trees, using their claw-tipped fingers to grip as they climbed or moved along branches. Their long tails could have helped them keep their balance. Some researchers hold that scansoriopterygids are the first non-bird dinosaurs to exhibit clear adaptations for climbing or living in trees.

Now, Yi makes three. Some of its fossilized bones are connected to each other as they would have been when the creature lived. But all the team had to work with were a few ribs, the neck and skull, and, of course, the two arms and hands.

Like its two relatives, Yi's hands had three fingers, with each hand's third finger longer that the other two. This helped separate scansoriopterygids from other maniraptors, a lineage of dinosaurs that includes the direct ancestors to birds.

Yi's third finger, however, was unusually long. And what put it in a league of its own was a slender length of bone or cartilage extending away from the wrists. Each of these slender bones were about as long as each hand's third finger.

Moreover the researchers detected sheets of soft, membrane-like tissue around the unusual, extended bone and around the fingers of both hands, as well as filament-like feathers.

The oddly placed bone sprouting from the wrist remained a mystery until one of the team members came across a reference to similar structures jutting from the wrists and elbows of flying squirrels.

That led to an aha moment: Yi must have had primitive wings, or at least some form of airfoil.

But did Yi fly or glide, and if so, how?

Based on one partial skeleton, it's hard to tell, suggests Kevin Padian, a paleontologist at the University of California at Berkeley, who was not a member of the team reporting the discovery.

"We can eliminate flapping because you need a very competent air foil," one that is strong enough and moves with enough precision to help produce lift, he says. Without some form of reinforcement – fingers in a bat wing or robust feathers, for instance – a membrane of skin won't do. The fossil provides no evidence for reinforcing.

It also appears gliding is out of the question, pending the discovery of more-complete specimens. Based on a reconstruction of Yi's form, the wings' center of lift would be too far forward of the animal's center of gravity. It could do little more than stall on launch – unless the wings had a swept-back shape.

Even the presence of the membrane doesn't help answer these questions because there isn't enough to give a sense for the full shape of any potential airfoil, Holtz adds.

"This puts us in a quandary," Dr. Padian says. "It looks like it ought to be involved in something like gliding, or parachuting, or something. The problem is that the rest of the animal doesn't look like it's built for this at all."