Creatures that lack legs tend to live in water, where locomotion is less restricted. Yet snakes learned to slither on land, not in water, according to a new study.

Using a combination of genetic data and fossils, researchers at Yale University reconstructed the common ancestor to all modern snakes. Their work, published May 19 in the journal BMC Evolutionary Biology, provides a new line of evidence that snakes lost their legs while living on land.

"People have debated snake origins for a long time, but this is the first time that we've done a comprehensive analytical study using computational methods to look at genetic, morphological and fossil data to look at how the early evolution of snakes progressed," study author Alllison Hsiang said in an interview. "Our study supports the idea that snakes evolved on land and not in the sea."

This first snake likely dragged its tiny, nonworking hindlimbs through dense vegetation such as in a rainforest. These hindlimbs had ankles and toes, though their bones were reduced in size to the point that they were no longer functional.

"These are a sign that these animals evolved from something that did have working hindlimbs," says Hsiang. "But since they no longer needed to use them, along their evolutionary trajectory these structures were lost or reduced."

Modern snakes are able to consume enormous prey relative to their own body size because of their specially constructed jaws. In order to for a boa constrictor to eat a deer, for example, or any other prey larger than its head, it must unhinge its jaws.

"But what we find is that the ancestral snake didn't have this ability - it evolved much later in the snake's evolutionary history," says Hsiang.

WIthout the ability to unhinge its jaws, the ancestral snake probably snacked on rodents and other small mammals. The study suggests that it was also likely a nocturnal hunter.

Hsiang emphasized the importance of looking at both fossils and genetic data in evolutionary biology.

"People would look at fossils of snakes and make a hypothesis about what's going on, but this is the first time that we're bringing analytical methods to bear on this data," Hsiang says. "I think it sets a precedent for really trying to answer these questions using computational tools rather than just making educated guesses."

The article in BMC Evolutionary Biology is available here.