Snakes lost their legs on land, not in the sea as previous scientists thought, according to new research by professor of biology S. Blair Hedges and postdoctoral fellow Nicolas Vidal.

"Everyone agrees that snakes evolved from lizards and are close relatives to lizards," Hedges said. "What is controversial is the specific group, whether on the land or in the sea."

Scientists estimate that snakes evolved 150 million years ago, Vidal said. For many years, they believed in the terrestrial or land theory, which says that snakes evolved from lizards that burrowed underground.

Then, seven years ago, the discovery of a fossilized marine snake with tiny hind legs fueled the marine hypothesis that snakes lost their legs in the sea. The marine snakes called mosasaurs are now extinct, but their closest cousins, the monitor lizards such as the Komodo Dragon, are still alive today.

Hedges and Vidal chose two genes and compared them in lizards and snakes to see how closely they are related. From statistical analysis of these gene sequences, they created a family tree.

"These two genes have accumulated different mutations in the different species," Vidal said. "These differences between the species are used to build the evolutionary trees. Snakes are not closely related to the monitor lizards, which means that they cannot have marine origins."

Hedges and Vidal's research is significant because of the comprehensive scope of the snakes and lizards studied. They collected DNA from 64 species representing all 19 families of living lizards and 17 of the 25 families of living snakes, Vidal said.

"We mostly acquired DNA from the soft tissues, blood, shed skin, scales and tips of the tails of the animals," he said. "A very small amount is needed, so we can get the sample without killing the animal."

Hedges and Vidal acquired samples from other colleagues, their own collections, fieldwork in South America and pet stores that trade exotic animals, Hedges said.

Some of the samples were difficult to acquire because some snakes and lizards are limited to certain parts of the world, such as remote islands, but it was crucial to their study.

"We felt it was important to analyze genes from all the lizard groups, because almost every lizard family has been suggested as being the one most closely related to snakes," Hedges said. "If we had failed to include genes from even one of the lizard families, we could have missed getting the right answer."

A possible theory to explain why snakes dropped their legs on land is that limbs are not conducive to their burrowing habits, Vidal said.

"The idea is that if you have limbs sticking out, it's difficult to get into small openings," Hedges said. "It's more advantageous, because it enables snakes to get more food and offers safety from predators."

Hedges and Vidal did not find the closest lizard relative to snakes, but they did succeed in eliminating the monitor and marine lizards as possibilities.

"This study doesn't answer all the questions," Hedges said. "In most fields, questions aren't answered all at once. They're answered in increments, and in this study, we've ruled out the marine theory, but that still leaves a lot of possibilities."

Hedges and Vidal said they hope to find the closest relative to snakes in the next year.

"The work by Dr. Hedges and Dr. Vidal is a tremendous contribution to the body of knowledge," said Bryan Fry, deputy director of the Australian Venom Research Unit at the University of Melbourne in Australia. "Their data provides strong evidence for the terrestrial loss of limbs."

A paper of Hedges and Vidal's research will be published in the May 7 issue of Biology Letters, a journal published by the Royal Society in the United Kingdom.