Snake Ancestors Lost Limbs on Land, Study Says

John Pickrell
for National Geographic News
February 11, 2004
Since the 19th century naturalists have debated whether the ancestors to modern snakes lost their limbs at sea or on land. Recently the discovery of early marine fossil snakes with tiny hind limbs reignited the controversy.

In addition, new genetic evidence may prove that snakes evolved on land.

"The big question has been, why would a four-legged animal lose its limbs and develop an elongated body?" said evolutionary biologist S. Blair Hedges of Pennsylvania State University. Hedges is co-author of the first reptile DNA study to include all living families of lizards and many snakes. That new analysis adds to mounting evidence that snakes are not descended from giant extinct marine reptiles.

"The jury is definitely still out on whether living snakes had a terrestrial or marine origin," commented Rick Shine, a snake expert at the University of Sydney in Australia. "But this new study is exactly the kind of research we need to clarify the evidence."

Sink or Swim?

One long-held theory is that snakes are closely related to some group of terrestrial lizards and lost their limbs on land. Many burrowing animals, from weasels to worm lizards, have smaller limbs today. "For animals wiggling around in small holes and crevices, it makes sense that limbs would get in the way," Hedges said.

Today many primitive snakes live in soil or leaf litter. Furthermore, many burrowing animals, such as the blind mole rat, have lost their eyes in the course of evolution. Snakes' eyes are so unusual that many experts believe they re-evolved from those of burrow-dwelling ancestors with much-reduced vision.

Another idea, first formally suggested by Victorian fossil hunter and evolutionary biologist Edward Drinker Cope (1849-1897), argues that snakes lost their limbs at sea and are closely related to the extinct marine lizards called mosasaurs.

Despite pronounced differences (body size for example), snakes share some skeletal features with mosasaurs, such as a loose jaw that allows the swallowing of large prey. Contemporaries of the last dinosaurs, paddle-limbed mosasaurs inhabited seas of the Cretaceous period, which ended about 66 million years ago. Mosasaurs grew up to 60 feet (18 meters) in length. Under this scenario, snakes are also related to mosasaurs' undisputed living ancestor, the fearsome monitor lizard.

Missing Link

For many decades the burrowing-ancestor theory had held sway. But in 1997, researchers described several marine snake fossils with tiny hind limbs. The discovery re-invigorated the marine theory. Paleontologists behind some of these fossils believed they were the oldest snakes yet discovered and a missing link with mosasaurs.

To arbitrate in the debate, Hedges and a colleague at Pennsylvania State University, evolutionary biologist Nicholas Vidal, compared the genetic similarity of 64 species, including all 19 families of living lizards, and 17 of the 25 families of living snakes.

Hedges said that while it's impossible to compare the DNA of mosasaurs with living snakes and other lizards, it is possible to compare the DNA of mosasaurs' surviving, undisputed relatives, the monitor lizards.

In the most comprehensive such study yet attempted, the pair collected difficult-to-obtain lizard DNA from sources as varied as the pet trade to remote research outposts of far-flung tropical islands. Using statistical methods to compare differences in the sequences of two genes—RAG1 and C-mos—among these species, the researchers demonstrated that monitor lizards are not directly related to snakes.

Rather, monitors, such as Indonesia's fast-moving predatory Komodo dragon (Varanus komodoensis), are closely related to other groups—including the glass lizards of Europe and North America—not believed to be snake relatives. The findings were recently published online in the science journal Biology Letters.

Because all lizards, apart from mosasaurs, are thought to have been land-dwelling when snakes evolved around 150 million years ago, "our study provides strong evidence that snakes evolved on land, not in the ocean," Hedges said. He noted, however, that further analysis is required to confirm which lizard group is most closely related to snakes.

More Data

"The new study is novel and interesting … [and] the most comprehensive yet," commented Olivier Rieppel, chair of the department of geology at the Field Museum in Chicago, Illinois. He noted that some of the proposed relationships in the study were unusual, a problem which might be resolved if a larger number of genes were considered.

"Vidal's and Hedge's analysis is a refreshing new insight on the subject," commented Hussam Zaher, evolutionary biologist and snake expert with the University of Sao Paulo in Brazil. "Their results are relevant and seem to reject the marine theory," he said. Zaher cautions that while the study is the most complete in terms of number of lizard families reviewed, he would prefer to see a larger number of the 4,700 or so lizard species examined in future analyses.

Zaher was one of a team who recently re-assessed the proposed "missing link" fossil snakes with tiny hind limbs. His team found compelling evidence that in fact these snakes are a side-branch of snake evolution, closely related to modern boas and pythons and not mosasaurs.

The National Science Foundation has recently awarded a Tree of Life grant to researchers at the Field Museum to study the reptile family tree. The upcoming study should, Rieppel said, "generate large amounts of further data to be brought to bear on the question of snakes' origins."

© 1996-2008 National Geographic Society. All rights reserved.