WonderQuest:  On the web since 1997...      

Snake legs that walk no more

Q: Do snakes have legs? I heard on the Discovery Channel that they have little bitty legs, like nubs. Is this true? (Rachel, Kankakee, Illinois)

A: Primitive snakes — such as, pythons and boa constrictors — do have nub-like legs beneath their skins and tiny, half-inch claws that protrude out above the nubs but nestle close to their bellies near the anus. Actually, even the nubs are not legs but rather a remnant of upper-leg (thigh or femur) bones. The males still use the spurs — but only during courtship and fighting — not to walk. No other snakes have legs.

Remnant of a hind leg still present in pythons. This electron micrograph shows the left limb bud (red) of a python embryo at 24 days of incubation. The wavy lines are embryonic snake scales. [Martin J. Cohn ©, University of Florida, used with permission]

The story of snake begins long ago (about 115 million years) in the early Cretaceous Period when dinosaurs were flourishing and flowering plants were developing. For a long time, we’ve known that snakes evolved from lizards but how they evolved is still unknown.

A long-standing dispute whether snakes evolved from marine or land lizards smolders on. Blair Hedges and Nicolas Vidal of Penn State University believe they settled the issue from DNA testing of almost every known family of snakes and lizards.

"Our results show clearly that snakes are not closely related to monitor lizards like the giant Komodo Dragon, which are the closest living relatives of the mosasaurs — the only known marine lizard living at the time that snakes evolved," Vidal says.

Land lizards are the ancestors, Vidal and Blair conclude since "all the other lizards at that time lived on the land."

"All lizards are terrestrial," counters biologist Michael Caldwell of the University of Alberta. "The only group to hop out of the water and become truly aquatically adapted is the mosasaurs and perhaps their close relatives, the snakes." Mosasaurs were gigantic aquatic predators with paddle-shaped limbs that apparently died out with the dinosaurs. Indeed, even the mosasaur ancestor was also a "terrestrial land-living lizard — kind of like the proto-whale. The mosasaur ancestor also started out on land and ended up in the sea," says Caldwell.

Hedges agrees that mosasaurs were aquatic lizards that evolved from land lizards. But disagrees that snakes are related to mosasaurs.

Fossil evidence may support the marine hypothesis — namely, that snakes evolved from mosasaurs. Caldwell and others have found fossil snakes that lived about 95 million years ago in a marine environment. Moreover, they had a "complete hindlimb that extended beyond the body wall" and, thus, may bridge the gap between 4-legged mosasaurs and legless living snakes, says Caldwell.

A 95-million year old snake (Pachyophis woodwardi) fossil found in marine rocks at Bosnia-Hercegovina. [Michael Caldwell ©, University of Alberta, used with permission]

Whichever evolutionary path snakes took — by land or by sea — snakes lost their legs.

If snakes came from land lizards, that suggests why limbs disappeared. Proto-snakes might have burrowed underground searching for food in small crevices. Limbs got in the way as they struggled through narrow tunnels. Not just limbs but even more so the "wide shoulders and pelvis" that supported their limbs, says Hedges.

A slender blindsnake (Typhlops biminiensis), a burrowing species — limbless, perhaps, for easier passage through narrow tunnels. [S. Blair Hedges ©, Pennsylvania State University, used with permission]

The proto-snake limbs shrank and eventually disappeared probably because of changes in the way the Hox genes function. The Hox genes, by the way, control vertebrae development in embryos and, furthermore, the pattern for the entire spine from head to tail.

"Examination of limb development in python embryos has revealed evolutionary changes in early limb bud formation that account for limblessness," writes Sean Carroll in his excellent book, Endless Forms Most Beautiful.

Consider first the forelimbs of an embryo. Hox genes within any animal control not only how an embryo develops its body parts but also where the parts sprout and grow. Different Hox genes control different parts.

Long ago, the Hox genes of limbed-python ancestors (lizards) apparently changed. The genes that controlled their formerly modest chest regions, in effect, told the embryo — grow a chest along the entire trunk and up into the head. Those genes expanded their zone of expression (region of influence) from up into the head to almost the proto-python’s tail. All the vertebrae that grow in this region bear ribs, which indicate that these are thoracic vertebrae.

The forelimbs of all four-legged vertebrates sprout at the boundary between neck and trunk vertebrae and that boundary — for the proto python — had, in a sense, shifted to the base of the skull since vertebrae with ribs grow that far. But, more importantly, the sharp neck-chest boundary had been eliminated! The vertebrae at skull base also had features of neck bones.

"So, the landmark for where to put the limb no longer exists" and no forelimbs develop, says Carroll.

A python embryo starts to sprout hind legs but then stops. Martin J. Cohn of the University of Florida and Cheryll Tickle of the University of Dundee investigated why.

It turns out that the python Hox genes fail to activate pathways (the apical-ridge and polarizing-region paths) that signal normal limb development. So growth stops. The hind limb bud lacks "key signaling proteins," says Carroll, describing Cohn and Tickle’s discoveries. This activation failure may "stem from changes in Hox gene expression that occurred early in snake evolution," say Cohn and Tickle.

Further Reading:

Nature: Developmental basis of limblessness and axial patterning in snakes by Martin J. Cohn & Cheryll Tickle. 1999 Jun 3;399(6735):474-9.

Sean B. Carroll. Endless Forms Most Beautiful. New York: W.W. Norton & Company, 2005.

Penn State University: Snake legs were lost terrestrially, S. Blair Hedges & Nicolas Vidal

University of Alberta: Marine snakes of the Tethyan-realm, by Michael Caldwell

(Answered June 10, 2005)