Birthplace breeders: look homeward, turtle.

Birthplace breeders: Look homeward, turtle

It might seem a simple task to track a reptile's journey from birthplace to mating site. But not when the creature is the green turtle, which doesn't start breeding until age 30. Researchers have tried to monitor its meanderings by tagging it with metal disks or wires, but the turtle's dramatic growth over the decades--from 4-inch infancy to 4-foot adulthood -- has stymied efforts to keep the tags in place.

So Brian W. Bowen, an evolutionary geneticist at the University of Georgia in Athens, turned instead to natural markers, analyzing mitochondrial DNA from eggs and hatchlings at four green-turtle breeding sites in the Atlantic and Caribbean. In the May 11 SCIENCE, he and his co-workers report that turtles from the four breeding sites tended to differ slightly in genetic sequence. The researchers note that genetically distinct populations of green turtles would complicate efforts to preserve this endangered species, since each subgroup would be unique and irreplaceable.

They say their finding lends credence to the "natal homing" theory, proposed in the 1960s, which holds that while turtles born in different regions may share common feeding grounds away from home, the animals part company at breeding time, each swimming hundreds or thousands of miles to breed and nest at its own birthplace. At the same time, the new work undercuts a competing theory, known as the social facilitation model, which contends that virgin female turtles randomly follow experienced breeders to a nesting site regardless of their birthplace. Such "social mixing" must be rare in green turtles, Bowen concludes, because widespread interbreeding of diverse turtle groups at each nesting area would have long ago smoothed out the genetic differences he and his colleagues detected.

Bowen, with co-workers John C. Avise of the University of Georgia and Anne B. Meylan of the Florida Marine Research Institute in St. Petersburg, offer several caveats to the interpretation of their findings. They note that turtles from two of the nesting sites--Florida's Hutchinson Island and Costa Rica's Tortuguero sanctuary -- had indistinguishable mitochondrial DNA sequences, a possible indication that some social mixing might have occurred between these two groups, or that the DNA assay was not sensitive enough to detect extremely subtle differences. In addition, they report that one of eight study turtles from Aves Island, off Venezuela, showed the same genetic pattern as the Tortuguero and Hutchinson turtles in their sample. And because turtles from the fourth breeding site, the remote Ascension Island in the South Atlantic, are not known to share feeding grounds with other animals, their distinct DNA pattern may reflect habitat isolation rather than an inherent avoidance of social mixing.

Zoologist David W. Owens of Texas A&M University in College Station, who helped develop the social facilitation model in 1982, observes that while the new study does not conclusively rule out his hypothesis, it does "seem to indicate there may be three distinct green turtle populations, even within the Atlantic-Caribbean regions."

A more recent study by Bowen's group strengthens that finding and broadens its geographic scope, Meylan told SCIENCE NEWS. Their preliminary analysis of a worldwide survey appears to indicate that most regional populations of green turtles are genetically distinct and return to the birthplace at nesting time, she says.

Such knowledge may influence future conservation efforts, Meylan asserts. "People think it's no big deal if one population of green turtles is wiped out, because it can be replenished by a neighboring rookery up the coast," she explains. "We're saying you can't expect that to happen. Biologists now think of the green turtle as one species, but there may be a number of different entities."