Public Release: 

Study shows top predator makes prey population vulnerable in catastrophe

Washington University in St. Louis

Darwin himself didn't quite state it this way, but it must have crossed his mind that "when the cat's away, the mice will play." Now, biologists at the University of California, Davis, and Washington University in St. Louis have completed a unique study of lizard populations on tiny islands in the Bahamas that shows what happens when a natural catastrophe devastates both the cats and the mice.

On islands devoid of cats, the mice rebound more quickly. Thomas W. Schoener, Ph.D., and David A. Spiller, of the University of California, Davis, and Jonathan B. Losos, Ph.D., professor of biology at Washington University in St. Louis were studying the effects of a large predator lizard species, the curly-tailed lizard, on both its prey, a smaller species called the brown anole, and on the entire food chain on 12 baseball-diamond sized islands in the northern Bahamas. Hurricane Floyd struck the area in the fall of 1999, drastically changing the experiment.

In 1997, just months after introducing the large predatory lizard, Leiocephalus carinatus, to islands on which the smaller lizard, Anolis sagrei, lived, the biologists had been stunned to find that anole populations were just half those on six other islands where the curly-tailed lizard did not exist; this difference presumably was a result of the much larger species eating the smaller one. In late 1999, two months after the hurricane, the researchers found that anole populations on the six islands with the predator were much more greatly reduced from pre-hurricane levels than were no-predator control islands.

One year later, the control populations had all returned to their pre-hurricane numbers, but most of the populations on the predator-present islands had failed to recover and several were extinct.

"The study shows dramatically that the presence of a top predator on an island affects the vulnerability of a prey population to a catastrophic event," says Losos. "The study is rare because it integrates two areas of ecological research: one studies the effect that rare catastrophic events have in determining the structure of a community; the other examines the effect of a predator on lower levels of the food chain."

The study was published in the July 5, 2001 issue of Nature. The work was supported by the National Science Foundation. Amazing to the researchers was the fact that there was any survival at all on the islands. Hurricane Floyd was a Category IV hurricane with maximum winds greater than 150 miles per hour that blew lizards off the islands or immersed them in a ten feet wall of water.

A previous study by the same authors, published in Science in 1998, had shown that a catastrophic hurricane could exterminate lizard populations. When they returned to the islands this time a month after Hurricane Floyd has passed, they expected to find the same result. Much to their surprise, not only were the populations not extinct, but all of the survivors were baby lizards that must have hatched since the hurricane.

They surmised that, as with the previous hurricane, although all lizards present on the island were swept away or drowned, lizard eggs must have been able to survive the hurricane and give rise to a next generation.

"l don't think anyone would have predicted that lizard eggs could survive immersion in saltwater for six hours," says Losos, who has duplicated the salty conditions in the laboratory and has found that eggs less than ten days old hatch well when put in salt water for that long.

The researchers have three possible explanations for the greater number of lizards on islands on which the predators were not present. One, on islands with the predator, the populations of the prey species already were significantly declining before the hurricane, making extinction easier.

Moreover, surviving curly-tailed lizards could have further reduced the anole populations after the hurricane. However, this explanation does not seem adequate because not all predator islands had curly-tailed lizards after the hurricane. Moreover, on those islands on which curly-tails were still found, the survivors were, like the anoles, hatchlings.

Although curly-tailed lizard hatchlings are larger than anole hatchlings, they probably are not large enough to eat other lizards. So, continued predation by curly tails after the hurricane probably does not explain the difference between the predator and no-predator islands.

A second explanation is that the anoles, forced off the ground and into the bush by the rampaging curly-tailed lizards, may have deposited their eggs in less secure places, such as trees rather than rock holes or other more protected sites. This would make the eggs deposited more vulnerable to hurricane winds and water.

The third explanation suggests that,because the anoles were forced to shift their habitat, moving far up into the bushes to avoid the curly tails, they may have had reduced hunting success or mating opportunities so that in the presence of the predator, they were producing fewer eggs, which in turn would result in fewer potential hurricane survivors.

"We have three hypotheses, but it's not clear if any one alone explains what happened," says Losos. "We speculate that the presence of a predator, besides reducing the population, also changes the ecology in such a way that the the anole populations are more vulnerable to a natural catastrophe, but we still don't know what is the exact causal mechanism."

Losos says that the three biologists will continue to monitor the islands and hope to restart the experiment in the spring of 2003. The intent is to study how the anole lizards adapt to the new habitats they must occupy in the presence of curly-tails.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.