Why aren't there more cannibals around?

Nutritionally, eating the neighbors makes a lot of sense. That's why David W. Pfennig has wondered for years why cannibalism is so rare among animals.

Now he and his colleagues have used naturally cannibalistic tiger salamanders to test the idea that the risk of disease limits the evolution of cannibalism. Their results provide the first laboratory demonstration that eating one's own species packs a greater risk of disease than eating a different species.

"Cannibalism is a good way of getting a nutritious meal," says Pfennig, an ecologist at the University of North Carolina at Chapel Hill. A person existing on, say, lettuce would have a much tougher time obtaining all the vital amino acids and other nutrients than someone routinely lunching on coworkers. Earlier research in the laboratory suggested that tadpoles and mosquitofish excel nutritionally when they eat healthy compatriots. "They grow better when they eat their own kind," Pfennig says. Also, "it's a great way to get rid of competitors."

From amoebas to humans, many animals can become cannibalistic under the right circumstances. Still, "it tends to be something that is not normally seen," Pfennig says.

For years, scientists have suspected that the risk of disease could limit the spread of cannibalism because pathogens that target the dinner target the diner, too. The theory seems to play out as predicted, Pfennig's team reports in the May Animal Behaviour.

He and his coworkers studied the effects of a serious hemorrhagic disease that sweeps through salamander colonies both in nature and in the laboratory. Five of 12 previously healthy tiger salamanders that ate sick salamanders of their own species died before metamorphosing. Other tiger salamanders flagged, but eventually recovered, after eating salamanders of another species suffering from what appeared to be the same disease.

The researchers also checked menu preferences of young tiger salamanders raised in crowded conditions. These youngsters had grown into so-called cannibal morphs, with big bodies and extra-wide mouths. When given a choice in the laboratory, 9 out of 11 ate salamanders of a different species and ignored their own.

The idea that pathogens limit cannibalism makes sense to James P. Collins, who has studied tiger salamanders extensively at Arizona State University in Tempe. He, Pfennig, and their colleagues have documented that salamander eggs are less likely to grow into cannibal morphs if they come from regions where the hemorrhagic disease is endemic.

Ecologist Mark A. Elgar of the Pierre and Marie Curie University in Paris notes that Pfennig is the first researcher to measure cannibalism costs directly in a nonhuman species. Observations of a Papua New Guinea tribe showed that its members caught the deadly disease kuru from eating human brains.

Pfennig's new work is "a nice, convincing, and original study," Elgar says. "I suspect the costs [of cannibalism] will vary between species, but risk of personal injury and pathogens are likely to be high on this list."