Beavers bite trees, benefit baby beetles.

At the first nibble by a plant-eating animal, many trees fight back by releasing noxious chemicals into their leaves. When a beaver chomps on a cottonwood, however, that strategy misfires. `The bitter compounds in the tree's regrowing shoots attract a leaf-eating insect.

Cottonwood leaf beetles prefer the leaves that sprout after a beaver has cut a tree down, report researchers from Northern Arizona University in Flagstaff. Beetle larvae that eat the new growth co-opt the nasty compounds to protect themselves.

"The thing that's neat about it is that it's a positive interaction," says Gregory D. Martinsen, a coauthor of the report in the January Ecology. "The beaver herbivory ends up benefiting the beetle."

Compared to the cottonwood's normal juvenile growth, the resprouts contain twice the concentrations of phenolic glycosides, aspirinlike compounds that repel other herbivores. Nevertheless, resprouts attracted 15 times as many adult Chrysomela confluens beetles, the scientists report. Beetle larvae were also more numerous on new growth. In an experiment, they grew faster and larger on the resprouts than on juvenile sprouts.

"In theory, the [change in] chemistry should be worse for the beetle," says ecologist Mark D. Hunter of the University of Georgia in Athens, "but the beetle does something very clever with that chemistry."

The larvae store the chemicals in their glands as a defense against predators, explains Martinsen. When disturbed, the larvae turn the glands inside out, exposing drops of foul-smelling, bitter-tasting fluid.

Ants, which eat beetle larvae, are common on cottonwood trees. Martinsen and his colleagues tested the potency of the chemical defense by placing larvae on ant mounds.

"The bravest ant attacks the [larva], and the glands come out. The ants literally do back flips to get away from it," he says. "This happens several times."

The researchers found that larvae which were fed resprouts could hold out against the ants 28 percent longer than larvae fed normal juvenile leaves.

Such studies of indirect effects, in which an interaction between two species affects a third, are a burgeoning area of research for ecologists.

"Although the idea that everything is connected to everything else has been around for at least a century," says J. Timothy Wootton, an ecologist at the University of Chicago, "good experimental evidence that that is indeed the case has only recently been available."