Can visiting a plant ruin an experiment?

Merely walking up to a plant and handling its leaves, which is routine for any researcher collecting data, may skew outcomes in studies of predators attacking plants, ecologists suggest.

For two species out of six tested, the amount of leaf damage from insects differed significantly depending on whether researchers handled plants once a week or left them alone, report James Cahill Jr. of the University of Alberta in Edmonton and his colleagues.

They linked one gentle hand-stroke a week to about a 60 percent jump in insect damage to the leaves of Apocynum cannabinum, or Indian hemp. In contrast, the same handling correlated with nearly one-third less damage to Potentilla recta, or sulfur cinquefoil.

"The biggest take-home message is that your being there matters," Cahill says. In the February ECOLOGY, he and his colleagues warn that their finding could have "potentially dramatic consequences for field biologists," pushing them to rethink the design of experiments.

"I think it should," comments herbivory specialist Richard Karban of the University of California, Davis. He calls this demonstration of a researcher effect "pretty exciting" since "the idea has bounced around, but not the evidence."

To test for such an effect, Cahill and his coauthors selected about 1,000 ankle-to-waist-high plants representing six species in a lush Pennsylvania field. The researchers avoided going near some of them but gave the others one gentle, barehanded stroke from base to tip each week.

Handling Kentucky blue grass (Poa pratensis) or the snapdragon-like butter-and-eggs (Linaria vulgaris) presented no hardship. But Cahill doesn't have fond memories of stroking Cirsium arvense and Solanum carolinese (Canada thisles and horse nettles). After 2 months, these three species looked much the same regardless of handling. Just why the other two species differed, "we have no idea," Cahill says.

"The fact that plants respond to touch and visitation shouldn't surprise anyone," says Jack C. Schultz of Pennsylvania State University in State College. "Because they can't run away, plants are exquisitely tuned and responsive to their environments." His lab has found that wind can trigger chemical-defense buildups in oaks, beans, and tomatoes.

Edward Farmer at the University of Lausanne in Switzerland compares the modern ecologist's task to that of a stealth aircraft evading enemy radar. To avoid detection by the plants and predators under study, however, Farmer says, "we have a lot more to learn about the surveillance systems" of these organisms.

The new work also raises questions about past experiments, but Karban isn't throwing away history yet. "Let's wait to see how widespread this effect turns out to be," he says.

As for the future, ecologists could find themselves commiserating with physicists over a research paradox. In 1927, Werner Heisenberg argued that measuring a particle changes its behavior, creating permanent uncertainty. Echoing this warning, Cahill and his colleagues called their paper "The Herbivory Uncertainty Principle."