To save gardens, ants rush to whack weeds.

Ants that grow their food have to weed, too. Now, the first detailed study of ants tending fungus gardens shows that whether the gardener has two legs or six, the chore looks much the same.

Like the best human gardeners, ants try to stop a weed invasion in its early stages, report Cameron Currie and Alison Stuart, now of the University of Texas in Austin and the University of Otago in Dunedin, New Zealand, respectively. Ants settle in one spot and weed, then move to the next spot, researchers report in the May 22 PROCEEDINGS OF THE ROYAL SOCIETY OF LONDON B.

"They're amazing weeders," Currie says.

Ants beat humans by some 50 million years in inventing agriculture (SN: 11/21/98, p. 334). Today, about 200 ant species cultivate a spongy lump of fungus inside their nests. Currie and Stuart focused on the elaborate gardens of the leaf-cutter ants, which use bits of foliage they collect to nourish their fungi.

Currie brought 15 colonies of Atta colombica ants from Panama into his University of Toronto lab. Each colony had its own soup-can-size fungal mass which workers cultivated with brute force and chemistry (SN: 4/24/99, p. 261).

Into one garden, Currie put just a nugget of another, invasive fungus. "I thought this was going to be exciting," he recalls. "But in about 10 seconds, along comes an ant and picks it up and carries it to the dump, and it's all over."

Currie and Stuart presented a bigger challenge by spraying the top of the fungus gardens with plain water or a solution of spores from one of two weed fungi. The tiniest workers, which do most of the gardening, quickly assembled.

"At first, they would lap up the droplets," Stuart says. "Within an hour, the surface was bristling with ants, but they didn't seem to be moving much."

With videotape and microscopes, she and Currie documented a largely stationary behavior that they call fungus grooming. Ants settled in a spot, patted the nearby garden with their antennae, and combed nearby fungal filaments with their mouthparts. The researchers suggest the ants were scraping up invader spores.

The ants then sacrificed parts of the crop, as if containing a plague, Stuart says. She saw an ant straddle a leaf snippet, rocking and tugging until it came loose. Bigger ants lugged the chunks to the dump.

The ants responded mildly to the water but more vigorously to spore solutions. A spray of Escovopsis, a dangerous pathogen specializing in ant gardens, mustered the most workers. They groomed the most intensely and hauled out the most trash. Trichoderma, an aggressive but more generalized invader, prompted less response. The weeders successfully eradicated Trichoderma from most colonies. Despite the extra effort, however, Escovopsis persisted.

"That implies that Escovopsis has devious ways of avoiding the grooming and weeding--it's an evolutionary arms race," muses Ted R. Schultz of the Smithsonian Institution in Washington, D.C. In recent studies of ant gardening, the power of pathogens to drive ant evolution has been surprising, he says.