Petite pollinators: tree raises its own crop of couriers. (This Week).

A tropical tree creates insect nurseries in its buds for miniscule pollinators, say German scientists. This novel strategy relies on thrips, insects rarely considered pollinators.

Darwin dismissed thrips as annoyances that fouled his pollination experiments, and since then biologists studying pollination haven't paid much attention to this dot of a creature. Yet a 2-millimeter-long Neoheegeria thrips serves as the main pollinator for a widespread Macaranga tree, report Ute Moog of the J.W. Goethe-University of Frankfurt in Germany and her colleagues in the January American Journal of Botany.

"Everybody in Southeast Asia knew Macaranga, but no one knew how they're pollinated," says Moog. She focused on Macaranga hullettii, which often lines roadsides and grows up to 20 meters high.

Male and female flowers grow on separate trees. Moog monitored trees in West Malaysia and found that the most common insect visitors to both sexes were thrips of an as-yet-unnamed species.

The vanilla-scented, greenish blooms sprout in bunches. Shields cover their reproductive organs, so wind pollination is unlikely. Before male flower parts mature, the buds loosen, and thrips slip into the pockets behind the shields. The researchers found eggs and larvae there and propose that thrips breed in the pockets, nourished by nectar from hairs on the flower. When Moog collected eggs and raised them, they matured in about the same time that a tree's buds take to open.

Moog found Macaranga pollen sticking to 13 percent of thrips she caught on female flowers. These flowers don't make pollen of their own, so it had to have come from another tree. Moog caught pollen-dusted thrips even on female trees at least 25 m from any male tree. "I think these small insects have more power than people think," Moog says. One thrips may not carry a lot of pollen, but a big tree can easily host more than a million thrips.

To test the thrips' importance in tree pollination, Moog bagged flowers in material that would let in windborne pollen but not even such a tiny insect. None of these flowers formed seeds. She bagged other flowers with coarser material that let in thrips but not giants like bees. These flowers prospered, setting 80 percent as many seeds as unrestricted flowers did.

The work "indicates how little we know about how natural forests maintain themselves," says thrips specialist Laurence Mound of CSIRO Entomology in Canberra, Australia.

"Thrips pollination has been widely overlooked," says Irene Terry of the University of Utah in Salt Lake City. Yet evidence has recently grown stronger that the small insects pollinate competently, she says. Last September, Shoko Sakai of Smithsonian Tropical Research Institute in Balboa, Panama, reported experimental evidence that thrips do the job for Castilla elastica, a tree once grown for rubber. In two other reports last year, Terry and Mound described thrips carrying pollen for palm-shape cycads.

"It has been an exceptional year for thrips-pollination research," Terry says.