Bumblebee 007: bees can spy on others' flower choices.

In a novel test of insect intel, researchers observed that bumblebees, which had spied on a worker bee from another colony feasting on unusual flowers, later tended to visit flowers of the same color.

This talent amounts to social learning, which is picking up a new behavior from observations of another animal, say the test's designers, Bradley D. Worden and Daniel R. Papaj of the University of Arizona in Tucson.

It's "an elegant experiment," comments Andrew Whiten of the Centre for Social Learning and Cognitive Evolution at the University of St. Andrews in Scotland. "[It's] the beet evidence I know of for observational learning in insects," he says. To date, research in such learning has focused on vertebrates.

In the 19th century, Charles Darwin proposed that honeybees forage on flowers that they see bumblebees visit. Worden says that his studies and one other group's investigations are the first to test such an idea.

In that other study, reported in the June 21 Current Biology, Lars Chittka and Ellouise Leadbeater of Queen Mary University of London describe a simple form of bee eavesdropping. The researchers found that placing a live "demonstrator" bumblebee on an artificial flower of one color usually inspired observer bumblebees to fly to faux flowers of the same color. When the researchers relocated the demonstrator to a flower of a different color, the observers tended to switch too. Foraging bees "pick each other's brains," says Chittka.

The new report from Worden and Papaj goes further by demonstrating social learning, Chittka says. The Arizona researchers allowed caged bumblebees that had never foraged among such flowers to watch for 10 minutes as trained workers from another colony visited either orange or green fake flowers.

Next, the researchers replaced those flowers with new ones free of any bee scent and arrayed them in a different pattern. Then, the investigators let loose a test bee that either had or had not observed trained worker bees. In a trial with demonstrator bees trained to land on orange flowers, the test bee was just as likely to visit an orange flower, regardless of whether it had observed the demonstrator. Bumblebees customarily respond readily to the orange color.

However, for green flowers, an unusual color choice for bees, seeing neighbors foraging increased by 50 percent the chance that a newcomer bee would visit a green flower.

In a similar experiment, Worden and Papaj replaced the live demonstrator bees with fake bees. "It was arts and crafts day in the lab," says Worden, who helped make the fakes from wire, glue globs, and salvaged wings. After observing the little models on green flowers, the observer bees more than doubled their preference for the odd color, the researchers report in an upcoming Biology Letters.