The colony also includes a queen, whose sole function is to produce thousands of other ants every year. Although we call her the queen, she has no authority over the colony.
So how does work get done? How can you have a war with no generals?
Gordon's earlier research had suggested that the ants communicate by smell, but Green — the lead author in the Nature report — wanted to take it a step farther.
Ants from all three classes — nesters, patrollers and foragers — were examined with a sophisticated piece of equipment. Gas chromatography revealed that each ant was coated with about 25 different hydrocarbons which emit slightly different odors.
"Hydrocarbons are simply molecules of hydrogen and carbon," Greene says. "There's nothing fancy about them. Yet subtle changes in the concentration of these relatively simple chemicals can produce very important and profound behavioral changes in ants."
It turned out that a patroller emits a slightly different gas than a forager, thus giving each a distinctive odor, at least to another ant. But is that really what sets the work day in motion? To find out, the researchers resorted to a clever bit of deception.
Tiny glass beads were coated with hydrocarbon that would give them the distinct odor of a patroller. Then the researchers hit nine different harvester ant colonies in Arizona, collecting the patroller ants early in the morning before they had a chance to return to the nest. Some 30 minutes later they dropped the tiny beads into the nest, one at a time, about 10 seconds apart.
The foragers brushed the beads with their antennae, and then took off for their day's work.
It didn't work when beads that smelled like foragers or nesters were dropped into the nest. It had to be patrollers, and the beads had to arrive at precisely the right frequency.
That proves, Gordon says, that the right smell, delivered in the right dosage, is all it takes to keep order in the system. The patrollers took off to do their thing without anyone having to tell them to do so.
It's not known yet whether other ant species use the same mechanism, she adds. Of the thousands of ant species all over the world, only about 50 have been studied extensively, Gordon estimates.
Who knows what dark secrets some of those may harbor? Perhaps some of them might help us understand how to build a robot that could make its own repairs, and reach its own decisions, and carry out tasks millions of miles away from home without someone pushing a single button.
Lee Dye’s column appears weekly on ABCNEWS.com. A former science writer for the Los Angeles Times, he now lives in Juneau, Alaska.