Ruvkun and other scientists choose to work with the laboratory roundworm, a tiny bacteria-eating soil-dweller, because the sequence of its entire genome is known and its 302-neuron nervous system has been described in detail.

But though it looks very different from most humans, the worm shares 40 percent of its genes with man, including almost its entire insulin pathway.

In their recent study, Ruvkun and his colleagues first used genetic tricks to deactivate genes that encode the proteins--known as receptors--that respond to insulin signaling in many worm tissues, creating a group of long-lived individuals.

What they did is like yanking parts out of an electrical appliance, explains Caleb E. Finch, an aging expert at the University of Southern California.

"You can modify a process by removing one of its controlling parts," says Finch. "It's like if you started pulling computer chips out of a radio."

Next, the researchers put the parts back in, but only in certain areas. In separate experiments, they turned insulin receptor genes back on in different worm tissues: first in the intestine but not the brain, then the brain but not the muscle, and so on.

The results were striking: While putting insulin receptors back in muscle and intestinal cells had only a small effect on longevity, replacing the receptors in the nervous system restored the worms' normal life span.