Researchers at Princeton University have evidence that adult monkeys produce a small but steady trickle of new nerve cells, a finding that challenges the classical assumption that after early childhood the brains of humans, apes and other primates can only lose cells, not gain them.

The finding raises the possibility that ways might be found to turn the trickle into a torrent, stimulating the replacement of cells killed off by such calamities as stroke or Alzheimer's disease. At the moment, however, it's unknown whether the new cells are even functional.

Large-scale turnover of adult brain cells occurs in birds, fish and some other lower vertebrates. And in mammals, some specialized cells, such as those involved in the sense of smell, may regrow. Until now, however, there's been little evidence that cells of the cerebral cortex--the vast outer mantle of the brain where thinking occurs--can be replaced.

"The assumption has been for over a hundred years that there are no new neurons added," said Charles G. Gross, a psychologist and co-leader of the study, which is reported today in the journal Science. "We have shown they are added, and to the regions of the brain involved in the highest cognitive functions."

During fetal life, nerve cells in the brain form in one place and migrate to distant regions, where they make connections to other cells. To see if that process occurs in adult life, neuroscientist Elizabeth Gould injected a dozen adult macaque monkeys with a biochemical taken up by newly formed cells. She followed that with markers that specifically identify nerve cells (also called "neurons"), and with tracers that sketch their paths.

The researchers found that neurons were being born continuously along the shores of the lateral ventricles, two huge lakes of fluid lying deep in the brain. The cells then migrated along well-known paths to the frontal, parietal and temporal cortexes, a trip that took more than a week.

Once in the cortex, the cells sent out "axons," the tendrils that actually establish the signal-carrying connections to other neurons. This suggests--but doesn't prove--that the new cells had found a useful place in the brain's long-established circuitry.

In order to test whether that's the case, Gross and Gould plan in future experiments to interrupt production of the cells--this can be done fairly easily--and see if an animal's behavior changes.

"We speculate that because we find them in these higher cognitive areas, the new neurons may play a role in learning and memory," Gross said.

The researchers also will give experimental animals various learning tasks and see if that alters the number of new neurons produced, or the destinations that the nerve cells seek. A previously published study by Gould showed that if you train rats in tasks that involve a structure called the hippocampus (which is not in the cortex), a larger-than-normal number of new cells migrated to the hippocampus.

Finding a useful role in the brain's densely packed circuits "is a formidable problem for the newly generated neurons," speculated Thomas M. Jessell, a neuroscientist at Columbia University. But the fact that evolution has preserved the brain's capacity to even form them is encouraging.

"Viewed optimistically, you'd think that if you take the trouble to make all these neurons, you'd want to put them to good use," Jessell said. "But you want to get the evidence that actually occurs before one gets too excited about the practical application of such observations."

What's clear is that any regenerative potential in the brain would have to be massively stimulated in some way to be of clinical use.

The fact that stroke victims don't recover neural function, or that the brains of Alzheimer's disease patients don't spontaneously repair themselves, suggests that even if what's happening in monkeys also occurs in human beings, it's not enough to overcome the brain's most devastating diseases.