lunging into the roiled waters of human intelligence and its heritability, brain scientists say they have found that the size of certain regions of the brain is under tight genetic control and that the larger these regions are the higher is intelligence.

The finding is true only on average and cannot be used to assess an individual's intelligence, said Dr. Paul M. Thompson, the leader of the research team and a pioneer in mapping the structure of the brain.

The measurement of intelligence has long been a controversial issue, and even more so the efforts to tease out the relative contributions of heredity and environment.

Dr. Bruce L. Miller, a neurologist at the University of California at San Francisco and an expert on brain changes in Alzheimer's disease, said Dr. Thompson's work was "an exciting study that starts to show there are some brain areas in which there are very significant genetic influences on structure."

And Dr. Robert Plomin, a psychologist who studies intelligence at the Institute of Psychiatry in London, said the high correlation found between the size of certain areas of the brain and general intelligence "does make it harder to dismiss intelligence as some meaningless construct, as some want to do."

Dr. Thompson, who is at the University of California at Los Angeles, uses a type of brain scanning called magnetic resonance imaging, which can show the difference between gray matter and white matter in the living brain. The gray matter consists of brain cells, while the white matter comprises the bundles of wiring with which the cells communicate with one another. The amount of gray matter is a measure of the number of brain cells.

The human brain seems to be divided into modules that perform separate tasks. The frontal lobes are involved in planning and risk assessment, while regions at the back of the brain handle visual processing. Dr. Thompson has tried to discover if the relative size of the brain's modules is under genetic control by studying how their size varies in twins.

With the help of colleagues in Finland, where a national registry of twins is maintained, he scanned the brains of identical and fraternal pairs of twins and measured the size of the brain modules. Qualities that are under genetic control show a characteristic pattern of varying hardly at all between identical twins, who have the same genes; quite a lot between fraternal twins, who share about half their genes; and a great deal between unrelated individuals.

The researchers had their computer draw three-dimensional maps of each subject's brain, and then color coded the modules' degree of heritability. In an article published in today's issue of Nature Neuroscience, they report that the quantity of gray matter in the frontal lobes was under particularly tight genetic control, as was a region at the side of the left hemisphere known as Wernicke's area, which is central to language.

Dr. Thompson's reason for probing the genetic control of brain structure was to uncover genes that might be involved in mental diseases that can be inherited, like schizophrenia and autism. But he and his colleagues also wished to understand the role of brain modules in healthy individuals, so they gave their subjects intelligence tests and found that intelligence was significantly linked with the amount of gray matter in the subjects' frontal lobes.

Dr. Thompson said the findings were "the first maps of the degree to which the genes control brain structure." There were only 40 subjects in his study - 10 pairs of identical twins and 10 pairs of fraternal twins - but the results gave "enough statistical power to identify the key brain systems," he said.

He expressed surprise that the amount of gray matter in the frontal lobes turned out to be correlated with intelligence in his study "because you wouldn't think something as simple as gray matter would affect something as complicated as intelligence." But the amount of gray matter, which is related to the number of brain cells, perhaps reflects something that bears more directly on intelligence, like the number of cell- to-cell connections, he said.

Dr. Plomin, who wrote a commentary on the study in the journal, said the larger volume of gray matter could be the cause of higher intelligence, or it could be the other way around - people with a stronger motivation, say, might exercise their brains harder and develop a higher density of neurons.

As brain-scanning studies like Dr. Thompson's become more refined, they raise the possibility that a scan could be used to gauge various elements of personality or behavior.

Dr. Thompson said he believed that as brain scans become increasingly informative they will raise issues of personal privacy just as genetic testing has done, and should be protected with similar safeguards.

The size of gray matter in the frontal lobes cannot be used to measure an individual's intelligence, he said. Some potential uses, such as scanning to compare the intelligence of different groups, would be unethical, he added. "It would be remiss to use technology developed for disease for those types of goals," he said.