Nearly 100 of the world's top scientists gathered at the National Institutes of Health last week to propose formation of an international team to map the locations of human genes -- a feat that is the biological equivalent of charting a vast, unexplored continent.

"The goal is understanding human disease," said Sir Walter Bodmer of London's Imperial Cancer Research Fund Laboratory. Bodmer chaired the human gene mapping meeting, sponsored by the Howard Hughes Medical Institute. A gene map of the human chromosomes, he said, would enable scientists to offer better means of disease "prevention and treatment."

Chromosomes are microscopic, rod-shaped bodies located in each living cell. Each chromosome contains one long strand of deoxyribonucleic acid, or DNA, the blueprint of life. DNA -- or a similar chemical, RNA -- is found in every living thing. DNA orchestrates every facet of development -- from directing which cells will become part of the brain or heart to determining hereditary characteristics such as sex, hair color, skin and eyes.

Segments of DNA that direct the creation of a single protein are called genes. If one of these genes is missing or altered, the result can be a serious illness such as cystic fibrosis.

The more information scientists have about genes and their locations, the better they can diagnose, treat and prevent diseases. Gene mapping and cloning are already allowing the prenatal detection of some 15 genetic diseases, including cystic fibrosis, Duchenne's muscular dystrophy, sickle cell disease and hemophilia.

Since 1978, scientists have been able to identify and clone, or genetically copy, some of the specific genes that cause these defects.

Work is also progressing on other diseases that are suspected of having a genetic basis, including heart disease and cancer. The goal is to treat these diseases by gene therapy -- that is, altering the gene to either prevent or treat a disease. But unless scientists know the location of specific genes and how they work, they are limited in the treatment they can provide.

Mapping specific genes in humans and in other closely related animals, including mice, will help scientists develop treatments for genetic disease, researchers say.

"We're just now becoming aware of the genetic basis for development," noted Dr. Frank H. Ruddle, a Yale University biology professor who participated in the meeting. "As we become aware of genetic mechanisms of development, it will give us insight into abnormal birth defects, genetics and the environment and it will give us insight into aging as well into neurogenesis development of the brain and nervous system and behavior."

Since the task of mapping the genetic structure is so complex and time-consuming, the scientists called for an international team of researchers from around the world who would work together on the project from various laboratories. Connecting these scientists would be sophisticated computer data banks to analyze and store information about the chromosomes.

But exactly who will coordinate and fund this project, which has been estimated to cost between $1 billion and $10 billion over the next decade, is uncertain.

"I'm in favor of the project," said Dr. James Watson, who shared the Nobel prize in 1962 for discovering the structure of DNA and now heads the Cold Spring Harbor Laboratory on Long Island. "But I think that I can say that everyone else at Cold Spring Harbor is against it."

The concern of many younger scientists, Watson and others told the meeting, is that research dollars -- already in short supply -- would be siphoned from other ongoing projects to fund the gene-mapping project.

Some initial organizational money may come from the Hughes Institute, the largest private philanthropic institution of any kind in the world. The institute already funds gene studies at Yale University, Baylor College of Medicine and the University of Utah. Hughes currently spends $3.5 million annually to fund the New Haven Gene Mapping Project at Yale.

"What might be an additional role for Hughes is to participate in the coordination of the project," said Dr. Donald S. Fredrickson, president and chief executive officer of the Hughes Institute. The topic will be addressed at an upcoming board of trustees meeting, Fredrickson said.

NIH and the Department of Energy also are likely to expand their already central roles in gene research. Among other research, NIH now funds GenBank, a data base of gene information, housed at the Department of Energy's Los Alamos National Laboratory in New Mexico.

Estimates of exactly how much time will be saved with an international scientific effort vary. If work continues to at its current speed, British scientist Dr. Sydney Brenner said, developing a gene map of the human chromosomes would take about "100 man-years of effort." With an international effort, the project "is well within one's ability to set up and do within a two- to three-year period."

"That means," said Nobel laureate Dr. Walter Gilbert of Harvard University, a molecular biologist and an advocate of the pro- ject, "that many of the benefits of having the physical map . . . would be available to us by 1988-90." This advance would enable researchers to identify functions of specific genes and "speed up all future human gene cloning tremendously." Gene clones -- artificially produced genes -- can be used for a variety of purposes in detecting and treating disease.

When the project is completed, experts estimate, it will take some 10,000 books to list the names and descriptions of the 3 billion chemical units that make up the 46 human chromosomes. By the time the project is finished, new technology will likely allow the information to be stored on optical discs that can be read by lasers -- similar to the way music is now recorded on compact discs. While most of the assembled researchers were in favor of the gene-mapping project, they agreed to put off for several years the complete sequencing of the entire set of human genes. Sequencing identifies the building blocks of the DNA which make up the genes, but doesn't tell what the individual genes do.Sequencing is similar to a satellite image of a continent that shows in detail such things as deposits of minerals, whether there is snow on the ground, if there is a sand dune present or a lake or mountain range. Chromosome mapping might be compared to identifying where the cities are and what their functions are.

By mapping known genes first, scientists say they can gain significant amounts of information at a fraction of the cost of sequencing. In the meantime, waiting a few years will allow rapidly evolving technological methods to catch up with the demands necessary for sequencing.

It seems important to "complete the gene map which is already being put together and then wait until the technology for sequencing gets better and cheaper and faster and be prepared to go on eventually to do the ultimate goal of sequencing," said Hughes Institute's Fredrickson.

In two to three years, said Dr. Leroy Hood of California Institute of Technology, a vast new array of technologies will be available to help scientists in their quest. These advances will allow sequencing to occur "10 times faster, 100 times more accurately and 100 times less expensively" than current methods, Hood said. "To do this right, you have to have technological developments."

"In the long run, if we spend five years to develop powerful new kinds of technologies, we're going to be in a position . . . to make a better assessment to the extent to which complete sequencing is necessary or desirable," he said.

As part of this effort, Hood unveiled the first automated DNA sequencing machine in June. The technique assigns a color to each of the chemical building blocks of DNA. Lasers then read the colors and relay the information to a computer.

"By getting together, we can hopefully achieve a great deal more than we can separately," Bodmer said.