The AIDS virus is revealing itself to scientists as a rather different and potentially more threatening germ than most researchers originally thought.

Instead of attacking its victims soon after it infects them, as most viruses do, the AIDS virus lies dormant for long periods, becoming a sleeping monster that may not fully awaken for many years.

The new view of the virus, emerging now because scientists have had several years to study its behavior, leads them to believe that the epidemic will not go away soon and that it will almost certainly continue for many years and perhaps for decades.

The new view also suggests that even if the spread of the virus were to stop, there could still be hundreds of thousands of new cases of acquired immune deficiency syndrome in coming years. Some AIDS researchers now warn that the U.S. health care system could eventually be faced with 10 or 20 times as many AIDS patients as the approximately 16,000 reported so far.

The reason for this more pessimistic outlook, which was a key topic recently at a major international scientific conference on Martinique, has gradually dawned on researchers and public health experts.

In the early days of the epidemic, federal health officials declared that the AIDS virus had an incubation period of up to six months. But, as time went by, the estimate has steadily lengthened. The current figure is seven years and all the experts say they believe the true figure is probably several years longer.

Further dimming the outlook is evidence that people who are infected can remain healthy and unaware that they are carriers but still be capable of spreading it.

"This is a new concept. The fact that the virus can be latent for a very long time makes it a much more serious threat in terms of public health," Dr. William Haseltine told participants at the conference, which brought together AIDS researchers from the United States, Europe, Africa and the Caribbean. Haseltine is studying the virus at Boston's Dana Farber Cancer Institute.

The longer a person can be a healthy carrier, Haseltine said, the more likely he is to transmit the virus to others.

Various reports at the conference suggested that the incubation period depends on the amount of virus already in the body and the speed with which the virus multiplies until it reaches a critical level. It is known that the virus preferentially infects certain blood cells called T4 cells, taking up long-term residence. T4 cells are an essential part of the immune system that usually become active only when the body is fighting infection.

Both French and U.S. researchers reported finding that while the AIDS virus can permanently invade an inactive T4 cell, it does not multiply until the T4 cell is activated to fight some other infection. Thus, as long as the infected person has no other infections or other demands on the immune system, the AIDS virus remains dormant.

Once the T4 cell is activated and the virus begins multiplying, the cell soon dies but also releases many new AIDS viruses that invade other T4 cells. As long as the body is fighting an active infection -- whether AIDS or anything else -- the virus can spread to more and more T4 cells. Over a period of years the number of infected T4 cells rachets steadily upward.

Eventually there comes a "last-straw" infection that activates so many infected T4 cells that a major part of the immune system is destroyed. The immune system becomes so weakened that infections are now free to ravage the body unchecked. It is these "opportunistic infections" that usually kill AIDS victims.

"What we need if we want to come up with a prognosis for the individual is some way to assess viral load," said Dr. Samuel Broder of the National Cancer Institute. Although there is a blood test that will reveal whether a person has been infected by the AIDS virus, there is currently no way to tell what proportion of a person's T4 cells are carrying a silent infection.

Because the blood test is used on all donated blood, researchers have been able to estimate that about 1.75 million Americans are already infected. Although public health officials once stressed that only a tiny fraction of these people could expect to develop AIDS, scientists at the conference heard preliminary indications that between 30 percent and 40 percent will eventually succumb. Privately, a few scientists predict that the level will be 60 percent.

If only one-third of those already infected eventually develop AIDS, there will be more than half a million cases.

Only a few months ago, health officials described persons who were positive on the test as merely having been "exposed" to the AIDS virus. This was because the test detects only antibodies manufactured by the body's immune system to attack the AIDS virus. Now it has been found that almost everyone who is positive on the blood test carries dormant AIDS viruses inside the T4 cells, where they cannot be seen by the immune system.

The most obvious factors that activate T4 cells are bacterial and viral infections. Because sexual activity transmits several infectious diseases, such as hepatitis-B, genital herpes and the papilloma virus that causes genital warts and cervical cancer, the degree of T4 cell activation is often linked to a person's range of sexual contacts. The more partners a person has, regardless of sexual preference, the more likely the person is to pick up a new infection that will give another boost to AIDS virus multiplication.

This new understanding, scientists said, may help explain the longstanding observation that the homosexual men who are most likely to develop AIDS are the most sexually active. It also suggests that appearance of AIDS among heterosexuals will be most frequent among those who have the largest number of partners.

Intravenous drug addicts, the other major risk group, are also exposed to many blood-borne infectious diseases.

Dr. Robert Gallo, the National Cancer Institute scientist who is widely regarded as the world's leading AIDS researcher, said infected persons could probably extend their life expectancy by keeping themselves in good physical health, maintaining resistance to infection.

Still, infections are not the only things that can activate T4 cells. Allergies can also stimulate the immune system. Even vaccination, which tricks the immune system into manufacturing antibodies, can be a risk for infected persons. Gallo said healthy carriers may wish to avoid vaccinations for diseases they are unlikely to contract.

Another risk for carriers, according to Dr. Peter Fischinger, deputy director of the National Cancer Institute, is pregnancy. He reported a study of 15 women who were healthy carriers and who became pregnant. Five of the women developed AIDS during pregnancy. Seven came down with lesser symptoms known as "AIDS-related complex," which is often a harbinger of "frank AIDS." Only three remained healthy through their pregnancies.

It is already known that infected mothers commonly transmit the virus to their babies. The disease is known to progress unusually fast in young children, perhaps because their immune systems are frequently activated by infections.

The new appreciation of how the virus multiplies puts a different light on the question of whether infected children should go to school. School, a notorious trading ground for minor infections, may be a far greater threat to a healthy child carrying the AIDS virus than that child is to the other children.