In a brilliant, furious burst between late 1982 and early 1984, the laboratory of Robert C. Gallo isolated the virus now called HIV, established its link to AIDS and developed a test to detect its presence in blood.

These achievements put Gallo and his co-workers at the National Institutes of Health at the forefront of the battle against AIDS and established their laboratory as one of the most famous and most productive in American medical science.

But over the past few months, these accomplishments have become the focus of one of the most extraordinary investigations in the history of modern science. In a probe that began last November and is expected to continue for months, NIH officials have conducted exhaustive, repeated interviews with Gallo and his co-workers and reviewed thousands of pages of documents to determine whether that burst of productivity was also the occasion of scientific fraud or, perhaps, of extraordinary carelessness.

Was the virus discovered by the Gallo laboratory actually no more than a copy of a virus isolated earlier at the laboratory of another researcher, the French scientist Luc Montagnier? And if so, was Gallo's "discovery" of the French virus strain an accident or part of a deliberate attempt to steal credit for discovering the cause of AIDS?

This is not the first time ethical questions have been raised about Gallo and the discovery of the AIDS virus. In 1985 Montagnier alleged in a lawsuit that his laboratory, not Gallo's, had been the first to isolate the virus. That suit was later settled after Gallo and Montagnier agreed to share credit for the discovery. The two then collaborated on a detailed history of who did what when and published it in a scientific journal. For most scientists, the matter was settled.

But last fall allegations of wrongdoing on Gallo's part resurfaced in a 50,000-word article by investigative reporter John Crewdson that took up 16 full pages of the Chicago Tribune. The article made new allegations against Gallo and, under pressure from congressional science watchdog Rep. John D. Dingell (D-Mich.), NIH officials felt compelled to look into the matter themselves.

The result is an inquiry without precedent in the history of the NIH, the nation's premier biomedical research center. No other investigation has taken so long, dealt with a scientific discovery of such importance or directly implicated so distinguished a researcher. Gallo is easily one of the country's most famous scientists, frequently mentioned as a Nobel Prize contender, and a man whose research publications were cited by other researchers publishing their own work during the last decade more often than those of any other scientist in the world.

Nor has any NIH investigation triggered so much rancor and controversy within the scientific community. The probe, in the nine months since it began, has been called unfair. Crewdson's article has been called too biased to be the basis for an official investigation. And Gallo's supporters, who count among their numbers some of the country's most eminent scientists, have begun to question publicly the wisdom of an inquiry that they say has seriously damaged the productivity of a scientist and a laboratory considered essential to fastest possible progress in the battle against AIDS.

"In an ideal world where a scientist is working on an epidemic involving millions of people, one might consider that historical questions be left for a time until the epidemic is over," said biologist Howard Temin, a Nobel laureate who teaches at the University of Wisconsin. "But, of course, that is not the way things are done in our society."

The origins of the controversy over the AIDS virus date to 1982, when two laboratories -- Gallo's at NIH and Montagnier's at the Pasteur Institute in Paris -- began working in parallel to track down the cause of the then-mysterious disease striking homosexual men. They examined blood samples of afflicted men and looked for some common contamination or infectious agent.

The first to publish results was Montagnier, who described finding a virus he dubbed LAV in an AIDS patient in May 1983. Montagnier, however, stopped short of claiming it was the cause of AIDS. In April 1984, Gallo went further. He announced that he had identified a virus -- which he called HTLV III -- and proposed it was the cause of AIDS. Gallo also reported developing a test to detect it in blood.

Within months, however, controversy erupted. Montagnier, it turned out, had sent Gallo a sample of LAV in September 1983. When Gallo's virus and Montagnier's virus were analyzed closely, they were found to be so similar genetically as to suggest that they came from the same blood sample.

This was a startling fact. As many laboratory technicians know, viruses from one blood sample can easily contaminate another. Moreover, laboratories have definite personalities, largely defined by the "lab chief," the position Gallo occupies. Some labs are careful, methodical, precise -- and slow to make progress. Others, like Gallo's, are busy, bustling places where scientists and technicians are driven hard to produce results fast.

In fact, Gallo's lab had suffered before from contamination. Did it happen again to Gallo? Did he discover Montagnier's LAV and mistakenly think he had stumbled onto something new?

These were the questions raised and supposedly resolved in the 1987 settlement of the French lawsuit. Gallo and Montagnier agreed to share equally the credit for discovering the AIDS virus. Part of the settlement was to create a new name for the virus. HTLV-III (human T-cell lymphotropic virus) and LAV (lymphadenopathy associated virus) became HIV (human immunodeficiency virus).

In his article, Crewdson suggested that not only had there been a contamination in Gallo's lab but also that it was deliberate. Gallo or one of his co-workers, it was strongly implied, had stolen the AIDS virus.

Crewdson presented no proof for his allegation. But, by examining key events from the period in painstaking detail, he built a provocative circumstantial case against Gallo. He tried to show that Gallo knew HTLV-III was the same as LAV and systematically attempted to deceive the scientific community into thinking that they were different.

Crewdson's article centers on two related issues.

The first is whether a cell culture in Gallo's laboratory was contaminated by a culture that came from the French laboratory.

The critical evidence here is the genetic fingerprints that were taken of the two virus isolates, which showed them to differ only by 1.5 percent. This is odd because HIV is a highly variable virus, constantly mutating even as it multiplies within a person. Most isolates of HIV that have been analyzed differ by 5 to 20 percent. For a virus like HIV, a difference as small as 1.5 percent is, in the minds of many, practically no difference.

The case isn't closed, however. The French virus and Gallo's isolate were taken from blood samples of two gay men who became infected in the New York area in 1977. That was early in the epidemic, when the virus had not spread far or mutated into as many forms as it has since. Could the two men, by some extraordinary fluke, have been infected by the same person?

At the same time there is new evidence that there may be some strains of HIV that don't vary that much. Washington University resercher Lee Ratner, for example, recently looked at three children infected with same tainted blood sample. Three years later, no pair of HIV isolates from the children differed by more than 0.2 percent. In other words, Gallo and Montagnier's sample could have come from men infected by a common source years earlier.

But is it? Because of the inquiry, Gallo could not be quoted in this article. But in previous accounts he has said that a lab worker infected with the strain he originally isolated has had little or no change in the variability of his virus over three years.

"It is certainly conceivable that they could in fact be different isolates," Ratner said. "It is very hard to make hard and fast conclusions based on variation data alone."

The second key issue is whether a contamination, even if it did happen, actually mattered. Gallo has publicly conceded that a laboratory mix-up was possible. But, he has asserted, the issue is irrelevant because during the critical 1983-84 period, his lab had identified numerous other strains of the AIDS virus, each of which was undeniably different from LAV and that any one of them sustains his claim to be discoverer of the cause of AIDS.

This claim, Gallo partisans say, renders moot the charges of fraud. Why would Gallo, they ask, deliberately steal Montagnier's virus if he had others of his own?

The article by Crewdson, who declined to be interviewed, disputes Gallo's claim. Crewdson said that his examination of notebooks and papers acquired from the Gallo laboratory through the Freedom of Information Act (FOIA) shows critical discrepancies between the sources of isolates given in scientific papers and the original notations in notebooks.

Crewdson's account makes much, for example, of Gallo's admission that the photographs of virus that he submitted to Science to illustrate his claim to have discovered the cause of AIDS in 1984 were actually photographs of Montagnier's LAV. Gallo long ago conceded the error, calling it a simple mistake. Crewdson suggests it was a deliberate deception, proof that at the time Gallo had no other viruses except LAV to photograph.

Crewdson also alleged that at key scientific meetings between late 1983 and early 1984, Montagnier described his success with LAV but Gallo did not mention his work on the AIDS virus. Yet, when the reports from each meeting were published later, Gallo's included discussions of his laboratory's isolation of HTLV-III.

All the questions are under consideration by the NIH inquiry, which is currently finishing its deliberations and is to present its findings to an outside advisory body named by the National Academy of Sciences.

The inquiry, NIH officials acknowledge, covers much of the same ground as the previous examination of Gallo's record during the 1987 lawsuit. But that investigation was done by lawyers, and this one is being conducted by scientists. The officials also said that Crewdson's detailed allegations and the continuing controversy required that NIH at least try to find out whether a full investigation was warranted.

"In some ways this is looking at the same issues as before," said Suzanne Hadley, director of the NIH Office of Scientific Integrity. "But we felt that it needed to be looked at entirely objectively and in a rigorous and thoroughgoing way and that's what we are going to do."

Hadley said that a decision about whether to proceed to an actual investigation of Gallo would be made this fall.

Whether matters will get that far, however, is an open question.

A panel of scientific experts assembled by the journal Science earlier this summer, for example, reviewed hundreds of newly released documents from Gallo's lab not yet seen by Crewdson and concluded that, contrary to the Crewdson piece, it does appear that Gallo had other isolates of the AIDS virus during the period in question.

Other scientists say Crewdson's reporting has serious gaps that biased his conclusions against Gallo.

According to scientists familiar with that period, the documents Crewdson obtained through the FOIA gave him only a partial and perhaps misleading picture of what happened. No documents are available under the FOIA from private contract laboratories, where a lot of the technical work under Gallo's direction was conducted. Nor was Crewdson able to get such access to the papers of Montagnier's laboratory in France, resulting in what some scientists say is a naive and much less critical approach to the allegations made against Gallo by the French.

Crewdson was unable to get many of Gallo's key collaborators -- and Gallo himself -- to cooperate with him in his research because of a strong belief within the scientific community that the article -- which Crewdson was known to be working on for years -- was purely an attempt to discredit Gallo. As a result, say some scientists familiar with the controversy, Crewdson had to rely on outside sources or second-hand evidence for crucial parts of his account.

Crewdson did not agree to be interviewed for this article. He referred questions to Chicago Tribune deputy managing editor Howard Tyner, who said that the idea that Crewdson's account was biased was "preposterous" and "may have very well started with Gallo himself."

Tyner said that the refusal of many scientists to cooperate with Crewdson "may have had an impact on the story" but, he added, "the number of sources that Crewdson had who did cooperate was very large and the result was a fair and accurate report."

Other scientists who dealt with Crewdson, however, said that he was so intent on "getting" Gallo that at times he appeared to venture into ethically questionable territory.

In one instance, for example, according to Mark Kaplan, a researcher at North Shore University Hospital in Long Island and an early Gallo collaborator, Crewdson called Kaplan after receiving copies of records on blood samples from an AIDS patient that Kaplan had sent to Gallo.

According to Kaplan, when he tried to avoid talking to Crewdson, Crewdson threatened to call up the family of the patient who had donated the blood, whose name had inadvertently not been blacked-out -- as is the custom -- before the document was publicly released.

Tyner said that Crewdson eventually did call the patient's family, and they had no objections to the patient's name being used.

Kaplan, however, said he was outraged at Crewdson's willingness to breach the promise of confidentiality given to all research subjects.

"As far as I'm concerned, his behavior was very corrupt," Kaplan said. "He had an axe to grind. . . . He was out to tear Bob down."

"You've got to remember that this thing called AIDS descended on us like a plague," said Duke University AIDS researcher Dani Bolegnesi, a close friend of Gallo. "Bob's lab jumped into this in full force in 1982-83 and really made a commitment to go after the solution. That wasn't easy. There were all kinds of theories as to what the cause was. Samples were coming in from all over the country. Analysis had to be carried out. It was a very confusing time. What Bob did was cut through a lot of the confusion, decide on one line of approach and say let's go. . . . "

"Crewdson tries to look at all this very analytically," Bolegnesi said. "He wants his i's dotted and his t's crossed. But you've got to remember the pressure on those working in the field to come up with an answer. That doesn't justify any mistake. But it was not the kind of experimental work that you could do in a relaxed atmosphere."

Since his article was published, Crewdson has continued to probe Gallo's personal financial arrangements, his involvement with French AIDS reseacher Daniel Zagury and his relationships with other National Cancer Institute researchers, looking for more evidence about what really happened in the Gallo laboratory between late 1983 and early 1984.

"We're not talking about an imprecise area of our lives," Tyner said. "Notes are supposed to be kept. Data are supposed to be kept. Crewdson's reporting is based as much as anything on lab notes and data which should be pretty clear."

Others aren't so sure.

"I was working one corridor over {from the lab where the virus was isolated} in those days," said Mandy Fisher, a former Gallo student now working at University College in London. She said a person could participate closely in the life of the lab, even attending the informal presentations that researchers routinely give in-house colleagues, and not detect any foul play. "And if I don't know what happened, then I don't know how Crewdson thinks he can know."