SCIENCE HAS accepted the possibility that the human immunodeficiency virus (HIV) that causes AIDS is a variant of a virus found in monkeys and chimpanzees. But no one knows how it jumped the species barrier to humans. I am strongly impressed by evidence that AIDS can be traced to one early polio vaccination program.

Many in the scientific community and the establishment media deride such a notion. But the record shows that by 1961, many scientists worried about the possible danger to humans of monkey viruses in polio vaccines, which are usually manufactured using monkey kidneys. Indeed, when scientists learned that a particular monkey virus -- called SV40 -- found in much polio vaccine could cause tumors in young hamsters, they quickly banned from further use any vaccine carrying SV40.

Polio vaccines, produced by culturing strains of polio viruses in primate cells in laboratories, are injected into or ingested by humans. The recipient's immune system makes antibodies that ward off the wild polio virus. Hundreds of millions of people have been immunized, perhaps history's most acclaimed public health effort.

The discovery that polio virus and other viruses could be grown in primate cell cultures was a key breakthrough in developing polio vaccines and won a Nobel prize in 1954 for researchers Frederick C. Robbins, Thomas H. Weller and John F. Enders, who used human tissues for their studies.

But it was later discovered that monkey kidneys used in vaccine production often contained previously unknown monkey viruses, some of which could infect people -- and in fact had done so. Researchers identified scores of simian viruses (SVs) in the kidneys of monkeys, commonly used to culture polio vaccine.

After SV40 was discovered, vaccine makers switched from Indian rhesus monkeys to African green monkeys. But in the early 1980s, researchers discovered that many such monkeys were infected with a retrovirus related to human immunodeficiency virus (HIV), the one that caused AIDS in humans. This retrovirus cousin of HIV, called simian immunodeficiency virus (SIV), could have been present in any vaccine made from the tissues of these monkeys before 1985, the year when sophisticated testing was instituted.

Could a vaccine containing an even closer relative of HIV have transmitted the AIDS virus to humans? And if so, would that transmission correspond with what is known about the early occurrences of AIDS?

With the help of Blaine Elswood, a 43-year-old AIDS treatment activist in San Francisco, I've found many clues suggesting such a possibility. Elswood's research has led me to a scientist whose early, experimental polio vaccination program in the former Belgian Congo (now Zaire) is today all but forgotten. After his vaccination program ended, he warned Congress about the vexing problem of monkey viruses contaminating the vaccines.

The clues include:

A 1989 article in the Journal of the Royal Society of Medicine which, while discussing a possible cross-species epidemic caused by a live-virus vaccine, asserted in part, "It would appear that the AIDS epidemic may be just one of the latest of several mammalian cross-species viral transfers triggered by the techniques of virology developed in the 20th century, which subsequently spread out of control in the new host species."

A 1989 letter from Japanese researchers to the journal AIDS noting that most live oral polio vaccines worldwide are still made in kidney-cell cultures from African green monkeys. They recommended that monkeys naturally infected with SIV should not be used to make vaccines. In Japan, they noted, only kidneys from monkeys free of SIV are used in polio vaccine production.

Elswood and Raphael Stricker of California Pacific Medical Center have co-authored a paper, recommended for publication in a science journal published by the Pasteur Institute in Paris, theorizing that Africa's AIDS epidemic was spawned by a contaminated polio vaccine administered from 1957 to 1960 to at least 325,000 people in Rwanda, Burundi and the former Belgian Congo. This is precisely the region where the AIDS epidemic rages most fiercely and from which many experts believe it spread.

That polio vaccine was devised by Hilary Koprowski of Philadelphia's Wistar Institute, who began the scientific race to develop live oral polio vaccine. Koprowski, a former vaccine researcher for Lederle Laboratories, was the first to administer live, weakened polio virus to human beings, initially in 1950.

Starting in 1957, Koprowski's Congo vaccines were the first ones administered to a large human population, sprayed into the mouths of hundreds of thousands of Africans. There was virtually no follow-up, which Koprowski blames on the Congo's subsequent independence and civil war.

When I tried out the Congo theory of the origin of AIDS on Gerald Meyers of Los Alamos National Laboratories, the chief federal expert in genetic sequencing -- the science of tracing a virus's evolution -- he conceded that at least the timing seemed right. He has computed that the common ancestor of the half dozen variants of today's primary AIDS virus, HIV-1, entered the human population about 1960.

Moreover, at least one polio researcher thought Koprowski's Congo preparation was contaminated -- though by what and to what effect (if any) is unknown. Albert Sabin, a legendarily careful researcher, reported in 1959 in the British Medical Journal that he had found an unidentified cell-killing virus in Koprowski's Congo vaccine. Koprowski has always disputed that, saying two other labs found his preparation free of viruses other than polio. Koprowski recently told me his were as safe as any of the other oral polio vaccines.

It is unclear what monkey species Koprowski used to make his two Congo vaccines. He first told me he had used African greens but in a later conversation said that while he could not document it, he suspected that he had first used rhesus monkeys -- which aren't a natural host for SIV. But he conceded that the kidneys were already removed when his lab acquired them, raising the question of whether he could have known the monkey species. Recently he has been quoted as saying he imported SIV-free monkeys from the Philippines.

On April 4, 1961, Koprowski wrote to the House health and safety subcommittee taking issue with a U.S. Public Health Service requirement that live polio virus vaccine be grown in monkey kidneys.

He suggested that human cell lines be used instead. "As monkey kidney culture is host to innumerable simian viruses, the number found varying in relation to the amount of work expended to find them, the problem presented to the manufacturer is considerable, if not insuperable," Koprowski wrote. "He is faced with the prospect of having to discard most of the manufacturing lots of vaccine . . . . As our technical methods improve we may find fewer and fewer lots of vaccine which can be called free from simian virus."

Initially, neither Koprowski nor Sabin thought the unknown viruses harmful. But a leading virologist of the time, Joseph Melnick of Baylor College of Medicine, told me last fall that the discovery of SV40 -- which caused cancer in baby hamsters -- had "scared the hell out of us."

Weller, one of the trio that won the Nobel prize for learning how to grow polio virus in tissue culture, recently told me of the discovery of SV40: "It {badly scared} all of us. Here was a virus -- we didn't know what it did in man -- that produced tumors in hamsters."

Although they switched monkey species, scientists continued to put unknown monkey viruses into the human population.

Weller said he thought there was a "pretty slim chance" HIV or a related retrovirus would be found in old polio vaccine stocks maintained by the Food and Drug Administration (FDA). Then, he said. "In science, nothing is 100 percent. I might well be wrong." Anthony Fauci, chief federal AIDS researcher, told me recently that my story of Elswood's hypothesis in the March 19 issue of Rolling Stone and a parallel theory written by Walter S. Kyle in the March 7 issue of the British medical journal Lancet had unleashed a "major firestorm" of controversy.

Kyle theorized that the AIDS epidemic among American male homosexuals could have been accidentally started in the mid-1970s by an experimental treatment for herpes lesions used in New York and California. The treatment: double doses, twice as often as used for polio vaccination, of the Sabin oral polio vaccine.

Kyle -- a lawyer who bases his theory on evidence obtained in discovery from polio compensation cases -- thinks the Sabin vaccine was contaminated with monkey retroviruses. A spokesman for Lederle Laboratories, the only U.S. manufacturer of oral polio vaccines since the mid-1970s, told me that since 1985, when sensitive new testing procedures were instituted, Lederle has sometimes found SIV in early stages of its vaccine production process. The spokesman said such contaminated materials are eliminated when found.

What about vaccine produced and administered before 1985?

The spokesman said that if you don't know something's there, you can't test for it.

The FDA's Division of Produce Quality Control has stored samples of polio vaccine since 1976. Fauci, director of both the National Institutes of Allergy and Infectious Diseases and NIH's Office of AIDS Research, told me: "If there are {polio vaccine samples} from back then, it would seem reasonable to go back and test them using our modern techniques." Indira Hewlett, FDA's senior scientist conversant with the test in question, the polymerase chain reaction, or PCR, agrees that such testing is in order and would be the best way to resolve the question. Melnick and Robbins also say testing the suspect polio stocks would be a good idea that could put the current controversy to rest.

But Louis Sullivan, secretary of health and human services, FDA Commissioner David Kessler and other senior officials so far have ducked the issue and declined my requests to interview them.

Whether or not the Wistar Institute itself evaluates the vaccines, the question could be settled by multiple PCR and other tests, performed in independent labs by investigators of unquestioned integrity and stature outside the United States -- preferably in England and Switzerland.

Tom Curtis, formerly of Texas Monthly, is covering this subject for Rolling Stone and the Houston Post.