Scientists have found surprising evidence that a "weakly contagious" human cancer virus may be present in as many as one of every 100 or 200 persons in the southeastern United States, "from Texas to Florida."

The virus lurks in many people but only rarely causes a cancer in this country. It may trigger no more than a few hundred cases a year, mostly but not entirely in blacks, of an often particularly virulent blood and lymph-gland cancer.

Such cases evidently exist "by the hundreds" in Japan, the Caribbean and probably Africa.

Either this virus--called "HTLV" for human T-cell leukemia-lymphoma virus--or a member of the same virus family is also "the best candidate yet" as a cause of acquired immune deficiency syndrome (AIDS).

These statements were made in interviews yesterday and in a series of articles today in the Journal of the American Medical Association.

Discovery of the HLTV virus was reported two years ago by Dr. Robert Gallo of the National Cancer Institute, who called it only a "suspected" cancer virus then.

Now the association is definite, according to articles by Drs. Douglas Blayney, William Blattner, Gallo and others in the AMA Journal and a report by NCI Director Dr. Vincent DeVita Jr., calling it "the first virus associated with a human cancer."

Scientists think that exposure to some other viruses, notably hepatitis-B and Epstein-Barr mononucleosis, is part of a chain of events that can cause cancer. But in no other case, DeVita's report said, "have virus particles actually been isolated from human cancer cells."

Gallo said that is because "infected cells tend to disappear" in cancer or AIDS.

But Gallo and others found either the virus or antibodies to the virus, evidence the body had been fighting it, first in four leukemia-lymphoma patients whose disease had characteristics of both leukemia (blood-system cancer) and lymphoma (lymph-gland cancer). Then they found it in still more patients.

In a series of recent studies, similar evidence was found in about two of 100 healthy persons in Georgia and about one in 200 in Florida.

Other "sporadic sampling," Gallo said, has shown that the virus is apparently endemic (always present in some persons) throughout the South and Southeast, with cases reported as far north as Michigan and New York.

There have been 24 such cases definitely diagnosed in the United States, about half of foreign origin, Blattner said. In parts of Japan and Caribbean, the disease is far more common, he said.

In Kingston, Jamaica, doctors studied the blood serum of every lymphoma patient other than those with Hodgkin's disease from January, 1982, to February, 1983. "Seventy percent were positive for this virus," Blattner reported. "Cases there were particularly virulent, with survival generally less than six months."

U.S. patients have done much better, and some may be cured, although Blattner said he is not sure why, because the Jamaican patients had similar treatment.

Japanese patients, he said, are found most often on Kyushu, the southernmost major island. "The bottom line," he said, "is that this virus is found in clusters. Also, people may carry the virus and remain, virtually all of them, healthy and happy throughout their lives."

Why some people contract cancer and others do not is unclear.

Gallo said he believes that "this virus is very poorly transmitted," requiring "intimate human contact and possibly transmission of whole cells, not just the virus , . . . by blood or other human fluids. Certainly, not everyone with the virus is going to get leukemia."