A young woman who died in Bangkok 10 days ago appears to have had the first confirmed case of person-to-person transmission of the "bird flu" that emerged in Southeast Asia last winter.

The woman is one of four members of an extended family who appear to have contracted the H5N1 avian influenza virus. At least two of them had contact with dead chickens -- the usual source of the rare infection -- but the 26-year-old victim had contact only with an ill child, not any birds, according to a World Health Organization report.

The finding is significant because the deadly H5N1 virus could become a public health threat when it gains the capacity to pass from person to person and not simply from bird to person.

Although such transmission seems to have occurred, there is no evidence it happens easily -- another condition necessary for the strain to become a major health threat. In the Thai case, the woman provided bedside care to her 11-year-old daughter, who also died.

"The hope is this is an isolated cluster," said Robert G. Webster, an influenza virologist at St. Jude Children's Research Hospital in Memphis. "If it goes to a second round from this family, then we are in trouble. If it goes from four to 16 people, then it's a great worry. Everyone is watching very closely."

Klaus Stohr, head of WHO's global influenza program, said an epidemiological investigation in the family's northern Thailand village points to this being a "non-sustained, inefficient, dead-end-street human-to-human transmission." He spoke at a briefing yesterday in Geneva.

Investigators believe the mother must have gotten the virus from her daughter because the mother lived in Bangkok, came to the village to care for her child, then returned to Bangkok, where she became ill and died on Sept. 20. She had no known contact with birds.

There are more than a dozen possible cases of H5N1 influenza cases under investigation in addition to the four in the family.

Most experts believe the world is ripe for a flu pandemic, an event that occurs about every 30 years. Typically, a new strain of influenza emerges from an animal source, gains the ability to infect people, then rapidly spreads around the world because nobody has immunity.

The first evidence that a strain of H5N1 flu could infect human beings came in 1997 in Hong Kong, when 18 people became ill during an influenza outbreak in poultry. Studies of antibodies in the blood of relatives and health care workers who had close contact with those people indicated that a few also became infected, although they apparently were unaware of it.

The strain in Thailand is descended from the Hong Kong H5N1 virus but is distinct in behavior and genetic identity. The biggest difference is its lethality.

In the Hong Kong outbreak, six of 18 patients died. Since last winter's outbreak in Vietnam and Thailand, 30 out of 42 people have died.

Nancy J. Cox, chief of the influenza branch at the Centers for Disease Control and Prevention, said throat swabs and other specimens from the Thai patients arrived in Atlanta yesterday. Over the next week or so, scientists will attempt to isolate the virus and grow a pure culture of it.

The main thing they want to learn is whether the virus is identical to last winter's strain, or whether it has evolved. They hope to get an answer by analyzing the virus's genetic fingerprint and its behavior.

In particular, they want to know whether an experimental vaccine being made from the winter strain is likely to prevent infection by the summer strain.

As part of planning for a possible H5N1 pandemic, the Department of Health and Human Services has contracted with two drug companies, Aventis Pasteur Inc. and Chiron Corp., to make batches of vaccine, which will be tested in people in late winter or early spring. Ultimately, 2 million doses would be stored for future use.

Who would get the vaccine in the event of an outbreak is a question the government is pondering. It is seeking suggestions from the public.

Last winter's H5N1 virus was sufficiently different from an H5N1 virus that caused two cases of human illness in East Asia in 2003 that a vaccine for one would not fully protect against the other.

If it turns out this summer's virus is different from last winter's -- which scientists do not expect to be the case -- work underway on a vaccine from the winter strain would not stop, but plans to make a new vaccine might begin, Cox said.