Washington ? Scientists have discovered possible bird flu in two wild swans on the shore of Lake Erie – but it does not appear to be the much-feared Asian strain that has ravaged poultry and killed at least 138 people elsewhere in the world.

It will take up to two weeks to confirm whether the seemingly healthy wild mute swans in Michigan really harbored the H5N1 virus or not.

On Monday, the Agriculture Department declared that initial testing had ruled out the so-called highly pathogenic version of H5N1 – but that they could have a relatively harmless, low-grade H5N1 strain instead.

That’s the suspicion, making Monday’s announcement almost a practice run for the day the more worrisome Asian strain actually arrives.

“This is not the highly pathogenic avian influenza virus that has spread through much of other parts of the world,” said Ron DeHaven, administrator of USDA’s Animal and Plant Health Inspection Service, adding, “We do not believe this virus represents a risk to human health.”

Monday’s announcement was the first reported hit from a massive new program to test up to 100,000 wild birds in an effort to catch the deadly Asian H5N1 virus if it does wing its way to North America, something the government thinks could happen this year.

Were the highly pathogenic H5N1 to be found in any wild birds here, that could trigger additional security steps to prevent infection of commercial poultry flocks, and even more intensive monitoring.

Wild birds, especially waterfowl, are flu’s natural reservoir – they carry a multitude of influenza viruses. Sometimes, those strains jump species, and if it’s a flu virus very different from one people have experienced before, a worldwide epidemic could result.

That’s why scientists have closely tracked the virulent H5N1 strain since it began its global march in late 2003. It is blamed for the death or destruction of millions of birds overseas. Virtually all the people who have caught it did so from close contact with infected birds or their droppings. But scientists worry that the virus eventually could mutate to become easily spread from person to person.

Last week, the government expanded the bird-testing program to encompass the entire nation, after initial sampling mostly in Alaska. Twenty mute swans from a Monroe County, Mich., game area were among the first new batches of tests – because, coincidentally, they were part of a state program to lower overcrowding of the nonnative species. That testing found the possibility of H5N1 in two of the swans.

Initial genetic testing ruled out the deadly Asian strain. In fact, USDA said the virus’ genes suggest that it is similar to a low-grade North American version of H5N1, a virus found here in wild ducks in 1975 and 1986 and on a Michigan turkey farm in 2003. Another similar version was detected last year in Canada, and scientists have thought it probably common in wild birds – but didn’t have the testing to prove it.

“This is no surprise,” DeHaven stressed.

Plus, all the swans appeared healthy, a good signal, he added. The virulent form of H5N1 usually rapidly sickens birds.

So why Monday’s announcement? To be open about all this testing, DeHaven said. And even low-pathogenic H5N1 requires monitoring, because it has the potential to mutate into the more virulent form, he added.

More important, “It was a real good test run of the system,” Dr. Willie Reed, director of the Michigan State University laboratory where the initial testing was done, said.