Nobody knows for certain whether it was the work of Harvard dons, World War II, stinky shoes or diligent researchers.

Whatever the case, a fungus that officials believe has subdued the gypsy moth now is well established in parts of the Northeast and mid-Atlantic, granting what one official called a "reprieve" to those regions' trees.

In the early 1980s, gypsy moths appeared likely to infest whole forests and gnaw off their leaves. But this year marks the fourth spring in a row that Virginia has not had major spray programs or any reported defoliation by the moth, circumstances also reported in other states.

That is a far cry from the days in the late 1980s and early 1990s when the Warrenton-based state control teams supervised the spraying of Northern Virginia counties in what appeared to be low-grade war.

The insect still is present in small numbers locally and is spreading north, south and west. But invasions are tamer than they were during the 1980s, when moths roared down the Blue Ridge Mountains.

"That fungus really did a number on them," said Frank Fulgham, supervisor of field operations for the Virginia Department of Agriculture and Consumer Services Office of Plant and Pest Services.

The fungus, called Entomophaga maimaiga, takes about a week to kill the moths during their caterpillar stage in the spring, entering through the skin and infesting the guts. Ann E. Hajek, a Cornell University professor and one of the lead researchers on the fungus, has proposed several theories about how it was spread.

Under one theory, a Harvard researcher who isolated the fungus introduced it in parts of the Northeast in 1910 and 1911.

Under another theory, returning troops from World War II brought back the fungus, thought to be native to parts of Asia. Still other theories have the fungus spores carried into this country on shoes or by air.

In 1985, 1986 and again in 1991, the U.S. Forest Service and the Virginia Department of Agriculture deposited the fungus in parts of Virginia.

Some speculated that it didn't damage the gypsy moth population until 1995 because the population wasn't large enough until then.

"You have to have a certain number for it to spread," Hajek said.

According to state figures, the first defoliation was reported in 1984, involving 374 acres. That year, 8,000 acres were treated. Five years later, 213,987 acres were reported defoliated and 345,102 treated.

In 1995, 839,854 acres were defoliated and 269,190 treated. The next year presented a startling development when there were no trees that appeared defoliated.

"It was just a tremendous relief," said Tom Finn, Warrenton-based supervisor for the state's Office of Plant and Pest Services. The population "just dropped overnight."

That relief has continued, and communities no longer have mass sprayings. Just two weeks ago, Finn said, Prince William County sprayed 1,330 acres, the first such treatment in Virginia since 1996.

But according to Tim Bridges, assistant director of the Fauquier County Department of Environmental Services, "the guard is always there."

Researchers and federal officials say it better be, since the nature of how the fungus spread and whether it will maintain itself is uncertain.

For now, the area faces a different menace: the tiny green worms, properly known as cankerworms, that are devouring trees in southeastern Fairfax County and in woodsy pockets across Northern Virginia.