AMHERST, Mass. — Researchers say a nearly forgotten fungus loosed nearly a century ago has saved a large swath of southern New England and eastern New York from voracious gypsy moth caterpillars for a second straight year.

“It’s such a brand-new phenomenon that we really know little about it,” said Joseph Elkinton, a professor of entomology at the University of Massachusetts and the state’s chief gypsy moth researcher. “We didn’t even realize the fungus existed until last year.”

Elkinton and his counterparts in Connecticut, southern Vermont and New York said that in many areas they had expected to be hard hit by the caterpillars this spring. Instead they found healthy trees and dead and dying caterpillars infected with the fungus Entomophaga maimaiga.

Up to 90 percent of the caterpillars in some areas of western Massachusetts that Elkinton checked were wiped out by the Japanese fungus. The fungus was introduced in Boston in 1910 in the hope of imposing a natural control on the gypsy moth.

The hairy brown caterpillars of the gypsy moth, a European species, have been munching on American oaks and other hardwoods since some escaped from a French naturalist who was experimenting with silk worms in Medford in 1869. Since then they have spread across the country to the Pacific. Major defoliation has been reported this summer in Pennsylvania, New Jersey, Virginia, West Virginia, Maryland, Ohio and Michigan, according to the U.S. Forest Service.

But while it took only 20 years after the moth’s release in Massachusetts for the first major infestation to be reported, it was more than 80 years before the fungus was noticed last summer by researchers in Connecticut and identified by Ann Hajek, of Cornell University’s Boyce Thompson Institute.

“Initially we thought it was because of the extremely wet spring last year,” she said. “But that hasn’t been true this year in New York and we are still seeing it.”

May was wetter across most of the region. But June, when the caterpillars began dying off from the fungus, was not, Elkinton said. Ronald Weseloh, a researcher at the Connecticut Agricultural Research Station, agreed.

“May may be the critical month,” Weseloh said.

“There’s no good answer” as to why scientists overlooked the fungus for so many decades, Hajek said. But without microscopic examination it is almost impossible to distinguish a caterpillar that succumbed from the fungus from one that died of a viral infection that kills off the creeping larvae when their numbers reach high levels.

“You almost have to know what you are looking for before you will find it,” she said.

“Last year was so wet and there were so many caterpillars dying from it that it almost came up and hit you in the face,” said Weseloh, one of the first researchers to suspect the fungus.

Its effect on the caterpillars may not be anything new.

Statistical studies of the rise and fall of gypsy moth caterpillar populations in Massachusetts, which has been charting their raids for more than a century, showed they do well in dry years and die off in wet ones. ” We never knew why, but the fungus may be the answer,” Hajek said.

Its potential use as a control, however, is yet to be fully explored.

Armed with soil infected with the fungus that was provided by researchers at the University of Massachusetts and some she collected in Westchester County, N.Y., Hajek introduced the fungus to an area around Ithaca, N.Y., where it killed off the unwanted caterpillars this spring.

“So we know it can be brought into a new area effectively,” she said. “But there is still a lot to be learned about its spread and effectiveness,” she said.

Last year researchers detected the fungus from southern Vermont down to New Jersey and west as far as New York’s Delaware River Valley. “This year I’ve gotten some samples from Delaware and Maryland,” she said, but none from central Pennsylvania, which has also been hit hard by the caterpillars.

“It seems to be spreading behind the moth. But it moves slower and doesn’t invade until some time after the first moth infestations,” she said.

“It may not be an important factor,” Elkinton said. “But certainly in rainy years it can make a difference and it would make a lot of sense to try to introduce it into areas where it does not now occur.”