Scientists in Montana may have found a cure for Dutch elm disease in the form of a tiny bacterium taken from the leaves of ordinary barley and other grains.

Injected into the trees, the bacterium seems to prevent the disease in healthy elms and to halt its spread in those already infected, according to Dr. Gary Stroebel, plant pathologist at Montana State University in Bozeman.

"We're very close -- we have one year of tests that looks very encouraging, but we're waiting to see what happens this season," Strobel said. But he and his sponsor, the Freshwater Foundation of Wayzata, Minn., have already applied for a patent on both the bacterium production method and the injection process, and the Ortho garden products company has applied to lease the patents.

Dutch elm disease, Asian in origin but first isolated in Holland in 1917, has marched across the nation, slaughtering millions of stately, gray-barked American elms since its arrival in New York in 1930. It is a fungus that is carried by a shiny, reddish-black dark beetle the size of a pinhead.

The beetle is thought to have arrived here in burl elm logs imported from Europe for furniture-making.

The fungus causes the tree to produce antitoxins that clog its own internal vascular system, preventing water from moving. The tree literally strangles, often in three weeks or less. The new bacterium, Strobel explained, is antagonistic to the fungus.

"It was a whole new approach that hadn't been tested before," he said. "It was a matter of finding the right microorganism, since not many are antagonistic to fungi. Apparently we found the right one, but we're still tentative."

The technique has been tried on only 35 trees in scattered locations nationwide, Strobel said, mainly in South Dakota and Rhode Island. He said it had proved 70 percent to 80 percent effective so far, but that it is still too early to know whether trees injected two years ago will survive this spring's expected upsurge in beetle infestation.

Some young trees three inches or so in diameter had resisted infection after receiving the shots, while other trees showing the characteristic half-dead crowns seemed to be recovering, Strobel said. Previously, one beetle could infect a tree, and infection always was fatal.

Field tests of the bacterium injections are planned this month in Minnesota, Ohio, Illinois, New York, Utah and the District of Columbia in addition to Rhode Island and South Dakota, he noted.

"If things look good this summer, it won't make Dutch elm disease a thing of the past, but we should be able to save some of the biggest and choicest trees," Strobel said.

Injection itself is not a new idea. Pesticide and fungicide injections were common and seemed to be making some headway in the early 1970s, but ran into problems of side effects and then resistance on the part of the fungus.

Efforts to control the disease began making headlines nationwide as early as 1934, since most of America's main streets were planted with the shady, arching elms. By the 1950s the disease had become an epidemic as the beetles grew resistant to the DDT sprays that had been the main line of defense.

By the 1950s the disease had spread to 30 states, and newspapers recounted heartbreaking stories of landmark elms withering one by one. The rate reached 400,000 trees a year in 1975, despite the expenditure of more than $100 million annually by state and local governments.

In recent years, most funding has gone to tree removal programs as researchers gave up on one control technique after another. The U.S. Forest Service spent $2 million a year in the last two years just to help states remove dead trees.

But the Freshwater Foundation raised $400,000 from Minneapolis-St. Paul businesses in August 1977 to fund three more research projects, according to Dr. Joseph Rossillon, its director. One was on the beetle in Syracuse, N.Y., another was on a synthetic sex-attractant approach in Minnesota, and the third on the fungus in Bozeman, Strobel's project.

All three projects will continue, Rossillon said, because Strobel's bacterium, although "terribly exciting" at the moment, may run into the problems of fungus resistance or other failures that have plagued the 45-year-old effort so far.

Strobel would not reveal the name of his microorganism for fear, he said, that others might learn to cultivate it before he publishes his results, which he expects to do in about three months.

He said, however, that it normally is found on the leaves of wheat, barley or oats, and also on the leaves of a pear tree. It can be grown cheaply in large vats by a fermentation process, he said, and can then be freeze-dried and packaged for shipment.

"If it works," he said, "the customer will just have to add water and stick it in the tree."