The Environmental Protection Agency has decided to allow release of a living, genetically engineered microbe into the environment for the first time. But while declaring the microbe -- designed to protect strawberries from frost -- safe enough to release, the agency has decided to make further tests of its risk, EPA sources said.

The antifrost work will be done by Advanced Genetic Sciences Inc. (AGS) of Oakland, which will place genetically engineered bacteria on strawberry blossoms in a California strawberry patch next month in hope that the microbes will block early frost damage.

The EPA considers the experiment safe even without additional testing, and is expected to announce its approval on Thursday. "We're doing additional tests not because we think there is a risk," an EPA spokesman said. "But we are talking about a first here. We thought it would be good policy to do all the tests we could."

The EPA decision is welcome news for biotechnology companies that are eager to use gene-splicing techniques to produce pesticides, disease-resistant crops, new breeds of farm animals, microbes to eat oil spills and other products with an estimated potential market in the billions of dollars.

The biotech industry hopes the EPA decision will clear the way for approval of other similar experiments and emerging products.

Word of the decision has already stirred opposition among environmentalists and renewed concern on Capitol Hill about the Reagan administration's regulation of biotechnology. Two congressional panels have scheduled hearings to examine the administration's methods of evaluating the AGS experiment and others that would release genetically altered organisms into the environment.

Few scientists believe the newly engineered microbe will create any environmental hazard. But some leading scientists, and EPA staff members as well, have told the agency that there are still unanswered questions about its risks.

Many who do not oppose this first release say nevertheless that the scientific work needed to make decisions in general about what is risky to release into the environment does not exist.

In reviewing the application, the EPA's Hazard Evaluation Division concluded that the new microbe will probably get outside the strawberry patch despite precautions and could survive indefinitely on plants outside the plot along with many other naturally occurring bacteria that cover leaves, according to EPA sources.

Thus the experiment will not be confined to the two-tenths of an acre as planned but could amount to an experiment on the whole Salinas Valley area around the test plot.

Martin Alexander of Cornell University, chairman of EPA's key advisory board on the topic, also told the EPA confidentially that the company should not be allowed to go ahead without more tests to determine what will happen to the microbe once it gets into the environment.

The agency concluded, however, that company tests showed that the new microbe is unlikely to replace any natural plant bacteria and thus is not likely to cause any harm.

Another key EPA adviser, Robert Colwell of the University of California at Berkeley, agreed but told the EPA that in the final analysis the agency would have to accept the company's assurances that the microbe, once outside the test plot, will do no harm to plants or animals.

In greenhouse tests, AGS President Thomas Dyott said the key fact was that the bacteria did not survive in large numbers for very long. After three months, the bacteria diminished to levels undetectable by the company's methods. Thus it is assumed that the "designed" bacteria are also unlikely to compete strongly with the naturally occurring bacteria outside the greenhouse.

The major tests added by the EPA in the last few days will monitor the air from a few feet to a few yards above the strawberries and gauge how many of the bacteria are carried by wind to areas outside the test plot.

EPA officials say the tests, considered to be relatively expensive, are a sign that the agency is trying to be as cautious as possible in the precedent-setting case.

In addition to the debate and criticism that EPA has encountered so far, the agency will face a lawsuit to block the experiment at the same time that AGS approval is announced, according to Jeremy Rifkin, head of the Foundation on Economic Trends.

The approval is "inappropriate and irresponsible," Rifkin said, because the EPA has not completed its own program to design tests to judge the risk of environmental release experiments.

In the worst-case scenario, critics envision the possibility that the microbe might escape the test plot and spread from plant to plant over a wide area, displacing natural bacteria of a similar type.

Many plants could then have longer growing seasons because of frost protection, and decay less readily. That would force changes in the habits of creatures from bacteria to animals that feed on the plants. Disruption of the food cycle or a change in pests could result.

Further, there has been a suggestion that if vast areas were sprayed with the new bacteria (which is a variety of Pseudomonas syringae), enough would rise into the air to alter an area's cloud and rain patterns.

What makes these scenarios unlikely, proponents say, is that the microbe won't multiply and become a major part of the general bacterial population, nor is there much evidence suggesting the worst cases are possible.

AGS has been waiting two years to test its pesticide while U.S. agencies, the courts, congressional committees and a White House task force studied the question of how the government should regulate the commercial products of gene-splicing.

Some members of Congress have doubts about the ability of the executive branch to assess the environmental risks of releasing genetically altered organisms.

Probably the next environmental release application before the EPA is one from Monsanto Co.

"We're very encouraged by what we see happening," Leonard J. Guarraia, director of regulatory and scientific affairs for Monsanto, said. The company wants to test a genetically engineered pesticide in the field. "The agencies are establishing a regulatory structure that is credible, science based, and asking the questions that need to be asked," he said.