Because Congress has not enacted legislation to control gene research, White House officials are asking American firms if they are willing to comply voluntarily with federal safety rules, presidential science adviser Frank Press disclosed yesterday.

In return, the chairman of a Senate Science subcommittee, Adlai E. Stevenson, accused the administration of "a lack of urgency" about the research-control bill it submitted in April.

And the Illinois Democrat said he was "skeptical" that voluntary compliance with safety precautions can work given "the temptations to which human flesh is susceptible" when "profit" and "commercial possilibities" enter the picutre.

The clash was the lastest in a seesaw effort to extend National Institutes of Health guidelines on "recombinant DNA" research to some of the nation's chemical and pharmaceutical giants. These guidelines are already compulsory for federally funded university research.

The research is designed to combine unrelated bits of genetic material to create new forms of life. Some firms hope to use the new genetric material to manufacture novel drugs or other products. Some scientists fear such efforts could create new disease organisms that would result in epidemics.

The administration's proposal in April was followed by House and Senate committee versions with differing provisions. Many scientists who originally sought some controls began lobbying against what they thought were onerous ones.

Yesterday press told the Senate subcommittee that his staff and the Commerce Department are "exploring the feasibility of voluntary compliance" only because "it appears certain now that no legislation will be passed" this year.

"We have talked" to some industrial officials, he said, and their assurances of compliance" make us ver comfortable" that "the public would be protected."

The NIH rules bar some experiments altogether - work with certain disease organisms, for example - and require that others be done only with organisms considered safe in sterile, enclosed laboratories.

The current administration view, said Press is that there is only a "small risk" of any untoward incident - for example, some new form of disease being set loose - although he acknowledged that such an incident, if it happened, might have "high consequences."

But another witness, anthropologist Margaret Mead, said "nobody knows what the risk are" yet, and all who speak of them speak in "ignorance." She pleaded for a slow approach and more limits on some experiments until more is known.

"We know similar guidelines" are often violated in many research and industrial fields, she said.

The subcommittee was told of an instance this year in which University of California at San Francisco researchers used an unapproved plasmid, or bit of DNA, in an attempt - later successful - to duplicate an animal insulin, gene.

The scientists destroyed their work to that date when they found that the plasmid was unapproved. But NIH first heard about it months later from a reporter. California scientist William Rutter said yesterday his group kept the information to themselves to avoid further "inflaming" the public over the genetic issue.

The plasmid the scientists destroyed was soon declared safe and approved by NIH. But Dr. Donald Fredrickson, NIH director, nonethless called the university's failure to inform NIH "disturbing" and "inexcusable."

Some environmental groups are expected to testify this week that such events are inevitable unless Congress passes a strong control bill with penalties for those who make their own rules.