A National Institutes of Health committee declared in effect yesterday that gene-splicing experiments are no more dangerous than other kinds of biological experiments, and voted preliminary approval of an end to federal regulation of the experiments.

The Recombinant DNA Advisory Committee of NIH, after five years of administering federal regulations, adopted a preliminary motion to end all the regulations on experiments that mix the genes of different creatures, and instead to establish voluntary guidelines like those governing other possibly hazardous experiments in biology.

The proposal will now be published and sent out for public comments before the committee takes a final vote in January.

The committee's action appeared to be the next-to-last step in ending the gene-splicing controversy, which began in 1973 when biologists themselves went on record as saying some splicing experiments might be hazardous.

The gene-splicing technology, invented that year, mixes the genetic material of two or more species. Biologists originally feared that adding foreign genes might give a microbe new properties that could change it from harmless to disease-causing.

After tens of thousands of gene recombinations, and some experiments designed specifically to test whether splicing is dangerous, the debate at yesterday's meeting centered chiefly on the social and political consequences of removing federal regulations, not on any scientific assessment of risks. There has been little public objection to such experiments except in Boston.

"There is no justification left, if there ever was one, no scientific justification for recombinant DNA rules ," said David Baltimore of MIT, who brought the proposal to the committee.

He said the public debate and furor over gene-splicing was appropriate, even though there might have been no scientific validity to the fears of biologists and the public. He said his proposal, which was approved on a vote of 16 to 3, still takes into account the "need to be sensitive to the political and social factors."

The approved proposal has several key parts: first, that the federal regulations become merely guidelines, without provisions for penalties if researchers violate them; second, that the guidelines ask researchers to use already-existing guidelines when working with hazardous organisms such as tumor viruses or organisms that produce poisons; third, that some experiments involving the most hazardous organisms worked with in any laboratories not be done at all except under unusual circumstances.

The NIH committee would be retained to oversee the guidelines and review experiments voluntarily referred to it by universities or industry. But the 200 or so local bio-hazard committees established under the NIH rules would no longer be required, though universities are expected to retain them for local oversight of biohazards.

One of the dissenters, Richard Goldstein of Harvard University, said he was "not convinced that there aren't any more risks." He named, as an example, reproducing the genes for making biological poisons, such as botulism toxin, in other microbes that have never had the ability to make such poison.

He acknowledged that his position over the years of the debate has changed very much, from great worry about the technology to a belief that the federal rules should be relaxed in great measure, except for a few possibly dangerous experiments.