A special panel of the National Academy of Sciences concluded yesterday that a Dallas police tape recording made at the time of President John F. Kennedy's assassination does not support the finding that there was a second gunman.

The study contends that the noises on the tape previously identified as gunfire "were recorded about one minute after the president was shot and the motorcade had been instructed to go to the hospital."

The report was the latest, but far from final, chapter in the unending controversy over the 1963 assassination in Dallas' Dealey Plaza. It flatly contradicted the experts who told the House Assassinations Committee in 1978 that their tests of the same recording showed "beyond a reasonable doubt" that a second gunman fired at the president from the area of the so-called grassy knoll.

By contrast, the new panel of experts, headed by Norman Ramsey of Harvard, said after a 1 1/2-year study that "The acoustic analyses do not demonstrate that there was a grassy knoll shot, and in particular, there is no acoustic basis for the claim of 95 percent probability of such a shot."

The noises on the tape, Ramsey added in a telephone interview, are "probably static."

The experts for the House committee had described the sounds as gunshots, three from the Texas School Book Depository behind the president and one from the grassy knoll to the right of Kennedy's motorcade.

"The chances of its being static or other noise are higher than previously estimated," Ramsey said. Higher than gunshots? "Yes, okay, higher than gunshots," he said.

The study came in for some instant criticism, and will undoubtedly come in for more in the interminable debate over who killed Kennedy. One of the experts for the House committee, James Barger, said one aspect of it seemed particularly "disingenuous."

The noises from Dealey Plaza at the time of the assassination were believed to have been accidentally recorded when a transmitter on a police motorcycle in the presidential motorcade stuck.

The new study made much of the fact that no police sirens can be heard on the recording for almost two minutes after the time the House experts delineated as that of the assassination.

"Many witnesses agree that sirens were activated shortly after the final shot and as the motorcade speeded up for its dash to Parkland Hospital," the report said. "The complete absence of siren sounds for two minutes is difficult to explain . . . . "

The report made no mention of the explanation given by the House committee: a United Press International photograph at the time showed that the Dallas police officer whose transmitter had stuck, H.B. McLain, had lagged behind after the shooting, still in Dealey Plaza alongside the grassy knoll while the motorcade sped off ahead.

Committee lawyers concluded that McLain then caught up with the motorcade, because siren sounds can be heard on the recording about 1 minute and 50 seconds after the presumed time of the shooting.

McLain has insisted that his siren was going, too, but the official police plan for the motorcade indicated that only three advance officers were to use "sirens when needed." McLain was not one of them.

The Ramsey panel said it also detected some "hold everything" talk on the recording--apparently the Dallas sheriff's instructions to his men in response to the assassination. The report pinpointed those words as having been uttered at the same time the gunshots were supposed to have been fired, clearly an impossibility.

Using another more distinctly enunciated segment of conversation from the two-channel recording, the National Academy's panel said that the supposed gunshots occurred after--not before--the Dallas police chief could be heard saying, "Go to the hospital."

As a result, the scientists concluded that the supposed gunshot noises "were recorded about one minute after the assassination and, therefore, too late to be attributed to assassination shots."

Other experts have suggested that a different but equally plausible alignment of who said what and when on one channel, in relation to the supposed gunshot noises on the other channel, would support the conclusion of the House committee's experts that the noises were gunshots.

One of the firms that did the House studies, Bolt, Beranek & Newman of Cambridge, Mass., said it has not had a chance to study the Ramsey panel's report in any detail. For now, the firm said simply that it "raises a scientific controversy about complex technical issues."