American experts on the identification of human skeletal remains say there could be many clues -- some conclusive, some only suggestive -- as to whether the bones purported to be Josef Mengele's are really his. But it is possible that all the evidence, taken together, will not prove the matter one way or the other.

Teeth offer the most reliable evidence, according to Clyde C. Snow, a University of Oklahoma forensic anthropologist who is one of the world's top experts. Because each person's teeth have distinctive patterns of ridges and grooves and because they are the hardest parts of the body, they change little with age. As a result, old X-rays or molds, if they are part of the German dental record that reportedly is being made available to the Brazilian investigative team, can be compared with teeth in the Brazilian skull.

"If you could get Mengele's dental X-rays, maybe from his days back in the German Army, you'd have it made with this skeleton," Snow said. Assuming that the dental records involve the same few teeth that remain in the skull, it would be possible to determine with certainty whether the dead man was Mengele.

Without adequate dental records, Snow said, other old X-rays, such as chest X-rays, can be compared with new X-rays of comparable bones to look for details of shape and internal structure that differ from person to person.

Without such highly individual information, Snow said, the best that can be done is to collect a number of specific features on the skeleton, forming a combination that would be highly unlikely to show up in two people.

Often the three easiest facts to establish, said Lawrence Angel, a forensic anthropologist at the Smithsonian Insitution's Museum of Natural History, are a skeleton's sex, age at death and stature.

Gender can almost always be established from the shape of the opening in the pelvic bones that acts as the birth canal in women. It is larger in women than in men. There are other shape differences in the pelvis, and the skulls of men and women differ in thickness, women's being the lighter.

Age is fairly easy to tell if the person was younger than the mid-50s. The chief clue is in joints between the closely fitting bones of the skull and of the pubic region. As people age, these joints tend to harden or fuse. After the mid-50s, however, further change occurs very slowly, and its rate is slowed even more in people in good athletic condition.

If the skeleton is really that of Wolfgang Gerhard, the man whose identity Mengele reportedly assumed just before the 1979 drowning, it would be that of a man only 53 or 54 years old. Mengele would have been 68 in 1979. The age difference may be too small to resolve using conventional methods.

However, a newer method has been developed by Ellis Kerley of the University of Maryland to use on older skeletons. Bits of bone are ground up and examined for special bone-forming cells. The older a person is, the fewer of these there are.

Height is a more straightforward matter. There are standard formulas for converting the length of leg bones or arm bones to stature. If upper and lower leg bones are complete, it should be possible to estimate the dead person's height at young adulthood to an accuracy of an inch or so. This could be compared with Mengele's military records.

A complete forensic investigation would also include a detailed examination of Mengele's old medical records and interviews with friends and relatives who could provide information about injuries that would leave lasting marks on the bones. Mengele, for example, is reported to have suffered a broken hip in Germany. Wilmes Teixeira, the internationally recognized Brazilian forensic pathologist leading the investigative team, has reported seeing an abnormality on the pelvic, or hip, bone that could be the remnant of this injury.

Methods also exist for determining the skeleton's blood type by analyzing the bone.

Still another comparison is possible if pathologists can get sharp photographs of Mengele. The skull can then be photographed from the same angle and the two images superimposed. Snow estimates that there is less than a 5 percent likelihood that the wrong skull will match up in contour with the photographed face.

These experts expressed high confidence in the ability of the Brazilian investigative team.

"If you get enough of these little points together," Snow said, "each one helps you whittle down the proportion of the population that your skeleton could represent and you come up with a composite that's too close to be coincidental."

But, Angel noted, the result could be a mix of matches and mismatches. "I think there's a good chance they will not be able to decide definitely," he said. "But then I have to confess that I'm biased. I hope he's still alive and that they catch him. He should be punished."