The idea of reading molecular signatures to ascertain the degree of kinship between species is simple: a common parent gives rise to 2 daughter species which, following Darwinian logic, gradually become unlike each other with each successive generation, acquiring genetic mistakes or mutations. The earlier the time of divergence, the greater the accumulated difference. And if the mutations accumulate regularly through time, they are manifested as changes.

Biochemists have a name for these changes -- the molecular clock. By measuring the degree of accumulated difference between 2 related species, the time since they diverged can be calculated. Molecular studies show that 99 per cent of the genes of humans and chimpanzees are identical.

While the idea of a molecular clock is relatively new, the idea of charting biochemical kinships among species was thought of over 80 years ago by George Nuttal, a British doctor who concluded as early as 1902 that immunological tests could be used to determine the degree of blood relationship between animals. But it was not until the late '50s that Morris Goodman at the Wayne State University School of Medicine in Detroit demonstrated the close genetic relationship between humans and African apes.

Then, in 1967, Berkeley biochemists Allan Wilson and Vincent Sarich announced that molecular evidence showed that humans and apes diverged some 5 million years ago, and not 15 million years, as most fossil-hunters believed. Supporters of the molecular method argue that it is superior to the study of fossils because it draws the family tree of living species. As Sarich has said, " I know my molecules had ancestors; the palaeontologist can only hope his fossils had descendants."

But critics insist that the molecular clock gives only part of the picture because molecules cannot reveal what the creatures looked like. They further argue that molecular methods are of no use in interpreting fossils, as there is no way of knowing whether an extinct animal left any living descendants. If it did not, its branch would not even appear on the molecular tree.

That latter argument has been answered by the work of Jerold Lowenstein, a specialist in nuclear medicine, who identified the infinitesimal amounts of protein that remain in ancient fossils. In 1982, he proved that Ramapithecus, long thought to be our earliest ancestor, was in fact an ancestor of orangutans.

The fossils were too old for him to extract any protein directly, so he ground part of the fossils into a fine powder and injected it into a rabbit. Even after a gap of 8 million years, the rabbits' immune system was apparently able to make antibodies against the proteineaous dust. When Lowenstein tested these antibodies, they reacted strongly to proteins from orangutans, nearly as strongly to gibbons and gorillas and less strongly to that of chimpanzees and humans. These results dismissed Ramapithecus from the long line of protohumans.