Human ancestors made ancient entry to Java.

A group of humanity's predecessors known as Homo erectus had reached eastern Asia by 1.5 million years ago and inhabited the region until about 1.0 million years ago, a new investigation finds.

The findings, based on chemical analyses of fossil-bearing soil on the Indonesian island of Java, lend weight to the theory that H. erectus ventured into eastern Asia long before most anthropologists traditionally believed. Geologist Roy Larick of URS Corporation in San Francisco and his colleagues focused on relatively undisturbed sediment layers that alternate with layers of volcanic ash. Other researchers have excavated H. erectus remains throughout these layers, which lie in central Java.

Until about a decade ago, anthropologists assumed from piecemeal evidence that H. erectus reached Java no more than 1 million years ago.

Now that the new findings push the occupation of Java by H. erectus back by 500,000 years, "the big question is, When did hominids first enter this area?" says anthropologist Russell L. Ciochon of the University of Iowa in Iowa City, who is a coauthor of the new study.

The researchers examined a geological deposit called the Bapang formation. It consists of stream-borne layers of earth interspersed with sheets of hardened volcanic ash. Eruptions of one or more nearby volcanoes produced the ash. Most Javanese H. erectus fossils have been found in this formation.

Larick's group measured the relative proportions of two forms of argon in volcanic material from throughout the Bapang formation. This technique yielded estimates of the time that has passed since the ash cooled.

Age estimates ranged from 1.51 million years for the lowest layer that the researchers sampled to 1.02 million years for the top-most sample. Progressive declines in estimated age from lower to higher ash layers strengthen confidence in the reliability of the results, according to the researchers.

H. erectus fossils that other researchers have found beneath the Bapang formation must be older than 1.5 million years, Larick and his colleagues say.

Another team, led by geochronologist Carl C. Swisher III of Rutgers University in New Brunswick, N.J., has used the same method to calculate age estimates for the Bapang formation. His findings agree with those of Larick's group.

Swisher described his work 2 years ago at a scientific meeting. But he won't submit his findings for publication until researchers resolve discrepancies between the new argon results and other evidence.

For example, volcanic ash placed at 1 million years old by both Swisher and Larick's teams has been estimated by other scientists to be about 800,000 years old. The latter age rests on measurements of paleomagnetic reversals, the rate of uranium fission in the sediment, and argon dating of meteorite fragments associated with the ash.

Swisher has also reported argon dates of more than 1.6 million years ago for H. erectus fossils found elsewhere on Java (SN: 3/5/94, p. 150).