A BATTERED old skull fragment from Java has reached across nearly 2 million years, figuratively speaking, to upset the current scenario of human evolution.
This is no ordinary fossil. It's the ``crown jewel'' of a collection that has defined the extinct species Homo erectus.
You'll find Java Man prominently placed in every museum display, video graphic, or textbook ``time line'' that depicts our ancestral tree. There he sits in the direct line of human development at the politically correct age of about 800,000 years.
How awkward, then, for two geologists now to show that Java Man is over twice that age and may not even be our ancestor after all.
Yet such is the power of radio-isotope dating, which, even when applied in a field as plagued with ambiguity as paleontology, Garniss Curtis could use to tell the press that Java Man's new age ``is a fabulously precise date.''
It's 1.81 million years, to be exact, give or take 40,000 years, at the Mojokerto site in Java, Indonesia. A second Java erectus site at Sangiran that Dr. Curtis and his colleague Carl Swisher have dated checks in at 1.56 million years with the same degree of precision.
The two geologists from the Institute of Human Origins in Berkeley, Calif., discussed their findings with reporters during the annual meeting of the American Association for the Advancement of Science (AAAS) here.
Their team's technical report appears in the current issue of the AAAS journal Science.
They explained that they date the erectus sites by dating the volcanic tuff in which the fossils were found. Volcanic crystals contain potassium, which has been decaying radioactively into argon at a steady known rate since the crystals solidified. That dates the tuff precisely.
Dr. Swisher noted that paleontologists still could question whether or not the fossils are the same age as the tuff. But he added that the question has already been well-explored since the fossil sites were first dated - incorrectly - using other means.
Accepting the new age in Java throws open the whole question of how and when human predecessors came out of Africa, as Swisher and Curtis explained.
A fossil site at Trinil in Java provides the official ``type'' definition of Homo erectus, they said. However, fossils identified as erectus also are dated as old as 1.8 million years in Africa.
The conventional scenario has erectus evolving for something like a million years in Africa, then moving out through Europe and Asia and evolving into one of our early ancestors.
But erectus in Java now is virtually as old as in Africa.
``So,'' Swisher says, ``the [old] scenario kind of breaks down.''
It opens such questions as:
* Are the African fossils truly erectus after all or a distinct species?
* When did a human predecessor come out of Africa?
* Does erectus fit in the human line, or is it a dead-end offshoot?
``We don't have a clue'' now as to what came out of Africa and when it came out, Curtis says. In fact, he adds, experts now are unsure of where erectus itself came from.