Homo erectus in Java: A 250,000-Year Anachronism
The story of human evolution has lately become as complicated as a Tolstoy
novel. It used to be a simple and lonely tale of one species living at one
time and evolving gradually into another species. Even the renowned
geneticist Theodosius Dobzhansky wrote in 1944 in the American Journal of
Physical Anthropology that "as far as is known, no more than one hominid
species existed at any one time level."
But lately, this tale has become thick with new subplots and characters.=20
The recent recognition of several different species of early Homo living=20
in Africa about 2 million years ago, and various forms of Australopithecus
before that, has made it clear that there were far more lineages in the
early history of the human family than previously believed. Now, it seems
that later chapters may also have to be rewritten to include at least one
more character: a relative who makes a surprise reappearance long after it
was presumed dead.
In a report on page 1870, an interdisciplinary team of scientists suggests
that one relative, H. erectus, was still alive in Java, Indonesia, as
recently as 27,000 to 53,000 years ago--at least 250,000 years after it was
thought to have gone extinct in Asia. If so, this remnant population of H.
erectus, a species that first appeared in the fossil record about 2 million
years ago, would have been alive when modern humans and Neandertals roamed
the earth. The team, led by geochronologist Carl C. Swisher of the Berkeley
Geochronology Center, came to this stunning conclusion after redating two
important fossil sites in Java. "If the dates are right, we have three
different species coexisting at the same time," says Chris Stringer, a
paleoanthropologist at The Natural History Museum, London. It also means,=20
says American Museum of Natural History paleoanthropologist Ian Tattersall,=
that "being alone on Earth is unusual."
The new dates also cast doubt on a theory, held by some
paleoanthropologists, that the Java people were the ancestors of the first
Asians, including the first Australians, who appeared in the fossil record
about 40,000 years ago. It is therefore no surprise that these findings are
proving controversial. "There is a real problem with the dates," claims
Australian National University paleoanthropologist Alan Thorne, who notes
that the site is notoriously difficult to date. And even if the dates hold
and others argue, they wouldn't necessarily rule out the Java-Australian
Swisher, however, is confident of the findings, which come mainly from a
fossil site on a terrace above the Solo River in Java, in the village of
Ngandong. This is where Dutch archaeologists discovered the faceless skulls
of 12 prehistoric humans between 1931 and 1933. Their identity mystified
anthropologists from the start: Over the years, individual skulls were
variously identified as that of a tiger, an ape, a Neandertal, and a modern
human. But in recent years, most anthropologists who have seen them agree
that the Solo skulls belong to H. erectus--making them the largest
assemblage of this early human ever found, says University of Florida
paleoanthropologist Susan Ant=F3n, a co-author of the paper.
It has been even more difficult to pin down their age. No teeth--the best
material for dating--were found with the skulls, and the keeper of the
fossils, Teuku Jacob of Gadjah Mada University in Java, has not allowed
chips to be removed from the skulls for direct dating. A French team did
date one ofthe skulls years ago using an experimental method that doesn't
damage the specimen. They came up with an age of at least 300,000 years,
says geologist Christophe Falgu=E8res of the National Museum of Natural
History in Paris, where the skull was dated. The group has had problems with
the dating, however, and has yet to publish results.
For the most part, geologists have had to use indirect methods to date the
fossils--by dating volcanic rock at the site or the teeth and bone of
animals found in the same fossil-bearing layer as the skulls. The results
have varied widely. A Japanese team in 1985 dated volcanic rock near the
site to about 250,000 years. But in 1988 came tantalizing hints that the=20
skulls might be much younger when a Dutch team published ages of about=20
100,000 years for nearby animal bone chips.
Swisher, who is systematically using state-of-the-art argon/argon methods to
redate volcanic rock at early human fossil sites in Indonesia, had hoped to
settle this question. First, he and other geologists ruled out the Japanese
team's date because it was on rock that, he says, "had nothing to do with=
skulls." After searching in vain for a layer of volcanic minerals closer to
the fossils, he turned to other material for dating. He dug a test trench at
the site to confirm that the half-meter layer of sandstone and clay cobbles
that had yielded the skulls was the only fossil-bearing layer, and collected
water buffalo teeth from the same level where the skulls were found. He also
collected bovid teeth from a nearby site, Sambungmacan, where remains
thought by many to be H. erectus have been found.
He sent those teeth, as well as bovid teeth that had been collected with the
human fossils in the early 1930s, to McMaster University in Hamilton,
Ontario, for dating by geochemist Henry Schwarcz and geologist Jack Rink.
Their lab is one of a half-dozen that uses the relatively new method of=20
electron spin resonance (ESR), which measures the electric charges induced=
in tooth enamel over time by radioactive materials, such as uranium, in the
surrounding soil. When the team came up with its surprising results--a
minimum age of about 27,000 years and a maximum age of 53,000 years--they
tried a different technique: uranium series dating, which measures the
clocklike radioactive decay of uranium to thorium in the teeth to date the
enamel. The results checked out.
Despite those precautions, some other geologists remain skeptical. The
animal teeth themselves contain high levels of uranium, which can throw off
ESR dating, according to studies by Falgu=E8res's group at a site in Italy.
Another concern raised by Thorne and others is that the animal teeth may not
be the same age as the skulls, because flooding or erosion at the site may
have washed older skulls out of their original resting place and into the=20
young fossil bed.
Schwarcz and Rink respond that there are geologic problems at the site in
Italy, where the volcanic minerals used to check the ESR dates might have
been swept into the site by water--and have been much older than the teeth.
As a result, there is no body of evidence to make them doubt the reliability
of using ESR to date teeth with high uranium at other sites. Also, they say
they got the same dates on all the teeth from the same layer at two sites--
with two dating methods. "We're very confident about these dates," says
Schwarcz. As for the flooding theory, Swisher's team points out that it's
hard to imagine how 12 crania and other human remains could have moved to
the same level and at two sites. Moreover, the uniformity of the dates and
earlier fluorine analysis of the hominid bone and bovid teeth show that they
had undergone the same amount of decay and, hence, were of similar age.
Nonetheless, say Swisher's collaborators, he kept trying to find alternative
explanations for the young age for H. erectus. "Every time the dates came
back, Carl [Swisher] would say, 'Are you sure these aren't H. sapiens?' "
says Ant=F3n. "And I'd go back to Indonesia to have another look." What=
saw convinced her that the Solo skulls had not been misidentified. "They are
unequivocally H. erectus," agrees Philip Rightmire of the State University
of New York, Binghamton, who examined them in March.
These findings may spell trouble for the regional continuity model of human
origins, which proposes These findings may spell trouble for the regional
continuity model of human origins, which proposes that H. erectus in Asia
was among the ancestors of modern humans in Asia. A competing--and more
widely held--theory is that modern humans evolved in Africa and displaced H.
erectus as they migrated around the globe. A key prediction of the regional
continuity theory is that the Java people were among the ancestors of modern
Australians. But, as Swisher points out, his new dates imply that the Solo
people may be younger than the first Australians, arguing against a direct
Thorne and University of Michigan paleoanthropologist Milford Wolpoff--a
leading proponent of the regional continuity theory--disagree. "There is a
great, long list of characters that are the same in the Solo skulls and the
earliest known human people from Australia," says Thorne, including the
unusually large size of the Solo brain cases. Thorne also argues that even
if the Java fossils are relatively young, they look so much like the=20
Australian fossils that they may have shared a recent ancestor.
To try to nail down the dates of the Solo people more firmly, Swisher is
planning to return to the Java site next year to look for organic material
to date with radiocarbon methods and soil carbonate for uranium series
dating. But until he finds such material, or until Jacob allows direct
dating of the human fossils with proven methods, Swisher says, "we have two
options: Sit on our data, or say that we've done the best we can with the
technology available to us and throw them out there for people to evaluate."
Given the startling nature of the data, there will be no shortage of=20
Volume 274, Number 5294, Issue of 13 December 1996, pp. 1841-1842
=A91996 by The American Association for the Advancement of Science.=20
THE ORIGINS SOLUTION