Thousands of years ago, in North America's past, all of its megafauna—large mammals such as mammoths and giant bears—disappeared. One proposed explanation for this event is that when the first Americans migrated over from Asia, they hunted the megafauna to extinction.These people, known as the Clovis society after a site where their distinctive spear points were first found, would have been able to use this food source to expand their population and fill the continent rapidly.Yet many scientists argue against this "Pleistocene overkill" hypothesis. Modern humans have certainly been capable of such drastic effects on animals, but could ancient people with little more than stone spears similarly have caused the extinction of numerous species of animals?Thirty-five genera or groups of species (and many individual species) suffered extinction in North America around 11,000 B.C., soon after the appearance and expansion of Paleo-lndians throughout the Americas (27 genera disappeared completely, and another 8 became locally extinct, surviving only outside North America).

Although the climate changed at the end of the Pleistocene, warming trends had happened before. A period of massive extinction of large mammals like that seen about 11,000 years ago had not occurred during the previous 400,000 years, despite these changes. The only apparently significant difference in the Americas 11,000 years ago was the presence of human hunters of these large mammals. Was this coincidence or cause-and-effect?

We do not know.Ecologist Paul S. Martin has championed the model that associates the extinction of large mammals at the end of the Pleistocene with human predation.With researcher J. E. Mosimann, he has co-authored a work in which a computer model showed that in around 300 years, given the right conditions, a small influx of hunters into eastern Beringia 12,000 years ago could have spread across North America in a wave and wiped out game animals to feed their burgeoning population.

The researchers ran the model several ways, always beginning with a population of 100 humans in Edmonton, in Alberta, Canada, at 11,500 years ago.Assuming different initial North American big-game-animal populations (75-150 million animals) and different population growth rates for the human settlers (0.65%-3.5%), and varying kill rates, Mosimann and Martin derived figures of between 279 and 1,157 years from initial contact to big-game extinction.

Many scholars continue to support this scenario.For example, geologist Larry Agenbroad has mapped the locations of dated Clovis sites alongside the distribution of dated sites where the remains of wooly mammoths have been found in both archaeological and purely paleontological contexts.These distributions show remarkable synchronicity (occurrence at the same time).

There are, however, many problems with this model.Significantly, though a few sites are quite impressive, there really is very little archaeological evidence to support it.Writing in 1982, Martin himself admitted the paucity of evidence;for example, at that point, the remains of only 38 individual mammoths had been found at Clovis sites. In the years since, few additional mammoths have been added to the list;there are still fewer than 20 Clovis sites where the remains of one or more mammoths have been recovered, a minuscule proportion of the millions that necessarily would have had to have been slaughtered within the overkill scenario.

Though Martin claims the lack of evidence actually supports his model—the evidence is sparse because the spread of humans and the extinction of animals occurred so quickly—this argument seems weak. And how could we ever disprove it?As archaeologist Donald Grayson points out, in other cases where extinction resulted from the quick spread of human hunters—for example, the extinction of the moa, the large flightless bird of New Zealand—archaeological evidence in the form of remains is abundant. Grayson has also shown that the evidence is not so clear that all or even most of the large herbivores in late Pleistocene America became extinct after the appearance of Clovis. Of the 35 extinct genera, only 8 can be confidently assigned an extinction date of between 12,000 and 10,000 years ago.Many of the older genera, Grayson argues, may have succumbed before 12,000 B.C., at least half a century before the Clovis showed up in the American West.