Read our COVID-19 research and news.

Third Time at Bat, But No Cheer for Mammoth-Killing Hypothesis

Proponents of the idea that an impact wiped out the mammoths and roiled early North American human culture have struck out, at least by baseball’s rules. Their third paper in a leading journal offering evidence of a devastating impact 12,900 years ago is, like its predecessors, failing to convince experts. The six experts who read the Proceedings of the National Academy of Sciences (PNAS) paper for Science either could find no convincing evidence for nano-scale diamond crystals of the sort an impact might produce, or if impact nanodiamonds are there, the researchers don’t see how they prove an impact. “I’m still not convinced diamonds have been found,” says research physicist Tyrone Daulton of Washington University in St. Louis in Missouri. “It could be [impact diamond], it could be something else.”

This PNAS paper, published 20 July online before print, is the third from the same group claiming evidence of impact nanodiamonds. The first claim, in an earlier PNAS paper and using carbon-13 nuclear magnetic resonance, proved to be baseless (Science, 7 March 2008, p. 1331). The second, in a Science paper using transmission electron microscopy (TEM), fared better, but most outside experts remained unconvinced that nanodiamonds had been found (Science, 2 January, p. 26).

In the latest PNAS paper, archeologist Douglas Kennett of the University of Oregon, Eugene, and 16 colleagues from 14 institutions report their discovery of nanodiamonds at a site on one of the Channel Islands off southern California from a stratum that recorded the demise of the pygmy mammoth 12,900 years ago. And not just conventional cubic nanodiamonds but also so-called lonsdaleite, an odd intermediate form made in the laboratory by shocking graphite into “hexagonal diamond.” A TEM analysis of the crystal structure certifies that lonsdaleite is there, the group writes, and lonsdaleite has only been reported to form under extreme conditions in the lab, in large impacts on Earth, or in outer space. Therefore, the group concludes, something struck graphite-rich rocks 12,900 years ago to create the lonsdaleite.

Many TEM experts disagree. They see shortcomings in the analysis by Kennett and colleagues: overexposed diffraction patterns, a lack of uncertainty estimates, a missing TEM diffraction ring, and a failure to use their analytical equipment in a mode that would have yielded definitive evidence, among others. The analysis “is just not quite good enough to unequivocally say yes or no,” says meteoriticist and microscopist Lawrence Garvie of Arizona State University, Tempe.

Cosmochemist Iain Gilmour of the Open University in Milton Keynes, U.K., writes in an email that while the analyses “look pretty reasonable” for at least some kind of nanodiamond being present, nanodiamonds have never been used as a diagnostic impact marker in far-flung debris unless accompanied by bona fide impact markers like enhanced iridium. Independent analyses have failed to find any iridium from 12,900 years ago. Next up in the mammoth wars may be not another paper but a frank exchange of views being arranged for December’s meeting of the American Geophysical Union in San Francisco, California.