Better for it. Hot acid has removed degraded parts of this zircon that would have skewed its apparent age.

Timing Is All in Mass Extinction

For geochronologists trying to place the blame for the world's all-time greatest mass extinction, the key is timing. Standard dating methods, however, have produced conflicting and often confusing results. Now, a team of researchers has used a new technique to bolster the argument that a volcanic catastrophe 250 million years ago may have wiped out 95% of the animal species on Earth.

Geochronologists date the Permian-Triassic (P-T) extinction by studying the vanishingly small isotopic remains of radioactive decay. The analysis involves measuring the isotopic ratio of uranium to lead in the mineral zircon, obtained from crushed rock samples. But not all ratios indicate the same geological age, and researchers must be somewhat subjective when determining which data to discard.

Roland Mundil of the Berkeley Geochronology Center in California set out to eliminate the "picking and choosing" from uranium-lead dating. Mundil and colleagues baked their samples at 850°C for 36 hours and then leached them with hydrofluoric acid under pressure at 220°C for 16 hours, with the intention of removing the parts that had lost lead over geologic time. This rugged pretreatment narrows the range of zircon ages from a single volcanic ash bed from about 20 million years to a few million years, with no picking and choosing. As reported in the 17 September issue of Science, the zircon data suggest that the P-T extinction happened 252.6 ± 0.2 million years ago--a time of massive volcanic eruptions in Siberia.

"It's an impressive piece of work," says geochronologist Michael Villeneuve of the Geological Survey of Canada in Ottawa. Mundil's technique "is a step in the right direction," adds geochronologist Samuel Bowring of the Massachusetts Institute of Technology. Even so, Villeneuve cautions that although the new treatment seems to remove much of the subjectivity of traditional approaches, it still leaves room for interpretation. "It needs a bit more proving out," he says.

Related Sites
Roland Mundil's site
The EarthTime geochronology project