"During the end-Permian extinction 95 percent of all species on Earth became extinct, compared to only 75 percent during the KT when the dinosaurs disappeared," says Dr. Lee R. Kump, professor of geosciences. "The end-Permian is puzzling. There is no convincing smoking gun, no compelling evidence of an asteroid impact."

Researchers have shown that the deep oceans were anoxic, lacking oxygen, in the late Permian and research shows that the continental shelf areas in the end-Permian were also anoxic. One explanation is that sea level rose so that the anoxic deep water was covering the shelf. Another possibility is that the surface ocean and deep ocean mixed, bringing anoxic waters to the surface.

Decomposition of organisms in the deep ocean could have caused an overabundance of carbon dioxide, which is lethal to many oceanic organisms and land-based animals.

"However, we find mass extinction on land to be an unlikely consequence of carbon dioxide levels of only seven times the preindustrial level," Kump told attendees today (Nov. 3) at the annual meeting of the Geological Society of America in Seattle. "Plants, in general, love carbon dioxide, so it is difficult to think of carbon dioxide as a good kill mechanism."

On the other hand, hydrogen sulfide gas, produced in the oceans through sulfate decomposition by sulfur bacteria, can easily kill both terrestrial and oceanic plants and animals.

Humans can smell hydrogen sulfide gas, the smell of rotten cabbage, in the parts per trillion range. In the deeps of the Black Sea today, hydrogen sulfide exists at about 34 part per million. This is a toxic brew in which any aerobic, oxygen-needing, organism would die. For the Black Sea, the hydrogen sulfide stays in the depths because our rich oxygen atmosphere mixes in the top layer of water and controls the diffusion of hydrogen sulfide upwards.

In the end-Permian, as the levels of atmospheric oxygen fell and the levels of hydrogen sulfide and carbon dioxide rose, the upper levels of the oceans could have become rich in hydrogen sulfide catastrophically. This would kill most of the oceanic plants and animals. The hydrogen sulfide dispersing in the atmosphere would kill most terrestrial life.

Kump and colleagues, Alexander Pavlov, University of Colorado; Michael Arthur, professor of geosciences, Penn State; Anthony Riccardi, graduate student, Penn State; and Yashuhiro Kato, University of Tokyo, are looking at sediments from the end-Permian found in Japan.

"We are looking for biomarkers, indications of photosynthetic sulfur bacteria," says Kump. "These photo autotrophic organisms live in places where there is no oxygen, but still some sunlight. They would have been in their hay day in the end-Permian." Finding biomarkers of green sulfur bacteria would provide evidence for hydrogen sulfide as the cause of the mass extinctions.

So, what of the 5 percent of the species on Earth that survived? Kump suggests that the mixing of the deep ocean layers and the upper layer was not uniform and that refugia, places where oxygen still existed, remained, both in the oceans and on land.