New fossils threaten an extinction theory.

Recent discoveries of long-gone marine invertebrates call into question the occurrence of a catastrophic global extinction hundreds of millions of years ago. The loss of diversity wasn't as widespread and didn't last as long as paleontologists had previously thought, several researchers now suggest.

The extinction during the Late Devonian period is widely considered one of the five massive extinctions in Earth's prehistory. The die-off climaxed at the end of the period's Frasnian stage, which lasted from around 385 million to 375 million years ago. The extinction began an extended interval of low biodiversity known as the Famennian stage. Coral reef ecosystems collapsed at the Frasnian-Famennian transition, and their gradual recovery took about 20 million years, until the end of the Devonian, many researchers say.

New data, however, suggest that at least some marine organisms rebounded quickly during the Famennian. Johnny A. Waters of the State University of West Georgia in Carrollton and his colleagues searched for fossils of echinoderms, a class of marine invertebrates, at Famennian sites in Australia, China, Morocco, and elsewhere. Their intent was to flesh out a fossil record previously known mostly from Appalachian and Swiss Alps deposits. Records at these sites suggested that the Late Devonian saw a prolonged drop in echinoderm diversity consistent with a mass extinction.

Waters and his colleagues identified many more echinoderm taxa in Famennian sediments worldwide than other paleontologists had presumed were there. "The Famennian was a time of major evolutionary innovation for echinoderms," says Waters. He suggests that reefs in many parts of the world rebounded quickly from a brief Frasnian extinction, but a bias toward research in North America and Europe obscured that pattern.

Previously studied reef devastation may have been limited to only one of Earth's hemispheres, Waters says. He and his coworkers reported their findings in Boston last week at the meeting of the Geological Society of America.

Some researchers remain skeptical. "Reefs really do disappear during the Late Devonian," says Paul Copper, a marine paleontologist at Laurentian University in Sudbury, Ontario.

As part of an international team compiling a database that describes biodiversity over Earth's past, Copper has determined that "the Frasnian-Famennian extinction was much worse than we thought." He suggests that echinoderms, which Waters investigated, were heartier than other reef-dwelling organisms and better able to survive the destruction of the coral.

Richard K. Bambach, a marine paleoecologist at Harvard University, comments that the spotty fossil record for echinoderms in the Famennian complicates use of that group as an index for general biodiversity. Nevertheless, he says, Waters' research helps take the edge off the Late Devonian extinction.

That the echinoderm class persisted in diversity well through the Famennian demonstrates that the loss of diversity in the Late Devonian "is not like end-Permian or end-Cretaceous," says Bambach, referring to subsequent mass extinctions. In those later episodes, as many as two-thirds of all living genera went extinct. Even with new types of organisms originating at slightly above-normal rates, net biodiversity plummeted during those periods, according to statistical analyses Bambach presented at the Boston meeting.

In the Late Devonian, by contrast, extinction rates were only moderately high, but very few new taxa arose to replace those lost, Bambach's research indicates. Rather than mass extinction, Bambach suggests, "mass depletion" might better describe the loss of diversity at that time.