David Perlman, Chronicle Science Editor
Thursday, March 10, 2005
Cycles of Extinction and Biodiversity. Chronicle Graphic
With surprising and mysterious regularity, life on Earth has flourished and vanished in cycles of mass extinction every 62 million years, say two UC Berkeley scientists who discovered the pattern after a painstaking computer study of fossil records going back for more than 500 million years.
Their findings are certain to generate a renewed burst of speculation among scientists who study the history and evolution of life. Each period of abundant life and each mass extinction has itself covered at least a few million years -- and the trend of biodiversity has been rising steadily ever since the last mass extinction, when dinosaurs and millions of other life forms went extinct about 65 million years ago.
The Berkeley researchers are physicists, not biologists or geologists or paleontologists, but they have analyzed the most exhaustive compendium of fossil records that exists -- data that cover the first and last known appearances of no fewer than 36,380 separate marine genera, including millions of species that once thrived in the world's seas, later virtually disappeared, and in many cases returned.
Richard Muller and his graduate student, Robert Rohde, are publishing a report on their exhaustive study in the journal Nature today, and in interviews this week, the two men said they have been working on the surprising evidence for about four years.
"We've tried everything we can think of to find an explanation for these weird cycles of biodiversity and extinction," Muller said, "and so far, we've failed."
But the cycles are so clear that the evidence "simply jumps out of the data," said James Kirchner, a professor of earth and planetary sciences on the Berkeley campus who was not involved in the research but who has written a commentary on the report that is also appearing in Nature today.
"Their discovery is exciting, it's unexpected and it's unexplained," Kirchner said. And it is certain, he added, to send other scientists in many disciplines seeking explanations for the strange cycles. "Everyone and his brother will be proposing an explanation -- and eventually, at least one or two will turn out to be right while all the others will be wrong."
Muller and Rohde conceded that they have puzzled through every conceivable phenomenon in nature in search of an explanation: "We've had to think about solar system dynamics, about the causes of comet showers, about how the galaxy works, and how volcanoes work, but nothing explains what we've discovered," Muller said.
The evidence of strange extinction cycles that first drew Rohde's attention emerged from an elaborate computer database he developed from the largest compendium of fossil data ever created. It was a 560-page list of marine organisms developed 14 years ago by the late J. John Sepkoski Jr., a famed paleobiologist at the University of Chicago who died at the age of 50 nearly five years ago.
Sepkoski himself had suggested that marine life appeared to have its ups and downs in cycles every 26 million years, but to Rohde and Muller, the longer cycle is strikingly more evident, although they have also seen the suggestion of even longer cycles that seem to recur every 140 million years.
Sepkoski's fossil record of marine life extends back for 540 million years to the time of the great "Cambrian Explosion," when almost all the ancestral forms of multicellular life emerged, and Muller and Rohde built on it for their computer version.
Muller has long been known as an unconventional and imaginative physicist on the Berkeley campus and at the Lawrence Berkeley Laboratory. It was he, for example, who suggested more than 20 years ago that an undiscovered faraway dwarf star -- which he named "Nemesis" -- was orbiting the sun and might have steered a huge asteroid into the collision with Earth that drove the dinosaurs to extinction.
"I've given up on Nemesis," Muller said this week, "but then I thought there might be two stars somewhere out there, but I've given them both up now."
He and Rohde have considered many other possible causes for the 62- million-year cycles, they said.
Perhaps, they suggested, there's an unknown "Planet X" somewhere far out beyond the solar system that's disturbing the comets in the distant region called the Oort Cloud -- where they exist by the millions -- to the point that they shower the Earth and cause extinctions in regular cycles. Daniel Whitmire and John Matese of the University of Louisiana at Lafayette proposed that idea as a cause of major comet showers in 1985, but no one except UFO believers has ever discovered a sign of it.
Or perhaps there's some kind of "natural timetable" deep inside the Earth that triggers cycles of massive volcanism, Rohde has thought. There's even a bit of evidence: A huge slab of volcanic basalt known as the Deccan Traps in India has been dated to 65 million years ago -- just when the dinosaurs died, he noted. And the similar basaltic Siberian Traps were formed by volcanism about 250 million years ago, at the end of the Permian period, when the greatest of all mass extinctions drove more than 70 percent of all the world's marine life to death, Rohde said.
The two scientists proposed more far-out ideas in their report in Nature, but only to indicate the possibilities they considered.
Muller's favorite explanation, he said informally, is that the solar system passes through an exceptionally massive arm of our own spiral Milky Way galaxy every 62 million years, and that that increase in galactic gravity might set off a hugely destructive comet shower that would drive cycles of mass extinction on Earth.
Rohde, however, prefers periodic surges of volcanism on Earth as the least implausible explanation for the cycles, he said -- although it's only a tentative one, he conceded.
Said Muller: "We're getting frustrated and we need help. All I can say is that we're confident the cycles exist, and I cannot come up with any possible explanation that won't turn out to be fascinating. There's something going on in the fossil record, and we just don't know what it is."