Read literally, the history of life derived from the past half-billion years of marine fossils says it takes 5 million to 10 million years for new species to begin replacing those lost during extinctions. That's bad news for a modern biosphere battered by a human-induced mass extinction. But now researchers have taken a second look at the fossil record after trying to remove some of its imperfections and have concluded that there's no Darwinian traffic cop holding life back.
The new reanalysis was a serendipitous affair. Harvard physics graduate student Peter Lu learned about the well-known marine fossil record compiled by the late paleontologist Jack Sepkoski while taking courses from evolutionary biologist Charles Marshall, also at Harvard. Sepkoski's raw data had already been analyzed by geoscientist James Kirchner of the University of California, Berkeley, and paleontologist Anne Weil of Duke University in Durham, North Carolina. Their results suggested that a postextinction world is environmentally inimical to life for millions of years.
Lu's former Harvard roommate Motohiro Yogo, now an economist at The Wharton School of the University of Pennsylvania, had another idea: Why not analyze the record of life using vector autoregression? That's a technique commonly used to forecast the performance of the stock market or the economy from past behavior. When applied to Sepkoski's raw record of when genera first appeared and last appeared in the record, it suggested the same evolutionary inertia Kirchner and Weil found.
The Harvard group then analyzed a modified version of the record. Paleontologist Michael Foote of the University of Chicago in Illinois had attempted to take account of known biases in the fossil record, such as the varying amount of exposed fossil-bearing rock found in different geologic time intervals. With such revisions, "the speed limit disappears," says Lu. In general, there's no delay between extinction and recovery, although there may be exceptions, such as after the great Permian-Triassic mass extinction. The findings appear in the 21 February issue of Proceedings of the National Academy of Sciences.
The new result "makes a great deal more biological sense than the prolonged delay" of recoveries, says paleontologist Douglas Erwin of the National Museum of Natural History in Washington, D.C. But Kirchner cautions that it's hard to sort out whether the finding is real or artificial. "The error bars can be large," he says. However it plays out, "this is the battle line for the next decade in paleontology," says paleontologist Steven Holland of the University of Georgia, Athens. "We're going to see a new wave of analyses that take incompleteness [of the fossil record] into account. Our view of evolutionary patterns is going to change."