Nov. 15, 2001
Vol. 21 No. 5

current issue
archive / search

    New analysis of fossil record answers scientists’ concerns over its quality

    By Steve Koppes
    News Office

    Paleontologists have long despaired of retrieving reliable ecological information from the fossil record. Remains can be destroyed, removed from their original habitats or mixed with later generations by natural processes. But a new analysis provides striking support for the fidelity of the marine fossil record and its ability to capture the nature of the living communities from which it comes.

    The new study, conducted by Susan Kidwell, Professor in Geophysical Sciences, and published in the Friday, Nov. 2 issue of the journal Science, is the most extensive and stringent analysis ever done to assess the quality of the fossil record. Its unexpectedly strong finding opens the door to studies in paleontology and conservation biology that scientists have been reluctant to undertake because of quality concerns, Kidwell said.

    Paleontologists and ecologists are now free to tackle some nagging questions that range from why the diversity of species is higher in the tropics than in temperate and polar latitudes, to whether or not abundant species really are more resistant to extinction than are rare species.

    This is a key finding because fossils are the only way scientists can reconstruct biological communities as they existed far back in time, before the advent of modern ecological or fishery studies, Kidwell said.

    “We have always used fossils to assess richness, that is, how many species existed at any one time and place. But we’ve never been sure how accurately those remains reflect the original relative abundance of species. Without that abundance data, we can’t say much about the nature of ecological communities.”

    But because of Kidwell’s analysis, this has changed. James Valentine, professor emeritus of integrative biology at the University of California, Berkeley, agreed that because of Kidwell’s study, scientists now might be willing to trust the relative abundance data of fossil populations.

    “Sue has shown that statistically, at least, we can well rely on them,” Valentine said. “A single locality may be weird, but if you’re looking at a lot of localities or studying a large fauna, your chances of recovering the original species abundances––relative species abundances, anyway––are really good.”

    Kidwell based her conclusion on a synthesis and statistical analysis of 19 detailed studies that compared the species abundance of live clams and snails to dead ones in 30 marine environments around the world. In these studies, marine biologists dredged bulk sediments from the top four to eight inches of the sea floor. Then they laboriously washed the sediments through a sieve, separated the dead clams and snails from the live ones, and compared how well the varieties and their numbers in the two groups agreed. The live organisms provide a way to check the accuracy of the burial record.

    Kidwell’s study showed a strong correlation between the numbers of dead and live specimens per species, ranging as high as 90 percent, depending on the methods used in field collection. It was the first time anyone had attempted to test the reliability of such data. This was partly because it was such a big job to pull the scattered sets of data together, but also because no one expected to get numbers this high, Kidwell said.

    “I’m delighted that the answer has come out so positively, but the important thing is to be able to put a hard number on how good it is, so we know what confidence we can have in the data,” she said. “This work shows that data from dead shells will be adequate for tackling a large number of ecologic and paleobiologic problems. It will be inadequate for others, but that’s really for the individual researcher to decide.”

    There are two possible reasons for the positive results, Kidwell said. First, the relative abundance of species may have remained largely unchanged over the period of time that these fossil graveyards had been accumulating, possibly for hundreds or thousands of years.

    Another explanation is that the information on species abundance in the fossil record actually reflects only the most recent period of input from the living community, Kidwell said. “So it doesn’t reflect time-averaged input over hundreds and thousands of years. Instead, it may only reflect the last few years or the last 10 years.”

    The latter explanation is the most likely to be true, Kidwell said. Radiocarbon dates show that although some shells in fossil graveyards are hundreds or thousands of years old, most are quite recent.

    “Most of those shells are lost. They’re recycled,” Kidwell said. “The shells are turned back into bits of calcium and carbon dioxide by boring organisms and by the leaching effect of acid waters. But what remains is apparently packed with information.”