Fossil-fueled argument

Paleontologists respond to molecular geneticists: Fossil record is good

A paleontological study in the Feb. 26 issue of Science shows that modern mammals probably originated no later than 65 to 75 million years ago, not 130 million years as indicated by molecular geneticists, and that the fossil record during this period is at least 10 times better than molecular geneticists have suggested.

PROVIDENCE, R.I. -- In an attempt to settle a controversy with molecular geneticists about the date of species origins, paleontologists have produced a new study showing that the fossil record can be tested to see how completely it preserves records of the species that were alive at the time.

The new study in Science combines data analysis of hundreds of early ancient mammal fossils with a mathematical model of species divergence to show that the fossil record is far better than some molecular geneticists have suggested and that it is unlikely modern orders of placental mammals arose much earlier than their oldest fossil records. Examples of modern placental mammals are humans, dogs, horses and mice.

In addition, the mathematical model could be used to determine the origin time of other major groups, such as flowering plants and birds, which molecular geneticists postulate may have originated tens of millions of years before their fossil records.

The first mammals are known to date from around 200 million years ago. Modern types of mammals did not appear until 65 million years ago, following demise of the dinosaurs.

A study published last year by molecular geneticists suggested that modern types of mammals first appeared and diverged into different lines of species around 130 million years ago. The scientists had sampled molecules of 20 species and had proposed a time of origin based on a molecular clock model of evolutionary change that not only doubled the evolutionary history of these mammals but suggested that the fossil record was unreliable in recording the origin of this major group.

Now a research team of paleontologists has responded to those assertions by studying the occurrence of fossils from at least 225 species of small ancient mammals. Using these data combined with a model of evolutionary branching patterns, the paleontologists mathematically modeled the probabilities of finding the supposed missing fossils of modern mammals. They demonstrated that modern mammals probably originated no later than 65 to 75 million years ago and that the fossil record during that time (the Cretaceous Period) is at least 10 times better than molecular geneticists suggested.

"The fossil preservation rate would have to be at least 10 times worse, and maybe up to 100 times worse, than molecular geneticists had claimed it was for the fossils of modern mammals not to have been found," said study author Christine Janis, professor of ecology and evolutionary biology at Brown University. Moreover, the probability that the fossils went entirely unpreserved is just .02 percent, she said.

"We can use real data and hard scientific calculations to estimate just how bad the fossil record would have to be for the fossils to be there and not be seen," Janis said. "The fossil record for that period is good enough for us to say that those species would most likely have been preserved if they had been there."

The study's lead author is Mike Foote, of the University of Chicago. The other authors are John Hunter, New York College of Osteopathic Medicine of the New York Institute of Technology; and J. John Sepkoski Jr., University of Chicago. Sepkoski and Janis, a husband and wife team, came up with the idea for the study. Hunter is a former Brown undergraduate who had conducted an honor's thesis in paleontology under Janis.

The measuring gap between molecular geneticists and paleontologists may be based on the geneticists' assumption that the rate of molecular evolution is constant, Janis said.

"The concept of a constant molecular clock has no empirical basis but somehow has become dogma," she said. "Maybe rates of molecular evolution can speed up at the start of a major evolutionary diversification, such as with the origin of the modern mammal orders. If you estimate divergence times based on constant rates of molecular change, then you may get estimations of origins that are way earlier than the first fossils."

Combining analysis of first-hand fossil records with modeling will now allow researchers to quantify "how bad" other portions of the fossil record supposedly are, Janis said. "Perhaps future dialogue between paleontologists and molecular geneticists will allow researchers to better determine what actually happened in evolutionary history," she said.