Paleontologist Malte Ebach and botanist Christopher Humphries call their technique "area cladistics." It uses an algorithm that, by combining geographical and morphological information, suggests the ancient locations of continents on which the fossils were found.

Ebach--who was invited to present the theory at a meeting on "Fresh Science" in Melbourne, Australia, last month--points to his study of trilobites, small marine invertebrates that lived about 590 million to 250 million years ago. He created a cladogram--a kind of family tree--after measuring features on 100 individuals from around the world. Then, assuming that similar animals lived closer together, he superimposed data on where they were found to estimate the position of the continents some 400 million years ago.

Ebach and Humphries, whose theory was published in the April Journal of Biogeography, say that trilobites living in what is now the southern and central United States were more like ones from eastern Australia than those in Africa or South America. This suggests that accepted maps of the ancient landmasses of Gondwana (covering South America, Africa, and Australia) and Laurentia (which included North America) are way off. "They moved right around the world 8000 kilometers west of where scientists originally placed them," Ebach says.

Paleontologist Bruce Lieberman of the University of Kansas, Lawrence, says the work reflects the increasingly quantitative direction of biogeography, which studies how evolution is related to changes in the face of Earth. But he's not planning to redraw any maps, saying there's "very good evidence"--both from paleomagnetic and fossil sources--that Australia was nowhere near North America at the time of trilobites.