Tiny tooth upends Australian history.

Australian paleontologists have discovered a fossilized tooth -- about as large as the capital letter starting this sentence -- that promises to overturn long-standing beliefs about the origins of kangaroos, koalas and the other strange marsupials found in the land down under.

Known for carrying their newborns in external pouches, marsupials have dominated Australian ecology since the fall of the dinosours. On other continents, though, marsupials were largely wiped out by placental mammals. Judging from that record, paleontologists have presumed marsupials to be an inferior group that succeeded in Australia solely because they were the only mammals present on the continent when it split off from Antartica and became an island. By that theory, the marsupials won Australia by default.

But the discovery of a 55-million-year-old s lar from a placental mammal shows that these animals did indeed reach Australia early on and competed with marsupials before the continent went its separate way, a group of paleontologists asserts in the April 9 NATURE.

"The age-old notion that marsupials are competitively inferior to placentals does not hold water. It looks like just the opposite -- that marsupials and placentals were together in Australia but that marsupials dominated and placentals went extinct," says Michael Archer of the University of New South Wales, one of the fossil's discoverers.

The molar bears a number of characteristics that led Archer and his colleagues to identify it as belonging to a placental mammal. A particular cusp, called the hypoconulid, sits in a back central part of the tooth. In marsupials, the cusp is displaced toward the tongue side. The fossil tooth also lacks another defining marsupials trait: a little shelf around the base of the tooth. These features were so striking that Archer's group identified the tooth as placental within a few seconds of finding it.

Studies with a scanning electron microscope and a laser scanning microscope revealed that the microstructure of the tooth's enamel also resembles that of a placental mammal.

The tooth belonged to an animal the size of a large rat, says Archer. It may fit in an order of mammals called the Condylarthra, which enjoyed great success in other parts of the world, ultimately giving rise to many different evolutionary lines. The researchers have named the new animal Tingamarra porterorum.

Prior to discovery of the tooth, the oldest Australian placentals (other than bats) dated to 5 million years ago -- a time when the northward-moving continent had drifted close enough to allow the passage of Asian rodents across the Malay Archipelago. Bats appeared far earlier in the Australian fossil record, and scientists have long presumed that they flew to Australia after it had separated from Antarctica. But Archer and his coworkers have found bat remains in the same deposit as the T. porterorum tooth.

The scientists say they cannot explain why marsupils dominated in Australia while suffering defeat on all other continents. Archer speculates that marsupials -- which give birth very soon after they conceive -- may hold a distinct advantage in an unpredictable climate. If conditions grew particularly bad, a marsupial mother could survive by abandoning her newborns, whereas a placental mother facing the same conditions might die because she carries her young internally for a much longer time.

Alternatively, chance may have decided the marsupials' fate, both in Australia and elsewhere. "Maybe it's just a lottery," Archer says.

Michael O. Woodburne, a paleontologist at the University of California, Riverside, comments that while the new-found tooth resembles those from placental mammals, "I would not say that a placental affinity has been proven." He cautions that researchers know very little about the marsupials from this time period, leaving open the possibility that the tooth belonged to an unusual maruspial.