In 1996, paleontologists made a startling discovery in northwestern Madagascar. Among dinosaur bones and sandy sediment there emerged a 167-million-year-old tiny jaw fragment with three teeth. It belonged to Ambondro mahabo, a species that was 25 million years older than any mammal of its kind ever found.
And it wasn’t supposed to be there. At the time, what was known of the fossil record pointed overwhelmingly to the conclusion that modern mammals’ forerunners arose in the Northern Hemisphere.
“The prevailing wisdom suggested that we shouldn’t find something like that from the time interval we were sampling, nor from the Southern Hemisphere,” said John Flynn, the paleontologist who led that dig and is now the Frick curator of fossil mammals at the American Museum of Natural History in New York.
It takes more than a single fossil to overturn an entire theory of evolution. But a review of existing fossil holdings published last year in the journal Alcheringa sought to turn decades of paleontological wisdom on its head. After an exhaustive study of skulls, jaws and teeth, a team of Australian paleontologists presented their conclusion that modern mammals originated in the Southern Hemisphere.
Their findings have set off an impassioned debate, revealing a North-South divide. Defenders of the Northern Hemisphere hypothesis highlight weaknesses they see in the latest findings. In response, supporters of the Southern Hemisphere origin, like Dr. Flynn, say it is time for paleontologists to grapple with the argument that their field’s understanding of natural history may be slanted toward the half of the world where scientists have carried out the most digs.
“In the Southern Hemisphere, these are just places that haven’t been explored by paleontologists,” Dr. Flynn said. “There has been a long-term, overall bias in the system toward a Northern Hemisphere perspective, partly because that’s where the scientists came from. And it leads you to interpret a lot of things in the light of that bias.”
At the heart of the dispute are the primitive early forerunners to modern placental and marsupial mammals. Known as tribosphenic mammals, they were “little shrew-like creatures that would have weighed about as much as a mouse,” said Tim Flannery, an independent Australian paleontologist and one of the authors of the recent review paper.
Although sophisticated for their time, they were a very basic version of mammals as we know them today. Dr. Flannery compared them to the Ford Model T “of modern or placental mammals.”
Dr. Flannery and company point to geographic arguments in favor of the idea that early mammals could have arisen in the Southern Hemisphere. The larger the land mass, the greater the likelihood of major evolutionary activity occurring. When mammals were emerging, Gondwana encompassed Africa, India, Australia and South America and was far larger than Laurasia in the Northern Hemisphere.
“A lot was happening there,” Dr. Flannery said, noting the emergence of songbirds and raptors on Gondwana during the age of dinosaurs. “We’ve just added this extra twist that we think the mammals were also evolving here.”
The early Southern Hemisphere mammals were unlike anything our planet had seen before.
“They had uniquely complex teeth that allowed the animal to puncture its food, crush its food, slice its food, all with the same tooth with different facets to it,” Dr. Flannery said. That gave them a great advantage over other organisms, he said. “When they got into the Northern Hemisphere,” he added, “they took off and became very diverse very quickly.”
The oldest tribosphenic fossil, from South America, dates back 180 million years, with a clear line of further tribosphenic fossils found in the Southern Hemisphere, including Ambondro mahabo, right through until 100 million years ago. “By this point, the teeth had become the kind of Swiss Army knife, all-functioning tool kit that mammal teeth became,” said Kris Helgen, chief scientist at the Australian Museum in Sydney and another author of the recent review paper.
It was at this time, too — between 100 million and 125 million years ago — that the first tribosphenic mammals appeared in the Northern Hemisphere.
Dr. Flannery and his co-authors argue that, having evolved in the south, tribosphenic mammals migrated into the north, island hopping between the two supercontinents.
According to Dr. Flannery, such an explanation fits with the theory that a new kind of mammal had been evolving in the Southern Hemisphere for millions of years before they suddenly appeared in the Northern Hemisphere.
“There’s nothing that’s clearly ancestral to these animals in the Northern Hemisphere, but in the Southern Hemisphere there are many,” he said.
Not everyone agrees. Zhe-Xi Luo of the University of Chicago is among the defenders of the existing hypothesis that tribosphenic mammals arose in the Northern Hemisphere. He said that the Southern Hemisphere origins hypothesis was “disadvantaged by missing out on a huge amount of data.”
Dr. Flannery and his co-authors, he argues, focus too closely on molar or tooth fossils at the expense of other parts of the mammalian anatomy. They also failed to consider fossils from all branches of the mammalian evolutionary tree. Further, Dr. Luo says, Dr. Flannery and his co-authors neglected to carry out a computational analysis of existing data. Such a statistical study requires the construction of a vast database of known fossils and the use of algorithms to compare anatomical traits. It can also enable paleontologists to reconstruct patterns of ancestry and, in turn, evolution.
Dr. Flannery, who has questioned the reliability of such databases, said the decision not to carry out such an analysis was deliberate and transparent. Such analyses result in double-counting of some elements, he said, and the database itself might be unreliable.
In Dr. Luo’s own work, he suggests tribosphenic mammals most likely emerged in China, independently of anything that was happening in the south. The southern tribosphenic mammals, he says, either died out or became monotremes, a family of mammals that includes the platypus and echidna.
Dr. Flannery and his co-authors also addressed the links between monotremes and tribosphenic mammals in a different paper last year. In that paper, they argued that monotremes belong to a separate branch of the mammalian evolutionary tree. “The monotremes have nothing to do with other modern mammals at all,” he said. “They’re an even more ancient lineage” — a conclusion that Dr. Luo strongly disputes.
Guillermo Rougier, a paleontologist at the University of Louisville and a peer reviewer of the paper by Dr. Flannery and his colleagues, offered a cautious endorsement of the Southern Hemisphere origin argument.
“It’s like a seesaw with a one-ton stone at each end, and then you put two grains of rice on one side,” he said. “You end up with a conclusion which is supported by one ton of evidence plus two grains of rice, but at the other end you have another conclusion which is supported by one ton of evidence.”
Neither side expects this paper to be the final word in the process of trying to reconstruct the mammalian past.
“Right now, it’s like finding a fossil with a long neck and making inferences that confuse a giraffe with the Loch Ness monster, because we don’t have enough information,” Dr. Rougier said.
Dr. Flynn said: “People think that in paleontology everything has been discovered. Nothing could be further from the truth.”