When the 4.4-million-year-old skeleton of a potential human ancestor was unveiled last year, the scientists who studied “Ardi” introduced her confidently as a denizen of the woods. Ardipithecus ramidus, they pointed out, had an opposable big toe and flexible hands to grasp branches of trees. What's more, the fossilized bones of at least 36 members of her species had been found with fossilized wood and seeds, as well as fossils of woodland creatures such as monkeys, parrots, and snails, suggesting that she lived in grassy woodland with small patches of forest.
But a group of eight geologists and anthropologists disagree with that conclusion. In a Technical Comment published in tomorrow's issue of Science, they argue that Ardi's habitat in Aramis, Ethiopia, was not grassy woodlands, but wooded or bushy grasslands. That may sound like a minor distinction, but it's critical to determining whether the earliest hominins began walking upright because environmental change forced them out of the woods. Plus, Ardi’s sylvan habitat has been used to falsify a long-standing hypothesis that suggested that upright walking evolved in savanna grasslands.
The dissenting researchers, who come from seven universities, propose that most of the terrain at Aramis was hot, dry savanna with scrubby bush and trees, and that the trees’ canopy covered no more than one-quarter of the ancient landscape. They base their conclusion on data from the ancient soils at Aramis published in October by Ardi's discovers, an international team known as the Middle Awash research group. Specifically, soil samples from nine fossil localities at Aramis showed that between 40% and 60% of the plants at the site were so-called carbon-four plants, such as grasses, suggesting that Aramis was too grassy to be called a woodland 4.4 million years ago. Instead, it was a “wooded grassland” or “grassland,” as defined by the United Nations Education, Scientific and Cultural Organization's definition for classification of African vegetation. A separate measure of aridity also suggests to the researchers that Aramis was so hot and dry that it could have supported at most a thin forest along a river or small woodland in an otherwise open bushland, says geochemist Thure Cerling of the University of Utah in Salt Lake City. Overall, the findings suggest that Aramis was undergoing a major ecological transition from woodlands to grasslands, and Aramis cannot be used to falsify the hypothesis that our ancestors began walking upright in grasslands rather than in the woods.
But the Middle Awash team says that the dissenters are being “extremely selective” about the evidence they cite. There was indeed a mosaic of open and woody grasslands along with more closed woodlands at Aramis, the Middle Awash researchers respond in Science—and they said so in their original papers. But within that larger landscape, Ardi’s “preferred habitat” was in the woodlands. The evidence for a wooded habitat that Ardi preferred within a larger landscape comes from “many, many different lines of evidence, just like the evidence at a crime scene comes from many different lines,” says paleoanthropologist Tim White of the University of California, Berkeley, co-leader of the Middle Awash research project.
For example, isotopes in the teeth of five members of Ar. ramidus and 172 teeth of two-dozen mammal species found in the same layer of ancient soil show that Ardi's diet was more like that of animals that browsed in woodlands than that of grasslands grazers. Furthermore, the shape of Ardi's teeth and jaw suggest that she was adapted to eat woodland fruit and plants, as a small species of arboreal baboon does today. The abundant signs of woodland animals, birds, plants, gastropods, and pollen found in Ardi's neighborhood, White says, clinch the case.
“We’re not focusing on the places Ardi would have seen during her life," White says. "Rather, what we’re trying to understand is which part of that environment this big primate preferred as its habitat—its address, where it lived.” And for the Middle Awash team, that address was in the grassy woodlands, not the wooded grasslands.