A publication of the Archaeological Institute of America

Special Introductory Offer!


This 4.4-million-year-old skeleton found in Ethiopia belonged to a primate that may have walked upright and had teeth similar to humans. But is she our ancestor? (Photo © T. White, 2008)

Fifteen years ago, a group of fossil hunters, including a UC Berkeley graduate student named Yohannes Haile-Selassie, crawled on their hands and knees across a patch of 4.4-million-year-old sediment in northeastern Ethiopia's Afar region. Moving shoulder to shoulder with the rest of the group, Haile-Selassie spotted a small, white fossil on the reddish clay. It turned out to be a hand bone, the first piece of a partial skeleton of a female Ardipithecus ramidus (nicknamed Ardi). She would have mainly eaten plant foods, and probably spent much of her time in the trees that covered the area when she was alive. She stood about four feet tall, weighed approximately 110 pounds, and had a brain the size of a chimpanzee.

Over the next three years, an international team of paleoanthropologists found about 130 pieces of the skeleton. Since then, researchers have studied and reassembled the bones in the hope of creating the most complete picture of someone who is becoming one of the most controversial figures in evolution. Their findings have major implications for understanding how the human race came to be: Why did our species evolve the ability to walk on two legs? What was our last common ancestor with chimpanzees? And, most critically, was Ardipithecus a direct ancestor of modern humans?


(Courtesy G. Suwa et al., Science)

Ultimately, the team discovered pieces of 35 Ardipithecus individuals scattered across roughly four miles of ancient landscape near an Afar village named Aramis. The fossils include a nearly complete hand and arm, as well as a lower jaw. But the partially complete Ardi skeleton has generated the most discussion, especially over a bone from the base of her big toe called the medial cuneiform. It shows that her toe would have stuck out from her foot like a thumb and that she would have been able to use it for grasping. "It really doesn't differ from apes, and that's the surprising thing," says University of California, Berkeley, paleoanthropologist Tim White. "It is fully apelike." White had been expecting a foot that looks more like one belonging to Australopithecus afarensis, the species to which the famous Lucy skeleton belongs. Afarensis lived about 3.7 to 3.0 million years ago and had feet that were much like our own, with toes that point forward. Ardi's apelike feet raise the question: Was she evolving toward walking on two legs? The team's anatomist, Owen Lovejoy of Kent State University, believes she was. Lovejoy points to several bones in Ardi's foot that he believes show it was becoming more like a modern foot, able to better withstand the pressures of bipedal movement. The shape of her pelvis also reveals she was becoming more effective at walking on two feet.

Some paleoanthropologists who are not part of the research team are skeptical of Lovejoy's interpretation. "It is hard for me to actually find a lot of characteristics that are not chimpanzeelike," says David Begun, a paleoanthropologist at the University of Toronto and editor of the Journal of Human Evolution, who notes that while Ardi's teeth are different from chimpanzees, he doesn't see signs that she was adapted to walking on two feet. Begun describes Lovejoy as a brilliant anatomist, but reserves final judgment until he can examine the fossils more closely.


Ardipithecus (center) had small canine teeth that resemble a human's (below) more closely than a chimpanzee's (far right). (Courtesy G. Suwa et al., Science)

"I see nothing in the foot that suggests bipedality," says paleoanthropologist William Jungers of Stony Brook University. He calls the reconstruction and conservation job that Lovejoy and White's team has done on the fossils "heroic." However, he is not convinced by Lovejoy's interpretations of the fossils, least of all the pelvis, which was badly fragmented. "That's really kind of a 3-D Rorschach test if you ask me," Jungers says, "I'm still not convinced that it's necessarily completely accurate." Jungers believes that Ardipithecus was capable of walking on two legs, but was better adapted to living in trees. "He is living in the pre-Ardi world, and he is thinking in the pre-Ardi world," Lovejoy replies. "What Ardi tells us is that we never looked like chimpanzees, we never walked like chimpanzees."

Lovejoy's analysis also shows that the bones in Ardi's hands and wrists were not adapted to knuckle-walking, a style of movement common to gorillas and chimpanzees, and assumed to have been used by the species that was ancestral to apes and humans. Lovejoy believes that because Ardipithecus had not evolved the hands and wrists of a knuckle-walker, the last common ancestor of humans and chimpanzees had a more primitive way of moving--it would have placed its hands palm down on the ground when it needed to walk, the same way monkeys do. If Lovejoy is correct, no human ancestor ever walked on its knuckles, and chimpanzees and gorillas each evolved that trait after they had separated from the human lineage between 12 and seven million years ago.

If this interpretation holds up, it could have a major impact on the field of primatology because it would overturn long-held ideas about how living apes can be used as models of humanity's earliest ancestors.

One anatomical trait that bears a clear resemblance to later hominins is Ardipithecus's teeth. Most male apes have large canine teeth that are used to fight and intimidate sexual rivals. In apes, the canines don't come in until a male reaches maturity, signaling that he is fair game as either a rival or a partner. The teeth of a male Ardipithecus, on the other hand, were small and dull and would not have provided any social signals, like humans.

Why Ardipithecus lost its long, sharp canines is a matter of debate, but Lovejoy does not believe it has to do with a change in diet. "It is in every way, unquestionably, a behavioral adaptation. It represents a behavioral shift, there's no doubt about it."

The idea that hominins might have evolved smaller canine teeth at the same time they were developing the ability to walk on two legs has some important implications for Ardi's behavior, according to Lovejoy. If a male Ardipithecus lacked the teeth to fight for mating privileges, he must have had a different method of attracting mates.


Ardi's fragmented pelvis made reconstruction difficult. As shown here, the pelvis would have let her walk upright without wobbling like a chimp.

Lovejoy thinks Ardipithecus females may have exchanged sex for food, which is a trickier mating strategy than it might seem. The male is risking a lot of his time and energy by providing food to a prospective mate so he has to ensure that the female doesn't mate with anyone else. "He has got to pick a female that doesn't show external signs of ovulation," says Lovejoy. If it were obvious that the female was ovulating, there would be nothing to stop another male from getting her pregnant. One important signal of sexual receptiveness in female apes is their mammary glands. "He has got to pick the least attractive female in the group," says Lovejoy, "and that's the one with [permanently enlarged] mammary glands, and the reason? Big mammary glands means she is not ovulating, she is nursing." This change in mating strategy may have also led to upright walking, according to Lovejoy, which would have allowed a male to carry food to a prospective mate. Not everyone is convinced. "I don't know why bipedalism has to have a social origin," says Begun. "Why are humans so special that social forces would be driving these huge changes in the anatomy, where that doesn't seem to be necessary for other animals?"

In some ways, Ardipithecus is an awkward fit for a human ancestor living 4.4 million years ago. In general, the overall body size of hominins tends to increase through time, and the size difference between males and females, called sexual dimorphism, becomes less. Just 200,000 years after Ardi lived, Australopithecus anamensis arrives in the fossil record, looking more like later human ancestors. Ardipithecus seems to be larger and there is less difference in size between males and females than there is in Australopithecus, which in some ways makes it more like its ape ancestors than Ardipithecus. "By showing that at 4.4 million years there is a hominin ancestor that is less dimorphic than Australopithecus is, I think, very exciting because that is not what you would have predicted," says Zeresenay Alemseged, a paleoanthropologist at the California Academy of Sciences. "We do not necessarily become more apelike or more primitive as we go back in time." Others see these differences as less exciting and more of a reason to doubt that Ardipithecus is a human ancestor. "The jury is still out on this. There has to be a series of important reversals just going from Ardipithecus to Australopithecus," says Jungers. "If Australopithecus is sexually dimorphic and Ardipithecus isn't, that's kind of peculiar." New fossil discoveries will be needed to settle the question of ancestry.


Ardi's apelike big toe could be used to grasp branches, but her rigid foot bones may mean she also walked on two legs.

In the 15 years since finding the first piece of the Ardi skeleton, Haile-Selassie has risen to become the head of the physical anthropology department at the Cleveland Museum of Natural History. He was also involved in the analysis of Ardi's habitat and biology. "We have opened up a huge forum for discussion about human origins," he says, "at the same time, we are going to have to rewrite our textbooks because we are finding early ancestors that we never expected to see in the fossil record." Part of those future textbook revisions may come from a site called Woranso-Mille in the Afar region where Haile-Selassie has uncovered the remains of a 3.6- to 3.8-million-year-old hominin that may reveal more information about Australopithecus anamensis. In the meantime, no one seems willing to say that Ardipithecus was definitively a direct human ancestor.

"I'd say I'm 50/50," says Lovejoy, chuckling."There is really no reason to believe that there couldn't have been another [hominin] which was more advanced and was the real ancestor of anamensis." Begun echoes that idea. "I do think it probably is a hominin; in other words, it was more closely related to humans than chimpanzees, but I'm not convinced it was a biped. I'm not convinced it looked all that different from a chimpanzee, and I'm not convinced that it's directly related to Australopithecines as opposed to something that branched off very early on...and then went extinct." But the debates are just getting started. "Controversy is going to happen almost no matter what," says White, "believe me, we haven't even seen the beginning of the controversy yet."