Welcome to the Family, Missing Link

Discovery of the fossil of an unknown human ancestor shakes up ideas of human evolution.

Let there be no more grousing about Take Your Kid to Work Day. When paleoanthropologist Lee Berger of the University of the Witwatersrand in South Africa took his son, Matthew, then 9, to the site near the Malapa caves north of Johannesburg where he and his team were hunting fossils in 2008, the boy wandered about 45 feet away from everyone else. After about a minute and a half, Matthew picked up a rock and called out, "Dad, I found a fossil!"

Berger assumed it was one of the common antelope fossils that litter the site, but when he strolled over to look at the rock Matthew was holding, he got the shock of his life. A clavicle—collarbone—was sticking out of the rock. Turning the rock over, Berger saw a jaw. Both bones clearly (to his trained eye, anyway) belonged to a hominid, or human ancestor. And as Berger and his team announced in a teleconference yesterday and in tomorrow's issue of the journal Science, it is no ordinary ancestor. The bones, and others that Berger and his team excavated after Matthew's serendipitous find, belong to two extremely well-preserved individuals, a woman in her late 20s and a 10- to 13-year-old boy. They represent a previously unknown human ancestor who lived between 1.95 million and 1.78 million years ago and have such an unexpected appearance that they will redraw the picture of human evolution like nothing since the 3-million-year-old Lucy.

Named Australopithecus sediba, the new species has a bewildering mix of features that show, as I wrote in our 2007 cover story on human evolution, that "evolution played Mr. Potato Head, putting different combinations of features on ancient hominids and then letting them vanish until a later species evolved them." In the case of sediba (the scientists have invited the children of South Africa to come up with a friendlier, informal, Lucylike name), some of the features seem to be too modern while others look almost too primitive for their age, with features that a simple, linear—and wrong—model of human evolution says should have vanished eons earlier.

For instance, sediba (the name means fountain or wellspring in the Sesotho language) had an extremely small head and brain—Berger called the poor things "pinheads." At 420 cubic centimeters in volume (compared to 510cc for the smallest known Homo brain), the head was smaller than Lucy's, who lived 1 million years earlier. Clearly, the path of human evolution did not march forward uninterruptedly from small-brained ancestors to large-brained ones. The pelvis and hip look more modern, indicating that sediba was an expert upright walker—bipedalism is a hallmark of being human—like early Homo species. But she also had arms so long they practically dragged along the ground: Berger likened the arms to those of a Miocene ape. Compared to Lucy, the face of the sediba was much more modern and less apelike (little of that jaw-jutting, overhanging-brow look, and with smaller teeth and a protruding nose rather than a flat, apelike one), as one would expect after a million extra years of evolution.

Sediba, said Berger, "is a highly transitional species" (he recoiled at the dated term "missing link," but to laypeople that's what this is) "with a mosaic of features shared by later and earlier hominids." The species had "made the committed change" to life on the ground, he said, "while maintaining a sort of reserve parachute by keeping these very long arms and legs for climbing trees." They were comfortable in both worlds—in the trees where Australopithecus evolved, and the savannah where Homo triumphed.

The mosaic features make placing sediba on the family tree a bit of a challenge. But the best guess is that the species is a direct descendant of Australopithecus africanusand an ancestor of Homo. But that's where things get messy. Several species of Homo roamed east and south Africa 2 million years ago; indeed, the modern era (since Neanderthals vanished 30,000 years ago) is an anomaly, points out anthropologist Ian Tattersall of the American Museum of Natural History, in having just a single species of Homo on Earth. Although it isn't possible to identify which species of Homoergaster, habilis, or rudolphensissediba is directly ancestral to, Berger says, it seems safe to say that Homo features emerged in fits and starts, two steps forward and one step back, rather than as a steady unbroken march to modernity.

The family tree is confused enough, Berger and his colleagues write, that sediba could equally well be an "ancestor for [Homo], or a sister group to a close ancestor that persisted for some time after the first appearance of Homo." (One example of genealogical uncertainty: some paleoanthropologists believe that Homo habilis and Homo ergaster are actually Australopithecus species.) Indeed, other experts don't think Berger got it right. Donald Johanson, who discovered Lucy in 1974, told me by e-mail from Africa that the fossils "are most interesting, but I find it curious that the authors point to so many anatomical features that indicate that the finds belong to our genus, Homo, yet they place it in Australopithecus, so I think they missed the boat here. The fossils are more likely to be some new species of Homo," in which case they join an already large club rather than standing at a unique junction between Australopithecus and Homo.

In addition, argues Johanson, who examined the fossils soon after Berger discovered them, "I do not see these fossils as evolving from Australopithecusafricanus, which I believe gave rise to A. robustus in south Africa. [Nor are the new fossils] the ancestor to all later Homo, as the authors believe, since we have evidence of Homo in eastern Africa at 2.33 million years." Exhibit A of that evidence for Homo beingolder than the new guys: an upper jaw 2.3 million years old that Johanson and his team recently unearthed from the same area where Lucy was found, and which "represents the oldest anatomical evidence, thus far, for Homo." (It is probably Homo habilis.) Berger's team is now excavating at least two more sediba specimens—an infant and another woman—who might reveal more about our family tree, including where the new guys belong.

The woman and boy found so far lay close together in what had been a cave. In a study describing the local landscape at the time sediba lived, geologist Paul Dirks, of James Cook University, and colleagues conclude that it was mostly grassy plains, cut by wooded valleys. Near the two fossils was a pair of "death chutes" leading from the ground into a deep underground cave. Dirks suspects that the woman and boy—perhaps mother and son—ventured down the chute to find water, perhaps during a drought, but were stuck when "some sort of calamity," as he put it—perhaps a sudden storm that triggered a mudflow—trapped them, and they died. The two bodies were then "washed into an underground lake or pool," said Berger. "They did not travel far, maybe a few meters, where they were solidified, as if thrown into quick-setting concrete." Since the skeletons were partly intact, they were probably only partially decomposed when they landed in the lower chamber, and probably died at about the same time—there to lie for almost 2 million years until a child happened along to find the newest member of the human family.