Opinion, arguments & analyses from the editors of Scientific American

Kenyan Fossils Rekindle Debate over Early Human Diversity


Koobi Fora fossils

The KNM-ER 1470 cranium, discovered in 1972, combined with the new lower jaw from Koobi Fora. The specimens are thought to belong to the same species. The lower jaw is shown as a photographic reconstruction, and the cranium is based on a computed tomography scan. © Photo by Fred Spoor

If I had to pick the hottest topic in paleoanthropology right now, I’d say it’s the origin and early evolution of our genus, Homo. Researchers know quite a bit about our australopithecine predecessors (Lucy and her ilk) and about later phases of Homo’s evolution. But the dawn of our lineage is cloaked in mystery. One question experts have long puzzled over is whether Homo split into multiple lineages early on, or whether the known early Homo fossils all belong to a single lineage. To that end, new discoveries made at the site of Koobi Fora in northern Kenya—one of the Leakey’s longtime fossil hunting grounds—are said to settle that matter in favor of multiple lineages. But some critics disagree.

The new finds—a partial face including almost all of the molars in the upper jaw, a nearly complete lower jaw and a partial lower jaw that date to between 1.78 million and 1.95 million years ago—bear on the identity of a famously enigmatic skull from Koobi Fora known as KNM-ER 1470. Ever since the discovery of the 1470 skull in 1972, researchers have struggled to place it in the human family tree. On one hand, at nearly two million years old it is the same age as H. habilis fossils from Koobi Fora and other locales in East Africa. The skull also shares some features in common with that species, which most researchers consider to be the founding member of Homo. On the other hand, 1470 is much larger than established H. habilis fossils, and differs from them in having a flat, long face, among other distinctive traits. Some experts thus assigned 1470 and some other fossils from Koobi Fora to a separate species, H. rudolfensis.

Meave Leakey and Fred Spoor

Paleontologists Meave Leakey and Fred Spoor collect fossils close to the site where the new face KNM-ER 62000 was found. © Photo by Mike Hettwer,, courtesy of National Geographic

But nailing down whether 1470 is a rogue H. habilis or a separate species has been tricky because no other skull shared that long, flat face and the specimen lacks teeth and a lower jaw to compare with other fossils. This is where the new fossils from Koobi Fora come in. In a paper published in the August 9 Nature, Meave Leakey of the Turkana Basin Institute in Nairobi, Fred Spoor of the Max Planck Institute for Evolutionary Anthropology in Leipzig and their colleagues report that the new face mirrors 1470’s face shape, although it is smaller overall. Inferring 1470’s upper jaw anatomy from the new face, the authors say the lower jaw fossils they found are good matches for the upper jaws of 1470 and the new face. (Scientific American is part of Nature Publishing Group.)

New mandible from Koobi Fora

The lower jaw KNM-ER 60000 after initial restoration but before the adhering matrix was carefully removed. © Photo by Mike Hettwer,, courtesy of National Geographic

“For the past 40 years we have looked long and hard in the vast expanse of sediments around Lake Turkana for fossils that confirm the unique features of 1470’s face and show us what its teeth and lower jaw would have looked like,” Leakey remarked in a prepared statement. “At last we have some answers.” The answers, in their view, indicate that 1470 and the new fossils represent a distinct human lineage from other early Homo fossils. This would mean that two Homo lineages lived alongside our ancestor H. erectus. H. erectus itself may have evolved from one of these two groups or another, as-yet-unknown group. The researchers did not formally name the new fossils from Koobi Fora, because of confusion surrounding the fossil that defines H. habilis, but they suggest that it may be appropriate to assign them to H. rudolfensis. Bottom line, they’re saying the fossils confirm that the non-erectus early Homo fossils in East Africa constitute two lineages, not one.

Although it may be hard to imagine sharing turf with another human species today, members of these ancient contemporaneous lineages need not have stepped on each other’s toes. In background materials distributed to the press the discovery team noted that chimpanzees and gorillas live in some of the same habitats. Both eat ripe fruit, but gorillas focus more heavily on tough vegetation than chimps do. “The early hominins [members of the human branch of the family tree] could have separated their neighborhoods in the same way,” the researchers explain. “They may simply have focused on different primary food items.” Exactly what these hominins were eating is uncertain, “but there are clues from the arrangement of the face and jaws that the newly described fossils, and the previously known [1470 skull], with their tall faces but shortened front tooth row, may have been focusing on foods that required chewing on the back teeth.” Analyses of the chemical composition of the teeth, as well as their wear marks, may yield further insights into what these hominins ate.

In an accompanying commentary Bernard Wood of George Washington University calls the new evidence for at least two parallel lineages in the early evolution of Homo “compelling.” Indeed he suggests that this chapter of our evolutionary history was even more complex than that. “My prediction is that by 2064, 100 years after [Louis] Leakey and colleagues’ description of H. habilis, researchers will view our current hypotheses about this phase of human evolution as remarkably simplistic,” he writes.

Other researchers are not convinced that the new Koobi Fora fossils show multiple lineages of early Homo co-existed. Adam Van Arsdale of Wellesly College, who has studied the 1.76 million-year-old H. erectus fossils from the site of Dmanisi in the Republic of Georgia, notes that in light of the considerable variation evident in the well-dated Dmanisi sample, the variation in the early Homo fossils from Africa can be accommodated by one species. In fact, the new Kenyan fossils show features in common with the Dmanisi ones, and thus help to link early Homo in Africa to H. erectus in Georgia, he says. In his view, all of these fossils—the habilis/rufolfensis ones and H. erectus--belong to one lineage.

“What the African assemblage lacks is a good sample from a single locality that shows variation. Instead you have lots of fragmentary, isolated specimens, all with temporal uncertainty, that show a huge amount of variation,” Van Arsdale explains. Whereas Leakey and Spoor see this variation as evidence for multiple concurrent lineages, “I tend to see this new evidence as making it harder to reject the idea of a single evolving lineage,” he says.

A more pointed criticism of the new study comes from Lee Berger of the University of the Witwatersrand in Johannesburg. Berger notes that in their paper Leakey, Spoor and their colleagues neglected to compare the new Koobi Fora fossils to Australopithecus africanus and A. sediba fossils from South Africa, which were contemporaries of early Homo from East Africa. (Berger led the team that discovered A. sediba, which was announced in 2010 and held up as a possible ancestor of Homo.) By ignoring those South African fossils, Berger contends, the team cannot rule out alternatives to their interpretation.

Berger also took issue with the team’s use of fragmentary material to argue its position. “All this paper does, unfortunately, is highlight the mess that the isolated and fragmentary East African record in this time period makes of the debate around the origins of the genus Homo, and it does little to illuminate the question,” he contends. Berger has previously argued that A. sediba, which is best known from two largely complete skeletons exhibiting a mosaic of australopithecine–like and Homo-like traits, demonstrates that evolution mixed and matched fossil human features in sometimes surprising ways, and that fragmentary remains therefore cannot be reliably assigned to species. “We and others have shown that you can't take isolated bits and force them into anatomical association. The [Koobi Fora] mandible goes with the maxilla? Where is the evidence for that,” he demands. “While we need more fossils like this, it’s not helpful to shoehorn them into debates they are not complete enough to be of use as evidence in.”

Spoor counters that he and his colleagues did include the South African fossils in their analysis, but that they excluded those comparisons from their report because their Nature paper focuses on the question of what the new fossils reveal about taxa of early Homo in eastern Africa. “A. africanus and sediba have nothing to say about that,” he asserts, noting that africanus and sediba have primitive faces, with “nothing specifically Homo-like in the skull.” He adds, “the interesting parts of A. sediba are in the postcranial skeleton.”

Suffice it to say, I doubt very much that we have heard the last of this debate. Stay tuned.

Update (8/9/12 at 9:53 A.M.): Paleoanthropologist Philip G. Rightmire of Harvard University sent the below observations about the new fossils. Like Van Arsdale, Rightmire has long studied the Dmanisi fossils, but he arrives at different conclusion about the Koobi Fora remains:

"It's my impression that the authors are on the money in attributing their material to the hypodigm including KNM-ER 1470. For a long time, this group was quite poorly documented and therefore enigmatic. The new facial parts duplicate some of the key features of the original (very distinctive) face, but at a much smaller scale. Sex dimorphism and individual variation within a single lineage seems to be the best explanation. Also, it's clear that this lineage differs from specimens such as OH 13, OH 24, and KNM-ER 1813 (attributed to Homo habilis). I'd say that there is a good case for the presence of two distinct Homo lineages along side Homo erectus. For me, an important question is which (if any) of these hominins took the first steps out of Africa, to establish settlements at localities such as Dmanisi. Our material is best described as early Homo erectus, but I think that the direct ancestors to the Dmanisi population were a more archaic form of Homo. The first hominins out of Africa may have been Homo habilis (not the group documented in the new paper). Or we may have to keep looking for an appropriate ancestor to Homo erectus. In any case, Homo erectus evolving in Asia probably dispersed only later (back) to Africa, and of course toward the Far East."


The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.

Starting Thanksgiving

Enter code: HOLIDAY 2015
at checkout

Get 20% off now! >


Email this Article