Archaeological evidence for carcass processing at Kanjera, Kenya, 2 million years ago.

Earliest unambiguous evidence for meat-eating by early hominins.

Modern humans are the only existent primates anatomically adapted for the regular consumption of significant quantities of meat. The human gut is reduced compared with that of other primates, a configuration more suited to a meat-eating diet than the predominantly vegetarian diet of other primates. Although crucial to many models of hominin evolution, however, the timing of and circumstances in which early hominins began to include significant quantities of meat in their diet remain poorly understood.

The earliest-known stone tools, from Gona, Ethiopia, are 2.6 million years old and are often taken to be early evidence for meat eating (Semaw, et al., 1997; Semaw, 2000). No hominin remains were recovered in association with the tools, but in 1999, anthropologists working at the nearby Bouri Formation reported the discovery of large mammal bones bearing cut-marks apparently made by stone tools, possibly as a result of dismembering and filleting carcasses. Animals appeared to have been defleshed, and their long bones broken open, presumably to extract marrow. The bones were found in association with 2.5 million-year-old australopithecine remains, thought to be of Australopithecus garhi (de Heinzelin, et al., 1999).

It has also been claimed that 3.39 million-year-old animal bones from Dikika, Ethiopia, show stone tool cut-marks for flesh removal, and signs of having been struck with hammerstones to extract bone marrow (McPherron, et al., 2010). In the absence of any associated tools, there is no way to tell whether the cut-marks were produced with specially-made tools or naturally-sharp pieces of stone. Some are sceptical and argue that as the bones were buried in coarse-grained, sandy deposits, it is likely that trampling by animals produced the marks (Domınguez-Rodrigo, et al., 2011).

Even if the above is accepted as evidence of carcass-processing by early hominins, it is too insubstantial to show whether these were one-off forays into meat-eating or part of a more substantial shift in hominin dietary adaptations. To demonstrate ‘persistent carnivory’ requires a geologically-stratified series of relatively large assemblages of animal remains, each showing extensive signs of persistent hominin activity. The sum of the assemblages must demonstrate that this activity persisted over the course of at least a thousand years (Ferraro, et al., 2013).

Although rather more recent than the above dates, such evidence has now been reported from Kanjera South, a small site located on the shores of Lake Victoria, southwestern Kenya (Ferraro, et al., 2013). Three excavations along 50 metres have yielded several thousand well-preserved animal remains, approximately 2 million years old, and associated with stone tools. There is a consistent record of hominin activities throughout the stratified sequence, which spans hundreds or possibly thousands of years.

The animal remains included gazelle and other small bovids, together with a smaller number of medium-sized bovids. The remains showed clear evidence of butchery by hominins in the form of cut-marks and damage caused by hammerstones. Patterns of tooth-marks made by carnivores such as lions and hyenas suggest that these animals only had access to the carcasses after the hominins had removed the bulk of the meat and bone marrow. Carnivores typically chew on the mid-shafts of long bones, but the percentage of bones that were so marked was low.

Small bovids are invariably wholly consumed by carnivores within hours of death, implying that the hominins acquired and butchered them very soon after death. A possible implication is that these animals were hunted rather than scavenged, and that Kanjera represents the earliest archaeological record of hunting activities by hominins.

The skeletal remains of the small bovids suggest that they were transported to the site for butchery more or less intact. However, in the case of the medium-sized bovids, head and limb parts predominate. These animals were too large to transport intact, so the hominins removed the limb parts, leaving the rest of the body behind. Although head contents are nutritious, they are difficult to exploit and would thus be ignored by other scavengers. They therefore represent a niche that tool-using hominins could exploit. It is therefore likely that hominins scavenged leftover head parts from carnivore kills and transported them to the site for processing.

The Kanjera data not only provides the required evidence of hominin meat-eating over a period of many centuries: it also provides clues about specific activities. Thus, it seems, the hominins obtained much of their meat by hunting small bovids, but they also scavenged medium-sized bovid heads as a separate by complimentary activity. The date of 2 million years ago is somewhere between 200,000 and 500,000 years earlier than the previous earliest evidence for persistent hominin carnivory.


1. Semaw, S. et al., 2.5-million-year-old stone tools from Gona, Ethiopia. Nature 385, 333-336 (1997).

2. Semaw, S., The World’s Oldest Stone Artefacts from Gona, Ethiopia: Their Implications for Understanding Stone Technology and Patterns of Human Evolution Between 2.6–1.5 Million Years Ago. Journal of Archaeological Science 27, 1197–1214 (2000).

3. de Heinzelin, J. et al., Environment and Behavior of 2.5-Million-Year-Old Bouri Hominids. Science 284, 625-629 (1999).

4. McPherron, S. et al., Evidence for stone-tool-assisted consumption of animal tissues before 3.39 million years ago at Dikika, Ethiopia. Nature 466, 857-860 (2010).

5. Domınguez-Rodrigo, M., Pickering, T. & Bunn, H., Reply to McPherron et al.: Doubting Dikika is about data, not paradigms. PNAS 108 (21), E117 (2011).

6. Ferraro, J. et al., Earliest Archaeological Evidence of Persistent Hominin Carnivory. PLoS One 8 (4) (2013).


Australopithecus sediba: a possible human ancestor

Australopithecus sediba is a possible human ancestor discovered in South Africa in 2010. The discovery was made at Malapa, a fossil-bearing cave located about 15 km (9.3 miles) NE of the well-known South African hominid-bearing sites of Sterkfontein and Swartkrans and about 45 km (28 miles) NNW of Johannesburg  (Berger, et al., 2010). It is situated within the Cradle of Humankind World Heritage Site. The recovery effort was led by Lee Berger, a paleoanthropologist at the University of the Witwatersrand, Johannesburg. The find was made when Matthew, Lee’s 9 year old son, discovered hominin collar bone embedded in a rock (Balter, 2010).

The find comprised two extremely well-preserved partial skeletons that were initially thought be somewhere between 1.78 and 1.95 million years old (Dirks, et al., 2010), later revised to 1.977 million years (Pickering, et al., 2011). These belonged to a juvenile male (MH1) aged 12 to 13 at time of his death and an adult female (MH2) (Berger, et al., 2010). They were found together buried in alluvial sediment, deep within the Malapa cave, part of an eroded cave system. Also found were the remains of wildcats, hyenas and a number of other mammals. On the ground above the cave are a number of ‘death traps’, or long vertical shafts. The smell of damp issuing from the shaft would have attracted animals. The pair – possibly mother and son – may have fallen to their deaths while searching for water. The sediments imply that subsequent high-volume water inflow, perhaps the result of a large storm, caused a debris flow. This carried the still partially articulated bodies deeper into the cave, to deposit them along a subterranean stream (Dirks, et al., 2010).

MH1 and MH2 were assigned to a new australopithecine species, Australopithecus sediba. The word ‘sediba’ means ‘fountain’ or ‘wellspring’ in the Sotho language. The more complete cranium of the juvenile MH1 has a capacity of 420cc, probably at least 95 percent of adult size. The remains share numerous similarities with Australopithecus africanus in the cranial vault, facial skeleton, lower jawbone and teeth, but there are also significant differences in the cranial, dental and postcranial anatomy. Homo-like features include smaller molars and premolars and less pronounced cheekbones. Certain features of the pelvis are similar to those seen in Homo erectus. The lower-to-upper limb bone proportions are also similar to those of later Homo, and unlike the more apelike proportions of Homo habilis. The anatomy of its hip, knees and ankles suggest that Australopithecus sediba was a habitual biped. Overall, it was claimed that Australopithecus sediba shares more derived features with early Homo than it does with other australopithecines. However, Berger was reluctant to place the new discovery within Homo, preferring to classify it as an australopithecine (Berger, et al., 2010)

The initial announcement of Australopithecus sediba attracted extensive news coverage, but not everybody was convinced by the claims made for it. Australian anthropologist Darren Curnoe was reported (MacKnight, 2010) as claiming that Australopithecus sediba is in the wrong place at the wrong time to be a human ancestor. He noted that Homo habilis emerged in East Africa well before the time of Australopithecus sediba. However, his argument does assume that Homo habilis is indeed an early human.  This may not be the case. It is also possible that at least some of Australopithecus sediba’s humanlike features could have evolved independently, and may not necessarily imply shared ancestry (Wood & Harrison, 2011).

Nevertheless, subsequent studies do support Berger’s initial claims. They suggest that aspects of the brain, dental morphology, pelvis, hand and foot of Australopithecus sediba could be interpreted as incipient humanlike features. A virtual endocast of the brain, obtained from synchrotron scanning, revealed an australopithecine-like size and pattern of convolutions. However, the orbitofrontal region showed possible development towards a humanlike frontal lobe. Possibly some neural reorganization of the brain preceded its later size increase in early humans (Carlson, et al., 2011).

The teeth of MH1 and MH2 are a mosaic of primitive and derived traits. Cladistic analysis of 22 dental traits suggest that Australopithecus sediba was a sister species of Australopithecus africanus(i.e. the two shared a common ancestor) and that the two were further evolved in the direction of Homo than were the australopithecines from East Africa (Irish, Guatelli-Steinberg, Legge, de Ruiter, & Berger, 2013). The lower jawbone morphology reduced dentition (especially canines and premolars) confirms that Australopithecus sediba was a distinct species to Australopithecus africanusand not merely a late-surviving form of that species (de Ruiter, et al., 2013).

The upper ribcage of Australopithecus sediba exhibits an apelike funnel shape, unlike the barrel shape associated with Homo. The funnel shape, as noted above, may be an adaptation to under-branch suspensory locomotion. The barrel shape may be associated with the increased chest volume and lung function necessary for endurance walking and running. The lower thorax, however, appears less flared than that of apes and more closely approximates the morphology found in humans (Schmid, et al., 2013). The spine is long and flexible, a form that has more in common with early Homo than with other australopithecines. Curvature of the lower spine is a hallmark of walking upright (Williams, Ostrofsky, Frater, Churchill, Schmid, & Berger, 2013).

The upper limbs were still predominantly apelike, suggesting the retention of substantial climbing and suspensory abilities (Churchill, et al., 2013). The hands show a mixture of australopithecine and human features. They retained adaptations for tree-climbing, but there was also a long thumb and shorter fingers. These suggest precision gripping of the type associated with tool manufacture and use (Kivell, Kibii, Churchill, Schmid, & Berger, 2011).

The pelvis and foot presented a mosaic of apelike and humanlike characteristics. These suggested adaptations to a more efficient (albeit not entirely human) form of bipedalism, at the expense of reduced arboreal efficiency (Kibii, et al., 2011; Zipfel, DeSilva, Kidd, Carlson, Churchill, & Berger, 2011). The bipedal mechanics differed from those reconstructed for other australopithecines, suggesting that there may have been several forms of hominin bipedalism at this time. The adaptations of Australopithecus sediba may have enabled it to both walk and climb reasonably well and thus survive in a dual arboreal/terrestrial world (DeSilva, et al., 2013).

Balter, M. (2010, April 9). Candidate Human Ancestor From South Africa Sparks Praise and Debate. Science, 328, 154-155.
Berger, L., de Ruiter, D., Churchill, S., Schmid, P., Carlson, K., Dirks, P., et al. (2010, April 9). Australopithecus sediba: A New Species of Homo-Like Australopith from South Africa. Science, 328, 195-204.
Carlson, K., Stout, D., Jashashvili, T., de Ruiter, D., Tafforeau, P., Carlson, K., et al. (2011, September 9). The Endocast of MH1, Australopithecus sediba. Science, 333, 1402-1407.
Churchill, S., Holliday, T., Carlson, K., Jashashvili, T., Macias, M., Mathews, S., et al. (2013, April 12). The Upper Limb of Australopithecus sediba. Science, 340.
de Ruiter, D., DeWitt, T., Carlson, K., Brophy, J., Schroeder, L., Ackermann, R., et al. (2013, April 12). Mandibular Remains Support Taxonomic Validity of Australopithecus sediba. Science, 340.
DeSilva, J., Holt, K., Churchill, S., Carlson, K., Walker, C., Zipfel, B., et al. (2013). The Lower Limb and Mechanics of Walking in Australopithecus sediba. Science, 340.
Dirks, P., Kibii, J., Kuhn, B., Steininger, C., Churchill, S., Kramers, J., et al. (2010, April 9). Geological Setting and Age of Australopithecus sediba from Southern Africa. Science, 328, 205-208.
Irish, J., Guatelli-Steinberg, D., Legge, S., de Ruiter, D., & Berger, L. (2013, April 12). Dental Morphology and the Phylogenetic “Place” of Australopithecus sediba. Science(340).
Kibii, J., Churchill, S., Schmid, P., Carlson, K., Reed, M., de Ruiter, D., et al. (2011, September 9). A Partial Pelvis of Australopithecus sediba. Science, 333, 1407-1411.
Kivell, T., Kibii, J., Churchill, S., Schmid, P., & Berger, L. (2011, September 9). Australopithecus sediba Hand Demonstrates Mosaic Evolution of Locomotor and Manipulative Abilities. Science, 333, 1411-1417.
MacKnight, H. (2010, April 8). Experts reject new human species theory. Retrieved September 12, 2012, from Independent:
Pickering, R., Dirks, P., Jinnah, Z., de Ruiter, D., Churchil, S., Herries, A., et al. (2011, September 9). Australopithecus sediba at 1.977 Ma and Implications for the Origins of the Genus Homo. Science, 333, 1421-1423.
Schmid, P., Churchill, S., Nalla, S., Weissen, E., Carlson, K., de Ruiter, D., et al. (2013). Mosaic Morphology in the Thorax of Australopithecus sediba. Science, 340.
Williams, S., Ostrofsky, K., Frater, N., Churchill, S., Schmid, P., & Berger, L. (2013, April 12). The Vertebral Column of Australopithecus sediba. Science, 340.
Wood, B., & Harrison, T. (2011, February 17). The evolutionary context of the first hominins. Nature, 470, 347-352.
Zipfel, B., DeSilva, J., Kidd, R., Carlson, K., Churchill, S., & Berger, L. (2011, September 9). The Foot and Ankle of Australopithecus sediba. Science, 333, 1417-1420.