Human evolution

Out of Ethiopia

Newly discovered fossils from Ethiopia provide fresh evidence for the 'out of Africa' model for the origin of modern humans, and raise new questions about the precise pattern of human evolution.

The idea that modern humans originated in Africa, with populations subsequently spreading outwards from there, has continued to gain support lately. But much of that support has come from analyses of genetic variation in people today1, and from fossil and archaeological discoveries dated to within the past 120,000 years2,3 — after our species evolved. Hard evidence for the inferred African origin of modern humans has remained somewhat elusive, with relevant material being fragmentary, morphologically ambiguous or uncertainly dated. So the fossilized partial skulls from Ethiopia that are described on pages 742 and 747 of this issue4,5 are probably some of the most significant discoveries of early Homo sapiens so far, owing to their completeness and well-established antiquity of about 160,000 years.

There are two broad theories about the origins of H. sapiens. A few researchers still support a version of the 'multiregional' hypothesis, arguing that the anatomical features of modern humans arose in geographically widespread hominid populations throughout the Pleistocene epoch (which lasted from around 1.8 million to some 12,000 years ago)6. But most now espouse a version of the 'out of Africa' model, although there are differences of opinion over the complexity of the processes of origin and dispersal, and over the amount of mixing that might subsequently have occurred with archaic (non-modern) humans outside of Africa2,7. Within Africa, uncertainties still surround the mode of modern human evolution — whether it proceeded in a gradual and steady manner or in fits and starts (punctuational evolution). Other questions concern the relationship between genetic, morphological and behavioural changes, and the precise region, or regions, of origin.

For instance, possible early H. sapiens fossils, dating from about 260,000 to 130,000 years ago, are scattered across Africa at sites such as Florisbad (South Africa), Ngaloba (Tanzania), Eliye Springs and Guomde (Kenya), Omo Kibish (Ethiopia), Singa (Sudan) and Jebel Irhoud (Morocco). But the best dated of these finds, from Florisbad and Singa, are problematic because of incompleteness and, in the latter case, evidence of disease. Meanwhile, the more complete or diagnostically modern specimens suffer from chronological uncertainties. So the most securely dated and complete early fossils that unequivocally share an anatomical pattern with today's H. sapiens are actually from Israel, rather than Africa. These are the partial skeletons from Skhul and Qafzeh, dating from around 115,000 years ago. Their presence in the Levant is usually explained by a range expansion from ancestral African populations, such as those sampled at Omo Kibish or Jebel Irhoud2,7,8, around 125,000 years ago.

The new cranial material from Herto, Ethiopia — described by White and colleagues4,5 — adds significantly to our understanding of early H. sapiens evolution in Africa. The fossils are complete enough to show a suite of modern human characters, and are well constrained by argon-isotope dating to about 160,000 years ago. Three individuals are represented by separate fossils: a nearly complete adult cranium (skull parts excluding the lower jaw), a less complete juvenile cranium, and some robust cranial fragments from another adult4. All display evidence of human modification, such as cut marks, considered to represent mortuary practices rather than cannibalism. Associated layers of sediment produced evidence of the butchery of large mammals such as hippopotamuses and bovines, as well as assemblages of artefacts showing an interesting combination of Middle Stone Age and late Acheulean technology5.

The morphology of the most complete of these three fossils helps to clarify the pattern of early H. sapiens evolution in Africa, as it shows an interesting combination of features from archaic, early modern and recent humans. The cranium is very large, but once the size is standardized, it shares with ancient African crania a wide interorbital breadth (the distance between the orbits of the eyes), anteriorly placed teeth, and a short occipital (the bone at the rear of the braincase). It also has a wide upper face and moderately domed forehead, as do the Skhul and Qafzeh fossils. Its low nose and face and flat midface are more widely shared early H. sapiens features, whereas other characteristics, such as its globular braincase, are typically modern. In the angulation and transverse ridge of the occipital, there is also an intriguing resemblance to fossils from sites such as Elandsfontein (South Africa) and Broken Hill (Zambia) that are often assigned to H. heidelbergensis or H. rhodesiensis. This may provide a clue to the individual's ancestors (Fig. 1). But overall, the fossil seems closest in morphology to particular crania from Jebel Irhoud, Omo Kibish and Qafzeh.

Figure 1: Origin of our species.

The figure shows the geographical and temporal distribution of hominid populations, based on fossil finds, using different taxonomic schemes. The new finds from Herto4,5 (H) represent early Homo sapiens. a, This reflects the view that both Neanderthals and modern humans derived from a widespread ancestral species called H. heidelbergensis2. b, However, evidence is growing that Neanderthal features have deep roots in Europe2,8, so H. neanderthalensis might extend back over 400,000 years. The roots of H. sapiens might be similarly deep in Africa, but this figure represents the alternative view that the ancestor was a separate African species called H. rhodesiensis. Different views of early human evolution are also shown. Some workers prefer to lump the earlier records together and recognize only one widespread species, H. erectus2 (shown in a). Others recognize several species, with H. ergaster and H. antecessor (or H. mauritanicus) in the West, and H. erectus only in the Far East8 (shown in b). Adapted with permission from refs 8, 11.

So White and colleagues' findings4,5 provide a plausible link back to more ancient African fossils, and forward to Levantine samples. They also raise questions about the overall pattern of modern human origins in Africa. Because of Africa's great area and still limited fossil record, it is uncertain whether the pattern of H. sapiens evolution there was essentially continent-wide, or was a more localized — and perhaps punctuational — process. The Herto finds shift the focus once again to East Africa. It seems from these crania and from possibly contemporaneous fossils, such as those at Ngaloba, Singa and Eliye Springs, that human populations of this era showed a great deal of anatomical variation. So, did the early modern morphology spread outwards from East Africa, perhaps gradually more archaic forms? Or could there have been an African version of multiregionalism, with modern morphology coalescing from various populations across the continent2,7,8? Only better samples and better dating of the African fossil record will help resolve these questions.

And what of the taxonomic status of the new finds? White and colleagues propose that, although measurements of the most complete fossil differentiate it from geologically 'recent' (that is, post-Pleistocene) H. sapiens, there is sufficient evidence to assign the material to this species overall, while naming a new subspecies, idaltu. However, in my opinion, the distinctive features described for H. sapiens idaltu might not be so unusual, and could probably be found in late Pleistocene samples from regions such as Australasia9.

Do the Herto fossils represent 'modern' H. sapiens? There is an ongoing debate about the concept of modernity, in terms of both morphological and behavioural characteristics2,3,7,8,10. Nevertheless, despite the presence of some primitive features, there seems to be enough morphological evidence to regard the Herto material as the oldest definite record of what we currently think of as modern H. sapiens. The fact that the geological age of these fossils is close to some estimates obtained by genetic analyses for the origin of modern human variation1 only heightens their importance.


  1. 1

    Ingman, M., Kaessmann, H., Pääbo, S. & Gyllensten, U. Nature 408, 708–713 (2000).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Stringer, C. Phil. Trans. R. Soc. Lond. B 357, 563–579 (2002).

    Article  Google Scholar 

  3. 3

    McBrearty, S. & Brooks, A. J. Hum. Evol. 39, 453–563 (2000).

    CAS  Article  Google Scholar 

  4. 4

    White, T. D. et al. Nature 423, 742–747 (2003).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Clark, J. D. et al. Nature 423, 747–752 (2003).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Wolpoff, M. & Caspari, R. Race and Human Evolution: A Fatal Attraction (Simon & Schuster, New York, 1997).

    Google Scholar 

  7. 7

    Lahr, M. & Foley, R. Yb. Phys. Anthropol. 41, 137–176 (1998).

    Article  Google Scholar 

  8. 8

    Hublin, J.-J in Human Roots: Africa and Asia in the Middle Pleistocene (eds Barham, L. & Robson-Brown, K.) 99–121 (Western Academic & Specialist Press, Bristol, 2001).

    Google Scholar 

  9. 9

  10. 10

    Lieberman, D., McBratney, B. M. & Krovitz, G. Proc. Natl Acad. Sci. USA 99, 1134–1139 (2002).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Rightmire, G. P. in Human Roots: Africa and Asia in the Middle Pleistocene (eds Barham, L. & Robson-Brown, K.) 123–133 (Western Academic & Specialist Press, Bristol, 2001).

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Chris Stringer.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Stringer, C. Out of Ethiopia. Nature 423, 693–695 (2003).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing