The earliest humans to leave Africa, in China

Since discovery in 2010 that remains of the genus Homo at Dmanisi in Georgia were about 1.85 Ma old several more instances of bones and stone tools a few hundred thousand years less than that age have turned up in China. All have been ascribed to H. erectus, although there are dissimilarities with African examples of the species and its predecessor H. ergaster. The technological breakthrough that led H. erectus/ergaster to knap the distinctive bifacial or Acheulean ‘handaxe’ was achieved at about the same time as the Dmanisi humans left Africa, yet there is no sign of such tools in eastern Asia until much later, most ancient artefacts there being of a more primitive, ‘Oldowan’ type. That is perhaps because more serviceable tools were fashioned from less durable materials than fine-grained rock that takes an edge. Maybe the skills were lost en route or the forebears of eastern Asian tool makers left Africa before the breakthrough. At any rate, the genus Homo is generally conferred on any being that had a tool-making culture, so that the presence of tools alone in a sedimentary deposit signifies that humans probably once inhabited that site. The earliest tools (3.3 Ma) from the Turkana area of Kenya were made half a million years before the first known appearance of well-documented remains of an un-named member of the genus Homo at  Ledi-Geraru in Afar, Ethiopia (2.8 Ma). At sites in Olduvai, Tanzania (1.9 Ma) and Turkana, Kenya (2.1 Ma) fossils of Homo habilis are found in association with ‘Oldovan’ stone tools.

Sites where early human fossils an tools have been found. (Credit: John Kappelman, Nature 2018; doi:10.1038/d41586-018-05293-9)

The latest development in the origin and wanderings of early humans has emerged from studies of a thick deposit of windblown silt or loess that makes up the Loess Plateau (Latitude 34°N) between the Yellow and Yangtze Rivers in central-east China. The loess is divided into several sequences by thin soil horizons (palaeosols). The entire stratigraphy contains tiny grains of iron minerals whose magnetic polarity was aligned with the Earth’s magnetic field at the time of deposition. This allows periods of normal and reversed geomagnetic polarity to be detected with considerable precision. Measurements have been taken at 10 cm intervals throughout the loess, to give an unbroken record of events throughout the Pleistocene Epoch that can be matched to a dated reference called the geomagnetic polarity timescale (GPTS). Palaeoclimate researchers have been able to show that the layers of loess correspond to successive glacial stages, whereas the palaeosol represent warm interglacials, exactly as recorded in sea-floor sediment profiles   A team of archaeologists from China and Britain have found primitive, Oldowan-type, artefacts in both the loess and palaeosol horizons at 17 different levels (Zhu, Z. and 10 others 2018. Hominin occupation of the Chinese Loess Plateau since about 2.1 million years ago. Nature, v. 559 advance publication online doi:10.1038/s41586-018-0299-4. See also). The artefacts are positioned at levels dated at between 1.26 to 2.12 Ma by the palaeomagnetic dating (from the Réunion to Cobb Mountain normally polarized subchrons).

Primitive stone tool (four sides shown) from the Loess Plateau of China. (Credit: Zhu et al./Nature 2018)

So, in both cool and warm conditions (34°N has cold winters today) toolmakers were regularly present in central, east China for almost 900 ka. The earliest must have made a 14 thousand km trek from tropical Africa across several climatic zones, and been physically, cognitively and technologically capable of surviving and reproducing for the one- to three-thousand years the journey must have taken (based on a dispersal rate of 5 to 15 km a year estimated from modern hunter-gatherers’ activities). Either there were repeated migrations of this scale or a pioneer population survived on or within reach of the loess steppe for hundreds of thousand years. The earliest emigrants would have been neither Homo erectus nor ergaster, for neither had evolved. Their age suggests that they may have been H. habilis, a view that has been expressed for the ancestors of the diminutive H. floresiensis known to have been present of the Indonesian island of Flores for around 700 ka. Until actual fossils are unearthed – not easy as the sequence is exposed in very steep slopes characteristic of dissected loess terrains – who the first occupants of China were remains mysterious. But one thing stands out: If early humans from that long ago could arrive, survive and prosper half a world away from their place of origin, then paleoanthropologists must consider the possibility of continual diffusion of the genus Homo away from its African origins once equipped with the ability to make tools. China may become the focus for early-human research as it became for that into the origins of birds and feathered dinosaurs.

You can read more about early humans and their evolution here.

A fully revised edition of Steve Drury’s book Stepping Stones: The Making of Our Home World can now be downloaded as a free eBook

Genus Homo pushed back nearly half a million years

Bill Deller, a friend whose Sunday is partly spent reading the Observer and Sunday Times from cover to cover, alerted me to a lengthy article by Britain’s doyen of paleoanthropologists Chris Stringer of the Natural History Museum. (Stringer, C. 2015. First human? The jawbone that makes us question where we’re from. Observer, 8 March 2015, p. 36). His piece sprang from two Reports published online in Science that describe about 1/3 of a hominin lower jaw unearthed – where else? – in the Afar Depression of Ethiopia. The discovery site of Ledi-Geraru is a mere 30 km from the most hominin-productive ground in Africa: Hadar and Dikika for Australopithecus afarensis (‘Lucy’ at 3.2 Ma and ‘Selam’ at 3.3 Ma, respectively); Gona for the earliest-known stone tools (2.6 Ma); and the previously earliest member of the genus Homo, also close to Hadar.

On some small objects mighty tales are hung, and the Ledi-Geraru jawbone and 6 teeth is one of them. It has features intermediate between Australopithecus and Homo, but more important is its age: Pliocene, around 2.8 to 2.75 Ma (Villmoare, B. And 8 others. Early Homo at 2.8 Ma from Ledi Geraru, Afar, Ethiopia. Science Express doi: 10.1126/science.aaa1343). The sediments from which Ethiopian geologist Chalachew Seyoum, studying at Arizona State University, extracted the jawbone formed in a river floodplain. Other fossils suggest open grassland rich with game, similar to that of the Serengeti in Tanzania, with tree-lined river courses. These were laid down at a time of climatic transition from humid to more arid conditions, that several authors have suggested to have provided the environmental stresses that drove evolutionary change, including that of hominins (DiMaggio, E.N. and 10 others 2015. Late Pliocene fossiliferous sedimentary record and the environmental context of early Homo from Afar, Ethiopia. Science Express doi: 10.1126/science.aaa1415).

Designating the jawbone as evidence for the earliest known member of our genus rests almost entirely on the teeth, and so is at best tentative awaiting further fossil material. The greatest complicating factor is that the earliest supposed fossils of Homo (i.e. H. habilis, H rudolfensis and others yet to be assigned a species identity) are a morphologically more mixed bunch than those younger than 2 Ma, such as H. ergaster and H. erectus. Indeed, every one of them has some significant peculiarity. That diversity even extends to the earliest humans to have left Africa, found in 1.8 Ma old sediments at Dmanisi in Georgia (Homo georgicus), where each of the 5 well-preserved skulls is unique.  The Dmanisi hominins have been likened to the type specimen of H. habilis, but such is the diversity of both that is probably a shot in the dark.

English: Cast replica of OH 7, the type specim...
Replica of OH 7, the deformed type specimen of Homo habilis. (credit: Wikipedia)

Coinciding with the new Ethiopian hominin papers a study was published in Nature the same week that describes how the type specimen of H. habilis (found, in close association with crude stone tools and cut bones, by Mary and Lewis Leakey at Olduvai Gorge, Tanzania in 1960) has been digitally restored from its somewhat deformed state when found (Spoor, F. et al. 2015. Reconstructed Homo habilis type OH 7 suggests deep-rooted species diversity in early Homo. Nature, v. 519, p. 83-86, doi:10.1038/nature14224). The restored lower jaw and teeth, and part of its cranium, deepened the mysterious diversity of the group of fossils for which it is the type specimen, but boosts its standing as regards probable brain size from one within the range of australopithecines to significantly larger –~750 ml compared with <600 ml – about half that of modern humans. The habilis diversity is largely to do with jaws and teeth: it is the estimated brain size as well as the type specimen’s association with tools and their use that elevates them all to human status. Yet, the reconstruction is said by some to raise the issue of a mosaic of early human species. The alternative is an unusual degree of shape diversity (polymorphism) among a single emerging species, which is not much favoured these days. An issue to consider is: what constitutes a species? For living organisms morphological similarity has to be set against the ability for fertile interbreeding. Small, geographically isolated populations of a single species often diverge markedly in terms of what they look like yet continue to be interfertile, the opposite being convergence in form by organisms that are completely unrelated.

Palaeontologists tend to go largely with division on grounds of form, so that when a specimen falls outside some agreed morphological statistics, it crosses a species boundary. Set against that the incontrovertible evidence that at least 3 recent human species interbred successfully to leave the mark in all non-African living humans. What if the first humans emerging from, probably, a well-defined population of australopithecines continued to interbreed with them, right up to the point when they became extinct about 2 Ma ago?

On a more concrete note, the Ledi Geraru hominin is a good candidate for the maker of the first stone tools found ‘just down the road’ at Gona!

Another candidate for earliest, direct human ancestor

The cranium of Malapa Hominid 1, Holotype of A...
The head of Australopithecus sediba. Image via Wikipedia

In May 2010 EPN commented on a new find from the famous fossil-rich caves of north-eastern South Africa; a new hominin species called Australopithecus sediba. At least one of them fell into a deathtrap shaft, died and remained unchewed without bones being spread far and wide. Inevitably, near-complete skeletons of individual hominins are soon pored over by dozens of specialists in human evolution, as they were for the much older Ardepithecus ramidus found in sediments of Ethiopia’s Afar Depression (see Early hominin takes over Science magazine in the November 2009 issue of EPN). Now there are two near-complete, well-preserved skeletons of Au. sediba and the palaeoanthropological world is agog. Dating to about 1.98 Ma the specimens represent the same time as do far less impressive remains of H habilis from Tanzania that were found with associated rudimentary stone tools. The first hint (just a fragment of upper jaw) of any remains that might be tagged ‘Homo’ dates to 2.3 Ma and is from Ethiopia, as are the first undoubted stone tools going back as far as 2.5 Ma, though lacking association with a maker.

Five consecutive papers on Au. Sediba occupy 22 pages in the 9 September 2011 issue of Science and make for startling reading. The first concerns the shape of its brain case, and therefore crudely its brain, discerned by tomographic X-ray scanning (Carlson, K.J. et al. 2011. The endocast of MH1, Australopithecus sediba. Science, v. 333, p. 1402-1407). It isn’t any bigger than that of other members of the genus but shows ‘some foreshadowing of the human frontal lobes’ and other shifts from the basic ape model that the authors imply are en route to human features. The next considers the two pelvis regions (Kibii, J.M. et al. 2011. A partial pelvis of Australopithecus sediba. Science, v. 333, p. 1407-1411); again australopithecine-like in the small size of the birth canal but with a hint of the S-shape of humans. Most astonishingly well-preserved are the fragile bones of a complete hand (Kivell, T.L.  et al. 2011. Australopithecus sediba hand demonstrates mosaic evolution of locomotor and manipulative abilities. Science, v. 333, p. 1411-1417), which convincingly shows the long thumb and short fingers (for a primate) that characterise Homo and are essential for a precision grip and making things. Actually, the thumb is longer relative to fingers (60%) than in humans (54%), but Lucy’s (Au. afarensis) was a closer match. No tools that such a hand might have created and wielded were found with the fossils. Then there is the foot (Zipfel, B. et al. 2011. The foot and ankle of Australopithecus sediba. Science, v. 333, p. 1417-1420), which, again, mixes human and australopithecine features, giving ‘a unique form of bipedality and some degree of arboreality’. The fifth paper (Pickering, R. et al. 2011. Australopithecus sediba at 1.977 Ma and implications for the origins of the genus Homo. Science, v. 333, p. 1417-1420) is as remarkable for the precision of U-Pb dating of speleothem (cave carbonates), which at 1.977+0.002 Ma far exceeds the workhorse Ar-Ar method used for most other hominins, as it is for the absolute age that precedes that of undisputed remains of humans.

In short, for Australopithecus sediba there is an embarrassment of riches unmatched until those of the 1.5 Ma old H. erectus (‘Turkana Boy’) found at Nariokotome in NW Kenya. To some extent this throws a flock of peregrines in among the palaeoanthropology pigeons, as an account of a meeting earlier in 2011, at which the bones were grandstanded, shows (Gibbons, A. 2011. Skeletons present an exquisite paleo-puzzle. Science, v. 333, p. 1370). Naturally, the authors are making the most of their material especially, it seems, its finder Lee Berger of the University of Witwatersrand, South Africa, the last author in all the papers. Comparisons with more australopithecine remains were said to be needed. The soon-to-be-famous hand has been said to be essentially like others from the same genus. While the remains of the creature’s pelvis could imply that its evolution was more driven by a need for efficient upright walking than to birth big-headed babies, the ankle shows a primitive trait that would have forced Australopithecus sediba to walk strangely as the heel bone is small and angled unlike that in human feet, which is broad and flat. But all the species’s features are combined in two near-complete individuals, whereas for the rest of its contemporaries, predecessors and near successors in time speculation is based on fragments of several individuals, none more so than in the case of the earliest agreed human, near contemporaneous H. habilis, which barely stands up to taxonomic scrutiny (Gibbons, A. 2011. Who was Homo habilis – and was it really Homo?  Science, v. 332, p. 1370-1371). Some would say that it was only the associated stone tools that assigned ‘Handy Man’ to more elevated status than slightly large-headed australopithecine. The fact is; stone tools were around since 2.5 Ma, at least in Ethiopia, and this newly found being could have handled them and even made them with its palpable dexterity. Finding tools and skeletons together is almost as rare as hens with teeth…