The earliest upright ape

Two decades ago the world of palaeoanthropologists was in turmoil with the publication of an account of a new find in Chad (see: Bonanza time for Bonzo; July 2002). A fossil cranium, dubbed Sahelanthropus tchadensis (nicknamed Toumaï­ or ‘hope of life’ in the Goran language), appeared like a cross between a chimpanzee and an australopithecine. The turmoil erupted partly because of its age: Upper Miocene, around 7 Ma old. Such an antiquity was difficult to reconcile with the then accepted ~5 Ma estimate for the evolutionary split between humans and chimpanzees, based on applying a ‘molecular clock’ approach to the difference between their mtDNA. The other point of contention was the size of Sahelanthropus’s canine teeth: far too large for australopithecines and humans, but more appropriate for a gorilla or chimp.

Cast of the reconstructed skull of Sahelanthropus tchadensis. (Credit: Didier Descouens, University of Toulouse)

In the absence of pelvic- and foot bones, or signs of the foramen magnum where the spinal cord enters the skull – crucial in distinguishing habitual bipedalism or being an obligate quadruped – encouraged the finders of a 6.1 to 5.7 Ma-old Kenyan hominin Orrorin tugenensis to insist that its skeletal remains – several teeth, fragments of a lower jaw, a thigh bone, an upper arm and of a finger and thumb but no cranial bones – were of ‘the earliest human ancestor’. In Orrorin’s favour were smaller canine teeth than those of later australopithecines. At the time of the dispute, centred mainly on absence of crucial evidence, doyen of hominin fossils Bernard Wood of George Washington University and an advocate of ‘untidy’ evolution, suggested that both early species may well have been evolutionary ‘dead ends’ (see: A considered view; October 2002). And there the ‘muddle’ has rested for 20 years.

In 2002 not only a cranium of Sahelanthropus had been unearthed. Three lower jaw bones and a collection of teeth suggested that as many as 5 individuals had been fossilised. A partial leg bone (femur) and three from forearms (ulna) cannot definitely be ascribed to Sahelanthropus but, in the absence of evidence of any other putative hominin species, they may well be. It has taken two decades for these remains to be analysed to a standard acceptable to peer review (Daver, G. et al. 2022. Postcranial evidence of late Miocene hominin bipedalism in Chad. Nature v. 608, published online; DOI: 10.1038/s41586-022-04901-z). The authors present convoluted anatomical evidence that Toumaï­’s femur, which had been gnawed by a porcupine and lacks joints at both ends, suggesting that it was indeed suited to upright walking. Yet the arm bones hint that it may have been equally comfortable in tree canopies. Yet it does look very like an ape rather than a hominin.

Much the same conclusion has been applied to Australopithecus afarensis, indeed its celebrated representative ‘Lucy’ met her end through falling out of a large tree ~3.2 Ma ago (see: Lucy: the australopithecine who fell to Earth?; September 2016). So, dual habitats may have been adopted by hominins long after they emerged. Yet Au afarensis was capable of trudging through mud as witnessed by the famous footprints at Laetoli in Tanzania. Only around 3 Ma has reasonably convincing evidence for upright walking similar to ours been discovered in Au africanus. The full package of signs from pelvis and foot for habitual bipedalism dates to 2 Ma ago in Au sediba. Even this latest known australopithecine seems to have had a gait oddly different from that of members of the genus Homo.

So, in many respects the benefits of full freeing of the hands to develop manipulation of objects, as first suggested by Freidrich Engels, may have had to await the appearance of early humans. Earlier hominins almost certainly did make tools of a kind, but the revolutionary breakthrough associated with humanity was more than 5 million years in the making.

See also: Callaway, E. 2022. Seven-million-year-old femur suggests ancient human relative walked upright. Nature (News)24 August 2022;

Handwerk, B. 2022. Seven Million Years Ago, the Oldest Known Early Human Was Already Walking. Smithsonion Magazine, 24 August 2022 (click the link ‘published today in Nature’ in 2nd paragraph to access complimentary PDF of Daver et al)

Tracing hominin evolution further back

The earliest hominin known from Africa is Sahelanthropus tchadensis, announced in 2002 by Michel Brunet and his team working in 7 Ma old Miocene sediments deposited by the predecessor to Lake Chad in the central Sahara Desert. Only cranial bones were present. From the rear the skull and cranial capacity resembled what might have been regarded as an early relative of chimpanzees. But its face and teeth look very like those of an australopithecine. Sadly, the foramen magnum – where the cranium is attached to the spine – was not well preserved, and leg bones were missing. The position of the first is a clue to posture; forward of the base of the skull would suggest an habitual upright posture, towards the rear being characteristic of knuckle walkers. Some authorities, including Brunet, believe Sahelanthropus may have been upright, but others strongly contest that. The angle of the neck-and-head ball joint of the femur (thigh bone), where the leg is attached to a socket on the pelvis to form the hip joint is a clue to both posture and gait. The earliest clear sign of an upright, bipedal gait is the femur of a fossil primate from Africa – about a million years younger than Sahelanthropus, found in the Tugen Hills of Kenya. Orrorin tugenensis was described from 20 bone fragments, making up: a bit of the other femur, three hand bones; a fragment of the upper arm (humerus); seven teeth; part of the left and right side of a lower jawbone (mandible). Apart from the femur that retains a neck and head and signifies an upright gait, only the teeth offer substantial clues. Orrorin has  a dentition similar to humans apart from ape-like canines but significantly smaller in size – all known hominins lack the large canines, relative to other teeth. Despite being almost 2 Ma older than Ardipithecus ramidus, the first clearly bipedal hominin, Orrorin is more similar to humans than both it and Australopithecus afarensis, Lucy’s species.

Near-complete skeleton of Oreopithecus bambolii from Italy (credit: Wikipedia Commons)

DNA differences suggest that human evolution split from that of chimpanzees about 12 Ma ago. Yet the earlier Miocene stratigraphy of Africa has yet to provide a shred of evidence for earlier members of either lineage or a plausible last common ancestor of both. In 1872, a year after publication of Charles Darwin’s The Descent of Man parts of an extinct primate were found in Miocene sediments in Tuscany and Sardinia, Italy. In 1950 an almost complete skeleton was unearthed and named Oreopithecus bambolii (see Hominin evolution becoming a thicket, January 2013). Despite dozens of specimens having been found in different localities, the creature was largely ignored in subsequent debate about human origins, until 1990 when it was discovered that not only could Oreopithecus walk on two legs, albeit differently from humans, it had relatively small canine teeth and its hands were like those of hominins, capable of a precision grip. Dated at 7 to 9 Ma, it may lie further back on the descent path of hominins; but it lived in Europe not Africa. Now the plot has thickened, for another primate has emerged from a clay pit in Bavaria, Germany (Böhme, M. and 8 others 2019. A new Miocene ape and locomotion in the ancestor of great apes and humans. Nature, online publication; DOI: 10.1038/s41586-019-1731-0).

Bones from 4 Danuvius guggenmosi individuals. Note the diminutive sizes compared with living apes (Credit: Christoph Jäckle)

Danuvius guggenmosi lived 11.6 Ma ago and its fossilised remains represent four individuals. Both femurs and a tibea (lower leg), together with the upper arm bones are preserved. The femurs and vertebrae strongly suggest that Danuvius could walk on two legs, indeed the vertebral shapes indicate that it had a flexible spine; essential for balance by supporting the weight of the torso over the pelvis. It also had long arms, pointing to its likely hanging in and brachiating through tree canopies. Maybe it had the benefit of two possible lifestyles; arboreal and terrestrial. Its discoverers do not go that far, suggesting that it probably lived entirely in trees using both forms of locomotion in ‘extended limb clambering’. It may not have been alone, another younger European primate found in the Miocene of Hungary, Rudapithecus hungaricus, may also have had similar clambering abilities, as might have Oreopithecus.

There is sure to be a great deal of head scratching among palaeoanthropologists, now that three species of Miocene primate seem – for the moment – to possess  ‘prototype specifications’ for early entrants on the evolutionary path to definite hominins. Questions to be asked are ‘If so, how did any of them cross the geographic barrier to Africa; i.e. the Mediterranean Sea?’, ‘Did the knuckle-walking chimps evolve from a bipedal common ancestor shared with hominins?, ‘Did bipedalism arise several times?’. The first may not have been as difficult as it might seem (see Africa_Europe exchange of faunas in the Late Miocene, July 2013). The Betic Seaway that once separated Iberia from NW Africa, in a similar manner to the modern Straits of Gibraltar, closed during the Miocene after a ‘mild’ tectonic collision that threw up the Betic Cordillera of Southern Spain. Between 5.6 and 5.3 Ma there was a brief ‘window of opportunity’ for the crossing, that ended with one of the most dramtic events in the Cenozoic Era; the Zanclean Flood, when the Atlantic burst through what is now the Straits of Gibraltar cataclysmically to refill the Mediterranean .

See also: Barras, C. 2019. Ancient ape offers clues to evolution of two-legged walking. Nature, v. 575, online; Kivell, T.L. 2019. Fossil ape hints at how walking on two feet evolved. Nature, v. 575, online; DOI: 10.1038/d41586-019-03347-0

Feet of the ancients

Cast of Footprints, Laetoli Museum
Cast of footprints, probably of Au. afrensis, from the famous trackway of Laetoli in Tanzania (Photo credit: GIRLintheCAFE)

Much of what palaeoanthropologists have surmised about the evolution of humans and their hominin forebears has come from fossils of their heads. Crania, jaws and teeth can reveal a lot about human ancestors and related species, and inevitably smart modern humans would dearly like to know how brainy and clever they were and when possible intellectual changes, such as the acquisition of language, might have taken place. But only the rest of the body gives us clues about what they did and potentially might have done. If, like Darwin, and following his lead Frederick Engels (, we believe that the single most important development was adopting an upright gait and thereby freeing the hands to manipulate the world, then fossil hands and feet are of very high importance. Yet they are among the most fragile appendages consisting of a great many separate bones, each being small enough to be transported by flowing water once soft tissues decay and a corpse falls apart. And they are easily bitten off by scavengers.  Heads are a lot bigger, heavier and robust, and being round and smooth, quite difficult for, say, a hyena or porcupine to gnaw. Moreover, disaggregated hominin foot and hand bones are not easy to recognise in fossiliferous sediments, especially if they have been scattered far and wide: the big prize being heads jaws and teeth, professional hominin hunters become expert at spotting them, but not necessarily the other 80% of skeletons.

Ardi (Ardipithecus ramidus)
Artists reconstruction of female Ardipithecus ramidus (Photo credit: Mike Licht,

So, the discovery of hominin hands or feet is a rare cause for celebration. A new partial foot has turned up in the hominin ‘bran-tub’ that is the Afar depression of NE Ethiopia (Haile-Selassie, Y. et al. 2012. A new hominin foot from Ethiopia shows multiple Pliocene bipedal adaptations. Nature, v. 483, p. 565-569) and has caused quite a stir. It is significantly different from the few other feet known from the hominin record. Moreover, it adds a sixth design to those already know, leaving out those of chimps, taken as likely to be similar to those of our shared common ancestor, Homo sapien, Neanderthals and H. erectus whose feet are much the same. While being easily distinguished from the feet of Homo species, those of australopithecines are sufficiently like them in basic morphology to suggest that Au. africanus and sediba both walked the savannas as upright as we do. But one of the earlier hominins, Ardipithecus ramidus, also from Afar but dated at more than 4 Ma, has provided an almost complete foot whose geometry , including a spayed-out, short big toe capable of grasping, almost certainly indicates that the creature was equally at home in trees as it was on the ground. Ardipithecus walked upright, but probably could not run as its gait placed the side of the foot on the ground, much like a chimpanzee, instead of proceeding heel-to-toe as we do (Lieberman, D.E. 2012. Those feet in ancient times. Nature, v. 483, p. 550-551). The new find seems similar, although better adapted for upright walking. Yet no other body parts have been found so it has not been assigned to a species, though it almost certainly represents a new one. The excitement concerns its age, which at 3.4 Ma is within the time range of Australopithecus afarensis, a family of which left the famous trackway at Laetoli in Tanzania whose foot prints strongly suggest full adaptation to human-like gait: walking, running and abandonment of partially habitual life in the trees.

It seems therefore that the multiplicity of co-existing hominins from 2 million years ago to very recently existed much further back in their evolutionary history. That raises several possibilities, among which is the possibility of repeated evolution of bipedality, hinted at by some similarities to the feet of modern gorillas in that of the newly found foot. Another implication is that simply being able to walk upright did not lead quickly to a tool-making ability because the earliest stone tools capable of cutting through meat, skin and sinew did not arise until 2.6 Ma. Like fossils of feet, those of hominin hands are extremely rare. The first crucial evidence of a hand with potential to manipulate objects delicately and with purpose is around 2 Ma, with the astonishingly well preserved hand of a young Au. sediba unearthed in South Africa ( Frustratingly, the 2.6 Ma tools are not associated with fossil hominins, and the Au. sediba skeletons had no tools.