Before Lucy: Earliest human ancestors

The term “Primate” is due to the 18th Century Swedish biologist Carl Linnaeus, representing his view that this group, which of course includes humanity, sat firmly at the top of creation’s tree.

To most people, the term “primate” means the anthropoids, i.e. apes and monkeys – creatures that bear more than a passing resemblance to ourselves. Nor do the similarities end there. Although we tend to think of ourselves as something apart from the rest of nature, characteristic human features such as intelligence, upright walking and our complex society all stem from our primate roots.

What is a Primate?
Curiously there is no universal agreement as to what unique traits can be regarded as defining features of the Primates.

Primates are one of 21 orders of placental mammal (Infraclass Eutheria). There are around 200 living species of primate, conventionally grouped as lorises, lemurs, tarsiers, New World monkeys, Old World monkeys, lesser apes, great apes and humans. They are typically arboreal (tree-dwelling) inhabitants of tropical and subtropical ecosystem.

Primate skeletal anatomy contains features characteristic of mammals in general but also some that are more specific to the primates. Like other mammals, primates have four types of teeth – incisors, canines, premolars and molars. Early mammals had three incisors (I), one canine (C), four premolars (P) and three molars (M) on each side of each of their jaws (or quadrant) – a “dental formula” of 3-1-4-3; but primates have lost the first two premolars (P1 and P2).

Most primates possess the following features (Klein, 1999):

1.Retention of primitive mammalian conditions (i.e. those possessed by early mammals) such as the pentadactyl (5 digits) limb and the clavicle (collarbone). Many mammals have lost some of these digits, e.g. horses (who have a single digit).

2.Grasping hands and feet with highly mobile digits including big toe and usually an opposable thumb.

3.The replacement (in all living species) of primitive mammalian claws with nails (evidenced in the fossil record by flattening of the underlying phalanges.

4.Convergent orbits (eye-sockets) producing overlapping fields of vision and permitting stereoscopic (3d) vision, together with the enhanced neurological apparatus for processing visual information, requiring enlargement of the visual cortex in the occipital and temporal lobes of the brain. The orbits themselves are (in living species) surrounded by a bony ring known as the post- or circum-orbital bar. Higher primates add a bony wall known as the post-orbital plate or septum separating the orbits from the skull behind.

5.The muzzle or snout is shorter than that of most mammals and the sense of smell reduced, with corresponding reduction of the olfactory regions of the brain.

6.A fully bony auditory bulla (middle ear) in which the ventral floor is composed of an extension of the petrosal bone that encloses the inner ear, or a combination of the petrosal bone and the ectotympanic bone or tympanic ring (across which the eardrum is stretched). In other mammals the floor of the bulla generally comprises an independent entotympanic bone; this has been lost in the primates.

7.Reduction of the number of incisors and premolars compared with basal (early) mammals and a relatively simple and primitive cusp pattern on the molars.

8.A unique sulcal (fissure) pattern on the surface of the cerebral cortex of the brain. Also, relative to body size, primates tend to have larger brains than other mammals.

Finally we should ask what is the defining feature of the primate survival strategy – the “killer app” or “unique selling point” to take a couple of crude analogies? Most mammalian (and indeed other) groups have one: the bats combination of flight and echolocation, the rodent ability to gnaw and eat just about anything, the ruminant digestive system and so on. The primate “killer app” is brain-power; primates have large brains for creatures of their size, even in relation to other mammals. Basically the key to primate success has always consisted of being smarter than the competition.

Classifying the Primates
Within the mammals, Primates are grouped with the colugos (Order Dermoptera), tree-shrews (Order Scandentia) and the extinct plesiadapids in Superorder Eurarchonta (“true ancestors”). The Euarchonta is part of a larger grouping, Euarchontoglires, which additionally includes rodents and rabbits.

The plesiadapids were primate-like mammals bearing some resemblance to squirrels. They are now usually classed as order Plesiadapiformes within Euarchonta, though some continue to regard them as a suborder of the primates.

The Primates were traditionally divided between prosimians and anthropoids (or simians). The prosimians comprise four living groups: the lemurs, the lorises, the aye-ayes and the tarsiers.

Lemurs have long tails that they use for balance, but they cannot use them for grasping, unlike monkeys. Their tails are also used as a form of communication, and for male “stink fights”. They have the characteristic primate opposable thumbs and long toes adapted for gripping tree branches, and nails rather than claws on all digits except the second toe of each hind foot, which has a claw used for grooming. Lemurs have a tapetum or reflective layer over the retina, to enhance night vision. Smaller lemurs tend to be nocturnal; the larger ones are diurnal. Studies suggest lemurs have trichromatic colour vision, though it is less sensitive than that of humans. Lemurs range in size from 30 gm to 10kg. They occur in the wild only on Madagascar and are thought to have “rafted” there from Africa after Madagascar broke away from the mainland. Safe from competition, they flourished and diversified.

The lorises are slim diurnal arboreal primates living in tropical central Africa and south and southeast Asia. They have large forward-facing eyes and small ears. The thumbs are opposable and the index finger is short. In common with the lemurs the second toe of the hind legs has a claw for grooming. Their tails are either short or completely absent. They range from between 17 to 40 cm in length and between 0.3 to 2 kg in weight.

The galagos and pottos are grouped with the lorises in the lorids. Like the lorids they are tree-dwelling, but they are nocturnal.

The aye-aye is an endangered species known only from Madagascar. It is the sole member of the Chiromyiformes and is a large (2.5 kg) nocturnal primate. The relationship of the Chiromyiformes to the lemurs and the lorises is uncertain; some make it ancestral to the other two groups.

Tarsiers are small nocturnal primates, now found only in Island South-east Asia, although they were once more widespread. They are characterised by enormous eyes, long hind-limbs and extended tarsus bones (which gives the group its name). They second and third toes of the hindlimbs have grooming claws. They lack a tapetum but they do possess a retinal fovea, suggesting their ancestors were either diurnal or cathemeral (active both day and night).

The taxonomic status of the tarsiers is controversial. They share many key features with the anthropoids, including a hemochorial placenta (i.e. one in which the mother’s blood is in direct contact with the chorion (the outermost of the membranes surrounding the fetus)), a feature not possessed by other prosimians.

Accordingly the tarsiers are now generally grouped with the anthropoids in Suborder Haplorrhini (“dry-nosed”) with the remaining prosimians being placed in Suborder Strepsirrhini (“wet-nosed”). “Wet-nosed” refers to the rhinarium, a moist hairless pad around the nostrils of the nose. This feature can be seen in cats, dogs, mice and indeed it is present in most mammals, including the strepsirrhini, but it has been lost by haplorrhini.

The anthropoids are divided into two groups – the Platyrrhini (“flat-nosed”) – New World monkeys, including marmosets and tamarins; and Catarrhini (“narrow nosed”) – Old world monkeys (cerapithecoids), gibbons, apes and humans (hominoids). The Platyrrhini, the Catarrhini and the anthropoids at large are all believed to be clades or natural groups. The status of the Haplorrhini depends on the affinities of the tarsiers – if any – to the anthropoids. By itself, the term “monkey” does not define a natural group, since the Catarrhini also includes apes and humans. Similarly the apes are not a natural group unless humans are also included.

The extinct adapids and the omomyids are usually assigned to the Strepsirrhini and Haplorrhini respectively. The omomyids are believed to be more closely related to the tarsiers than to other prosimians and to form part of the Haplorrhini, albeit as an outgroup. The adapids are often claimed as ancestors to the lemurs, which seems plausible.

These groupings reflect supposed evolutionary relationships; the supra-ordinal classifications relate to early divergences of the placental mammals, probably occurring in the Cretaceous; the sub-ordinal classifications relate to more recent divergences; for instance the branch between the New and Old World monkeys probably occurred about 40 million years ago (mya).

Trying to represent these relationships within a traditional Linnaean taxonomy is problematic and can only be accomplished by use of a bewildering multiplicity of subdivisions of the main categories. It is far better to try to describe the above groups as clades, or branches on an evolutionary tree.

Origin of the Primates
Primates are of course mammals, a class of vertebrates that evolved from the therapsid or mammal-like reptiles during the Triassic period 220 mya and soon diverged into a number of lines including the therians, which turn includes the marsupials (metatherians) and placentals (eutherians). The placentals and marsupials diverged early in the Cretaceous period 125 mya; the earliest currently known placental is Eomaia scansoria (“climbing dawn mother”) discovered in Liaoning Province, China and described in 2002 (Ji et al, 2002). Early placental mammals were similar to modern animals making up the now-abandoned Order Insectivora, the most primitive mammals (moles, shrews, hedgehogs, etc) but now known to be polyphyletic (i.e. lacking a common ancestor that is part of the same group). These creatures tended to be small and were probably nocturnal – both advantages in a world dominated by dinosaurs.

The long reign of the dinosaurs came to an end at the Cretaceous/Tertiary (K/T) boundary 65.5 mya, almost certainly as a result of an asteroid that struck Earth leaving a massive crater near what is now the town of Chixulub in the Yukatan Peninsula in Mexico. The impact would have triggered tsunami, earthquakes and volcanic eruptions. Material ejected into space by the impact would have re-entered the atmosphere, triggering firestorms across the globe. Finally the dust lofted into the upper atmosphere would have produced an “impact winter” lasting for up to a decade after the impact. Under these conditions, it is little surprise that 50% of the world’s species became extinct.

Nobody doubts that the mammals were the main beneficiaries of this event, but there are three models for their diversification: the Explosive model, the Long Fuse model and the Short Fuse model.

The Explosive model – which is the traditional position – postulates that there was a mammalian “big bang” immediately following the dinosaurs’ disappearance; an evolutionary rush to fill the vacant niches. On this picture, the divergence of the mammalian orders from each other and subsequent diversification within these orders occurred mostly or entirely after the K/T boundary.

The Long Fuse model also places most of the diversification within each mammalian order after the K/T boundary, but in this scenario, the orders themselves diverged from each other well back in the Cretaceous.

Finally, the Short Fuse model postulates even within the various mammalian orders, some diversification had begun to occur over 100 mya, long before the end of the age of the dinosaurs.

These models should not be regarded as mutually exclusive and really mark three points on a continuum of possible models of mammalian diversification.

In 2003 a study was carried out (Springer, Murphy, Eizirik & O’Brien, 2003) applying Bayesian statistics to both genetic and taxonomic data, with constraints based on fossil data. Overall, the results supported the Long Fuse model. The primates diverged from other mammalian groups 85 mya, and the strepsirrhines and haplorrhine diverged 77 mya. The 85 mya date for primate emergence is 17 million years before the earliest known fossil of a possible primate, a creature known as Purgatorius, to which we now turn.

Purgatorius: the first Primate?
Purgatorius, known from a single tooth molar dated to the Late Cretaceous 67 mya and from more comprehensive Palaeocene remains, is the earliest possible primate known from the fossil record. Fossils are known from North America. Although its exact affinities are not known, dental evidence does strongly suggest a link with the primate order. Purgatorius had primitive molars with high cusps, three incisors, a relatively large canine, and four premolars in each quadrant (i.e. on each side of upper and lower jaws). The molars of later primates are more bulbous, none have more than two incisors, and only the more primitive retain four premolars; monkeys apes and humans have lost the two most mesial (i.e. towards the midline). Purgatorius is believed to have been about the size of a mouse or a small rat. It was almost certainly nocturnal, with large eyes adapted for night-vision. It possessed a shrew-like snout and had claws rather than nails. Its diet would probably have consisted of insects and fruit.

Purgatorius survived the K/T boundary event and it was sufficiently generalised in its anatomy to have given rise to the later Eocene primates, but there is no reason to suppose it actually did so.

The Plesiadapids
The Palaeocene (65-56 mya) is the first period of the Cenozoic era, directly following the extinction of the dinosaurs. At that time, the Americas were separated by ocean, but North America was connected to Europe. Climate was much warmer than today, with tropical and subtropical forests in middle latitudes. Palaeocene primates in the fossil record are best known from the Euroamerican land bridge.

The plesiadapids were the most abundant primates (or near-primates) of the Palaeocene and are known from the Palaeocene and Eocene epochs. They ranged in size between mice and squirrels, bearing some resemblance to the latter.

The plesiadapids thrived during the Palaeocene in North America and Eurasia, and some made it into the Eocene, but all had become extinct by the Oligocene, possibly due to competition from rodents. They probably arose in North America, later crossing the land bridge to Eurasia. Like Purgatorius they had claws rather than nails and possessed a generalized, un-evolved skeleton and an elongated skull in which the orbits (eye sockets) were confluent with the temporal region (sides and base) rather than being separated by a bony bar, as is the case with later primates. The affinity of the plesiadapids to later primates – if any – is not known with any certainty. Unlike Purgatorius they possessed dental specialisations including large procumbent central incisors (possibly used to grasp food), a reduced number of lateral incisors, anterior premolars or both (Klein, 1999). No later primate possesses these features, suggesting the plesiadapids died out without issue. Regardless of whether or not you consider them to be true primates, the pleasiadapids are what cladists call an “outgroup” in relation to (other) prmates, i.e. less closely related to other members of a clade.

Adapids and Omomyids
The Palaeocene is followed by the Eocene (56-34 mya). Early in this period the land connection between Europe and North America was broken; however the mild climate persisted. Forest vegetation, increased rainfall and hotter conditions had spread from the equator to the poles. Two diverse primate groups – the adapids and the omomyids – become very common in the early Eocene fossil records of North America, Europe, Asia and possibly Africa where Altiatlasius koulchii, known from 60 million year old teeth discovered in Adrar Mgorn in Southern Morocco, has been assigned to the omomyids. Both groups possessed features associated with living primates such as grasping hands and feet with digits tipped by nails instead of claws, and a complete postorbital bar. There was a shift in emphasis from sense of smell to vision.

The adapids tended to be relatively large, with a typical body mass of 1kg. Characteristic features included small orbits, suggesting a diurnal lifestyle and cheek teeth adapted for diets consisting of leaves and fruit. Postcranial remains (i.e. those below the cranium) suggest an arboreal lifestyle. They resembled lemurs but lacked some of the specialisations of living lemurs and lorises. They retain four premolars on each side of the jaw, whereas lemurs have only three. They had generalised lower incisors and canines; in lemurs these are elongated and protruding to form a dental comb. That the adapids were ancestral to the lemurs is widely but not universally accepted. That they were ancestral to the anthropoids seems less likely, though it cannot be ruled out.

By contrast the omomyids were smaller (below 500gms), with much larger orbits, suggesting a nocturnal habit. Their cheek teeth were adapted for a diet primarily of insects. They possessed elongated tarsal bones, similar to those of the tarsiers. It is a popular view that they were ancestral to the tarsiers and/or the anthropoids; again, though, this is not universally accepted.

Were early primates diurnal?
The commonly-held view that early primates were nocturnal rests largely on the fact that the majority of living prosimians are so, and the large orbits of many fossil forms, suggesting that they were also.

We have already seen that there is reason to believe that the tarsiers evolved from diurnal ancestors, and this is supported by a study of the gene sequences of opsins in primates, which rejects the nocturnality – or at least exclusive nocturnality – of ancestral primates.
Opsins are light-sensitive proteins found in retinal photoreceptor cells. Trichromatic or colour vision requires three types of opsin, sensitive to short, medium and long wavelengths. However colour vision is not is particularly useful for nocturnal animals, and has been found that in nocturnal animals either the genes coding for short wavelengths or those coding for medium/long-wave opsins rapidly pick up deleterious mutations, rendering the opsins themselves non-functional and giving the animal only monochromatic (“black-and-white”) vision. Because the “bad” opsin genes do not in such cases affect the organism’s survival, there is no Darwinian natural selection acting to eliminate them.

For any species, this mutation would be expected to occur at the same rate across successive generations, and on the nocturnal picture the opsin genes in all nocturnal primates would be expected to have undergone similar degrees of deleterious mutations, reflecting similar times of divergence from the last common diurnal ancestor (presumed not to be a primate).

However this prediction was not borne out by the study, which showed considerable variation in the degree of genetic defects found across a range of prosimians, indicating different time periods of deleterious mutation for different lineages, and suggesting different diurnal ancestry for each.
This in turn implies that the common primate ancestor of all of these lineages must have been diurnal, unless each lineage independently went through a phase of diurnality, before reverting to nocturnality, which seems unlikely (Tan, Yoder, Yamasita & Li, 2005).

Origin of the Anthropoids
As we have seen, the Springer, Murphy, Eizirik & O’Brien study gives a date of 77 mya for the divergence of the Haplorrhini from the Strepsirrhini. Other estimates range from 90 to 63 mya.
The origins of the anthropoids within this group are incompletely understood, with four competing theories of anthropoid phylogeny: 1) they evolved from Eocene adapids; 2) they evolved from Eocene omomyids; 3) they evolved from tarsiers; 4) they diverged from other haplorrhines early in the Cenozoic, and form a sister group to the omomyid/tarsier clade.
There are also four major competing views concerning the geographic origins of the anthropoids: 1) anthropoids originated in the circum-Tethyan region (the shores of the Tethys Sea, an ocean separating the supercontinents of Laurasia and Gondwana); 2) anthropoids originated on the Asian continental landmass; 3) anthropoids originated as part of a southern continental fauna, centred on Indo-Madagascar; or 4) anthropoids originated in Africa.

A review of existing data and theories about anthropoid origins was published in 2005 (Miller, Gunnell & Martin, 2005). Morphological, molecular and biogeographic evidence was considered. The review rejected ancestry with any known primate group. It claimed the anthropoids have existed as long as any known primate group, and that the time-depth is too great for their ancestry to be reliably determined, though its authors did not rule out one of known groups could be more closely related than the others.

This is a rather extreme view: the last phylogeny of the four above seems to be a less extreme position (Beard, K. C., Krishtalka, L. & Stucky, R. K., 1991). On this picture, there is a relatively early split in the Haplorrhini between the anthropoids and the common ancestor of both the omomyids and the tarsiers.

The Miller, Gunnell & Martin review also considered geographic origins of the anthropoids, and came down in favour of either an African or Indo-Madagascan origin.

Eosimias: Dawn Monkey from China
Currently, the earliest known anthropoid is Eosimias sinesis (“dawn monkey from China”), which is believed to have lived 45 mya. A fragment of a lower jawbone and foot bones were unearthed in China in the 1990s by joint American-Chinese expeditions led by Dr. Christopher Beard of the Carnegie Museum of Natural History. Eosimias is believed to have weighed no more than 100 gms, about the size of a pygmy marmoset. Beard and his colleagues claimed the remains suggest a Middle to Late Eocene emergence in eastern Asia of the mosaic of traits leading from primitive to anthropoid physiology. It shares derived features with undoubted anthropoids, including a 2-1-3-3 dental formula, a single-rooted anterior lower premolar (P2), shortening and crowding of the third and fourth premolars (P3 and P4) and labial expansion of the crowns. However, the two halves of the jawbone are not fused at the symphysis (cartilaginous joint), unlike all later anthropoids.

The anthropoid status of Eosimias was supported by a study published in Nature in 2000. A parsimony analysis of eleven tarsal characters using the PAUP 4.0 (Phylogenetic Analysis Using Parsimony) computer program suggested that Eosimias is a sister group to other anthropoids (Gebo, Dagosto, Beard, Qi & Wang, 2000).

Later Eocene Anthropoids
From Myanmar (Burma) come fragmentary jaws believed to date to 44-40 mya, representing at least two possible anthropoid species – the gibbon-sized Amphipithecus mogaungensis and the less well-known Pondaungia cotteri.

Amphipithecus had a number of features linking it with the anthropoids: the jawbone was deep compared with molar crown height and remained deep all along the jaw (in prosimians it is shallower, especially towards the front); the jawbone was fused at the symphysis (unlike Eosimias) and reinforced by an inferior and a superior transverse torus (a horizontal shelf-like thickening of bone; when this feature is present in prosimians they have only the inferior transverse torus); the second molar was parallel-sided (in common with anthropoids) and not narrowing towards the front (unlike the prosimians).

In common with Eosimias the dental formula is 2-1-3-3. The molar crowns were relatively flat, with low blunt cusps, suggesting a diet focussed more on leaves and fruit and less on insects than is typical for prosmimians; its premolar morphology was close enough to that of Eosimias to suggest descent; its overall form was sufficiently generalised for it to be ancestral to both the Catarrhini and the Platyrrhini (Klein, 1999).

Fossils of the Fayum Depression, Egypt
One of the most productive fossil sites for early anthropoids is the Jebel Qatrani Formation in the Fayum Depression, near Cairo, Egypt. The sheer diversity of the Fayum taxa is strong evidence for an African origin and early radiation for the anthropoids. It also supports the greater time depth for the anthropoids than was once supposed, as suggested in Miller, Gunnell & Martin review.

The site was originally attributed to the early Oligocene, but is now believed to straddle the Eocene/Oligocene boundary, 34 mya. Late Eocene material includes partial jaws, skull fragments and limb bones and has been assigned to four anthropoid genera (Catopithecus, Proteopithecus, Serapia and Arsinoea), all believed to have been small, weighing less than 900 gm, roughly the size of a squirrel monkey. Their premolars and molars resemble those of later Oligocene anthropoids.

A nearly complete skull of Catopithecus reveals clear anthropoid features including fused frontal bones, a postorbital plate separating the orbit from the braincase and an ectotympanic bone fused to the margin of the auditory bulla. However its brain was smaller than that of a modern anthropoid of the same size. Proteopithecus retained the 2-1-3-3 dental formula of Amphipithecus, the formula still found in living platyrrhines, meaning it may lie near the common ancestry of both the platyrrhines and the catarrhines. On the other hand Catopithecus has lost a premolar and has the 2-1-2-3 found in later catarrhines.

Both Catopithecus and Proteopithecus are tentatively assigned to the propliopithecids, a catarrhine/proto-catarrhine family known mainly from the Oligocene. However the affinities of these taxa to later groupings are very tentative and the 2-1-3-3 dental formula of Proteopithecus makes such an association questionable.

The Platyrrhine/Catarrhine split
New World monkeys show up in the South American fossil record from 25 mya. Nobody knows for certain how they got there but rafting from Africa likeliest explanation. This theory suggests that the ancestors of the New World monkeys (and presumably other animals) were swept out to sea by flash floods, together with mats of vegetation on which they were able to survive until the currents brought them ashore in South America.

Although it might seem more feasible to suggest that the New World monkeys reached South and Central America via North America, there are two objections. Firstly there is no fossil evidence to suppose that any anthropoids ever lived in North America prior to the arrival of humans; secondly the Americas were still separate and actually further apart than South America was from Africa and South America. Furthermore there were probably chains of islands stretching between the two continents, so series of shorter rafting events could have taken place. Notably some rodents are also believed to have reached South America from Africa.

Estimates based on molecular data suggest the split between the Platyrrhine and Catarrhine occurred approximately 40 mya. Although the platyrrhines are believed to have originated in Africa, fossil evidence is lacking. Proteopithecus has many features resembling both fossil and present-day platyrrhines, but these appear to be pleisiomorphies, i.e. features shared with anthropoids predating the platyrrhine/catarrhine split (Miller & Simons, 1997).

Into the Oligocene
The Oligocene (34-23 mya) is the third period of the Cenozoic Era. At the transition to the Oligocene global temperatures dropped sharply as the configuration and topography of the continents changed, altering oceanic and atmospheric circulation. The cooler, dryer conditions resulted eventually in the disappearance of subtropical forest from middle latitudes. It was during this period that the promsimians went into decline and became confined largely to nocturnal niches, suffering from both loss of habitat (in Europe and North America) and competition from the anthropoids (in Africa and Asia). Only in Madagascar – which the anthropoids never reached – did the prosimians continue to flourish.

Roughly 1000 primate specimens have been recovered from the Oligocene layers of the Fayum, of which all but a handful are anthropoids. Twelve species in six genera are generally recognised: Qatrania, Apidium, Parapithecus, Propliopithecus, Aegyptopithecus and Oligopithecus. These are grouped into the parapithecoids (Qatrania, Apidium and Parapithecus) and the propliopithecoids (Propliopithecus, Oligopithecus and Aegyptopithecus). Not all lived at the same time and some may have evolved from others.

By the standards of present-day anthropoids, the Oligocene Fayum taxa were small, ranging from 300gms for Qatrania to 3-4 kg (Parapithecus) up to 5.9 kg (Aegyptpithecus, some species of Propliopithecus). They were still larger than any modern insectivorous primates, suggesting they were mainly fruit or leaf eaters. The majority had short molar shearing crests, suggesting they were fruit eaters. Most were probably arboreal (by analogy with similar-sized living primates). The limb bone morphology suggests they were either quadruped-climbers (Propliopithecus and Aegyptopithecus) or quadruped-leapers (Apidum).

The parapithecoids possess the 2-1-3-3 dental formula of platyrrhines. They are probably basal to the anthropoids as a whole: their origins predating the platyrrhine/catarrhine split. It seems quite likely that they represent a separate and extinct branch of the anthropoids and are not ancestral to either the platyrrhines or the catarrhines.

The propliopithecoids have a 2-1-2-3 dental formula, suggesting a link with the catarrhines. Skull and postcranial evidence, in particular from Aegyptopithecus, suggests that they lie close to the ancestry of both the Old World Monkeys and the hominoids, but predate the split between these two groups.

Cerapithecoids and Hominoids
The penultimate split in the primate line is that between the cerapithecoids (Old World Monkeys) and the hominoids (apes and humans). Just when the split occurred is not known for certain due the poor fossil record. However the split must have occurred between the time of Aegyptopithecus (33 mya); and that of the first generally accepted apes Morotopithecus (21 mya) and Proconsul (20 mya). A commonly-accepted date is 23-25 mya (Late Oligocene/Early Miocene), based on molecular data, though one study using a technique known as maximum likelihood-based quartet analysis places it much further back, at approximately 30 mya in the Early Oligocene. Maximum likelihood-based quartet analysis is a computer-intensive statistical technique that has found widespread use in calculating divergence times (Steiper, Young and Sukarna, 2004).

The Early Miocene
The Miocene (23 – 5.33 mya) is the fourth period of the Cenozoic Era. By this time the configuration of the continents was approaching that today, and many familiar geographic features formed as a result of continental collisions during the Miocene. These included the Himalayas, the Tibetan Plateau, the Ethiopian highlands and the Rift Valley. The Ethiopian highlands interrupted the eastwards flow of rain, creating a rain-shadow across East Africa, which saw a significant rainfall reduction. This effect was exacerbated by the uplift of the Himalayas and the Tibetan Plateau.

From the earliest Miocene, 23 million years ago, the rainforest belt covering Africa had been breaking up into distinct ecological niches with increasing patches of woodland/grassland interrupting the rainforests.

As Africa’s northward drift brought it into contact with Eurasia, the Tethys Sea closed, causing southern Eurasia to become cooler and dryer. Equally important, the new land connections permitted faunal exchange between Africa and Eurasia, and among the migrants were the hominoids, who radiated through the forests of Eurasia, from Spain to China.

In eastern Africa, primates inhabited tropical forest and woodland. Most were hominoids with cerapithecoids being infrequent. Cerapithecoids only predominated in the less forested regions such as northern Africa.

A number of early Miocene hominoid species are known from fossil deposits in eastern Africa. The best known is Proconsul but at least five other genera are known: Dendropithecus, Micropithecus, Morotopithecus, Afropithecus and Turkanapithecus.

The ape-like Proconsul, discovered in 1909 and named in 1931, appeared in Kenya about 20 mya and possessed some Old World monkey features such as short forelimbs and a deep, narrow chest. Its forelimbs were adapted for walking on palms rather than the knuckle-walking of modern chimps and gorillas. In common with modern apes it lacked a tail and had a slightly larger brain, relative to its size, than a monkey. Proconsul was a forest-dwelling arboreal quadruped, spending much of its time in the trees. Its facial anatomy and low-cusped thin enamelled molars suggest that its diet consisted largely of soft fruit. It lacked the enlarged fruit-peeling incisors of chimps and orang-utans and the high-cusped leaf-chewing molars of gorillas. It was sufficiently generalised to be close to the common ancestry of all three. Proconsul comprised 3-5 species, ranging in size from 11 kg (roughly the size of a gibbon) to 87 kg (orang-utan sized). The name “Proconsul” is a reference to it having evolved before chimpanzees; “Consul” was a common name for chimpanzees at that time, including one at London Zoo.

Out of Africa – and back again!
Kenyapithecus africanus, a hominoid that appeared around 16-15 mya, is often touted as the direct ancestor of the great apes and humans. Its more immediate descendents could have included a number of later Eurasian Miocene hominoids: Ouranopithecus, Sivapithecus and Gigantopithecus. Kenyapithecus had a robust jawbone, enlarged upper premolars and thick enamel – features that are certainly steps in the direction.

By this time, beginning 17-18 mya hominoids were migrating across the newly-established land bridge into Eurasia where, as already noted, they flourished. At least seven genera probably existed: Dryopithecus (14-8 mya, known from central and western Europe), Pliopithecus (16-11 mya, known western and south-central Europe and southern China), Ouranopithecus a.k.a. Graecopithecus (9.6-8.7 mya, known from Greece), Sivapithecus (12.5-7 mya, known from Siwalik Hills on India-Pakistan border and in Turkey), Gigantopithecus (6.3-0.5 mya, Siwalik, western China, Vietnam [largest primate ever, bigger than a gorilla]), Lufengpithecus (8-7 mya, known from Lufen in southern China) and Oreopithecus (9-8 mya, known from Italy).

These apes ranged in size from gibbons to chimpanzees, though the aptly-named Gigantopithecus was even larger than a gorilla and is the largest primate ever to have lived. They were further evolved in the direction of modern apes. Dryopithecus, for example, had a shortened inflexible lumbar vertebral (waist) skeleton, broad flat thorax (chest), scapulae (shoulder-blades) behind the thorax (rather than alongside), long arms that could rotate around the shoulder joint, implying upright (orthograde) posture (as opposed to prograde), and able to hang below branches or climb hand over hand – it is the oldest hominoid for which this characteristic ape-like posture and locomotion can be implied.

Others, such as Ouranopithecus had thick enamel, permitting it to eat nuts, seeds, tubers and other hard grit encrusted food, which became common in relation to fruit as woodland replaced forest. Faunal remains suggest it occupied open woodland and savannah and had to spend significant amounts of time on the ground to get at its preferred foodstuffs. Its teeth resemble those of the later australopithecines, suggesting it is close to the line leading to humans. Unfortunately no post-cranial remains have been found.

Sivapithecus also had thick enamel and lived in open woodland. It had many features in common with orang-utan and is thought to be ancestral. If so, it suggests orang-utans had split from the line leading to African hominids by 12 mya.

If any of these Eurasian apes was the common ancestor to modern-day gibbons, great apes and humans, how did we end up with gibbons and orang-utans in Asia, and gorillas, chimps and humans in Africa? The poor hominoid fossil record in Africa after 13 mya has led some to propose scenarios in which hominoids left Africa, but their descendants later returned.

One possible scenario is that the ancestor of the gibbons left Africa, leaving behind the ancestor(s) of orang-utans, chimps and gorillas. Subsequently the ancestor of orang-utans dispersed to Asia.

A second scenario, involving fewer assumptions, is that the ancestor of all of the above (possibly Kenyapithecus) dispersed out of Africa and gave rise (via one of the genera described above) to the gibbons and great apes in Eurasia. One group later moved back to Africa and begat the gorillas, chimps, and humans. This move back to Africa would have happened after the gibbon/great ape divergence, but before the emergence of Sivapithecus. This “Out of Africa and back again” scenario seems highly plausible.

If this second scenario is accepted, then the immediate ancestor of the hominids emerged from Eurasia 14-15 million years ago and joined a general faunal migration into Africa, taking advantage of the new cooler conditions at the expense of the endemic fauna.

Finally, about 6 million years ago, the line diverged again, with one line leading to chimps and the other to the australopithecines and, ultimately, to humans.

Nakalipithecus nakayamai
However recent discoveries (Kunimatsu, Yutaka et al., 2007) have cast doubt on the “Out of Africa and back again” scenario. A jawbone and teeth recovered from Nakali, Kenya have been assigned to a new genus of great ape, Nakalipithecus nakayamai. N. Nakayamai is described as a large hominoid with dental size corresponding to female gorillas and orang-utans and it has similarities to Ouranopithecus. It appears to be slightly older than Ournaopithecus (9.9-9.8 mya, versus 9.6-8.7 mya). Another fossil great ape, Chororapithecus abyssinicus has also recently been discovered in Ethiopia (Suwa et al, 2007).

These findings together with the long-established existence of a third Late Miocene hominid, Samburupithecus kiptamali, discovered in 1982 in the Samburu Hills, northern Kenya, weaken the view that hominoids disappeared from Africa in the late Miocene. Although it is not necessarily the case that any of these apes are the last common ancestor of humans and African apes, it may well be the case that they are close relatives.

Clearly further fossil evidence is required, and all that can safely be said at the present time is that the matter is far from settled.


Beard, K. C., Krishtalka, L. & Stucky, R. K. 1991: First skulls of the early Eocene primate Shoshonius cooperi and the anthropoid–tarsier dichotomy. Nature (349), 64–67 (1991).

Cameron, D and Groves, C 2004: Bones, Stones and Molecules: “Out of Africa” and Human Origins, Elsevier Academic Press.

Colbert, E and Morales, M 1991: Evolution of the Vertebrates (4th edition), John Wiley & Sons, Inc.

Gebo, Dagosto, Beard, Qi & Wang 2000: The oldest known anthropoid postcranial fossils and the early evolution of higher primates, Nature (404) 16 March 2000 pp 276-278.

Groves, C 1991: A Theory of Human and Primate Evolution, Clarendon Press Oxford.

Ji et al 2002: The earliest known eutherian mammal. Nature (416), pp 816-822.

Klein, R 1999: The Human Career (2nd edition), University of Chicago Press.

Kunimatsu, Yutaka et al. (2007): A new Late Miocene great ape from Kenya and its implications for the origins of African great apes and humans. PNAS 104(49) pp.19220–19225 (December 4, 2007)

Levin, R and Foley, R 2004: Principles of Human Evolution (2nd edition), Blackwell Publishing.

Miller, Gunnell & Martin 2005: Deep Time Depth and the search for Anthropoid Origins, Yearbook of Physical Anthropology 48:60–95 (2005).

Miller, E & Simons, E 1997: Dentition of Proteopithecus sylviae, an archaic anthropoid from the Fayum, Egypt, PNAS Vol. 94, pp. 13760–13764, December 1997

Springer, Murphy, Eizirik & O’Brien 2003: Placental mammal diversification and the Cretaceous-Tertiary boundary, PNAS February 4, 2003 vol. 100 no. 3 1056-1061.

Steiper, M, Young, N and Sukarna, T 2004: Genomic data support the hominoid slowdown and an Early Oligocene estimate for the hominoid–cercopithecoid divergence PNAS December 7, 2004 vol. 101 no. 49 17025.

Suwa, Kono, Katoh, Asfaw & Beyene 2007: A new species of great ape from the late Miocene epoch in Ethiopia, Nature (448), 23 August 2007, pp 921-924.

Tan, Yoder, Yamasita & Li 2005: Evidence from opsin genes rejects nocturnality in ancestral primates, PNAS October 11, 2005 vol. 102 no. 41 14715.

© Christopher Seddon 2008


Author: prehistorian

Prehistorian & author