ALL that palaeoanthropologists have to show for more than 100 years of digging are remains from fewer than 2000 of our ancestors. They have used this assortment of jawbones, teeth and fossilised scraps, together with molecular evidence from living species, to piece together a line of human descent going back 5 to 8 million years to the time when humans and chimpanzees diverged from a common ancestor. The big question now is are there any serious breaks in the lineage? Do the fossils we have discovered represent all or only some of the species of hominids that existed?
A good way to judge this is by comparing evidence from the past with current observations. For example, we can predict how many species of hominids were likely to be alive at any given time if we know how widespread they were and how much territory a similar modern species occupies. Combine these findings with estimates of how long each species survives before going extinct, and you get an idea of the total number of hominid species that existed over the past 5 or more million years or so.
But there is a problem. To interpret the scant evidence about human evolution scientists must group hominids into species and genera. How can they know where one species ends and another begins? Fossil remains don’t tell us which groups of individuals were interbreeding – a characteristic often used to distinguish one species from another. So classification is partly subjective. People tend to be either “lumpers” – who group together as many creatures as possible – or “splitters” who create new categories whenever they see significant differences in the samples. Fortunately, there is some consensus between the two.
Advertisement
Virtually all scientists in this field now agree that Charles Darwin was right – human beings, the Hominidae, did descend from African apes. The earliest known hominid fossils are all from Africa, with the oldest dating back around 4.4 million years. Less ancient finds show hominids spreading through Europe and Asia, then Australia and the Americas. These show that hominids became steadily less apelike until, around 100 000 years ago, there appeared people who could have mingled in a modern crowd without causing alarm.
All palaeoanthropologists recognise at least two genera – the earlier Australopithecus, exclusively from Africa, and the later Homo, found first in Africa but then globally. But this is where the consensus breaks down. Australopithecus includes two forms known as “gracile” and “robust”. The first have lightweight lower jaws and smoothly contoured skulls whereas the second have thick jaws with massive grinding teeth and sometimes crested skulls. For this reason, some people recognise two hominid genera as well as Homo -classifying the graciles as Australopithecus and the robusts as Pa ranthropus, a separate genus.
Separating hominid fossils into species is an even more contentious business. First there is a practical problem. Because good specimens are extremely rare, paleoanthropologists lack reliable evidence to help them distinguish one species from another. For example skulls which give most clues, are rarely found intact.
Little and large
Secondly, there is a conceptual problem. Hominids are primates, and modern primate species are extremely variable. Many, such as baboons and gorillas, are sexually dimorphic – the males are much larger than the females. But in some species, including gibbons, the sexes are a similar size. So if a collection of fossils falls into two categories, one bigger than the other, do you have two species or one that is sexually dimorphic?
Over the years, biologists have classified hominids in scores of different ways. This partly reflects genuine disagreement over the data and is partly a matter of taste. At the extremes, some lumpers recognise fewer than 10 species while splitters divide the known fossils into dozens of species. Bernard Wood, professor of anatomy at the University of Liverpool, has now steered a rigorous course between the two extremes.
Wood has looked at living species and identified those characteristics that vary greatly within a single species and those that do not vary much. For example, while the breadth of faces may vary enormously within a species, the heights of the orbits of the eyes and details of tooth shape do not. So fossils that vary greatly in tooth shape are likely to represent more than one species. Despite this exacting approach there will always be cases where the evidence is not clear cut. In these situations, Wood is a “splitter”.
Out of Africa
The table summarises, in his own words, thinking to date. Two things set this classification apart from many others. First, Wood recognises the third genus Paranthropus. In addition, his work supports the splitting of hominids into some new species. Within Paranthropus he recognises three species – two from East Africa and one from southern Africa. Notably, too, he divides H. habilis – the oldest acknowledged member of our own genus – in two, to give a more strictly defined H. habilis and a new species, H. rudolfensis characterised by a broader face, heavier lower jaw and larger grinding teeth. Wood also divides H. erectus into H. ergaster – the first hominid to spread out of Africa to Eurasia – and H. erectus, which evolved in Asia. Finally, he recognises three species among our closest ancestors. From “archaic” H. sapiens comes, H. heidelbergensis – the ancestor of both the Neanderthals, H. neanderthalensis, and ourselves, H. sapiens.
But does this really complete the list of all the species of hominid? There seems to be no way of knowing. Fossilisation is rare and even if the entire planet were excavated some species may have been lost without trace. Yet common sense suggests that we must already have uncovered most of them. The hominid species we know are large, and fossils have been found all over Africa and Eurasia. It seems unlikely that other species could have existed alongside them and then disappeared without trace.
Robert Foley, a bioanthropologist at the University of Cambridge has used a more theoretical approach to the problem. He also assumes that the split with chimpanzees came around 7.5 million years ago and then reasons that the number of hominid species existing since then depends on two things – how many new species evolved in that time and how many would have lived at any one time. His answers come from other animals that we know more about – living primates, cats, bears, pigs and the fossil record of mammals as a whole.
First, Foley showed that during the Tertiary – the period since the extinction of the dinosaurs 65 million years ago – mammal species tended to last for about one million years before becoming extinct or evolving into something else. Assuming hominids follow this pattern, Foley predicts around 7.5 successive waves of species since we diverged from chimpanzees. But how many species would there have been in each wave?
Vital clues come from studies of the catarrhines – Old World monkeys and the great apes. Gorillas, bonobos and other chimps between them occupy nearly 3 million square kilometres of Africa, which means one species on average per million km2. The 16 species of Cercopithecus – the genus that includes vervets and Diana monkeys – also average about 1 million km2 each. In fact, this figure holds good for catarrhine species as a whole. Foley then assumes that the pattern can be extended to include the first hominids – australopithecines, paranthropines and early Homo. These species were strictly African, between them occupying around 5 million km2, so at any one time around five species would have existed.
Wood’s hominid classification has five species living side by side only once in the past 4 million years or so. Briefly, around 2 million years ago, P. robustus and P. boisei shared the woodland and steadily expanding grassland of south and east Africa with H. habilis, H. rudolfensis and possibly the first H. ergaster. Even so, the number of known hominid species is clearly about right. A. afarensis and A. ramidus are unlikely to have been the only hominids living between 4.5 and 3 million year’s ago, and there should have been at least one more species before A. ramidus if we split from chimpanzees much before 5 million years ago. But, in general, Foley’s work confirms the common sense view – that we do know most of the species of hominids before they migrated out of Africa.
This is where the theory seems to break down. There is a huge discrepancy between the number of later hominids and what we would expect from living catarrhines. By at least 1 million years ago, Homo had spread through Africa deep into Asia – a total area of around 25 million km2. Why then is there no evidence of more than 20 hominid species instead of the two – H. ergaster and H. erectus – that Wood recognises? And where are all the predicted species from around 100 000 years ago when hominids occupied 30 million km2?
Foley has an explanation. H. erectus and H. ergaster developed a liking for meat. The australopithecines probably had a diet with meat providing only around 5 per cent of their daily calorie intake. Paranthropines were probably even more vegetarian. But H. ergaster and H. erectus were scavengers and possibly hunters with meat supplying 20 per cent or more of calories. As carnivores they would have needed much more space. Once again, moderns species give an idea of how big each range might have been. Bears, for example, average nearly 11 million km2per species, the predatory leopard covers a colossal 23 million km2 and pig species average around 7.5 million km2 each. Extrapolating from these findings, we can be reasonably confident that we have identified most of the more recent hominid species.
So, it seems we do have a fairly complete picture of the human family tree. According to Foley there are probably no more than 20 species in all. Wood divides the fossil record into a definite 13 and possibly as many as 18. The next step is to decide how these groups interrelate, and here again palaeoanthropologists disagree. What does seem likely, though, is that there will be no more major surprises about the origins of humans.
Boning up on dates
THERE are fewer than 20 published fragments of the oldest known hominid, Australopithecus ramidus. These probably come from just half a dozen individuals which, I believe, represent a single species. The evidence does, however, suggest that A. ramidus should be assigned to a new genus – its teeth are significantly smaller and the enamel thinner than those of later australopithecines. Earlier this year a new genus, Ardipithecus, was proposed.
Next comes A. afarensis, as exemplified by the famous “Lucy”. Recent studies reveal as much variation in the lower jaws and limb bones of A. afarensis as in any modern species. But the pattern of differences is not particularly striking – the only clear variation is in size. Few living species are separable by size alone so it seems unlikely that A. afarensis is more than one species. The size variation is probably explained by differences between males and females and by the way the species evolved.
The most recent of the gracile australopithecines is A. africanus. Fossil specimens include a minority that have larger teeth and jaws. Some anthropologists argue that these more robust individuals belong to a different species but I feel the difference is not significant enough. So A. africanus remains as one species.
The established robusts, which I and others place in the genus Paranthropus, come from both southern and East Africa. Some palaeoanthropologists feel that fossils of the southern African P. robustus really represent two species – P. robustus from Kromdraai and P. crassidens from Swartkrans. I think there are too few Kromdraai specimens to be sure. I do, however, agree that the East African specimens probably represent two species – the older P. aethiopicus and its even more robust descendant P. boisei. Despite the fact that they resemble one another, the southern and East African robusts may not even be closely related. If this idea gains support then the East Africans should take the generic name Zinjanthropus, which Louis Leakey originally proposed for P. boisei.
When the oldest member of our own genus. Homo, was discovered in 1964, anthropologists did not exactly welcome it. Some believed H. habilis was just another australopithecine or an early H. erectus. My work seems to confirm H. habilis as a distinct species. The foramen magnum – the hole at the base of the skull where the spinal column is attached – is further forward than in australopithecines. Further evidence comes from the teeth, which are narrower than those of australopithecines but not as small as in H. erectus.
More excitingly, I have shown that the fossils now ascribed to H. habilis probably represent two species. The specimens with broader faces, heavier lower jaws and larger grinding teeth fall into H. rudolfensis – a name suggested by a Russian anthropologist. But as a final complication, H. habilis and H. rudolfensis may not be true Homo at all. Their teeth are large relative to their body size and their skeletons suggest they had limb proportions closer to the australopithecines than to true Homo. Perhaps, H. habilis and H. rudolfensis should be placed in a new genus.
After H. habilis comes H. erectus. Again, fossils ma, more correctly be divided in two. The earlier kind which arose in and then migrated out of Africa could be reclassified as H. ergaster, leaving H. erectus in its new, narrow sense to describe those descendants of H. ergaster who migrated into Asia. Then H. ergaster would be our true ancestor. Evidence for this includes a sharp shelf at the back of the skull where the neck muscles attach. This is well defined in H. erectus but much less pronounced in H. ergaster and the earliest H. sapiens.
Finally, our own species, H. sapiens. Fossil remains are traditionally lumped together as “archaic” H. sapiens – including the famous neanderthals of Europe and the Near East who are placed in their own subspecies H. sapiens neanderthalensis – and modern humans who form another subspecies H. sapiens sapiens. I question this classification. “Archaic” skulls are very different from our own – more robust with large faces and teeth and conspicuous brow ridges. I classify H. neanderthalalensis as a species in its own right and place the rest of the archaics in a separate species. H. heidelbergensis. The latter arose in Africa and spread throughout the world. Both Neanderthals and modern humans are then seen as separate offshoots of H. heidelbergensis and classified as full species.