Ancient History & Civilisation


On Their Own Two Feet

Not everyone agrees about when, exactly, the family Hominidae came into existence—that is, when the last ancestor of Homo sapiens lived that was not also the ancestor of one or more of the great apes. Partly this is because the ancient fossil record of the hominids is sparse; partly it is because what there is of that record is difficult to interpret; and partly it is because there is no precise agreement right now on how much time the molecular (DNA) differences that have so far been measured among the living hominoids (humans and the greater and lesser apes) tell us has elapsed since our lineage went its own way. Still, we have made progress. In 1950, nobody had the slightest idea in calendar terms about how far back into time the hominid family could trace its roots. The techniques necessary for making an estimate in years simply weren’t there. But in the 1960s, after the arrival of chronometric dating methods, it came to be widely believed that some fragmentary fossils from India and Kenya, 12 to 14 million years old and known variously as Ramapithecus and Kenyapithecus, might be the remains of a human precursor.

Even as this notion was weakening under the onslaught of new fossil discoveries, scientists in the emerging field of molecular systematics (in which molecular structures rather than anatomical ones are compared in order to determine zoological affinities) made an astonishing counterclaim, arguing for a much more recent point of hominid emergence, perhaps as little as 5 million years ago. In the last quarter of the twentieth century there was some convergence of such estimates, mostly toward the shorter end of the scale, with the paleontologists abandoning the notion of extreme hominid antiquity, and the molecular systematists easing up in their insistence on its great youth. Most observers, whatever kind of data they are dealing with, are relatively content at present with the notion that the last common ancestor of human beings and of one or more of the apes lived around 7 million years ago, give or take a million years or so. But this is a fluid number and not one that is likely to solidify any time soon.

Not very long ago there were no fossil contenders for hominid status that dated to more than 3 to 4 million years ago. Now, thanks to active fieldwork and some remarkable discoveries, there are several candidates in the 4-to-7-million-year range. Still, the picture remains a little murky, not least because we are not entirely sure what to expect that our earliest ancestor looked like. In considering this matter, paleoanthropologists have traditionally started by contemplating themselves. We human beings differ from our closest living relatives in a variety of respects, and during the last century or so several different human peculiarities have been taken as the defining characteristic of humanity.

Among the most obvious distinctive human characteristics is our large brain, three times the volume (even relative to body size) of that of any ape. Early paleoanthropologists were particularly entranced by this symbol of human superiority, so much so that almost all of them were prepared to be taken in by the Piltdown hoax, sprung in 1911. The supposedly very ancient skull found at Piltdown, in southeastern England, was eventually exposed as a fabrication that combined a recent human braincase with a modern ape jaw. But for the almost half-century before the fraud was exposed, this ‘‘specimen’’ stood as powerful testimony that an enlarged brain had been the key human feature from the very beginning—even once evidence began to accumulate that it had not been.

Once the enlarged human brain had lost its glamour in this regard, scientists began to look elsewhere for the human hallmark. Our precision grip (key to the venerable notion of ‘‘man the toolmaker’’) and our very small canine teeth (great apes’ canines are rather large, especially in males) were both considered and ultimately rejected as uniquely diagnostic criteria. Researchers eventually focused their attention on our upright, two-legged posture, which is nowadays almost universally considered to be the defining characteristic of the human lineage. Nothing that was not an upright biped could be considered hominid. Of course there was a logical flaw here, for our expectation is no more than an assumption. What we need to do is to demonstrate that a fossil candidate for hominid ancestry is not excluded from that position by any of its characteristics, not to demonstrate that it has passed some predefined threshold that is based on a derived characteristic of later hominids. Navel-gazing aside, though, the search for the first hominid has in practice boiled down over the past few decades to the search for the first upright biped. And the problem has become that few if any of the fossils recently claimed to be very early hominids have (at this writing) a clearly demonstrable bipedal form.

This cranium of Sahelanthropus tchadensis, a putative early hominid from Chad, in central-western Africa, is between 6 and 7 million years old; it is currently the most ancient claimant to membership in the hominid family. Courtesy of Michel Brunet.This cranium of Sahelanthropus tchadensis, a putative early hominid from Chad, in central-western Africa, is between 6 and 7 million years old; it is currently the most ancient claimant to membership in the hominid family. Courtesy of Michel Brunet.

The very earliest fossil to have been described as a hominid is a cranium (skull minus the lower jaw) that was found in Chad, in central-western Africa, a discovery announced in 2002. It is believed to be about 6 to 7 million years old. Not only is this an extraordinarily early date for a hominid, but the specimen comes from a decidedly unexpected place: almost all other early African hominids have been discovered thousands of miles to the east, in the Rift Valley region of eastern Africa, and in South Africa.

Sahelanthropus tchadensis, as the skull has been named in reference to where it was found, is surprising in its morphology, too. To give some context, when you compare the skull of, say, a chimpanzee to that of a human being, you first of all notice that the relationship between the facial skeleton and the braincase is totally different in the two species. In the chimp, the facial skeleton is large, projects forward prominently, and contains big jaws and teeth. It dwarfs the small braincase that sits behind it. In a gorilla, the braincase viewed from the side looks a bit bigger in proportion to the face than a chimpanzee’s, but only because a large flange of bone (called the sagittal crest) projects vertically along the midline of the skull, making the braincase look bigger than it is. This ridge is there to compensate for a shortage on the small skull surface of muscle attachment area for the huge jaw muscles. In the human skull, in contrast, the small, flat face and jaws are tucked beneath the front of a huge, balloon-like braincase. The effect could not be more different.

In light of these comparisons, Sahelanthropus is odd. Its face is massive, but flat, with an oddly ‘‘modern’’ look to it, while its tiny braincase is very apelike, even bearing a trace of sagittal crest. It bears rather small canine teeth, and its describers have found evidence of a rather forward-placed foramen magnum. This last feature is the large hole in the base of the skull through which the spinal cord joins the brain; it is typically found beneath the skull in species with upright posture, whereas in four-legged animals it points more directly backward. Naturally, the discoverers of Sahelanthropus find hominid resemblances here, although these can be disputed. It is altogether an extraordinary specimen. So how does Sahelanthropus compare with other very early supposed hominids?

In the case of the other 6-million-year-old fossil candidate for classification as a hominid, it’s a little hard to say. This is because the poorly known Orrorin tugenensis, discovered in 2000 in the Baringo Basin of northern Kenya, consists so far mostly of postcranial bones—that is to say, bits of the body skeleton. The bones in question are mostly parts of a couple of femora (thigh bones) and part of a humerus (upper-arm bone). And although there is nothing to dispute the assertion by the fossils’ finders that the leg bones show features associated with upright walking, the parts really needed to confirm this have so far not been found. The few known teeth of Orrorin, described in 2001, are also not easy to interpret. The premolar and molar (chewing) teeth of other early hominids tend to be rather large, yet these are fairly small; so is the one known canine tooth, but in shape it is considered to be rather chimpanzee-like.

The picture is muddied yet further by another claimed early hominid also described in 2001. This is Ardipithecus kadabba, a name given to some fragmentary fossils from sites in Ethiopia dated to between 5.8 and about 5.2 million years ago. The A. kadabba scraps include a foot bone that is thought to indicate bipedality. But even if this is accurate, we should be wary of concluding that Ardipithecus was bipedal in any familiar way. The describer of a later (about 4.4 million years old) species of Ardipithecus, A. ramidus, warns that anyone wanting to find an analog for the way it walked should ‘‘check out the bar scene in Star Wars.’’ The A. ramidus fossil material also includes teeth that are rather atypical for hominids. However, it has been said to represent an upright biped because it includes a fragment of cranial base that apparently shows a forward-positioned foramen magnum.

Where does all this leave us? We have a very motley assemblage of purported early hominid material from the period between 6-plus and 4.4 million years ago, and it may be significant that Ardipithecus has been compared with chimpanzees and Sahelanthropus with gorillas. But if all of these forms, or even some of them, are genuine hominids, they establish that from the very beginning the history of the human family has not been the single-minded slog from primitiveness to perfection so beloved of the devotees of the evolutionary synthesis. Rather, it has been a history of evolutionary experimentation, a process of exploration of the many different ways that there evidently are to be hominid. This is an important lesson for us to learn. The fact that Homo sapiens is the only hominid species on the Earth today makes it easy to assume that our lonely eminence is historically a natural state of affairs—which it clearly is not.

So what set this process of evolutionary experimentation in motion? Episodes of diversification within groups of organisms, often known as adaptive radiation, are frequently spurred by changes in the environment. And it appears that the hominid radiation was no exception. During most of the Miocene epoch, which ended about 5.2 million years ago, the African continent, in which the hominid family emerged, was largely covered by forests of various kinds. In these forests had flourished a diverse variety of hominoid primates, that is, members of the group from which both human and ape ancestors emerged. About 10.5 million years ago, polar cooling and a seasonal decline in rainfall toward the equator began to affect the African forest cover, leading to the gradual breakup of dense forests and the consequent spread of more open woodlands and grassy areas. Along with this change the diversity of the forest-living Miocene hominoids began to dwindle, and it is probably no coincidence that the hominid family began to establish itself just as more open habitats were becoming a significant part of the African landscape.

Clearly, though, hominids did not simply emerge out of the forests and onto the open savanna in one fell swoop (indeed, they could not have done so, for classic treeless Serengeti-type savannas were still very far in the future). Rather, they embarked on a long period of exploration of the possibilities offered by the new and expanding forest-edge and woodland habitats. The fossils of other mammals found along with those of the early hominids seem to confirm this preference for woodland environments, which have their own distinctive animal communities, although archaic hominid fossils have been found in contexts that indicate both relatively dense forest and quite open conditions. Possibly it was the exploration of varied habitats that was responsible for the apparent diversity of the earliest hominids.

The oldest hominid that we know for sure walked upright, at least when on the ground, is Australopithecus anamensis, a species known from a small sample of fossils from the sites of Kanapoi and Allia Bay in northern Kenya. Nearly all of these fossils date from between 4.2 and 3.9 million years ago, and one of them consists of pieces of tibia (the lower-leg bone) that show clear signs of upright posture. When apes amble along on all fours, their legs go straight down to the ground from their hip joints, rather like table legs do. This is fine while the apes are supporting their weight on four limbs, but is a bit of a handicap when they try to walk on two legs because they have to swivel the outer leg around their center of gravity to take each step forward, swaying the body sideways in the process.

A highly speculative phylogenetic tree of the family Hominidae, containing most of the fossil hominid species recognized by recent scholars.

A highly speculative phylogenetic tree of the family Hominidae, containing most of the fossil hominid species recognized by recent scholars. Dotted lines represent possible pathways of ancestry and descent, while solid lines connect the oldest and youngest current records of each species. Time runs along the vertical axis; horizontal arrangement is arbitrary. © Ian Tattersall.

In contrast, an upright biped like ourselves has upper legs that slant inward toward the knee from the hip joints. In this way, with each stride the weight of the body is transmitted straight forward as the feet move close by each other, with no awkward sideways movement. Part of the apparatus needed to accomplish this lies in the knee joint, the surface of which is oriented at a right angle to the shaft of the tibia rather than with a sideways cant as in the apes. In the A. anamensis tibia, the part that contributes to the knee joint has the same orientation as in a human tibia, a pretty firm indication of upright posture. And there are equivalent indications in the ankle joint.

In those fossil bits that are known, A. anamensis is fairly comparable to A. afarensis, the best-known of all of the several early bipedal hominid species allocated to the genus. The most famous fossil representing the latter species, and probably the most famous hominid fossil of all time, is ‘‘Lucy,’’ the partial, but nonetheless unusually complete, skeleton of a tiny (and thus presumed female) individual who lived 3.18 million years ago. Discovered in the mid-1970s at Hadar in Ethiopia, Lucy is one of many fossils thought to belong to this species that have been found at sites as far from Ethiopia as Tanzania and possibly Chad, and that date from about 4 to 3 million years ago. Among these other fossils are two fairly complete skulls from 3-million-year-old deposits at Hadar, as well as postcranial bones that nicely complement what we know from Lucy herself. One remarkable find, from a 3.4-million-year-old stratum also at Hadar, is the ‘‘First Family,’’ the fragmentary remains of as many as 13 individuals who may have died together in a natural catastrophe such as a flash flood.

From the resulting aggregate of fossils, we have a pretty good picture of what A. afarensis looked like and a lot of information on which to base guesses about the way in which these creatures moved around (which doesn’t, of course, mean that all paleoanthropologists are in agreement on the matter!). The size range among the bones of mature A. afarensis is particularly striking and implies that males were a great deal larger than females. Lucy herself probably stood little more than three feet tall, whereas males may have been a foot taller. Estimates of body weight vary; males may have weighed up to about 100 pounds, and females may not have exceeded 60 pounds.

The first thing you might notice about the skeleton of A. afarensis is its wide, shallow pelvis, which at first glance seems to be proportioned rather like our own. It certainly contrasts dramatically with the long, narrow pelvis of the quadrupedal apes. The pelvis of A. afarensis is not that of a quadruped that wore its innards slung as in a hammock beneath the spine. Instead, these organs were supported from below by the bony bowl of the pelvis (though not as effectively as in Homo sapiens). The broad, shallow pelvis therefore bespeaks an upright posture, though it doesn’t tell us much about whether that posture was adopted mainly in the trees or on the ground.

In terms of moving around, the ape pelvis has a form that gives the thigh muscles their greatest mechanical advantage when the hip is flexed. In contrast, the human hip is arranged so that speed and the range of available movement are emphasized, particularly when the leg is extended straight out. The A. afarensis pelvis lies clearly on the human side of this divide, but it is not identical to our own. The ball-and-socket hip joint, for example, has a rather small surface area, which concentrates (rather than diffusing, as in humans) the force generated when the foot hits the surface being walked on. And the pelvis itself is remarkably wide and flaring, with numerous anatomical details that are not matched in any living form. Few would disagree that the A. afarensis pelvis shows a radical reorganization in the direction of uprightness when contrasted to the presumably more ancestral condition of apes, but its combination of features leaves much room for debate on exactly how the species moved around.

The hip joint of A. afarensis may leave questions unanswered, but the knee joint is more conclusive. The knee joint of Lucy and her kin was clearly that of an upright biped, whose thighs converged from the hips to the knee, just like ours and those of A. anamensis. This can be seen most dramatically in the distinct angle formed between the horizontal knee-joint surface and the inwardly angled axis of the femur shaft. The tibia went straight down from the knees to the feet, which would have passed close together when walking. Overall, though, the legs were shorter than ours relative to body size, and the bones of the feet in these archaic hominids do not tell a simple story. The rear of the foot is relatively short like ours, and it has features in common with later humans that indicate a restricted ability to move beyond the fore-and-aft plane. In front of the ankle, in contrast, the foot was longer than ours, especially in its frontmost part, where the bones of the toes can be described as particularly apelike.

How about the rest of the body?

The arm bones of A. afarensis show both apelike and humanlike characteristics, and the arms themselves are longer than ours in comparison with the legs, though most of this disparity seems to be due to the shortness of the legs. The shoulders are narrow, and the rib cage is very unlike ours.

Instead of being essentially cylindrical in shape when seen from the front, it tapers dramatically outward from top to bottom, as ape rib cages do. Viewed from the top, though, it is shallow from front to back as ours is, rather than being deep like that of a quadruped. The spine itself is composed of vertebrae with long projections for muscle attachment, indicating a relatively massive musculature. The muscles in this area of the body are important in locomotion among both quadrupeds and bipeds, though, so this doesn't help us much in determining posture.

However, a telling indicator lies in the weight-bearing central parts of the back vertebrae. In A. afarensis, these are small relative to ours (and to those of apes); but in one related species, at least, the vertabrae show evidence that the spine (in side view) had the double curve that is another characteristic of our upright posture.

So what do all these conflicting indicators add up to in telling us how A. afarensis got around? There has been a great deal of debate on this subject, with some paleoanthropologists emphasizing the evident specializations for bipedality that can be seen widely through the skeleton, and others placing more importance on the features retained from a tree-living past. However, some consensus seems to be emerging between the extremes. Researchers have reported that, particularly in relatively open environments, chimpanzees tend to hold their torsos upright while foraging in trees, and many think that hominids evolved from species that did the same thing with even greater frequency. On the ground the primarily quadrupedal chimpanzees fold their hands so as to bear the weight of their upper body on the outsides of their knuckles and have thereby been able to retain the long hands that are so useful to them in grasping tree branches. But, predisposed as they almost certainly were to holding their bodies upright anyway, the ancestral hominids took a different tack as the African forests began to fragment, walking upright on two legs as they moved across the ground. 

The skeleton of ‘‘Lucy’’ (from 3.18 million years ago), who was only about three feet tall. Courtesy American Museum of Natural History.The skeleton of ‘‘Lucy’’ (from 3.18 million years ago), who was only about three feet tall. Courtesy American Museum of Natural History.

This history resulted in animals that were not as agile in the trees as apes or as efficient on the ground as we are. Nonetheless, the have-your-cake-and-eat-it-too adaptation exemplified by A. afarensis evidently served this species and its relatives well, for it endured as a stable anatomical complex for several million years. Clearly, these early hominids were quite comfortable in the expanding forest fringe areas that offered the resources of both the deep forest and the more open woodlands. Occasionally they evidently ventured entirely into the open, as shown by the astonishingly preserved 3.5-million-year-old bipedal trackways of Laetoli, in Tanzania.

One intriguing suggestion is that, during these early times, hominids got their start as omnivores by using their arboreal skills to steal the antelope carcasses that leopards—denizens of the woodland and savanna— regularly stashed in trees precisely so that they would not be stolen while their owners were away roaming over the landscape. Chimpanzees are known to hunt monkeys and small antelopes, so there is no reason to suspect that the very earliest hominids would have been unfamiliar with the advantages of a high-protein diet.

Accordingly, right from the earliest days of their discovery, our ancient ancestors were interpreted as hunters, with an intrinsic propensity for violence. After all, human beings have historically been very successful hunters, and even chimpanzees hunt occasionally; so shouldn’t the early ‘‘bipedal apes’’ have been hunters as well? Not necessarily. In the last half million years or so of human evolution, hunting has undoubtedly been critically important to the hominid way of life; but before that, the picture is much less easy to interpret. Early authors

suggested that ancient hominid fossils and the animal bones found with them were the remains of the hunters and their victims, respectively. But in the 1980s, the paleontologist Bob Brain pointed out that the whole assemblage looked like the remains of leopard and hyena prey. Indeed, Brain found one australopith skull bearing puncture marks that were almost certainly made by the canine teeth of a leopard. And in their recent book Man the Hunted, the anthropologists Donna Hart and Bob Sussman have argued that being prey species shaped the early hominids far more than the occasional hunting of a hare would ever have done.

Hart and Sussman point out that early hominids, coming to the ground as their formerly forest habitat fragmented, were ecologically edge species, flourishing in those areas where the forest gave way to woodland and grassland. And today’s most successful edge primates are not the apes but the macaque monkeys of Asia, adaptable generalists who live in large groups that usually split up into smaller subgroups for foraging. They are behaviorally flexible and omnivorous, and they tend to return to home bases each night. They are also subject to quite high levels of predation, which has a major influence on their group organization and movements.

While they are closer human relatives than macaques are, today’s apes are very differently adapted from early hominids, and Hart and Sussman conclude that ecologically the macaque analogy may be a better one. So they propose that early hominids may have lived in multi-male, multi-female groups of variable size that split up during the day’s activities, but re-formed at night at well-protected home bases, sleeping on cliffs and in the trees, a preference that fits well with their anatomies. The early hominids would have been omnivorous, eating fruit, herbs, roots, and the occasional insect or lizard. As in macaques, females formed the social core of the group, which was always vulnerable to predators. Males, who are reproductively more expendable, served as sentinels, and indeed it may have been the threat of predation in their new habitat that formed many of the behaviors of our small and relatively defenseless early ancestors. This is additional reason to believe that, while they may have preferred to move on their hind limbs over the ground, the early hominids had not emancipated themselves entirely from the trees. Indeed, it is very likely that at night these small-bodied and largely defenseless animals regularly took shelter in the relative safety of trees, cliffs, and other places accessible only to climbers.

The perennial question of ‘‘why bipedality?’’ has most frequently been posed in immediate functional terms, rather than in terms of the structure of the ancestral form from which the first hominid bipeds were descended. Paleoanthropologists have regularly tried to identify the “advantage” that assured the eventual triumph of bipedal hominoids in non-forest environments. It has been suggested, for example, that the key factor was the freeing of the hands that bipedalism allows. Once your hands are not committed to supporting your body weight, they are available to be modified and used for other purposes, such as carrying or manipulating objects. Similarly, it has been pointed out that by standing up you can see potential dangers at a greater distance. Or maybe bipedal locomotion was simply more efficient than quadrupedalism over open ground.

Some years ago the paleoanthropologist Owen Lovejoy caused quite a stir by suggesting that the success of the early bipeds was due to a reorganization of reproductive activity that increased the rate of production of offspring. Lovejoy pointed out that modern humans are unique among hominoids in two important ways. First, males have no way of knowing when females are ovulating (and thus ready to reproduce); and second, particular males and females tend to become long-term reproductive pairs. These traits, he thought, had roots deep in the hominid past. From the beginning, bipedalism freed the hands of females to carry extra babies around. However, the consequently limited mobility of the females required them to bond with males who would then use their freed hands to bring them food they had obtained. Of course, the only way for males to be certain that the infants they fed were their own was to develop pair bonds with certain females. And from the female point of view, constancy of male interest could be ensured only by the development of highly visible secondary sexual characteristics, such as prominent breasts, which serve as constant attractants, replacing the cyclical swelling around the genitalia that had previously served to attract males by advertising ovulation.

The key to the success of this strategy, Lovejoy believes, is that the energy saved by non-foraging females could be invested in extra reproductive effort. This hypothesis emphasizes bipedality as an adaptation for increasing reproductive fitness rather than as an efficient means of getting around or shedding heat, and it neatly links our peculiarities of locomotion, reproduction, and social organization. However, it has been convincingly contested on a whole host of grounds, among them that the great disparity in body size between males and females of Australopithecus afarensis is typical of polygynous hominoids (among whom males constantly compete for females) and is the reverse of what is seen in the only other pair-bonding modern hominoid, the gibbon. The reproductive-advantage idea is a good story, but it reminds us that we should always be wary of stories that do not fit all the facts. Nevertheless, even though we cannot observe long-extinct hominids in action, it would be unwise for us to forget that their behaviors must have been critical ingredients of their successes and failures.

One particularly intriguing suggestion about the reasons for early bipedality involves the regulation of body and brain temperature in treeless, unshaded environments. In the tropics a major problem once you move away from the forest is the heat load imposed by the strong sun overhead. Shedding this heat is important, particularly for the brain, which can be damaged quickly by overheating. If you stand up, you minimize the heat-absorbing surface area that you present to the sun, even as you maximize the area of your body available to lose heat by radiation and by the evaporation of sweat. And the taller you are, the more you can benefit from the breezes that blow above the level of the surrounding vegetation. In sum, there are plenty of potential benefits from an upright posture on the ground. As to the most important of them, take your pick. But the critical thing to remember is that once you have stood upright, all of these potential benefits—and all potential liabilities, too—are yours. The crucial factor is standing up in the first place. And for a newly terrestrial hominoid, the most significant element here was almost certainly having had an ancestor that already favored holding its body upright.

Bipedal though they might have been on the ground, though, these early hominids would hardly have qualified for the epithet ‘‘human.’’ In particular, their skulls were still effectively those of apes, housing apesized brains in tiny braincases in front of which large faces projected aggressively. This conformation is quite the opposite of that of later hominids, in which we see ever smaller faces that eventually became tucked beneath the fronts of larger, rounder braincases. The long faces of apes have a lot to do with the long tooth rows contained in the upper and lower jaws. Modern apes have quite wide incisor teeth at the front of the mouth, flanked by substantial, pointed canine teeth that project far beyond the level of the other teeth in each tooth row.

This is true of both sexes, but in apes the canines of males are relatively much larger than those of their female counterparts, even in relation to their larger bodies. In animals with large canines there is a gap (known as a diastema) between the side incisor and the canine in the upper jaw. This allows the jaws to close fully, as the lower canines fit into the gaps. Continuing along the tooth row toward the rear, we can see additional distinctions between apes and humans. The lower first premolar of an ape has a single point (cusp); in humans, in contrast, this tooth often has two cusps, which is why dentists commonly refer to our premolars as ‘‘bicuspids.’’ The three molar teeth behind are relatively elongated in apes, yielding long, parallel-sided tooth rows, quite different from the short, rounded rows of teeth seen in Homo sapiens.

Contrasting shapes in the pelvises of a chimpanzee (left), Australopithecus afarensis (center), and a modern human (right) show us that on the ground Australopithecus was a biped.

Contrasting shapes in the pelvises of a chimpanzee (left), Australopithecus afarensis (center), and a modern human (right) show us that on the ground Australopithecus was a biped. While different in many details from that of Homo sapiens (right), the Australopithecus pelvis is broad and flaring like that of the human, and it contrasts strongly with the long, narrow pelvis of the quadrupedal ape. Courtesy Peter Schmid.

Like its body structure, the dentition of A. afarensis shows a mixture of similarities to both apes and humans. Presumably, the ape resemblances of A. afarensis represent retentions from an ancestral condition that was common to both forms. In particular, the teeth of A. afarensis were large, except for the canine. Nonetheless, this tooth still projected somewhat beyond its neighboring teeth, required a small diastema in the upper jaw, and had some of the pointy shape of an ape canine. In addition the enamel covering the teeth was thick, a characteristic of most early hominids, though not of Homo sapiens. This is a feature thought to reflect a dietary shift away from soft fruits and toward tougher foods such as tubers.

Despite certain humanlike features, though, many paleoanthropologists like to refer to early hominids such as A. afarensis as ‘‘bipedal apes.’’ There is plenty of justification for this in terms of the behavioral capacities we may infer for them, for the making of stone tools was still far in the future when A. afarensis frequented the African forest edges and woodlands. And there is very little reason to suppose that this species and its like represented any significant cognitive refinement over what we see in the apes today. It’s important, though, not to underes timate the mental qualities of the apes—and by extension, of the early hominids. Apes show remarkable, if limited, powers of intuitive reasoning, as well as a striking ability to communicate their emotional states and to understand the motivations of other individuals. They even develop local ‘‘cultural’’ traditions involving the transmission from one generation to the next of learned behaviors such as cracking nuts on stone anvils and ‘‘fishing’’ with sticks in termite mounds. Indeed, many primatologists think that the capacity for culture in this restricted sense is a basic great-ape trait, and if so, we have even greater reason to believe that the apes can give us a general picture of the apparently quite impressive intellectual starting point of our own lineage.

But whether or not this turns out to be the case, it is still important not to view early hominids simply as junior-league versions of ourselves: implicitly, creatures striving to become us. Equally clearly, these ancient relatives did business in their own unique ways and weren’t apes, either. But one of the ways in which A. afarensis and species like it seem to have been significantly closer to apes than to us was the speed with which they developed from infancy to maturity. Young apes grow up much more quickly than young humans do; a male chimpanzee is reproductively mature at about six to seven years of age, for example, whereas a male human takes twice as long, or longer. This prolonged maturation process—which, it is important to note, extends the period of social learning—expresses itself among other things in the rate at which the permanent teeth erupt. It has been shown that the earliest hominids matured quite rapidly, at rates probably comparable with those of apes. A relatively rapid developmental process may, indeed, have characterized hominids until quite a late stage in their evolution.

Australopithecus afarensis, though a good example of its group, is only the best known of several species that were traditionally classified in the subfamily Australopithecinae of the family Hominidae. This subfamily is nowadays implicitly taken to include all of the extinct hominids, with the exception of those allocated to the genus Homo—which raises problems of definition that have yet to be adequately addressed. There is also, inevitably, some argument as to whether this group deserves the status of subfamily; there is, after all, debate even over the level at which Hominidae itself should be recognized. Most scientists thus currently prefer to use the more informal term ‘‘australopiths’’ for this group, and we’ll do so here.

The australopiths have been known since 1924, when the first such specimen, described under the name of Australopithecus africanus, was found in a lime quarry in South Africa. This specimen consisted of the skull of a very young individual, which immediately introduced problems because young apes and humans resemble each other in skull proportions much more than adults do. What’s more, even as an adult this child would have possessed a rather small brain, and at the time paleoanthropology still remained largely under the sway of the large-brained but fraudulent Piltdown specimen. It would be another quarter century before it became generally accepted that the most ancient hominids had not been distinguished from other primates by the big brain we so prize in ourselves today.

Numerous finds in the 1940s and subsequently, however, have demonstrated that the South African australopiths were no mere localized curiosity. Indeed, in the period between about 4 and 1 million years ago at least eight australopith species, all African, are now routinely recognized in the genera Australopithecus and Paranthropus (though sometimes the genus Australopithecus is used to include both). In the welter of new species the long-standing distinction made between the so-called robust australopiths, with relatively heavily built skulls, and the more lightly built graciles is gradually yielding to a recognition that a much more complex branching pattern of descent probably characterized the australopiths during their long tenure on Earth.

There is as yet no consensus view of the relationships among these early hominids. But at the moment many are happy to look upon the 4-million-year-old A. anamensis as a ‘‘stem’’ species, which most likely gave rise fairly directly to our old friend A. afarensis, known from between about 4 and 3 million years ago. An approximately 3.5-million-year-old fragment of lower jaw from Chad has been called A. bahrelghazali, but many scholars consider this to be a central-western African version of A. afarensis. If the distinction between gracile and robust forms is an accurate one, it was shortly before 3 million years ago that the gulf began to develop. Australopithecus africanus is the classic example of the gracile forms and is found in sites in central southern Africa that are hard to date but that are believed to fall in the period between a little more than 3 million and a little less than 2 million years ago.

In contrast to the Homo sapiens, or modern human skull (left), with its balloon-like braincase and tiny face, both the chimpanzee (right) and the Australopithecus (center) skulls exhibit small braincases and large, protruding faces.

In contrast to the Homo sapiens, or modern human skull (left), with its balloon-like braincase and tiny face, both the chimpanzee (right) and the Australopithecus (center) skulls exhibit small braincases and large, protruding faces. Photo by K. Mowbray, AMNH.

A very recent find of an as-yet incompletely excavated skeleton from very early levels at Sterkfontein, the classic A. africanus discovery site, is at least 3.3 million years old and most likely represents a distinct species antecedent to A. africanus. From within the time span of A. africanus comes the Ethiopian species Australopithecus garhi, named in 1999 from a handful of fossils that included an upper jaw with rather large chewing teeth. These fossils mystified their discoverers to such an extent that they left open the question of whether their new species might anticipate Paranthropus or Homo, or whether it might even be a late version of A. afarensis, which seems the most plausible option.

The robust forms are typified by Paranthropus robustus, a species from South African sites probably dating to between about 2 and 1.5 million years ago, and by the so-called hyperrobust Paranthropus boisei from sites in eastern Africa dating from 2.2 to 1.4 million years ago. All australopiths have large chewing teeth, but those of the robusts are truly massive, with premolars of molar-like proportions. In contrast, there is significant diminution of the incisor and canine teeth, which are tiny. The huge molars rapidly wear flat and are implanted in massive jaws. Most scientists see in these fossils evidence that a group of australopiths departed from the omnivorous ancestral condition and embarked on a lifestyle that involved processing large quantities of tough vegetal foodstuffs or perhaps even invertebrates. The massive chewing apparatus needed to accomplish this dietary shift is accompanied, among other things, by the presence of sagittal cresting, whereby the rear centerline of the braincase is marked by a thin vertical ridge of bone. The robust lineage can be traced back to at least 2.5 million years ago, when the species Paranthropus aethiopicus showed up in eastern Africa, and some scientists have even argued that A. afarensis shows features foreshadowing the robusts. Unlike the later and evidently more specialized robusts, which had quite flat faces, the earlier P. aethiopicus possessed a rather projecting snout and fairly substantial front teeth.

Overall, then, the australopiths were a diverse group indeed. With the exception of the highly specialized later robusts, most of them probably had fairly varied diets, eating pretty much whatever food they could lay hands on, although microscopic examination of the teeth reveals wear surfaces textured rather like those of frugivores or omnivores, and one study of bone chemistry suggests that A. africanus was already consuming substantial quantities of meat. Hunting in itself would probably have been nothing new for a hominoid—some chimpanzees hunt from time to time, sometimes quite frequently. These remote precursors of humans probably scavenged most of their animal protein, however, and it is highly unlikely that they ever pursued anything larger than small prey. With the possible exception of the robusts they all probably had broadly similar lifeways. But it is hard to avoid the impression that these various different types of australopiths were busily exploring the options offered by the range of new habitats made available by the climatic changes affecting their continent. We can thus look upon the multiplicity of australopith species as the outcome of a set of evolutionary experiments that was made by a special kind of hominoid learning to cope with new habitats. And it was out of this process of experimentation that the ancestors of our own genus, Homo, somehow emerged.

If you find an error please notify us in the comments. Thank you!