By 5 Ma, the hominid populations had split into those that began to walk erect and those that continued to be knuckle-walkers. The adaptation to bipedal walking appears to have occurred over a long period of time, because the earliest hominin fossils with preserved wrist joints, from Ethiopia and Kenya, indicate they retained knuckle-walking adaptations (Richmond & Strait, 2000), like gorillas and chimpanzees. Because gorillas and chimpanzees are both knuckle-walkers, but chimpanzees are more closely related to humans, parsimony dictates that the preferred hypothesis be that knuckle-walking was the locomotion of the last common ancestor of humans and chimpanzees. Thus the presence of leftover knuckle-walking features in the earliest hominins was predictable. Whether this implies that they still, on occasion, actually walked quadrupedally on their knuckles — or that these features are just relics of a knuckle-walking past, not yet lost by the new bipeds — is debatable.
In 1993, a joint American, Japanese, and Ethiopian paleoanthropological team working in Middle Awash, Ethiopia, discovered what they claimed to be the oldest fossil hominin remains thus far known (T.D. White et al., 1994). The species occupied the African landscape around 4.4 Ma and in terms of time are very close to the split between the chimpanzee and earliest hominins, as calculated from molecular clocks (WoldeGabriel et al., 1994; T.D. White, 2002). Though first classified as Australopithecus ramidus, these fossils have now been allocated to their own genus, Ardipithecus, and very recently an even more primitive form has been described from deposits 5 Ma that is supposed to be a subspecies of the same species, A. ramidus kadabba (Haile-Selassie, 2001; T.D. White, 2002). The teeth of Ardipithecus are generally similar to those of the chimpanzee (including thin molar enamel), but the canines are shorter, though still sexually dimorphic. In the features of the rather fragmentary cranium, however, it resembles later australopithecines; its cranial base is said to be short, implying that the head is balanced "on top" of the spinal cord and needs less muscle support and that it walked upright. We must admit, however, that this is difficult to judge because no detailed illustrations of the basicranium have yet been published. Whether Ardipithecus really shares any derived features with later hominins is difficult to tell. If the description is accurate, then it certainly does; but without detailed figures and descriptions of these important specimens, nothing more can be said.
With all the significance that has been attached to bipedalism, a word of caution is in order. 9-7 million years ago, on the Tyrrhenian Island (as it then was) in Italy, lived a small hominoid of uncertain but definitely non-hominin affinities: Oreopithecus bambolii. Many aspects of its anatomy, not least its bizarre dentition, place it outside the range of modern hominids, and whether it can be viewed as a hominid at all has been disputed. But of one thing there is no doubt — the structure (including the microstructure of the cancellous network) of the pelvis shows that it was upright and bipedal (Rook et al., 1999). Bipedalism can and did evolve more than once, and with some, at least, of the same osteological modifications. We need a total morphological pattern, not only bipedalism, to identify a hominin.
The current archeological and paleontological evidence suggest that Ardipithecus preferred a forest habitat; the fossils are found with a typical forest fauna (WoldeGabriel et al., 1994; Andrews & Humphrey, 1999; Denys, 1999). This is reinforced by its implied dietary adaptations. For example, the thin molar enamel and small molar teeth suggest that its diet was based more on leaves, possibly soft fruits, and other soft vegetative material. If Ardipithecus did walk bipedally, then it was doing so before the climate shift (around 2.5 Ma) that led later hominins to occupy a more open habitat. Thus early bipedalism evolved in the forests, suggesting that it was not an adaptation for occupying the savanna (Andrews & Humphrey, 1999; Feibel, 1999; Foley, 1999).
The early development of bipedalism, then, seems not to have been an adaptation to life in the savanna as such but was perhaps associated with the need to move from one forest patch to another (Rodman & McHenry, 1980; Cameron, 1993a; Cameron & Groves, 1993). While bipedalism is an inefficient system for short-term sprinting, it is very efficient for longdistance travel, though its earliest form, as observed in the australopithecines, is said to be a "cheap way of moving" (Taylor & Rowntree, 1970; Fleagle, 1999). Walking upright will also mean that the body is a smaller target for the sun, and thus the body will remain cooler than if walking on all fours, where much more of the body is exposed to the sun (Wheeler, 1984, 1991, 1993; Foley, 1987; Aiello & Dean, 1990). Richard Leakey has suggested, in lectures, that in open dry forests, a large-bodied ape is forced to be terrestrial yet may still depend on an arboreal diet, such as seed pods. To reach them, it is forced to rear up. This explanation for bipedalism strikes us as a most perspicacious piece of lateral thinking. Nor can we exclude the Aquatic Ape Hypothesis (AAH). Elaine Morgan has long argued that many aspects of human anatomy are best explained as a legacy of a semiaquatic phase in the proto-human trajectory, and this includes upright posture to cope with increased water depth as our ancestors foraged farther and further from the lake or seashore. At first, this idea was simply ignored as grotesque, and perhaps as unworthy of discussion because proposed by an amateur. But Morgan's latest arguments have reached a sophistication that simply demands to be taken seriously (Morgan, 1990, 1997). And although the authors shy away from more speculative reconstructions in favor of phylogenetic scenarios, we insist that the AAH take its place in the battery of possible functional scenarios for hominin divergence.
Almost 300,000 years after the disappearance of Ardipithecus from the fossil record, the first of what we call the australopithecines appears, Australopithecus anamensis. This species first appears in deposits at Kanapoi, south of Lake Turkana, Kenya, dating to around 4.2 Ma and disappears from the fossil record around 3.8 Ma (M.G. Leakey et al., 1995; C.V. Ward et al., 2001). Certain anatomical features of the jaw joint, as well as the increase in molar size and enamel thickness from that observed in Ardipithecus, suggest that it employed a very different pattern of food processing and/or food types. Although the little we know of the skull of this species indicates it is primitive, with rather large canines and narrow, rectangular jaws, the leg bone fragments, which consist of parts of the knee and ankle joints, are more similar to members of our own lineage, Homo. Indeed, if only the leg bones had been discovered, it is likely that they would have been allocated to a species of Homo. And it is significant that as long ago as 1967, the lower end of a humerus, part of the elbow joint, was discovered at Kanapoi, and for a long time (until excavations at Kanapoi were reopened in the 1990s) it remained rather a mystery — a "curiously modern-looking" elbow joint older than the more primitive australopithecines (see Senut & Tardieu, 1985; M.G. Leakey et al., 1995).
Doubts still remain. Andrews (1995) has suggested that the postcranial bones and the skull parts of this hominid are from two distinct species: that the skull and teeth remains, which were found in geological deposits dating a little earlier than the recovered leg bones, belong to an extinct great ape, while only the leg bones, he argues, should be allocated to this new species. There is little in the preserved anatomy of this species that suggests a close relationship to later hominins. Certainly its cranial morphology is primitive, and even if the leg bones do belong to the same species, we know, or think we know (from Ardipithecus and probably Orrorin), that bipedal locomotion had already had a long history, of a million years or more. And let us always bear Oreopithecus in mind.
As with Ardipithecus, A. anamensis appears to have occupied a gallery forest, although other areas from this time, which also contained specimens of this species, are known to have been more open, wooded or largely bushland (M.G. Leakey et al., 1995; Andrews & Humphrey, 1999).
Perhaps the best-known Pliocene hominid is the world famous "Lucy" skeleton found by Johanson's group in the early 1970s, dating to between 3.6 and 2.9 Ma (Johanson & Taieb, 1976; Kimbel et al., 1994; T.D. White, 2002). "Lucy" and her kind were originally named Australopithecus afarensis (Johanson et al., 1978), but Strait and Grine (2001) have allocated them to the genus Praeanthropus (see also Strait et al., 1997) (Figure 3.2), originally erected by Senyurek in 1953 for the Garusi maxilla, which is now catalogued as Laetoli Hominid 1. Although Senyurek called LH1 Praeanthropus africanus, the International Commission on Zoological Nomenclature has recently suppressed this usage of the name africanus and placed afarensis on the Official List of Names in Zoology. We provisionally accept this revision, but will return later to the whole question of taxonomy.
Lucy was just over a meter tall (about 3M feet), with long arms, a short trunk, and short legs. Taking her skeleton together with the more fragmentary remains of other individuals, we can say that the foot bones of Pr. afarensis are generally human-like, but the ankle joints are chimpanzeelike in their overall flexibility (see Richmond & Strait, 2000). This has caused great controversy. To some experts, it suggests that tree climbing was still an important ability; and those who argue for tree-climbing point out that the toe bones are curved and strong, the ribcage is funnel-shaped and apelike rather than barrel-shaped and human-like, and the arms are strongly muscled (Susman, 1979; Stern & Susman, 1983; Susman et al., 1984; Heinrich et al., 1993). To others, however, these features are simply primitive traits left over from a more arboreal ancestry. The shape of the pelvis of Pr. afarensis indicates very clearly a bipedal gait, and the femur is valgus (meaning that it is orientated inward at the knee, which is typical of humans, whereas in apes it is more vertical); in particular, the big toe is not markedly — or at all — divergent, as one would expect if it were doing much climbing (Lovejoy, 1974, 1981; T.D. White, 1980; T.D. White et al, 1983; see also partly Robinson, 1972). Footprints of Pr. afarensis were discovered at Laetoli during the mid-1970s. Although their interpretation is still very controversial, the one thing that all agree upon is that they show a big toe aligned with the other toes, with only a slight gap (M.D. Leakey & Hay, 1979; Day & Wickens, 1980; T.D. White, 1980). A third school of thought accepts that Pr. afarensis was basically bipedal but finds it difficult to overlook altogether the survival of these apelike, presumably arboreal, features (B.A. Wood, 1992). Wood suggests that Pr. afarensis was neither predominantly arboreal nor fully bipedal (B.A. Wood, 1992). And then we have to come to terms with the recent revelation that studies of the wrist bones also suggest that some form of knuckle-walking may have been involved (Richmond & Strait, 2000) (or were the knuckle-walking adaptations, too, a mere survival from a primitive ancestor?).
One very obvious feature of Pr. afarensis is the large difference in body size between males and females. Males average in height 4 feet, 10 inches and in weight 143 lb, while females average in height only 3 feet, 3 inches while only weighing 66 lb (Stringer & McKie, 1996). Thus females average only 80% of male height and only 46% of male weight. This degree of sexual dimorphism, while common in the great apes (orangutans and gorillas but not chimpanzees) and many monkey species, is unusual for hominins. So marked is this size dimorphism that one of us (CPG) had previously been reluctant to accept such an interpretation, preferring to believe in the coexistence of a small and a large species (Groves, 1989a), but now tends, still not without some misgivings, to accept the homogeneity of the species.
There is no evidence of tool making by Pr. afarensis, and its brain was little or no bigger than that of a chimpanzee. Unlike Ardipithecus, but like members of A. anamensis, the molar teeth in Pr. afarensis were large in size, and the molar enamel was thick, suggesting that it was eating tougher food types, including hard fruits and possibly nuts.
Around 3.3 Ma some populations of Pr. afarensis in east Africa were living near a large lake, surrounded by forests and bushland. Over the next 500,000 years, the region fluctuated in climate and habitat types, the lake receded and then expanded, and the fauna shifted from a more closed forest type to a more open grassland type and then back again to a closed forest type. When Pr. afarensis disappeared from the fossil record, around 3.0 Ma, savanna dominated the region. By 2.8 Ma, however, the region had reverted to the bushy, forested conditions that held sway almost
500,000 years earlier (Potts, 1996; see also Andrews & Humphrey, 1999; Feibel, 1999).
Contemporary with Pr. afarensis in Ethiopia is the recently discovered australopithecine from Chad, some 2,500 km west of the Rift Valley (Brunet et al., 1995, 1996). The specimens were originally thought to represent a western population of Pr. afarensis, but in 1996 they were allocated to a new species of Australopithecus, A. bahrelghazali. Very little is known of this species at present, and whether its ascription to Australopithecus is justified can only be answered with the discovery of more complete material. Like A. anamensis and Pr. afarensis, it has thick molar enamel, but its mandible is of a lighter construction and its premolar cusps are less developed than in Pr. afarensis (Brunet et al., 1995). The deposits containing A. bahrelghazali reflect a lakeside environment with both perennial and permanent streams, and a vegetation mosaic of gallery forests and wooded savanna with open grassy patches (Brunet et al., 1995), rather like that occupied by earlier and contemporary proto-australopithecines of east Africa.
The most recent Pliocene "australopithecine" species discovered, Australopithecus garhi, appears in the fossil record of the Middle Awash of Ethiopia around 2.5 mya (Asfaw et al., 1999). Its discoverers suggested that this is the likely direct ancestor to Homo. In overall anatomical features (Figure 3.3) it is similar to Pr. afarensis, including a similar cranial capacity, around 450 cc, its projecting facial profile, and, if the limb bones are correctly associated, its body proportions with typically short legs and a short trunk but relatively long arms. Yet it has some features that are said to be significantly different from those of the earlier australopithecines. These include the forward positioning of its cheekbone, a more oval premolar shape, and huge premolars and molars. Actually, it shares no unique characters with Homo, and we suspect that it represents either a species of Praeanthropus (Pr. garhi) or perhaps a new genus. It certainly cannot be considered a species of Australopithecus (see Strait & Grine, 2001; Cameron, in press b). It has been suggested that it made stone tools (Heinzelin et al., 1999): Animal bones with butchering marks (cut marks) as well as some primitive stone tools have been found close by, though whether these were actually the work of members of the "garhi group"
remains problematic, for early representatives of Homo are also found in this region dating to around the same time or just a little later (Kimbel etal, 1997).
Was this article helpful?