The presence of these large molar teeth means that the tooth-rows of A. afarensis are long, as those of apes are. But largely because the canine teeth are small, the dental arcades are slightly curved in outline and lack the aggressively parallel arrangement exhibited by the apes. The dentition of A. afarensis is thus rather like the cranium itself in possessing features reminiscent both of apes and of later hominids.
A recent sophisticated study of the wear produced by chewing on some A. afarensis molar teeth suggested that, while members of this species probably preferentially sought out soft fleshy fruits to eat when obtainable, where such foods were unavailable they would have gone after tough, brittle foodstuffs like nuts, seeds, roots, and the underground runners of grasses. This would have made them considerably more omnivorous than today’s apes are, and it would be consistent with the typically quite heavy wear that we see overall on A. afarensis teeth. A diet of this kind suggests a pretty generalist adaptation to a habitat that spanned the spectrum from closed forest to open woodlands.
The Hadar desert badlands have produced an incredible quantity of hominid fossils over the years, and many localities there have yielded fossils of Australopithecus afarensis. Undoubtedly the most extraordinary of these localities is a spot known as AL 333 at which, in 1975 and later, researchers unearthed a trove of some 240 fossils representing the remains of 17 hominid individuals—and, most unusually, fossils of very little else. How these bones came to be buried where they are is a puzzle. They are all broken up, which is consistent with their having been transported from somewhere else by water. But why were they concentrated in one place? They weren’t accumulated by a scavenging agent such as hyenas (which are famous for transporting hominid cadavers to their dens) because, although broken, they show no signs of gnawing—and a sluggish river channel is, in any case, hardly the spot you’d expect to find a hyena den. So there is a bit of a mystery here; and it’s important that it be solved eventually, because despite the fact that the fine-grained sediments in which the bones were enclosed are typical of those laid down in a slow-moving river, one suggestion is that these are the remains of an entire unfortunate social group that was swept up in a single catastrophic event—maybe a flash-flood—that happened at some time between 3.18 and 3.22 million years ago. And if all of the individuals—nine adults, three adolescents, and five juveniles—actually belonged to one social unit, then all of them must have belonged to the same species.
This is not otherwise a sure bet because, although all the comparable fossil parts from Site 333 look basically the same, the size range among them is huge. Still, despite all the uncertainty surrounding how the fossils came to be jumbled together in one spot, the current majority opinion is still that all of the Hadar hominids—including Lucy, who is as small as the smallest of the 333 specimens—belonged to the single species A. afarensis, which must consequently have varied greatly in size. The most plausible explanation for the large size range among members of the same species is that males were very much larger than females, comparable to what we see among gorillas today, and not at all like chimpanzees and bonobos, in which sexes differ much less in size.
LAETOLI
Around the time the first Hadar discoveries were being made, another group of paleontologists was hard at work at the site of Laetoli, a thousand miles to the south. Laid down in the Tanzanian portion of the Rift Valley near the well-known site of Olduvai Gorge, the geological layers at Laetoli are slightly older than those at Hadar, running from about 3.5 to 3.8 million years old. Between 1974 and 1979 the broken jaws and teeth of three hominid individuals were collected at various Laetoli localities, but the site is most famous for the numerous animal trackways discovered there beginning in 1976. These include footprint trails left by hominids who, some 3.6 million years ago, had walked across a layer of wet cement-like volcanic ash that subsequently hardened. This was an extraordinary find. We can be confident that Lucy walked upright; but we must always remember that this is not something we can observe directly in the bones. Rather, we have to infer it from Lucy’s anatomical structure. A footprint, however, is different, in that it is truly fossilized behavior. And the trackways at Laetoli are as eloquent of bipedality as it’s possible to get. At one site an arrow-straight double trail of prints some 80 feet long, more or less like those anyone might leave walking along a wet beach, attests clearly to a purposeful bipedal gait. What is unusual is that the Laetoli environment at the time these prints were made was quite open; the hominids were slogging across a flat plain largely devoid of trees, and they must have felt pretty vulnerable as they did so. But they were heading directly for the Olduvai Basin, only a few miles away, which at that time would have offered all the hospitable resources of a forest surrounding a shallow lake.
The footprints themselves are clear evidence of bipedality: there is no indication that the hominids steadied themselves using their forelimbs, and the way in which weight was transmitted from one end of each print to the other seems to reflect the way we walk—which is to say, it went from the heel, along the side of the foot and across the ball, with a final thrust concentrated on the big toe. This was not the lurching gait of a bipedal bonobo. The feet that made the prints were structured essentially like ours, with longitudinal and transverse arches and a short big toe set in line with the others. The short distances between successive footfalls suggests that even the bigger individual was of fairly diminutive stature, although it seems the pair was not moving very fast—hardly surprising, given the slushy surface across which they were making their way.
While there is nothing to cast doubt on the bipedality of these 3.6-million-year-old hominids, there has been some debate about the exact gait they employed. Did they, for example, fully extend the knee with each step? Or did they retain some vestige of the bent-kneed gait that today’s apes use when moving upright, and which, at some remove in time, the hominid ancestor must also have employed? A recent experimental study, using human subjects moving both straight- and bent-kneed, has confirmed that if you don’t fully extend your knee, the impressions your toes make in wet sand are deeper than those made by your heel. And the Laetoli prints clearly show heel and toe depressions that are about the same depth, arguing for a straightened knee. Clearly, in these footprints we have evidence of a serious biped.
The scientists who carried out the experimental work suggest that adopting upright locomotion on the ground allowed the Laetoli hominids to increase their ranging distances without expending extra energy, during a period when the forest was diminishing. Indeed, it’s very unlikely that any hominid could have made a decent living in the rather barren ancient environment adjacent to the trackways, making it all the more plausible that the prints in the wet ash had caught them in the act of aiming straight for the forests that lined the nearby Olduvai Basin.
Just who those bipeds were is another matter. Not far from the footprint tracks at Laetoli are rocks of about the same age that yielded the handful of hominid fossils already mentioned. In an unusual collaboration, the scientists who initially studied the Hadar and Laetoli specimens eventually decided that they were all from the same new sort of hominid. This new species, Australopithecus afarensis, was named for the Afar region of Ethiopia in which Hadar is situated, and from which most of the fossils in question came. But under standard zoological procedure, every new species has to be based on a “holotype,” a single specimen to which every other individual assigned to that species has to be compared. And to emphasize their conviction of unity, the scientists chose a lower jaw from Laetoli as the holotype of A. afarensis. Not everybody found this appropriate, though, as some scientists felt that they could discern evidence for more than one species of hominid just at Hadar, let alone at the Ethiopian and Tanzanian sites together.
At present there is a state of uneasy truce, with most paleoanthropologists willing to accept at least provisionally that the known bones and teeth, at least, can be assigned to the same species. But the association of A. afarensis
and the footprints is much more actively debated. Perhaps a majority of paleoanthropologists is willing to believe that individuals of A. afarensis made the Laetoli trails; but at least a substantial minority thinks that the fossil foot bones from Hadar indicate a foot far too long and primitive to have produced the strikingly modern Tanzanian footprints. If the majority is right, then we will have to accept that the Lucy’s arboreal adaptations, and her broad pelvic proportions, were indeed compatible with remarkably humanlike bipedality. But the jury is still out; and all we know for sure right now is that somebody was out there strolling upright through the Tanzanian Rift 3.6 million years ago.
DIKIKA
Only a few years ago, the word “Dikika” was in almost nobody’s vocabulary, but now it’s one of the hottest buzzwords in paleoanthropology. During the glory days at Hadar in the 1970s, and again in the 1990s, everybody was too busy to look south across the Awash River to the more or less equivalent deposits of Dikika. But when the investigation of those rocks finally began at the start of this century, they proved to have a dramatic story to tell. First, some scrappy bits of tooth and jaw attributed to Australopithecus afarensis showed up; but these were soon overshadowed by the discovery of the crouched partial skeleton of a three-year-old juvenile. Thought to be female, the skeleton was soon baptized with the informal name “Selam” (“peace”). So well preserved was Selam that it seemed the infant was snatched from its group by floodwaters and almost immediately buried whole in soft mud, some 3.3 million years ago. And this poignant Pliocene misfortune proved a bonanza for the paleontologists, who found that the Selam fossil preserved elements that were not included, or were more poorly preserved, in the extensive A. afarensis collections from north of the Awash. Among these elements are a hyoid—the bony portion of the Adam’s apple—that resembles that of an ape rather than a human, and a complete scapula (shoulder blade) that is unexpectedly reminiscent in overall shape of a gorilla’s. Selam has the ankle of a biped, but the carrying angle between her femur and tibia is not marked, confirming that this is a feature that has a strong component of behavior in its development—carrying angles do not properly develop in modern humans who spend their entire lives in wheelchairs.
One big difference between apes and humans is that apes develop to maturity much faster than we do, depriving them of the extended childhood that provides us with so much opportunity to learn. On her own, Selam cannot do much to demonstrate how quickly individuals of A. afarensis developed; but the expectation would certainly be that she lay on the apish side of the curve. The fossil has yet to be fully freed from the hard rocky matrix that envelops it, but as far as we can tell, her upper body structure confirms the generally arboreal features found in adult A. afarensis. Not only does Selam’s scapula show a shoulder joint that was oriented largely upward, as a good climber needs for holding the arms above the head, but her hand also shows features associated with climbing. Ape and human hands may look superficially similar, but they are in fact constructed very differently. Apes have thumbs that are short in relation to the fingers, and the hand is long, with its major axis in line with the arm. The ape hand is more the hand of a powerful grasper than of a dexterous manipulator; and it is the kind of hand that you want if you are going to spend most of your time clambering around in the branches of trees. In contrast, the major axis of the modern human hand goes across the palm, and the thumb is long and can be opposed precisely to any of the other, shortened, digits. Selam’s digits appear to have been long and curved.
The skull of “Selam,” the infant skeleton found at Dikika, Ethiopia. Despite its tender age of only three years, this tiny 3.3 million-year-old skeleton has yielded a wealth of information about the structure and development of the species Australopithecus afarensis. Courtesy of Zeresenay Alemseged.
To complete this image of a hominid that was not completely wedded to the ground, CT scans of Selam’s ear region suggest that the semicircular canals of her inner ear resembled those of apes and other australopiths. These canals are important organs of balance, and their orientation reflects not only the way the head is habitually held, but how well it is insulated from movements of the spine on which it is poised. Selam’s semicircular canals are reported to resemble those of apes and other early bipeds, suggesting that although her species may have been an upright walker, it was not suited for fast running—an activity in which it is important to maintain a fairly constant head position despite the gyrations of the body below.
But Selam is not the only surprise from Dikika. In mid-2010 the research group there came up with something that was even more remarkable. Strata dated to just under 3.4 million years ago yielded four superficially unimpressive fragments of mammal bone that proved, on close inspection using a scanning electron microscope, to bear markings that archaeologists suggested were of the kind produced only by stone tools. To understand the significance of this finding, you need to bear in mind that 3.4 million years ago is 800 thousand years before we have any evidence of stone tools themselves. The very first stone tools we know of are reported from not very far away along the Awash Valley, but are a mere 2.6 million years old. Yet stone tools are very durable things—carnivores don’t chew on them, and under most circumstances they preserve in the record indefinitely. If ancient hominids at Dikika were clobbering one rock with another to produce sharp cutting flakes so they could butcher carcasses, where are those flakes? And where are the “cores” with the flakes removed from them? It’s not as if paleoanthropologists haven’t thoroughly scoured the Dikika and Hadar landscapes for interesting objects of this age.
There are several possibilities as to why no stone tools have been found in these regions. One is that the paleontologists had the wrong “search image,” and simply were not finding such implements this far back in time. But even the most primitive deliberately made tools show distinctive signs of manufacture, and it’s unlikely that over many years experienced searchers would totally miss pieces of stone with obvious modifications. Alternatively, the Dikika researchers suggest that the very early history of stone tool-making showed very low “intensity”: i.e., just one flake was removed per core, so that each core would show little sign of modification while the flakes themselves were rare. Another possibility is that the scratches are in fact “trampling marks” produced by the sharp hooves of grazing mammals that had stepped on the bones. But, perhaps most likely, the australopiths had simply used naturally broken stones for butchery. Experimental archaeologists have now shown that it is indeed possible to dismember a mammal carcass using stones of the kind that are routinely fractured against each other while being swept down rivers. Such pieces don’t have the razor-sharp edges of deliberately manufactured stone tools, but they can nonetheless do the job.
Still, whatever exactly happened, under the microscope two of these slivers of bone (a piece of rib and a fragment of femur, one from an animal the size of a cow, the other the size of a goat) not only show “cut-marks” of the kind that are produced by slashing with a sharp tool while the bone is still fresh, but they also bear scratches and pits such as those made when a piece of fresh bone is scraped or bashed with a hard and pointy object. So here is a strong suggestion that early hominids, presumably Australopithecus afarensis, were indeed butchering the carcasses of large animals out there on the bushy Dikika landscape, some 3.4 million years ago—even if they weren’t, strictly speaking, stone toolmakers.
UP THE RIVER
Dikika presumably has plenty more surprises in store for paleoanthropologists. Doubtless so also has an area, some distance upriver, that is known as the Middle Awash Valley. This region occupies a unique position in paleoanthropology. It has not produced fossils as lavishly as Hadar has, but it has yielded hominid remains that range from Ardipithecus kadabba at 5.8 million years ago, to the very earliest days of our own species Homo sapiens a mere 160 thousand years ago. No other place in the entire world registers events in hominid evolution over such an enormous span of time. The 4.12-million-year
-old Ethiopian Australopithecus anamensis fossils are from the Middle Awash, and from relatively close by comes a geologically younger partial hominid skeleton assigned to Australopithecus afarensis, from a 3.58-million-year-old locality known as Woranso-Mille.
The Woranso-Mille bones are substantially earlier than any of the A. afarensis fossils from Hadar, and they are comparable in age to the Laetoli scraps. Unfortunately the skeleton lacks a skull or teeth, but the skeleton is said to be broadly similar in preserved portions to the later and smaller-bodied Lucy. A scapula is quite well preserved, and although its shoulder joint portion does seem to have been quite upwardly oriented, it is unlike its counterpart in the Dikika child in bearing no particular resemblances to its equivalent in any African ape. As for the lower part of the body, the Middle Awash researchers think that that their bigger-bodied specimen is more relevant than the diminutive Lucy for assessing exactly how A. afarensis walked, as smaller subjects weigh less and therefore require fewer specializations to support their body weight. This is actually pretty arguable, for one thing because it’s hard to imagine a pelvis and leg structure more suggestive of bipedality than Lucy’s. But it is certainly good to have a larger counterpart skeleton. The Woranso-Mille individual sadly lacks a complete leg, but by the team’s estimate this hind limb had been relatively a bit longer than Lucy’s. If this is indeed the case, then the Woranso-Mille fossil might fit a bit better than the Hadar materials with the sort of hominid that made the more or less contemporaneous Laetoli prints. Frustratingly, it has no foot bones, leaving lots of room for speculation on this point.
Masters of the Planet Page 6