Lone Survivors (40 page)

A recent example of such work is the study by Michael Hammer and his colleagues. They used DNA sequences from sixty-one noncoding regions in the autosomal genome of three sub-Saharan populations (the Mandenka of West Africa, Biaka pygmies from central Africa, and San Bushman from southern Africa) to test for signs of archaic admixture in Africa itself. They concluded that African populations also contain about 2 percent of ancient genetic material, and this was input some 35,000 years ago, not from Neanderthals or Denisovans but from an unknown archaic population within Africa itself, which might have been separate from the modern human lineage for some 700,000 years. Three separate regions of DNA showed deep divergences and one seemed to have introgressed from an ancient and now-extinct population in central Africa. Such a deep time scale could suggest that early groups of
heidelbergensis
survived in Africa alongside the evolving lineage of modern humans.

So let us briefly review what we know from fossils about human species about 200,000 years ago. In East Africa, we see the first known signs of the modern human pattern of a high rounded skull, while in Europe, Neanderthal evolution was well under way. We have virtually no data on who was living in western Asia at that time, while the more ancient Narmada skull from India suggests that a descendant of
Homo erectus
or
heidelbergensis
might have been living there. China seems to have been occupied by descendants of
Homo erectus
or
heidelbergensis
too, but because of uncertain dating we cannot be sure whether the Ngandong fossils indicate that
erectus
was still surviving in Java. But
Homo floresiensis
was presumably well established in its long and lonely residence of Flores. That seems to represent a relatively clear picture in Africa and Europe, if imperfect elsewhere. But let's home in on the African record in a bit more detail, and with a more critical eye than I have employed so far.

As we saw, Omo 1 at about 195,000 years and the Herto crania at about 160,000 years seem to establish an early modern presence in Ethiopia, and the Guomde fossils from Kenya might even take that presence back to 250,000 years ago. The fragmentary but more primitive-looking Florisbad skull from South Africa is about that age, but there are other fossils from East Africa that do not fit the pattern so well. These include Omo 2, with its more primitive braincase shape and angled back, and a still undated archaic skull from Eliye Springs in Kenya. In my view the Laetoli H.18 (Ngaloba) skull from Tanzania, which is thought to be only about 140,000 years old, is not a modern human, and I'm doubtful whether the similarly dated but pathologically deformed skull from Singa in Sudan is fully modern either. When we move across the Sahara to Jebel Irhoud, I doubt the modernity of the fossils from there too, even if the synchrotron did show that the child's jaw displays a slow and modern growth pattern in its developing teeth. Those fossils are currently dated to about 160,000 years now, although there are indications that they might be even older than that.

In my early analyses of the cranial shape of these African fossils, I considered them separately, which planted the seeds of a view that specimens like Omo 1 and Jebel Irhoud (if ancient enough) might be part of our ancestry. However, as I became more confident about the nature of the African sample and its dating, I started to lump all the late archaic specimens together in my comparisons with samples from elsewhere, such as Skhul and Qafzeh, the Cro-Magnons, and the Neanderthals. I increasingly came to the realization that this was a mistake, because it obscured the considerable internal variation in that African sample.

African variation 195,000 years ago: Omo Kibish 2 (
left
) and 1.

Now, a similar conclusion about its variability was reached by the anthropologist Philipp Gunz and his colleagues, using geometric morphometrics. They measured the shape of the upper part of the skull in a sample of fossils ranging from
erectus
to Cro-Magnons and in a global collection of modern humans. They found, as I did, that African specimens like Irhoud, Omo 2, and Ngaloba occupied the middle ground between archaics like Neanderthals and
heidelbergensis
and the moderns, including Cro-Magnons and Qafzeh. But the African fossils showed more shape variation from each other than any equivalent group of fossils in the entire sample, whether modern, Neanderthal,
heidelbergensis
, or
erectus
. Might this be telling us something important about modern human origins in Africa? I have argued that
heidelbergensis
in Africa, represented by fossils like Bodo (Ethiopia), Elandsfontein (South Africa), and Broken Hill (Zambia), probably evolved into modern humans in some part of the continent, and I have wavered over whether this transition was a gradual or a rapid event. The general consensus has been that the transition was gradual, and that dating fossils like Broken Hill to about 500,000 years ago allowed plenty of time for the required major evolutionary changes in skull form (and probably body form, brain, and behavior) to occur.

But even the existing fossil record can still surprise us. Fifty years after Darwin published his cautious prediction that Africa would turn out to be the continent of our origin, the Broken Hill skull was fortuitously discovered, and so began the process of proving he was right. But the skull was not immediately recognized for what it was, and might easily have been lost to science. On 17 June 1921, miners quarrying a small hill full of metal ore near the town of Broken Hill (now Kabwe), in what was then Northern Rhodesia (now Zambia), uncovered a skull coated in brown sediment. Its huge staring eye sockets apparently scared them so much that they all ran off. Their supervisor, a Swiss miner called Tom Zwigelaar, was somewhat braver and got someone to photograph him at the site of discovery holding the relic.

The Broken Hill cranium (“Rhodesian Man”) was the first important human fossil ever found in Africa, and even now it is one of the most impressive. It resides in a metal safe outside my room at the Natural History Museum in London and is one of the treasures of the Palaeontology Department—shiny brown and beautifully preserved, with massive brow ridges glowering over those empty eye sockets. It was first put into its own species—
Homo rhodesiensis
—by Arthur Smith Woodward of the museum in 1921 and has been named and renamed ever since. In 1930 it was described by the Czech-American anthropologist Aleš Hrdli
č
ka as “a comet of man's prehistory” because of the difficulty of deciphering its age and affinities. Despite its completeness and apparent primitiveness, its exact place in human evolution still remains unclear, because it has never been properly dated. When I was a student, it was still being used as evidence that Africa was a backwater in human evolution, because such a primitive specimen was living there only 50,000 years ago, while much more advanced humans had evolved in Europe and Asia. Now, as I said, it is usually placed with fossils like Bodo and Elandsfontein as representing the African component of our ancestral species
Homo heidelbergensis
from about half a million years ago.

As a small boy, visiting with my parents, I can vividly remember seeing the Broken Hill skull (or rather a plaster cast of it) on display at the Natural History Museum and being intrigued by its primitiveness and mystery. Ever since, I have nurtured the hope that I could help to place it definitively in the story of human history, either as an ancestor or as a distinct species that had died out without contributing to our evolution. I studied it for my Ph.D. in 1971, and I have regularly included it in my analyses of fossils, making it a central part of my concept of
Homo heidelbergensis
as a species that represented the last common ancestor of modern humans and Neanderthals. But without knowing how old it was, its precise place in human evolution has remained elusive and seemingly beyond reach, given the complete destruction of the site from which it came. But at last this is proving possible—and the results are surprising me as much as anyone else. For at least fourteen years prior to the discovery of the skull, miners had been digging through a fifteen-meter-tall column of fossilized bones, and because these were heavily impregnated with mineral ores, they were throwing them all into the smelter—I'd rather not think about what might have been lost!

After its discovery, other fossil human and animal remains and artifacts were recovered in the site and around the locality by people like Aleš Hrdli
č
ka and Louis Leakey, including from the mining dumps and the miners' huts, but only two human bones were found close to the skull, and at the same time. These were a long and straight shinbone (tibia) and the middle part of a thighbone. The latter find had a particularly interesting history. It was found by a Mrs. Whittington, who happened to be visiting her sister, whose husband worked at the mine. She was obviously an adventurous woman and was lowered down on a rope to collect it, but it was then virtually forgotten until Desmond Clark negotiated its transfer from a Rhodesian museum to London in 1963. Two other intriguing nonhuman finds also proved to be important in unraveling the lost history of the Broken Hill skull. One was a thin and mineralized yellowy brown silty deposit, which the miners collected because they mistakenly thought it was the mummified skin of Rhodesian Man. The other was a mass of tiny bones found around the skull and even cemented inside it. Originally thought to be those of bats, these actually represent the bones, jaws, and teeth of various small mammals, and they provide important information on the age of the skull and on where it originally lay in the cave.

The skull itself shows a strange combination of features. On the one hand, the brain size is only just below the modern average, but on the other, the face is big and the braincase shape decidedly primitive—long and low with enormous brows (
monstrous
was the word that Hrdli
č
ka used), with an angulated back to the skull and a transverse bony ridge reminiscent of
Homo erectus
. The cheekbones are not hollowed out as in modern humans (although a second upper jaw fragment found elsewhere in the site does show this feature), and the teeth are riddled with disease to an extent unusual for an early human: many are decayed and some of their roots are abscessed.

There are several other curious features, including a small and nearly circular hole in the left side of the braincase. Over the years, this has been suggested to be from a spear point, a lion's canine, or even primitive surgery. But not long after I joined the Natural History Museum, I learned of an entirely new idea. A British newspaper was serializing a book called
Secrets of the Lost Races
and requested permission to print a picture of the Broken Hill skull. When I asked what caption would accompany the illustration, I was told that it would say that this was the skull of a Neanderthal shot by an alien's bullet 100,000 years ago! I pointed out that the fossil wasn't really that of a Neanderthal, that it was probably much older than 100,000 years, and that a bullet hole would probably have been accompanied by radiating cracks; I also asked, what self-respecting alien would be using something as primitive as bullets? Nevertheless, it was agreed that the newspaper could have its photo if it included the statement that recent research suggested the hole showed signs of healing and was probably caused by disease emanating from within the braincase. Of course such scientific data didn't suit the paper's agenda, and it included a drawing of the skull instead, leading me to suffer several frustrating weeks as members of the public telephoned, wrote, or even turned up unannounced at the museum asking to see “the Neanderthal shot by the spaceman”!

The tibia found with the skull represents a tall individual of about 180 centimeters, but for a presumed male
heidelbergensis
, he was probably not that heavy at around seventy-five kilograms, compared with estimates for the fossils from Boxgrove (tibia) and Bodo (skull) of over ninety kilograms. So Rhodesian Man probably had the tall and relatively slender build we might expect for an inhabitant of the drier tropics today, although more robust and muscular. The tools from the site are a varied bunch, and none can be closely associated with the skull except for a round stone found with the femur fragment and which, along with others from the site, has been interpreted as a projectile, a pounding stone, or even a bola (a hunting or herding weapon formerly used in South America, consisting of balls connected by string or rope and which are thrown to entangle the legs of the target animal). Other artifacts included flakes, scrapers, and even some possible bone tools, but none of these look very ancient, suggesting a maximum age of perhaps 300,000 years. While some of the animal bones found around the site may have been those of prey, no studies so far have found convincing evidence of butchery, and given the complete destruction of the site, it is impossible to tell whether the bones and stone tools come from inhabitants of the cave(s) in the mined hill or found their way in through some other means.