Lone Survivors (11 page)

Nearly twenty Neanderthals have now been CT scanned to reveal their inner ear anatomy, and each semicircular canal is subtly distinct in size, shape, and orientation when compared with ours. What makes this discovery particularly intriguing is that the canals in the assumed ancestral species
Homo erectus
, and in early modern human fossils studied so far, are more like our own, so it seems to be the Neanderthals that are the odd ones out. But fossils in Europe that might be ancestors of the Neanderthals, such as the skulls from Steinheim and Reilingen in Germany, show an approach to the Neanderthal conformation, suggesting that the distinctive pattern could have evolved in Europe. But why?

One possibility is that the shape of the semicircular canals is reflecting something else, such as overall skull or brain form, and it's true that the Neanderthals do have some distinctive features in the shape of their temporal bones—the bone around the ear region on each side of the skull. Another possibility is that it reflects some form of adaptation—perhaps climatic—but against this, modern humans from cold climates show no significant differences from moderns who live in hot climates. The scientists who have conducted the most comprehensive studies, including Fred Spoor, argue that a plausible explanation lies in the essential function of the semicircular canals: to control the movement and rotation of the head. Although the exact mechanisms of interplay between the head and neck and the semicircular canal system are poorly understood, the Neanderthals had shorter but bulkier neck proportions than modern humans, which might have affected movements of the head, if it was more deeply buried within powerful shoulder and neck muscles. In addition, the Neanderthals had a more projecting back to the skull, a flatter base to the braincase, and a more projecting face, particularly around the nose, all of which might have made a difference to head movements all the way from less strenuous activities like walking through to highly energetic running or hunting.

One of the first fossils to reveal these unusual Neanderthal inner ears forms part of the collections at the Natural History Museum in London. This is the rather fragmentary Devil's Tower child's skull (from its large size, probably a boy), found together with animal bones and stone tools beneath the sheer north face of the Rock of Gibraltar during excavations in 1926. It consists of three skull bones, half of an upper jaw, and most of a lower jaw, with a mixture of milk teeth and still-forming permanent teeth. In modern children, where the birth date is uncertain, or an unknown murder victim needs to be identified through forensics, the best way to estimate age is from their teeth. This method was applied to the Gibraltar fossil, and it was straightaway evident this was a child less than six years old in modern terms, as the first molar tooth was not yet ready to erupt. Studies of the fossil in 1928 suggested from the dental maturity that he was actually about five years old at death, but to judge from the voluminous skull bones, his brain size was already slightly larger than the modern average. Until 1982, everyone assumed that the bones belonged to one child, but in that year the anthropologist Anne-Marie Tillier suggested that while most of the bones did indeed represent a child of about five, the temporal bone was from a different and less mature child of about three years old at death.

In the 1970s new microscopic techniques became available to study the microstructure of teeth, and earlier suggestions that human tooth enamel contained daily “lines” of growth began to be studied as a means of estimating the length of time it took a tooth to develop, and hence of potentially gauging the age at death of a child. These daily lines are grouped and expressed on the surface of the front teeth as transverse ridges, or
perikymata
(from two Greek words meaning “around” and “a wave”), each of which represents about eight days of growth. In the 1980s, using a scanning electron microscope, I collaborated with the paleoanthropologists Tim Bromage and Christopher Dean, and subsequently also with the primatologist Bob Martin, to estimate the probable age of the Devil's Tower child from its well-preserved upper central incisor, and to study its growth and development. By counting the perikymata and adding a few months to represent the small amount of root growth, we estimated the age at about four years. We also used a rare and important collection of human skeletons from the crypt of Christ Church at Spitalfields in the City of London, with actual age at death recorded on coffin plates or parish records, to test the perikymata method. We found that it worked well as an estimator of age in the children who had been buried there. In addition, I studied the temporal bones of the same children's skulls to assess whether a temporal bone as immature as the one from Devil's Tower could belong with the other bones and teeth of that child. The results were clear: both the teeth in the jaws and the temporal bone came from a child of about four years at death, and thus there was no reason to dissociate them on grounds of differential maturity. However, because the temporal bone was from the other side of the skull to the equivalent parietal bone, the two could not actually be directly articulated to prove they belonged together.

Nevertheless that “fit” was demonstrated a few years later when the CT experts Christoph Zollikofer and Marcia Ponce de León used the technique to reveal further anatomical data and to produce a three-dimensional reconstruction of the whole skull, showing that the temporal bone undoubtedly belonged with the other remains. They not only mirror-imaged the missing parts from the preserved portions but were also able to complete a hypothetical whole skull by “importing” elements from other Neanderthal children who had the appropriate parts preserved, adjusting their size virtually to complete the fit. To test the method, the researchers also virtually disarticulated a modern child's skull of comparable maturity and demonstrated that they could re-create it very accurately, using only the portions preserved in the Gibraltar child.

Having re-created the Devil's Tower child's skull digitally and on-screen, they could also remake it physically, using a technique called
stereolithography
. This technique was developed for industrial purposes to test the fit of parts with each other, and rather than carving out or molding a shape, objects are built up via the consecutive solidification of thin layers of a light-sensitive liquid resin. It is magical to watch the process—an ultraviolet laser beam, guided by the digital CT data, gradually materializing a solid object out of a pool of transparent resin. A skull or jaw can be re-created, one thin layer at a time, as the beam flickers across the resin, causing the liquid to progressively set. This replication method has many advantages over conventional molding and casting: it causes no damage to the surface of valuable fossils since it is noninvasive, it is remarkably accurate and nondistorting, and internal structures such as air spaces and unerupted teeth can be replicated and made visible if the transparent resin is left uncolored.

But this was not all that was revealed. The boy's teeth (including those still unerupted in the jaws) were also studied in great detail, and a feature that had been noted in previous research was given special attention. The front teeth in the two halves of a lower jaw are usually mirror images of each other in terms of their positions and orientations, but in the Devil's Tower specimen, some on the right side seemed out of place. The CT images clearly showed that this boy had suffered a fracture of the lower jaw earlier in life but had survived, allowing the injury to heal quite well, and so this was unlikely to have been the cause of his early death. As already mentioned, he was large-brained, and the CT reconstruction also allowed an accurate estimate of his brain size, which would have been between 1,370 and 1,420 cubic centimeters, with a little more growth to come—a volume already comparable with those of European men of today.

There has been much discussion about how Neanderthals grew up—whether they matured at the same pace as we do today—and the Devil's Tower child has become an important part of the discussion. Apes have rapid brain growth before birth and relatively slower growth in the years immediately after, while we have rapid brain growth both before and after birth. At birth, allowing for body size differences, human babies already have brains that are one third larger, relatively, than those of apes, but by adulthood our brains are three times larger. The fact that we must grow our brains so much after birth is largely dictated by the limits imposed by the size and shape of the birth canal of the human pelvis, and it's likely that there is a limiting threshold of about five hundred cubic centimeters, after which a substantial period of postbirth growth in brain volume would be required.

This threshold must have been reached during the time of
Homo erectus
, which means that
erectus
babies probably had extended periods of immaturity compared with apes, during which the brain could continue to grow at a fast rate. For example, estimates suggest that compared with our landmarks for the average eruption age for the first, second, and third molars of about six, twelve, and eighteen years,
erectus
may have had a timing of about five, nine, and fifteen years respectively. But that eruption sequence marking important stages in childhood, adolescence, and the beginning of adulthood would still have been far more prolonged than in the chimpanzee, whose molar eruption ages are about three, six, and ten years.

Essentially apes have an infancy of about five years, after which they have about seven years of adolescence and are then projected into adulthood, whereas modern humans have two extra phases inserted between infancy and adolescence: childhood (about three to seven years) and a juvenile phase (between about seven and ten years). In these phases the child is still dependent on support from its mother and older kin, for protection, for learning, and for food to grow and fuel an energetically demanding large brain. The fact that our children grow so slowly spreads out the energetic costs of rearing them and may be an important contributing factor in the greater number of children that
Homo sapiens
parents can sustain, compared with apes. And recent studies have shown that although adult human brain size is essentially achieved by the age of eight, the brain continues to wire up its connections and cross-connections right through adolescence, when in humans there is still much to learn culturally and socially. In addition, we mature much later than the other apes, with an adolescent phase lasting between the ages of about ten and eighteen. Neanderthals with their large brains must have had long childhoods too, although, as we shall see, there is some evidence that they reached adulthood earlier than the average for humans today—not surprising, and perhaps even essential, if most adults were likely to die before they reached forty (see chapter 6). So their learning processes would have been extended too, even if not quite to the extent we find in our species, and their brains may have had to grow to their large size at a slightly faster rate and over a shorter period—which perhaps explains some aspects of their diet. Whether their large brains endowed them with an intelligence like ours is another fascinating question.

The brain and head size of
Homo sapiens
at birth are right at the limit of what is practicable for the human birth canal to withstand, and medical science may be required to assist in difficult deliveries, taking over the role of midwives in traditional societies. There are a few poignant Cro-Magnon burials of women with seemingly newborn babies, attesting to the difficulties of birth 30,000 years ago. A notebook account of the 1932 excavation of the much more ancient Tabun Neanderthal woman's burial, in what was then Palestine, mentions the skeleton of a fetus tucked against the side of her body. Sadly, these enigmatic remains were never described, and we do not know whether this was a mistaken identification or if the material was too fragile to recover from the hard cave sediments. But the woman's skeleton did survive and is curated at the Natural History Museum, representing the most complete female Neanderthal skeleton yet described (others from the Sima de las Palomas in Spain are in the process of study and publication by Erik Trinkaus and colleagues).

The CT experts who worked on the reconstruction of the Devil's Tower child's skull also worked on reconstructing the pelvis of the Tabun woman. In the absence of the putative remains of the baby that was found with it, they instead reconstructed the fragile skeleton of a newborn Neanderthal child buried at Mezmaiskaya in Crimea, and in a spectacular demonstration of the power of CT technology, they combined the two in order to investigate Neanderthal obstetrics. They discovered that the child's brain size was about four hundred cubic centimeters, typical of a newborn today, but the skeleton was already much more strongly built. In testing the birth process, it was apparent that the slightly wider pelvis of the Tabun woman should have eased labor. However, the baby's skull was already Neanderthal-shaped, with a longer head and a more projecting face, suggesting that the birth process would have been as difficult for the Neanderthals as for us, involving the same unique (to humans) twisting of the baby's body during delivery.

In another CT study of the Tabun woman's pelvis, this time without a direct newborn comparison, the paleoanthropologists Tim Weaver and Jean-Jacques Hublin came to rather different conclusions, arguing that the Neanderthal birth process would not have been the same as ours. The modern birth canal is widest across its breadth higher up, but then alters downward to become widest front to back, which is why our babies generally change position as they descend. However, in their reconstruction Tim and Jean-Jacques found that the Tabun birth canal was wide across its breadth throughout, and thus the Neanderthal birth process would have been simpler than ours, without the need for additional rotations, and perhaps less dangerous. We
Homo sapiens
have narrower pelvises than either our Neanderthal cousins or our African predecessors, for reasons that are still unclear, but this evidence suggests that the change in our hips led to new evolutionary demands, probably requiring both biological changes in the process of delivery and social changes in the level of support needed for modern human mothers giving birth.