The Sediments of Time

by Meave Leakey

  The fact that the last H. habilis lived for so long alongside H. erectus within Africa but had died out whereas H. erectus conquered all of the world it could reach by land makes one wonder what erectus had that habilis didn’t. While we’ve speculated that persistence hunting was the key, we won’t know for sure until somebody is lucky enough to stumble across an H. habilis skeleton. We still know frustratingly little about this elusive and mysterious hominin, and for now, our story continues with the evolution of H. erectus into Homo sapiens.

  18

  Through Thick and Thin

  One day in 1980, two visitors arrived out of the blue at the National Museum carrying a loaded old kikapu basket of woven palm leaves and insisted on seeing Richard urgently. Upon admitting them to his office, Richard had a few choice words for the meteorologist and his wife, for their kikapu was chockablock with sturdy black fossils of every description. “What on earth are you doing with all those fossils!” he snapped. “Don’t you know that’s illegal and wrong?”

  The couple had been exploring the western shore of Lake Turkana near a lush, palm-laden oasis called Eliye Springs. They obviously didn’t know about Kenya’s Antiquities and Monuments Act, which prohibits the collecting of any fossils without a permit. Right at the bottom, unwrapped and unprotected, sat a rather impressive skull. It bore all the hallmarks of a collection of Middle Pleistocene hominins from both Europe and Africa that are early versions of Homo sapiens.

  These skulls have even more pronounced handlebar brows than any species known from earlier times. They also have a large nasal aperture and a low thick boned cranial vault. They lack the hollow seen in our cheeks (in scientific parlance called the canine fossa) so their cheeks probably would have looked stuffed as though they got caught in the kitchen before dinner. Some of them also have a vestige of the erectus occipital torus. Moreover, their brains have finally begun to enter the “human” zone: the largest of these skulls from Africa, the Bodo cranium from Ethiopia, has an estimated brain size of 1,300 cc.

  The average brain capacity for H. sapiens is 1,350 cc. The skull in the kikapu was complete but for its handlebar brows and its upper jaw. Richard could see right away what an unusual hominin this was. He didn’t know what to do first—castigate his excited visitors even more severely or thank them grudgingly while admitting to the huge excitement welling up inside him. Here, perhaps, was Kenya’s first example of a very early Homo sapiens. Because they were of a scientific bent, the couple had not only brought the bones to the museum but were also able to give us a remarkably accurate description of where they had found them, and they had left a conspicuous marker behind on the beach.

  Richard and I were soon flying to Turkana, Alan Walker in tow. We had asked Kamoya to meet us there with a vehicle and mark out a makeshift airstrip for us to land on. But as we flew north, huge banks of storm clouds were gathering in all directions. We dodged them with increasing trepidation as large sheets of water accumulated below us and we passed river upon river in full spate. Kamoya would be lucky to get across all of these normally dry rivers—let alone the mighty Turkwell, which lay between the main road and Eliye Springs. Sure enough, as we flew over the Turkwell at Lodwar, we made out a familiar Land Rover with Kamoya beside it waving energetically at us in dismay from the near bank of the swollen river. Richard, rather typically, had no intention of turning around after having got this far. “Well, we’ll just find our own place to land and walk from there,” he said. I was rather alarmed about this, since every potential exposed flat area that I could see was heavily waterlogged, and if a plane sinks into soft ground on landing, it digs its nose in the sand and cartwheels over the propeller. But being able to arrive within metres of the place he wishes to go with nothing but his highly attuned sense of direction to guide him is one of Richard’s most useful—and sometimes most annoying—talents, and we soon found ourselves safely on the ground a short walk away from a well-marked hole in the sand exactly as the couple had described. The hole from which the skull had been extracted was very close to the shoreline, and the only reason that these bones had been exposed was because the lake was unusually low that year. Poor Kamoya had driven in great anticipation all the way to the Turkwell in vain, but at least we had not been likewise thwarted. And I suppose there is some consolation in that he was saved the backbreaking trouble of preparing an airstrip!

  Since the brow ridges appeared to have come off very recently in a clean fresh break, we hoped to be able to find them at the site. We also wanted to determine the age by looking at the geology. What we found was rather disappointing. There were bones everywhere—but the brow ridges were not among them. All the bones we saw were very rolled and weathered, and had clearly eroded out of the ground elsewhere and been redeposited in the current location much later. We still don’t have a date for this specimen, but its features suggest that it is one of these early modern humans and that it lived anywhere between 600,000 and 200,000 years ago.

  Only two of these curious chimeras, which are transitional between H. erectus and modern H. sapiens, have been found in Kenya. The second was discovered by Wambua in 1976 on the other side of the lake at Ileret. This skull has been dated at 300,000 years and looks really modern in spite of being really rather old. A 270,000-year-old femur that Kamoya found close to the skull is similarly modern-looking although it is notable for its thick-set and robust proportions.

  The fossil trail of our ancestors is decidedly feeble between H. erectus and H. sapiens, and its interpretation is made more difficult by the absence of associated volcanic material suitable for dating. Although at least six alternative dating methods have now been developed for this time interval, they are prone to error in various conditions. The oldest evidence is a handful of skulls of early Homo sapiens from around Africa that range in age from about 700,000 years to 200,000 years. Like we saw in the herring gulls of northern Europe and the anamensis-afarensis lineage, we have a series of gradually evolving fossils that grade from typical H. erectus to anatomically modern H. sapiens with no obvious point when one clearly transitions into the other. As a result, many names have been given to these specimens, and their true affinities are unclear.

  Nevertheless, these skulls have been named Homo heidelbergensis in Europe after the first of these, a mandible, was found at the village of Mauer near Heidelberg, Germany, in 1907. In Africa, these archaic hominins are often grouped under the name Homo rhodesiensis after the most complete and best-known example was found in 1921 in Kabwe in what was then Rhodesia (now Zambia). It is by no means clear, however, that H. heidelbergensis and H. rhodesiensis are two different species. What is clear is that by 200,000 years ago Homo sapiens had evolved to become anatomically identical to our own modern form.

  Another feather in Kamoya’s cap is that he was the one who found the earliest currently known anatomically modern Homo sapiens in 1967. At the time, however, nobody apart from Richard recognised its age or its import. Richard had invited Kamoya to join his first expedition to the Omo River, which he was leading on Louis’s behalf. The expedition was a joint collaboration with several teams from Kenya, France, and the United States, and Richard found himself allocated with the most logistically challenging of all the areas they planned to prospect. Most of his problems revolved around the Omo River, which was at that time seething with enormous and ferocious crocodiles. On one occasion, Louis visited both the American and Kenyan sectors to see some fossils and was obliged to go up the river to reach the Kenyan camp. And he reported counting no less than “598 crocs, none of them less than seven or eight feet long, and some nearly twenty feet.” Richard and his field crew had to cross the Omo to reach large parts of their survey site, and in all probability, they were the team with the greatest chance of overcoming the formidable obstacles involved, though I believe that Richard felt at the time that he hadn’t been dealt a particularly good hand. The raft they had built to ferry the cars across the river was a ramshackle affair, but it would have been perfectly adequate
if Louis had provided an engine with enough horsepower to contend with the strong currents and the lightning speed of the attacking crocodiles.

  Richard was disappointed to find few fossils of any import in the limited area of older sediments allotted to their group and turned reluctantly to the far younger Kibish Formation, which he was less interested in. But here they were soon rewarded with Kamoya’s discovery of an almost complete skull and a partial skeleton. Soon afterwards, another member of their team, Paul Abell, found a second skull from sediments of the same age on the opposite side of the river. At the time, these clearly identifiable remains of Homo sapiens were dated at 130,000 years based on the uranium-series dating of snails found in the same level. This was much older than 60,000 years, which was the age when people used to think Homo sapiens first appeared.

  The full significance of this find would be understood only much later when Frank Brown returned to the Kibish with primate expert and palaeontologist John Fleagle and re-examined the sites and geology of these two ancient sapiens. More than three decades after his student days on the original Omo expedition, Frank’s understanding of the Omo-Turkana basin geology had radically improved. He had questions to settle on several fronts. Sceptics had questioned whether the two finds were really the same age because Abell’s skull looked more primitive than Kamoya’s. The original age had also been suspect, and some considered it to be too old. Frank and John were able to locate the exact site of both discoveries, and they even retrieved more parts of Kamoya’s skeleton that had eroded out of the ground in the intervening years. Frank found a tuffaceous layer just below the level of the hominins that had gone unnoticed in 1967 and another new tuff some way above the hominin layer. From these two tuffs, the age of the hominins was securely sandwiched between an older limit of 198,000 years and a minimum of 104,000 years.

  With about fifty metres of sediments between the two tuffs, the big challenge was to narrow this wide age range down and pinpoint a precise date. Frank correlated the known dates of the two tuffs with well-dated sapropels and palaeomagnetic reversals in the Mediterranean cores so he could count the bands marking wet spells when the lake level rose and match them to the Mediterranean sapropel sequence. The result was an astonishing age of 195,000 years. This makes the Kibish fossils the oldest modern human known in the world. With these reliable dates, the Kibish finds supplanted Tim White’s remarkable 1997 discovery of one immature and two adult H. sapiens in Herto, Ethiopia, that had hitherto been considered the oldest known example—between 154,000 and 160,000 years.

  We are now into such a recent part of our prehistory that we can more precisely track the climate as it is minutely recorded in ice that has not yet melted since it first formed. Antarctica has been covered in ice permanently for at least the last five million years. This ice builds up incrementally in successive layers each winter season, and each summer, dust from the exposed land below the ice cap is whipped up by strong winds and settles in a layer over the last season’s snowfall, and these layers accumulate in massive ice sheets.

  The very oldest layers get so compressed that they melt under the combination of the huge pressure exerted by the miles-long mass of frozen water above and the thermal heat rising from the ground below. Slices of time are therefore removed from the bottom, and the record is preserved only for the relatively recent past. Still, we have a year-by-year record in the ice for hundreds of thousands of years. The ice cores allow us to build a picture of our recent climate in much more precise detail than we can for the time of H. erectus when we have only the foraminifera and other deductive clues in deep-sea cores to rely on. The fine-textured ooze of sea cores is formed through such slow deposition that the resolution of detail is quite low—and a single worm can disrupt the work of hundreds of thousands of years of accumulation!

  One of the oldest ice cores, which goes back at least 740,000 years and is a mind-boggling two miles long, was recovered from the Antarctic ice sheet from a site called Dome C. Extracting these ice cores is no mean feat. In his book The Weather Makers, Tim Flannery rightly calls the extraction of the Dome C core as one of science’s greatest triumphs and describes the inordinate challenges involved.

  The drill site was bitterly cold: -58°F at the beginning of the drilling season and -13°F in the middle of the Antarctic summer. The drill itself is just four inches wide, and as it grinds its way downward, a slender column of ice is separated and drawn to the surface. The first mile was especially difficult, for there the ice is packed with air bubbles, and as the core was drawn up, these tended to depressurize, shattering the ice into useless shards. Worse, ice chips can clog the drill head, jamming it fast.

  The very air bubbles that make extraction of the cores such delicate work are part of the reason that the ice cores are such a good record of climate change. It all has to do with the way the ice sheets form. The top layer of the ice sheet is fresh winter snowfall. Immediately below this loosely packed snow is a layer that has been newly transformed into exquisite ice crystals. Farther down the ice sheet, these crystals become compacted by the weight of the layers above until, approximately one thousand years after the snow first fell, they have become recrystallized into solid ice (which resembles the ice cubes we make in our freezers for drinks). Tiny bubbles of air become trapped in the new ice when these snow crystals recrystallize and encapsulate a micro sample of the atmospheric air at the time that they formed. Because the chemical composition of atmospheric air depends on local temperature conditions, analysis of the carbon dioxide levels, the oxygen isotope ratios, and the proportion of methane that is trapped in the tiny bubbles gives an accurate barometer of the climate conditions under which they formed. Layer by layer through the ice sheet, the composition of the air in the bubbles varies, tracking the rise and fall of both atmospheric temperature and sea levels.

  The dust separating each ice layer also provides valuable information about annual cycles. During colder glacial periods, there is less precipitation around the world and the dust layers are thicker—and the amount of dust is a proxy for windiness. One of the most well-known ice cores, the Vostok core from Antarctica, is 3,600 metres long and covers a time interval of 420,000 years. Four complete glacial and interglacial cycles are recorded in the Vostok core, each with a periodicity of 100,000 years. The core clearly shows that the windiest, dustiest time is at the height of glacial periods when temperatures are at a minimum. The data are consistent with other evidence that shows that the global climate is cold, arid, and windy during glacial maxima and that there are more grasslands and expanded deserts and less forest cover. Since the work at the Vostok ice core site, scientists have recovered longer sequences—800,000 years of climate history preserved at Dome C and a tantalizing 2.7-million-year sequence.

  At first, climatologists concentrated on extracting and analysing ice cores from the polar regions. But to get a truly global climate reconstruction, the ice cores from tropical glaciers also needed to be part of the picture. As you can imagine, extracting a glacial core from the top of a mountain is even more challenging than in Antarctica because of the dangers of working at high altitude for prolonged periods and the logistics of getting heavy machinery and a source to power the drill up precipitous mountain slopes—not to mention the difficulties involved in bringing the heavy cores down without melting them. There is an extraordinary account in Thin Ice of Lonnie Thompson’s lifelong endeavours to get ice cores out of the tropical glaciers of mountains before they are all gone. This book by Mark Bowen vividly details many of Thompson’s greatest triumphs, including the successful extraction of an ice core from the Guliya Glacier on the Tibetan plateau that goes back 760,000 years. Over several decades, Thompson’s late friend and colleague Bruce Koci developed lighter and better equipment to use at the top of tropical mountain glaciers. Thompson now has cores from Latin America, China, Bolivia, Alaska, and Africa. He is credited as the man largely responsible for accumulating much of our body of knowledge on low-latitude, high-altitude ice co
res. And ice cores permit us to reconstruct the global climate at the time H. sapiens evolved and moved out of Africa.

  When we compare the data from both alpine and polar ice cores, it is clear that the greenhouse gases in the atmosphere—carbon dioxide and methane—have varied in lockstep with the cycles of glaciation and deglaciation and that these are heavily determined by the Milankovitch cycles. The ice cores also show very sharp transitions from cold to warm periods, with sudden temperature changes of twelve degrees Celsius, whereas cooling happened much more gradually. Although far more detailed than the sea cores, the ice cores are in agreement about global climate swings. The composite picture that we can build shows that Homo sapiens, like Homo erectus, evolved during a glacial-interglacial icehouse world—one that was both more pronounced in the extremes of cold and warm and more erratic.

  This information now makes it possible to trace the last chapters of human evolution with a vastly improved understanding of the climate. H. sapiens evolved just under 200,000 years ago, and this coincides with a warmer interglacial period that started some 245,000 years ago and ended some 185,000 years ago. Then the ice returned for 55,000 years. All this leads to one unavoidable conclusion: for the large part of the evolution of H. sapiens, the whole planet was perishingly cold and the tropics were parched, arid lands. During the most arid phases, as evidenced by cores taken from Lake Malawi, vegetation was so sparse that the usual bushfires did not occur, for there is a marked decrease in charred particles and very little pollen present. The level of Lake Malawi, which today is a staggering 706 metres deep, was at times reduced to a depth of 125 metres, and the salinity and alkalinity rose dramatically.