The Sediments of Time

by Meave Leakey

  We extracted the skull and carefully packed it into a box for the journey back to camp. The following days were spent sieving the slope below and digging a small excavation to make sure there were no more pieces still hidden in the ground. We then took several sacks of the sieved sediment back to camp, and this was washed with lake water and sieved again. Meanwhile, I reconstructed as many of the broken pieces as I could to make the specimen stronger before taking it back to Nairobi. This initial reconstruction revealed an intriguing face that seemed unusually long and flat with rather small teeth. Was this really A. afarensis? And if it wasn’t, could the damaged bones tell us what we needed to know? I couldn’t wait to find out.

  It was a long time before we could resolve the question as to what this skull represented. After our return to Nairobi, Christopher Kiarie, who is an excellent preparator, spent nine painstaking months removing the sandstone rock grain by grain from the bone. As more of the pieces were cleaned, we were able to start reconstructing the skull. The more we put together, the more unusual the face became. It didn’t look like we had ever encountered one of its kind before, and we were impatient to begin the study and make comparisons with other known specimens. We enthusiastically began a scientific study of the skull as well as the other hominins we had found at Lomekwi. A new long-time collaboration also began with Fred Spoor, an excellent Dutch anatomist and palaeontologist, who was at the time working at University College London before moving to the Natural History Museum. Fred had first worked with us in the field at Kanapoi, and his cautious and meticulous approach was perfectly suited to the controversial task ahead of us. What was this flat-faced hominin, and how did it fit into the evolutionary picture that had been built up gradually over many decades of discovery?

  When the Lomekwi skull was finally prepared, Louise, Fred, and I arranged to visit Addis Ababa to make comparisons with similar-age fossils that had been found in Ethiopia. The first order of business was to ascertain if our flat-faced specimen could be a variant of A. afarensis, since the sediments at Lomekwi covered the same time interval as those from which A. afarensis derived. Tim White and Don Johanson were generous in allowing us to look at their specimens, and Berhane Asfaw and Bill Kimbel, who were both in Addis at the time, willingly shared their fossils and ideas with us. Berhane and Bill are both old friends whom we have known for many years. Bill has an extensive knowledge of both A. afarensis and early Homo, and we frequently bounced ideas off him. We greatly valued his insights and the scientific discussions with Berhane about our Lomekwi skull that made our visit to Addis all the more fruitful. We took numerous measurements and observations of all the afarensis specimens in the Ethiopian collection. We became increasingly excited, as there were apparent differences even before we could fully analyse the data.

  If our flat-faced friend was not A. afarensis, we needed to rule out affinities with any of the other known species of the same time interval. Later that year, the three of us travelled to South Africa and studied the specimens of A. africanus and Paranthropus robustus housed at Witwatersrand University and Pretoria’s Museum of Natural History. Back in Nairobi, we included detailed comparisons with the Kenyan collections in the Nairobi National Museum.

  Although Kiarie had done a superb job of cleaning the fossil, there was nothing he could do about the fact that the skull was distorted. The entire skull was slightly off-centre, and numerous hairline cracks crisscrossed its surface. These cracks had filled with matrix as the bone became fossilised and had distorted the specimen. This made comparisons much more difficult and interpretations more tentative. But whilst the distortion of the cranial vault was so extreme as to conceal its real shape, the face did not appear to have been as seriously affected. Nevertheless, if we were going to name a new type of hominin, we needed to be sure that our measurements were not off because of these distortions.

  We were lucky that technology invented for medical purposes works perfectly well for fossils too and that we could use CT scans to look at the inside of the cranium to determine the distortion and then use computer-imaging technology to reconstruct it with a far greater degree of accuracy than would otherwise be possible. We could also take scans of the tooth roots normally hidden from view. The CT results confirmed our initial measurements, and the most outstanding observation was that A. afarensis was very different from our Lomekwi skull.

  We immediately started thinking about what this skull meant. Having spent the last decade trying to find diversity among the hominins between four and three million years ago, we were now faced with the usual problem of how much variation should reasonably be accommodated in a single species. If Lomekwi represented something different, we would have to be very sure that our analysis was faultless before we redrew the family tree. We were fortunate that both A. afarensis and A. africanus have very large known samples. It would be relatively straightforward to take a series of measurements, analyse these statistically, and see if the Lomekwi skull fell within the range of variation seen in either one of these species.

  Fred Spoor compiled a large database of all the relevant characters and measurements of each specimen that we had examined in Ethiopia, South Africa, and Kenya. Along with the digital images we had taken of many of the relevant fossils, this database enabled us to assess the morphology of the new fossil skull character by character. To our delight, the data showed the Lomekwi skull to be significantly different from all australopithecines in a number of ways.

  Most incongruous was the flat face of our new Lomekwi skull, and we soon found ourselves calling it Flat Face. In particular, Flat Face had big deep cheeks extending from below the eye socket towards the teeth, which created a tall malar region. The snout was also less primitive looking in that it didn’t protrude nearly as much as that of the more apelike afarensis (in other words, its face was less prognathic). When we took a line from just below the lower rim of the nasal opening directly down to the central incisors and measured the angle between this line and the one formed by the sockets of the upper teeth, we found that Flat Face was less apelike and more humanlike (it showed less subnasal prognathism). In later hominins, including humans, this angle is much greater.

  But what did the differences in facial shape mean? Evolutionary changes in facial shape are driven largely by dietary requirements, and the unusually small teeth of our new fossil supported such a hypothesis. Our work at Lothagam indicated that new dietary opportunities had opened at the end of the Miocene. Although many herbivores then became grazers (needing bigger, thicker, taller, or serially developed teeth), others had evolved mixed feeding strategies, and some had remained browsers—this herbivore diversity in size and ecology was mirrored by the large number of different carnivores. How had early hominins responded to these new opportunities? If these new feeding niches could also be exploited by early hominins, we should not be surprised to find differences in the dentition and shape of the face among different lineages of hominins. The distinctive facial features of Flat Face and its small teeth could indicate a different diet than that of A. afarensis.

  There were additional differences between Flat Face and its contemporaries. Flat Face had terribly damaged teeth that had broken off and left only a few fragments in place, and these could tell us very little. However, there was one upper second molar that was complete enough to be measured although it had a large matrix-filled crack. After correcting for this crack, we found that this tooth was smaller than any other upper second molar we had measured for our database. To be sure, we rechecked our measurements on this tooth through a microscope and on computer images.

  The size of the opening of the ear was also very small, similar to what we had seen in the Kanapoi temporal bone of A. anamensis and Yohannes Haile-Selassie’s new anamensis male skull from Ethiopia. So besides being different from A. afarensis in the shape of the face and the size of its teeth, Flat Face also retained some primitive traits that had disappeared in A. afarensis.

  As we completed our comparisons, we had to de
cide where Flat Face belonged. The more we looked at the evidence, the more difficult it became to incorporate this specimen in any of the known species. A. afarensis was the most likely given its contemporaneous age, but Flat Face was clearly different. We felt that this was not a good fit, and we began to think of our skull as a new species. But how did Flat Face fit into the human family tree?

  If we were to name a new species, we would also have to decide if Flat Face belonged to a known genus. Naming a new genus would shake up the family tree far more than the announcement of a new species, and our working assumption had always been that Flat Face belonged to one of the four known genera: Ardipithecus, Paranthropus, Australopithecus, or Homo. The problem was that the data did not sit comfortably with this assumption, since, with only one exception, none of these genera included species with a similar long flat face. Flat Face did not show any of the strongly primitive characters of Ardi­pithecus, so it was easy to exclude it from that genus. And Flat Face had almost none of the specialised features of the robust australopithecines grouped together under Paranthropus. Paranthropus does have a flat face, but it is different: it looks as though the nose had been punched hard and pushed in so the face is “dished.” Plus, Paranthropus has huge megadont cheek teeth that could not be more different from the small teeth of the new fossil from Lomekwi. We could find no grounds for including it in this genus unless it could perhaps be an ancestral species for this group, which we considered unlikely because of its completely different facial morphology. The only possible place within an existing genus was either in Australopithecus or Homo.

  With one exception, Flat Face also lacked the specialised features associated with our own genus. Of all the skulls in the collections, Flat Face seemed to most resemble a specimen we had found in 1972, the skull called 1470 that I had the pleasure of putting together fragment by fragment at Koobi Fora with an infant Louise. This extraordinary fossil still remains very much an enigma. It had a very similar face with little facial or subnasal prognathism. At 1.9 million years (almost 1.5 million years younger than Flat Face), 1470 is considered to be an early example of Homo, and the remarkable likeness gave us great pause. Was the Lomekwi skull ancestral to 1470 and the Homo line? As we shall shortly see, the assigning of 1470 to Homo habilis is clouded in controversy as well.

  The only other available genus was Australopithecus. Should we then attribute Flat Face to Australopithecus since that was the only genus believed to have been living at that time, or should we emphasize the magnitude of the differences by naming a new genus? The problem with this was the absence of megadont cheek teeth that characterize the australopithecines. Unable to decide, Fred and I wrote two papers that argued for each of these solutions. Put this way, the answer was clear. The paper attempting to incorporate Flat Face as a species of Australopithecus was unconvincing whereas the alternative paper describing a new genus provided a logical argument. So we took the plunge and named both a new species and a new genus. We gave our new skull the name Kenyanthropus platyops: the flat-faced man from Kenya.

  Flat Face was splashed across the front pages of the world’s leading newspapers, and in spite of our trepidations, the paper was generally well received by our colleagues. The significance of the find was far-reaching. Just as Louis had predicted all those years ago and as I had long believed, diversity in the fossil record did indeed extend to at least 3.3 million years ago. Moreover, A. afarensis was not necessarily the ancestor of all later hominins. Later hominins may have had their ancestry in Kenyanthropus or Australopithecus or in some other as yet undiscovered genus or species. The few sceptics who disagreed with us were most concerned about the possible effects of the distortion of the skull on our assessment of its shape. The strongest disagreement came from colleagues such as Tim White, who firmly believes that human evolution was a single evolving lineage with no diversity at this point in time.

  Dissention in science is often a positive force that stimulates further research and new discoveries. Our discovery had upturned conventional wisdom and led to new questions. But in order to prove our conclusions and convince the sceptics, we needed additional—and undistorted—specimens. This, of course, is easier said than done and remains a challenge for the future.

  My family in 1958.

  Courtesy of the Epps family

  Me with Richard’s dog Ben and our two pet otters awaiting a lake crossing to Koobi Fora, ca. 1970.

  Bob Campbell, courtesy of the Leakey family

  Counterclockwise from right: Nzube, Richard, Kamoya, and a heavily armed and curious local herdsman stare at the complete skull KNM-ER 406 exactly as we discovered it on our memorable camel trip in 1969.

  Bob Campbell, courtesy of the Leakey family

  Richard and I inspect some of the many hundreds of fragments of 1470 in 1972. I spent hours gluing this specimen together, usually with Louise keeping cool in a basin of water at my feet.

  Bob Campbell, courtesy of the Leakey family

  Not wanting to miss out on the excitement in the field, I took Louise (left) and Samira with me to Koobi Fora within weeks of their births, and throughout their childhoods they regularly joined us in the field.

  Bob Campbell, courtesy of the Leakey family

  The camels proved to be highly contrary and single-minded, requiring a considerable amount of coaxing and coercing in 1969. Much to Nzube’s relief (and ours) we soon substituted them for a more reliable, mechanical conveyance.

  Bob Campbell, courtesy of the Leakey family

  The enigmatic Flat Face, at last cleaned of matrix, presented us with a set of completely new features at an early stage of human evolution, which led us to announce a new genus and species, Kenyanthropus platyops, in 2001.

  Bob Campbell, courtesy of the Leakey family.

  Before satellite technology transformed our means of communication, Kamoya needed all the equanimity he could muster as he spent many frustrating hours trying to make radio contact with Nairobi through the control tower.

  Bob Campbell, courtesy of the Leakey family

  Before the advent of GPS, I spent an inordinate amount of time on elevated plateaus trying to orient myself against our aerial photos.

  Courtesy of Louise Leakey

  At Kanapoi, and later at Lomekwi, we used “hill crawls” on particularly prolific areas to localise our sieves and excavations around points where we found hominins. We recovered both a deciduous molar and a capitate this way—the very bones we needed the most.

  Courtesy of Louise Leakey

  Fossil hunting is an acquired skill. In addition to perseverance and a large element of luck, it is essential to have a mental template of the various bones. We found a far greater variety of fossils after I taught the team about the morphology of modern mammals, which expanded their search-image library.

  Courtesy of Louise Leakey

  Our scenic lakeshore campsite at Lomekwi came to an ignominious end on the very first night in 1998 when a heavy storm upended all our tents, including the all-important store tent.

  Courtesy of Louise Leakey

  At Kanapoi, we conducted some of our most extensive and time-consuming sieves of my career, moving mountains of dirt by dental pick and paintbrush, and checking all of the sieved soil for fragments of fossil bone.

  Courtesy of Louise Leakey

  Back in the lab in 2000, Fred Spoor and I compare our specimens to the casts of other hominid finds to ascertain what we had found.

  Courtesy of Louise Leakey

  Louise refuels the plane at Ileret with help from our camp manager, Mutuku, ca. 2011.

  Mike Hettwer / hettwer.com

  The field crew relax in whatever shade they can find during the hottest part of the day before prospecting again in the late afternoon. Because ambient temperatures fall so little at night, Turkana is one of the hottest places on earth.

  Mike Hettwer / hettwer.com

  An example of early aerial photographs with fossil finds marked on them. Before GPS, a pinprick with a
n associated number written as small as possible on the back was the only way to map the location of our discoveries. All of this data has now been digitised, which makes it considerably easier to analyse.

  Courtesy of Meave Leakey

  In 1951, Mary Leakey painstakingly traced the evocative rock paintings from 186 sites in Tanzania, and thirty years later, I was privileged to help her compile a book of the lifelike images and scenes depicted in this ancient art. In this image, figures excitedly dance and cavort around an elephant they have caught in a trap: a poignant image of daily life that gives a unique window into the behaviour of our ancestors.

  Courtesy of Meave Leakey

  The construction of permanent facilities by the Turkana Basin Institute dramatically altered our ability to conduct long-term fieldwork. The Turkwell campus, pictured here, is situated on the banks of the Turkwell River.

  Courtesy of Louise Leakey

  Once Louise completed her studies, she joined me in leading the Koobi Fora Research Project. She proved an invaluable addition as well as a welcome companion, as shown here in 2002.

  Courtesy of Josephine Dandrieux

  Samira’s daughter, Kika, together with her cousins Seiyia and Alexia, illuminated for me the vital (and highly enjoyable!) role played by grandmothers in human evolution. A cornerstone of our modern reproductive strategy is to have grandmothers help rear the next generation, which allows for a shorter interval before the next child.