Surprisingly young age of the Homo naledi fossil site


Update #2 to Human Origins: How diet, climate and landscape shaped us

The age of the fossil-rich Homo naledi fossil site within the Rising Star cave system in the Cradle of Humankind World Heritage Site near Johannesburg, South Africa was frustratingly unknown until recently. The exceptionally large number of fossil bones scattered on the cave floor proved difficult to date, but an international team of scientists has applied several dating techniques (optically stimulated luminescence, uranium/thorium and palaeomagnetism) to constrain the age of the deposit to between 414 and 236 thousand years old (Dirks and others, 2017). This age is supported by the estimated age of several fossil teeth from the deposit, independently dated to between 335 and 183 thousand years old using a combination of uranium series and electron spin resonance (US-ESR) methods. The US-ESR methods directly date the fossils, but rely on a number of complex model assumptions that result in a large amount of uncertainty. Although not terribly precise, these ages are considered fairly robust and are surprisingly young for fossils whose features (or traits) suggest that they are much older. Prior to the age determinations, many estimated that the fossils, given their mix of australopith and early Homo features, would date to around when our genus Homo first appeared in the fossil record between 3 and 2 million years ago (Ma). How, then, is such a young age explained for fossils that retain so many old features?

A mosaic of features

One of the striking aspects of the Homo naledi fossils is the odd mix or mosaic of features they display. The amazingly rich fossil find from the Rising Star cave system includes a total of over 1500 bones from the Dinaledi Chamber alone (Berger and others, 2015), with more bones recently discovered in the separate Lesedi Chamber (Hawks and others, 2017). All of the bones recovered from both caves are considered to belong to Homo naledi and comprise a minimum of 15 individuals. Such a large number of bones provides a fairly complete skeleton of Homo naledi who stood 1.4-1.6 m high and weighed 40-55 kg. One of the most notable features is the small size of the skull, having an interior volume (endocranial capacity) of between 460 and 610 cc. This range in volume is based on three skulls and is intermediate between the mean skull size of the australopiths and the earliest Homo species for which skulls are available that date to around 2 Ma. Such a small skull suggested that Homo naledi represented one of the earliest members of our genus Homo, which branched from the australopiths around 2.8 to 2.3 Ma, based on fossil teeth and jaws from East Africa (Villmoare and others, 2015). In contrast, its foot shares many features similar to ours. Comparing Homo naledi to other species is difficult to do because for most other species there are simply not enough fossil bones to compare. However, overall, Homo naledi much more closely resembles early Homo (H. habilis and H. erectus) than it does us (H. sapiens) or our predecessor species (‘archaic’ H. sapiens). The close resemblance to early Homo suggested to many that the age of the deposit would be similar in age to when early Homo appeared circa 2.5 Ma and not the reported age of less than 0.5 Ma.

Hawks et al 2017 H naledi skull   Hawks et al 2017 H naledi endocranial LES1

Skull of Homo naledi (LES1) from the Lesedi Chamber (scale bar 5 cm) and the digital reconstruction of the endocranial volume of 610 cc, scale sphere is 10 mm (Hawks and others, 2017).

The most likely explanation is that Homo naledi represents one of the earliest members of our Homo lineage and that it managed to retain many features of early Homo up until at least 414 to 236 thousand years ago. While retaining many of its early features, it also appears to have acquired features that closely resemble later features, which either evolved independently (convergent evolution) or were acquired through interbreeding (hybridization) with other, later-evolved Homo species (but probably not our species, which only appeared by around 200 to 150 thousand years ago).

What is remarkable is that Homo naledi persisted for so long in a region occupied by other, later-evolved Homo species. For example, the skull from the Florisbad fossil site is thought to represent our predecessor species and was likely contemporary with Homo naledi, the two living within several hundred kilometres of one another. One possible explanation of their co-existence is that they occupied distinctly different habitats. Although their feet and aspects of their hands are similar to ours, Homo naledi’s fingers are curved. Curved fingers suggest that they were adapted for living in and moving about in trees. Therefore, they may have resided within heavily treed habitats, such as forest canopies, whereas our predecessor species was living primarily in more open grassland and savannah habitats. Although such niche partitioning may explain the co-existence of different species of Homo, it is remarkable that a species with so many early features, including such a small brain, managed to survive until just prior to when our species Homo sapiens evolved onto the scene.

Hangers on

Homo naledi is not the only member of our Homo lineage who managed to persist over such a long period of time. Homo erectus, for example, managed to survive in Asia long after they had become locally extinct in Africa, surviving up until around 300 thousand years ago in China and Java. And, in many respects, the suite of unusual features of Homo naledi reflect those of Homo floresiensis, the ‘hobbit”, which also retains features of early Homo (H. habilis and H. erectus) and lived up until just 50 thousand years ago on the island of Flores (Sutikna and others, 2016). The persistence of H. floresiensis on the somewhat remote island of Flores seems more plausible than a group within the African continent, but perhaps this reflects the fact that groups in Africa were adapted to specific habitats that effectively isolated them from other groups. Population densities were likely low and, along with the diversity of habitats, may have facilitated the survival of earlier groups for long periods of time. More recently evolved species might have been widely dispersed in part because of their big brain, sophisticated tools and control of fire, but were perhaps spread thinly enough over the landscape to have permitted pockets of earlier evolved groups to hang on. What the hobbit and Homo naledi seem to indicate is that among the more recently evolved members of our lineage, older groups managed to persist, either within distinct habitat or niche holdouts  ̶  perhaps culturally as much as physically isolated from other groups. The fossil record is so limited it may have hidden from us or we may have tended to underestimate the amount of variability in features, such as brain and body size that existed in the past.

Homo naledi culture?

Typically an archaeological site has hundreds to thousands of stone artefacts but very few if any fossil bones of those who made the artefacts. And whether or not any bones of the makers of the stone tools are found, it is common to find the bones of other animals at many archaeological sites. Hence, the Rising Star cave sites present the most unusual case: many bones of Homo naledi not in association with any stone tools or other cultural artefacts, nor any other large animal bones. Hence, although we know a lot about what Homo naledi looked like, we have very little idea of what they made or what other animals they lived among.

This lack of context makes it difficult to know much about the habitat in which they lived and how they lived. They most clearly did not live in the deep caves where they ended up as fossils. It is conceivable that they fell into or were washed down into the caves through surface openings connected to deep cave chambers. But in that case we would expect other large animals to have fallen in or been washed into the caves along with them. Some have proposed that they were intentionally disposed of into cave openings and that this disposal may indicate a type of ritual burial (Berger and others, 2017). Ritual burial is a cultural behaviour that has so far only been associated with our species, with the earliest hints (mortuary defleshing) dating to around 160 thousand years ago, and proper burials with grave goods not until around 100 thousand years ago. It is possible they disposed of their dead into the caves as a form of good housekeeping, but it is hard to imagine that small-brained Homo naledi had the mental ability to practice ritual burials, and evidence to substantiate ritual disposal into the caves remains lacking.

The complete absence of stone tools makes it difficult to know what stone tools, if any, were used by Homo naledi. They possess the wrists and hands capable of making and using stone tools, but did they? The age of the site falls within the Middle Stone Age (MSA), a time of regionally diverse stone tool industries throughout Africa. Some have suggested that Homo naledi may have been the maker of some of these stone tools (Berger and others, 2017), but so far there is no direct evidence to support this idea. The appearance of stone tool use has been dated as far back as 3.3 Ma and associated with australopiths having brains similar to or smaller sized than Homo naledi, but these early stone tools are a far cry from the diversity and sophistication of MSA stone tools that existed throughout much of Africa by 300 thousand years ago. Hopefully future finds of Homo naledi in association with cultural artefacts will shed some light on where and how they lived.

Lomekwian tool           MSA tools across Africa

A 3.3 Ma Lomekwian stone tool (left; Harmand and others, 2015) and regionally diverse MSA stone tools throughout Africa 300-100 thousand years ago (right).

Further Reading

Berger. L., and others, 2017. Homo naledi and Pleistocene hominin evolution in subequatorial Africa. eLife 2017;6:e24234. DOI: 10.7554/eLife.24234

Berger, L., and others, 2015. Homo naledi, a new species of the genus Homo from the Dinaledi Chamber, South Africa. eLife 2015;4:e09560. DOI: 10.7554/eLife.09560

Dirks, P., and others, 2017. The age of Homo naledi and associated sediments in the rising star Cave, South Africa. eLife 6:e24231. doi: 10.7554/eLife.24231

Harmand, S., and others, 2015. 3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya. Nature 521, 310–318.

Hawks, J., and others, 2017. New fossil remains of Homo naledi from the Lesedi Chamber, South Africa. eLife 6:e24232. DOI: 10.7554/eLife.24232

Sutikna, T., and others, 2016. Revised stratigraphy and chronology for Homo floresiensis at Liang Bua in Indonesia. Nature 532, 366–369.

Villmoare, B., and others, 2015. Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia. Science 347, 1352–1355.

New ages from Jebel Irhoud, Morocco

Update #1 to Human Origins: How diet, climate and landscape shaped us

John S. Compton

The date

New fossil finds and a better constrained age for the Jebel Irhoud fossil site in Morocco shed new light on our evolution. The new ages indicate that the site is approximately 315 thousand years old (315 ka), or nearly twice as old as the previous estimated age of 160 ka for the site. The new, older age was obtained by applying the thermoluminescence dating method, which estimates the amount of time lapsed since stone flint tools found from the same layer as the fossil bones had been heated by fires made by those living at the site. The thermoluminescence method measures the amount of light given off by the stones tools (the amount of luminescence) as they are heated to higher temperatures. The idea is that the fires at the site were hot enough to completely erase any luminescence trapped in the stones. Once cooled, and so long as they were never exposed to the heat of a fire again, the stones gradually accumulated stored luminescence through time from natural radiation received from the surrounding sediment. The present-day background radiation dose of the deposit was measured from where the stone tools were recovered and was assumed to have remained more or less constant back through time. Heating of the stone flints in the lab causes the stored luminescence to be released, and the amount released could be measured. Dividing the total released luminescence by the radiation dose rate measured at the site allows an age to be calculated.

Luminescence dating, both thermoluminescence and optically stimulated luminescence (or OSL), has revolutionized our ability to date archaeological sites beyond what was previoulsy limited to the last 50 thousand years by the radiocarbon method. The drawback to luminescence dating is that it too has limits in how far back it can reach because the amount of luminescence that can accumulate is limited. Once the stones have acquired maximum luminescence they can acquire no more, making it difficult to apply the method to samples older than 200 ka. So, the new age determined at Jebel Irhoud by thermoluminescence in excess of 300 ka is remarkable.

The other drawback with the thermoluminescence method is that the dates are not nearly as precise as those obtained by the radiocarbon dating method. The weighted average age of 14 samples from the fossil-bearing layer at Jebel Irhoud is 315 ka, but this age is associated with a fairly large uncertainty, with a 68% probability that the age of the site is between 349 ka and 281 ka, and a 95% probability that the age of the site is between 383 ka and 247 ka. The age range of 383 ka to 247 ka is consistent with both the range in age of a human tooth recovered from the site determined by a different method known as electron spin resonance (a far more complicated method than thermoluminescence that I will not try to explain here), and the known age ranges of fossil animals found at the site. Therefore, although not terribly precise, the new age is a huge improvement on previous ones for the site and, as we shall see, the new age provides valuable insights into the interpretation of the Jebel Irhoud fossils and their possible connection to other fossil sites in Africa.

The fossils

The Jebel Irhoud site has yielded a rich human fossil assemblage comprising skulls, jaws, teeth, leg and arm bones from at least five individuals, including a child and adolescent. Although the original location of all the fossils is not known, they are all assumed to have come from the same layer at the site associated with the new date of circa 315 ka. The face, jaw and teeth are considered to be similar to our own, although the jaw and teeth are unusually large, and the size of the browridge is variable. The most significant difference appears to be in the shape of the skull, being lower in height and more elongate (less bulbous) than ours (see figure below). The features of the skull resemble those of other skulls of similar age, such as the 260 ka skull from the Florisbad site in South Africa. These differences in the shape of the skull suggest that those living at Jebel Irhoud, and more broadly throughout Africa from around 330 to 230 ka, were not yet fully us (Homo sapiens).

The line of descent in our Homo lineage is thought to be from early Homo species (Homo habilis, for example) to Homo erectus to Homo heidelbergensis to our predecessor species, and finally, to us Homo sapiens. As I discuss in my book, the range of variations within a species and the complexity of speciation in a large continent such as Africa, makes it difficult from the limited number of fossils available to delineate species clearly. Some experts, the ‘lumpers’, view the changes in our lineage as gradational and refer to the Jebel Irhoud fossils as ‘archaic’ Homo sapiens. Other experts are ‘splitters’ and argue for stepwise evolution of distinct species. I believe there are enough physical differences and culture differences (see below) to support an intermediary species in the evolution of H. heidelbergensis to H. sapiens, an intermediary species that would take the place of the lumpers’ archaic Homo sapiens. However, this intermediary species lacks a generally accepted species name, and so I refer to it as our ‘predecessor’ species.

To my view, the new age indicates that the fossils at Jebel Irhoud represent some of the earliest members of our predecessor species and not, as implied by many of the news stories, that our species Homo sapiens now has an age range that extends back to around 300 thousand years. The oldest yet recovered fossils of our species Homo sapiens, ones that represent anatomically modern humans (AMHs) remain those found in East Africa that date to between 200 ka and 150 ka. The earliest members of our species most probably descended from our predecessor species, who were widespread throughout Africa by around 280 ka as represented by the fossil skulls found at Jebel Irhoud in Morocco, Florisbad in South Africa and Laetoli and Ileret in East Africa.

predecessor skulls.pngFossil skulls dated between 315 and 260 thousand years old that are possibly representative of our predecessor species intermediate between H. heidelbergensis and H. sapiens (from left to right: Florisbad, Laetoli (Smithsonian Institution) and Jebel Irhoud (Natural History Museum, London)).

Herto 1.png   Herto 2.png  modern skull 1.png  modern skull 2.png

Fossil skulls of our species Homo sapiens from 160 thousand years ago (Herto, two images on left; photos by David Brill ( and modern (two images on the right). Our species is largely defined by its bulbous shaped skull.


Cultural differences

The arrival of our predecessor species is associated with the major transition from the Earlier Stone Age to the Middle Stone Age (MSA), a transition marked by smaller, more regionally-diverse stone tools. Jebel Irhoud is now the oldest site known having a direct association of fossils with MSA tools. MSA tools are widespread throughout Africa from around 300 ka to 230 ka, but most often not associated with human fossil remains. Some of the MSA stone points recovered from Ethiopia have been interpreted from their edge damage to have been used as thrown spears by at least 279 ka. The stone points from Jebel Irhoud have not yet been interpreted as having been thrown as spears (javelins). Those living at Jebel Irhoud were competent hunters based on the animal bones found. It may be that they made effective use of sharpened wooden spears without stone armatures, something their predecessor Homo heidelbergensis was doing.


spear tips.png    San throwing.png

Obsidian projectile spear tips from Ethiopia dating to at least 279 ka (Sahle et al., 2013) and a photo of a !Kung San throwing a spear (photo courtesy of Neil Roach).


The presence of burnt bones and charcoal suggests they had control of fire. However, they do not appear to have used fire to intentionally heat the stones (pyrotechnology). The earliest evidence for intentional heating of stones to improve their work-ability dates from 164 ka at the Pinnacle Point site in South Africa, presumably made by H. sapiens. Heating was probably not necessary in the case of the raw stone material available at Jebel Irhoud and the heating of about a third of the stone tools there was likely because they inadvertently ended up beneath where later fires were made.

The other cultural artefact to appear for the first time in the MSA is ochre, an iron-rich rock used for, among other things, symbolic body painting based on the specific collection of the reddest coloured stones. The earliest use of ochre is in East Africa, but no ochre has been reported from the Jebel Irhoud site. The absence of thrown stone-tipped spears and ochre at Jebel Irhoud may indicate that these cultural innovations were only developed later among groups of our predecessor species. Alternatively, these cultural items may have been present among groups in sub-Saharan Africa but were lost by groups too small to sustain these cultures after they had expanded into North Africa where soon became isolated from other groups.

ochre.png  ochre powder.png

Ochre stones and grinding ochre into a red powder.


Speciation events

Do the fossils at Jebel Irhoud suggest that the geographical region of origin of our predecessor species was North Africa? I argue in my book that the Maghreb, located at the northernmost tip of Africa was a potential geographical region of origin for species within our human lineage. This was based primarily on the periodic isolation of the Maghreb, separated from the rest of Africa and Eurasia by the Sahara-Arabian Desert and the Mediterranean Sea. Relatively small groups living in isolation in the Maghreb over long periods of time may have evolved away from other groups. These substantial physical barriers would have made exchange highly unlikely, except during relative brief periods when the Sahara-Arabian Desert ‘greened’ by receiving enough rainfall for the transformation of the desert into grassland and lakes.

These greening events may have made it possible for the periodic mixing of previously isolated groups living in the Maghreb, sub-Saharan Africa and Eurasia. We know the Sahara-Arabian Desert greened most recently 9 to 6 ka, with large impacts on the movement and interaction of people throughout the region. There is also good evidence of widespread greening 130 to 120 ka go associated with the movement of our species out of Africa into the Levant, and possibly into North Africa as indicated by the first appearance there of distinctive Aterian stone tools. What the groups represented by Jebel Irhoud fossils had evolved into by the time of the 130 ka greening event and to what extent they may have intermingled with Homo sapiens coming from sub-Saharan Africa is unknown. There were other greening events, such as the one associated with when modern people left Africa as part of the Great Expansion 60 thousand years ago, but the timing and extent of past greening events remains poorly known.

The interglacial period that occurs within the dated range of the Jebel Irhoud fossil site of 383-247 ka that is most likely to have had a significant greening event is Marine Isotope Stage (MIS) 9  ̶̶   roughly 330 ka. The fossils at Jebel Irhoud may represent early members of our predecessor species who evolved there among small, isolated populations of H. heidelbergensis and then spread out into the African continent with the greening of the Sahara associated with the MIS 9 interglacial period. Alternatively, our predecessor species may have evolved from H. heidelbergensis somewhere in sub-Saharan Africa and moved into the Maghreb during the MIS 9 or another greening event. Unfortunately, the age resolution of archaeological sites and number of sites are not sufficient to determine at this stage where in Africa our predecessor species evolved.

MIS 9.png

Ochre, javelins and diverse Middle Stone Age regional cultures are associated with the evolution of our predecessor species from H. heidelbergensis in Africa. The age range of 383-247 ka of the Jebel Irhoud site places it around the time of the MIS 9 Interglacial period when a greening of the Sahara may have allowed movement of groups living in North Africa and sub-Saharan Africa.


What we know from the dated MSA stone tool assemblages in North Africa, East Africa Rift Valley and South Africa is that our predecessor species was widespread throughout Africa by around 280 ka, but from which of these areas they evolved into us Homo sapiens remains unresolved. In my book I suggest that the currently available information favors South Africa as our species’ geographical region of origin  ̶̶  check it out and see if you agree.


Further Reading

Richter, D. et al., 2017. The age of the Jebel Irhoud (Morocco) hominins and the origins of the Middle Stone Age. Nature

Hublin, J.-J. et al., 2017. New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiens. Nature

Stringer, C. and Galway-Witham, J., 2017. On the origin of our species. Nature 456, 212-214.

Gibbons, A., 2017. Oldest members of our species discovered in Morocco. Science 356, 993-994. [doi: 10.1126/science.356.6342.993]

Compton, J.S., 2016. Human Origins, How diet, climate and landscape shaped us. Earthspun Books,

Sahle Y., Hutchings W.K., Braun D.R., Sealy J.C., Morgan L.E., et al., 2013. Earliest stone-tipped projectiles from the Ethiopian Rift date to >279,000 years ago. PLoS ONE 8(11): e78092. doi:10.1371/journal.pone.0078092.