The new Dmanisi skull

Screen Shot 2013-10-23 at 7.24.41 PM
Georgian National Museum

The skull of an early Homo individual, discovered in Georgia in 2005 and presented on October 18 in the scientific journal Science, is truly a wonderful find. Not only was it recovered relatively intact and complete, but together with four other cranial fossils found at the same site, it constitutes a unique window into how much early Homo individuals could vary from one another.

  • Dmanisi site and fossils

The Dmanisi site is located in Georgia and is the earliest known hominin site outside of Africa. Argon isotope dating indicates that the site was occupied between 1.77 and 1.85 million years ago. So far, Dmanisi has been a real treasure trove for paleoanthropologists: besides the new skull described in Science this week, the site has yielded four other relatively complete crania, in addition to other skeletal fossils and artifacts such as crude stone tools.

What makes Dmanisi really special for the human fossil record is that it provides researchers with a set of fossils from the same time and place, creating a unique window into the biological variation of early Homo individuals. By contrast, most African fossils of about the same age (1.8 million years old) come from different sites and from a larger period range.

The five relatively complete Dmanisi skulls represent five individuals of different sex and age (adolescent, adult, late adult). From the context in which they were found (site, geological strata, other animal fossils, etc.), it is thought that these individuals were dragged down in the den of carnivorous animals over a time period of a few hundred centuries. On the whole, the Dmanisi skulls have features similar to those of contemporaneous African fossils that have been attributed to the species Homo habilis and Homo erectus. Some of their cranial features also resemble H. erectus specimens found in East Asia.

  • “Skull 5” and what it tells us

The complete skull now described in the Science article by Lordkipadnize and colleagues, called “Skull 5”, is made of a cranium discovered in 2005 (called D4500) and its associated mandible discovered earlier in 2000 (D2600). From its morphological features, the researchers describe Skull 5 as a male, with a large and prognathic (forward-projecting) face, small brain and heavily worn anterior dentition indicating it might have been used for activities other than chewing, such as gripping.

With a brain volume estimated at 546 cm3, Skull 5 is slightly above the range for the earlier genus Australopithecus but at the lower end of the estimated variation for H. habilis (modern humans have an average brain volume of 1350 cm3). The researchers also estimated the stature and body mass of the individual from other skeletal fossils probably associated with Skull 5, placing it at 146-166 cm and 47-50 kg, which is within the range of variation known for early Homo.

Given its completeness, Skull 5 provides the first evidence of how the face (including the mandible) of adult early Homo was oriented and positioned relative to the braincase. Until now, the anatomical connections between the neurocranium (braincase) and the face in adult individuals were a matter of debate, as most specimens found in Africa and East Asia were incomplete, and the more complete fossils were those of adolescent individuals (who have more orthognathic faces, that is, not as much forward-projecting, and lightly built superstructures compared to adults). Skull 5 now gives the complete picture of an adult male with a relatively large and prognathic face.

Interestingly, the researchers also found some similarities between the maxilla of Skull 5 and that of the 2.33 million-year-old early Homo specimen found in Hadar, Ethiopia. The Hadar specimen has been proposed as the oldest Homo specimen, but Australopithecus sediba (found a few years ago in South Africa and dated as 1.98 million years old) has also recently been proposed as a potential ancestor of the Homo lineage. The authors of the present Science paper argue that the affinities existing between the Hadar fossil, other early Homo specimens from Eastern Africa, and now Skull 5, go against the latter scenario. However, paleoanthropologist John Hawks suggests in a post blog that most of the shared features between Skull 5 and the Hadar specimen may also be shared with the South African Australopithecus sediba specimen (which was actually not included for comparative analysis in the Science article).

  • The Dmanisi fossils, “normal” variation, and consequences

As interesting as Skull 5 can be on its own, the Science article also looks at the variation observed among the five relatively complete skulls that have been recovered at Dmanisi so far, and at what it may mean for the way paleoanthropologists have been considering other early Homo specimens in Africa. This is the part that has garnered the most attention from the press (and that has perhaps been hyped a bit too much in the process).

Lumpers and splitters

It all comes down to the difficulty of knowing whether the morphologic variation one observes when looking at two different fossils represents interspecific variation (the two fossils are from different species) or intraspecific variation (the two fossils are individuals from the same species, who may differ for reasons such as age or sex, or because they belonged to populations existing in different places or at different times). It is especially an issue when the fossils one tries to compare have been discovered in different places and/or come from slightly scattered time periods (not to mention the fact that most fossils are only fragments). As a consequence, paleoanthropologists may use different approaches in their analyses, which in turn may lead to different conclusions: some paleoanthropologists may favor “lumping” (considering the fossil specimens as only one species and attributing variation to inter-individual/inter-population differences), while others may favor “splitting” (considering the fossil specimens as different species and attributing variation to inter-species differences).

Dmanisi: one or more species?

As I have said earlier, the Dmanisi site is unique in that it has yielded five relatively complete skulls that come from a single locality and were deposited within a rather narrow range of time. It therefore provides a unique window into what the normal variation in populations of early Homo may look like. Indeed, the single time and place makes it difficult to argue for the presence of multiple species (though still possible). The researchers used a computer-based analysis of morphometric parameters to compare the range of variation found among the five Dmanisi skulls with the range of variation observed in living chimpanzees and in living humans, and found that the shape variation among the Dmanisi skulls was approximately equivalent to the shape variation existing among living chimpanzees and among living humans. They therefore concluded that the variation observed in the Dmanisi fossils could very well represent normal variation within a single species (H. erectus), with most of said variation being due to variation in age and sex among the fossil specimens.

H. habilis, H. erectus, H. ergaster, H. rudolfensis: over-taxonomizing?

Fossil specimens of early Homo found at different sites in Africa and roughly contemporaneous of the Dmanisi fossils (about 1.8 million years old) have been categorized in four different species: H. habilis, H. erectus, H. ergaster, and H. rudolfensis. However, it turns out that the variation in these African early Homo specimens is within the range of the variation observed at Dmanisi. The authors of the Science study therefore suggest that the morphological diversity seen in the African fossil record around 1.8 million years ago may rather reflect variation between populations of a single evolving lineage, H. erectus, rather than represent multiple paleospecies.

The hypothesis of a single lineage rather than several species branching off has been proposed before. For example, in an article published in the scientific journal Evolution earlier this year, Van Arsdale and Wolpoff suggested that the variability present in the African early Homo fossil record could actually reflect a lower degree of variability in any time slice coupled with evolution over time, and therefore represent a single evolving lineage rather than multiple species.

So, in light of the new Dmanisi set of fossils, it seems that paleoanthropologists may have previously been “over-taxonomizing”, that is, relying too much on species identification and inter-species variation to account for the variation observed in the early Homo fossil record. Maybe the African early Homo specimens (or at least some of them) should all be lumped into one species. However, as the authors of the Science article themselves point out, thinking of the different early Homo as different populations within a single H. erectus evolving lineage rather than as different species does not overturn the current understanding of human evolution, contrary to what some titles in the press would have you believe. They write:

“The hypothesis of phyletic evolution within a single but polymorphic lineage raises a classificatory but not evolutionary dilemma”.

Of course, this would not make for a catchy title, now, would it?


If you want to know more about the Dmanisi find and what it means, from the point of view of experts in the field, I can recommend the two following blog posts:
– John Hawkes:
– Adam Van Arsdale:

1. A complete skull from Dmanisi, Georgia, and the evolutionary biology of early Homo. Lordkipanidze D, Ponce de León MS, Margvelashvili A, Rak Y, Rightmire GP, Vekua A, Zollikofer CP. Science. 2013 Oct 18;342(6156):326-31. doi: 10.1126/science.1238484
PMID: 24136960
2. A single lineage in early Pleistocene Homo: size variation continuity in early Pleistocene Homo crania from East Africa and Georgia.
Van Arsdale AP, Wolpoff MH. Evolution. 2013 Mar;67(3):841-50. doi: 10.1111/j.1558-5646.2012.01824.x
PMID: 23461332

ResearchBlogging.orgLordkipanidze D, Ponce de León MS, Margvelashvili A, Rak Y, Rightmire GP, Vekua A, & Zollikofer CP (2013). A complete skull from Dmanisi, Georgia, and the evolutionary biology of early Homo. Science (New York, N.Y.), 342 (6156), 326-31 PMID: 24136960


6 thoughts on “The new Dmanisi skull

  1. thereviewer March 3, 2014 / 3:27 pm

    You’ve covered this story really well here. It’s a really interesting development and hopefully will serve to concentrate efforts on uncovering more fossils of this time period to solve the conundrum!

    • Aurelie March 4, 2014 / 12:31 pm

      Paleoanthropology is not my area of expertise, but I was at the time following a MOOC on human evolution and I really got into it. I just saw the new development of the story on your blog, that is, that there may be two lineages present at Dmanisi, interesting!

      • thereviewer March 5, 2014 / 4:20 am

        It’s a really captivating subject area isn’t it! The only drag is that there is always debate and argument, rather than accepting these finds for what they are, marvellous! I guess that’s what makes science a healthy discipline and maintains public interest!

      • thereviewer March 6, 2014 / 4:15 pm

        Well you could have fooled me, you wrote that like a pro! Was that the John Hawks MOOC? It sounds/sounded (?) good but I didn’t hear about it until the other day, I presume it’s a little too late to sign-up…

      • Aurelie March 7, 2014 / 3:44 pm

        No, it was last semester, a MOOC led by Adam Van Arsdale (via edX). I did not know about the one by John Hawks, maybe you should sign up anyway, even if you don’t have time to finish the course it may be interesting anyway.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s