One of my favorite paleoanthropological sites is Dmanisi, in the Republic of Georgia. It is the oldest securely dated hominid site outside Africa (just under 1.85 million years ago), and the hominids found there display a neat mix of primitive Homo habilis and derived H. erectus features. I consider myself lucky to have had the opportunity to excavate at Dmanisi last year, and to return to Georgia (lamazi Sakartvelo! [I hope I translated that correctly]) for more fieldwork next month.
Recently, Reid Ferring and others (2011) described the results of excavations of M5, a section of the site a bit aways from the area where the hominids were found. M5 is pretty cool because it presents a nice geological “layer cake,” as Ferring described it to us: each of the strata (different layers of deposition) are nicely and evenly stacked on one another. Check out the labeled layers on the right of the figure, from Ferring et al. 2011:
This is in stark contrast to the jumbled strata (like ‘spaghetti’) where the hominids were found. In geology and archaeology, there is a general “law of superposition,” which states that the lowest layers in a sequence would have been deposited earlier than the layers above them. The A sediments at Dmanisi, as seen in the figure above, are thus older than the Bs. Hominids have only been found in the B sediments. But work at M5 has shown that stone tools are found in the older A sediments, meaning that hominids arrived at the site and used it continually, beginning just after 1.85 million years ago.
Tools from the site differ between the older A and slightly later (still older than 1.75 million years!) B sediments in both material and manufacture. As they say in the paper (p. 2/5), a major difference in tool manufacture between the strata A and B occupations could be that during the earlier A times, “either cores were more intensively reduced or selected flakes were made elsewhere and carried to the site.” I’m not sure why this may be, but it is neat that within a fairly narrow time span, researchers can see habits change in our early ancestors.
The authors also note that the older tools from A sediments indicate “that Eurasia was probably occupied before Homo erectus appears in the East African fossil record” (from the paper’s abstract). If only hominids also came out of the A sediments! The News is touting this as meaning H. erectus evolved in Eurasia and then some members of the ‘new species’ moved back into Africa, but I don’t think this is necessarily the case. The Dmanisi hominids are described as H. erectus, but lack some key H. erectus apomorphies (most notably a large brain size) and really look pretty similar to contemporary hominids in Kenya (such as KNM-ER 3733) and Tanzania (such as OH 16). Plus, the E. African hominid fossil record around 1.9 million years ago leaves some tantalizing hints at hominids more erectus-like than habilis-like, such as the ER 2598 occipital fragment.
So while Dmanisi definitely demonstrates the presence of hominids outside Africa earlier than most well-accepted “Homo erectus” (or “ergaster”) fossils in E. Africa, I don’t think they necessarily indicate that the species arose in Eurasia. Rather, what the fossil record likely shows is the evolution of populations of early Homo, in Africa and Eurasia, toward the more ‘advanced’ H. erectus we know and love (due to gene flow w/in a widespread species, rather than parallel evolution of similar traits in different species).
Reference
Ferring R, Oms O, Agustí J, Berna F, Nioradze M, Shelia T, Tappen M, Vekua A, Zhvania D, & Lordkipanidze D (2011). Earliest human occupations at Dmanisi (Georgian Caucasus) dated to 1.85-1.78 Ma. Proceedings of the National Academy of Sciences of the United States of America PMID: 21646521
Lawn Chair Anthropology 
I'm always interested in Dmanisi. Mostly because one of my Professors was involved in the initial description of the Dmanisi mandible back in the mid-nineties, which by the way was found by some Archeologists from the University of Mainz (the one where I study). Basicly this is the "we did something great too!" -story, we get told every now and then over here when it comes to paleoanthropology.Anyways, it'd be interesting to see a bigger picture on the distritbution of those early homo populations across Africa and Eurasia at this specific time. Right now, there a some huge geographical gaps, so I don't think that there's enough evidence for making any kind of assumption on how those early populations are related to each other and so forth.Have some fun while your're over there! How did you get this position by the way?
Yeah, it'd be great to fill in the gaps between Georgia and Egypt. Or the Rift Valley and Egypt, for that matter. I think it's funny that maybe the most common discussion in paleo fits the pattern of "the study counters the common belief that [insert species name] evolved in [place/time] instead of [another place and time]. Not that it's not important, but just there are neater questions out there.I got invited to Dmanisi because I was a teaching assistant for one of the organizers of the field school. When I found out he was working on that I asked if he needed any help, and I guess I did a well-enough job as his TA before that he said yes.