About 800 to 950 thousand years (ka) ago the earliest human colonisers of northern Europe, both adults and children, left footprints and stone tools in sedimentary strata laid down by a river system that then drained central England and Wales. The fossil flora and fauna at the Happisburgh (pronounced ‘Haze-burra’) site in Norfolk suggest a climate that was somewhat warmer in summers than at present, with winter temperatures about 3°C lower than now: similar to the climate in today’s southern Norway. At that time the European landmass extended unbroken to the western UK, so any hunter-gatherers could easily follow migrating herds and take advantage of seasonal vegetation resources. These people don’t have a name because they left no body fossils. A group known from their fossils as Homo antecessor had occupied Spain, southern France and Italy in slightly earlier times (back to 1200 ka). Since the discovery of their unique mix of modern and primitive traits, they have been regarded as possible intermediaries between H. erectus and H. heidelbergensis – once supposed to be the predecessor of Neanderthals and possibly anatomically modern humans (AMH). Since the emergence about 10 years ago of ancient genomics as the prime tool in examining human ancestry the picture has been shown to be considerably more complex. Not only had AMH interbred with Neanderthals and Denisovans, those two groups were demonstrably interfertile too, and a complex web of such relationships had been pieced together by 2016. But there has been a new development.
Population geneticists at the University of Utah, USA, have devised sophisticated means of making more of the detailed ATCG nucleotide sequences in ancient human DNA, despite there being very few full genomes of Neanderthals and Denisovans (Rogers, A.R. et al. 2020. Neanderthal-Denisovan ancestors interbred with a distantly related hominin. Science Advances, v. 6, article eaay5483; DOI: 10.1126/sciadv.aay5483). In Earth-logs you may already have come across the idea of the ancestral ‘ghosts’ that are represented by unusual sections of genomes from living West African people. Those sections seem likely to have resulted from interbreeding with an unknown archaic population – i.e. neither Neanderthal nor Denisovan. It now seems that both Neanderthal and Denisovan genomes also show traces of such introgression with ‘ghost’ populations during much earlier times. The ancestors of both these groups separated from the lineage that led to AMH perhaps 750 ka ago. Rogers et al. refer to the earliest as ‘neandersovans’ and consider that they split into the two groups after they entered Eurasia, at some time before 600 ka – perhaps around 740 ka. This division may well have occurred as a result of a population of ‘neandersovans’ having spread over the vastness of Eurasia and growing genetic isolation. The reanalysis of both sets of genomes show evidence of a ‘neandersovan’ population crash before the split. Thereafter, the early Neanderthal population may have risen to around 16 thousand then slowly declined to ~3400 individuals.
However, the ‘neandersovans’ did not enter a new continent devoid of hominins, for as long ago as 1.9 Ma archaic H. erectus had arrived from Africa. Both Neanderthal and Denisovan genomes record the presence of sections of ‘super-archaic’ DNA, which reflect early interbreeding with earlier Eurasian populations. Indeed, Denisovans seem to have repeated their ancestors’ sexual exploits, once they became a genetically distinct group. From the ‘ghost’ DNA fragments Rogers et al. conclude that the ‘super-archaics’ separated from other humans about two million years ago. They were descended from the first ‘Out-of-Africa’ wave of humans, represented by the fossils humans from Dmanisi in Georgia (see First out of Africa, November 2003 and An iconic early human skull, October 2013 in Earth-logs Human evolution and migrations). A measure of the potential of novel means of analysing available ancient human DNA is the authors’ ability even to estimate the approximate population size of the interbreeding ‘super-archaic’ group at 20 to 50 thousand. Long thought to be impossible, it now seems possible to penetrate back to the very earliest human genetics, and the more DNA that can be teased out of other Neanderthal and Denisovan fossils the more we will know of our origins.
See also: Gibbons, A. 2020. Strange bedfellows for human ancestors. Science, v. 367, p. 838–839; doi:10.1126/science.367.6480.838