By Phil

In my absence a lot has occurred in researching African substructure since the Ballito boy paper a while back.

I’ve somewhat touched upon this subject previously in two relevant articles, one of which I will provide a short update on regarding “cranks” in population history, including Bruce Fenton, in the future.

The structure of this article will be outlining the general and particular ancestral groups linked to modern Sub-Saharans, briefly putting them in context of OOA, and attempting to approximate them in the form of fossilized specimens. The ancestry analysis will start from the most recent data to the most ancient data discussed (that is shortly before the Holocene towards the branching of H. sapiens from other relatives from H. Ergaster from about 0.5-1 mya) as it will only be more complex from then on.

The first layer will be the finding from the Mahgreb remains that date about 18k, showing profile that is approximately two thirds Near Eastern hunter gatherer, 1 third Sub-Saharan but showing no particular affiliation to a modern sample. It is only slightly shifted towards the Hadza, perhaps suggesting the ancient Eastern African cluster that preceded Basal Eurasian. This shows two things, one the antiquity of back migration to Africa and the oldest DNA sample that is directly linked to modern Sub-Saharans. This could possibly be part of a previous human culture, which may have been part of the migration that lead to modern West Africans. I, however, hesitate to suggest they are direct ancestors. It provides a rough portrait of ancient North Africa nonetheless, a location I particularly figured to be relevant for clues on modern sub-saharans given the different geography of North Africa at the time of the “Wet Sahara“.

The next finding is the break-down of DNA among Western (Yoruba and Mende), Eastern (Nilotic), and Southern (Khoi-San) Africans. The findings show that, in accordance to the Ballito Boy study, Western Africans are a mixture of Ancient East Africans and Basal African Sapiens, Khoi-Sans are derived from the basal Branch at around 250k-350k, and Nilotics are mostly of the 70k-80k branch. However, what it shows is that these populations exchanged genes, particularly between Eastern and Southern Africans. In the case of West Africans, the Yoruba more so then the Mende. I suspect that this could explain the Igbo from the African Genome project showing signs of hunter-gatherers that are more like the Khoi-San than Pygmies at low percentages.

( 1/22/20 Edit: This recent study breaks it down wih Hunter Gatherers having roughly 4% while farmers have 5.8% of a similar AMH species that diverged slightly after neanderthals. Kabwe may be the best comparison).

Archaic DNA is where it gets interesting however. I’ve already discussed the 140k-150k admixture event that lead to modern variation of MCU7, but two canididates have came up. One, predictably by this point, is an archaic human ancestry at about 8% in Yoruba, using the same methods to detect 2% archaic DNA in pygmies, dated around 460k-540k , roughly after the divergence of Sapiens from Neanderthals based on another study. The first study however distinguished it from that of the Pygmies, which seems to be older compared to the estimates in this study. The dating of the Pygmy’s admixture is also set around 30k, while the same study dates that of the average Sub-Saharan at 9k. The TMRCA is place at around a median of 1.0 mya based on the overall candidate loci, this seems to align nicely to the age of the MCU7 phylogeny in Africans compared to that of Eurasian Archaics. This is however older than the admixture found in Hammer’s study, suggesting late Erectus. This can be resolved by the conclusion of Hsieh’s study. That is, archaic admixture in Africa was found to be weak, though continuous. This could mean that while penetrating new ecosystems, Hominid interaction what short but often, accumulating alot of layers in small degrees, preserved due to advantageous traits as seen in Neanderthal’s immune system genes. These genes were in turn a trait of the common ancestor that Sapiens lost.

It is therefore possible that archaic admixture could contain different archaic admixture within it based on this behavior. A previous study’s results of loci with ages similar to that of the Neanderthal divergence in hunter gatherers supports this, being somewhat intermediate between the MCU7 population and the Basal African. This particular one, however, may very well be tied to a larger finding.

A revised study finds archaic African admixture prior to the split, yet it isn’t African exclusive and seems to have occurred prior to the OOA split. It is only a preprint, but it aligns with what we may have assumed given how long archaic and modern linegaes have mixed. However, as opposed to the continuous flow inferred from pygmies, this particular event was strong and rapid. The researchers likewise note on page 4, though against models of an TMRCA after that of Neanderthals, do not rule it out as a scenario in African substructure. This acknowledges the point made by Razib awhile ago that substructure in Africa reflects the complex development of hominids there. This even pertains to even the Sapiens specific line.

I will now briefly outline the following fossils relevant to modern African population substructure. At the most diverge, early homo found at the Ishango area. I would like to mention that I doubt it mated with Sapiens directly, as the sediment analysis would suggest. Rather, I believe this specimen would genetically contain the variant of genes multiple studies have found to be under selection in African hunter-gathers. The Late Erectus found in Hammer or the Rhodesiensis-like specimen in the initial ghost population models for the Yoruba and Mende I believe were the actual populations that interacted with Humans while possibly carrying these genes.

Late Erectus, or Ergaster, then is represented by Tighenif or Rabat. The latter species can be presented by Bodo or Kabwe. The Kabwe, however, seems to be within Erectus s.l based on internal anatomy than a Petralona equivalent to Sapiens as Bodo is suggested. That is, relative to Bodo or Florisbad, it lacks specializations towards Sapiens. The Ceprano studies however show it nonetheless to be more within the range of overall morphology towards Middle Pleistocene Hominids rather than Ergaster.

(Edit: The Post- LCA admixture could be represented by the two specimens above as well, and I would likewise add a Sapiens Intermediate as well, Florisbad. This status seems to have been replicated here, here, here, here and here. The similar Eliye Springs skull bridges the gap towards Sapiens.)

The next specimen would be the Ceprano skull, the best morphological node for the species Sapiens, Neanderthal, and Denisovans for their LCA. This is significant as it thoroughly rebukes Fenton by his first mistake, dismissing the morphological data in Western Eurasia and Africa. It is shown to be distant from Asian Erectus and seems to represent a “Homo Erectus Sensu Lato” that developed derived traits that would be continued into a trajectory seen in Africa. This is supported by Mounier’s analysis that hypothesized the LCA to be closer to African than European samples. Likewise, his analysis with Manzi shows Erectus/Ergaster tendencies in morphology in partiuclar areas rather than Neanderthal. Manzi also acknowledges a potential morphological link in Africa as well. Therefore we can attribute the LCA detected in DNA in Africans and non Africans by the revised preprint to the above specimen.

(Edit: Manzi has additionally, without confusion, located such an African ancestor to the Hiedelbergensis morphology of Ceprano at the Gombere site.)

Basal Humans, likely the migratory population 148k years ago, would be presented by the Iwo Eleru skull. Jm8 mentions of their relation to similar specimen and DNA sequences found. Likewise, they seem to be responsible for A00.

My previous article has previously mentioned the resulting phenotypic diversity in Modern Africans by way of their skulls. However, I will briefly touch upon some recent examples from my bin of sources.

Klasies River 100k : More so “modern” than Florisbad but shows no particular phenotype towards Khoisan.

Border Cave 100k-50k: Speculated once to be linked to modern Khoi-san, however modern studies shows that it is only apparent much later. General considered “Modern“.

Lukena Hill Crania 20k: Similar to late Pleistocene North Africans, rare morphology in modern samples. Best represents pre-holocene humans prior to gracilization along with 33k Nazlet Khater. The latter skull could likewise conform to Khoi-san ancestry prior to Gracilization in the Holocene.

Matjes River Crania 9k: Joins other Holocene skulls displaying Khoisan morphology.

Nakuru IX: Odd skull with not much literature, somewhere from southern Africa and dated around 17k, yet aligns with Bantu. This book, which provides a very close continuum in morphology that I’ve been following, dates it as Holocene with some Khoi-san samples. This suggest that at some point it was re-grouped.

Concluding remarks, I have left out alot of other significant details that could give direction in what to investigate in future findings. I encourage that readers go over the original articles themselves to notice other significant findings, such as the relevant positively selected features associated with them. I, may likewise, touch upon the sequences of morphology, paleo-environments, and archaeology in the future. For the time being, we have acquired both game-changing evidence of ancient substructure and a refined continuum of homo evolution in recent years.