An Investigation into the "Mysterious" Mesolithic Maghrebi populations

The present discussion takes off from the trail of a previous one posted here on the examination of the "Mechta-Afalou" and the European "Cro-Magnons" as elements of a monophyletic unit constituting a taxonomic "type". That previous discussion had put a spotlight on the futility of such a thesis, which varied quite widely in its specifics as one went from the subjective taxonomic regime of one researcher to the next. In that same discussion, the following was noted:  "the Mechtoid concept and its supposed relationship with Cro-Magnon, is nothing more than another Eurocentric way of trying to relate North Africa to Europe...essentially a *wishful* desire to see north Africa as more an extension of Europe than the actual continent [Africa] its attached to".

Indeed, there has been a long indirectly-stated desire to make the coastal northern piece of the African continent an extension of Europe, and territory which has traditionally been THE domain of European-descended peoples than of autochthonous Africans, which are generally treated as "newcomers" to the region. In other words, some sort of a "natural" apartheid was put in place in the late Upper Paleolithic until some time in the early Holocene, preventing autochthonous Africans from inhabiting the place, while the same region was supposedly hospitable to "European immigrants". This theme blatantly makes its reappearance in another study that has fairly recently enjoyed elevated circulation and has become a citation-favorite within online 'white' supremacist cult gatherings, undoubtedly attributable to the French-published study becoming more accessible as English-translated versions became more available on the net. The study in question comes from Kefi et al., under the heading: Mitochondrial Diversity of the Population of Taforalt (12,000 years b.p. - Morocco): A Genetic Study Approach to the Peopling of North Africa, 2005.

These authors make it a point to demonstrate that their Taforalt samples were virtually untainted by "sub-Saharan" African genetic infiltration, essentially presenting a scenario whereby the supposed Taforalt "foreign settler" community were living in a virtual world of a natural apartheid, whereby the authochthonous African populations for some odd reason, steered clear from the region at all cost, which as it turns out, was at least hospitable enough for a "foreign settler" community, but not Africans themselves. This is at least what these authors, and others like them, will have people buy into. One might as well say that there was a wall of a "security beam" network barricading the entire length of coastal northern Africa from autochthonous Africans, which would burn them if they dared cross to the "other side", hence preventing them from finding a haven in coastal northern Africa and keeping them at bay in the Sahel or "sub-Saharan" Africa. The authors tell readers that essentially all but a single individual or two has possible "sub-Saharan" African ancestry, saying in the abstract that:

Genetic and anthropological studies have revealed the complexity of the peopling of North Africa. For the purpose of contributing to a better understanding of the establishment of populations in this Mediterranean region, we proposed the study of the mitochondrial diversity of the Ibero-Maurusian population of Taforalt (12,000 years - Morocco). Ancient DNA was extracted from thirty specimens excavated from the EpiPaleolithic cave of Taforalt. We amplified and sequenced the HVS1 fragment of the mitochondrial DNA control region. The genetic diversity of the population of Taforalt shows that they are not individuals related to sub-Saharan individuals. The hypothesis of a sub-Sudanese [likely a typo in the original publication, and probably meant to be read as "sub-Saharan"; however, arguing against a possible typo, is the fact that the term is used repeatedly. This notwithstanding, one can't help but get the impression that this word is treated as if it were synonymous with "sub-Saharan African"]  origin of the Ibero-Maurusian males is rejected. Our results are consistent with a local origin of the population of Taforalt and genetic continuity in North Africa.

While the authors speak of a "local origin" of the EpiPaleolithic Taforalt group in question, this should not be mistaken for a purely autochthonous African origin in its coastal areas in the north, as 1) any autochthonous African population would have to be related to "sub-Saharan" populations, even if they don't descend from some established group of the latter but ultimately draw from common ancestral gene pools specific to Africa, and 2) the authors go onto tell the reader that pretty much "all the sequences of the Taforalt specimens display haplotypes belonging to Eurasian haplogroups", which going by even their own report, is a stretch of the real state of affairs, whatever the degree that one might reckon it to be:

The 21 specimens analyzed are sufficient to give a good idea of the characteristics of the Taforalt population. None of these specimens are related to the actual sub-Sudanese specimens (lack of haplogroup L, that groups haplotypes typically Sub-Saharan African). The hypothesis of an origin of Ibero-Maurusians south of the Sahara is therefore not supported by our results, showing a population of Mediterranean type rather in the North of Africa at least until 12.000 years. Our results support work based on dental, cranio-facial morphometrics and post-cranial analysis showing divergences between Ibero-Maurusian populations and their contemporaries in Sudan (Bermudez de Castro 1991, Irish 2000).The presence of a sub-Saharan component in the actual north African populations is likely attributable to migrations later than 12.000 years BP.

All the sequences of the Taforalt specimens display haplotypes belonging to Eurasian haplogroups. Among the haplogroup majority, H is considered as originating in the Near East, the emergence of this haplogroup dated to about 35,000 years, while haplogroup JT, also originating in the Near East, has an age of up to 50,000 years; finally, the native haplogroup U6 of northern Africa is purposed as being contemporary with haplogroup JT.

When the authors say above that "none" of their Taforalt specimens were related to "sub-Sudanese", likely a euphemism for "sub-Saharan" Africans, they were not being frank, according to specifics of their very own report. From the rather ambiguous assignments that they provide the reader throughout their report, it is estimated that at least one specimen could very well be a carrier of a hg L marker. The remains of this one individual with plausible "sub-Saharan" African ancestry, appears as follows:

*Click on all images for greater resolution.

This individual—designated as Taf VIII—apparently displays the usual host of cranio-facial characteristics that Eurocentric-minded folks have traditionally attributed to the "caucasian" archetype, yet we are passingly alerted about the very real possibility that individual could well have been a carrier of the hg L marker, which as posted above, the authors essentially equate with "sub-Saharan" Africans. This fellow will have not looked out of place with the rest of the group, whom we are told, were generally carriers of "Eurasian" mitochondrial markers; not at all. Why is this important? It is only significant to the extent that such canio-facial manifestations had formed the basis for preconceived nationalistic-driven Eurocentric arbitrary taxonomic constructs like "Caucasoid" , "Mediterranean", "Negroid", etc, whereby the "Caucasian" or "Caucasoid" taxonomic unit is supposed to embody a standalone monophyletic taxonomic unit, specifically as one "race" among the several human "races". This basic ideology is what has plagued coastal northern African paleontological record, and even why we are having to discuss topics such as the one at hand. These cranio-facial features have prompted several advocates to treat coastal northern Africa as the traditional bio-historical domain of "Caucasians", ultimately from Europe and/or the so-called "Near East", that only relatively recently, got infiltrated by people from south of the Sahara. This very idea is in fact bluntly stated in the extract cited above; one need not go too far, to look up an example. Honestly, if the cranio-metric patterns of these EpiPaleolithic northern Africans were deemed to have stereotypical "Negro" traits, would we be having this discussion; would Eurocentric ideologues be just as eager to mystify them and entertain the idea that they could have been "European" emigrants; would we be having people like Kefi et al. saying that they had no relationship with "sub-Saharan Africans" and that the bulk of their mtDNA markers are "Eurasian"? Something to definitely think about.

Meanwhile, as noted moments ago, the integrity of Kefi et al.'s mtDNA-sequence reports is hampered by several noticeable loose ends that need to be ironed out, not the least of which is the glaring extreme ambiguity. Caramelli et al. [2003, 2008] who go out of their way to brief the reader on their even more meticulous quality control measures [than those supposedly taken by Kefi et al.] to ensure authentic DNA reporting, still suffered informational inadequacies, let alone that of Kefi et al.'s comparatively far less stringent quality control regime. Indeed Kefi et al.'s confession about having to grapple with DNA contamination meets expectations, given the insufficient quality control measures detailed by them. DNA of their DNA handling crew infiltrated that of several of the PCR fragments, generating sequences similar to those of the DNA crew members. Of course, Kefi et al. reassure their readers that this problem was subsequently contained, claiming to have shelved the affected specimens, and relieved them of any further analytical use. Who is to say that other contaminating exogenous DNA did not infiltrate sections of the remaining DNA fragments? After all, one or a couple of sequences [16126C and 16355T for example] named in the contaminating DNA also reappeared in several other DNA sequences, aside from the three specimen fragments eliminated from the analysis.

In fact, while the authors proclaim to presumably gotten around to correcting the mutational artifacts of exogenous influences, the following revelations point to unnatural mutations that made their way into the PCR fragments:

Table 7 groups the polymorphisms observed for 23 specimens, mutations in this table are checked several times on different sequences. The mutations are considered authentic when they are found on independent PCR [fragments] of two different DNA extractions performed on the same specimen. In certain cases we noticed for the same individual, different sequences from PCR of independent extractions. As regards the specimen Taf V-18 for instance, mutations 16126C-16304C-16311C are spotted on the sequences of two independent products of PCR, acquired from a first extraction, while for the second extraction, mutation 16311C is not found. So sequence considered for the specimen Taf V-18 is the following: 16126TC-16304C. The non-reproducibility of certain mutations is also observed in the case of specimen XXIV Taf (Table 8). Indeed, the 16129A mutation is observed only once, it is therefore excluded.

The investigation of the specific root cause of these exogenous and inconsistent sequences from the very same specimens seem to have been brushed aside, and instead, the authors hoped to get around the problem by simply eliminating sequences that don't repeat themselves in the PCR products of two "independent" DNA extractions from the same specimen. If repeated contamination somehow excluded one or another contaminating sequence or mutation in one or the other of the two "independent" extractions from the same specimen, how would that have then been remedied by the authors, short of simply eliminating a sequence that does not reappear in one of the two "independent" DNA extractions? These inconsistencies are reflective of intrusive foreign DNA elements in the affected DNA extractions of designated individual specimens. It should be noted that the authors named three specimens earlier, which they told the reader that they eliminated from the analysis due to "contamination" of DNA from exogenous DNA from the research crew; these were namely, specimens Taf VIII-18, Taf XVII-2, and Taf XVII-3. Interestingly, samples from these very same three specimens reported larger-than-average [particularly, for ancient DNA fragments] quantities of replicated DNA strands, as the authors themselves admit, as follows:

For all successful PCR, the quantities of estimated DNA amplifications vary from 5 to 15 ng per 8 μl of product of PCR deposited on agarose gel. Except for for the three specimens Taf VIII-18, Taf XVII-2 and Taf XVII-3 for which the quantity of DNA estimated is from 20 to 30 ng (per 8 μl of product of PCR deposited on agarose gel). These quantities are two times the amounts normally obtained by amplifying ancient DNA. In addition, the PCR products were obtained at the first attempt of amplification reaction. These findings resemble those observed during the amplification of recent DNA [of the research team members], suggesting contamination with recent DNA.

The three above-mentioned were eliminated from further use; yet PCR products from the same specimen but of two "independent" PCR trials with inconsistent sequences were retained for the final analysis. Could it be because, if these items too were eliminated, then the authors would have been left with very little DNA material to work with?  

The authors point out:

Could the determination of the sequences of the specimens of Taforalt resolve the problem of the nomenclature of the subjects of this ancient population?

Taf V-5, Taf V-18 and Taf V27 have different mitochondrial sequences (Picture 7). These are therefore of distinct individuals. So, the Arabic numeral points out the number [specimen designator] assigned to subject buried in the burial. This is confirmed by the analysis of the sequences of Taf XXV-3 and Taf XXV. These last specimens have different polymorphisms (Picture 7), and therefore those of two different subjects. Specimens Taf V-5, Taf V-7 and Taf V-20 have the same mitochondrial sequence. Buried in the same burial, these three specimens could have links of consanguinity according to the maternal line. For studying the mitochondrial diversity of the Taforalt population, we adopted the same selection criteria as those we use in the study of the recent populations, and therefore excluded the related individuals.

Thus we considered a single sequence among those of the three specimens: Taf V-5, Taf V-7 and Taf V-20. We then have the sequences of 21 specimens instead of 23. The sequences of these 21 specimens represent 13 different haplotypes.

The authors started out with 31 EpiPaleolithic specimens, and they tell the reader that they were left with working with 26 specimens after eliminating certain named specimens for one reason or the other, which was then reduced to first 24 and then 23 specimens. Yet, we come to learn that the even those 23 specimens were reduced further down to just 21 specimens that were used in the final analysis. In the above, the authors named yet another two specimens, both alleged to be closely related to one specimen retained in the final analysis; the names of the three related specimen were given above as follows; Taf V-5, Taf V-7 and Taf V-20. One of these, we are not told specifically which one, was to have been considered in the final analysis, while the other two [again, we are not told which two of the three closely related specimens] would have been eliminated from the study.

Perhaps a relatively minor issue, it is of note that Taf V-5's and Taf V-20's  fragment sequences start earlier and end earlier than that of Taf V-7, with the former's sequences between positions 16054 and 16317 having been compared against those of the Cambridge Reference Sequence, while the latter's was read from 16081 and 16404; what if mutations in positions prior to 16081 in Taf V-7's case and after 16317 in either Taf V-5's or Taf V-20's case were different from those of either Taf V-5 or Taf V-20, and V-7 respectively; could there be mutations here that could drastically alter what the authors would call the "most likely haplogroups" that these specimens' markers fall into? Something to ponder, but at any rate, along with the earlier named three contaminated specimens, the aforementioned two closely-related specimens would have amounted to five specimens being eliminated, reducing the 31 specimens down to 26. However, the authors provide the reader with a table consisting of 23 individual specimens, and all three of those closely related individuals were in it! What they did not tell the reader, nor did they identify them by name or tag, and hence, possibly leaving an unsuspecting reader scratching his/her head, is that five more specimens were excluded from the study, which were not included in the aforementioned table. These undeclared left-out specimens are namely; Taf VI-9, Taf XVII-18, Taf XIX-7, Taf XXI, and Taf I—missing specimens not named! The reader is not offered explanation on why these were not made part of the study, and so left on his/her own, to wonder what might have been wrong with them. They could have been damaged, degraded or contaminated; any or a combination of any of these could have affected sequences of those specimens. Had the authors therefore eliminated those aforementioned specimens with inconsistent and therefore dubious sequences, they would have been left with fairly small amount of individual DNA material to work with, and even then, the results would not be unequivocal, given that a good deal of their overall sample size would have been purged from the final analysis; the integrity of the so-called "good" DNA would have been put to question as well. Through it all, the authors could not even get themselves to firmly assign the sequences into a specific haplogroup set.

Several of the variant hypervariable segment positions noted for the different sequences, and therefore offered as some sort of haplogroup vetting-gauges, thereby narrowing assignment of said sequences down to one or couple of suspected haplogroups, turned out to be types that are actually shared across different super-haplogroups. One wonders if some of the sequences could well not be artifacts of exogenous mutations, like those accompanied by hydrolytic deamination, upon the prospect of further investigation.It may well boil down to a matter of a select combination of designated mutations rather than standalone mutations that provide a better measure of pinning down the most likely haplogroup at hand, but those shared mutations across different super-haplogroups in of themselves offer little use in predicting the most likely haplogroup that a specimen falls into. To demonstrate with examples, let's take Kefi et al.'s use of the HVRI segment position "16126C". Now these authors use this mutation to predict specimen Taf XXV 3's most likely assignment into haplogroup JT.

The transition 16126C on its own offers little in way of predicting the most likely haplogroup-affiliation since its presence spans multiple super-haplogroups; for instance, the presence of this transition could just as well make either of the super-haplogroups L3 and L1 the most likely lineage of specimen Taf XXV 3, and yet, Kefi et al. personally preferred to single out haplogroup JT, for an identification with hg L3 or hg L1 would contradict the authors' stated position about the lack of maternal relationship between Taforalt specimens and sub-Saharan Africans. Similarly, specimen Taf 55-IB supposedly reported the mutation represented by position "16239T". The authors subjectively preferred hg H as the most likely haplogroup lineage of that specimen, yet the same specimen could just as likely belong to either hg L2 or hg L3. Although the combined presence of "16126C and 16355T" could be relatively rare in the mitochondrial organelle bearing African-specific hg L lineage, both these mutations have occurred more than once in the African hg L gene pool, and so, it is not inconceivable that a specimen, Taf 55-I in particular, could just as well fall into either an African-specific hg L3 lineage, hg L2 lineage, hg L1 or possibly L4. The cases of heteroplasmy that the authors observed simply compound the uncertainty of their haplogroup assignments. Kefi et al.'s heteroplasmic mutation of "16298 T/C" reported for specimen Taf V 27, could just as well assign that specimen's lineage into hg L3, where the individual 16298C mutation is known to occur. In relation to this, the combined presence of "16179T and 16298T/C" could well assign Taf XIX a into haplogroup L3, wherein both mutations occur in L3h1 sister lineages, even though Kefi et al. gave their personal liking to hg V as the possible haplogroup; hg L1 could be a relatively more distant possibility, wherein the mutation 16179T also occurs. "16124T/C" and "16239T" could readily place specimen "Taf VI-10" in hg L3.

The positions "16172C" and the aforementioned "16126C" could place a specimen (Taf XXIV) in a rare L3h1b1 marker, and likewise, Taf V19E in either some L3h1b1 derivative, L1a subclade, or even M1 subclade, which all have variants bearing the 16172C mutation, assuming that Kefi et al.'s reports for either specimens doesn't involve exogenous mutations, and that homoplasic mutational events took place across hgs L3h1b1, L1a, U6, M1 and possibly, per Kefi et al.'s reckoning, JT in the HVR1 control region. The individual "16233T" mutation occurs frequently in different African-specific L type super-haplogroups, though it also occurs within M and N sub-haplogroups to a comparatively smaller degree. This could place Taf VIII in either of these super-haplogroups. That several of these alternatives had not been considered by Kefi et al., especially where singular or duo "characteristic" mutations are concerned, suggests that the authors could not resist cherry-picking in order to align results with a preconceived perspective. Moreover, Kefi et al. rely too heavily on the HVR1 section of the mitochondrial genome, which is prone to homoplasy across divergent monophyletic units, and on a inadequate number of selections at that. Is it any wonder then, the authors couldn't even firmly discern if certain markers were hg U6, hg JT, or if certain markers belonged to hg U or hg H; likewise, they couldn't discern whether certain markers were to be placed in hg L, M or N. Indeed, one even questions how much faith the researchers themselves placed on their findings as each and every such assignments was followed up with a question-mark!

Even the figures that the authors post in relation to specific haplogroup components doesn't add up, in light of the ambiguity characterizing their specimen assignments to lineages; they note:

42.8 % (9/21) of the Taforalt population have a sequence identical to CRS (Cambridge Reference Sequence, Anderson and al. 1981) reference sequence. These specimens could be classified in haplogroups H or U. Four specimens (19 %) of the Taforalt population have the 16126C mutation. These haplotypes might belong to haplogroup JT. The polymorphisms observed for specimens Taf 55-IB, Taf V26, Taf VI-10 and Taf XVI a2-19 (4/21-19 %) orient the classification of their haplotypes in haplogroup H.

Two specimens of the Taforalt population: Taf XXIV and Taf VI9E (9.5%) have the 16172C mutation in their sequence. Taf VI9E and Taf XXIV could be classified into haplogroup U6 (Maca-Meyer et al. 2003). Specimen Taf VIII has a sequence that differs from the reference sequence (CRS) by the substitution 16223T. The haplotype of this specimen may belong to haplogroups L3, M or N. Two specimens (9.5%) of the Taforalt population could be classified into haplogroup V. We observed the presence of three cases of heteroplasmy (Taf VI-10, Taf V-27, Taf XIX). These heteroplasmies concern positions 16298 and 16124 where we clearly observed the superimposing of the curve of the nucleotide T with the curve of the nucleotide C (to see chromatogram in Figure 4).

The figures given by the authors above, by way of distinct designated individual nucleotide positions, inadvertently head-counts at least one individual twice, unsurprisingly—which amounts to the authors having confused themselves by the ambiguity of their very own specimen assignments into haplogroups, wherein the same individual was counted for both hg JT and hg U, instead of either one of these two haplogroups but not both of them at the same time! This resulted in the authors counting specimen Taf XXIV for both hg JT and hg U6 at the very same time, thereby assigning "4/21" individuals into hg JT and "2/21" individuals into hg U6.

While Kefi et al. claimed to have used all the necessary precautionary measures in their DNA extraction and amplifications thereof, the above-mentioned issues of contamination, ghost-mutations, and the elimination—which itself was a fairly conservative choice on the authors' part, as other specimens which were plagued with questionable and inconsistent sequencing reports, by authors' very own admission, were not put aside—of a good chunk of the original sample size put to question, the overall informative integrity of the study, and they demonstrate that the quality control of the DNA analysis was not meticulous enough, not leaving out the fact that the authors' results thereof were not duplicated in two or more different labs by different research teams.

The authors made quick references to other disciplines whose findings they claim that their results reinforce, such as that of cranial and post-cranial analysis. They claimed that data suggests that "Mechtoid" characters did not completely disappear but were found in 3% of Maghrebi populations and "higher" in Canary Island populations, citing old anthropological works...

Our results reinforce the anthropologic suggest that human type Mechta el-Arbi has not disappeared completely. The characters are méchtoïdes found in 8% of protohistoric burials, in 3% of current populations of the Maghreb and higher percentage among island populations Canary (Ferembach 1962 Camps 1989). Our results give weight to the dentistry, and lithic, craniometric suggesting regional continuity North Africa since the Pleistocene terminal (Lubell et al. 1984, Irish 2000).

Brace et al.'s cranial analysis from the same year, 2005, head on contradicts Kefi et al.'s position, wherein they note, as cited herein before, that:

It is clear, however, that they [the Basques] do not represent a survival of the kind of craniofacial form indicated by Cro-Magnon any more than do the Canary Islanders, nor does either sample tie in with the Berbers of North Africa as has previously been claimed (37, 44-45)... 

To test the analysis shown in Fig. 3, Cro-Magnon, represented by the x in Fig. 4, was removed from the European Upper Palaeolithic sample and run as a single individual. Interestingly enough, Cro-Magnon is not close to any more recent sample. Clearly Cro-Magnon is not the same as the Basque or Canary Island samples.

In a plot wherein the original "Cro-Magnon" specimen, the "Old Man", was run separately from other Paleolithic European specimens, the recent Maghrebi populations are visibly quite distant from both the said Cro-Magnon and the EpiPaleolithic Maghrebi specimens; see:

Note the positions of Cro-Magnon I, the Algerian Neolithic, the pooled "Taforalt and Afalou", the recent Algerians, Tunisians, and a pooled "Berber" sample.

Kefi et al. reference Irish (2000) presumably to buttress their claim that the EpiPaleolithic Moroccan specimens were not related to "sub-Saharan" Africans. Irish himself, in the work "The Iberomaurusian enigma: north African progenitor or dead end?", 2000, claims:

Iberomaurusians from Taforalt in Morocco and Afalou-Bou-Rhummel in Algeria, Nubians from Jebel Sahaba in Sudan, post-Pleistocene Capsians from Algeria and Tunisia, and a series of other samples were statistically compared using 29 discrete dental traits to help estimate diachronic local and regional affinities. Results revealed: 

(1) a relationship between the Iberomaurusians, particularly those from Taforalt, and later Maghreb and other North African samples, and (2) a divergence among contemporaneous Iberomaurusians and Nubian samples. Thus, some measure of long-term population continuity in the Maghreb and surrounding region is supported, whereas greater North African population heterogenity during the Late Pleistocene is implied.

Similarities in dentition between the Maghrebi specimens is understandable when one considers similarities in the socio-economies of the different contemporaneous EpiPaleolithic and early Holocene groups of the Maghreb, likely brought about by acculturation due to relative geographic proximity and through possible interactions. The same applies to the chronologically later so-called "Capsian" groups of Algeria and Tunisia, who could have attained certain cultural influences from "Ibero-Maurusian" groups of those regions, like for example, the dental mutilation of the front incisors; this tradition is noticeably less frequent in the "Capsian" groups than the "Ibero-Maurusian" groups, but occurs there nonetheless. Still, similarities in dental traits has no bearing on cranio-metric revelations that said coastal northwestern specimens do not form some homogeneous cranial type, which a number of researchers had been compelled to acknowledge, despite attempts to force them into a preconceived taxonomic type(s) spanning several of some or the other of the groups that Irish mentions above.

For instance, as pointed out here before, Briggs came up with four "types" from his analysis of EpiPaleolithic and early Holocene Maghrebi specimens [including the so-called "Ibero-Maurusians" and so-called "Capsian" groups], three of which were described as "Mediterranean" sub-types, while the remainder was reckoned to be a mixture of these "Mediterranean" sub-types; these "types" were named respectively as follows: "Type A", "Type B", "Type C", and "Type D".

Chamla on the other hand, came up with two primary "types" and a derivative type namely, the "Mectha-Afalou", the "Mediterranean" types (the "Proto Mediterranean"), and the "Mechtoids" respectively. The "Mechta-Afalou" were associated with the "Ibero-Maurusian" industry, while the latter two were associated with the "Capsian" industry, with the Mechtoids being representative of the Upper Capsian industry [which is interesting, considering that Chamla saw them as the "gracile" version of the "Mechta-Afalou" type, whom as noted, had been associated with "Ibero-Maurusian" industry]. The "Mechta-Afalou" were considered to be generally more robust than the latter Capsian groups.

Recalling another example, noting differences in the EpiPaleolithic and early Holocene cranial collection, see:

As we can see in the above, estimation is made that both specimens (which had all been indiscriminately brought under the "Mechtoid" banner) with no supposed discerning "Negro" qualities and those with "Negro" qualities had been found specimens extending from the Maghreb to Sudan. The questions that arise from these observations, is that if the specimens in question represented Europeans presumably from the Iberian region, then are we to assume that they were responsible for spreading the just-mentioned diversity to fairly far flung areas from their Iberian point of origin, extending all the way to the eastern corner of Africa—i.e. as far as Sudan, having been able to successfully crossover from the Sahara to the Sahel, yet at the same time, autochthonous Africans somehow could not crossover the Sahara and penetrate the Maghreb; are we to assume that sections of these "Iberians" naturally had "Negro" cranio-metric qualities about them, while others hadn't? If not, then how would the "Negro" cranio-metric qualities have been attained? Yet, how is it that despite such dichotomy [between "Negroid" and "non-Negroid" qualities] in the so-called "Mechtoid" samples, some insist that these specimens represent a taxonomic type? And for those who might insist that only the "non-Negroid" specimens represented the "Iberian" immigrants, it should be noted that even those so-called "non-Negroid" elements are in of themselves clearly quite varied. This leads to the question: Where does one come across the same variety in the Iberian peninsula itself, presumably predating the African examples, as that which was supposedly put in place by a would-be "Iberian settler" immigrant community(s) in northern Africa? Likewise, what cultural precedents of the so-called Mechtoid varieties extending from Taforalt to Jebel Sahaba does one find in Iberia? And by "culture" here, reference is being made the culture in its wholesome, as opposed to just some isolated superficial parallels in a few stone tools here and there. These questions are generally greeted with much silence, yet one does not cease to come across those who speak of EpiPaleolithic Maghreb being the exclusive territorial domain of some "Iberian" settler community.
 
The excessively long history of anatomically modern human occupation of the Maghreb predating the emergence of the so-called "Ibero-Maurusian" populations shatters Eurocentric myths about EpiPaleolithic Maghrebi specimens representing some "Iberian" settler community that would have been isolated from autochthonous Africans from "sub-Saharan" Africa. Certainly, the Aterian populations do not represent a European settler group, nor do those from the Irhoud sites. Both the Aterian and Irhoud individuals are deemed to be "modern humans", although still retaining a few so-called "plesiomorphic" traits. If these folks did not come from Europe, as all indicators point out, not to leave out the fairly obvious one, which is that their presence in the Maghreb even predates the settlement of Europe by anatomically modern humans, then surely the source of these humans would have had to have been "sub-Saharan" Africa. Some try to lend some modicum of rationality to north African "Iberian" settler-group, by speaking of their supposed LGM (Last Glacial Maximum) arrival, and hence, the reason for their presence in northern Africa—i.e. the Maghreb serving as a LGM refuguim.

Obviously, the Aterian specimens are certainly not representative of such immigration, since they are associated with Maghrebi complexes that predate the LGM, nor are the EpiPaleolithic so-called "Mechtoid" specimens, which are generally dated onwards from a period when the LGM was winding down.

Several publications have noted archeological and/or temporal gap between the Aterian groups and the EpiPaleolithic "Ibero-Maurusian" ones in the main locations where the former had been identified; this time gap would essentially cover nearly the entire period of the LGM! The only location where such archeological/temporal gap between the two groups is filled with continuous occupation, as an exception, is the Haua Fteah cave in Cyrenaica, Libya; this has been predicated on the understanding that layers extending from recent era through to the EpiPaleolithic paleontological series, and thereafter, through the Upper Paleolithic Dabban industry to the Libyan pre-Aurignacian complex ("Mode 4 industry") and then the more older Paleolithic periods [i.e. the Levalloiso-Mousterian/Aterian Paleolithic industry], had all been identified respectively at the Haua Fteah site. It is more than questionable that such paleontological and archeological record has its precedence in the Iberian peninsula.

Aterian cranial specimens from Dar es-Soltane are generally reflective of the upper layers (chronologically more recent layers) of the Aterian industry, but here are some images of at least one specimen—Dar es-Soltane 5; Dar es-Soltane 5 [left/top] and Dar es-Soltane 5 sketch [right/below]:

Immediately above, is a "Capsian" site skeletal remain; note the missing incisors, and compare its nasal opening with the Aterian specimens above, and also the following Taforalt specimen.

At a glance, the nasal widths of these specimen (Aterian specimens in particular) appear moderate, if not considered to be "narrow", and not too far off from the Taforalt specimen posted earlier [also shown below]—Taf VIII:

Taf VIII's nasal opening certainly does not give an impression of a low bridge and wide nose. Given a look at the nasal openings of both the Aterian individuals and that of the Taforalt specimen, the latter's (the Taforalt specimen) example does not appear to be inconsistent with the Saharan patterns throughout that regions entire paleontological history.

Nasal opening aside, note the Aterian specimens more robust traits, notably for example, the brow ridges.

Finally, the issue of the post-cranial data that Kefi et al. note in a passing: Perhaps these authors were alluding to findings like this, by T. Holliday, which suggest differences between the Jebel Sahaba specimens and the Mesolithic "Ibero-Maurusian" specimens.

Body proportions of the Jebel Sahaba sample.

TRENTON W. HOLLIDAY1.

1Department of Anthropology, Tulane University. 

The Epipaleolithic site of Jebel Sahaba (Sudan) was discovered in 1962, ca. 1 km from the east bank of the Nile, and ca. 3 km north of Wadi Halfa (the site is now submerged beneath Lake Nasser/Nubia). From 1962-1966, a total of 58 intentionally-buried skeletons were uncovered at the site. Diagnostic microliths suggestive of the Qadan industry as well as the site’s geology suggest an age of 14 – 12 ka for these burials. In this study, the body proportions of the Jebel Sahaba hominins are compared to those of a large (N = ca. 1100) sample of recent human skeletons from Europe, Africa, and the north circumpolar region, as well as to terminal Pleistocene “Iberomaurusian” skeletons from the northwestern African sites of Afalou (Algeria) and Taforalt (Morocco), and Natufian skeletons from the southern Levantine sites of El Wad and Kebara. Univariate analyses distinguish Jebel Sahaba from European and circumpolar samples, but do not tend to segregate them from North or Sub-Saharan African samples. In contrast, multivariate analyses (PCA, PCO with minimum spanning tree, NJ and UPGMA cluster analyses) indicate that the body shape of the Jebel Sahaba hominins is closest to that of recent Sub-Saharan Africans, and different from that of either the Natufians or the northwest African “Iberomaurusian” samples. Importantly, these results corroborate those of Irish (2000), who, using non-metric dental and osseous oral traits, found that Jebel Sahaba was most similar to recent Sub-Saharan Africans, and morphologically distinct from their contemporaries in other parts of North Africa. This study was funded in part by NSF (grant number SBR-9321339).

Both the "Ibero-Maurusian" specimens and the Jebel Sahaba specimens generally have tropical body proportions, or else a fairly sizable segment of their sample sizes bear such proportions, where limb and/or trunk-limb proportions are concerned. Perhaps a hint on this in the above extract, is the fact that Jebel Sahaba could not be segregated from either the "Ibero-Maurasian" and the recent "sub-Saharan" African samples through the "univariate" analysis; whereas it was effectively distinguished from the recent European and circumpolar samples. It is possible that the parameters for body linearity or mass may well have placed either the Natufians and the "Ibero-Maurusian" samples in "intermediate" ranges. Should this be the case, the finding would contradict Kefi et al.'s supposition about a lack of relationship between the Taforalt specimens and "sub-Saharan" Africans, be it tentative evidence of the Taforalt populations exchanging genes with geographically proximate "non-African" groups, or micro-evolutionary response to the sub-tropical environment near the Atlas Mountain chain region. Certainly the Natufians are no stranger to being recognized as at least immediate partial descendants of Africans, ultimately from "sub-Saharan" Africa. Recalling Holliday's work here, the Upper Paleolithic European specimens differed from the later Paleolithic Europeans, in that the latter, while still retaining tropical body proportions, developed shorter limbs, thereby reflecting change, likely brought about the extreme cold temperatures associated with the last glacial periods. The remains of Taforalt specimens, on the other hand, are reportedly suggestive of tall individuals; rarely do tall individuals exhibit "shortened" limbs.

According to material referenced by Hershkovitz et al. 1995, the limb-proportion means of a pooled EpiPaleolithic northern African sample fell right within the tropical African range [85% for crural index and 77% for brachial index]. Now, since this is a pooled group from a wide geographical expanse, spanning that from Jebel Sahaba to Moroccan territory—i.e. west to east or vice versa, which had all previously been indiscriminately included in the so-called "Mechtoid" taxonomic-construct, it might well be instructive to keep in mind that this pooling would naturally affect the overall mean. The individual indexes actually ranged from 82-88. More of the individual Jebel Sahaba samples are inclined to be in the higher-end ranges of the pooled samples with greater probability than those of their Maghrebi counterparts, considering that aforementioned data on "multivariate" examination on body shape, and the former's relative geographical proximity to the tropical areas. Obviously, 82% index is an example of a low value generally out of the range of the tropical African mean, but in order for the mean to be within the range of the tropical African mean, there would have had to have been enough individual indexes in the pooled sample with great enough values, so as to bring the average to the stated value, which is similar to one that Holliday observed for "sub-Saharan Africans" elsewhere; see: Body Proportions—Trenton W. Holliday - 3

It is not inconceivable that Mesolithic Maghrebi groups [who do not appear to be ancestors of recent Maghreb Tamazight or "Berber" speaking groups based on cranial findings and genetic material] may have interacted and exchanged genes with geographically proximate groups that "back-migrated" into the African continent, just as the case seem to be in that region today, though evidence is wanting in that regard, but the idea that they were "isolated" settler immigrant groups untainted by genetic exchange with autochthonous Africans, particularly "sub-Saharan" Africans, is unsupported by the preponderance of available evidence, both from an archeological and bio-anthropological standpoint. As it stands, the burden lies on advocates proposing thereof,  to present substantive multidisciplinary evidence that unquestionably establishes EpiPaleolithic Maghrebi as nothing more than settler immigrant "Europeans"!

*Keep an eye on possible updates in future. Update posted 4/16/2013 (see link below).

 Update and additional information!

________________________________________________
*References:

— Kefi et al. (2005), Mitochondrial Diversity of the Population of Taforalt (12,000 years b.p. - Morocco): A Genetic Study Approach to the Peopling of North Africa.

—Hershkovitz et al. (1995 ), Ohalo II H2: A 19,000-Year-Old Skeleton From a Water-Logged Site at the Sea of Galilee, Israel. 

—Caramelli et al. (2008), A 28,000 Years Old Cro-Magnon mtDNA Sequence Differs from All Potentially Contaminating Modern Sequences. 

—Brace et al. (2005), The questionable contribution of the Neolithic and the Bronze Age to European craniofacial form. 

—T.W. Holliday (1999), Brachial and crural indices of European late Upper Paleolithic and Mesolithic humans.

—Personal notes: 2006, 2010.

*Further reading: Following Trails of the Cro-Magnon