Cf. http://racialreality.blogspot.com/2010/03/richard-lynn-on-italian-iq.html

Coriscans at the MLDP come out as 85% North-Italian 15% Sardinian, they have little to do with Sicilians or Cypriots..They have in fact one of the highest percentges of the subclade of R1b U152

Conroy, 1/8 is too low in her case to make a good inference about the genetics of Corsicans, as the rest of her ancestry comes from Caucasoid populations too, thus making it impossible to discern which parts of her genome comes from Corsican ancestors with confidence (it is worsened by the fact that we don’t know jack shit about the overall autosomal genetics of Corsicans, so there is a vicious circle here).

My eldest daughter is partly Corsican – actually 1/2 Irish, 1/8 Breton, 1/8 Normandy, 1/8 Lombard, 1/8 Corsican, and in most analyses, her Breton and Normandy seem to merge with the Irish to give some sort of generic British, while her Lombard (Swiss border, Lake Como) and Corsican give some sort of Southern Italian. So by a process of elimination, the Corsican element must look like Cypriot or something?!

Here are her Dodecad K12B results:

Single:
[1,] “French_D” “4.7444″
[2,] “French” “5.2463″
[3,] “Mixed_Germanic_D” “6.8205″
[4,] “Kent_1KG” “8.3709″
[5,] “Dutch_D” “8.5303″
[6,] “English_D” “9.0515″
[7,] “CEU30″ “9.057″
[8,] “Cornwall_1KG” “9.7711″
[9,] “British_D” “10.4655″
[10,] “British_Isles_D” “10.5988″

Mixed-Mode:
[1,] “21.6% C_Italian_D + 78.4% English_D” “0.4769″
[2,] “16.9% S_Italian_Sicilian_D + 83.1% Kent_1KG” “0.511″
[3,] “22% Greek_D + 78% British_D” “0.5131″
[4,] “16.9% Sicilian_D + 83.1% Kent_1KG” “0.5201″
[5,] “23.6% O_Italian_D + 76.4% Kent_1KG” “0.5405″
[6,] “21.6% C_Italian_D + 78.4% CEU30″ “0.577″
[7,] “75% English_D + 25% O_Italian_D” “0.686″
[8,] “82% English_D + 18% S_Italian_Sicilian_D” “0.7213″
[9,] “18% Sicilian_D + 82% English_D” “0.7215″
[10,] “20.3% C_Italian_D + 79.7% Kent_1KG” “0.7479″

Whereas HarappaWorld has her as follows:

Single:
[1,] “french_hgdp_28″ “3.1544″
[2,] “utahn-white_1000genomes_100″ “7.8712″
[3,] “british_1000genomes_99″ “9.1146″
[4,] “n-european_xing_25″ “9.4165″
[5,] “utahn-white_hapmap_18″ “9.6669″
[6,] “hungarian_behar_19″ “11.1076″
[7,] “orcadian_hgdp_15″ “11.9383″
[8,] “slovenian_xing_25″ “13.0643″
[9,] “spaniard_behar_12″ “15.0464″
[10,] “spaniard_1000genomes_98″ “16.143″

Mixed-Mode:
[1,] “73.1% british_1000genomes_99 + 26.9% tuscan_hapmap_102″ “0.8487″
[2,] “73.4% british_1000genomes_99 + 26.6% tuscan_hgdp_8″ “0.8647″
[3,] “73.8% british_1000genomes_99 + 26.2% tuscan_1000genomes_11″ “0.9465″
[4,] “67.5% orcadian_hgdp_15 + 32.5% tuscan_hapmap_102″ “1.0734″
[5,] “39.6% italian_hgdp_13 + 60.4% orcadian_hgdp_15″ “1.175″
[6,] “67.8% orcadian_hgdp_15 + 32.2% tuscan_hgdp_8″ “1.2094″
[7,] “30% italian_hgdp_13 + 70% utahn-white_1000genomes_100″ “1.2255″
[8,] “68.3% orcadian_hgdp_15 + 31.7% tuscan_1000genomes_11″ “1.3501″
[9,] “66.8% british_1000genomes_99 + 33.2% italian_hgdp_13″ “1.4398″
[10,] “90.3% french_hgdp_28 + 9.7% mordovian_yunusbayev_15″ “1.4592″

Of course ancient Tuscans (aka Etruscans) colonized the East coast of Corsica in times of yore.

Even if there was such an error in the sampling process, why in Piedmont and Liguria and not in the even more industrial and immigrated region of Lombardia, correctly mentioned in your article (Milan), as the epicenter of industrial development and immigration (or also Emilia-Romagna and Veneto)? There should not be this kind of distinction, yet there is.

Of course the migration could be from, say, 300 or 500 or 2000 years ago but I can’t fathom it in all the historical period. Neither in the prehistorical one admittedly (except Cardium Pottery Neolithic possibly) but that is a blurrier zone.

Just to connnect the dots a little, the Mozabite people are a Berber ethnic group living in Southern Algeria (which is in the Northern Sahara desert). The Berbers, in turn, are indigeneous Saharan people who were traditionally herders, whose indigeneous languages are one of the main language families within the Afro-Asiatic language family. While their lifestyle is similar to that of Arabic herders, their indigeneous language (spoken until sometime after Arabic was introduced in the 8th century) and genetics are very distinct. They show considerable genetic continuity to the hunter-gatherers of North Africa ca. 12,000 years ago.

The implication would genetic ancestry contributions to the Berber people in Northwest Africa, from Neolithic people who pre-date Indo-European expansion, who have common origins with the Sardinians, either across the Strait of Gibraltar or via Egypt and/or the North African coast.

An exchange across the Strait of Gibraltar (or by sea in the Western Mediterranean) would seem to be a more likely source for this component than a North African coastal route given (1) the existence of some other Y-DNA (e.g. certain clades of Y-DNA hg E in Southern Europe), (2) mtDNA ties back and forth between Spain and Northwest Africa (e.g. mtDNA haplogroup V), and (3) the lack of strong Sardinian-like signal between North Africa and the West Asian Highlands (Anatolia, the Caucasus and Iran).

Given the lack of Basque-like languages in North Africa (often seen in cases of genetic ancestry contributions in connection with conquest and superstate rule), and the lack of much Y-DNA haplogroup G or haplogroup R1b in Mozabites, my guess would be that copper age bride exchange in connection with trade across the Straight of Gibraltar would be a most likely source of the Sardinian-like genetic component seen in the Mozabites.

While the Roman Empire extended to North Africa providing a potential means of gene exchange, a gene exchange with Berbers in this era seems less likely because the Berbers aren’t sufficiently similar to Italians (the core of the Roman empire) genetically to be a good fit for that.

i have friends who live in the piedmont. i have been to turin. it was heavily settled by workers from southern italy, including sicily. this is well known

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=6&cad=rja&ved=0CEoQFjAF&url=http%3A%2F%2Funinomade.org%2Fwp%2Fwp-content%2Fuploads%2F2012%2F05%2FNordisti-contro-Sudisti.-Internal-Migration-and-Racism-in-Turin-Europe-1950s-and-1960s.pdf&ei=tRpSUNDADYe6iwLE7YCwDQ&usg=AFQjCNEqm5hGnduxqh2wFZfDYJGb1Rcw6g&sig2=1sZQyW2VgdYYG1OdN8B-YQ

Corsicans are one of the number one populations whose overall autosomal genetic results I have been looking forward to seeing for years for the same reasons, but somehow no population geneticist seems to be interested in them, at least not as much as you and me.

“What I’m saying is that one run of ADMIXTURE may tell you that K = 4 is the best fit, but another run may tell you that K = 6 is the best fit”.

You’re probably right but, looking at the various K values in fig S6, they are all roughly the same for Italy, with most of the changes taking place in the small Levantine slice and stuff that looks more like noise than informative, right? I totally agree with not reifying the clusters into “real populations” without at least much more extensive testing.

“People whose geographic origins are in the north often cluster with southerners (i.e., the southern cluster), but people whose origins are in the south rarely seem to cluster with northerners”.

Maybe the most interesting of all. My first thought after reading your comment was: “wait, if they are admixed, they won’t show as southerners anymore but as central italians or whatever else in between, there must be true genetically South Italian communities in Northern Italy that keep the overall affinity”.

Luckily fig. S2 illustrates the same graph by official regions such as Piedmont, Veneto, etc. much smaller and more precise than the tripartite N-C-S division. And what do we see? That it is only people from Piedmont and Liguria (and not Lombardia, Veneto or even the tiny Alpine region of Val d’Aosta) who do that. They cluster with Southern Italians, with Central Italians (one Lombardo also here but too exceptional to matter) and even with “admixed Sardinians” (those Sardinians who tend towards the Mainland genetic makeup).

Would it be only Liguria, we could think of Genoese mariner and trade relations maybe but Piedmont? The only element in common those two regions have is that they were both territory of the historically documented ethnicity of the Ligures, which also extended into SE France, East of the Rhône, being probably a “mountain refuge” population versus the Celtic advance.

This is most interesting and probably more correct than your notion of “intra-Italian migrations”, unless of course we imagine that Ligurians were migrants from the South or something like that. I can find no better explanation than an ancient Ligurian link.

… “may be that in the Copper Age the western Mediterranean was dominated by a Sardinian-like population, which later was displaced and assimilated by newcomers”.

North Africans have some 30% of their mtDNA and some 10% of their Y-DNA of likely Iberian and certainly SW European origin. Austosomal analysis also give them strong Iberian component. I think this should be Paleolithic, because we know that Berbers (Libu, Numides, Mauri) dominated North Africa west of Egypt since at least 5800 years ago (Ancient Egypt arises) and probably since Neolithic (Capsian Neolithic) or even the Late Paleolithic (Early Capsian culture). There’s no other wave until the Arab invasions (excepted more localized influences mostly limited to parts of Tunisia: Phoenicians, Romans, Vandals), so the Iberian (SW European) substrate should be pre-Neolithic, because there was an obvious E1b-dominated (African Afroasiatic) wave on top of it.