Italian R1b1 - The Central Italian Refugium

[johnraciti]

Did Italian R1b's originate in the Iberian Peninsula or originate during the Last Glacial Maximum in an Italian refugium?

If Italian R1b's originated during the Last Glacial Maximum in an Italian refugium, this would mean that they could be native to the area, and that it is possible that they moved south to Sicily.

[F.E.C.]

John,
until now, the only two Italian paternal lines tested for S21 (mine and Mike's) are positive.
The S21 marker could characterize the ones whose ancestors overwintered in an Italian refugium during the LGM (http://archiver.rootsweb.com/th/read...-02/1139789318).
More Italian R1bs' samples are necessary to support this theory: why don't you take into consideration to test for this marker?

Francesco

[YCCHgI]

Yes!

My ancestors come from a south-central Italian town high up in the mountains (an isolated place). Historically, my ancestors' town was almost completely free of invasions. At any rate, I recently tested positive for Hg I1b2 (which some now have renamed).

The relevance is: according to Rootsi, et. al., my subclade of Hg I waited out the LGM with the R1b bearers. Consequently, there is a high correlation of Hg I in the lands where there is high R1b.

(As a matter of fact, my clade only exists in the extremes of Western Europe: If you throw out Sardinia, like Cavalli-Sforza does, because of pronounced "founder effect," my subclade exists only in Spain, France, Ireland, Portugal, Sweden and Italy.

My point is, barring of course a recent infusion from one of these lands, the presence of I1b2 in isolated Italian towns could well indeed help prove your point. If it is a "tracking device" for certain R1b lines, it could show small LGM refugia populations in Italy. Or, they certainly headed there shortly thereafter.

IMO, it would certainly make sense.

[Downer101]

Yes, some R1b went to Italy, look at this from my Genographic account. I was predicted to be S21+ too, but I need to be certain by taking the test.

[vineviz]

Did Italian R1b's originate in the Iberian Peninsula or originate during the Last Glacial Maximum in an Italian refugium?

The picture of R1b in Italy quite complex.

The frequency of R1b in Italy is very high in areas of the north where migrations from other populations is known to have been very pronounced. The frequency of R1b in areas that are more isolated from continental Europe (like Liguria or southern Italy) tend to have dramatically fewer R1b folks.

It is certainly possible that there was R1b living in Italy during the LGM period, but that population has almost certainly been overlayed with later migrations that included R1b settlers from other areas of Europe.

But it is not inconceivable that all or most of R1b came into Italy post-LGM. I'd say the question is interesting, but still wide open.

[cacio]

vineviz:

that's a very interesting point. btw, where did you get the data on Liguria? (I am quite interested myself because my parents come from the coast very close to Liguria).

It must also be said that R1b is present throughout Italy in large numbers, and it is the most common haplogroup almost everywhere except may be the extreme south (Sicily) - so the multiwave hypothesis with some R1b's from before the LGM seems more plausible. But then we know there were all those migrations, from the Indoeuropeans (who went all the way down to Sicily), to the Celts in the North. In this sense, your project, with many R1b from Central-Southern Italy, is quite relevant, because these must reflect a pre-Celtic presence of R1b in Italy (the Celts only occupied the NW).

Another question of course is what really overwintered in Italy. I1b, G?

cacio

[MMaddi]

Cacio,

Vince and I have had an e-mail exchange in the last day or two about the subject of R1b in Italy. I brought to his attention that Leo Little, who is very knowledgeable about R1b in Europe and analyzing haplotypes, has put together a table from the YHRD database, which you can see here - http://freepages.genealogy.rootsweb..../yhrd-2007.htm

This table gives a breakdown by region of those haplotypes in the database that have the values DYS392=13 and DYS=23. The first marker value isolates R1b and the second marker value is a non-R1b modal value for what's called Frisian R1b. Frisian R1b haplotypes have been found to be a strong indicator for S21+ status. In turn, the present understanding of S21 is that it represents deep ancestry in the area around the North Sea (northern Netherlands, northern Germany, Denmark and southern Norway).

Looking at the table, you can see that Sicily, which has a smaller percentage of R1b than northern Italy (as you point out), is high is Frisian-looking haplotypes. Both Sciacca and Sicily unspecified in this database are above 40% for this haplotype, while most of Europe is below 30% and most of the non-Sicilian areas of Italy are 30-35%.

The other non-Sicilian area of Italy in the 40%+ range is Liguria. I pointed this out to Vince. So this is interesting, although I don't know what would account for this. Vince may have something to say about this.

I think that Italian paternal lines will be important for R1b SNP testing to understand R1b in Europe and the question of which haplogroups and subclades were in which refugium. The fact that the Sicily Project has an R1b1b and it looks like there is another one in the Italy Project indicates there are R1b subclades found in Italy which are rare in other parts of western Europe. Finding them and accounting for their ancestors' migrations from Central Asia will yield a lot of knowledge for population geneticists. At least that's my belief.

By the way, with 97 yDNA results in the Sicily Project, we have 25 R1b members, just above 25%.

Mike Maddi

[vineviz]

The other non-Sicilian area of Italy in the 40%+ range is Liguria. I pointed this out to Vince. So this is interesting, although I don't know what would account for this. Vince may have something to say about this.

I don't have a whole lot to say about it, except that some unpublished data I have been working with lately suggests that R1b frequency is pretty low not only in southern Italy (Sicilia, Calabria, Basilicata, Puglia, etc) but also in some northern regions like Liguria and Venetia.

[cacio]

Mmaddi and vineviz:

thanks for the info, very interesting and suggestive.

How reliable is DYS392=13 as a sign of R1b? I am asking because certain percentages of DYS392=13 seem a little low in the North and Central Italy (19% Veneto? 25% Latium?), while it seems OK if not high in the South, in view of Vineviz's blog and Di Giacomo's paper. May be Italian versions of R1b do not have it? If I recall Di Giacomo (which I don't have with me right now), the North seemed to have had > 50% R(1b).

Anyway, the table and your observation suggest a couple of things.
1) Sicily and the Normans. I did not believe that the Normans could have had a large genetic impact. But then if the Frisian variety of R1b is there, may be that was the case. I remember also that in (Capelli?) paper the Palermo area had a significant fraction of I1a (but Sciacca would be in another part of Sicily).
2) Veneto and Liguria. Liguria is a little strange because so much of it is on the coast. I think I remember that Di Giacomo also found Genova with less than the expected share of R1b. But of course, as a big port on the Mediterranean, this may not be a surprise. However, the Di Giacomo paper also shows a huge fraction of R1b in Garfagnana. Garfagnana is a mountain region in Tuscany, but close to Liguria. Before the Romans, it was inhabited by a tribe that the Romans classified as Ligurian (rather than Celt), although the relation between Celts and Ligurians is open to discussion. As for Veneto, it wasn't really occupied by the Celts (who were in Lombardy, Piedmont and Emilia). Venice itself, again, as a trading center, could be special. Di Giacomo I think has data on a mountain valley in Trentino (which is presumably similar to Veneto? at least, certainly it was not celtic) with very high percentages of R1b. Similarly, it would be interested to see how Veneto relates to the Balkans, with their I1b, E3b's and the like.

cacio

[MMaddi]

Mmaddi and vineviz:

thanks for the info, very interesting and suggestive.

How reliable is DYS392=13 as a sign of R1b? I am asking because certain percentages of DYS392=13 seem a little low in the North and Central Italy (19% Veneto? 25% Latium?), while it seems OK if not high in the South, in view of Vineviz's blog and Di Giacomo's paper. May be Italian versions of R1b do not have it? If I recall Di Giacomo (which I don't have with me right now), the North seemed to have had > 50% R(1b).

That's a good point about 392=13, which I hadn't thought about. My 392 is actually 14, not 13, so if I were in the YHRD database, Leo's search wouldn't have turned up my haplotype in his R1b sample. Here's another chart Leo put together, at http://www.geocities.com/null439/R1b_Distribution.htm, which shows the distribution, on ysearch I believe, of marker values for 6,600 R1b haplotypes for every one of the first 37 FTDNA markers. The chart shows that about 87% of the R1b's have 392=13 and about 11% have 392=14. Looking at the 25 R1b's in the Sicily Project, 21 have 392=13 and 4 have 392=14, not far off from the figures for Leo's table. Perhaps Vince can take a look at DYS392 for the R1b's in the Italy Project and detect something significant.

Anyway, the table and your observation suggest a couple of things.
1) Sicily and the Normans. I did not believe that the Normans could have had a large genetic impact. But then if the Frisian variety of R1b is there, may be that was the case. I remember also that in (Capelli?) paper the Palermo area had a significant fraction of I1a (but Sciacca would be in another part of Sicily).

Having read up on the history of Sicily in the Middle Ages, I always have been hesitant to attribute a large genetic imact to the Normans on Sicily. This is based on them ruling the island for only about 125 years and the fact that, from what I've read, not many Norman men actually ended up settling in Sicily. I have read that in Piazza Armerina the Normans brought in Lombards to colonize the town. That would be another source of R1b, but that's just one town. David Faux of Ethnoancestry believes that much of the R1b in Italy is due to the settlement there by the Lombards. And then there's the short period of rule by Frederick Hohenstaufen and his father and sons, which would be more potential German genetic input. I don't know how much genetic impact all these factors together had on the level of R1b in Sicily, although it is suggestive that the Frisian R1b level seems higher than the rest of Italy and many places througout Europe.

Mike

[cacio]

Mmaddi:

thanks again for the info. 392=13 then does seem a good sign overall. But who knows if Italian R1b have the same proportion. ie it may be that Anglo saxon R1b have 80% 392=13, but Italians have, say, 50%.

As said, my prior on the Normans was as yours, but then one wants to explain I1a's and these "Frisian" R1b (if they are actually so).

Regarding the Lombards I find it hard to believe the hypothesis that the Lombards (in the sense of germanic population) explain most of R1b in Italy. It is remarkable how little influence they had on culture and language. It would be surprising they had instead such a large influence on the genetic component. One must of course consider the fact that btw the fall of the roman empire and the late middle age there weren't only the Lombards. Italy was invaded (and ruled by) Goths, and many other Germanic tribes went through in the period. But that doesn't explain the South. We should instead observe a gradient from NE to S, the route taken by the Lombards. And the component then should be highest in the NE, like Friuli and Veneto (for other characters, like blondness, this actually seems the case).

Incidentally, regarding the (limited) "Lombards" in Sicily - it is Lombards in the sense of modern Lombardy region (ie, they came from Northern Italy). Around the 12th-13th century, I don't think there were any distinctly Lombards left in Italy. (Of course, according to the previous hypothesis, Lombardy in the XII century would have been inhabited by the male descendants of the Germanic Lombards).

cacio

[F.E.C.]

Mmaddi and vineviz:

thanks for the info, very interesting and suggestive.

How reliable is DYS392=13 as a sign of R1b? I am asking because certain percentages of DYS392=13 seem a little low in the North and Central Italy (19% Veneto? 25% Latium?), while it seems OK if not high in the South, in view of Vineviz's blog and Di Giacomo's paper. May be Italian versions of R1b do not have it? If I recall Di Giacomo (which I don't have with me right now), the North seemed to have had > 50% R(1b).

Anyway, the table and your observation suggest a couple of things.
1) Sicily and the Normans. I did not believe that the Normans could have had a large genetic impact. But then if the Frisian variety of R1b is there, may be that was the case. I remember also that in (Capelli?) paper the Palermo area had a significant fraction of I1a (but Sciacca would be in another part of Sicily).
2) Veneto and Liguria. Liguria is a little strange because so much of it is on the coast. I think I remember that Di Giacomo also found Genova with less than the expected share of R1b. But of course, as a big port on the Mediterranean, this may not be a surprise. However, the Di Giacomo paper also shows a huge fraction of R1b in Garfagnana. Garfagnana is a mountain region in Tuscany, but close to Liguria. Before the Romans, it was inhabited by a tribe that the Romans classified as Ligurian (rather than Celt), although the relation between Celts and Ligurians is open to discussion. As for Veneto, it wasn't really occupied by the Celts (who were in Lombardy, Piedmont and Emilia). Venice itself, again, as a trading center, could be special. Di Giacomo I think has data on a mountain valley in Trentino (which is presumably similar to Veneto? at least, certainly it was not celtic) with very high percentages of R1b. Similarly, it would be interested to see how Veneto relates to the Balkans, with their I1b, E3b's and the like.

cacio

These are all good points cacio. I can't see why the South should reflect Italy's aboriginal Y-dna more faithfully than the Centre and the North.

As for Liguria and Venetia then, you're correct when you say these regions had very big ports. Especially Genoa and Venice were the capitals of maritime empires, so I wouldn't be so sure to find more "indigenous" dna there than, say, in Lombardy.

On the contrary, I find very interesting the case of Garfagnana. There are still small and isolated communities in the Appennines and that's were, in my humble opinion, you would have more chances to trace "native" dna.

I don't agree with you only when you say that Trentino (or, even more meaningful in NE Italy is the case of Friuli Venezia Giulia) wasn't inhabited by Celtic tribes in ancient times.

[F.E.C.]

Regarding the Lombards I find it hard to believe the hypothesis that the Lombards (in the sense of germanic population) explain most of R1b in Italy. It is remarkable how little influence they had on culture and language.

Sorry but I really don't agree on this as well. There are so many examples of words and cultural characters of clear Germanic descent in the lands of Italy once inhabited by the Lombards.
300.000-500.000 men and women in a country of 8-9.000.000, as Italy was fifteen centuries ago, don't vanish in oblivion.

[F.E.C.]

First of all I make this clear: what I'm going to say isn't meant to refer to anyone participating to this very interesting topic.

Maybe I'm paranoic but there's something that really really really annoyes me: I've like the feeling that someone's got a problem with R1b1c (or any of its subclade) being aboriginal to Italy. I just don't understand what makes the Italy R1b refugium theory so unlikely!

Sometimes they say: yes, maybe few R1b's overwintered in some place other than Spain but the alternatives are most often the Balkans or even the Black Sea! What is wrong with Italy?

[cacio]

F.E.C.:

I hope more research will be done in isolating an Italian refugium and its characteristics. People certainly lived in Italy during the LGM. It may have been the case that the UK was populated from males moving up from Spain, but Italy must have had an autochtonous component, whether they then moved north or not. (in fact, the alps may make it difficult to move north).

As for the Lombard contribution, I agree that they did have some impact. But I would still classify this impact as small in linguistic terms. There are certainly germanic words (bank etc), but these are a minority, most words remained solidly latin. If there had been a significant genetic component from the Lombards and in general from Germanic populations, I would have expected the language to have changed as well. I also believe that the Lombards tended to remain separate from the rest of the population, so that intermarriage was very infrequent. Even taking your numbers on population, that only makes 5% of the overall population.

As for Veneto and Trentino, I know little about their history, so you should correct me on this. In pre-Roman times, Veneto spoke a local language unrelated to celtic. Trentino has Rhaetic inscriptions, a language that was probably not indoeuropean and that sounds related instead to Etruscan. Instead, the NW of Italy had celtic inscription (Leponti etc.), which seem to suggest that the celtic presence in NW predates the historical accounts of invasions from Gaul. I am also not aware of any historical account of celtic invasions into the NE, as far as I know, the Celts were in Piedmont, Lombardy and Emilia. (whose dialects are indeed related to each other and not so related to Veneto).

cacio