GENEALOGY-DNA-L ArchivesArchiver > GENEALOGY-DNA > 2010-09 > 1284052441
Subject: Re: [DNA] 2 routes for P312 and U106 ?
Date: Thu, 9 Sep 2010 13:14:01 -0400
I read all replys so far and I wanted to comment on this Lawrence's point.
The center of maximal variance would then be quite distant from France. I think some more discussion is needed on what really the maximal variance is pointing to. I would say it's close to the first place where expansion took place.
It might be that P312 were next to U106 at the very beginning, before any real expansion or close to the beginning of expansion. Data are sparse for U106 and would be below the level of sampling for P312 around Poland . Then a fast migration (I mean in 1 or 2 generations) from this common center to western Europe (possibly France as indicated by the variance map) would have brought a small group a P312 to a new territory where expansion of this branch took place.
I didn't mention it in the first post but Beaker's settlements do not form a continuous geographical set. Rather it looks like a network with hotspots at the nodes (my way of picturing it). May be the success of P312 in western Europe is some kind of deal with the people (farmers) in place against the new thread coming (precisely the northern IE wave). What , in my view, favors the move of a small group of P312 to the west is that large groups (corded ware) made their way very slowly.
Lawrence wrote :
> The map will then show U106 emanating from Poland, which easily had the
> highest variance after Estonia. Slovakia was next after Poland. The total
> number of samples across Estonia, Poland, and Slovakia was 30.
> It is all too tempting to throw out data that disagrees with one's
> preconceptions. There are two unbiased ways to deal with small samples:
> 1) Throw out _all_ samples that are too small. In this case, we could
> simply throw out every sample of 10 or less. Of course, not much would be
> left, and Slovakia would have the highest variance among the surviving data.
> 2) Consolidate small samples into larger regional ones, as I did above. If
> we consolidate Estonia, Poland, and Slovakia into an Eastern European
> category, it easily has the highest U106 variance in Europe.