GENEALOGY-DNA-L ArchivesArchiver > GENEALOGY-DNA > 2006-11 > 1164067636
Subject: Re: [DNA] Sykes' data shows E3b absent entirely from"CentralEngland"
Date: Tue, 21 Nov 2006 00:07:16 +0000
-------------- Original message --------------
From: David Faux <>
> I entirely agree. This is a fascinating finding. E3b has a number of
> subclades so it would be interesting to see what is really to be found in that
> remote corner of Wales. It would not surprise me at all to learn that what we
> are seeing is the genetic residual of the Roman settlement of Chester (the Y
> component at least - perhaps mtDNA should be tested as well).
> At some point we could test the available samples for specific subclade, but
> the phenomenon is so intriguing that you are right, a new sample, say at least
> 50 but 100 would give better confidence in the findings - whatever they might
> I've always thought the E3b's in Abergele might have had something to do
> with the Roman legions stationed at nearby Chester. I'd love to know if the
> results would be replicated in a fresh, larger sample.
> ----- Original Message -----
> From: "David Faux"
> Sent: Monday, November 20, 2006 8:43 PM
> Subject: Re: [DNA] Sykes' data shows E3b absent entirely from
> > Immediately to the west of Mercia is North Wales. Looking at Weale et als.
> sample, 39% of those in Abergele were Hg 21 which I believe is E*(xE3a), in
> other words most likely E3b. Mind you the sample size was small, and a good part
> of the diversity (22%) was seen in one haplotype, but the other 17% is spread
> out so I don't think we are seeing a founder effect entirely - the authors talk
> about "high genetic islolation and drift" in relation to the sample. This one
> needs to be satisfactorily explained. Good luck.
> > David Faux.
> > Steven Bird wrote:
> > I've been examining and plotting E3b haplotypes from Dr. Sykes' data,
> > corresponding to the OGAP. He classifies a region as "Central England,"
> > which appears to correspond roughly to the ancient Anglo-Saxon kingdom of
> > Mercia. (Offa was the most famous king of this kingdom; the one who
> > constructed Offa's Dyke.) The southern portion of "North England" on Sykes'
> > map appears to overlap with the northernmost portion of the Mercian kingdom,
> > but most of "Central England" appears to lie within the boundaries of what
> > once was Mercia.
> > What is interesting is the complete absence of E3b in this region, according
> > to the Sykes survey data of 193 haplotypes in Central England. The region
> > is surrounded by "North England," "East Anglia," "London," "South England,"
> > and "Wales" (the data set does not distinguish between South Wales,
> > Mid-Wales and North Wales). Within these regions, E3b appears at a
> > frequency of 2-5%. Capelli showed similar percentages in specific locations
> > surrounding this area, from 3-6% of the population, although he managed to
> > overlook the Sykes "Central England" area entirely.
> > There is a very low incidence of E3b in Ireland (0.44%), and 0% found north
> > and west of the "Loch Lomand" glacial line between Argyll and Hebrides, and
> > Grampian and the Highlands. Sykes found 3 samples out of 202 in the
> > Northern Islands, and Capelli found 0% in the Orkneys.
> > What is the explanation for this donut hole in the E3b distribution of
> > modern England? It can't be due to Neolithic settlement patterns, or the
> > Norman settlement pattern (which was all over this part of England). It
> > could very likely be due to displacement of local population by Mercian
> > invaders (in the 6th century). The difference between the absence of E3b in
> > Central England and the higher than average presence in the regions
> > surrounding Central England suggest such a displacement.
> > In fact, the ONLY event that would seem to me to account for the
> > displacement of E3b from this particular region of England would seem to be
> > the Anglo-Saxon invasion. Are there other historical events in Central
> > England which could explain such a pattern?
> > Steve Bird
> To unsubscribe from the list, please send an email to
> with the word 'unsubscribe' without the
> quotes in the subject and the body of the message