View Full Version : Welsh Y-Chromosome Haplogroups

10-12-2019, 01:24 AM
I have good reason to believe my y-chromosome line is Welsh, from the Borders/West Midlands area, i.e., the old Welsh kingdom of Powys, and the tribal lands of the Cornovii or maybe the Ordovices or the Dobunni. My terminal SNP is FGC36981, but I think it arose in North America, maybe with my second great grandfather. The broader haplogroup is characterized by the SNP BY166 (Z18021 at 23andMe) which is found among the following surnames:


The surnames listed above that don't appear to be Welsh are borne by men who don't match others with their surname but do closely match other men with Welsh surnames on that list. Due to privacy concerns I cannot reveal anything more specific.

The phylogenetic pedigree is R-M207, R1-M173, R1b-M343, L754, L389, P297, M269, L23, L51, P310, L151, P312, Z290, L21, DF13, Z39589, DF41, FGC5572, BY168, BY166 (and FGC36974, FGC36982, FGC36981 for me and my closest kin).

That's strictly from memory, so if I left anything out, let me know.

What other y-chromosome haplogroups do you believe are strictly or mostly Welsh? Be as specific and detailed as possible.

10-12-2019, 03:42 PM
Guess no one is interested in Welsh y-dna haplogroups.

10-12-2019, 08:18 PM
I am, just don't know of any specifically Welsh haplogroups. I hope others contribute.

10-12-2019, 08:39 PM
I am, just don't know of any specifically Welsh haplogroups. I hope others contribute.

Same. I wouldn't be surprised if Welsh haplogroups were shared throughout the periphery of the Irish Sea region.

10-13-2019, 01:17 PM
I am, just don't know of any specifically Welsh haplogroups. I hope others contribute.

Probably it's going to take going pretty far out on the branches, perhaps to the twigs to get the kind of geographic resolution needed. Obviously NGS testing with Family Tree DNA's Big Y-500 or Big Y-700 or the equivalent is required.

However, I do think R1b-L371 is almost exclusively Welsh, and I believe its discovery predates the advent of Big Y testing.

My own subclade, R1b-BY166 (R-Z18021 at 23andMe), is almost exclusively Welsh (judging by surnames and by our members who can get their pedigrees across the Pond), with one exception thus far: a man with a Welsh surname whose mdka was born in Wiltshire in SW England not all that far from Wales.

We appear to be a Borders group, so it wouldn't surprise me to see more mdkas born in what is now England, but in that part of England that was once part of the old Welsh kingdom of Powys. For example, I have among my closest matches at 67 markers a couple of men with the Welsh surname Beddoes. However, that Beddoes family is spread into Shropshire within a few miles of the Welsh border. (Thus far, I can't convince any of the Beddoes to join our project. The admin of the Beddows/Beddoes Project has inculcated in them the fear of sharing their information and consequently an absolutely stupid mania for privacy that defeats most of the purpose of y-dna testing.)

Earl Davis
10-13-2019, 07:26 PM
I suspect my BY3290 might be Welsh. My often changing estimate is that it was formed between 300 and 900 Common Era. So far we have 7 Davis, 1 Thomas, 1 Bailey, 1 Griffiths, 3 Wynne, 1 Beddies, 1 Powell, 1 Rice, 1 Parker, 1 Young, 1 Kelsall, 1 Parker, 1 Lentle that I know of. Most trace their known ancestry back to either Wales, West England [Westerness] or South West England. There is at least one ringer in the mix whose know ancestry from Switzerland.

10-13-2019, 08:58 PM
I believe S27900, which is part of the P312 subclade ZZ37/38, is primarily Welsh.

Since surnames were adopted relatively late in Wales, I have noticed most of the Welsh have STR matches with a variety of Welsh surnames.

EDIT: I corrected the SNP to S27900. The sequence is P312>ZZ37>ZZ38>Z29645>S27900. The list of those who have tested positive for this marker can be seen in section F2b of the P312 Project YDNA results. This appears to be a large group, as most of them have a large number of 67 marker matches, and they are predominantly Welsh surnames.

10-14-2019, 10:08 PM
. . .

Since surnames were adopted relatively late in Wales, I have noticed most of the Welsh have STR matches with a variety of Welsh surnames . . .

That's right. My own subclade is pretty tight knit in terms of STRs. The Welsh patronymic naming system accounts for that.

10-15-2019, 05:00 PM
One of my Vance surname project subgroups all descend from an earliest known ancestor in the early 1700s who came to the US from Ireland.

However their Y-DNA connects probably just a few generations before that to the Bassetts of Wales, who are descended from Normans into Wales in the 11th century. While I’m sure the Y-DNA isn’t proven back to the Normans, this same line appears to be the predominant Welsh Y-DNA line for the Bassett surname at least for those who have tested so far.

This line is I-M170>P215>CTS2257>L460>P214>S2525>L38>S2606>S24121.

So the operating theory at the moment is that this Vance subgroup represents a NPE off the Welsh Bassett line.

One of the interesting things about this line is an apparent consistent innate resistance to Y-DNA STR mutations. Between the Vances and Bassetts they’re all no more than a genetic distance of 3 going back to common ancestors who have to be at the latest in the 1600s.

However there hasn’t been much SNP testing among either the Bassetts or Vances in these groups. We only know them down to S24121, though that will change eventually.

10-16-2019, 06:10 PM
That's right. My own subclade is pretty tight knit in terms of STRs. The Welsh patronymic naming system accounts for that.

Exactly. The S27900 group has a very strong STR signature.

For anyone interested in Welsh surnames, I highly recommend The Surnames of Wales by John and Sheila Rowlands (1996). It is essentially an encyclopedia of the origin and distribution of surnames in Wales.

10-22-2019, 01:58 AM
R-A228 marks another Welsh cluster. Around 3% of men from Wales or with patrilineal Welsh ancestry belong to this haplogroup, while it's very rare outside of Wales.

Here's R-A228 in Alex's tree:
R-P312/S116 > Z290 > L21/S145 > DF13 > Z39589 > Z251/S470 > FGC13899 [> A248] > A228

Over 100 likely or confirmed A228+ members are represented in the FGC13899 project:

I believe the MRCA for A228 as a whole lived very roughly 1500 years ago. There are currently three known major branches (marked by A241, A959, and A17480), each apparently with a TMRCA (again, very roughly) something like 1200 years.

Geographically, A241 appears to be predominantly associated with North Wales, and A959 with South Wales (but neither branch is exclusive to either region). A17480 is the smallest of the three branches, and I don't think there's enough data right now to say what geographic trends may be like for it within Wales. A17480 appears to be present in both N. and S. Wales, and probably also England. At least one descendant of this branch also made it to Sweden.

10-22-2019, 05:44 AM
A few years back, I put together a set of around 1600 haplotypes (by merging 67 marker haplotypes from the WalesDNA project, the Welsh Patronymics project, and various surname projects for typically Welsh names) and looked for clusters that appeared to be relatively young and associated with Wales.

Here are the clusters I found, along with their frequency at the time in the combined WalesDNA and Welsh Patronymics haplotypes (not including the additional haplotypes from individual surname projects), some of the other names the various clusters had been given, and the SNPs they appear to be associated with.

These appear to be the largest clusters with origins in the past 1500 years or so and strong associations with Wales:

1. The L371 cluster (3.5%) L371
R-P312/S116 > Z290 > L21/S145 > DF13 > Z39589 > L371/S300

2. TYPE42 / pa-39214 (3.4%) S27900
R-P312/S116 > Z40481 > FGC84729 > ZZ37 > ZZ38 > Z29644 > S27900

3. 251-1130 (2.7%) A228
R-P312/S116 > Z290 > L21/S145 > DF13 > Z39589 > Z251/S470 > FGC13899 [> A248] > A228

5. TYPE14 / d-5311211 (1.8%) FGC17099
R-P312/S116 > Z40481 > ZZ11 > DF27/S250 > ZZ12 > FGC78762 > ZZ19 > Z31644 > Z42772 > A17603 > FGC17112 > FGC17099

7. The largest Welsh haplogroup E cluster (1.4%) BY5965
E-V13 >> E-BY5965 (> FGC14091)

8. The largest Welsh haplogroup G cluster (1.4%) FGC490
G-Z6748 > FGC490 (> Z40857)

Some clusters that look like they have significant (though, in some cases, far from exclusive) associations with Wales but that are probably less frequent in Wales than the six clusters above:

6. R-L21 DF49+ DF23+ WALES (Old Group 4) 49-7623-z15WSW (1.6%) DF23+ M222- >> Y31595
R-P312/S116 > Z290 > L21/S145 > DF13 > Z39589 > DF49/S474 > Z2980 > Z2976 > DF23 > ZP112 > Y31591 > Y31595

9. I1-AS4 (1.4%) FGC41708
I-M170 > M253 > DF29 > Y2592 > CTS6364 > CTS10028 > S6346 > FGC10430 > FGC10445 > FGC41708

11. 11_13Combo / R-L513 (1.4%) It appears the haplotypes I'd included in this grouping don't all actually form a single recent cluster. But most are L513>S6365>CTS6621 [Norman/Cambro-Norman?]:
R-P312/S116 > Z290 > L21/S145 > DF13 > L513/S215/DF1 > S6365 > L705

12. Type 34 (1.2%) M323 [Flemish?]
R-U106 > Z2265 > BY30097 > Z381 > M323

14. L1335+ WALES (Old Group 2 Cadwgon) / Wales 2 (1.2%) FGC15565
R-P312/S116 > Z290 > L21/S145 > DF13 > Z39589 > L1335/S530 > Z16450 > FGC15565

16. 5494-5561-z450611 (1%) A1289
R-P312/S116 > Z290 > L21/S145 > DF13 > FGC5494 > FGC5561 > A1290 > A1289

17. Type 26 / R-DF21 21-641-z1311 (1%) BY35673
R-P312/S116 > Z290 > L21/S145 > DF13 > DF21/S192 > S5488 > Z16294 > BY11121 > ZZ21 > BY3827 > BY18887 > BY35673

18. Type 39 p-1515 (0.6%) PH611
R-P312/S116 > Z40481 > FGC84729 > ZZ37 > ZZ38 > Z39300 > BY15581 > PH611 (> Y77705)

Probably old and large Germanic clusters that have some Welsh descendants but don't necessarily have any special association with Wales (also, given the possibility of NPEs and the fact that many patronymic Welsh names are also English names, the Family Tree DNA haplotypes I was using may not reflect actual Welsh frequencies for these clusters):

4. The largest I1 cluster, probably Ken Nordvedt's AS2 (2.2%) Z138+

13. Z2 cluster (1.2%) U106>Z381>Z301>L48>Z9>Z2>CTS10893>FGC924>FGC918

Predominantly Irish clusters (given the possibility of NPEs and the fact that some Gaelic names were anglicized into Welsh-looking ones, the Family Tree DNA haplotypes I was using are probably not going to reflect actual Welsh frequencies for large Irish clusters very well):

10. M222 (1.4%)

15. Type 35 / R-CTS4466 (1%)

10-22-2019, 06:55 AM
Two more branches that have significant associations with Wales:

R-P312/S116 > Z290 > L21/S145 > DF13 > FGC5494 > FGC5561 > FGC7448 > FGC5496 > FGC5521 > Y18844 > Y18846 > FGC43417 > FGC43418 > Y34193

CTS10571 / S4000
R-U106/S21 > Z2265 > Z381/S263 > S264/Z156 > FGC70809 > Y3966 > A10645 > S3997

And a two other possibilities:

R-P312/S116 > Z290 > L21/S145 > DF13 > L513/S215/DF1 > S6365 > Z16385 > BY363

R-P312/S116 > Z290 > L21/S145 > A7905 > A7906 > A7900

01-02-2020, 09:12 PM
There is a subclade of DF19, if you look at the right side of the R-P312/S116 > DF19/S232 > Z302/S233 that might fit the bill:


^^ I imagine the "American" one told some scribe his unspellable/unpronounceable Welsh surname with an aside that "It's Welsh," and the person just gave up and wrote down "Welch." :lol:

01-02-2020, 10:21 PM
I've meant to post here for some time. I'm interested in my grandfather's R-L371 on my mum's side, which is largely confined to Wales (TMRCA 350 ybp, YFull). As some of you might remember from our miscellaneous Welsh thread, I discovered this by testing my late mother's first cousin at YSEQ after matching him on MyHeritage. His family had left Wales before he was born so we hadn't known he existed until his reply back to my match inquiry. I thought I'd never find a surviving male on that line (there's another definite common Y descendant but he's never replied to me or any match who I'm in contact with) and he also supplied me with family trees and anecdotes, as well as the only photo any of us have seen of my great grandfather. I treasure that.

I worked out from the raw file that he was R1b but U106-. With my limited budget I took a punt on the L21 panel test after some welcome advice from rms2. It paid off and matched my own completely non-scientific intuition because Gramps was always the quintessential Welshman for me. :)He was a Jones but the marker is shared with names including Rhys and Pugh. I don't set any store by the sadly now defunct BritainsDNA's view that L371 (S300) comes from "ancient Welsh-speaking kings (http://www.walesonline.co.uk/lifestyle/showbiz/dafydd-iwan-discovers-impressive-celtic-7829517) or war lords who ruled the southern counties of Britiain". But in any case it's my single most precious DNA discovery.

Incidentally, one of only two or three really keen and knowledgeable autosomal matches I have is on that side. Together we've traced our Jones line back to a 1770 baptism, which is good going for South Wales. We share gg grandparents and found their 1862 marriage certificate, marked with two X's. It's that kind of thing that keeps me on the DNA hunt.

Edward J
01-03-2020, 02:30 PM
Interesting that one of the few FTDNA groups that I am grouped in is the Walsh/Welch Surname project (due to STR matching) in J Lineage 01. Not sure if I really have any connection to this group, hard to tell by looking at the STRs (w/o any further full sequence data).