PDA

View Full Version : Thinking Out Loud: Using Geography and Higher Order Matches to Trace the Homeland



rms2
09-28-2013, 05:04 PM
Okay, since this is the Personal Genetics subforum, I'm going to talk about my own results, which may get the attention of one person: me. Anyway, here goes.

At 111 markers, the only close match I have with a solid paper trail to the Old World traces his ancestry to the little village of Llanafan Fawr, Powys, Wales (http://goo.gl/maps/lvfDF) in the 17th century. Llan is Welsh for church, so the name means St. Afan's Church. The surname of this match is Samuel, and he is six away at 111 markers.

At slightly further away, eight and nine off at 111 markers, I have two with the surname Webb, one of whom traces his y-dna mdka to Gloucester (http://goo.gl/maps/hxxn5), also in the 17th century.

At 67 markers one of my closest matches, 65/67, is a Beddoes, which is a Welsh surname, but he himself was born in Worcester (http://goo.gl/maps/82QzK). Mr. Beddoes told me his family, as far as he knows, has always lived in Shropshire (http://goo.gl/maps/NFc5b), which is in the West Midlands, right on the Welsh border.

As you can see from looking at the map of Britain, none of these places is far from any of the others. One of them is in Wales, and the others are in far western England, very near the Welsh border. Since two out of the three surnames are Welsh, and my own surname, Stevens, can also be Welsh, I strongly suspect my own y-dna ancestry is Welsh. I also have a 60/67 match with a Price, which is a Welsh surname.

In my haplotype cluster, which is solidly DF41+ (P312>L21>DF13>DF41), only one other person thus far can trace his y-dna mdka to the Old World. That is Self, kit 53479, who traces his y line to Melksham, Wiltshire, England (http://goo.gl/maps/tF0FR), c. 1620. There is a Welsh name, Selyf, but I am not sure that is the source of the surname Self. Although a member of my haplotype cluster, Self is 12 off me at 67 markers and 18 off at 111. Anyway, Melksham in Wiltshire is likewise not too far from the points of origin of my other matches. They are all pretty western and either in Wales or close to it.

Samuel, Beddoes, and Price are Welsh surnames. Webb is an English surname. Self is English but could be Welsh (Selyf). My own surname, Stevens, can be either Welsh or English.

Wales or western England? It's not a slam dunk either way, but I am leaning toward a Welsh origin for my immigrant y-dna ancestor. Wish I knew for sure.

alan
09-28-2013, 05:17 PM
I think the possibility of British elements that became English later culturally is high once you get into western and northern England.

rms2
09-28-2013, 05:24 PM
I think the possibility of British elements that became English later culturally is high once you get into western and northern England.

I agree. Welsh is probably right in my case regardless of which side of Offa's Dyke my ancestor came from.

Maybe time and a few more matches will tell the tale.

AJL
09-28-2013, 05:38 PM
rms2: I have the same issue with one of my early colonial lines named Grannis. My ancestor of this name came quite early to Connecticut (by about 1649) and his name may be fabricated since it doesn't seem to have come from anywhere... One of my distant Grannis cousins is kit 78473 in the L21 Project though, and I believe his DNA matches suggest the progenitor was from around Wales (his closest matches appear to be Owens and Adams, both dead-ended in North Carolina).

rms2
09-28-2013, 05:45 PM
My two closest matches at 111 markers, 4 and 5 away, have the surnames Stephens and Stevens, respectively. Stephens is my closest match at 4 off at 111. He is dead-ended in Caswell County, North Carolina, 18th century. My Stevens match, 5 off at 111, is dead-ended in Indiana in the early 19th century. Fun matches to have, and I am glad for them, but they haven't been much help in crossing the Pond. Thus far, Samuel (105/111) and Beddoes (65/67) are my best trans-Atlantic tickets, and both of them have Welsh surnames.

AJL
09-28-2013, 05:53 PM
I do wonder if this could explain why I have so many Family Finder matches in NC with no ancestors from there. A quick backgrounder here:

http://www.northcarolinahistory.org/encyclopedia/114/entry/

rms2
09-28-2013, 05:57 PM
I do wonder if this could explain why I have so many Family Finder matches in NC with no ancestors from there. A quick backgrounder here:

http://www.northcarolinahistory.org/encyclopedia/114/entry/

Wow! Thanks for that information. I had heard there were Welsh immigrants in North Carolina, but I did not know the particulars.

rms2
09-28-2013, 06:04 PM
One problem for people of Welsh y-dna ancestry is the Welsh patronymic naming system. Under it, a large number of men with the same y-dna grandfather could all have different surnames. For example, say a man named Hugh had four sons: John, Owen, Thomas, and William. All of them would be known by the surname ap Hugh (or Pugh), for son of Hugh. Their sons, however, would have the surnames ap John (Jones), ap Owen (Bowen), ap Thomas (Thomas), and ap William (Williams). Four different surnames, but the same y-dna ancestry. The confusion would only multiply with more sons, grandsons, and great grandsons, etc., and repeats of given names.

AJL
09-28-2013, 06:09 PM
Excellent point -- unless two brothers immigrated at the same time they won't likely share a surname. So you have an Adams and an Owens who are first cousins even, and you've lost sight of the common ancestor.

rms2
09-29-2013, 11:54 AM
For those of us who are brick-walled in North America or Australia or New Zealand, etc., taking close, higher-order (67-marker and 111-marker) matches and plotting their ancestral origins on the map of Europe (or Asia or Africa) is the best way to get some insight into a possible ancestral homeland. Waiting around for that one perfect match - the one who shares not only your surname but can tie you into a detailed and well-documented pedigree back to the Old Country - won't get most of us very far.

Anyone who does not have at least 67 markers would be best advised to get them as soon as possible, and 111 markers are even better for this purpose.

For those guys who have 67 or 111 markers but do not have any close matches yet, be patient, and maybe try to recruit some folks with your surname for y-dna testing, especially people who are either from the Old World or who have a solid paper trail to the Old World.

Please use this thread to tell how you have used geography and higher order matches to attempt to pinpoint your own ancestral homeland. We can all learn from each other.

rms2
09-29-2013, 12:24 PM
rms2: I have the same issue with one of my early colonial lines named Grannis. My ancestor of this name came quite early to Connecticut (by about 1649) and his name may be fabricated since it doesn't seem to have come from anywhere... One of my distant Grannis cousins is kit 78473 in the L21 Project though, and I believe his DNA matches suggest the progenitor was from around Wales (his closest matches appear to be Owens and Adams, both dead-ended in North Carolina).

I took a look at him, AJL. I see what you mean. I would recommend he go out to 67 markers and on to 111 markers. Geno 2.0 or, better yet, Chromo2 would also be a good idea. He doesn't have any close matches who have tested positive for anything downstream of L21 yet.

AJL
09-29-2013, 04:05 PM
Thanks, I appreciate your having taken the time to look over his kit. I'll try to get my cousin to upgrade. It would be nice at least to define a further SNP.

rms2
10-04-2013, 11:22 PM
Here's a link to a map of Britain showing the distribution of the surname Stevens in 1881: Stevens 1881 (http://gbnames.publicprofiler.org/Map.aspx?name=STEVENS&year=1881&altyear=1998&country=GB&type=name).

Take a look at the difference when the spelling is Stephens: Stephens 1881 (http://gbnames.publicprofiler.org/Map.aspx?name=STEPHENS&year=1881&altyear=1998&country=GB&type=name).

Since my closest 111-marker match (107/111) thus far is a Stephens, I think that may have been the original spelling, and the Stephens map seems to fit my matches better than the Stevens map. This site (http://thewelshsurnameshop.com/?page_id=47) says many Welsh Stephenses who went to America changed the spelling to Stevens, and I know I have read that somewhere else before, too. But they got Meic Stevens' stage name wrong: it's Meic Stevens (http://en.wikipedia.org/wiki/Meic_Stevens), not Meic Stephens.

AJL
10-05-2013, 12:42 AM
Very interesting, thanks!

rms2
10-06-2013, 12:57 AM
I noticed that this Stephens 1881 map (http://gbnames.publicprofiler.org/Map.aspx?name=STEPHENS&year=1881&altyear=1998&country=GB&type=name) shows the surname at its most frequent in the postal area based in Llandrindod Wells, Powys, Wales (http://goo.gl/maps/3YDmL). That is only about five miles (around 8 km) from Llanafan-fawr (http://goo.gl/maps/cWaa0), which I mentioned in my first post as the ancestral homeland of one of my closest 111-marker matches.

rms2
10-06-2013, 12:16 PM
Okay, I know I am rattling on about my own genealogical quest, but here's more geography-related rattling. I mentioned that I have a 65/67 match with a Beddoes, which is a Welsh surname. Here is the Beddoes 1881 map (http://gbnames.publicprofiler.org/Map.aspx?name=BEDDOES&year=1881&altyear=1998&country=GB&type=name). It has considerable overlap with the Stephens 1881 map (http://gbnames.publicprofiler.org/Map.aspx?name=STEPHENS&year=1881&altyear=1998&country=GB&type=name) and the Samuel 1881 map (http://gbnames.publicprofiler.org/Map.aspx?name=SAMUEL&year=1881&altyear=1998&country=GB&type=name) (Samuel is a 105/111 match of mine).

rms2
10-06-2013, 07:03 PM
I'm wondering if anyone else has used this matches+geography method to try to patch together a basic idea of his or her y-dna ancestral homeland.

When I started, I didn't know what to expect. I thought I was some kind of Germanic or the descendant of a Viking (yes, I was a "Viking wannabe"). At first I held onto that idea and resisted evidence to the contrary. After awhile, however, if one is going to be at all truthful with himself, he has to face facts. In my case, the preponderance of the evidence pointed away from the Germanic/Viking thing and toward the West and the "Celtic Fringe", especially Wales. It wasn't what I had wanted, but the truth has a way of forcing itself on you. Now I like the whole idea and embrace it. Might as well like what you get; the alternatives are misery and self delusion.

AJL
10-06-2013, 09:08 PM
I'm wondering if anyone else has used this matches+geography method to try to patch together a basic idea of his or her y-dna ancestral homeland.

Yes, very much.

My Y paper trail only goes back about 200 years, to northeast Belarus. But no members of my group have a very long paper trail.

In 2007-2008, it was generally thought (or I thought) that pretty much all R1a in Europe was Slavic Eastern European (Russian or similar).

STRs were not that informative. By 2009, though, there were SNPs to differentiate R1a-M458 from other subclades. Now, with a very intricate R1a SNP tree, my best guess is my Y line was near Alan/Sarmatian/Sabir areas as of 1500 years ago, and was in the area of northeast Poland/Lithuania/Belarus by about 1700.

rms2
10-07-2013, 11:17 AM
My wife's father was from Gomel, Belarus.

AJL
10-07-2013, 09:53 PM
Very small world!

rms2
10-08-2013, 11:42 AM
Very small world!

Indeed it is.

This thread has caused me to realize, if I didn't already, how much I have benefited from y-dna testing. Before I got involved in y-dna testing, I had pretty much the same paper trail I have now: I still cannot seem to get past my third great grandfather on my y-dna line. But now I am able to leap past needing to know the actual names and other personal details of my individual ancestors and at least have some idea where my immigrant y-dna ancestor came from. That is huge.

Don't get me wrong: I still want to get beyond my third great grandfather and West Virginia (part of Virginia then) in 1804 and discover the names and other important details of as many of my ancestors as I can. But finding out about my regional and ethnic heritage is very important to me. I have wanted to know about that since I was a small boy. Now dna testing has made that possible. I do not regret even one cent I have spent on it (well, there were a few negatives for SNPs that weren't so cool, but that's a minor complaint).

rms2
10-10-2013, 11:26 AM
I was reading that the Welsh county of Powys, in which Llanafan-fawr is located, and which seems to be the hub of all of those surname maps I posted earlier, in ancient and medieval times included Shropshire and other parts of what is now England adjacent to the current Welsh border.

This is from the Wikipedia article on The Kingdom of Powys (http://en.wikipedia.org/wiki/Kingdom_of_Powys):




The Kingdom of Powys was a Welsh successor state, petty kingdom and principality, that emerged during the Middle Ages following the Roman withdrawal from Britain. Based on the Romano-British tribal lands of the Ordovices in the west and the Cornovii in the east, its boundaries originally extended from the Cambrian Mountains in the west to include the modern West Midlands region of England in the east. The fertile river valleys of the Severn and Tern are found here, and this region is referred to in later Welsh literature as "the Paradise of Powys".


749 750

AJL
10-10-2013, 10:04 PM
I suspect you're on to something here. If there were more people from there tested you might have a very informative match.

(As a side note, I've had some fun hikes in the Brecon area, though I've spent more time in Carmarthen and Ceredigion. Great hiking in all those spots, if you don't mind getting rained on...)

rms2
10-10-2013, 11:51 PM
I suspect you're on to something here. If there were more people from there tested you might have a very informative match.

(As a side note, I've had some fun hikes in the Brecon area, though I've spent more time in Carmarthen and Ceredigion. Great hiking in all those spots, if you don't mind getting rained on...)

We are thinking of making a trip to Wales in April during Spring Break (I'm a teacher). I've heard from a friend there are still many Stephens/Stevens folks in the area of Llanafan-fawr. I'm wondering if any of them would be willing to submit a y-dna test if I paid for it. I would start with 12 markers because of the cost. If I got any good matches, I could upgrade them.

AJL
10-10-2013, 11:56 PM
Excellent idea, keep us posted!

I hear that Welsh people often have pretty good genealogical records simply because with a surname shortage, you don't want to marry your Davis/Price/etc. cousin unawares.

rms2
10-11-2013, 11:14 AM
Last night on Facebook I joined the Powys Historical Society group and posted a plea for y-dna testing, especially for men with the surname Stephens or Stevens. No response yet, but I am hopeful.

I also took a look at pages connected to Llandrindod Wells (the county seat) and the surrounding area. I found that Stephens is apparently a common surname there. I was tempted to message a young man with the surname Stephens and offer him a free y-dna test, but I figured, given the state of things in the world today, he would probably think I was some kind of predatory weirdo, so I didn't.

AJL
10-12-2013, 12:26 AM
That's probably wise. Posting the offer on a genealogically related website makes sense, but out-of-the-blue offers for DNA testing elsewhere may seem like entrapment for a civil paternity suit or something. :)

rms2
10-15-2013, 09:26 PM
I started a Facebook page for the subject of genetic genealogy in Powys, Wales:

Powys, Wales, DNA (https://www.facebook.com/groups/249550651864348/)

I don't know if anyone besides me will care, although we already have a few members. B)

I'm hoping to attract a few natives and convince them to have their y-dna tested.

kaybee
10-15-2013, 11:26 PM
That is a good idea, and gives me some food for thought!

avalon
10-16-2013, 02:10 PM
I started a Facebook page for the subject of genetic genealogy in Powys, Wales:

Powys, Wales, DNA (https://www.facebook.com/groups/249550651864348/)

I don't know if anyone besides me will care, although we already have a few members. B)

I'm hoping to attract a few natives and convince them to have their y-dna tested.

Good luck in your search for closer matches in Powys. According to the "Surnames of Wales" by John Rowlands, Stephens was a late patronymic surname in Wales. As a forename it was found in small traces in 15C Brycheiniog which is now part of Southern Powys. It does appear most frequent in mid and south Wales.

One problem you might find with recruiting Y-DNA in Powys is that it is the most sparsely populated region of Wales and Stephens folk will be vastly outnumbered by the more common Welsh surnames.

The other issue is that Eastern Wales, including much of Powys, has been subjected to a lot English influence in recent centuries, so much of the modern y-dna in places like Llanidloes, Welshpool, Newtown etc will be of English origin. Of course, there will be Native Welsh Y-DNA in Powys but it is likely rarer than in Western Wales.

ADW_1981
10-16-2013, 06:46 PM
Why is it that the L21 guys find considerably closer matches than us P312 guys (either DF27*, Z220..etc?)
What are the geographic implications with this? I wonder if it's because you guys have been on the islands longer than we have? I hate to use the word bottleneck, but it seems there is a very mild one from a YDNA standpoint.(?) Where as the older branches such as P312-DF27 were more spread out on the European continent with no close relatives at all, or at least the odds would be for close relationships to die out due to other male competition.

What would the implications be? Close y-str matches to other men with different surnames may imply an older heritage in Britain or Ireland? I don't see any reason why there would be sexual selection of L21 over P312 per-se.

rms2
10-17-2013, 12:13 AM
I think L21 is actually the youngest or one of the youngest of the P312 subclades, so naturally we have not had as much time to develop genetic distance as the rest of you have.

Also, there is British Isle bias in the database to account for.

In the case of guys of Welsh ancestry, the Welsh patronymic naming system accounts for some of the close matches.

rms2
10-17-2013, 12:17 AM
Good luck in your search for closer matches in Powys. According to the "Surnames of Wales" by John Rowlands, Stephens was a late patronymic surname in Wales. As a forename it was found in small traces in 15C Brycheiniog which is now part of Southern Powys. It does appear most frequent in mid and south Wales.

One problem you might find with recruiting Y-DNA in Powys is that it is the most sparsely populated region of Wales and Stephens folk will be vastly outnumbered by the more common Welsh surnames.

The other issue is that Eastern Wales, including much of Powys, has been subjected to a lot English influence in recent centuries, so much of the modern y-dna in places like Llanidloes, Welshpool, Newtown etc will be of English origin. Of course, there will be Native Welsh Y-DNA in Powys but it is likely rarer than in Western Wales.

Glad to see you back!

Where the heck have you been lately?

avalon
10-17-2013, 06:51 PM
Glad to see you back!

Where the heck have you been lately?

Work and family have kept me away recently but I browse the threads when I can!:beerchug:

rms2
10-19-2013, 01:22 PM
Work and family have kept me away recently but I browse the threads when I can!:beerchug:

Well, I'm glad you're around. It's good to have at least one other person with an interest in Wales.