Something is puzzling about Matches with Other Surnames

[jablair]

John & Richard,

Perhaps I can give some examples of why you might match 25 for 25 with someone with a different surname, using my own Blair study. As in the case of many Scottish surnames, the Blair surname has "mixed" with other surnames.

The Blair surname goes back to approximately 1165 when King William I of Scotland conferred to a Norman, Jean Francois, the title Barony of Blair. Jean Francois' son adopted the surname "de Blare" and became the Blair of Blair. The title Blair of Blair continued through a direct line of male Blair descendants until about 1700 when the only surviving child of the Blair of Blair was a daughter Magdelene. Magdeline married William Scott, who adopted the Blair name and coat of arms and became the Blair of Blair. Any descendants of William Scott Blair would carry the DNA of Scott though they carry the name Blair.

In 1600 Alexander Blair, a direct descendant of John Blair (the Blair of Blair) and Grisel Sempill, but NOT heir to the title, married Elizabeth Cochrane, the only daughter and heiress of William Cochrane of Cochrane. Alexander adopted the name and arms of Cochrane in lieu of Blair and became the ancestor of the Earls of Dundondonald. Any descendants of Alexander Blair Cochrane would carry the DNA of Blair though they carry the name Cochrane. Their DNA would also link them to the original Blair of Blair.

These are just two case of name adoption in Blair surname. I'm sure further research would reveal others.

Obviously, as Project Coordinator for the Blair DNA Project, I am most interested in any DNA studies performed by Cochrane or Scott, as matches to these names could provide valuable insight into Blair genealogy.

There is also the case of adoptions. I am a 25 for 25 match with a participant with the surname Philbrick. In our case we know that Philbick's great grandfather was born a Blair but adopted by Philbrick. We also know with almost certainty who our common ancestor is. If Philbrick did not know is great grandfather was adopted we might not have made the connection.

Another participant in our project does not have a Blair surname (I'll call him Smith). His gr-gr-gr grandfather was born out of wedlock and was given his mother's maiden name even though his father was a Blair. Smith's DNA will be Blair. If he searched for matches with other Smith's he might never find any.
In both of these two "non-paternal" events the participants knew of there existence. But how many other participants in DNA projects have similar "non-paternal" events that they are NOT aware of?

John, you say you have "very extensive and well documented records over the last 12 generations" and that you "Have no doubt about the paternity of any of those in my documented records." I'd say you are much luckier than 99% of the rest us tracing our ancestors, considering how sketchy most of the older records are. But even if you are correct about the last 12 generations, what about generation 13 or 14 or 15. Are you certain that no "name change" or "non-paternal" event did not occur?

Considering the Blair name goes back about 30 generations and all that has happened with just one line of the name how can you be certain that similar changes did not occur with your surname?

John

[AngelaC]

Re: Re: Re: "Something os puzzling about Matches with Other Surnames".

Originally posted by JohnSanford
Hello Angela,
I have very extensive and well documented records over the last 12 generations of my children's ancestry. Have no doubt about the paternity of any of those in my documented records. Would like to understand better the 'Merit' and actual value of these DNA tests for Genealogy purposes. Just how does this information add to, improve, and enhance our family history searches since about 1500 AD. Sincerely, John D. Sanford, of CA.
-

Hi John, like JA Blair, I think a good way of showing you what this DNA testing can do, is to provide some examples of why I am getting some of these genetic testing done, and what I hope to get out of it for my family tree. My mention of paternity testing was more for Richard, since I think he used an example of people without a surname, using these tests to find out what their paternal family surname might be.
Firstly, if you haven’t done so already you might like to look at: JA Blairs website, its quite good. http://blairgenealogy.com/dna/dna101.html - it probably explains the science behind the tests better than I have.

At the moment I am getting my mitochrondrial DNA (mtDNA) tested, and I am getting my paternal line tested - as I am female and don’t have a Y-chromosome, so my father kindly agreed to have his DNA tested. Both samples are in the lab at the moment.
I am getting my mtDNA tested out of anthropological curiosity, in my example it probably won’t specifically answer any family tree questions. The earliest ancestor on my direct maternal line (my mothers mothers,mothers,mothers, mothers, mothers mother) lived in Perthshire, Scotland in the mid to late 1700’s. The DNA sequence change (mutation) rate on mtDNA is very slow, so the last time there would have been a change in my ancestral line in the DNA monitored by the tests might have been thousands of years ago. There is a very slim chance that I might get a database match with a relative. When you send off a DNA sample to the lab, you are able to choose whether you want to allow people who match your results to be able to contact you. If you sign this release form, then you will be able to contact those people you have an exact match to (but of course, only those that have chosen to have their contact details available to matching people).
If I match any contactable people in the database, we might compare our family trees, and there might be a small chance that we share a common ancestor - in that case both of us would have found a long-lost third, fourth, fifth or sixth cousin. However, it is more likely that the family relationship between myself and the matching people is far too far in the past to make any matches on a family tree, - in which case I’ll be satisfied just with just having my anthopological curiosity answered. There have been a lot of anthropological based studies looking at the different proportions of mtDNA haplotypes in different populations, which reflect very ancient population migration patterns. I won’t get into this in more detail yet (but if you’d like to know, let me know). I’ve been reading a lot of the scientific papers on what proportions of haplotypes are found in different populations, and it suddenly seemed weird to not know what haplotype my mtDNA is.
An example of how mtDNA might help in a family tree sense is the fact that one of my great great great great great grandmothers was a Jamaican of mixed heritage. We cannot say for sure if this was African or Native American. Unfortunately she only had sons - but if she had a daughter, we could have tried to track down a female line descendant (ie her daughters daughters daughters daughters daughter) to see if their mtDNA matched either a African Haplotype or Native American.

I am getting my paternal Y-chromosome line DNA tested, initially to answer some basic questions about the origin of my family surname. My family surname is Cone, which broadly has three origins - one origin is Jewish (from Cohen), one origin is Scottish, and one origin is German. My earliest 100% confirmed paternal ancestor came from Suffolk, England. We don’t currently know which of these large groups the Suffolk Cones are related to, or whether they are a fourth group. I am currently in the Coon surname study which encompasses related surnames of the possible German origin, so in a few weeks I’ll see how closely there is a match.
If my fathers Y-Chromosome origin is European in origin (Scottish or German), the haplotype will most likely be R1b or R1a, if it is Jewish in origin then the haplotype would be J. If my fathers Haplotypes turns out to be J (especially if the markers match the Cohanim haplotype), then I would probably leave it at that, and conclude that my Cones were originally Cohens. On the other hand it is probably more likely that my Cones are European in origin.
If there is a huge difference in the 25 marker test, to others in the surname study I can perhaps safely assume that my paternal line is not related to the German Cones - but I’d keep an eye on the study as they have more people tested, since they might have other “strays” tested that are not related to the main family line tested, but might match my fathers Y-DNA line. If the match is off by a few markers, then I’d perhaps be related to the German Cones, but at that point I wouldn’t have any definite answers of how related my fathers line would be to the main members of the study (and I would have no answers about how related my paternal family line is to any of the Scottish Cones).
At this time I’d have to have a sit down, and see if my father has any matches in the database. On average, a complete match in the 25 marker test means a 50% likelihood of being related in the last 7 generations [see bottom of posting for explanation of figures]. We might be 7th cousins, or in some cases the relationship might be further back. If we are related within the last 400 years, we will likely have the same surname. However, if their common ancestor was before surnames, then we might have completely different surnames. If any of the mismatched surnames are similar like Coan, Coon, Cowan, Cohen etc., this might reflect changes in spelling over time). For any surnames/surname variant matches, we might compare the geographical origins of our ancestors, and look at our respective family trees. We might discover a common ancestor, or might not. If we cannot find a common ancestor, we might conclude that, yes we do have a common ancestor but concede that we’ll never know where the exact link is. If, however, my father does not have any exact matches, then I’ll have to have another sit down and decide what to do next. For instance it might be that there has been a change in one of the markers in the last few generations. In this kind of situation, it’s easier to figure out how people are related if you are part of a large surname study, and the more people that are tested for the surname study, the clearer the relationships will become, and where in the tree a marker change occurred. I might decide to find other Cones who might like their DNA tested as well (for instance it would be good to get other people from the same documented Cone line as my father tested, and some of the American Cones of the Scottish Cone origin, as well as some Coans whose ancestors used to be Cone in Suffolk. If my father doesn’t have any close matches within the surname study that we have joined, I might have a chat with the group administrator, and we might decide whether to include them in his study, or whether I should start a separate study. As I say, I’ll know better what I’m doing when my fathers results come in in three weeks……

Hopefully this been helpful ,

Angela .

[The next paragraph goes over some of the science and statistics that the scientists use to come up with the “chance of being related” figures, - I’m not sure how understandable the following paragraph is to the average person, so you might want to skip it.]
This “50% likelihood of being related in the last 7 generations” figure has been calculated by the Scientists using statistical probabilities based on the average marker change/mutation rate. On a Y-chromosome there are thousands and thousands of different genes. In the large part, a mans Y-chromosome is identical to their fathers - however, every generation there are very very small changes to the genetic information. The scientists calculate a change/mutation rate of 0.002 for the genes/markers they use for these family tree studies (other genes have different change rates) - which means in any one gene/marker examined a change will occur from father to son on average once every 500 generations. This also means, if we could test 500 markers/genes (of a type suitable for genealogy), on average there will be a change in one marker/gene between father and son. What does this mean for the 12, 25 & 37 marker tests,..well (assuming my calculations are correct - It’s been 14 years since I last did statistical probability calculations in high school!!).
For a 12 marker test, a change will be seen in any of the markers on average once every 41 generations (97.6% chance that father & son will match), For a 25 marker test, a change will occur on average once every twenty generations (95% chance that father and son will match), For a 37 marker test, a change will occur on average once every thirteen generations (92.6% chance that a father and son will match).
The tables and charts that Max referred you to, use the same mutation rate to calculate the probability of being related to a person with an exact match within a specific number of generations since common ancestor.

[JohnSanford]

AngelaCP = Thank You for Sharing great Information, with us!

-
Angela, Your responses have been very good. You apparently are well informed about the value of, and the use of, the different kinds of DNA Testing. You have helped me a whole lot, by your sharing, and I am sure you’ve helped many others who look at these messages, to much better understand these services. You have helped me a lot to understand the tests, and your enthusiasm for them, and I thank you for it. We see that hundreds of viewers have seen your messages, indicating that many others probably benefit from your information, too. Sincerely, John Sanford, of San Jose, CA.
- -

[AngelaC]

Re: AngelaCP = Thank You for Sharing great Information, with us!

Originally posted by JohnSanford
-
Angela, Your responses have been very good. You apparently are well informed about the value of, and the use of, the different kinds of DNA Testing. You have helped me a whole lot, by your sharing, and I am sure you’ve helped many others who look at these messages, to much better understand these services. You have helped me a lot to understand the tests, and your enthusiasm for them, and I thank you for it. We see that hundreds of viewers have seen your messages, indicating that many others probably benefit from your information, too. Sincerely, John Sanford, of San Jose, CA.
- -

Thank you

[JohnSanford]

Re: Something is puzzling about Matches with Other Surnames

- - I have been interested in your ‘Something is puzzling . .”, thread, since it was first posted 3 months ago. How do you feel about it now after the time has passed and many knowledgeable persons have responded to that situation and the question it suggests? Have you learned more about it now, and are you convinced that the tests are in fact correct, useful, and helpful to you, to further develop your own family records? There have been many fine responses among the 30+ already noted from experienced persons who were willing to share information. The thread was viewed over 1938 times with some interesting responses. I found it most interesting, and wonder if you may still be ‘puzzled’ ? Sincerely, John Sanford, of CA.

Originally posted by Richard
Something is puzzling me about matches with other surnames.
I get my Y chromosone from my father and he get his from his father and so on... And the purpose and point of the Y- DNA test is to see if you match other individuals with the same surname. So how can I match other males with a different surname? . . . . . I just want to know about matches with other surnames.

[Richard]

Re:Something is puzzling about Matches with Other Surnames

John,

You ask me if I'm still puzzled about matches with other surnames?

My answer is, Yes, and I probably always will be on that point.

You ask me if I have learend more about it now(I assume you mean matches with other surnames)?

My answer is, Yes, but what I have learned still doesn't thoroughly explain matches with other surnames.

You ask me if I'm convinced that the test are in fact correct?

My answer is, Yes, if an individual goes all the way to 37 markers and has records to substantiate the results.

You ask me if I find they are useful(I can only answer for me)?

My answer is, No, because I have found no match that is within my own paternal line as of this date.

You ask me if they have helped with my own family records?

My answer is, No, because my records are complete and the results thus far have not yielded anything that can be added to my family records.

Richard

[AngelaC]

Re: Re:Something is puzzling about Matches with Other Surnames

Hi Richard,

I'm just curious to know what your results were, and which marker test you took - if you fall into what is known as the "atlantic modal haplotype" that might explain a few puzzling things for you. The Atlantic modal haplotype is based on 6 of the DYS markers, and falls within the Haplogroup Rb1.

The "Atlantic modal haplotype" is extremely common, - and results can be a bit harder to interpret because there are so many people with it.
For people who fall into the Atlantic modal haplotype, the 12 marker test will probably only tell you whether you are related in a anthropological sense (eg. over the last few thousand years).
Quite often people will match at the 12 marker test (or be one marker off), and then find out when they upgrade to a 25 or 37 test that they are not related in a genealogical sense.


Angela

[Richard]

Re:Something is puzzling about Matches with Other Surnames

Angela,

I don't know if I can post my actual results on the forum?

What you have stated in your post is correct about the Atlantic modal haplotype.

On a 12 marker test I am as common as the day is long. And when I got to the 37 marker test I am the oddest cookie in the jar.

I'm not trying to be critical but that is the way it turned out.

I also would suspect that there are going to be a lot of individuals that probably fall into the Atlantic modal haplotype. It apparently is an extremely large group.

But as far looking at things from an anthropological sense compared to a genealogical sense that is something that could very well be the problem.

I understand things from a genealogical sense but I definitely don't understand things from the anthropological sense.

Richard