1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Only registered members can see all the forums - if you've received an invitation to join (it'll be on your My Summary page) please register NOW!

  3. If you're looking for the LostCousins site please click the logo in the top left corner - these forums are for existing LostCousins members only.
  4. This is the LostCousins Forum. If you were looking for the LostCousins website simply click the logo at the top left.
  5. It's easier than ever before to check your entries from the 1881 Census - more details here

DNA Strategy

Discussion in 'DNA Questions and Answers' started by DavidK, Dec 9, 2019.

  1. DavidK

    DavidK LostCousins Member

    About a year ago, I purchased an Ancestry atDNA test. At the time, I was focussed on my maternal grandfather’s line, and the test produced some spectacular results.

    Turning more recently to my paternal grandfather, Edward Kendall, whose family history first got me hooked on genealogy, I was dismayed to find very little in the way of DNA matches. I had build an extensive tree of the Kendall families of Ringwood, Hampshire, and over the years, made contact with several third cousins who were descended from the siblings of my GGF, Edward’s father Charles.

    Of the three known 3rd cousins who had tested at Ancestry, Two matched as 5th-8th cousins, and the other was not a match at all. Three more Ancestry users, who connect to my tree as 4th cousins, also matched as 5th-8th. I began to doubt my tree.

    I had read Peter’s DNA Masterclass, and noted a 2% chance of no DNA match between 3rd cousins, but it wasn’t a huge consolation, as I was getting poor matches with *all* my known cousins, and zero "above average" matches.

    So I got in touch with my half 2nd cousin, Susan, who is descended from my grandfather’s half sister, Annie Julia, and sent her an Ancestry test kit. The DNA match was 199 cM across 8 segments, well above the average (73 cM), and high in the expected range (0-341 cM) for this relationship.

    I felt better about the connection to my GGF Charles, but now wondered about the link to his parents, John and Elizabeth. Charles led a complicated life, and the BMD records are convoluted. Perhaps I’d connected my GGF Charles to the wrong parents?

    Following Charles’ untimely death in 1910, Annie Julia appeared in the 1911 census, living with Charles’ sister Elizabeth and her family, but listed as “Cousin” instead of “Niece". On the other hand, another of Charles’ children appeared with his sister Alice, correctly listed as “Niece”, and two more with Elizabeth’s married daughters, correctly listed as “Cousin”. Three out of four ain’t bad. The historical record seems circumstantially credible!

    My next step was to get in touch with another 3rd cousin, Darren, who is a direct male descendant of my GGF Charles’ brother John. John is also the GGF of one of my low-matching 3rd cousins. I sent Darren a Y-111 kit, and upgraded my own from Y-37 to Y-111. Results are due in January.

    That was all before i joined this forum.

    In retrospect, I wonder if the estimates provided by Y-111 will be accurate enough to confirm a 3rd (vs 5th-8th) cousin?

    I know that Peter advises using DNA to knock down brick walls, and it becomes clear after re-reading his Masterclass a couple of times that it’s highly random, but it’s so tempting to seek confirmation!

    On the other hand, many years ago, when I was researching the Kendall families of Ringwood, I was unable to connect three main branches. So I located male-line descendants of the other two branches, and arranged Y-37 tests, which indicated we were related. I’ve upgraded them all to Y-111 which, if nothing else, will hopefully provide greater accuracy to that exercise.
  2. peter

    peter Administrator Staff Member

    The figures you've quoted are not for a half 2C but for a half 2C1R.
    If you want to know how closely someone is related to you, use autosomal DNA, not Y-DNA. Y-DNA will tell you whether you share a common male ancestor but unless you have an exact match at 37 markers it's unlikely that upgrading will provide sufficient resolution.
    There are three random elements: how much you have inherited from the common ancestor(s), how much your cousin has inherited, and what the overlap is between them. You can't change any of those factors, but what you can do is change who tests.
  3. peter

    peter Administrator Staff Member

    This is what FTDNA show for a 37 marker match with no differences

  4. DavidK

    DavidK LostCousins Member

    Ah, quite right. The match with Susan of 199 cM across 8 segments is still well above the average (117 cM), and right in the middle of the expected range (9-397 cM) for half 2C.

    It turned out, after I got Susan's results, that her son had already been tested. I hadn't recognized his name in my match list, and the estimated relationship was only 4th–6th cousin. Looking at the match of 57 cM across 3 segments, he's below the average (73 cM), and low in the expected range (0-341 cM) for a half 2C1R -- which rather proves your point about DNA, as he's definitely Susan's son!

    I was reluctant to send Darren an atDNA kit because of the poor- and non-matches with the other 3rd cousins. My thinking was to look at the Y-DNA result (since he's a direct-line male, willing to test), and then if it were exact, to look at an atDNA test.

    The other Y-37 tests I upgraded are for distant cousins, with genetic distances of 1 and 2 respectively. I wrote a blog article here about the testing we did in 2010, and was told that upgrading the tests would improve the estimates.
  5. peter

    peter Administrator Staff Member

    When you say that they're distant cousins, do you mean that they're documented cousins, or that you've only found them using Y-DNA, and have yet to find the paper trail? I can see that in 2010 you were lacking evidence, but things may have changed since then.

    If there is a paper trail that shows that you share a direct paternal line ancestor then the Y-37 tests are sufficient to verify this - upgrading the tests won't tell you anything more about your relationship to them. Even if you're still where you were in 2010 it is unlikely to help since even with a genetic distance of 3 at 111 markers there is a wide range of possible relationships:


    What might help is autosomal DNA testing, because that's not just focused on a single line. If you can show that one of your Y-DNA relatives is descended from one of the wives of your male line ancestors you'll really have something to work with.
  6. DavidK

    DavidK LostCousins Member

    Those are the same figures I see at FTDNA for my cousin who has a genetic distance of 1.

    In my tree, I've traced his line back five generations without a connection to mine, and when I adjust the FTDNA TiP report accordingly, I get different figures (which I can't seem to screenshot here) that indicate lower probabilities (eg., 83.49% at generation 8).

    That seems counter-intuitive.
  7. DavidK

    DavidK LostCousins Member

    Back in 2009-2010 I built a tree containing every Kendall in all the records from Ringwood, Hampshire, and painstakingly connected them. I came up with three main families (and lots of stragglers) who could not be connected by the records. I then found male-line descendants of each (including myself), and had them all tested.

    Indeed. It may be worth taking another run at the records. The last time I was in Ringwood (about 2009) I discovered that there was a register at the local chapel that dated back to the 1800s. I tried without success to get a look at it.
  8. peter

    peter Administrator Staff Member

    A reminder that earlier I wrote:
    Bear in mind that you could be the one who hasn't inherited very much DNA from the common ancestor - getting someone else from your part of the tree to test, ideally someone from the previous generation (and as distantly related to you as possible), could make a world of difference.
  9. DavidK

    DavidK LostCousins Member

    My grandfather had several half siblings, so one of their children would seem to fit the bill. Is that correct?

    Looking through my tree, there really is only one possible candidate. A half 1C1R born in 1952. The others are all much older (1915, 1925, 1928) so unlikely to still be alive. Of course, locating living people in the UK is extremely difficult (I've been trying to locate three 2nd cousins from West Midlands, with no success at all).

    I also have a few of my father's siblings, but they're more closely related, of course.

Share This Page