DNA puzzle

and in #5

Now I'm confused.

I did mention in my initial post that I could see how D was related to A.

However, while my sister has a match with D, I don't and nor does our 3rd cousin E. And none of the matches my sister shares with D are a match with A.

Not at all - in #12 I wrote "after all my investigations into the various available trees an undocumented relationship is looking like a distinct possibility", although at the same time I am not ruling out the possibility of a documented connection.

My reply to Jorghes was in response to her question about comparing trees, and naturally this was my first approach to solving this mystery. That D and A are 3rd cousins doesn't necessarily mean that my sister is related to each of them in the same way, particularly as all the other matches she shares with D are not shared matches with A.

 

I agree. Look carefully at the addresses of people in the relevant branches of the trees you're comparing. I 'solved' one of my DNA mysteries when I discovered people across trees who were close neighbours (living next door but one to each other) in the 1911 census.

 

Yes, the 20cM limit can be a bit frustrating sometimes, but I always do it both ways in situations like this. That is, I go through each shared match in turn and check to see who their shared matches are, and that picks up some matches that doing it the first way misses.

But I totally agree with what you say about not being a match doesn’t mean you are not cousins.

I can certainly try and follow up more on D but unfortunately, although it gives me an area of A’s tree to focus on, I then have the problem of where to focus on in my tree, as I don’t know how, or whereabouts in my tree, I am related to D, except that it is almost certainly not where I have been focussing my attention so far.

Sorry, I feel like I’m being awkward here, but I am not trying to be so - it’s just how it is. But I will have another go at trying to fathom out my connection with D or at least try to narrow it down.

 

Is the connection between A and D something that you can follow a little bit further?

I have plenty of DNA results that may share DNA with my siblings, but not with me, that's the nature of DNA. The ones I find suspicious are the ones which may share DNA with me or my siblings, but with neither of my parents.

 

I did pretty through investigations into the trees of both D and A about 3 years ago, but I've fished out all my notes from that time to check through again. Back then I had no shared matches with A and my sister had only D, so it was the obvious place to start. I was then mostly looking for a documented connection as there was no other evidence to suggest I shouldn't be, although from previous experience, I always keep other possibilities in mind. And you have to start somewhere when considering an undocumented connection - otherwise the possibilities are too numerous.

My investigations back then revealed that A and D were descended from a full brother and sister, and that the sister (D's ancestor) married a man with a relatively unusual surname. The matches my sister shares with D do not get top marks for useful trees or responding to messages, but from what I could determine, the shared matches shared ancestry with D via the husband with the unusual surname - hence were not related to A.

Having read the DNA article in the latest newsletter, it may help if I clarify two things here.

1) I am very well aware that we cannot blindly assume that everything written in official records is correct and, as it happens, around 25 years ago I started writing an article, based on my own experiences, of the numerous different ploys our forebears used to cover up their indiscretions. And, as well as deliberate cover-ups, there are also many genuine errors in the records.

2) One of the two main reasons I did a DNA test was to check the validity of my tree, including checking out all my 'cover-up' theories. Mostly, my DNA has confirmed my suspicions that some of my ancestors had been somewhat economical with the truth. And referring to what Helen wrote above, it's not just the neighbours that need to be checked out - boarders, lodgers and housekeepers are also key players to consider.

 

Oh yes, I can confirm that from my own research!

 

The article in the recent newsletter - What's in a name? One member's DNA journey - does provide a very good illustration of how to investigate certain unexplained DNA matches, and this is the sort of approach I would have followed in similar circumstances.

However, I feel the situation I've described here is not really that similar, which may be why similar methods don't seem to be working very well for me:

1) I don't have any obvious nearish areas of my tree which are unsupported by either my DNA or my sister's DNA or, mostly, both. Two of my gt gt grandmothers were illegitimate with fathers unknown, but otherwise all my gt gt grandparents, and most of my 3 x gt grandparents, are well supported by DNA matches with documented cousins. The least well supported at gt gt grandparent level is my paternal grandmother's paternal grandfather, where I have matches supporting his maternal ancestry but haven't yet identified any clear ones on his paternal side.

2) I don't have a cluster of unknown matches associated with my unexplained match A, nor do I have any other relatively close matches (greater than c 35 cM) that I cannot assign to a particular grandparent, although there are a few without trees where I cannot pinpoint an exact connection. I do have clusters of more distant unknown matches, including one quite large cluster - but I don't share any of the matches in that cluster with my sister, and those that have trees show ancestry almost exclusively in the US which makes it trickier to check and extrapolate their trees.

The net result of this is that I don't have any clear areas in both my tree and A's tree to follow up on. If I run with my 3rd cousin E also sharing unexpectedly high DNA with A, then I don't know where to search in A's tree. If instead I narrow down the search in A's tree to the line shared with D, then I don't know where to look in my tree.

 

I guess the thing here is that I don't really know if the match is important or not. As it's the closest by far of any match for which I have been unable to identify a connection, it has been niggling me since I first got my DNA results over 3 years ago, and I thought it was about time I had another look.

And now that I have some shared matches, and knowing that a 3rd cousin shares a similar amount of DNA with the match, I had more focus in doing this. There was (and maybe still is) the possibility of breaking down the brick wall of the unknown father of our illegitimate shared gt gt grandmother.

However, it has seemed to me that the discussions here were nudging me towards the distinct possibility that one of my nearish ancestors might not be the child of their documented parent(s), and naturally I wasn't going to just ignore that without investigating further - building a tree that is as accurate as possible has always been a top priority for me.

I am still not disregarding this possibility, nor am I disregarding other types of undocumented relationship, but without much clearer evidence to the contrary, I think my tree is more likely to be correct as it stands.

 

I'm not sure 'importance' has to be the only factor for investigating a DNA match. This unidentified close match was intriguing me - more so now than 3 years ago for reasons described above - and I wanted to see if I could now get any closer to determining our connection. Doing so once every 3 years doesn't seem to me an excessive amount of time to invest in indulging my curiosity, and something useful might have emerged from it. Also, the main reason this particular investigation has taken up rather too much of my time was because I allowed myself to be diverted into something of a wild goose chase, when logically I already knew an NPE (or similar) in this area of my tree was somewhat unlikely (#14).

It may be worth reiterating for the benefit of others in a similar situation - an unexpected and unidentified close-ish DNA match does not of itself indicate there is an issue with your documented tree (such as an NPE), so unless there is other evidence to suggest something is amiss don't waste time looking for one. Most likely there isn't an issue in your tree but even if there is, there is little chance of finding it unless you can significantly narrow down the possibilities.

 

I thought this might be a good time to give an update on this puzzle. To recap briefly, I have an unidentified DNA match (A) with whom I share around 135cM, and with whom my 3rd cousin also shares a similarly high amount of DNA. This 3rd cousin shares my gt gt grandparents Thomas & Jane.

I haven't looked much more at A, but two new DNA matches shed a bit more light on things. Firstly, I have another documented 3rd cousin match descended from Thomas & Jane, and she too is a match with A, though I haven't been able to establish how much DNA they share. But that makes 3 of us (plus my sister) descended from Thomas & Jane via 3 different siblings, all related in an unknown way to A.

The other new match is a documented 3rd cousin once removed, descended from Thomas's sister, who is likewise a match with A, sharing around 40cM.

Taken together, these new matches suggest to me that an NPE between me and Thomas and Jane now seems even less likely, but also that A is more likely to be related to me via Thomas than via Jane.

As well as this, I have been looking in more depth at my DNA matches who appear to be in the 'Thomas & Jane' area of my tree. I have so far identified 17 independent* matches (9 of those between 20 & 77cM) with documented cousins related to me via Thomas - 3 descended from Thomas, another 3 from Thomas's parents and the rest from his grandparents (via at least 5 of their children).

From all of this, it seems to me pretty likely that Thomas is genuinely my ancestor and that my documented tree in his line is correct. So if match A does indicate an NPE in my ancestry, I'm pretty confident it does not involve Thomas or his father.

* Independent indicates that where 2 or more matches are closely related to each other (eg siblings) I have counted them together as just one match

 

Nearly two years on, I thought it might be worth adding an update here which I think more or less concludes my puzzle (this puzzle being summarised at the start of my previous post). New matches bring new information, and most importantly here, that my mystery match A is related to both of my gt gt grandparents Thomas and Jane.

To cut a long story short, and all things considered, it seems my initial suspicions of a male in my tree being responsible for an NPE in A's were correct. I had wondered if Thomas might be the man in question, but now I know that A is also related to Jane, one of their sons is more likely. If so, this would make us 2nd cousins once removed, which is one of the possible relationships indicated by the amount of DNA both I and my 3rd cousin share with A.