This post is intended to get you to think about the possibilities in your own Tree – it is not about the term NPE (there are many variations); and it is not about the historical percentage (there are NPEs, but not many good studies on the NPE percentage). For this blog, I’m using NPE to mean an Ancestor who by most accounts is in our Tree, but it turns out that “Ancestor” is not our biological Ancestor.
Three assumptions for this post (in order to get to the thought process):
1. NPE is a paper-trail/genealogy Ancestor, who is not the biological Ancestor.
2. The rate is 2% – there are many studies (cf: https://isogg.org/wiki/Non-paternity_event )
3. I’m going to count each “box” in our Tree as an Ancestor (recognizing that there may be some Pedigree Collapse and/or Endogamy)
The mumber of our Ancestors doubles every generation. At 8 generations back we have 256 Ancestors at that generation and 510 total Ancestors all together. At a 2% NPE rate overall, that would mean about 10 of our Ancestors were NPEs. If you read the ISOGG article above, you’ll find a slew of reasons for NPE (in addition to infidelity). As a genealogist with many Colonial Ancestors, I believe the instance of one or both parents dying in those times, and someone else raising the orphaned children as their own (with no paperwork) was more prevalent than we know. And probably many other variations…
So if the paper-trail doesn’t help, how would we ever know? I think our DNA Matches can sometimes give us a clue.
From my experience tracking Matches (see: https://segmentology.org/2021/12/19/segmentology-common-ancestor-spreadsheet/ ), I can tell you there is a clear correlation between family size and number of DNA Matches – the larger the number of surviving children (with families of their own) the more DNA Matches we’ll have. And the obverse: with a “skinny” family of only one child, we get no cousins, and if there are two children (one is our Ancestor), any Matches would descend from the other child, and there would be far fewer Matches. Some other factors may also come into play, such as distance back to the Ancestor and recent immigration, so we must take those into consideration as well.
So here are two scenarios to think about:
1. A large family with relatively few descendant Matches – maybe one or both of the “parents” are NPEs.
2. A Surname comes up repeatedly in Matches – say within a Shared Match Cluster, or a Triangulated DNA Group. I now have two of these – in each case virtually all of the Matches are from the same Tree. Clearly, I should descend from this Surname – I just need to find out how (and if it involves an NPE or a Brick Wall).
As I said at the top of this post, this is just to get us all to think – to keep an open mind – to follow the DNA signals – and to understand that we probably have some NPEs in our Tree. Watch for the clues….
[30B] Segment-ology: How Many NPEs Do You Have? – by Jim Bartlett 20220325
segment-ology 
Jim, please rephrase your explanation of the “clear correlation between family size and number of DNA Matches,” specifically the “skinny family” example. I’m not in agreement with “…a ‘skinny’ family of only one child, we get no cousins…”
LikeLike
Gregory – a family of only one child – that child has no siblings – cousins are descendants of two siblings. I don’t understand how I could have any cousins from the stated, “skinny” set of parents. I can have cousins from the next generation up or down, but not from this generation… (I think) Jim
LikeLike
I agree, cousins are descendants of two siblings. “A family of only one child” does not make sense…are you saying, if neither father nor mother of a family have siblings, their children will not have first cousins? If yes, I agree.
Greg
>
LikeLike
Greg, My father was an only child – so I have no first cousins. Although, in most cases in our ancestry, it was unusual for a married couple to have only one child, it did happen sometimes. When two of my Ancestors had only one child, that child was my Ancestor. Using Ahnentafel numbers my Ancestors 26 and 27 had only one child: 13. So I have no cousins from Ancestors 26 and 27. I have lots of cousins from 52 & 53 and 54 & 55, and from 13 who married 12. I once got a ThruLine with a “cousin” back to 26 & 27, but quickly determined the TL was incorrect. In hindsight – perhaps I shouldn’t have mentioned it, because it doesn’t happen much (but, it is one of my Quality Control Checks). Jim
LikeLike
After nearly 50 years of genealogy and 15 years of DNA I have been expecting an NPE in my direct line somewhere. So far the score is zero confirmed, but some suspects still not cleared.
There was someone chasing my family who was looking for a couple with almost the same first names as some ancestors, but the family he really needed was one suburb away.
In my lines there is a cousin match that works if the mother is NPE, and the child has come from the father’s brother and his wife. (There may have been problems around the time of birth and temporary care became permanent, perhaps. There is no other likely link for that size cM match.)
Two of my lines of 16 great great grandparents have zero DNA matches so far, across all the major labs. They each have at least one member who is very highly documented, to the extent that their stories at one time made them household names. Most of their descendants feel they know their stories very well, so maybe they see no need for a DNA test.
Furthermore, one of these lines definitely had a surname change in two successive generations. The second surname change was trivial, but the first was disruptive. And worse, the original surname was a very, very frequent one in that area. I do have some matches to that surname but the Common Ancestor seems to be from before 1650 and for most matches before 1600.
The other line has extensive conventional genealogy due to one man marrying a second time bigamously, and the two sets of descendants trying to make sense of this. But no DNA match that I can find. At least not that far back. So maybe there is an NPE in that line. I need to persuade a few of the documentary matches to test their DNA to check.
LikeLike
Christopher – Thanks for sharing. There are many stories out there – most of them may never be resolved. But the ones that involve the Ancestors of a genetic genealogist (or Segmentologist;>j) stand the best chance of discovery and resolution. The DNA is a very powerful tool, but we still usually need some other records/stories to fill in the blanks. Jim
LikeLike
NPE or bad luck?
My 3rd great-grandfather, Clark Smith Sr, is my brick wall. On paper, I found a 4th cousin, Dorothy, with a Clark Smith Sr common ancestor, but we aren’t DNA matches. Okay, that can happen around 50% of the time. So, then I started testing more people, 3 of my siblings, a first cousin and a third cousin. None of us match Dorothy, but three of us have a small DNA match with Dorothy’s 2nd cousin Bart.
It would be nice to have a DNA Painter like tool that would tell us to keep testing, or accept that it is an NPE and try something different.
Cheers,
Jim Smith
LikeLike
Jim, With what you present, I’d say NPE – for sure. That is presuming USA ancestry. For my mostly Colonial Virginia ancestry, I have an average of 40 Matches with Common Ancestor for each of my 4C Ancestors – more for 5C Ancestors. See my post: https://segmentology.org/2021/12/19/segmentology-common-ancestor-spreadsheet/ I have compiled over 4,000 Matches with Common Ancestors – each Match with at least 3 of their descendants from the CA. At Ancestry you can search for a Surname, even SMITH, with a specific State and county (usually narrows it down). Given average family sizes for the times, you should get a lot of Matches with Common Ancestors.
Also be sure to have most of “Clark Smith Sr” children in your Tree to help AncestryDNA find ThruLines – even in Private Trees.
If you try both of these and don’t start building a list of CAs, I’d try something different. Jim
LikeLike
Hi Jim,
Good article, good comments. The NPEs are always interesting.
A while back I gave a webinar for the Legacy people based on an NPE that I had found in my family tree in 1870. This was really a textbook case and it made for a good story. I called it “DNA Opens Pandora’s Box.” I have given that same presentation to several different genealogy societies, and every time I do there are always several people who come up to me afterward and tell me the stories of their NPEs.
Jim Baker
Rocklin, CA
LikeLike
Jim, I’ve had similar experiences. What is your opinion: NPEs overall are more or less than 2%? Jim
LikeLike
I think your 2% is a good number, based on my own group.
LikeLike
I always wondered why I had so many “cousins” (over 5000) but no surname matches at the Y-67 (FTDNA). When I did the Big Y-700 I decided to confront the issue. I now have the NPE narrowed down to a set of brothers, and probably one quite specifically. I did this by using finding a “current surname” cousin I could take way back and using records of colonial times and settlements to identify the circumstances of the “event” – almost with certainty. I’m still collecting data and working with a cousin of the “new surname” to make it more and more certain 🙂
Bill
LikeLike
Bill, A great research effort. BZ. One of my favorite stories about the BARTLETT Y-DNA Project c2002 – four researchers who descended from 4 brothers (about 1800 time frame) all got Y-DNA tests. Three were exact matches. Long story short, one matched the farmer next door…. Jim
LikeLike
Jim. It is likely, but you need not descend from a surname that comes up repeatedly. It could be the surname of the husband of a sister or niece of your ancestor. In either case, research into that surname would be worthwhile as it could add a new branch to your tree.
LikeLike
Louis, I know to be careful. And I build the Trees and keep notes. Note that the husband, or other relationship, of a bio-Ancestor, would be a good thing – it would really narrow down the search field. The cases I have so far show me with dozens of Matches, who descend from multiple children of a couple. In such a case the DNA could be coming from him or her or both. If it’s not too far back (which is subjective depending on the cM threshold you’re willing to use), there should also be Match-cousins to his parents, or hers, or both. This is one way to add confidence… I believe almost all of my Ancestors within about 8 genenerations back (if they had several children), will have passed down sufficient DNA to me to find some Matches (or maybe to me and my sibling…) That’s about the area the “Porcupine” chart begins to kick in… Jim
LikeLike
for one of my lines that I have a unknown ancestor I do have Y tests that confirm the surname and the immigrant ancestor. Possibly Y tests could be used more to verify a NPE.
LikeLike
RH, Again, I agree with you. I have determined (through testing and by examination of the Surname Projects at FTDNA) the Y-DNA signature for a number of my surnames – my goal is to determine as many as I can. In at least two cases they show that is not true all the way back in a genealogy time frame. In most of the cases, it shows the Y-DNA *is* true for the genealogy timeframe. By the same token, I’m also always on the lookout for all-female lines and mtDNA “signatures” – somewhat harder, but I’m still collecting them… Jim
LikeLike
but if the female line has a brother and male to male descendant then a y test could be done. yes there is mtdna but it changes so little I wonder how good a mtdna would be. Or maybe you think that is not a problem ?
LikeLike
RH, Both the Y and the mt change very, very little over a genealogy timeframe of say 10 (or even 20) generations. The problem is finding two all-male or all-female lines (from different children of the Ancestor) down to living people today. We need to lines for the “triangulation” back to the Ancestor. Easier with the Y, but still it’s hard to find… Jim
LikeLike
I have a thought but no proof. I tend to think that NPEs are not evenly spread. I think they may occur more in some family groups than others.
LikeLike
RH, I tend to agree. I have a multi-great uncle, who was a bachelor until his 50s – we’ve now found multiple cases where he was the probable father….
LikeLike
I am thinking more social and economic factors might be some reasons for NPE’s. Families tend to share the same environment so might be why some families may have more occurrences.
LikeLike