Select Page

One wife or two

Reader David Condit has a mystery in his family. The evidence is clear that his 2nd great grandfather had nine children. What isn’t clear is how many wives he had. There’s a family story that the mother of the last three children was a different woman — a later wife — than the mother of the first six children.

David himself is descended in a direct female line from a daughter who was one of the older children, and he has cousins a generation younger who are descended in a direct female line from a daughter who was one of the last three children. So, he asks, can DNA testing tell whether there was one female ancestor or two, and are he and his cousins close enough in generations to his 2nd great grandparents and to each other for the testing to give them answers?

The answer is a qualified yes.

Let’s deal first with the kind of testing needed here. Remember that there are basically three types of DNA tests and only one is impacted very much by how many generations have passed since the likely common ancestor or ancestors lived.

That test — the one where generations matter — looks at autosomal DNA (atDNA)1 and it’s used primarily across genders to locate relatives — cousins — from all parts of a family tree.2 The likelihood of a match drops off from generation to generation, so where first and seconds cousins have an extremely high chance of matching each other, third cousins will match only 90% of the time, fourth cousins only about half the time, and fifth cousins only about 10% of the time.3 Because these test takers already know they’re cousins, this test isn’t the right one for this question.

Another type of DNA testing looks at Y-DNA, the type of DNA passed in a direct paternal line from father to son to son.4 Because Y-DNA doesn’t change much from generation to generation, the number of generations usually doesn’t matter a great deal in Y-DNA testing, so that’s not an issue here. But because we’re looking at a female ancestor, who doesn’t have any Y-DNA to pass on to her descendants, this test won’t help here at all.

So the focus here is on mitochondrial DNA (mtDNA), the type of DNA passed from a mother to all of her children but that only daughters can pass on to their children.5 Like Y-DNA, mtDNA changes very little from generation to generation, so that people many generations apart can still match on this test.6 And as long as David and his cousins — male or female — can trace descent through an unbroken female line from the wife (or wives) of David’s 2nd great grandfather, the mtDNA test has a good chance of telling them what they want to know.

Now, assuming that all of the children of David’s 2nd great grandfather had one mother, then all descendants in the direct female line would all have the same mtDNA. That would include David, a 2nd great grandson whose mtDNA would be the same as his mother, grandmother, great grandmother and 2nd great grandmother. And it would include his cousins, male and female, whose mtDNA would be the same as their mother, grandmother, great grandmother, 2nd great grandmother and 3rd great grandmother.

Here’s what the descendancy of that one woman’s mtDNA would look like with David being the final male on the left hand side and his cousins being the final male and female on the right hand side:


Because all of the mtDNA flowing from the one single wife of David’s 2nd great grandmother would be the same down these generations, those tested — the ones at the bottom on each side — would show the same mtDNA.

But if only the first six children of David’s 2nd great grandmother were the children of a first wife, and the last three children were the children of a second wife, then each of those two women would have her own mtDNA and would pass that on to her children. Each of those women’s daughters would then pass the individual mtDNA on to their children and so on.

Here’s what the mtDNA descendancy would look like in that case, with the first wife being the ancestor of the people on the left hand side (David again being the final male on that side) and the second wife being the ancestor of the people on the right hand side (his cousins again being the final male and female on that side):


The differences in that case are pretty stark, and in the usual case mtDNA will readily show the differences between the descendants of children of two different mothers.

So why is the answer a qualified yes? Because there is one possibile situation where an mtDNA test could result in misleading results. That’s the situation where the mother of the older children is related to the mother of the younger children in their direct maternal line. A classic example is where the second wife is a sister to the first wife. In that case, here’s what the lines would look like:


This would also be the case where the two women were cousins, each descended in a direct female line from a common female ancestor.

That’s why DNA always, always, always has to be used in tandem with traditional paper-trail genealogy. It works with it, not instead of it.

Let us know what you find out!


  1. ISOGG Wiki (, “Autosomal DNA,” rev. 8 Feb 2012.
  2. See ISOGG Wiki (, “Family Finder,” rev. 8 Feb 2012; and ISOGG Wiki (, “Relative Finder,” rev. 8 Feb 2012.
  3. Judy G. Russell, “Autosomal DNA testing,” National Genealogical Society Magazine, October-December 2011, 38-43.
  4. ISOGG Wiki (, “Y chromosome DNA test,” rev. 26 Jun 2012.
  5. ISOGG Wiki (, “Mitochondrial DNA,” rev. 30 Jul 2010.
  6. See generally “How many generations back does my mtDNA test trace?,” mtDNA Results FAQ, Family Tree DNA ( : accessed 7 Jul 2012).
Print Friendly