DNA Testing and Distant Cousins

The story about my great-grandmother, Elizabeth (Lizzie) Dunzinger Panther is that she was born in 1854 in New York City and after the death of her parents when she was two years old, her grandparents brought her to live with family in Burlington, Des Moines County, Iowa. I found her in the 1855 New York State Census as one of two children of Andrew and Fanny Dunzinger along with her sister Mary A. Dunzinger. It says she was born in 1854.

She's found in the 1870 US Federal Census living as a servant in Burlington in the Charles and Walburga Wagner household. I found her in the 1905 Iowa State Census where it states that she had been in Iowa for all except two years of her life.

Andrew Dunzinger is found prior to immigration to America, in Wemding, Bavaria, as is the Wagner family. So logic follows that this would mean that the family she was brought to live with would have been the Wagners.



I don't have access to the records in Wemding. I paid a researcher to research for me in the Wemding documents, which are now found in Eichstatt. The researcher that did this work for me says she can't find a connection between the families in Wemding.

A while back, I received communication from a descendant of Charles and Walburga Wagner. We did a DNA test to confirm that we are related. Well, the DNA results came back. According to the DNA results, we are not related. What a let down! Really, it shouldn't be a let down. It is new information and we should be happy to know for sure. However, we were so sure we were related, we thought we were just making it official.

But do DNA results that don't show a relation really mean we're not related? Not necessarily. DNA can prove a positive but it can't prove a negative. All it can do is make a suggestion of that negative fact.  But what does that mean and how does it work?

DNA inheritance is completely random. I believe many people think that DNA inheritance is thought of like a soup where all the components of the parents are mixed thoroughly and each child is given a ladle of a diffusion of their parents DNA. This is why people say, "I'm three quarters German and one quarter Irish" and then their DNA shows something completely different. DNA inheritance is better thought of as a pie where the pie representing each parent is sliced into random sized slices anywhere from the tiniest sliver containing a singe nucleotide pair on up to a large chunk making up nearly half of a chromosome. These randomly sized slices are taken from one parent, totaling exactly half a pie.

Pick any number of pieces that make up half a pie. That is what a child gets from one parent.

Then, for the pieces that a child doesn't get from one parent, he gets those pieces from the other parent. Two different children of the exact same parents, in theory, could have zero DNA in common. This is extremely unlikely to happen, but it is theoretically possible. So imagine in a few generations where two children of the same set of parents share less than half of same DNA. Within not too many generations, their descendants will likely share none. Of course, luck could work out the other way so that very very distant cousins could continue to share a bit of their DNA from their common ancestors. This is why the DNA testing companies, when they see a small segment that matches between two individuals, will say that you're likely 5th to distant cousins.

What are the odds of you matching cousins of varying degrees?
First cousins will virtually always show as a match in their DNA results. Let's call their percentage 100%. Even second cousins have a greater than 99% chance of showing up as matches to each other, and third cousins have a 90% chance.

Once you get to fourth cousins, the likelihood drops off rapidly. You have a little more than a 50% chance of matching a fourth cousin, and your odds drop all the way down to just over a 10% to match your fifth cousin. By the time you reach your sixth cousin, you're actually very unlikely to match them, with less than a 2% chance.

So, in the Dunzinger/Wagner connection, if we assume that Andreas or Fanny Dunzinger were first cousins with either Charles or Walburga Wagner, since I am four generations down from Andreas and Fanny, the likelihood of my matching this Wagner descendant is about around 10% if we are of the same generation descent. I believe that this Wagner descendant is at least one generation more removed, which means we actually have closer to a 2% chance of showing as a match.

So, just because we don't show up as a DNA match to each other, that doesn't necessarily mean we're not related. Maybe more Wagner descendants will eventually be tested. If so, we may yet prove a connection to the family. Until then, or until we find a documented relationship, we'll have to work under the assumption and possibly begin to investigate the possibility that they were just good friends, or, what I view as extremely unlikely, that Lizzie was part of the Orphan Trains that brought orphans from the east coast to the midwest. I believe this is unlikely because the Wagners and the Dunzingers are documented to come from the same town in Bavaria.

--Matt