Documenting my family's past for future generations. My family tree includes the Smith/Mansell families of Alabama and Oklahoma, the Castle/Day families of Kentucky and Oklahoma, the Wheat/Ming families of Texas and Oklahoma, and the Bell/Roberts families of Mississippi, Tennessee, and Oklahoma.

Tuesday, September 17, 2013

Questions and More Questions


My brother and I--about 1960
My brother and I don’t look anything alike. I certainly didn't inherit his athletic body type. According to Family Tree DNA, however, we are full siblings, not that I had any doubt. It has been most interesting to compare our matches with cousins from different branches of our family. I had even wondered how productive it would be to have him take the Family Finder test, because I assumed our results would be pretty much the same. Not so. While we match many cousins to the same extent, we have found that his matches with some are much greater than mine.

At Family Tree DNA you find out right away how many total centi-Morgans (units of distance along a chromosome) you have in common with your match, how long your longest stretch of identical DNA is, and their estimation of how many generations away the two of you share a common ancestor (MRCA—Most Recent Common Ancestor.) You can then upload your results to Gedmatch and compare with cousins who tested at other testing companies, such as Ancestry.com or 23 or Me.

For both my brother and me, one of our very closest matches is to a Castle cousin. But while I have a total of 83.33 cM’s in common with her, my brother has a whopping 127.71. My brother’s longest strand with her is 51.61; mine is 55.03. I’m finding out that’s a large strand. FTDNA and Gedmatch identify longest strands down to 7 cM’s as being from probable relatives. The fact that we have such large identical strands with our Castle cousin shows that we are pretty closely related. In fact, I know that we are 3rd cousins. Our great-grandfathers were brothers.

It hasn’t always been so easy to figure out how I am related to my FTDNA matches. I have even shared emails with a couple of my matches, and we finally gave up when we couldn’t figure out where our families link. My largest match at 129.91, estimated to be a 2nd cousin, is Herbert Archie Miller. I have absolutely no idea who he is, and at a 2nd cousin level, I really should know him. Not only is his name not familiar, but I also do not recognize a single surname in his list. Another of my largest matches at 118.34 is a Huff cousin. My brother does not match him to this extent, but he matches another Huff cousin at 103.21 total cM’s, and a longest strand of 39.25. Interestingly, my match with this cousin does not even show up on FTDNA because nowhere do I match him at more than 7 cM’s.

With the very hard work of my Huff cousin, Barbara Joiner, we have been able to use our FTDNA and Gedmatch results to test some family relationships. For example, we know that our great-great-grandmothers, both probably raised in the home of William and Susannah Huff and identified as sisters on many trees, are not. How do we know this? Because Barbara, great-great-granddaughter of Ellender Huff, has a mtDNA haplotype of J; my mtDNA haplotype, inherited from my great-great-grandmother Elzina, is T.

I also know that somehow I have Pharris DNA that Barbara does not have, because I match several Pharris cousins from Jackson County on FTDNA and Ancestry.com. Is this because Elzina’s mother was a Pharris, perhaps an earlier wife of Sam Huff whose marriage to Lucinda Hardcastle occurred when he was 38? Or was Elzina’s mother a Huff who was never married to the Pharris man who became Elzina’s father? The answers in DNA only lead to more questions.

It’s amazing how many Huff cousins have tested their DNA—probably because, like me, they have come to a brick wall with the Huffs of Jackson County. Barbara has made a chart, which now includes at least 30 cousins, and compares everyone’s total cM’s, longest strand, and MRCA. Most of us have a MRCA of 6 generations back, so we think we may all have common ancestors in Leonard Huff, born 1721, and his wife Elizabeth Stout.

Both FTDNA and Gedmatch will let you compare your DNA with your cousin’s, chromosome by chromosome. The task now is to chart all these matches. Eventually, with enough cousins from different branches, including the families that married into the Huffs, and enough work, we should be able to identify exactly which stretches of DNA on which chromosomes are Huff DNA, or Pharris, or Roberts.


That’s why I’ve been especially excited that I have discovered two new cousins that descend from Caleb Roberts and his wife Sally Huff. You can easily see that they are related to my brother and me on both the Huff and Roberts sides because it looks like our MRCA is closer than it really is. (A little factoid I learned from my more DNA-educated Huff cousins.) The task now is to chart our matches from Chromosome 1 to Chromosome 22 in an effort to find areas where we all (my brother, me, and our two new cousins) match, but areas that do not match Barbara and the other purely Huff cousins. Maybe we can then say that this is Roberts DNA.