Not All Daughters Lead to Matrilines

 

It seemed like such a straightforward, streamlined solution: to determine which of John Carter's children belonged to which of his wives, start with the daughters. Then, since my fifth great-grandfather John Carter lived and died in Virginia in the 1700s, trace the descendancy of his many daughters, seeking to focus specifically on the daughters' daughters' daughters, on down to the present time, at which point we could find willing Carter descendants of those matrilines to take a mitochondrial DNA test and help us compare results.

There's a problem with such an approach. Yes, of course, there is that one issue of finding eligible descendants who would be willing to participate in such a DNA project. But that is not the problem which concerns me. Where I'm stuck is this: not all ancestral daughters would eventually lead to matrilines. Somewhere in that long line of generations, John Carter's descendants could experience the exact opposite of what genetic genealogists call "daughtering out."

While I'm still stymied with actually documenting the genealogies of all of John's children, I did find one way to do a quick and dirty assessment of such a situation. I turned to the ThruLines tool at Ancestry to see whether, even at the level of fifth great-granddaughter, my autosomal DNA test could reveal genetic connections to this distant Carter ancestor.

At the level of fifth great-grandparent, that is the most distant level the ThruLines readout will extend. Sure enough, there in my results was an entry for my own DNA matches who join me in linking back to John Carter. 

There were seventeen of those Carter DNA matches. The closest of them shared a puny nineteen centiMorgans with me—an understandable amount, considering the distance of the relationship—and the smallest shared amount registered at only six cMs, hardly a measure generating confidence.

Still, my purpose was to do a quick assessment of my theory that finding matrilines direct from the current generation back to the Carter wives might be a challenge. Looking at the proposed lines according to ThruLines, here is what I found: 

• One match to Mary Beverley Carter, through her son Richard C. Stevens
• Five matches to Margaret Carter (half sister), all through her son Carter Marshall
• Six matches to Sarah Carter (half sister), four through son John Sutton, two through son Norborne Sutton

As you can see, none of those matches came strictly through a matrilineal line of descent, foiling our project's purpose. 

Looking even further, checking the ThruLines entries specifically for the wife of John Carter from whom I'm descended, Hannah Chew, there are a total of fifty nine DNA matches currently entered in this category. For one of them, Carter daughter Judith showed six matches to me, but three come through Judith's son Joseph Sutton, two from son Stephen Sutton, and one more from son John Sutton. Not even among these can I find lines of descent from any of John Carter's granddaughters. I, myself, am such an example, as the matriline leading from Hannah Chew to me stops with my second great-grandfather, Thomas Taliaferro Broyles.

In a general exploration of all of John Carter's grandchildren, I spotted further problems from documentation. Some daughters did have daughters of their own—but of those I've found so far, some died young, or, having survived and married, had only sons. I did find one daughter who had a daughter—but I find no record that she even married, let alone had daughters of her own.

Granted, ThruLines relies on the trees of those who tested their DNA at Ancestry. There can be mistakes. But even the old published genealogies some people rely on contain errors, too. Seeking actual documentation, as we've already discovered, can reveal details not included in some genealogies.

To complicate matters, researching ancestors from the 1700s is far easier when we are looking for ancestral grandfathers than when we are looking for their wives. Women from earlier centuries seemed almost invisible, adding this one challenge to our quest to trace the matrilineal descent of John Carter's wives. Still, if someone doesn't build the tree, we'll never arrive at the answer we're seeking.

Two Margarets and Two Marshalls

 

What do you do when two children in the same ancestral household have been given the same name? Apparently, in some cases, the one easily gets confused for the other. 

This week, as I turn my attention to those DNA matches who claim the same John Carter as my fifth great-grandfather of colonial Virginia, I've been working my way through the lines of descent of John's children represented among my matches. Since I descend from Margaret Chew Carter, John's daughter by his second wife, Hannah Chew, I was rather dismayed to discover that John already had another daughter by that same name. This research, I thought, will not be easy.

Sure enough, that same fact stumped another researcher over one hundred years ago: Joseph Lyon Miller, who wrote the 1912 genealogy, The Descendants of Capt. Thomas Carter of "Barford," Lancaster County, Virginia. It was he who, a couple years before publishing his book, realized his error in confusing the one Margaret Carter for the other. This he noted in an article appearing in the October, 1910, edition of The William and Mary Quarterly, and correctly identified Margaret Chew Carter's husband as Zachariah Taliaferro. The other Margaret, according to Joseph Miller's article, was wed to a Captain John Marshall.

This discovery brings difficulties of its own, for John Marshall is not only a fairly common name, but also one of distinction in the early history of our country. In fact, searching through trees of matches at Ancestry, I was surprised to see one subscriber had appended a portrait of John Marshall for that other Margaret's husband—but it was the wrong John Marshall. Sure enough, double checking that for myself, I saw the familiar portrait did belong to Chief Justice John Marshall, whose parentage—and name of spouse, incidentally—did not line up with what the Miller book indicated.

As for the John Marshall supposedly married to John Carter's older daughter Margaret, I've found very little on him so far. The Miller book noted that he was "of Caroline" and that he must have been deceased before 1794, as that was the year a legal document was drawn up, including Margaret's name and indicating that she was a widow. The only other note about Margaret in the Miller book was that she had a son named Horace.

So, when I turned to my ThruLines readout at Ancestry.com to check all my DNA matches from Margaret's line, I was hoping at least one of the four matches there would descend through her son Horace. Wrong. Not a one of those four descended from a son named Horace. Instead, I got four matches descending from a son of Margaret said to have been named Carter.

While a given name like Carter might be a promising discovery, considering that was Margaret's own maiden name, I could find no record of such a name in that immediate family line. I couldn't find Margaret's husband's will—in itself a hunt and peck proposition, since all I know about his date of death is that it had to predate that 1794 document drawn up for Margaret as widow, plus her brother and sister-in-law. The actual date of death could have been years before that point. 

It is in research situations like this when I appreciate the sharing opportunities offered by the community surrounding genealogical services. While Ancestry.com is just this past year ramping up their effort to encourage subscribers to collaborate, over the years, I have seen other subscribers share documents they've discovered during genealogy road trips and other research efforts. As I examined the other trees, searching for someone—anyone—who might have discovered a record regarding this mystery Carter Marshall, I ran across a three page document scanned and uploaded to another subscriber's tree.

The document, labeled by the subscriber as "Carter Heirs' Survey," was supposedly from Caroline County, Virginia, location of the deceased Captain Marshall. The first of the three scanned pages had the heading, "April Court 1801," and whatever ledger the scan was drawn from had labeled that item as page 274. Alas, thinking I could simply scoot over to FamilySearch.org and search through Caroline County records, even those way markers offered me no help.

Still, the court document—showing as plaintiffs a listing of several men by the surname Marshall, including one "infant" Horace—named Margaret Marshall, "executrix of John Marshall," as defendant. Could this be the right Margaret? And the right John Marshall? At least the case was said to appear in Caroline County, home of the John Marshall who was our Margaret's husband, the entire county of which, at the time, had a mere seventeen thousand residents. What were the chances?

The fact that this Margaret was labeled as executrix and not administratrix tells me I need to get back to that hunting and pecking once more; there's a will to be found out there somewhere in Caroline County. And, if this posted document is any indication, John and Margaret Marshall certainly had more than one son. Perhaps those DNA matches leading back to one Carter Marshall are on the right track, after all.