Select Page

And, of course, the beginning…

It began as one of The Legal Genealogist‘s earliest DNA-can-help-with-this mysteries.

The post ran on the first of July 2012 — that’s eight-and-a-half years ago — and I was “Looking for an Alabama relative”: someone willing to take a particular DNA test to help identify the mother of my Margaret Battles Shew.1

Margaret, you may recall, is my third great grandmother. And the problem is that we weren’t 100% sure which of her father’s two wives was her mother.

William Battles married Kiziah Wright in Oglethorpe County, Georgia, in December 1818,2 but within a few years had run off with Ann Jacobs, the woman who became wife #2 on Christmas Day 1829.3 The sordid details are outlined in Kiziah’s divorce petition in 1824 — a case dismissed for want of prosecution in 1829, apparently when Kiziah died4 clearing the way for him to marry Ann.

Margaret’s birth year in the records is all over the map. The 1850 census shows her as age 23, so born around 1827 — after William ran off with Ann.5 But the 1860 census pegs her age as 38 (birth year around 1822)6 — and that could easily be well before William ran off with Ann.

Now whoever Margaret’s mother was, my folks are in an unbroken matrilineal descent: on the chart, my maternal aunt is tester 2 — descending from Margaret’s mother to Margaret to Martha Louise to Eula to Opal and then my aunt.7

mtdna results

So this is the perfect case for DNA — and specifically mitochondrial DNA (mtDNA), the kind that’s passed from mother to all her children but that only her daughters pass down. All we need is a documented descendant of wife #1 and a documented descendant of wife #2 to be willing to test their mtDNA to compare it to ours.

Sounds so easy…


Not even a little bit.

First off, unless we descend from wife #1 Kiziah, she didn’t leave any known descendants — nobody we knew to compare to in her line. Her divorce papers didn’t mention children at all. So on that side we had to go back a generation to her mother, find other daughters, and come back down. But wife #2, Ann, had a bunch of daughters on the 1850 census, so that was a great starting place.

Except it wasn’t. They lived in a county where the courthouse burned twice. No marriage records to identify the married names of Julia or Samantha or Charlsey. I kept moving the issue to back burner while periodically asking for help. Like the 2012 blog post, that didn’t work.

Then it was one of my Christmas wishes in 2013.8


I tried again in early 2014,9 after identifying Charlsey’s married name. I blasted everyone with a family tree naming Charlsey, and finally finally finally got lucky. A lovely woman responded that she didn’t know anything about Charlsey, but her husband descended from Julia…

Yep. By the end of that year, we had one big piece of the puzzle. On 2 December 2014, we got my cousin Michael’s DNA results: he’s tester 1 on the chart and he’s both an autosomal match and a perfect mtDNA match. Case closed, right?

Not so fast. We also had to rule out the possibility that Ann and Kiziah shared a common female ancestor further back in time. Which meant chasing a test taker on Kiziah’s side. After half-heartedly working on that from time to time, I made it a New Year’s resolution for 2019: to finally determine — if it was at all possible — the identity of the mother of Margaret (Battles) Shew.10

Step by step, through 2019, I tried to find testers who descended from Kiziah Wright Battles’ mother, Lucy Jones Wright.11 And by the end of the year, I was about to admit defeat. Between daughters who only had sons, and daughters who didn’t marry, and daughters who married men named Brown or Jones, and descendants who didn’t want to test…

But I am stubborn. I’m German on my father’s side, Scots-Irish on my mother’s side. That’s pretty much the definition of stubborn. So, every so often, I’d pick it up and try again.

And finally, late in 2020, things started falling into place. A lovely lady who descends from Lucy Jones Wright’s daughter Nancy agreed to test, and she’s tester 3 on the chart. And when her results came in, she was a perfect match for another gentleman whose descent can be traced back to Lucy Jones Wright, shown as tester 4. Because of minor glitches in their paper trails, and to put this beyond doubt, I asked one more person to test — a descendant of a different daughter of Lucy Jones Wright.12 She’s tester 5 — and her results just came in. She matches the other Lucy Jones Wright descendants, and none of them match us.

On my side, mtDNA haplogroup H3g, matching the documented descendant of Ann Jacobs. And all of the descendants of Lucy Jones Wright — who would have the same mtDNA as any female line descendant of wife #1 Kiziah — mtDNA haplogroup H3h. Totally different and distinct subclades (defined as “a subgroup of a subgenus or haplogroup”13) of H3.14

H3g split off from the parent group H3 about 6,000 years ago, give or take a few centuries, and it has specific markers that have to be present for someone to be designated as H3g. H3h split off from H3 about 5,700 years ago, give or take a few centuries, and it also has specific markers that have to be present for someone to be designated as H3h.15

Kiziah’s mtDNA, which would be the same as her mother’s, is H3h. That’s not us.

Ann’s mtDNA is H3g. That is us.

Welcome to the family tree, fourth great grandmother Ann Jacobs Battles.

The case of Margaret’s mother is at an end.

And — sigh — just beginning… who were Ann’s parents?

Cite/link to this post: Judy G. Russell, “Finding Margaret’s mother: the end,” The Legal Genealogist ( : posted 31 Jan 2021).


  1. See Judy G. Russell, “Looking for an Alabama relative,” The Legal Genealogist, posted 1 July 2012 ( : accessed 31 Jan 2021).
  2. Oglethorpe County, Georgia, Marriage Book 1: 61, Battles-Wright, 12 December 1818; “Marriage Records from Microfilm,” Georgia Archives Virtual Vault ( : accessed 31 Jan 2021).
  3. St. Clair County, Alabama, Marriage Record 1: 53, Battels-Jacobs, 25 Dec 1829; digital images, “Marriage records (St. Clair County, Alabama), 1819-1939,” ( : accessed 31 Jan 2021).
  4. Blount County, Alabama, Circuit Court Minutes B: 373-375 (1829); Circuit Court Clerk’s Office, Oneonta, Ala.
  5. 1850 U.S. census, Cherokee County, Alabama, population schedule, 27th District, p. 136 (back) (stamped), dwelling 1055, family 1055, Danl Shew household; digital image, ( : accessed 31 Jan 2021); citing National Archive microfilm publication M432, roll 3.
  6. 1860 U.S. census, Cherokee County, Alabama, population schedule, p. 315 (stamped), dwelling 829, family 829, Margaret Shoe household; digital image, ( : accessed 31 Jan 2021); citing National Archive microfilm publication M653, roll 5.
  7. Yeah, I’m in that group too, but I have one oddball result called a heteroplasmy that my aunt doesn’t have, so we use her as the exemplar. See Judy G. Russell, “The myth of the GD0,” The Legal Genealogist, posted 16 Nov 2016.
  8. See ibid., “All I want for Christmas: DNA,” The Legal Genealogist, posted 22 Dec 2013.
  9. See ibid., “Picking your Battles,” The Legal Genealogist, posted 23 Mar 2014.
  10. See ibid., “DNA resolutions for 2019,” The Legal Genealogist, posted 30 Dec 2018.
  11. See ibid., “Keeping that DNA resolution,” The Legal Genealogist, posted 28 Apr 2019; “Finding Margaret’s mother, part 2,” posted 18 May 2019; “Finding Margaret’s mother, part 3,” posted 25 May 2019; “Finding Margaret’s mother, part 4,” posted 29 June 2019; “Finding Margaret’s mother, part 5,” posted 7 July 2019; “Finding Margaret’s mother, part 5a,” posted 10 Aug 2019; “Finding Margaret’s mother, part 6,” posted 16 Nov 2019; “Finding Margaret’s mother, part 6,” posted 16 Nov 2019.
  12. See ibid., “Finding Margaret’s mother, part 7,” posted 3 Jan 2021.
  13. ISOGG Wiki (, “Subclade,” rev. 21 July 2013.
  14. See generally Maciamo Hay, “Haplogroup H (mtDNA),” Eupedia ( : accessed 31 July 2021).
  15. See generally Mannis van Oven, “mtDNA tree Build 17 (18 Feb 2016): subtree R0,” ( : accessed 31 Jan 2021).
Print Friendly, PDF & Email