Using DNA To Solve A Family Mystery: Sarah Ann Eva Vaughan, wife of John Lafayette Vaughan

For the last several months I have been working to solve a mystery in our family that it was believed only DNA could unravel. It was a question none of the Vaughan family historians have been able to answer: Who was Sarah Ann "Eva" McCarty, the first wife of our ancestor John Lafayette Vaughan? Where did she come from? Who were her parents? [For a biography of John Lafayette Vaughan, please visit here: http://thesaltofamerica.blogspot.com/2012/08/john-lafayette-vaughan-belle-starr.html]

These questions were never able to be answered during the lifetime of my beloved cousin, Colleen Vaughan Allen. I believe she would be very proud that I believe I have solved the mystery beyond a reasonable doubt.

We have only ever had two records of this woman: her marriage record and her headstone. We knew she was the mother of Mattie Mae Vaughan, William "Bill" Riley Vaughan, and an unnamed infant whom she died giving birth to. But that was the extent of our knowledge. Even in Census records, the space for Mattie's and Bill's mother's birthplace would be filled with "United States", not even a particular state. So it appears that even the woman's husband and children knew virtually nothing about her.

Here is the marriage record for John Lafayette Vaughan and S. A. McCarty, the original of which is in my possession:

And this is her headstone, courtesy of FindAGrave.com contributor MillieBelle:

Eva's headstone at Choate Prairie Cemetery, west of Inianola, Pittsburg County, Oklahoma.

One will notice that the names on the marriage record and the stone don't match. Where does "Eva", fit into "S. A.", you might ask? Well, that has been a part of the mystery. Family records have said that "S. A." stood for "Sarah Ann", but there was no primary source for that listed; it is possible that Mattie, Bill, or John gave that name to relatives who inquired about John's first wife, but no genealogical record located actually states "Sarah Ann". The marriage and the headstone are the only two items with her name. But this procured another question for my aforementioned cousin Colleen Vaughan Allen, who was long the family authority on the Vaughan family history, a title she inherited from her daughter Lou Ona Vaughan White, daughter of John Lafayette Vaughan by his second wife: Could S. A. and Eva have been two different people?

Colleen certainly thought it was possible. John could have married S. A. and then married Eva, and they could be two entirely different persons and we just may not have a marriage record. This is hardly unheard of; even today, the marriage record for my great grandparents, Sampson Lafayette Vaughan and Lillian Auston, continues to elude us despite their marriage occurring in the last hundred years, roughly around 1922. So this was a perfectly possible scenario, but we knew so little about this woman that we could not say for certain.

Fast forward to this year and I have procured a DNA sample for Norma Marie Vaughan, daughter of Bill Vaughan and last living grandchild of our mystery woman (or one of them, if there were indeed two woman). Once Ancestry.com had completed their testing and shared the results, I began to comb through DNA matches. Most of her closest matches were what I expected: a lot of Vaughans, a lot of Hamiltons, and a fair few Browders. Norma's mother was Lola Mae Hamilton, daughter of William and Margaret Browder Hamilton. So I was seeing a lot of cousins connected to her 3 known grandparents as expected.

There was a conspicuous absence among her matches: Persons with the surname McCarty in their families. If Norma's grandmother's maiden name was McCarty, one would expect to see some of that family among her matches. But in searching her matches' trees, there was one Very High match with that surname in the tree, a handful of "High" matches, and then mostly "Good" and lower matches. Her closest match with that name in their tree was a man who matched at 48/2; his McCarty family came to Texas by way of Tennessee. This McCarty family was not connected to any of the McCarty families of the next 3 highest matches, and none of them were connected to each other either.

I was side-tracked for a while exploring her closest match, a woman named Staci, who did not know who her biological father was. Eventually we figured it out. That saga can be read here: http://thesaltofamerica.blogspot.com/2017/04/a-father-found-but-lost-again-dna-story.html

So since I wasn't finding anything by searching the surname McCarty, eventually I began sorting through matches find people who were not related to the Vaughans, Hamiltons, or Browders. After a while, I finally found a few. Most of them were also related to each other, so that told me I was looking at cousins from the family of Norma's unknown grandmother.

These matches led me to a man named Bird Lewis Stafford of Dallas County, Missouri.

Headstone of Bird Lewis Stafford, courtesy of contributor Genealogy Researcher on FindAGrave.com.

Nearly all of the matches I was finding that actually had trees could be traced back to this man. I zeroed in on the 8 closest Extremely High confidence matches (though there were several Bird Lewis Stafford descendants who matched with varying degrees of confidence) and began to gleam what I could from the information I now had.

As you can tell just by this chart, Bird Lewis Stafford had quite a few children. As I researched him, I found that he had at least 11 children that grew to adulthood--so not even including the children that died in infancy. These matches represented descendants of 5 of these 11 children, and all from his first wife, Lucy Parker. (Two of Bird's children that grew to adulthood were from his second marriage.)

From here, I needed to pull out the handy DNA Detectives relationship chart:

Based on the shared centimorgans between Norman and that matches, I knew her relationships to the matches would most likely be in groups F, G, and H. 

Going back to what little we knew about Eva, I knew she was likely born about 1874-1875 based on her headstone stating she was aged 23 years when she died in September, 1897. Bird died in 1867, meaning he was not Eva's father. Eva's age ruled her out as being the child of Bird's two youngest daughters who were born in 1864 and 1868. Her age along with Norma's genetic distance from the seven 3rd great grandchildren and one 2nd great grandchild of Bird Lewis Stafford appeared to rule her out as a grandchild of Bird as well. Based on her age and DNA matches, it appeared most likely that Eva was a great grandchild of Bird Lewis Stafford by one of his 9 adult children that grew to adulthood from his first marriage.

This would presumably make Norma a 4th cousin (or closer) to 7 of the 8 matches, and a 3rd cousin once removed (or closer) to S. C. Since Eva could be descended from one of the five children of Bird from whom these 8 matches descended, it was possible that some of the matches were even closer. The lowest 5 of the 8 certainly appeared to fit in Group H--3rd cousin once removed or other distant (4th) cousin. At 75/4, K. L. could be on the highest end of Group H, or right in the middle of Group G. 

At this point my progress stalled. It was May, and I had a lot on my plate, so I set this project aside for a while. I hoped that maybe in a few months Norma would have more matches that would indicate a likely connection to one of the nine children.

Now it's August and I check again. She had no new matches in this family as best I could tell, so I decided to look harder at a few other intriguing matches.

In looking at her top 4 matches in the group of 8, the two highest descended from Cynthia and the next two highest from her brother Lewis. What was interesting is that Cynthia and Lewis married a pair of relatives: Asberry/Asbury Benjamin Chapman and Matilda Cheek, respectively. Asberry was the son of William Chapman and Mary Bond. Matilda was the daughter of James Cheek and Elizabeth Chapman, another child of William Chapman and Mary Bond. So Asberry was an uncle of Matilda. Since Norma shared the most DNA with these linked families, I decided to look focus on their respective descendants.

Headstone for Louis/Lewis Stafford, son of Bird, and his wife Matilda. Courtesy of contributor Pat Faulkner on FindAGrave.com.

First, I looked through their respective families. Lewis (also spelled Louis) was younger than Cinthia. He was still having children in the mid-1870s when Eva was born, and genetically we had ruled Eva out as a grandchild of Bird. This seemed to rule Lewis out as Eva's father. Cinthia and her husband married about 1849, and had five children between that time and Asberry's death in the Civil War in Memphis in 1863. Eva could not be the child of the youngest member of this family, Elias (b. 1862), but could be the child of any of the remaining four, though Martha (b. 1858) was less likely because she would have been only roughly 16-17 at the time of Eva's birth. That would be uncommon but not impossible.

Second, I checked her DNA matches for additional Chapman descendants. In more than one case, I had to dig into matches' family trees to find the connection. These people had trees with people in them, but had not connected to the Bird Stafford or William Chapman families, so I had to do some digging myself. I did that once already with B. H., a match mentioned above who was the highest match to Norma in this family. I found four additional interesting matches:

The first of these matches, Robert, is intriguing because despite being a full seven generations removed from Bird, he still shared enough DNA with Norma that he could be in Group G or even Group F, and he is a descendant of Cynthia. The next match is a lower-end four generation match to a descendant of Bird's son Lewis again. The 3rd match, V. J. F., is intriguing because she is not a Stafford at all--but she is a Chapman from the same Chapman family as the aforementioned Asberry (husband of Cynthia) and Elizabeth (whose daughter Matilda married Lewis). She descends from another sibling of this pair, William Chapman, another child of William Chapman and Mary Bond. V. J. F. is four generations removed from this couple. The fourth match is yet another descendant of Cynthia Stafford and Asberry Chapman. 

These four matches combined with the previous eight convinced that I needed to be looking hard at the family of Cynthia Stafford and Asberry Chapman. The match with a Chapman who is NOT also a Stafford, and the match with Robert Y. with 70 cm despite 7 generations removal from Bird told me this is the right direction to be looking. I decided the most logical step would be to start with the eldest child of Cynthia and Asberry. And it is fortunate that I did, because I believe it saved me a lot of time.

Mahala Jane Chapman was the eldest child of this family. She appears as an 8 year old in the household of A. B. and Cynthia Chapman in 1860. 

1860 Census - A. B. Chapman household - Dallas County, Missouri - Courtesy of FamilySearch.org

Some Stafford/Chapman family records have her year of birth as 1850. Her headstone also has this year. But her Census records reflect a year of birth between 1864-1857, and this one indicates closer to 1851-52. I believe this one is likely the most accurate given that she is clearly older than her brother William, who is traced easily, but not born by the 1850 Census. 

On 16 Jul 1868, Mahala married a man named Silas R. Busby, or at least that is how his name appears in the marriage record and 1870 Census. By 1880, his name appears to be listed as "Billy S.", but his wife's name and age and the eldest child of the family confirm that "Silas" and "Billy" are the same. His name could have been William Silas, or William Silas R., or some other combination of those names. I cannot locate any records on him prior to his 1868 marriage or after the 1880 Census, so I cannot determine what his proper name should be. But since two records indicate "Silas R.", that is how I will refer to him.

And when I found the family on the 1880 Census, there it was: the answer I had been seeking.

1880 Census - Silas/"Billy" Busby household - Sebastian County, Arkansas - Courtesy of FamilySearch.org

The 3rd child down you'll see her. "Sarah A. E.", age 5. Now am I crazy, or is that "Sarah Ann Eva", born 1875? The "A." kind of looked like an "H." though, so I looked for other examples on the page to compare the census-taker's writing to.

1880 Census - Sebastian County, Arkansas - Courtesy of FamilySearch.org

So here you'll see the "A" in Alexander and the "H" in Henry each underlined in red. Based on comparing these, it is clear to me that the letter after "Sarah" in the above image is indeed an A and not an H.

It would appear I have a match. Sarah A. E. Busby would be a great grandchild of Bird Lewis Stafford as the daughter of Mahala Jane Chapman, daughter of Cynthia Jane Stafford. The age matches, and the name matches right down to her being able to sign as "S. A." and go by the name "Eva". 

Now I needed to see if this relationship gelled with the DNA matches. In looking closer at the tree of Robert Y., I suddenly had an explanation as to why he matched Norma so closely despite the 7 generation removal from Bird Stafford--Robert also descends from Mahala Chapman Busby. He descends from her daughter Mary Cynthia who married James Joseph Phipps. If Norma's grandmother is actually Eva "Busby", then she is three generations removed from Silas and Robert is five generations removed from Silas. That would make them 2nd Cousins Twice Removed. And what does the DNA Detectives chart say? Well, it says that 70 centimorgans shared is certainly enough to fall into Group G, indicating a possible relationship of 2nd cousins twice removed. 

So now I needed to go through all of the other 11 DNA matches I have been looking at see if the possible paper relationship between Eva/Norma matches with each of them.

Please note that other than V. J. F., who is not a Stafford, all of these relationships were calculated based on the matches' mutual relationship to Bird Lewis Stafford. However, in the cases of those who are descended from Lewis Stafford are also descendants of the same Chapman family as Cynthia's husband, so Norma would theoretically share more DNA with those persons because they share Stafford and Chapman DNA, not just Stafford DNA.

But as you can see across the board ALL 12 DNA matches fit when comparing centimorgans shared to relationships on paper. Three 3rd cousins, one 2nd cousin twice removed, three 3rd cousins once removed, and five 4th cousins all match Norma appropriately. It would be preferable to have some more DNA matches from the Busby family to fully 100% confirm this relationship, but in this case all of the circumstantial evidence genetically is very strong.

The last thing for me to do was to see if there was any more circumstantial evidence outside the realm of DNA to help support the assertion that S. A. "Eva" 'McCarty' is really a BUSBY. I found only one thing.

In the 1900 Census, both Eva's widow John L. Vaughan and her presumed mother Mahala Chapman can be found in the Choctaw Nation of Indian Territory. But the Vaughans are in modern-day Pittsburg County, and Mahala, with her new husband William McDonald, is in modern-day Bryan County, near the current town of Calera. This Census says Mahala has had only one child, and that only child is still surviving. I believe this was an error and was referring only to her current marriage; she had only one child with William McDonald.

In 1910, Mahala is still in Bryan County, widowed again, and living with her now-adult daughter from the second marriage. This Census record states she was the mother of 9 children, 2 of whom are still living. One of those two would be the child from her second marriage she is living with. The other would be the aforementioned Mary Cyntha Busby Phipps, the ancestress of DNA match Robert Y. I could find virtually no information at all on any of the other children listed in the 1880 household for Mahala and Silas/"Billy" Busby. This indicates that Mahala's daughter "Sarah A. E." is indeed deceased, which aligns with "our" Eva having died in 1897. While that is not the strongest piece of evidence, as piece of this entire puzzle it does support the final conclusion.

That final conclusion is this: Based on the DNA evidence indicating that Norma Marie Vaughan is certainly a descendant of Bird Lewis Stafford, and is most likely a descendant of William Chapman, and those men have a mutual descendant named Sarah A. E. Busby who is the same age as our Sarah Ann Eva, grandmother of Norma Marie Vaughan, and there does not appear to be a 2nd woman with the same name that "could be" that woman, then it follows that Sarah A. E. Busby is most likely Sarah Ann Eva McCarty Vaughan. McCarty was most likely a previous married name before she married John Lafayette Vaughan, as DNA evidence does not indicate that she is a biological McCarty.

I believe all of the above-listed evidence supports this conclusion. I would be glad to hear from anyone that can provide further evidence to support my assertions, or I would be equally interested if someone could submit evidence that contradicts my conclusion if such evidence exists. Until that time, I believe it would be safe to list the first wife of John Lafayette Vaughan as Sarah Ann Eva Busby, born about 1875 in Missouri, died 29 Sep 1897 and was the daughter of Silas R. Busby and Mahala Jane Chapman.