Serendipity & Edney
- Oct 23, 2025
- 10 min read
Updated: Nov 16, 2025
A lot of genealogy research can be tough if you don’t have famous ancestors. Breaking down walls for common last names like Brown, White, and Thomas can tax your research skills and dedication. However, when a wall tumbles down, you are often rewarded with several generations and many new family names and lines. It feels like a wonderful achievement.
Finding a maiden name for your female ancestors can be especially difficult. If the woman married before the 1850 federal census, good luck! That is when the census started listing names of everyone in the household. The relationship such as wife, son, daughter, mother-in-law etc. was not included until the 1860 federal census so you could only guess if the people were related in 1850. If the woman you seek was already married and not living in the household of her parents, you have some work ahead of you. Often marriage records are not available either. So how do you break down that wall? I will show you how I did it for one of my ancestors.
This is about my maternal 3rd great grandmother Edney Marshall. She and Richard Marshall were the parents of Martha Jane, who married John Thomas. Their son was Albert Almerine Thomas, who married Cornelia Brown. They are the parents of nine children, including my grandfather Paul Brown Thomas.
Edney Marshall’s maiden name was unknown to my family. There was some thought that it was Duvcott or something similar, but no one really knew. It was said that she might have been French. Federal census records indicated she was born in South Carolina. Each of the four federal censuses has a different birth year, 1802, 1806, 1812, and 1815. Her husband Richard was born in 1796 with only one census reporting 1795. They probably married around 1830 because her first known child was born in 1831. If she was really born in 1812, she would have been about 18 when she married. So, we can probably disregard the other birth years and say she was born around 1812 plus or minus a few years. We don’t really know though without more evidence.
Sometimes, a mother-in-law is living with the family on a census, so I looked for that possibility. None of the four federal census records had an elderly woman living with Richard and Edney and none listed a mother-in-law.
After looking unsuccessfully for birth, marriage, and death records for Edney, my next step was to look for death certificates for her children and hope one or more would list her maiden name as the mother. She had five known children—three daughters and two sons. All were born in Georgia per census records. There is no information on the death of one son, but I suspect he died in the Civil War. The three daughters all died in Alabama and dates are known but no death certificates are available. One daughter was on a death and burials list, which included her mother as Edney with no last name.
So, one son left. William W. born in Alabama in 1836, 1837, or 1839, according to the three census records with his parents. Searching for his name and birth information brought up one match for a William Marshall who died in San Diego, California in 1918. This match was on the 1900 census in California with his wife Laura V. (born in Texas) and 19-year-old daughter Florence A. He was born in April 1837 in Georgia and both parents were born in South Carolina. He could be the son, but he is a long way from the rest of his family. William and Laura married in 1872 according to the 1900 census. So, where was our William previously if this is him?
Searching further yielded an 1860 census with a William Marshall 24, living alone in Randolph, Georgia. However, there is a D W Marshall, 39, living with a family next to him. So, this is probably not our William. Also, D W Marshall is not the correct age to be his uncle born about 1801. We don’t really know though if they are family. I couldn’t find any other good prospects for the 1860 census.
I found a census for 1870 for a William Marshall, 35, living in Falls, Texas. He is living in the home of a physician with other boarders. Also listed are 2 children named Marshall--Benjamin, 12, and Hildegard, 3, both born in Georgia. Are these his children?
Next, the 1880 federal census shows a William in Runnels, Texas where he and his wife Laura are listed with three children (Eolean 13, Mary 3, and Lora 11/12). The daughter Eolean was born in Georgia in 1876 and must be from a previous marriage since she was born before William and Laura married (as revealed on the 1900 census). She could be Hildegard from the 1870 census. Benjamin would have been 22 and might have been on his own by 1880. Laura is shown as 27 and born in Texas.
So, it appears William and his daughter Eolean Hildegard had moved to Texas where he met and married Laura. It is unfortunate that a census was not found showing the mother of his children in Georgia. (I do learn later that her mother's surname was Davidson, but have not found any further information on the mother.) So, all we know is that William and daughter Eolean and, perhaps a son Benjamin, were in Georgia around 1867, her birth year. Something happened to her mother and William decided to move to Texas with her. We still have no evidence that he is the child of Richard and Edney.
We now have reason to believe that the William on the Texas census and the California census is probably the same individual. This means he moved to California from Texas sometime between the dates of the 1880 and 1900 census enumerations. Now, it must be verified that our William and the William that died in California in 1918 are the same person.
The California Death Index lists a William Marshall born 1837, died December 23, 1918, in San Diego. A Burial Permit document shows his full name as William Wilburn Marshall, but there is no family information. His death certificate is available for a fee. So, I sent for a genealogy death certificate. When the certificate arrived, it confirmed his father’s name was correct as Richard Marshall, but his mother was only listed as Mitchell. The informant was William Wilburn’s wife Laura. I wasn’t sure if she really knew his mother’s name since she never lived in Georgia, probably never met her, and didn’t list the first name. I wondered how she would know the last name and not the first. Seemed unlikely. Sometimes, the informant information is incorrect. So, I was unsure since Mitchell was not a name even close to what the family thought. It is entirely possible there was more than one William Marshall with a father named Richard as well. Pretty common names. I couldn’t find any other verification for Mitchell. Nothing I found supported this name.
I added the death certificate and name Mitchell on my Ancestry tree as an “or” possibility in February 2017. I searched for records with that name on Ancestry with no success. Then I moved on to chasing other ancestors and kept Mitchell in the back of my mind.
Sometimes, it just takes a little serendipity.
Flash forward to 2021. I decide to look at Irish DNA matches for my mom on MyHeritage. I found a match named Lorna that had a reasonable size match with a small tree available. I looked at her tree and she had the name Mitchell on it. Gladys Lily Mitchell with a father “Mitchell ?” Nothing more. Something clicked, and I thought of that Mitchell on the death certificate years ago. What if it was correct? Could this be? It is a long shot, but I have a feeling. The Nancy Drew in me must investigate. So, I trace Lorna’s tree back for her without her knowledge. Her Mitchell line goes back to Roscommon, Ireland from her current Dublin location. I find some interesting stories along the way. Are these Mitchells our connection? I need a lot more to know. I look at shared matches and find some with Mitchell on their trees.
It is time to see if I can find more for Edney so I use the search on MyHeritage, having recently extended my membership to include data search capability. It seems that was a good decision. I enter Edney Mitchell born around 1812 in South Carolina. Up pops one family, and it includes an Edney! No one knows anything more about her and some aren’t even sure she is female. I laughed because I always thought Edney was a nickname being used for Edna but it may actually be her name. I am excited to find her parents are listed as Allen Mitchell and Drucilla Lee. Now to verify this information. There are two good methods to do this. First, I can build a tree for this family by tracing each of her possible siblings down using records. The second is to see if our family has DNA connections to the siblings’ descendants.
I add Allen and Drucilla and all their children to my Ancestry tree to see what DNA ThruLines and hints will be generated. I am most interested in seeing the ThruLines because the DNA matches found by Ancestry’s algorithm might find that my family’s DNA matches some of Allen and Drucilla’s children. Ancestry will also provide lineages that I can then verify with records. It will take 24 hours for Ancestry to work its ThruLines magic. While I wait, I go back to MyHeritage.
In MyHeritage, I copy the DNA matching segments for Lorna and my maternal relatives and paste them into DNAPaint, a wonderful site that enables a user to track shared DNA segments. So much easier than doing it on your own on spreadsheets. You can assign an ancestor or an ancestor couple to segments as you identify this information. After I paint Lorna’s segments into my existing profiles, I am delighted to find she matches on the lines previously identified for Richard and Edney Marshall. Mitchell is looking correct so far. Still a long way to go to verify this.
I look at DNA matches that triangulate with Lorna and my maternal relatives (including me). (A match triangulates if you compare each of the three or more as separate groups of two to verify they match each other on the same segment. This proves that the segment was passed down to each person by the same ancestor couple.) If they have trees, I search for Mitchell or other names in common with my maternal family. I add these matches to DNAPaint. I check to see if they match with the Marshall and possible Mitchell lines. They do. This is great progress! Now, I must tie all these families together somehow on the Mitchell lines in their families and mine. Building out my tree and theirs is the only way to accomplish that. I work on that process until the Ancestry ThruLines appear for Allen and Drucilla Mitchell.
When the ThruLines for Allen and Drucilla are ready, I am excited to find many good DNA matches on lots of the possible siblings’ lines for Edney. Now, I must evaluate each of those to verify the trees and lines are correct. You take nothing for granted here. Many times, the people on the ThruLines are generated by Ancestry algorithms that search trees for possibilities. Sometimes they are correct but can also be incorrect. The match can be on a different family line altogether. (AncestryDNA does not provide a chromosome browser, which would identify where matches share DNA on each chromosome. This is what the other sites provide so that people can see the segments shared.) So again, I begin verifying each person on each line, looking at documents. If they are correct, I add them to my tree as well. This goes well, and I find that many of the family moved to Tuscaloosa, Alabama. Although Edney eventually ends up in Alabama, she is never in Tuscaloosa. We have matches on most of the children. Valid documents for their family lines are found! It looks like this is Edney’s family!
Next, I look at Edney’s grandparents on her paternal side. I cannot find much on her mother Drusilla. Many of the trees have William McLin Mitchell as the grandfather on the paternal side. There are not many documents, so I examine the DNA ThruLines. I confirm 4 DNA matches through their trees, examining documents up to William McLin Mitchell. Some of these matches come through two of William’s great grandchildren, Permelia and John Polk Mitchell. Three of the matches descend from 3 different children of John Polk. I find records to confirm the lines for the descendants as they are projected by Ancestry. I have good matches with them. I continue to check the other matches for descendants from other siblings. One of the matches I have already verified is a descendant, Pruitt, is also on MyHeritage and is a good match with the Irish Lorna. Lorna and Pruitt triangulate with my maternal relatives. Good preliminary evidence that the lines are connected. However, finding that connection will be more work! DNA can only get you so far back, so now records are important.
Using various sources, including a Mitchell of Castlestrange genealogy; a Dictionary of the Landed Gentry of Great Britain and Ireland; a DAR Lineage Book of the Charter Members; US and International Marriage Records; multiple types of England, Ireland, and Scotland records; The Belfast Newsletter 1738-1925; and the NUI Galway Landed Estates Database, among others, I trace the family lines back to Castlestrange, County Roscommon, Ireland in the mid 17th century when a Thomas Mitchell settled there from Scotland. The line from America appears to converge with the line from the Irish DNA match with Thomas being the MRCA (most recent common ancestor). His son Thomas is the line to Dublin and his son John is the line to America. Edney Mitchell’s line is from my mom’s paternal side. Seems we can conclude our family is Scots Irish! Interestingly, Thomas may have come from the same area of Scotland as my mom’s maternal side of the Richards/Watson. That confirmation will take more work, and I save it for later!
This was solved using a combination of written records, DNA, and an abundance of determination. Sometimes, knocking down a brick wall is easier. Sometimes, it seems that wall will never go down. This time. It helped to have an Irish DNA match to tie the families together. This was a lot of work but is probably my biggest success since it took the family line back to a foreign country. Usually, my family is still in America, having arrived so long ago.
Some interesting facts along the way:
Edward Mitchell, Esq., of Castlestrange, my 1st cousin 8X removed was the High Sheriff of Castlestrange in 1886. This is interesting since many of Edney’s descendants were on the police force and one was a police chief.
Solved by research/DNA January 2021
Comments