Mary M Howard: DNA Project to Confirm that her Parents were James Seth Howard and Harriet C Weeks
Wow! What a journey this has been! For over a year, I’ve dedicated hundreds of hours to researching and verifying the family of my third great-grandmother, Mary M. Howard Stephens. You’ve been with me every step of the way, and I’m excited to share the progress we’ve made!
Let’s recap the key details about Mary M Howard Stephens:
Mary M. Howard was born on May 1, 1848, in Hillsborough County, Florida.
She married Wilson E. Stephens on March 25, 1869, in Manatee County, Florida.
She passed away on April 12, 1879, in Manatee County, Florida.
Here’s what we’ve discovered so far:
Documentary evidence presents a strong circumstantial case that Seth and Harriet C. Howard are Mary M. Howard Stephens’ parents.
DNA analysis further supports this connection, showing strong genetic links between the descendants of Mary M. Howard Stephens and the descendants of Seth and Harriet Howard’s other children, where direct documentary evidence exists.
To strengthen the DNA evidence, researchers must also demonstrate relationships to descendants of Seth and Harriet Howard’s siblings. However, we initially didn’t know the parents or siblings of Seth Howard or Harriet C. Howard. So, we set out to discover them. As a result, we successfully identified three brothers for Harriet C. Weeks Howard.
We then uncovered the father of Harriet C. Weeks and her siblings through further research.
By diving deep into the pension application of Harriet C. Weeks’ proposed brother, John J. Weeks, we were able to confirm that the same John J. Weeks, who lived in Marco, Lee County in 1900, was also the same John Weeks who lived on Chokoloskee Island in 1876 and in Hillsborough County during the 1840s and 1850s. This confirmation was based on his Mexican War Survivor’s application, Indian War Survivor’s application, and his widow’s pension application.
Finally, a review of John J. Weeks’ baptism record provided the name of his mother, giving us a likely candidate for the mother of Harriet C. Weeks as well.
So, what’s left? We need to revisit Step 3 above and perform the DNA analysis between the descendants of Harriet C Weeks and the descendants of her proposed siblings. I’m thrilled to show you the results of this project!
Seven siblings have been identified for Harriet C Howard. If the descendants of these 7 siblings can be confirmed to have an atDNA connection to the children of Seth & Harriet Weeks Howard, then Seth & Harriet can be conclusively confirmed to be the parents of Mary Howard.
The land entry case file for John Weeks, Sr confirms through direct evidence that he had at least 8 living children at the time of his death in 1844:
This study will focus on Harriet’s brothers and their descendants: William, Madison, Thomas, and John J Weeks. Documentary evidence exists to confirm a relationship between Harriet’s husband, Seth Howard, and these 4 men. DNA points in the direction of her 3 sisters, Vicey, Selia, and Mary, but I have been unable to confirm these sibling relationships with direct or circumstantial documentary evidence.
For this DNA project, I used my great-uncle’s kit since he is the closest generation to John Weeks Sr. and Margaret Bennet Weeks among the kits I work with. I couldn’t use my father’s kit for this study because he is related to the Weeks family through both his mother and his father.
In this analysis, I will refer to my great-uncle as MMH1. My first step was to review MMH1’s ThruLines on Ancestry.com to see what results he was generating.
Total autosomal DNA matches with tester MMH1 descending from John and Margaret Weeks as suggested via ThruLines on ancestry.com. Results accessed on Sep 27, 2024.
As you can see, there were 274 DNA matches that ThruLines calculated to be descendants of John and Margaret Bennet Weeks who were MMH1’s 3rd great-grandparents.
Autosomal DNA matches with tester MMH1 descending from John and Margaret Bennet Weeks by each proposed child of the couple as suggested via ThruLines on ancestry.com. Results accessed on Sep 27, 2024. Modified by author to exclude a “proposed” child not listed within the Land Entry Case file for John Weeks, Sr’s heirs.
Of those 274 DNA matches, 189 DNA matches were from the Harriet C Weeks Howard’s line of descent. So, that leaves us with 82 DNA matches from the five remaining siblings!
The next step in the project was to locate the closest DNA match to MMH1 for each proposed sibling of Harriet C Weeks Howard and, thus, uncle for Mary M Howard Stephens. Once located, confirm the relationship using documentary evidence. The following limitations must be acknowledged:
Descendants of Celia Weeks could only be confirmed up and to their relationship to Celia. I have been able to confirm that her maiden name was Weeks. However, I have been unable to locate solid documentary evidence to tie either her or her husband to John and Margaret Weeks.
Descendants of Thomas Weeks do not appear to exist. At the time of his death in 1855, his probate file notes that he has a child. No guardianship record has been located for this child. Thomas Weeks’ widow, and the mother of the unnamed child, got remarried and is on the 1860 census (and subsequent censuses) with children all born after the date of her marriage to her second husband. Sources and confirmation of these facts can be found in my publicly available tree on ancestry.com entitled “Sunni’s Family Tree” for username sunnisearchangel. The following profiles contain the relevant information: Thomas Weeks 1824-1855, Mary Jane Burnett 1838-1930, and Michael Alderman 1836-1864.
Descendants of Vicey Weeks do not show up within MMH1’s ThruLines, however, I have access to another kit (more on this later) whom I am calling JJW1. She does show a direct line to a Victoria Weeks, however, research has not been completed to confirm this relationship.
William Weeks was married twice, so I wanted to confirm a line of descent from at least one child from each marriage. But it should be noted:
William had a son named Thomas J Weeks who lived in Lithia, Florida and married a woman named Martha. Thomas J and Martha Weeks had a son named Francis M Weeks. Some DNA “matches” show up in the ThruLines for Frank M Weeks’ son, Peter Weeks. However, Peter Weeks was adopted by FM Weeks when he was 5 years old in 1929. His birth name was Leo Sanchez Martin. Admittedly, these DNA matches may exist because Frank M Weeks and Leo Sanchez Martin were related in some way, but I have not attempted to confirm this. For this study, I am excluding any descendants of Peter Weeks, adopted son of Frank M Weeks.
William had a son Peter P Weeks. All documentary evidence points to him having no living offspring. So, all ThruLines attributing descendancy through Peter P Weeks should be excluded.
William’s daughter Harriet Weeks married her cousin George W Weeks so all descendants from this union should be excluded from this study.
Madison Weeks was married twice, and this can get a bit complicated, so bear with me.
First, I want to confirm a direct line of descent from at least one child of his first marriage. However, I’ve only been able to confirm three children from this marriage: Harriett, Rebecca, and John. Rebecca married Joseph Lafayette Williams, and they had a son, Joseph Edward Williams. J.E. Williams later married his step-first cousin, Mary Anderson, who was the daughter of DeLaura Pate, Madison Weeks’ stepdaughter. This relationship is more significant than it seems, as I’ll explain further.
Madison’s second marriage was to Nancy Mary Ann Davenport Pate, with whom he had three sons. Now, here’s where it gets tricky: Nancy was a half-sister to Wilson Elihu Stephens, the husband of Mary M. Howard Stephens—the very woman we’re focusing on for this DNA project! This means my great-uncle, MMH1, will match the descendants of Madison and Nancy Weeks’ children because Nancy was his half-second great-aunt. As a result, all descendants from Madison’s second marriage need to be excluded due to the multiple-relationship overlap. While I can’t use MMH1’s autosomal DNA (atDNA) for this analysis, I can use the atDNA results of JJW1.
John J. Weeks was married twice and had one paramour. Based on the documentary evidence, it appears he didn’t have any living children with his first wife, Deborah Tanner. He then had two daughters with his paramour, Sarah Mercer Raulerson. After Sarah passed away while giving birth to their youngest daughter, John married Sarah’s daughter from her previous marriage to Jacob Raulerson.
This situation requires careful navigation due to the multiple intertwined relationships.
I want to show a confirmed line of descent from at least one of his daughters with Sarah Mercer Raulerson.
I want to show a confirmed line of descent from at least one of his children with his “stepdaughter”, Lizzie Raulerson.
Special care was given to ensure that no pedigree collapse was included in this study, however, it may need to be revised if any additional pedigree collapse is found later. As described above, these were very small communities, and the families are extremely intertwined.
The following is a chart showing the line of descent of each DNA match analyzed. Their relationship to MMH1 in the blue boxes. I also had the privilege of collaborating on JJW1’s match list (she is the 2nd great-granddaughter of John and Maragaret Bennet Weeks through their son John J Weeks). If one of the descendants from the other proposed siblings also matches with JJW1, I have included that information in a small yellow colored box below the blue box. And as described above, sometimes, I can ONLY use her results for this study because MMH1’s results are tainted due to multiple relationships and/or pedigree collapse.
The following chart shows and analyzes the shared centimorgans of autosomal DNA shared between MMH1 and the other test takers who are documented to descend from the other proposed children of John and Margaret Bennet Weeks.
I have included whether the shared autosomal DNA between MMH1 and each of his cousins falls within one standard deviation of the mean, as this is standard practice for this type of analysis. However, because MMH1 has so many DNA matches, this measure is less significant than it would be if we were focusing on a single DNA match to solve a genealogical 'brick wall.' Additionally, I found at least one other documented relationship from the same presumed sibling of Harriet C. Weeks Howard that did fall within one standard deviation of the confirmed relationship with MMH1. The two relationships in the chart that fall below one standard deviation from the mean were only identified because I also had access to the results of test taker JJW1.
Autosomal DNA inheritance, except in parent/child relationships, is random and does not follow a strict linear pattern. At the third and fourth cousin level, it is common to have documented cousins who do not appear as DNA matches. Therefore, while some documented relationships fall outside one standard deviation from the mean, each assumed sibling relationship to Harriet C. Weeks Howard is supported by at least one documented relationship that DOES fall within one standard deviation from the mean. These relationships can thus be considered confirmed through documentary evidence and validated by DNA research and analysis.
The following chart shows and analyzes the shared centimorgans of autosomal DNA shared between JJW1 and the other test takers who are documented to descend from the other proposed children of John and Margaret Bennet Weeks.
The DNA analysis for JJW1 and her matches shows results similar to those found when reviewing the results for MMH1 and his matches. Although some documented relationships fall outside one standard deviation from the mean, each presumed sibling relationship to John J. Weeks is supported by at least one documented relationship that falls within one standard deviation of the mean. Therefore, these relationships can be considered confirmed through documentary evidence and validated by DNA research and analysis.
By combining:
the circumstantial documentary evidence showing that Seth Howard and Harriet C Weeks Howard are the likely parents of Mary M Howard Stephens b.1848 d. 1879 m. to Wilson Elihu Stephens in 1869 in Manatee County, FL - along with -
the DNA study completed in January of 2024 which shows a DNA relationship between MMH1 to at least one descendant (confirmed through documentary evidence) of every child of Seth and Harriet Howard - along with -
this DNA study which shows a DNA relationship between MMH1 and multiple descendants of 3 of the 7 siblings of Mary M Howard Stephen’s mother, Harriet C Weeks Howard,
we can now conclude that Mary M Howard Stephens is the confirmed daughter of James Seth Howard and Harriet C Weeks Howard.
Additionally, we have direct evidence that John J Weeks, Madison Weeks, Harriet Weeks, and Thomas Weeks are siblings and that their father is John Weeks, Sr. We have direct evidence that Margaret Bennet Weeks is the mother of John J Weeks. It is likely that Margaret Bennet Weeks is also the mother of Harriet C Weeks Howard. Documentary evidence of Margaret’s relationship to her parents and her siblings will need to be obtained and a DNA study performed for us to conclude this relationship between Harriet C Weeks Howard and Margaret Bennet Weeks.
In conclusion, there is no direct evidence naming James Seth Howard and Harriet C. Weeks Howard as the parents of Mary M. Howard, who was born on May 1, 1848, in Hillsborough County, Florida, married Wilson Elihu Stephens on March 25, 1869, in Manatee County, Florida, and died on April 12, 1879, in Manatee County, Florida. However, a close analysis of circumstantial evidence suggests a relationship to Seth and Harriet Howard. This is further supported by shared DNA matches with descendants of their other children, for whom direct evidence exists. When these findings are combined with a DNA study showing a direct genetic link to three of Harriet C. Weeks Howard’s siblings, the case that Mary M. Howard is their daughter becomes strong and conclusive.
