A Case Study in Genetic Genealogy —
Many genealogists have had to climb the dreaded “brick wall”, but few have had to deal with a completely missing branch because of an adoption where there are no adoption records, no birth certificate, no living immediate family members, and no idea of the biological parents’ location, names, or history.
The family tree of my wife, Lynn Francis, is fairly well documented going back several generations. But there was one branch we were not able to complete due to an undocumented adoption. This is the story of how we found Lynn’s great- grandmother solely through autosomal DNA.
Starting with a Dead End
Through online and family records, we were able to find most of Lynn’s ancestors several generations back, but there was one branch with a dead end. Lynn’s maternal grandmother, Olivia June Bowne, was adopted in St. Louis, Missouri around 1905. We had the names of her adoptive parents, Samuel and Mary Bowne, her brother Charlie, who was also adopted and her birth date in 1902, but that was all we knew. Family lore states that when the Bownes adopted Olivia June, they also decided to take home Charlie, who was her slightly younger but not related. We had absolutely no idea who Olivia June’s biological parents were.
Searching on Ancestry, we found Samuel Bowne was listed alone, in the 1903 St. Louis, Missouri, city directory.1 In the 1908 city directory, Mary Bowne was listed as his widow.2 To date, no record of Samuel’s death has been found. A Social Security application and claim record for Olivia June listed her parents as Samuel Bowne and Mary C. Edmonds.3
In the 1910 US census for Pleasant, Garfield County, Washington, we found Mary Bowne, age 48, listed as a servant for a Canadian gentleman, Stewart Vallen, along with her adopted children Olivia June, age eight, and Charlie, age seven.4 It indicated she was widowed and had no children of her own. The census states both Olivia June’s and Charlie’s birthplace as Missouri, and that they were boarders.
The 1920 US Census for Toutle, Cowlitz County, Washington, found Olivia June, age 18, as the wife of Walter Hankins, Lynn’s grandfather.5 The census listed both Olivia June and her parents’ birthplace as Missouri.
So, at this point, we know when Olivia June was born, who her adoptive parents were, approximately when she moved from St. Louis to Washington, and when she married Lynn’s grandfather, Walter Hankins. But we still had no idea what her birth name was, when she was adopted, where she was born, or the identity of her biological parents. These were the questions we set out to answer.
One possibility we had to consider was that there were numerous orphan trains coming to St. Louis around that time. Lynn’s grandmother could have been born anywhere and brought to St. Louis on an orphan train. Missouri has strict laws when it comes to adoptee identification, but prior to 1917, St. Louis adoption records were maintained by the Recorder of Deeds. Pre-1917 St. Louis City Deeds of Adoption are public records and are recorded with property transactions in land record deed books. But according to the St. Louis City Recorder’s website “when recorded, a Deed of Adoption was indexed by date of recording not date of deed”.6 In other words, it could be that a 1902 Deed of Adoption will be found in a 1950 Land Deeds Index. When we contacted the St. Louis Recorder of Deeds, they were able find a record of adoption for Olivia June’s adopted brother Charlie. Charlie was adopted in August of 1903 for the consideration of one dollar. His biological mother was Grace Landwich of Youngstown, Illinois. Samuel and Mary Bowne of St. Louis were the adopters.7 The St. Louis Recorder of Deeds was not able to find any records for Olivia June. Curiously enough, the 1910 US census indicates Charlie’s parents’ birthplace as Illinois, while listing Olivia June’s parents’ birthplace as the United States. We began to wonder if she really was adopted in St. Louis.
DNA Revelations
We figured the best chance for answering these questions would be through DNA. Lynn took the AncestryDNA test in August of 2016. As you might expect, after a few weeks we found a lot of third and fourth cousin matches. But the matches by themselves were not going to identify her biological great-grandparents if we didn’t even know their names.
We first eliminated cousins who were not on Lynn’s mother’s side of the family, but the remaining matches weren’t going to help us unless we had some idea of what Lynn’s biological great grandparents’ tree looked like.
One thing that would improve our chances would be to get Lynn’s mother’s DNA. Her mother is in memory care facility with moderate dementia, so we were not sure we would be able to get her to take a test. Spitting in a tube was out, but on a pre-Christmas visit in December of 2016, Lynn was able to get a swab of her mother‘s DNA with her cooperation and send it off to Family Tree DNA (FTDNA).
When we received the test results, we downloaded Lynn’s mother’s raw DNA and uploaded it to Gedmatch.
After several days of playing with spreadsheets and triangulating cousins, we were able to make a few educated guesses for potential DNA matches related to Lynn’s grandmother, Olivia. To be related to Lynn’s unidentified biological great-grandmother, they had to share matching DNA segments with both Lynn and Lynn’s mother, thereby eliminating DNA cousins on her father’s side.
But we still didn’t know which of the remaining DNA cousins were related to Lynn’s grandmother Olivia June or her grandfather Walter Hankins. It still seemed like an impossible task.
A Possible Cousin
The most promising DNA cousin was a second to fourth Cousin of Lynn’s mother. They shared 91 centimorgans (cM) across 4 DNA segments, a significant match. This cousin was both related to Lynn and her mother and was an X- match for both of them.8 (At the time we ignored the X-match, but later discovered it was an important factor in helping identify Lynn’s great-grandmother). This cousin, LNorth, was kind enough to grant us access to her ancestry tree. Of course none of the names in her tree sounded familiar since we didn’t know the biological maiden name of Lynn’s grandmother. We only knew Lynn’s grandmother was born in 1902, so we were probably looking for birth parents that were born between 1865 and 1885, assuming they would have had a child between ages 17 and 37.9 Unfortunately, LNorth’s tree consisted of very large families that had quite a number of people who were born between 1865 and 1885. And LNorth could be a descendant of Lynn’s grandfather, Walter Hankins, sending us on a wild goose chase.
Somehow, we needed to refine Lynn’s DNA cousin list to a handful of cousins on her mother’s side so that we could narrow the focus.
Half-Sister Helps With More DNA
As luck would have it, Lynn’s mother has a younger half-sister, Billie Jean, and they share the same mother, Olivia June. If we could get Billie Jean’s DNA, we could cut our work in half again. If a DNA cousin matched Lynn’s mother and her aunt Billie Jean, they had to also be biologically related to Lynn’s grandmother, Olivia June.
Lynn approached her aunt Billie Jean in 2018 about getting a DNA sample. She is a retired nurse, and she was excited that we might eventually be able to identify her grandparents.
Our first puzzle was deciding which test should we send to Billie Jean. After some thought, it made sense to send Billie Jean an AncestryDNA kit. Even though Lynn’s mother’s DNA was not on AncestryDNA, we could find matches that were in common with Lynn and her aunt Billie Jean. We could also take the raw DNA and upload it to FTDNA and Gedmatch, where all three tests could be compared.
Six weeks later, we received Billie Jean’s DNA results. Since Bille Jean was listed in Lynn’s family tree, we attached Billie Jean’s DNA results to Lynn’s family tree on Ancestry. Within a few days, we had hundreds of DNA matches.
Starting Over
At this point we threw away all our prior assumptions and started over. We were looking for someone who matched both Lynn and Billie Jean or Lynn’s mother and Billie Jean. We looked for matches on AncestryDNA, FTDNA and Gedmatch.
We weren’t surprised to see several of the same names we’d seen before, but many of the prior names we could eliminate because if they didn’t match both Billie Jean and Lynn’s mother, we knew they were likely to be on Lynn’s grandfather’s side of the family.
One name stood out on the Ancestry and FTDNA testing sites. It was LNorth again. The same promising cousin we found earlier, but this time we knew it was a valid match on Lynn’s grandmother’s side. On both Ancestry and FTDNA, LNorth matched Lynn, her mother, and Billie Jean, and on FTDNA they all had matching X chromosomes.10
FTDNA results showing the overlapping segments
Name | Chr | Start | End | cMs | SNPS |
Billie Jean | 20 | 5879473 | 62382907 | 88.77 | 9926 |
Lynn’s mother | 20 | 11244 | 62382907 | 104.2 | 18396 |
LNorth | 20 | 5879473 | 45913931 | 48.85 | 10492 |
Billie Jean | X | 122405854 | 149248804 | 45.49 | 3382 |
Lynn’s mother | X | 102656602 | 154570039 | 79.42 | 6942 |
LNorth | X | 102656602 | 124067035 | 25.24 | 2550 |
The above chart reveals the FTDNA results showing the overlapping segments on Lynn’s mother’s chromosome 20 and X chromosome.
We started looking for other cousins that had a match on Lynn’s maternal side who were also a match to LNorth. We found two additional candidates in the second-to-fourth cousin range that matched Billie Jean on AncestryDNA, Christine and Hughess. LNorth identified Hughess as her first cousin once removed. Looking at their trees on Ancestry, we discovered one surname in common, Kendall. Without too much effort, we found they all had a common ancestor named Henry Kendall, born in Pennsylvania in 1801.11
Mirror Tree
From here we created a mirror tree for Henry Kendall and his descendants. In this mirror tree, we looked for all of his descendants that were the correct child-bearing age in 1902. It was a big family, but we were able to whittle the list down to 24 individuals, all grandchildren of Henry Kendall. Although Henry had eight children, the three DNA matches seemed to be descendants of just two of Henry’s children, Samuel Kendall and Jacob Kendall. Jacob Kendall had four children born between 1874 and 1887. Samuel Kendall had nine children born between 1870 and 1889. Sticking with our original assumption that Olivia’s parents were born between 1865 and 1885, we were able to now focus on just 13 of Henry’s grandchildren.
We began researching each one of these thirteen grandchildren looking for any clues. Most of Henry’s grandchildren were raised in the towns of Cobden or Makanda in Union County, Illinois, about 110 miles from St. Louis. Being the largest town in the vicinity, St. Louis was probably frequented often by the Kendall family. But we found no evidence of any of them living in St. Louis, except for Mary Emma Kendall, born in Union County in 1877.
In the 1900 U.S. census for Rich, Union County, Illinois, we found Mary E. Kendall, age 21, single, living with her parents, Jacob and Delila[h] Kendall.12 In the 1910 U.S. census for St. Louis, Ward 27, we found Mary E. McGuire.13 The 1920 U.S. census for St. Louis, Ward 27, listed her mother Delilah Kendall, confirming that we had the correct Mary E. (Kendall) McGuire.14
We don’t know exactly when Olivia June was adopted, but we know it was around 1905, when Olivia was a toddler. Mary Emma Kendall married Orley McGuire in St. Charles County, Missouri, 10 Jun 1905.15
We contacted Lynn’s DNA cousin, LNorth, to see if she could help. She had many photographs and newspaper articles of the Jacob Kendall family, but nothing that would help confirm that Olivia might be the child of Mary Emma.
Going back to the DNA, we refined our mirror tree by trying to identify exactly how the three matching cousins fit into the tree, assuming that Mary Emma was Olivia June‘s mother. Completing the tree and charting their assumed DNA relationships confirmed that Mary Emma was the most likely candidate. We were able to place all three cousins into our relationship chart. One was a third cousin once removed, a descendant of Samuel Kendall; one was a half third cousin, also a descendant of Samuel Kendall; and one was a second cousin once removed, a descendant of Jacob Kendall and the great-granddaughter of Orley P. McGuire and Mary Emma Kendall. More importantly, each of these cousins had the expected amount of DNA in common with Lynn, her mother, and her aunt Billie Jean. We were also able to confirm that they should all have an X chromosome match, since the tree showed there were no two consecutive male descendants between Lynn and her DNA cousins.
This looked very promising, but we felt we needed one more close DNA connection to prove Mary Emma was Lynn’s great-grandmother.
Yet Another Cousin
In communications with DNA cousin LNorth, she put us in contact with one of her cousins, Donna, who has been on the genealogy sidelines for a while, but was interested in getting involved again. Donna was Mary Emma’s granddaughter. Donna didn’t have an ancestry account, nor had she ever submitted DNA to a testing center. After explaining to her the possible relationship between Olivia June and Mary Emma, she was excited about the possibility of finding a new cousin and purchased the AncestryDNA kit that same day.
Mapping Donna’s position in our relationship chart and mirror tree showed Donna to be a half-first cousin to Lynn’s mother and Billie Jean. If they match with the correct amount of DNA, that will help prove, from a DNA perspective, that Mary Emma is Lynn’s great-grandmother. We determined that the range of DNA required to be a half-first cousin should be between 137cM and 856cM.16
Conclusion
About six weeks later, the results were posted on AncestryDNA. After uploading to Gedmatch, we find a 449cM shared match with Billie Jean over 21 DNA segments, well within the expected range for a half-first cousin.
We are confident that we found Lynn’s great-grandmother based on the results of our DNA analysis and the supporting genealogical records.
Chart of the DNA family members and cousins
Name | Relationship | Expected DNA Range | Shared Total cM | Largest cM | X Shared Total cM | X Largest cM |
Mom | Self | |||||
Lynn | Daughter | 3330-3720 | 3587 | 281.5 | 196 | 196 |
BillieJean | Half-sister | 1317-2312 | 2016 | 141.3 | 81.3 | 44.1 |
Donna | Half 1C | 137-856 | 528 | 63.9 | 74.9 | 46.4 |
Christine* | Half 2C1R | 0-341 | 155 | |||
LNorth | 3C | 0-217 | 91 | 54 | 28.2 | 28.2 |
Hughess* | 3C1R | 0-173 | 30 | |||
*All results are from Gedmatch except Christine and Hughess were matches to Billie Jean on AncestryDNA. The Shared Total cM were provided by Ancestry. LNorth identified Hughess as her 1C1R. Donna identified Christine as her 1C1R. |
Lynn’s family tree showing DNA cousin relationships
As expected, Lynn, Lynn’s mom, Billie Jean, and Donna’s DNA contain many overlapping segments on several chromosomes. Using DNA Painter, we show the segments that overlap Lynn’s mom’s DNA on chromosomes four and five.
Olivia June Bowne Hankins Smith
Mary Emma Kendall McGuire (Courtesy of Donna)
About the Mary Emma (Kendall) McGuire family
Mary Emma Kendall‘s father, Jacob Kendall, was born in Ohio on 28 August 1843.17 He was a farmer18 and served as a private in the Union Army 67th Regiment, Ohio Infantry.19 Jacob married Delilah Hock on 2 March 1873 in Union County, Illinois.20 Delilah was born on 26 October 1846 in Pennsylvania.21 Jacob died on17 June 1919 in Illinois,22 after which Delilah moved in for a time with Mary Emma’s family in St Louis Ward 27, St Louis County, Missouri.23 Delilah died on 11 November 1929 in Union County, Illinois.24
Jacob and Delilah Kendall had five children:25
- Charles M. Kendall was born 8 July 1874 in Makanda, Union County, Illinois; he died on 5 October 1909 in Union County, Illinois; he married Minnie Mae Kimmel in 1897; he was a preacher.
- Mary Emma Kendall was born 21 August 1877 in Union County, Illinois; she died of peritonitis after a ruptured appendix on 9 March 1925;26 she married Orley McGuire on 10 June 1905 in St. Charles County, Missouri. Mary Emma and Orley McGuire had two children, Jacob K. RussellMcGuire and Mary Janet McGuire.27
- Ora Ann Kendall was born on 23 December 1879 in Progress, Union County, Illinois; she died on 28 January 1906 in Makanda, Union County, Illinois at age 26; she married Edward E. McGuire on 9 November 1901.
- Harry Edward was born in 1883; he died in infancy.
- Lydia Pearl Kendall was born on 28 September 1887 in Progress, Union County, Illinois; she died 2 August 1905 in Union County, Illinois at age 17.
We can speculate, but we may never know the reason Olivia was adopted. Now we move on (hopefully) to discover Lynn’s great-grandfather.
Download the full article, with photos, as published in the GFO Bulletin: Finding Lynn’s Biological Great-Grandmother
- U.S. City Directories, 1821-1989; Ancestry.com. St. Louis, Missouri, 1903: image 130 of 1382, p. 253 of Gould’s Saint Louis Directory for 1903. ↵
- U.S. City Directories, 1821-1989; Ancestry.com. St. Louis, Missouri, 1908: image 117 of 1025, p. 242 of Gould’s Saint Louis Directory for 1908 (title page missing). ↵
- U.S. Social Security Applications and Claims Index, 1936-2007; Ancestry.com, 2015. ↵
- 1910 U.S. Census, Pleasant, Garfield, Washington, p. 17B, Olevia Bowne; digital image, Ancestry.com citing NARA microfilm publication, roll T624_1654, FHL microfilm 1375667. ↵
- 1920 U.S. Census, Toutle, Cowlitz, Washington, p. 1B, Olevia Hankins; digital image, Ancestry.com citing NARA microfilm publication, roll T625_1920. ↵
- St. Louis Recorder of Deeds (www.stlouiscityrecorder.org/archives/adoptionrecs/pre1917adoptions.html). ↵
- St. Louis City Deed of Adoption, Recorder of Deeds, August 1903, p. 507-508. ↵
- X-matches and other tools are available through FamilyTreeDNA. For more information, see the X-chromosome inheritance chart. Also, X-Chromosome Matching at Family Tree DNA on the blog DNAeXplained by Roberta Estes (https://dna-explained.com/2014/01/02/xchromosome-matching-at-family-tree-dna/). ↵
- U.S. Social Security Applications and Claims Index, 1936-2007; Ancestry.com, 2015. ↵
- AncestryDNA does not provide chromosome segment information or X-DNA information, the segment segment information for chromosome 20 and the X chromosome were taken from FTDNA. ↵
- Find A Grave, database with images (http://www.findagrave.com), memorial 122948376. ↵
- 1900 U.S. Census, Rich, Union, Illinois, p. 3, Mary E. Kendall; digital image, Ancestry.com, 2004. ↵
- 1910 U.S. Census, St. Louis Ward 27, St. Louis City, Missouri, p. 4A, Mary E. McGuire; digital image, Ancestry.com citing NARA microfilm publication, roll T624_819, FHL microfilm 1374832. ↵
- 1920 U.S. Census, St. Louis Ward 27, St. Louis City, Missouri, p. 20B, Mary E. McGuire; digital image, Ancestry.com citing NARA microfilm publication, roll T624_958, FHL microfilm 1374832. ↵
- Missouri State Archives; Jefferson City, MO, USA; Missouri, Marriage Records, 1805-2002, p. 152; Ancestry.com, 2007. ↵
- Blain T. Bettinger, The Shared cM Project – Version 3.0, August 2017. ↵
- Illinois, Deaths and Stillbirths Index, 1916-1947; Ancestry.com, 2011. ↵
- 1900 U.S. Census, Rich, Union, Illinois, p. 3, Jacob Kendall; digital image, Ancestry.com, 2004. ↵
- U.S. Civil War Soldiers, 1861-1865; National Park Service, M552 roll 58; Ancestry.com, 2007. ↵
- Illinois, Marriage Index, 1860-1920; Ancestry.com, 2015. ↵
- 1900 U.S. Census, Rich, Union, Illinois, p. 3, Delila Kendall; digital image, Ancestry.com, 2004. ↵
- Illinois, Deaths and Stillbirths Index, 1916-1947; Ancestry.com, 2011. ↵
- 1920 U.S. Census, St. Louis Ward 27, St. Louis City, Missouri, p. 20B, Delila Kendall; digital image, Ancestry.com citing NARA microfilm publication, roll T624_958, FHL microfilm 1374832. ↵
- Illinois, Deaths and Stillbirths Index, 1916-1947; Ancestry.com, 2015. ↵
- Most of these details came from records provided by Donna, the granddaughter of Mary Emma Kendall. ↵
- Obituary and cause of death provided by granddaughter Donna. ↵
- 1920 U.S. Census, St. Louis Ward 27, St. Louis City, Missouri, p. 20B, Delila Kendall; digital image, Ancestry.com citing NARA microfilm publication, roll T624_958, FHL microfilm 1374832. ↵