A Double Mystery

This mystery was a pleasure, and a challenge, to solve. At first glance, it appeared to be an easy case, as there were many close matches, but in the end even “easy” can be deceiving.  All names and some identifying descriptions, such as ages and places have been changed.

Please note: The full report also included all sourced materials, including death certificates, census reports, newspaper clippings, etc. It also included family group sheets, charts of DNA matches, and other historical information as evidence of these findings.

Objective:
To determine Evan Thompson’s biological father.

Background:
In 1997, Evan Thompson was born in Richmond, CA to a 37 year old woman named Emily Thompson. Emily passed away when Evan was a child, and never told him who his father was.  In 2017, Evan took an Ancestry.com DNA test in hopes of learning more about his heritage and in 2019 turned to Family DNA Finder for help untangling his DNA matches.

Findings:
When I first looked at Evan’s Ancestry family DNA matches, I was able to quickly divide the majority of the matches into two groups: 1) Maternal & 2) Unknown, but presumably Paternal. I first marked each of the maternal results that he recognized from his mother’s family with a pink dot designating it as being from his mother’s side. Then I looked at each of those family members’ shared matches with Evan and marked them also with a pink dot. Eventually I had about 30 people marked as being probable maternal side matches. I then looked at Evan’s list of DNA matches and marked anyone who didn’t have a pink dot with a dark blue dot. This was the first division between maternal and assumed paternal results. A pink dot meant maternal side; a dark blue dot meant paternal side.

My next step was to look at his highest “dark blue dot” matches, which were in the 572 cM-200cM range. The two highest matches were Harvey (572 cM) and Cletus (374 cM). These two had a lot of DNA matches in common with each other, which led me to believe that they were from the same side of the family. I took a look at all of their shared matches, and marked them with a red dot. I then looked at the family trees of the shared matches and found a set of common ancestors, Scott Purdue and Bertha Vinson. Using genetic genealogy charts, I could tell that Harvey & Cletus’s cM levels would have them be about two to four generations away from Evan. Scott Purdue and Bertha Vinson could be Evan’s grandparents or great-grandparents. I created an extensive family tree for Scott Purdue and Bertha Vinson, looking at each of their 6 known children and each of those children’s children, etc. The information was found using Ancestry databases & records, other family trees, obituaries, newspaper archives, social media posts, and other online records. I found that Evan’s high matches, Harvey and Cletus, were first cousins, and their grandparents were Scott Purdue and Bertha Vinson

I then looked at the next set of Evan’s matches that a) were not maternal matches, and b) also did not match the Purdue/Vinson family. This was so I could perhaps find more shared sets of ancestors that would eventually divide into paternal grandparents or great-grandparents, etc. Evan had quite a few clusters of matches that were relatively high in cMs, meaning we wouldn’t have to look too far back to find a common ancestor. Right away I found a common ancestor couple named Joseph Graca and Maria Chavis. I did the same as I did for the Purdue & Vinson couple, and created family trees for all their descendants, siblings, parents, and grandparents.

As I followed the Graca & Chavis family, I found they had two daughters and one son. The son and one daughter had descendants who tested on Ancestry and who matched Evan in the 300 cm range, which made it less likely that they would be Evan’s parent or grandparent. I took a closer look at the remaining daughter, Theresa Graca, and found a few things that seemed to stand out. One, her age was in line to be a grandparent to someone born in 1997. Second, she had lived in a county that was geographically close to where Evan had been born. Third, she had a son whose age was in line to be Evan’s father. But more investigation needed to be done before making a determination.

Over to the Purdue and Vinson family, they had 6 children, and all but two had stayed close to their home state, Texas. The two who hadn’t, Gerome and Phillip, had moved to California around the same time. I decided to look closer at those two.  They both were similarly aged, and could both be old enough to have a grandson born in 1997. They both lived in Northern California, though one, Phillip Purdue was a couple of hours away from Richmond, CA. The other brother, Gerome Purdue, had lived in a county adjacent to Richmond. Phillip had been married for a long time, and all family trees linked to Gerome showed he had never married. This still didn’t mean that either of them, or even their siblings, could be ruled out as a grandparent to Evan.

I spent some more time on Theresa Graca’s immediate family and history. She had been married twice, and had two daughters and a son. The son, Sam McKay, seemed like an excellent candidate to be Evan’s father. He hadn’t married, he lived near Richmond, CA, he was the right age, and his relationship to those who had DNA tested from the Graca family fell within the expected range for him to be a father to Evan. But there was an issue. In order to be Evan’s biological father, he would also need to have ties to the Purdue/Vinson family who Evan had a lot of close matches with.

I was able to formulate a family tree for his father’s side, the McKays, and there was no connection to the Purdues or Vinsons. This left me with two possibilities: a) he wasn’t Evans birthfather, b) he wasn’t actually his presumed father’s son.

I decided to take each possibility one-by-one. If Scott McKay wasn’t Evan’s birthfather, then who was? I looked closer at Theresa Graca’s records. She had passed away several years prior. This meant that she had a social security file that was indexed on Ancestry. One line was tucked in it, just a small footnote: in 1962 she had changed her name to Theresa Purdue.

There was no marriage record found, no divorce record, no internet or newspaper notation somewhere that made any mention of a marriage between Theresa and a Purdue. All we had was one small clue hidden away in an index.  

Evan and I decided to reach out to Theresa’s children to see if they would be willing to shed some light for us. We had gone as far as the records and DNA could take us. My theory was that Theresa and Mr. Purdue (and I had to make an educated guess that it was Gerome Purdue, based on his brother being married at the time and other siblings living many states away), had placed a baby boy for adoption, and that son was Evan’s birthfather. The other possibility was that Mr. Purdue was the biological father of Sam McKay, but the timing wasn’t perfect on that, with him being born several years before Theresa changed her name. That wasn’t out of the realm of possibility, but it was a theory that I’d return to later if my original theory was wrong.  

We contacted Theresa’s daughter, Kelly, and she very graciously gave us some very important information. Yes, her mother, Theresa Graca, had placed a son for adoption around 1962. She didn’t know much more than that, as her mother had told her this 40 years ago, and then passed away. She did not recall a Mr. Purdue, not as a step-father or other presence in her life (she was under five in 1962). She had found one piece of paper where her mother had disclosed she had changed her name to Purdue, but nothing other than that.

I took that information and started looking at the California Birth Index. I searched for a boy born between 1960-1963 in Contra Costa County with a mother’s surname as Graca. I found one born in May, 1962. He was listed without a first name, just his mother’s surname and a father’s surname as McKay – not Purdue. I was able to connect with a person who had more information from the California Birth Index than I did, and who could check to see if a California birth certificate had been re-issued in another name, but the same certificate number, as is done in California adoptions.

The information came back – the boy I found born in May 1962 had been adopted and his new name was Travis Jessup.      

My next step was to get any contact information I could for Travis to give to Evan. Unfortunately, I soon found out that Travis had passed away.

I gave Evan all the information I could about Travis, including about his adoptive family. We found out that Travis had a daughter who was a bit older than Evan and lives not very far from him. Travis’ (adoptive) brother was able to give a photograph of Travis to Evan, and pass on some of his story.  All of the information the brother knew from the adoption lined up with what we had found out. Evan is now in contact with his biological half-sister.

Conclusion:
This began very straight-forward, with lots of high DNA matches with wonderful, accurate family trees on Ancestry which pinpointed the most recent common ancestors of Evan’s biological father. What we weren’t expecting was to find that it was in essence a double mystery – an unknown father who had unknown parentage himself. Nonetheless, using DNA and intensive genealogical research, the search was completed successfully.  

An 87 Year Old Question

I had the honor of helping solve an 87 year old mystery this past year. The following is a modified report presented to my client, who was born in 1932. All names and some identifying descriptions, such as ages and places have been changed.

Please note: The full report also included all sourced materials, including marriage licenses, death certificates, census reports, newspaper clippings etc. It also included family group sheets, charts of DNA matches, and other historical information as evidence of these findings.

Objective:
To determine Lorena Candela’s biological father.

Background:
In 1932, Lorena Candela was born in Modesto, CA to a 17 year old woman. Lorena was never told who her father was, and assumed she would never find the answer. In 2017, Lorena took an Ancestry.com DNA test in hopes of learning more about her heritage and in 2019 turned to Family DNA Finder for help untangling her DNA matches.

Findings:
When I first looked at Lorena’s Ancestry family DNA matches, I was able to quickly divide the majority of the matches into two groups: 1) Maternal & 2) Unknown, but presumably Paternal. I first marked each of the maternal results that she recognized from her mother’s family with a pink dot designating it as being from her mother’s side. Then I looked at each of those family members’ shared matches with Lorena and marked them also with a pink dot. Eventually I had about 45 people marked as being probable maternal side matches. Nearly all the people who had pink dots had some Italian ethnicity showing in their Ancestry results, which helped me know I was on the right track, as Lorena’s mother was Italian. I then looked at Lorena’s list of DNA matches and marked anyone who didn’t have a pink dot with a dark blue dot. This was the first division between maternal and paternal results. A pink dot meant maternal side; a dark blue dot meant paternal side.

My next step was to look at her highest paternal matches who were not her immediate family (because her son & granddaughter who had also tested would have both a pink & dark blue dot since they share DNA with both of Lorena’s maternal & paternal sides).

The two highest matches were Denise (229 cM) and Michael (222 cM). These two had a lot of DNA matches in common with each other, which led me to believe that they were from the same side of the family. I took a look at all of their shared matches, and marked them with a red dot. I then looked at the family trees of the shared matches and found a set of common ancestors, Ethan Wilson and Ida Keane. Using genetic genealogy charts, I could tell that Denise & Michael’s cM levels would have them be 3rd cousins or closer to Lorena. This meant that their shared ancestors, Ethan Wilson and Ida Keane, could be Lorena’s grandparents, great-grandparents or great-great grandparents. I created an extensive family tree for Ethan Wilson and Ida Keane, looking at each of their 8 known children and each of those children’s children, etc. The information was found using Ancestry databases & records, other family trees, obituaries, newspaper archives, social media posts, and other online records.

I then looked at the next set of Lorena’s matches that a) were not maternal matches, and b) also did not match the Wilson/Keane family. This was so I could perhaps find more shared sets of ancestors that would eventually divide into paternal grandparents or great-grandparents, etc. I found clusters of matches who shared sets of common ancestors, but because their shared cM with Lorena was relatively low, it wasn’t completely obvious how they all fit in. The highest match who didn’t match Denise or Michael was Raymond at 96 cM. I was able to build a family tree for him and several of his shared matches with Lorena and found that many of them had a shared ancestor with the name Apfel. I marked them all down with a light blue dot, even though I wasn’t quite sure which Apfel ancestor they all came from.

After Raymond, the next set of matches that didn’t match him or Denise or Michael, were DNA matches who had Gregory Lloyd as an ancestor. I marked them with an orange dot.

I found more clusters, and color coded them, but it still wasn’t clear how they all fit in.

I went back and decided to focus on Ethan & Ida Wilson’s son, Jefferson Wilson. A few pieces of information immediately came to light. First, Jefferson Wilson had married a daughter of Gregory Lloyd & Josephine Apfel, named Joan Lloyd. If you recall, two of my separate clustered groups had Gregory Lloyd & the Apfel family as ancestors. So here were two separate families that had DNA matches to Lorena coming together. Then I looked at Jefferson & Joan’s offspring to see if anything stood out. They had four children, but only one lived into adulthood, a daughter named Rebecca Wilson. Rebecca had married a man named Jacob Waters. They had two sons, but one died when he was 4 years old. The other son was Jason Waters. I decided to look at Lorena’s DNA matches and see if any of them had Waters in their family trees. I found 6 that I could directly link to Jason Waters’s great-grandfather Nathaniel Waters, plus dozens more that also were shared matches with those 6 in the cluster. Then I looked at who Jason’s other paternal relatives were, and found that his paternal grandmother was Christa Kirk. I looked to see if Lorena had any DNA matches with Kirk descendants, and found at least 5 that I could find a direct connection to Christa Kirk’s father Harrison Kirk. So, to recap, out of Jason Waters’ 4 sets of great-grandparents (Waters, Kirk, Wilson, Lloyd), Lorena had multiple DNA matches with descendants of all in the 2nd-4th cousin range (229 cm-40 cM).

With this information at hand, I went to look at Jason’s history to see if there was a chance that he could also have been in Modesto in the late part of 1931 or early 1932. Right now, there isn’t anything that links him to Modesto, however, he moved from Kentucky to Sacramento by the time the 1930 census was taken. In 1932, when Lorena was born, he would have been 19, and Lorena’s mother was 17. The geography isn’t a perfect match, but all of the DNA links to his family living in Kansas, Kentucky, Arkansas, and Oklahoma, and only this family unit, with this set of diverse ancestors, moved to the Central Valley, which Modesto and Sacramento are part of.

Conclusion:
I can find no other instances of members of this Lloyd family and this Wilson family marrying, nor of this Wilson & Waters families marrying or producing offspring. Since Rebecca Wilson, an only child (past childhood), & Jacob Waters, an only child, did not have any other known offspring that lived past childhood, there is only one likely option to be Pat’s biological father, and that is Jason Waters.