Elizabeth Smallwood (DuBois), "Widow Fuller" - Finally solving the mystery of Elizabeth Fuller

===Case study: the curious case of Elizabeth Fuller===

In 1785, an ancestor of mine by the name of Manly Smallwood married one Elizabeth Fuller, in Philadelphia. This was well documented, as was their family. The problem was that nobody knew where Elizabeth Fuller had come from. Some web trees said she was born in Waterford, now Camden County, NJ, to a father named Samuel Fuller. And that was all anyone knew. Nobody could find any record of a Samuel Fuller, or any other male Fuller of the right age for that matter.

I spent much time looking at various possibilities for Fullers within Philadelphia. There were Fullers who lived there; one Germanic family originally surnamed Foeller had a daughter who married in the same church as Elizabeth and Manly, but she already had a sister named Elizabeth who married a Tannenberg. I found an Episcopal Fuller family whose father had died young and named both sons in his will, but no daughter. I looked for evidence that a Fuller had wound up in Philadelphia as the result of the nascent Revolutionary War, and came up empty. DNA was clearly the only option for resolving this.

About two years after I'd started with DNA analysis, at the point where I'd finally learned how to build a genome map for myself, I stumbled across a chunk of father's side DNA that had multiple matches. One of these matches was familiar because I already was related to her in another way, and she had an identical overlap with a different match that turned out to be from Texas. So I built the Texas tree out, and found that it had one strand that went back to the Morgan and Avery families in Connecticut. I already knew that the first match had New England ancestors, so I began poking around to see how these might be related. And I found errors. Lots of them. I found so many that I basically needed to rebuild the entire Sweet family tree from Lycoming County, Pennsylvania, because someone had simply waved their hands and assumed a connection where there was no actual reason to believe there should be.

When I was done, I knew where the Sweets of Lycoming County had actually come from (Steuben County, New York). I didn't know the maiden name of Joseph Sweet's wife Eliza though. But I was looking for a Morgan or Avery connection, so I looked for evidence that either one of these had had a presence in Steuben County at the right time. It turned out that the Avery family did - and I was able to conclusively rule out any Morgans. This was looking good.

As a final exercise, I took a third DNA match that overlapped these two, and discovered that he, too, was an Avery descendant. That was enough to prove it, as far as I was concerned.

Was this enough to solve the mystery of Elizabeth Fuller? Well, there was an Avery-descended family, the Daniels, who went south from Groton CT to Woodbridge, New Jersey. Even better, a Daniels had married a Fuller at one point. This was looking exciting, even though I still couldn't figure out how an unrecorded Fuller could have gotten to Philadelphia. It was now at least plausible that one had gotten as far as Woodbridge, New Jersey.

But here, once again, the trail ended. While I was fairly certain Elizabeth's mother was a Daniels by birth, I still couldn't plausibly invent a Fuller father out of nothing. It was time to look at a different chunk of DNA.

The thing about DNA is that you can't always just go looking for the chunk that will give you the answers you seek. It's not a predictable process. Basically, you stumble upon it, and that can be today, or next week, or next year. You don't know it until you explore it. In this case, I was lucky; only about six weeks elapsed before I found the next revealing DNA section that would finally unlock Elizabeth Fuller's mystery.

For this chunk of DNA, I had three trees, once again. One tree was a person based in western New York who I could easily determine was an Avery. So maybe this DNA was Avery too? I set out to confirm or disprove this. The next DNA match was from the deep South, and their tree required complete construction, which took quite a while. I became pretty stuck in Onslow County, NC, where he had an ancestor named "Sarah Debose", only there were no Debose families I could find anywhere in that area. But I did find a North Carolina will for an Abraham Dubois, silversmith of Philadelphia, who had the will written up while in North Carolina on business related to his sons. Some of the sons apparently had married there and had children, and my guess was that Sarah Debose was one of them. I was able to learn that Abraham had not died in North Carolina after all, but had returned to Philadelphia where he remarried after his first wife, a Cheesman, died. His second wife had appealed to the courts when Abraham finally died for real in 1804, because she'd given him a number of additional children in the interim.

Looking at the families around for Abraham, I found that there were no less than two other known Abraham Duboises in the Philadelphia area at the time. These were descendants of three Dubois brothers who had come south from their original home in Ulster County, New York, and founded a Huguenot church in Salem County, New Jersey. One brother returned, but two of them had many children in Salem County. Still, since there was no Avery connection, and no connection to the other DNA matches I could find, I went on to another of the DNA matches for this chunk of DNA: a doctor from Texas.

This doctor had a family that was readily traceable with census information back to Tennessee. One of the parents stated she was born in Kentucky. I looked for persons with that surname in Kentucky and was able to connect her to a family on the Ohio River. This family named a son with a middle name "Van Meter", which tied the mother back to eastern Kentucky and a Van Meter family there. These Van Meters had come from Ulster County, New York, and were Dubois descendants.

So two matches out of three were Duboises. Excitedly, I looked at the first match, and she too was explicitly and firmly a Dubois. Case solved! Elizabeth Fuller was, somehow, a Dubois.

It was at this point that I started looking for evidence of Duboises in Elizabeth's life. The attendance records for the church in Philadelphia Elizabeth and Manly attended were available online. I looked for Dubois church attendees in that list - and found one: Elizabeth Dubois, who attended through 1797 and then ceased. This could well have been Elizabeth's mother, a widow at that point. It was clearly hanging together. The only remaining part was figuring out which Dubois was the father, which Daniels was Elizabeth's father, and how Elizabeth had gotten the maiden name "Fuller".

At the beginning of this exercise, I had looked for Fullers in Philadelphia. A John Fuller had died there in 1767, describing two young sons in his will. One was named "Samuel". He was born at just the right time to be Elizabeth's first husband. Presumably he died after they'd been married only a short while - perhaps the Revolutionary War had something to do with that, since Samuel's brother was a sergeant in that conflict. Elizabeth Fuller then remarried to her deceased husband's comrade in arms, Manly Smallwood. She was a widow at age 20. But her maiden name was probably Dubois.

Jonathan Daniels, who married Mary Potts (the Avery descendant) was born in New London, Connecticut. He married there, had several children there, and then began a slow trek through Connecticut and New York into Middlesex County, New Jersey, where he died in 1764, leaving an extensive will. Many of his children accompanied him to Woodbridge. Elizabeth's mother was probably the daughter of one of them. But which one?

Dates ruled out the later children. And when I looked at the earlier children that had gone to Woodbridge, I found that Benajah Daniels (who was Jonathan's executor) had a family that was very poorly described. Not only that, but web trees persisted in giving him a daughter named "Elizabeth Rose", only they gave her a birthdate that was impossible given the dates of Benajah and his wife. I began to suspect that somebody had misheard or misread "Elizabeth Debose" as "Elizabeth Rose" and had managed to confuse generations of genealogists in that way. This was an excellent place for Elizabeth Fuller's mother to have come from.

For Elizabeth's father, I looked at sons of both Salem County Dubois brothers carefully. One son was said by web trees to be "Charles Abraham" or "Abraham Charles". The conjunction of these two names was strikingly original, and I wondered at first if someone had inadvertantly merged together twins. But it turns out that this mysterious "Abraham Charles" was said to be born on two different dates, exactly three years apart. This could have been a typographical error, or it could in fact represent a situation were there were two sons who were confused because they were born exactly three years apart. Even more interestingly, a woman named Elizabeth Preston was recorded to have married Charles Dubois. If Abraham Dubois married Elizabeth Daniels, the confusion would have been understandable.

===Confirmation===

With this hypothesis, the resolution of the Elizabeth Fuller mystery was within sight. All that remained to be done was confirm it.

One way to confirm it had already happened: records were located which hinted that Elizabeth Dubois may have been Elizabeth Fuller's mother. On top of that, I found a death record in Philadelphia for Elizabeth Dubois which was consistent with the church records. But unfortunately it did not state the name of Mrs. Dubois's spouse, only that she was a widow.

The true confirmation therefore had to come from DNA. This work is still in progress. For the Daniels side, a half-dozen matches have been linked back to the Daniels family, or at least to Woodbridge or other places along Jonathan Daniels' migration route. The Dubois side has not yet begun; there aren't as many matches.