Ah, the census. The glorious census. It’s such an important document in genealogy, I would venture to say even a foundational document. But its flaws are many. Those who rely on only the census to reconstruct their ancestral families do so at the risk of recreating a family inaccurately. I’ve discussed this topic before, but this lesson deserves repeating. Often. Take for example the case of my ancestor Lydia.
Lydia Wilson married Daniel Waters 23 July 1879 in Somerset County, Maryland. In 1880, the couple is shown in the census with their young child, John. (image above)
By 1900, Daniel has presumably died and Lydia, with new husband Edward Cottman, was enumerated with 4-year-old daughter Mamie in the household, along with sons Cranston, James and George, who are surnamed Waters (they are mistakenly marked as boarders).
In 1910, Lydia and Edward are enumerated with James (marked as a stepson) and Mamie:
In 1920, daughter Mamie remains in the household with 67-year-old widowed Lydia:
All told in these four census records, we count four children.
But the careful examiner of the 1900 and 1910 censuses will notice that Lydia has birthed six children, six of whom are living. Who are the other two children? Would you have noticed or missed that?
Well, lucky for us, Lydia left a will. In that will she names all six of her children: Mamie, George, [John] Cranston, James, Fannie and Allena. Of course, that record cannot be found online. Deed records for Lydia’s land would have also uncovered the names of her children, and usually, those records also are not online.
But which kids were fathered by which of Lydia’s husbands?
The 1900 census enumerates Lydia’s three sons with the surname “Waters” and her daughter Mamie as a “Cottman.” If we assume for now that those names are accurate, that still leaves us daughters Allena and Fannie to properly place in the family.
In county marriage records, we find that Lydia’s daughter Fannie Waters married the Rev. Fred Gillis. Fannie and Fred later lived in Smyrna, Delaware, as mentioned in Lydia’s will. Is that marriage record online? No.
Lydia’s daughter Allena was trickier. Like her sister Fannie, she is never enumerated in the household with her mother Lydia. And the reference to Allena “Waters” in Lydia’s will would easily leave one to believe that she was the daughter of Lydia’s husband Daniel Waters. But she wasn’t.
Allena married a man named Emory Graham Waters, and the couple can be followed in census records in Somerset through 1940. Philadelphia, Pennsylvania records the marriage of this couple in 1903. Thus, Allena’s married name was Waters. Allena’s surname (according to her marriage record) to Emory was…..Wilson. If you’ll recall, that was her mother Lydia’s surname. But….did Lydia have her out of wedlock? Certainly not impossible, but let’s take a closer look.
I’ve emphasized in this blog before the need to examine original records, and for us to not be satisfied with the indexed records we find in online databases and books. We should always—always—strive to retrieve the original record and examine it for ourselves. Examining the original marriage record for Lydia and Daniel, we find Lydia Wilson noted as a widow. That’s an important clue that tells us two things. There was a marriage before Daniel Waters and “Wilson” was not Lydia’s maiden name.
A search of original records in Somerset County record no previous marriage for Lydia. So I thoroughly researched all of her children; I retrieved the records of their marriages, deaths, land purchases, etc. And finally, only in daughter Allena’s death certificate do we find what couldn’t (or rather, hasn’t yet) be found anywhere else—the name of her mother’s first husband: Alfred Wilson.
This previous marriage is especially hard to discern because Allena, as the only child of that union, is not living with her mother in 1880 as we would have expected. I do find her in Philadelphia in 1900, working as a servant:
This example aptly illustrates the absolute folly of depending on census records to tell you the complete story of a family. The census must be used in conjunction with other records. And those other records are mostly not online. You’ve got to either write to a repository (courthouse, archives, vital records office) to obtain copies of other types of records or go there in person to do research. I am fortunate to live in Maryland, so I’ve spent many days at the State Archives and was able to pull these records myself.
A cursory examination of the census records alone for Lydia’s family would have left us to believe that Lydia had two husbands and four children. But analyzing and reasoning through the evidence helped us to properly orient Lydia’s family, and uncover her first marriage.
By the way, Lydia’s maiden name was….. Bowser:)