Three Chances — Maybe

 

The research task this month has been to follow the matriline leading from my husband to his mother, to her mother, and on up the line from daughter to mother through the generations. The reason for this search has been to sketch out the possible other descendants of that as-yet-unknown matriarch to determine the connection between my husband and his mtDNA matches.

Finding yet another ancestral sister in that matriline has opened up possibilities to help determine just how these mtDNA matches connect to my mother-in-law's line. This month, we discovered another sister of my mother-in-law's second great-grandmother, a woman born in Ohio about 1821 by the name of Rosanna Jackson. Having found her, I discovered Rosanna had married Walter Mitchell, and the couple eventually had three daughters: Mary, Martha, and Sarah. But before you assume that gives me three chances to find the path to those matrilineal matches, there is one caveat I need to introduce: youngest daughter Sarah was most certainly daughter of Rosanna's husband Walter Mitchell, but we'll have to look closer at the records to see whether she was actually Rosanna's own daughter.

Why? Because about the time Sarah was born, Rosanna had died. At least according to her headstone, Rosanna had died in Iowa on October 14, 1862. According to the earliest census in which I can find Sarah's entry, the U.S. Census for 1870, Sarah was listed as eight years of age, giving a year of birth around 1862. But in that enumeration, we also realize that the wife in the Mitchell household was not Rosanna, but another wife by the name of Bridgett. Since Bridgett was sixty eight years of age at that same time, it is unlikely that she would have been the mother of Sarah, but a lot can happen in the undocumented years in between decennial enumerations—not to mention reporting errors and other possible recording mishaps. I'd feel more confident in that assertion of Sarah's birth to Rosanna in 1862 if I could find records affirming that maternal connection.

Another problem with the assumption that we have three chances to locate possible matrilineal matches through Rosanna's line lies with the identity of oldest daughter Mary. While Mary appeared in the 1860 census as an eight year old in the Mitchell household in Chickasaw County, Iowa, she was not listed in the subsequent census. True, by the time of the 1870 census, Mary could have been married and starting a household of her own, so it is no surprise to see her missing from her father's household. But there was one other troubling clue, and it came inscribed on her mother's headstone.

If Mary Mitchell, born in Ohio in 1852, was one and the same as the "Mary G." listed on Rosanna Mitchell's headstone, she apparently died in December of 1862. Thus, no possibility of any children—let alone daughters—from Rosanna's daughter Mary. Admittedly, as many families of the time did, Walter and Rosanna could have named a subsequent daughter by that same name, and if so, we can still search for a descendant named Mary. I'll leave that possibility open, but I tend to doubt that was so, based mainly on Rosanna's own death about that time.

Those discoveries leave us with two main tasks in searching through this matrilineal connection: determine whether baby Sarah was indeed daughter of Rosanna, and search for signs of any other sister named Mary Mitchell. Once we clarify those two details, we'll be off to start building a descendancy chart for each of Rosanna's remaining daughters, whoever they turn out to be.