My Old Ninety-Six, South Carolina family was significantly impacted by three pulses of post 1865 migrations out of the South.
The first pulse came quickly upon the heels of the Civil War, with groups of ancestral kin leaving South Carolina for Ohio (Elyria, Ohio in particular), North Carolina (Winston-Salem, Asheville, and Raleigh in particular, as well as Halifax, Wake, and Buncombe Counties in general), and Arkansas. Smaller groups rode out to Kentucky, Virginia, and Tennessee during the same time period.
The second migratory pulse occurred during the WWI era, with large groups of kin heading north to Michigan (mostly Detroit), Illinois (Cook County), Pennsylvania (Philadelphia, Pittsburgh, as well as Chester County), Washington D.C., Maryland (Baltimore), Delaware, New York, and New Jersey. A second pulse into North Carolina also occurred during this period.
The third great movement of family occurred during WWII and continued through the 1950’s. Their ports of call mirrored those in the second pulse, with the addition of Massachusetts, Georgia, and, to a lesser extent, California, as places for relocation.
Military draft cards were essential to unlocking the mystery of which family lines moved elsewhere in the US – and when they moved there.
One clue to the extent of these migrations within my family is the sheer number of entire family groups who seemingly fell off the face of the earth after the 1870 Federal Census, and accompanying state census records for the same time period. They had to have gone somewhere. The question was: where? I bounced a few ideas around about how to find these missing families. Then it hit me: WWI and WWII draft cards.
My reasoning was pretty straightforward. The women in the family who were caught up in these mass migrations out of the South would be nigh on impossible to find first due to name changes via marriage. This is especially true for my South Carolina ladies, who came from an enormous interconnected family that used the same dozen or so names for their daughters. Sure, I might get lucky and find the right Janie Gilchrist in Buncombe, North Carolina straight away. The reality, however, is that I would need to research at least a half dozen or so Janie Gilchrist’s born in the Old Ninety-Six region who popped up in Buncombe County. Add the complexity that some of these Janie Gilchrists will have parents who haed similar names, well, trying to separate each Janie from all the others becomes a hurculean task.
It was easier to track the males in the family for the simple reason that their names rarely changed. I began with my Mat(t)hews / Mathis / Mathes lines, and rolled this approach out to my Holloway and Peterson lines. Each of these three family lines were huge. No, seriously, these were among the largest families in Old Ninety-Six. Tackling these three lines first would answer questions about where cousins and other members from their extended families also moved to.
The image below shows how I began my search using military draft cards.
I did a very general search on the surnames Matthews Mathews Mathis Mathes (I am going to collectively refer to them as Matthews). And yes, Ancestry returned an enormous list of men. Which makes sense. This was an enormous family. Faced with a breathtaking list of men – many with the same or similar names – I concentrated first on those I either immediately recognized, or those I could relatively easily figure out in relation to my family tree. I needed to do these first in order to reduce the number of men I was faced with. I looked for a few fundamental things to attach the right record to the right man. These cards also answered a basic question: the Matthews surname variation each man chose, which was also adopted by his descendants. I have men who born a Matthews yet died a Mathis. Knowing the preferred spelling enabled me to find them in various records.
Draft cards told me:
- They often told me what part of the Old Ninety-Six region, and adjacent counties, they came from (e.g. Edgefield, Greenwood, McCormick, Abbeville, Greenville, Newberry, Barnwell, Aiken, etc. Knowing where they were born enabled me to zero in on specific Matthews family groups;
- Middle names or initials as unique identifiers. When you’re faced with dozens of William, Willie, Wiley, Bill, Billy, and Bill Matthews, anything that distinguishes one from the dozens of others with the same, or similar, name is crucial; and
- If they were residing with a relative, did I already know who that relation was? A parent, sibling, cousin, or aunt or uncle, can make the correct identification of an individual a much easier process. The name of a spouse can help. This isn’t necessarily the case for Old Ninety-Six. I recall looking at eight WWII draft cards for men by the name of William Matthews (or a name variation) married to a Lula, who was born in Edgefield, SC – and all of them were living in Philadelphia.
Once I had reduced the list of Matthews men as per the above bullet items, I began to slowly, methodically, chip away at the remaining Matthews men who were left in my draft card search. This is where things became time consuming. You really have to love genealogy to go this route.
One by one, I did a research workup on all the remaining Matthews men in the draft card list. The critical pieces of information were dates of birth. Be advised that the year of birth given on draft cards can vary from 1 to 5 years from the actual year of birth. The important thing to note is that the day and month don’t change.
Armed with specific birth dates, I could usually find a Social Security Application record for these men. From there, I could usually find a death record with the name of one or both parents, if their names where known, as well as their respective places of birth. Next, I looked for obituaries, which tended to give the names of parents and siblings; including the places where that person’s surviving family members were living at the time of that person’s death.
Death certificates also offered further information via the informant, if the informant was a family member. If the informant was a sibling, cousin, aunt, uncle, or in-law, you now know where that family member was living at that time.
Then, and only then, did I hit the Federal Census returns. This is genealogy retrofit style. You work from the known, add as much information as you can, then work backwards until you find the family line of the person you are researching, and carry that line back to an established line in your tree. Thankfully, in my case, that backwards journey takes between 1 to 3 generations to make that connection. Again, this is the blessing of having a large tree. And yes, there are times I hit a brick wall when I work with a research line in this way. However, sooner or later, the crucial missing puzzle piece is found. Or someone who knows that line inside and out, usually a descendant, will reach out with the missing information I need. So I don’t sweat these temporary brick walls too much.
This research approach worked around 6 out of 10 times. This is where having a large family tree (nearly 100,000 individuals) has come into its own.
I know my Old Ninety-Six family. I understand the various themes that run through my family’s history. It was rare for an ancestor or kinsperson from South Carolina to ride out on their own. No, my people moved in groups. More often than not, entire extended families just upped sticks and moved together. Draft cards began to demonstrate this. If I had one or two family members removing themselves from Greenwood, SC to Philadelphia, the chances were high other family members did too. I was able to pick up the thread for a person’s parents, siblings, cousins, etc. I could find them in Philadelphia too. More often than not, they were living in close proximity to the person I was researching.
In my next post, I’ll share how using military cards and obituaries in tandem can yield some stunning results – and smash through some brick walls.