A distant cousin
I was recently contacted by a distant relative on 23andme, based on a shared last name in our family trees (you can list a set for others to see) and a shared fragment of DNA. We were able to trace our connection to ancestral siblings born in the 1720′s (eight generations between me and the parents of these siblings!). My distant cousin told me more about that branch of the tree — the mother of both siblings can be traced back another three generations to Constance Hopkins, a 14-year-old passenger on the Mayflower and daughter of Stephen Hopkins. There were only 102 passengers on the Mayflower, which sailed in 1620, and half of them died the first winter. My great-great-great-great-great-great-great-great-great-grandmother Constance survived, however, and had a dozen children with Nicholas Snow.
Genealogy just got real
One amazing thing about this connection is that the shared DNA proves, beyond reasonable doubt, that our genealogies are solid for every link in our genealogies descending from those 1720′s siblings! Non-paternity events are common enough to make me suspicious of family trees tracing into the distant past: with five father-to-child links in those eight generations, how sure could I have been that all those links were honest? If you assume a non-paternity rate of 10%, there’s only a 60% chance none of those generations had a little “shenanigans” (.9**5 = 0.59). A more optimistic non-paternity rate of 4% raises the estimate to 80% (.96**5 = 0.82), but you can see why it’s hard for me to get too excited about these things.
Now this genealogy is real to me in a way it wasn’t before — before this distant cousin contacted me it was just a hypothetical set of historical records, but now I can be quite confident that the links back to that pair of siblings are a true history.
That leaves three unconfirmed generations between me and and Constance, only one of which was a father-to-child link (potential non-paternity event). It sounds like I can be fairly confident in stating I am a descendant of an original Mayflower passenger!
What are the chances of that anyway?
The Mayflower passengers were a tiny number of people …but… as Constance’s own records show, those that survived were extremely prolific. According to Wikipedia, between 1640 and 1790 the population of New England grew from ~13,700 to ~900,000 with almost no influx of immigration.1 Assuming an average of 30 years per generation (which is true for my own ancestry to Constance Hopkins) that’s five generations: each individual in 1790 had about 32 ancestors from 1640. That’s a bit high — mixing isn’t perfect — and I’m not sure we can trust that *no* immigration occurred. To be conservative let’s assume there’s effectively an average of 24 ancestors from 1640′s New England for each individual in 1790.
How many people in 1640 (out of 13,700) were Mayflower passengers or descendants? I’m guessing that the 50 survivors of the 1620 trip could’ve grown to 100 settlers & descendants by 1640. That means about 0.73% of New Englanders in 1640 were Mayflower descendants.
Putting these numbers together, we would predict 1 in 6 New Englanders in 1790 would have at least one of their twenty 1640′s ancestors from the Mayflower (1 – 0.9927**24 = 0.16 =~ 1/6). The total US population at the time was 3.9 million, so implies around 1 out of 26 US residents at that time was a Mayflower descendant (6 * 3,900,000 / 900,000 = 26).
At this point I’ll give up on guesswork. Immigration starts to play a stronger role in the growth of the United States after this point, but those immigrants mixed with the existing population. With seven more generations worth of mixing, the fraction of Mayflower descendants could easily be higher than it was in 1790. I’m guessing a fair number of people in the US can trace some ancestry to the early New England — and if they can, I think there’s a good chance they have an ancestor descended from that tiny initial group that came over on the Mayflower.2
1The Wikipedia paragraph referenced gives the 1790 population as 700,000, but I decided to use the slightly higher numbers from the table below (“Estimated Population of American Colonies 1620 to 1780″). 1790′s number of 900,000 is based on an estimated growth rate of 2.32% per year, which is based on the New England population growth from 1740 to 1780 in that chart.
2As I have been recently reading 1493, I’m keenly aware of how these settlers can also be described as invaders who took advantage of a collapse in the native populations. I don’t think Plymouth and the first Thanksgiving should be romanticized, nor should I feel unfair guilt over it (it’s hardly my fault, and most of the native collapse happened before they got there), but the history of Plymouth is a bit more real to me now.