Saturday, March 16, 2013

Flaw in Melungeon DNA Test

             Flaw in Melungeon DNA Test
             This blog post is in responce to the Huffington Post article found here.
I have taken an autosomal DNA test and it came back mostly Caucasian, but we do have some American Indian and some sub-Sahara African DNA as well. Our triracial identity is confirmed. My ancestors were NOT a part of that study, however we lived there, on Copper Creek, Scott Co., Va just like they did and we attended "Stoney Creek Primitive Baptist Church" just like known Melungeon families did, and our closest neighbors were Gibsons descended from a known Melungeon family. My response to that article is below.
 Assume 30 American Indians, full blood, 15 male, 15 female. They are surrounded by 80 percent  Caucasians, and 20 percent African Americans. Assume 20 percent of the American Indians marry others with Indian blood.  Of the remaining 80 percent, 60 percent marry a Caucasian and 20 percent marry an African American. How many generations will it take for people of straight male or straight female line of American Indian descend to disappear?
These will be recorded (m,f) by race. 30 Indians. First generation – 20 % of 30 is (0.2)*30= 6 of these 30 marry other American Indians. These families would be (I,I), (I,I), (I,I) leaving 24 American Indians. SIxty percent of these 30 marry Caucasians, and 20 percent of these 30 marry African Americans. 60% of 30 is (0.6)*30= 18 Indians marrying Caucasians. 20% of 30 is (0.2)*30= 6 American Indians marry African Americans. Assume half male and half female. After one generation we have the following unions.
If we consider the first variable to be male and the second female, which can be represented as (m, f), and replace M and f with the race of th individual as W (Caucasian), B (African), and I (American Indian), we have the following. If we are interested in the American Indian component, we have;
(I,I), (I,I), (I,I) ó (3,3) ó (3 Indian males, 3 American Indian females)
(I,W), (I,W), (I,W), (I,W), (I,W), (I,W), (I,W), (I,W), (I,W)  ó (9,0) ó (9 American Indian males)
(W, I), (W, I), (W,I), (W,I), (W,I), (W,I), (W,I), (W,I), (W,I) ó (0,9) ó (9 American Indian females)
(I, B), (I,B), (I,B) ó (3,0) ó (3 American Indian males)
(B, I), (B, I), (B, I) ó (0, 3) ó (3 American Indian females)
Of the original 30, only 15 have Indian blood on the father’s side for 2 generations, and only 15 have females on 2 generations of the female side. It is important to know that while there is NO new infusion of American Indian blood, yet the infusion of both Caucasian and African blood is inexhaustible.
We can extrapolate that since after only one generation of mixing race, only 50% of the original American Indian y-chromosome is left, that 50% will be lost also in each generation. Likewise, only 50% of the mtDNA, which we inherit from our mothers, will remain from the American Indian side of the lineage. So we can extrapolate further generations –
1st generation – 100%
2nd generation – 50 %
3rd generation – 25%
4th generation – 12.5%
5th generation – 6.125%
6th generation – 3.0625%
In the year 1800 these families were considered mixed-bloods. If we consider 4 generations per century, by the year 2000 we have 8 generations passing since the year 1800. So let’s add 2 more generations.
7th generation – 1.53125%
8th generation – 0.765625%
Oh yes, don’t  forgot, these people were already mixed when they were first recorded about the year 1800, so let us add one more generation:
9th generation – 0.3828125, or slightly more than one third of one percent of the descendants of the original 30 American Indians would still retain the American Indian markers of their y-chromosonal DNA, or the mtDNA markers from an American Indian ancestor. To determine a value we might expect from these original 30 Indians, just multiply (number of American Indians) by the probability that their Y-chromosome or mtDNA is preserved, and after 9 generations we get (30 persons)* (0.3828125)= 1.14843750, or about one of those original 30 might still retain that original information. And lo and behold, look at the Sizemore surname, it DOES retain the Y-chromosome DNA marker of an Indian ancestor. The laws of Probability Theory don’t lie. Conversely, if there is one marker that exists from an isolated community, there could be another thirty families that have lost that marker. And that is exactly what we have.
The key to understand this decline is the limited numbers of American Indians. While each generation there is a new infusion of Caucasian and sub-Sahara African DNA, the American Indian DNA was from a few original donors, only. Through the generations, there was never a new infusion of those markers, so they continued to decline, generation after generation.


  1. Just found your blog. Like you I have been searching for my elusive Southeastern American Indian ancestors. I have Bass mixed blood Indian-European (African?) ancestors so I can trace my family from late 1600's in Bertie County, NC through south,southwest route to my birthplace in Darlington County, South Carolina. My grandmother has undeniable NA features with white skin. Her mother had even more obvious features but without the pale face. (Oops! Bad pun.) I do suspect great grandmother was Catawba. Well anyway
    will follow your blog! Lynne G

    1. Sorry I haven't answered. My Byrum's (great grandma Elizabeth Byrum Hawkins) came from Bertie County, NC! They went to Texas just before the Civil War, and my direct line to Oklahoma.

  2. Hey Lynne, if you see this late comment, we're cousins on the Bass side (my line is John Bass 1673 to Isaac Bass, Jr. 1770).

    Hey Vance, I wonder if our families are connected as well, perhaps through those Melungeon folks. Although I've never had a DNA test (someday, if the budget budges enough), I expect my results would be similar to yours--mostly Caucasian, with some American Indian and Sub-Saharan African. I don't know much about my father's folks (recent immigrants--Scandinavian and German), but I've traced my mother's Southern family way, way back, and besides lots of English, Scottish, and Scots-Irish who came to Virginia, the Carolinas, and New England, I discovered French and Dutch who came to New Amsterdam and Barbados, some Mississippi Choctaws (no paper trail, but they are the Indians I'm surest about), some alleged Cherokee and Creek ancestors, and those mysterious Melungeons, specifically, the families Bass, Bunch, Turner, and possibly Riddle (my Riddles are literally riddles). Those lines are rumored to be part Cheroenhaka, Chowan, Saponi, and more, but who knows? The spotlight has been has been on those infamous Bunch E1b and Bass A1a y-DNA results, and we know they are only a small part of the story.

    I have photos of some of my Mississippi part-Choctaw ancestors (Slades and Rouses) and they confirm what the paper trail does not. In me, there's not much to see, looks-wise--perhaps the profile of the nose and forehead, the teeth, the skin that never turns completely pale, even in mid-winter (though unlike Mom, I freckle). Friends have guessed that I am part-Italian or Jewish, but that's because I grew up near Boston, where such ancestry is more common. There's definitely something there that's not Anglo/Celt/German/Scandinavian. My late mother said I am 1/64 Indian, "probably Cherokee," but I long suspected there was more to our story--and so there is.

    I'll be following the blog and posting more of my typically long-winded comments. Happy hunting, cousins! --Amy

  3. The Riddles, Turner's, Bunches are definitely Eastern Siouan. Basses might be as well. But you need to research YOUR Riddles, Basses, Bunches, and Turners, and make sure they are related to those of known Eastern Siouan heritage. Look at my most recent blog entries (as of mid Feb 2014)