The End and the Beginning

When I ordered my records from the State digital archives I also ordered several other records from Madison County. From the Anderson Library I ordered some obituaries I thought might belong to my family and I also ordered death records from the county using the forms I got from the Madison County Historical society.

The Health department kindly mailed my original forms back to me in the self addressed stamped envelopes I sent them. I'm not sure if there was some error in communication or if there is some message there.

I did however get the obituaries from the library and they were most informational. One lists Rosa with several children and the other references Levi without naming him.

This is a nice little obit listing the children and even her sister who I now know is an Eaton, but sadly not her husband. The next excerpt is a bit more ...tabloid, but it does contain some good information for me.

Disappeared! How, why? What really happened? The text is so vague, what is "these 20 years ago"? Does it mean 1886? Were there no police or no sheriff in this town? What was this lawsuit? If he disappears in 1886 how can I pick him up in a veteran's roll in 1890 in Ohio? Why is his pension request only closed in 1902?

"A later day will have to be awaited in order to solve the mystery."

Butler County Pennsylvania

I did a search of the Indiana State Digital Archives and found a civil war record for Levi Thompson. His civil war records have been the most effective way to track him (as in the 1890 veteran's roll in Ohio). Unlike his other documents from the National Archives or his online indexes, this one listed some more interesting personal information. Including the thing that made me change the secondary title on this blog.

Eyes hazel; hair dark; height 5'10"; complexion dark; nativity Butler Co Pa. Not just a state or several states, but a COUNTY! So his wife was right. He's from Pennsylvania. Having spent more than a year trying to find such a minor piece of information...well it makes the smallest victory exciting.

Finally some concrete reason why he's not with the other Thompson families in the area and  a big bolster of support for those Pennsylvania genetic relatives.

I was also sent the information on other Thompsons in the area as well. At least two other Thompsons served in the vicinity. Both were 19 at the time, both born in Indiana, both named John and I've been able to find both in the 1850 census. None with Levi though. Without some further proof, it appears that he is unrelated to the other Thompsons serving with him, although the physical descriptions of the other Thompsons are very similar to him. 

For reasons I can't really defend logically, finding a place for Levi has been very important to me. I have a place. He does too.

Hollingsworths

As I've contacted genetic relatives and asked permission to view their trees, I've made it a habit to try to check them against each other to see if they have any matches. I have a lot of matches in the Pennsylvania area, and sorting them out can be pretty hard, not least because they are so interrelated. I began with just one genetic relative who happened to be from Pennsylvania and also happened to have Thompsons. She actually has two distinct Thompson families. One family beginning with Robert Thompson born 1786 in Lancaster Co PA and the other beginning with James Thompson born 1668 Wicklow Ireland (but moved to New Jersey).

Previously I had gathered a list of surnames surrounding the Lancaster PA Thompsons to see if they appeared fairly often among my genetic matches. Many of them did. When I find a pattern in my matches in simple surname searches at 23 and me, I like to search through the trees I have shared to see if there is anything there.

What I've found is that the New Jersey Thompsons marry into the Hollingsworth family right there in the early 1700s with James and his wife Ann Hollingsworth. I also found that a second Genetic match had Hollingsworths in her tree right there at the same point. The second match was related to Randal Malin who marries Mary Hollingsworth (sister of Ann).

What really blew me out of the water was that a third genetic match was related to the Hollingsworths at Mary Hollingsworth born 1816 in Illinois.

The other odd irony is that the Hollingsworths seem to be active right in the same area of Pennsylvania and Delaware that the New Sweden colony was. Some were even married in the old Swedes church.

Of course there are no Hollingsworths in my tree that I know of (but that's what I know). I seem to be related to people who are related to Hollingsworths, but I do not genetically match with two Hollingsworths I've seen at Gedmatch.dom.

They may not be my family but clearly they are to be watched.

http://familytreemaker.genealogy.com/users/d/a/r/Earl-L-Darrah/GENE3-0005.html

Finns, Swedes and Danes?

The long haul to figuring out my Y DNA results has left the playing field for Thompson ancestors more open rather than closing it up. I'm R1b-U106 which is pretty popular anywhere around the North and Baltic Seas. My closest Y DNA matches at Ancestry.com are from Denmark and England. YHRD has Denmark and Germany, SMGF has Anglo-americans, Swedes, Germans and Russians. That's a lot of ground to cover and these people have been mixing and matching for centuries.

In looking for some information on the Corson family (higher matches at Ysearch) I got embroiled in New Amsterdam and associated colonies. Since any of these places is reasonable, I did a search for Thompson and Sweden and found an interesting article on Thomas Jacobsson the Finn and  his Thompson descendants. The article follows several generations of Thomassons/Thompsons in the New Sweden Colony: http://www.colonialswedes.org/forefathers/Jacobsson.html

These are Finns who lived in Sweden and moved to the (eventually) Dutch colony of New Sweden at the junction of White Clay and Red Clay Creeks. The article was interesting in that these Thompsons sort of melt into the fabric of Pennsylvania, Delaware and the Carolinas, blending in with other Thompson families and becoming "lost".

It was an interesting possibility but really any "Thomas" who had sons from Ireland to Prussia would be a good candidate. So I put the article away.

Then I had my tests done at 23 and me. A friend who is also doing family research told me I should load my results into Gedmatch.com which is a free gedcom and genome comparison tool. All free things cost somebody something and I'm sure this is no different as the person developing it is donating all their time and effort, let alone the costs of hosting.

Anyway I did load my results and (as with 23 and me) there were no really close matches. Gedmatch.com will return results you wouldn't normally see at 23 and me. These would be the smaller and older matches that many people disregard. Among my matches, was one email address with a .SE extension and given that Danes are at the top of my Y DNA list and Sweden is not too far away, I decided to contact her.

She didn't have a whole lot to go on, because Swedes switched last names every generation, but she did have a large "Forest Finn" family that she could give me last names for. I had to look up Forest Finns. That got me thinking again about that article I had read so many months before. So I shared the Thomas Jacobbson article with her to see what she thought.  She immediately recognized a name from the article as a relative of hers, a Mulikka or Moulica. She told me that the Forest Finns often intermarried.

That led me to do a quick study of my smaller matches at Gedmatch.com and I noticed quite a few .FI email addresses.

So here's where all this could make total sense. The Forest Finns had a kind of diaspora leaving the Swedish empire and were on the move within the empire in the 1500s. My Y DNA matches are no closer than the 1500s and are "sons" appearing in Denmark and England. It's possible that a common Forest Finn ancestral family left the Baltic and headed for points west depositing DNA along the way. Also since my Thompsons may have been from Pennsylvania and these Finns disappear into Pennsylvania there is a good case for matching proximity. I also seem to be related to at least four other people with ties to the same New Sweden colony.

Here's where it all falls apart. I don't know which side of my family these Finns fall into. It is most likely a grand coincidence that my genetic matches with Swedes and Finns would turn up a Thompson. After all I am completely missing my German Schultz family after my great grandmother. There are hundreds of possibilities there for intermingling with Swedes and Finns. I also don't know the last names of many of the mothers in my tree, any one of which could provide me with Baltic DNA.

Without a clear paper trail, this is just more conjecture. It's worth following up on as one more possibility in a sea of possibilities.

Genome Analysis

A bit of a warning here. When analyzing the genome of a person like me, it helps to realize that this is not the genetic analysis of a single person but the combination of everyone I'm related to for several hundred years. Everyone..even the ones I don't know about.

23 and me provides some of this. My Y is R1b-U106 which we've discussed at length here. My MTDNA (super ancient all-mother stuff) is T2a. Wide spread in the old world but not very prolific. It appears to offer some protection against alzheimer's but comes with a downgrade in fertility.

They also provide some broad generalizations. Like the major category of your ancestry:

 

Okay so 100% Europe is not the most exciting thing. Their examples show awesomely diverse people with oranges and greens. I've got the gray box.

Here is 23 and me's Global Similarity for me:

Northern European. Okay that's probably what anyone could expect.

They also have an Ancestry finder, which is based on the responses of other people to a general ancestry survey. These results represent the people who are related to me who have all four grandparents from the same country. I made the amount of DNA necessary smaller to get the most results.

If I leave the settings at the default it's UK, Ireland, Germany and France..which you could pretty well get from my tree.

There are people who will also run analysis of your results. I contacted Doug McDonald and he was kind enough to run an analysis of my genes. Here is what he came up with. 

Mostly Europe a little Middle East, some Africa and a tiny splotch of America. Doug said the African is real but probably beyond the genealogical timeframe.

This next graphic is interesting. It's a scatter plot of nations and types of people. It doesn't represent my nationality, just my similarity to other nationalities. I am most like:

The melty spot between France and England appears to be me. Keeping in mind that this represents the total of my parts and also that the data for Germany is missing. Germany won't give away it's genome apparently. 

Here is the map version of a similar represntation:

The green dot is me. Right there in the Channel. I asked for some explanation of what I was seeing and Doug said it was English or British with significant input from the continent. That sounds about right.

The final bit of analysis I did was with Dodecad Oracle which does a different analysis along the same lines as Dr. McDonald's analysis. Here are those results:

DodecadOracle(c(11.7,49.0,26.9,0.2,8.6,1.2,0.6,0,0,1.4,0.4,0),k=30)

      [,1]               [,2]    
 [1,] "CEU"              "5.3498"
 [2,] "N._European"      "6.8162"
 [3,] "Argyll_1KG"       "7.5783"
 [4,] "Orcadian"         "7.7582"
 [5,] "Orkney_1KG"       "8.1615"
 [6,] "German_D"         "8.3337"
 [7,] "French"           "11.0648"
 [8,] "French_D"         "11.1261"
 [9,] "Mixed_Germanic_D" "13.625"
[10,] "Slovenian"        "14.5148"
[11,] "Dutch_D"          "14.8358"
[12,] "Kent_1KG"         "17.2383"
[13,] "Hungarians"       "19.0526"
[14,] "British_Isles_D"  "19.3073"
[15,] "British_D"        "19.3129"
[16,] "Cornwall_1KG"     "19.968"
[17,] "Swedish_D"        "22.1075"
[18,] "Irish_D"          "22.6407"
[19,] "Norwegian_D"      "23.3146"
[20,] "French_Basque"    "23.3369"
[21,] "Portuguese_D"     "23.5818"
[22,] "N_Italian_D"      "24.7998"
[23,] "Spaniards"        "24.8212"
[24,] "Tuscan_H"         "25.3811"
[25,] "TSI"              "25.5292"
[26,] "Tuscan_X"         "25.6874"
[27,] "IBS"              "25.9553"
[28,] "Spanish_D"        "25.9719"
[29,] "FIN"              "28.684"
[30,] "North_Italian"    "29.1846"

Of course my first question is: What is CEU? it turns out it's white people from Utah. The listing runs from group I'm most like to least. The way I read this, I'm most like white people from Utah..fair enough. Next I'm Northern European..check. Next Argyll, which I'm going to guess is people from Argyll Scotland. Argyll is a bit of a melting pot of Scots, Irish and Norse. That's probably close enough. Orcadians and Orkney...Orkney is also a melting pot of Scots and Norse, then on down to German and French etc..All reasonable and pretty much in line with what Doug McDonald has to say about me. 

That is me and several thousand of my ancestors in a nutshell.

Giving it all I've got

Genetically anyway.

I've been using my 23 and me account along with the vast sources of the internet to contact and confer with genetic relatives. It's been awesome and eye opening.

I was spoiled by success right away. My first contact through 23 and me was related to the Hathaways and I was able to triangulate with another genetic relative who matched me in the same spot on the same chromosome and also was related to the Hathaways. So that part of my genome is "Hathaway".

Here is the gentle reminder I get from 23 and me almost daily..."not always so". The things I think I know are not always the way things are. Genes care little for nice clean family trees or social rules or boundaries and borders between political states.

I am also reminded that, although I'm a Thompson, I'm also a Hathaway..as much as a person named Hathaway today is. I'm a Williamson and a Paull and a Dartt. I'm a Clarke and a Hutchinson and I'm a Sutherland.

So far, most of my genetic relatives are fairly distant. Falling into the 1700s. None much closer than that. Everyone so far at 23 and me shares less than 1% of my genome, yet I've been able to confirm that my Seelye tree is fairly accurate and so are my Hutchinsons. So..it would appear that my parents are my parents and that my grandparents are probably my grandparents...Of course there are no Thompson matches and no matches with any family related to them.

I could despair about that but before we go there I need to point out that my genetic matching has only been as good as my family tree. Genetic matches, for me anyway, are not often direct and they are hundreds of years in the past. So if you and I don't have a tree going back at least to the 1700s, we are unlikely to find each other. Now we can get an idea of why I don't have any Thompson or Williamson matches...I wouldn't know one if I was staring right at it.

There are tantalizing clues though. I seem to be related to a lot of families from North and South Carolina. I'm also related to a cadre of Pennsylvanians. I've got people from Scotland and Ireland as well as Canada. I'm also oddly related to Ukrainians, Latvians and Lithuanians and several German families in Russia. Somewhere in the more remote past I'm related to Swedes, Finns and Danes and some segment of Ashkenazi Jews. The most remote match has been a man from Morocco.

Whatever ideas I have about my family and it's identity, they have proven to be just that. Ideas.