James Lewis Armstrong, Jr. - :Lineage from John Lewis Armstrong to William (Christie's Will) Armstron was messed by a member.

Not a line I have worked much on. Made a Discussion in 2018 asking for help but have pretty much left it on the back burner. Bird is difficult. In this timeframe, many surnames are difficult lol. I am glad to see the work done on the Armstrong families and happy to add what I can but it is not much as it is beyond the fifth great grandparents distance for me and I work with autosomal DNA unless I have access to Y or mitochondrial which this is neither for me. I know that many have worked on ancient lines using DNA and comparing at a lower threshold can be very fruitful for known segments/shared SNPs.
We rely on the paper trail and the skill of genealogists, pro and amateur/hobbyists, to put the pieces in place when the records are scant or missing. I think most know this is my mode. At this time, I have looked to discern why these names were in my tree and am working from the bottom up for my Emersons. I am seeing an in law relationship with some on my father’s side but have shared that as well. I do believe Armstrong is a direct/ancestral grandparent relationship for me but I am not an Armstrong researcher and the revision tabs will reflect that I have not touched most of these and “mine” only minimally.
Let me know if there is something I should be doing other than what I am presently focused on which is trying to fill in and make sense of the confusion between two brothers (Henry and Isaac)

Private Thanks, it was your post that motivated me to look for the original manuscript online, after all. Sorry I neglected to remember and make honorable mention of your part in this collaboration. (Thanks to Erica for recently referring to your post.)

In reply to Erica Howton at 10:53 PM: Could you give me a page reference in the Chronicles? I can't work out which James you are referring to.

(1) Clogher is in County Tyrone. It's about 5 miles from Brooksborough. Distances are small in that part of the world.
(2) Cynthia isn't going through James of Orange County. Totally different path, and one that hasn't yet come under challenge. She just happens to have a very similar DNA overlap with the match that does go through James of Orange County.
(3) No, I agree that indentured arrival is indeed incompatible with Brooksboro origin. Yes, I'm throwing that out. The hypothesis is that there were two different James Armstrongs - one who was an indentured servant who paid off his debt in Delaware, married Mollie Bird, and lived and died there. The other arrived the standard way originally to Chester County PA, married there, had a family, and then went to SC. We have no idea who the second James married. And there is no proof whatsoever that these are the same James, at least as far as I can tell.

In other news, as asked, I researched the mystery people in the tree of the DNA match whose Armstrong origins are under attack. The ancestor I believe you wanted researched is this one:

Ambrozel McMahon

Her family is, as I feared, something of a mess. I did figure out what happened here, I believe. Margaret Russell, daughter of Joseph Russell and Mary Cox, married a wealthy North Carolinian from Carrabus County who told her, incorrectly, that he wasn't married. Three years later he returned home. He left no trace of himself in either Kentucky or Tennessee but fathered two children. Margaret Russell apparently realized there was a problem at some point and married (this time for real) to Samuel Shaw, in Washington County TN. Eventually they settled in Hamblen County, a few miles from where her birth family settled. She'd not seen anything wrong here because her father, Joseph, was often away for years (didn't show on the 1850 census with his wife and children, but died after 1880 in Texas).

Anyhow, Lawson Frye was of solidly Germanic descent - no Armstrongs there. For Margaret Russell, I'm stuck with ancestry for her parents Joseph Russell and Mary Cox. They came from Orange County, North Carolina, and Mary was Joseph's second marriage, but nothing further have I found. Frankly, neither Cox nor Russell sounds like it is likely of Irish descent either, but I can't prove it.

So in science we cannot necessarily prove things, but we CAN disprove them. Ways to disprove my hypothesis as stated above:

(1) Mollie Bird died in Orange County, SC. If that can be shown, hypothesis disproved.
(2) There was no James Armstrong that died in Delaware or wherever Mollie wound up.
(3) Mollie's father died in Orange County, SC.
(4) James's children or grandchildren were clearly named after people in Mollie's family, e.g. "Daughter Bird Blah" or after her father or mother.

Are any of these true? If they are, I'll abandon the idea that James of Orange County never married Mollie Bird. But I still need an Armstrong link for my DNA match and that seems like a challenge from what I see, so the Irvins would be the only way possible. Are you ready for that?

'Two daughters married men with Welsh surnames - Lloyd & Allen.'

Allen is in the Clogher records. And there were Lloyds too. See page 171:

https://www.ancestry.com/discoveryui-content/view/100507:62077?tid=...

For Allen, there's a page and a half for them, starting on page 10.

And, no Russells in the Clogher book at all. But there are early Coxes - two of them, from the 1720s. They married Hamiltons and Strongs.

FWIW, I don't find any "Mebanes" in the Clogher records. Either they were Presbyterian exclusively, or they weren't Irish, or it was spelled differently. I tend to think they were Scots but not certain.

'There’s a grand daughter Mary Tinnin In James’ 1794 Will. So she’s must be related to Alexander Mebane’s mother, Mary Mebane'

GENI has Mary Tinnin already. She was Mary Irvin before she remarried.

She's a daughter of William.

Here's Mary Tinnin. Second husband Alexander Tinnin. Mary ‘Polly’ Erwin

Now, back to the Paxtang Armstrongs.

First, we should not make the assumption they're one family. They could well be members of several related families. Have a look at the About section for this James, who is the son of John of Brookesboro that we'd earlier discussed. His elder brother William had an interesting chunk in the "about" section, which referred to James: James Armstrong .

Does this sound at all like this guy? James ‘of Milford’ Armstrong, Ill

So I don't see a son William or a daughter Jeannette or a brother William, not in Paxtang, but he does have a wife who went by "Janey". Mistaken family affiliation, perhaps, in Pennsylvania? Might cut down on the size of that family...

FWIW, the dates he has are set by two considerations: (1) having adult or near-adult children with him in 1732, and (2) making sure Irish naming conventions aren't violated.

(Elder brother William, by the way, went to South Carolina eventually, according to his profile notes. They are sourced.)

Well.
The FTDNA spreadsheet doesn't help much because the James we are interested in isn't an ancestor of anyone tested via Y-DNA at FamilySearch, near as I can tell.

But this is a remarkably consistent group, unlike (say) the Taylors. You can tell by looking at the columns to the right of "Haplotype". Variation there is tiny. Many of the haplotypes given must be subclades of R-M269.

So if by chance we run into a test done by a male descendant of James, perhaps then we can finally convince Erica that autosomal DNA matters too. ;-)

The marriage record exists for a Thomas Armstrong and Jane Duvall dated 22 Jun 1797 Franklin County VA. Not her?

'If you can create a study report and get sign off from a noted genetic genealogist'

This requires some definition. Who is noted, and who isn't? Aren't you basically insisting that money changes hands? That's a non-starter for 99.9% of people and you folks are undoubtably aware of it.

My daughter has a PhD and is in internship for a medical program in pediatric genetics. Does she count? Her PhD thesis required lots of genome work. She knows genetics quite thoroughly, and publishes papers readily. No? Then who? You see the problem. You have to define the authority before you can use that excuse.

If it's mainly the write-up you need, and signoff from a geneticist, THAT I can arrange.

This may not be the forum for discussion of what you can learn from AT DNA and what you can't. But I can state a simple rule, and this seems like what you are referring to:

- If you want to prove that two people are related, and they are more than five generations apart, AT DNA is hit or miss.

This is very different from the kind of analysis I do for GENI relationships. In that case, I *start* with a known overlap in a specific place in the genome. So it isn't a question of proving that people are related. It's a question of showing HOW they are related given that we already know they are. This is absolutely NOT limited to 5 generations - it goes back theoretically indefinitely - although eventually the length of a match in the DNA just gets too short and DNA testing companies won't recognize it. The limit is about 9 generations, in my experience - I've never analyzed a chunk of DNA that came from further back than that, and that far is rare.

The analysis involves finding the list of people who share that specific overlap with me and building their trees, and then finding the commonality. It takes at least one congruence to construct a hypothesis, and at least two matching congruences to confirm the hypothesis. That still doesn't constitute 95%+ proof, but if you get three or more - that's about as good as you can ever do.

The hypothesis comes in the following form: "This chunk of DNA comes from THIS specific pair of ancestors". Obviously when you include those ancestors you also include THEIR ancestors, and so on.

Sources of error here are simple:

(1) Tree errors. If the trees are wrong, so are the conclusions.
(2) Different sets of shared ancestors. Picking among several possibilities requires more matches to analyze to settle it. This problem gets worse and worse the deeper you're trying to go, obviously.

Y-DNA and MT-DNA are much simpler to analyze but they are only useful in limited situations.

I'd be happy to help develop standards GENI can use to decide if AT matches are indicative of anything, or not. But of course my degrees are in engineering and computer science and if you need credentials in the field that's harder to arrange except through consulting.

I did create a project but I tried too hard to make it accessible to just anyone, I think. It's written to be helpful, not to be a standard of evidence.

I will first of all start over on that project, and include the same broad reasoning and technique just provided here. I will then create a derivative discussion (?) on this particular case, including all the reasoning and data, tied to this particular case. Getting that reviewed by "expert" peer reviewers that carry enough weigh with the GENI staff is where I don't know what exactly I will do but I will consider multiple options, including involving my albeit related daughter :-)

What I had thought we could work on collaboratively was something akin to a flowchart. It basically would give curators a tool to use to assess whether a DNA match had been assessed and evaluated correctly against other matches that same individual had. In other words, establishing guidelines for evaluating DNA matches. I mean, this is what tech support people use and the kinds of things they are trying to support are highly complicated. They don't need to be able to write the flowcharts themselves to use them though.