Matrilineal Monday: Who the Hell is Patsy?

I've written four gigantic DNA-related genealogy posts recently, and not one has crawled out of the draft box. (Nor have any of their off-topic asides, so there goes my public chuffery at having Steve Martin reply to me on Twitter.) As I start typing, I can't help but wonder if this one will end up published or if I'll just rant myself into a mood and burn out at 2000 words, still unsatisfied. 

I'm going to use a grumpy genetic genealogy moment from yesterday to fit one of Geneablogger's daily prompts: Matrilineal Monday.

As discussed at excruciating length in play-by-play detail (but again, not posted, so I'll try not to sink back into it), many people who have taken autosomal DNA tests don't know what the hell they're doing. Ignorance is understandable when you're new, but even in long-time customers I'm seeing so many people who don't understand how to use the test to find their missing ancestors.

One popular variation is the "I don't see any surnames in your profile that I recognize, so bye." (DOOR SLAMS SHUT.) It doesn't matter that we share DNA and are almost definitely cousins. It doesn't matter that those unfamiliar surnames are, quite possibly, unfamiliar because they belong to missing branches in the person's family tree that said person wants desperately to know about. Nope, said person just walks away from the match. Delayed gratification is, apparently, for pussies.

Other than craving a pie chart of their ethnicity (which so many people misread), I don't even know why those people test. I've seen this phenomenon everywhere, too: Ancestry, 23andMe, and FTDNA on Mike's account (my results should be there soon). It's a baffler.

Yesterday a very nice man replied to my 23andMe request to share genomes. (Because, unlike at FTDNA, you have to beg each and every match to let you see which segment you have in common. That is a slam against 23andMe, but don't worry - I have slams for each company. Alas, that blog post never made it past my howls of frustration. Just please trust that I'm not in FTDNA's or anyone's pocket.) It's not like we were sharing health information. I just wanted to know on which chromosome we matched, and what the start and end points of the segment were.

The man denied my request, all while pleasantly chatting about my upcoming move to Australia (it's in my profile), and it was clear that he made a sincere attempt to look through my tree for names in common. Like I said, nice guy.

But... he's also a guy with five surnames in his profile. Five. Hundreds of possible surnames within the autosomal timeframe of 10 generations, and he's dismissing me because I don't have the five he knows (or focuses on).

Also? The man's a Brit, descended from Brits. Me, I'm nearly totally colonial. I have relatively few lines that left the United States in the past 300 years. That narrows things quite down a bit.

If he would just allow sharing so I could see the segment we match, then I could label that segment as "probably English." It's something. I've run with less.

Later, if I ever identified through triangulation which ancestor passed along that segment to me, then I could write this guy back and tell him something new about his family. I'd like that. I love sharing information. (You've seen how much I type here. And no, I'm not scaring people like this gentleman off with overly wordy messages... mostly because there's a 500-character limit. Ha!)

As a parting remark/justifcation, the man said something bizarre. "Besides, I think H1 covers half of Europe."

What he was referring to is my maternal haplogroup. Just to refresh the poor soul reading this at 2 a.m. on a frozen computer screen after Googling her way here by mistake (I can think of no other excuse to still be reading), three DNA tests are the go-tos for genealogists:

  1. yDNA test: for males only, testing the direct paternal line (dad's dad's dad, etc.)
  2. mtDNA test: for males and females, testing the direct maternal line (mom's mom's mom, etc.)
  3. autosomal test: for males and females and the current HOTNESS in genealogy, letting you see chromosome snippets you share with distant (or not-so) cousins, helping you work back to a common ancestor

I haven't had a proper mtDNA test (aka a "full sequence," which currently will show you other maternal line descendants within 16ish generations), but 23andMe does a minimal mtDNA test that lets you know your haplogroup, aka the folks your mom's mom's mom's mom's (and so on)'s people were running with (tens of) thousands of years ago.

So, my maternal group is H1, and like all the H-groups, it is indeed one of the most common mtDNA haplogroups in Europe.

But, it's also just one line from a long, long time ago. To dismiss finding a common ancestor among many recent lines because of this single maternal line is incredibly misguided and silly.

Fine, we don't match on the maternal line. How about, oh, the other thousand possible surnames within 10 generations? Each with their own lines leading to haplogroups that could be anything?



A sensible person would've ended the post there. I did my bitchin', and I discussed a matrilineal topic. Done and done.

But we all know this blog is fueled by a need to keep my typing fingers out of trouble, so let's describe my matrilineal line and ask some Google-bait questions that may just lead to answers.

Ok! My Girl's Club Roster runs as follows:

There's me, Shari (duh)

daughter of

Sandra (Sandy to her friends, Alessandra in her eBay life)

daughter of 

Frances (who would smack me across the face if I told you her actual first name, even though I think it's pretty)

daughter of

Nora (that's "Nanny" to me, and bless her for asking to take a photo of all four generations of us together)

daughter of 

Delie ("Grandma Trimble" to my mother and grandmother, "Sister Trimble" to those she nursed, "a Langwell" to those who nodded knowingly, and "Ida Adelia" on her headstone)

daughter of

Mary Jane

Here's where we slip just past living memory. Did anyone ever think to ask Nanny (or Grandma Trimble, who lived until 1956) about her grandmother? Probably. Did whatever she say survive to the present day? Seems not. Not even Nanny's younger sister, alive until 2003 with a keen genealogist for a daughter, recalled anything about her maternal grandmother.

Even if the line so far hadn't been known through personal experience, plenty of birth and death certificates back it up. (You have to love the way FamilySearch has images of all Texas death certificates through 1976 - and some into the 1980s - available for free.)

We know that Mary Jane is Grandma's mother not just because they're together in the 1880 census, but because her son said so on Delie's death certificate. Of course, informants are sometimes a bit off or completely wrong, but usually they're right, plus Grandpa Trimble was still alive to consult.

P.S. If someone shows you a photo of Thomas and Mary Jane Langwell, that's not them. I share genomes with the person who knew the photo's owner; it's a different pair of ancestors (not mine), and apparently some copy 'n paste clickologist got twisted things up along the way. Alas, the only known photo we have of Mary Jane is her cemetery marker.

Carrying on: Mary Jane was the

daughter of


Specifically Malinda Roberts, née Smith. When playing the board game of genealogy, drawing a new "Smith" line should send you back 11 spaces and require that you pay $50 to every other player. If they married a Jones, you go straight to Genealogy Jail and weep for three turns.

(Three-minute intermission while I ponder the niche marketability of a genealogy board game.)

Death certificates in Oklahoma started in 1908 but weren't reliably filed until the 1930s. I asked for MJ. Oklahoma shook its panhandle-shaped head. So, we can't look at Mary Jane's death certificate to see her mother's name.

The Smith and Roberts lines intermarried quite a bit, and their descendants hold reunions to this day in Arkansas. They (in their oft-repeated, no-patient-zero, but also no-genealogy-agenda way) say that Mary Jane who married Thomas Langwell was a daughter William and Malinda Roberts. I believe them for the following reasons:

  1. In the 1880 census, Thomas and Mary Jane lived next to/near John C. Roberts, John W. Roberts, Jasper M. Roberts, Samuel J. Roberts, Andrew J. Roberts, John Slape (married to Louiza Katherine Roberts), Layton Monroe Roberts, and Thomas Heathcoat (married to Anna Roberts) - all sons, nephews, or sons-in-law to William and Malinda.
  2. The Mary J. in the 1880 census was born circa 1846 in Tennessee, parents born in Tennessee. In the 1900 census she was said to be born in October 1843 (doesn't match her November 1845 stone) in Tennessee, parents born in Tennessee. In the 1910 census, she was born circa 1844 in Tennessee, parents born in Tennessee.

    Guess who was born in Tennessee and had a daughter named Mary Jane born circa 1845 in Tennessee, according to the 1850 census and the 1860 census? William and Malinda.

That's a lot of quacking. This judge rules in favour of the duck.

Malinda almost made it to the 1880 census, which would've been ever so nice, since it was the first census to ask people about their parents' birthplaces. (Note that recent censuses do not. ANGRY BEAR WAVES PEDIGREE CHARTS AND GROWLS.)

Actually, she still made it into a supplementary document: the 1880 Mortality Schedule. In 1880, census takers also asked questions about who in the household had died in the previous 12 months. The schedule notes that in Boone County, Arkansas, "Malindie Roberts" passed away at age 64 in February 1880 from typhoid fever. She had lived in the county 20 years, was born in Tennessee, and her parents were born in Tennessee.

Wow. I sure do love that 1880 Mortality Schedule!

Unfortunately, "knowing" that her parents were born (or said by someone else - we don't know if it was kin or a neighbor or what - to be born) in Tennessee doesn't help as much as we would like. Tennessee only became a state 16 years before Malinda came along. So, while we can make logical guesses where to look for official records of her parents' origins (the Carolinas, Virginia, Georgia...), we don't know really know from this document where they were born.

However, modern genealogists say that Malinda's the daughter of



I have never-ever-ever-ever seen a proper source for this. It wouldn't have to be an official document; a near-contemporary account would do.

Check every online genealogy, and you'll see people citing other people. Check print genealogies, and no one cites anyone.

In the Summer 1986 issue of Ansearchin' News (scanned and visible for free at, on page 60 (image 14 of the PDF), a transcription of records from the family Bible of John C. Roberts, Malinda's double-nephew, is supplied. (Malinda's sister married Malinda's husband's brother. John is also the one living right next door to Mary Jane and Thomas in that 1880 census.)

However, a bit of background on the family is also supplied in that AN article, and it's not clear which items - if any - of the lead-up to the transcriptions are based on information from the family Bible.

What it does say, though, is this: "It is believed five children of William and Mary Roberts married children of John and Jane Smith."

Jane? What happened to Patsy?

Also, we don't know who believed it, or why. You won't find John and Jane on any online genealogy today. Everyone is convinced that it's John and Patsy. Again, I don't know why.

Maybe if I worked harder and reinvented the wheel, I'd find out. Or I could just go do the 1,240,133 things that are more pressing, or less expensive, or more rewarding, or more needed, or something, and simply curse the ^&%! jerkwads who don't cite their sources. (Their real sources. "A tree I saw on" or "an uncle who has this line back to the 1400s, trust me" is not a source.)

Oh well. I know nobody owes me anything. Still, it would be nice if people in a research-based hobby were more diligent and generous with their research. Yes, it super-sucks when you work hard to add to the scholarship and credit gets stripped away and the data twisted, but wouldn't you rather help all the nice people than hoard the info away from the poopieheads? Don't you wish someone had handed you a few breaks instead of making you reinvent the wheel? (And, odds are, didn't someone do that at some point?)

I can see the case for Patsy. There was a Patsy (Haggard) who married John Smith, and they lived in Tennessee. Malinda's sister named a daughter Patsy.

Yeah, that's all I've got.

And that's not too good.

Now, I'm not saying it's not Patsy. In fact, I have a few DNA matches at that descend from Patsy's father, theoretical grandfather, and theoretical great-uncle. If I had to bet, and splitting the bet wasn't allowed, I'd probably bet on Patsy. Probably.

(Since it's, which unlike FTDNA and 23andMe still doesn't allow people a way to compare segments, we have no way of determining if our DNA connections are the same as our paper trail connections. It's maddening. I'm usually a fan of, but this set-up is a huge mess that promotes sloppy genetic genealogy and very dangerous assumptions.)

Anyway, who was Jane? Where did that come from? Is she Patsy Jane? I've seen some people refer to Malinda's mother as "Martha Patsy" - have they seen something, or are they just assuming that Patsy was a nickname for Martha, as was the norm then?

(If so, why not just say "Martha" with Patsy labeled as a nickname? Why "Martha Patsy"? Would you believe - of course you would - that one of those recent not-published blog posts was all about how I hate when genealogists write "Mary Polly"? Polly is a nickname for Mary! It's not her bloody middle name! I even linked to a Monkees video and mentioned Mary Poppins. PASS THE CHOCOLATE; LIFE IS HARD.)

"Matrilineal Monday: Who the Hell is Patsy?" Well, as far as I know, it's all down to DNA at this point. All I need is for people to believe in science, not in reaffirming what they already know.

(Maybe not a board game. Maybe something more like Oregon Trail. Man up and ford the symbolic river, people!)

P.S. Yes, THE Steve Martin. (That I squee'd a little over a celebrity reply on Twitter only accents my crow's feet.) I bet he wouldn't hold back his genomes or his sources.

29 January 2013 |






Carnival Elation (2009)
Carnival Splendor (2009)
Carnival Spirit (2010)
Carnival Spirit (2011)
Carnival Splendor (2011)
Norwegian Pearl to Alaska (2012)