Author Topic: My g-g-grandfather had kids with three different women - DNA strategy?  (Read 961 times)

Offline ali.aitch

  • RootsChat Extra
  • **
  • Posts: 18
    • View Profile
I am trying to find my g-g-grandmother Mary Johnson using DNA, but my g-g-grandfather had children by three different women.
My g-g-grandfather John Horlock (b1824, Horsley GLS) married Sarah Hammont and had two sons with her in Chelsea (1851 census.) He was next discovered living with my g-g-grandmother Mary Johnson in Islington (1861 census) just before the birth of my g-grandfather Tom Sansome Horlock. Mary was calling herself Mary Horlock, the wife of John, aged 23 and born in Somerset. (They didn't marry, I assume Sarah was still alive. John and Sarah's two sons were recorded on census night 1861 as staying at a homeless shelter in Saffron Hill, Clerkenwell.) They then had a daughter together, Emily Jane. Tom and Emily's birth certificates both give Mary's maiden name as Johnson, as does the birth certificate of a child who died in infancy. By the census of 1871 John was living with another woman called Mary with two additional children, Martha and Kate. They were both described as his daughters, with new-Mary calling herself John's wife, aged 38, also born in Somerset. Martha and Kate were not registered in the name Horlock; I assume they were the children of the man new-Mary was living with before, and were registered in his surname. When John and new-Mary had yet another two children, William and Mary, their birth certificates give Mary's maiden name as Poole. Martha died in childhood and Kate took up with one Henry Sessions, who spent so much of his life in prison that they did not have children.
I have taken a DNA test with Ancestry and have a sprawling tree there. Of all my matches, I can find only one who is a certain, documented descendant of Emily Jane. I'll call him Alf. When I look at the DNA matches which Alf and I have in common, those I am able to build a tree for all turn out to be descendants of Mary Poole through William and Mary, her children with John. (There are a couple of our shared matches who have no tree or clues, so I can't trace their ancestors.)
I can find plenty of matches who trace back to Tom Sansome but they too, of course, have John's DNA as well. I need to find a match who is decended from a relative of Mary Johnson, or from a child she had with another man, in order to eliminate John's DNA and find out who Mary was. (Is my logic correct here? Looking at DNA matches gets me back, at best, to people who are descended from John and any one of three or more women.)
I have built trees for all the Mary Johnsons born around 1838 in Somerset from entries I can find on the 1841 and 1851 censuses. They all died too soon, married elsewhere or emigrated. I have messaged all the no-tree people but got no joy there. I have access to the DNA matches of my sister, my cousin and a niece on Ancestry (all descended though Tom Sansome.) I have transferred my DNA results to MyHeritage, familytreeDNA and GEDmatch but am unsure how to use their tools and they can't rival Ancestry for the huge number of trees and DNA test takers. I've tried to find out who Mary Poole was on the off-chance that she and Mary Johnson knew each other back in Somerset - she, too, has eluded me.
I have been looking for Mary Johnson since the arrival of the internet. I thought that DNA would crack the case, but now I'm struggling with the DNA side of things. Can anyone suggest a way forward, please? On Ancestry.co.uk I am ali_aitch1 and the tree is called Seeking Mary Johnson. Thanks so much for reading. Alison x
Horlock Gloucester/London
Applegate Norfolk/London
Sparrow London
Jaye London
Spillman London
Wood London

Offline Cas (stallc)

  • RootsChat Marquessate
  • *******
  • Posts: 6,992
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #1 on: Saturday 18 May 24 20:20 BST (UK) »
I am no DNA sleuth but will try and give a few thoughts on how I handle my matches.

I now and again do a search on my DNA page for the Surname (middle box) I am researching at that time. I will do just a surname and try with and without place (right box). Looking into the peeps that pop up, they are all DNA connections, that may or may not be what line I am researching.

This can bring up the smaller matches as shared DNA with matches does not show under 20cms.

I colour code family, which may have more than one colour if they overlap. Some have a few colours so I know where the family connection is. I can then bring up the connections by colour.

I would try with Tom, give the match a colour for John...if any other matches do not have any shared matches with him give a colour for Mary and try and build that way.

Hope this makes sense, am sure others with more knowledge will add to help

Cas
Census information is Crown Copyright www.nationalarchives.gov.uk

Squire/Thomas/Williams/Bowen/Lewis/Davies/Jones/Rees/Morgan/Lloyd - Glamorgan
Lewis/Davies - Breckonshire
Davies/Roderick - Myddfai Carms
Thackwell/Thomas - Hereford/Monmouthshire
Shoemac/Squire/Keirle/Small - Somerset
Berry/Baggot/Lee/Clayton - Lancs
Yelland/Bray/Trethewey - Cornwall
Baggot/Hurley/Keaveny/Shiel/Flynn - Ireland

Offline ali.aitch

  • RootsChat Extra
  • **
  • Posts: 18
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #2 on: Sunday 19 May 24 02:29 BST (UK) »
Thank you for this, Cas. I'm not sure I really understand what you mean by "try with Tom" - I know I can search my *matches* for a surname (in this case Horlock) but there doesn't seem to be an option to search with both first name AND surname. I have found quite a lot of *trees* with Tom in, but they all either go no further back than Mary Johnson, or they incorrectly give Mary Poole or Sarah Hammont as his mother, or they nominate a Mary Johnson who I have eliminated by building their trees. I think I have to sleep on it, I might get what you are saying after a good night's kip. (by the way, love your peeps who pop up - was it Harry Enfield's Greek bloke in Saturday Night Live?)
Horlock Gloucester/London
Applegate Norfolk/London
Sparrow London
Jaye London
Spillman London
Wood London

Offline Cas (stallc)

  • RootsChat Marquessate
  • *******
  • Posts: 6,992
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #3 on: Sunday 19 May 24 05:49 BST (UK) »
I am not sure I am the person to try and help with DNA, am still on a learning curve myself. There are tools such as DNA painter etc that others can advise you on, just put a post up to ask.

I find it a help sometimes to narrow by searching maternal or paternal lines. Ancestry has this by parent 1 or parent 2.

What I think I meant to say was start with the matches you have for Tom. The DNA going back from him would include both paternal/maternal sides if people have tested.

Colour code say blue for known matches to father John. Any matches that do not have any shared matches with John, colour code say red for mother Mary. Any that you are unsure of which side they belong, colour code say yellow. I have found this useful to check all matches against each other. I also right myself notes that state matches with so and so - or possible? (I write a given surname) Notes to self can help not having to go over same details if there are a number of matches. (Just saying It works for me)

Each DNA match connects to a shared set of grandparents or grandparent going back, the higher the Cms the closer the match usually.  I would not worry about trees unless there is a DNA connection, many have lots of errors.

I know who Harry Enfield is, but can't recall watching him...not sure were the word peeps came from, but I use it now and again..never really thought about it lol.

Sorry not to have been more help, sometimes DNA can be waiting game until you get a match and it starts to fall into place.

Good luck & happy hunting :)

Cas
Census information is Crown Copyright www.nationalarchives.gov.uk

Squire/Thomas/Williams/Bowen/Lewis/Davies/Jones/Rees/Morgan/Lloyd - Glamorgan
Lewis/Davies - Breckonshire
Davies/Roderick - Myddfai Carms
Thackwell/Thomas - Hereford/Monmouthshire
Shoemac/Squire/Keirle/Small - Somerset
Berry/Baggot/Lee/Clayton - Lancs
Yelland/Bray/Trethewey - Cornwall
Baggot/Hurley/Keaveny/Shiel/Flynn - Ireland


Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,490
  • Computer incompetent but stiil trying
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #4 on: Sunday 19 May 24 06:32 BST (UK) »
I have a tip for the no tree no contact matches

Also using colour code . The highest unknown match .I allocated a colour and then use that colour for all their shared matches with you . Some of the smaller shared matches may have  trees or be from a particular region or ethnicity

I've just colour coded green for Irish ethnicity and red for Scottish .for someone I'm helping
  don't know which side their ethnicity is from but  sometimes it helps .

Having access to cousins &  sisters DNA also helps . check which of the smaller shared matches they match too.

I always check The younger generation's matches too. Sometimes a no match can be an indication of a more distant relative .

Good luck
Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,490
  • Computer incompetent but stiil trying
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #5 on: Sunday 19 May 24 06:37 BST (UK) »
Btw do you have Sarah HAMMONTS. Parents and siblings & their spouses  can you look at trees with them on and see if any owners indicate they have done DNA.

Useful for elimination . Also if they are an older generation
You could message and ask if they match any Horlock descendants of the first union .
Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,490
  • Computer incompetent but stiil trying
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #6 on: Sunday 19 May 24 06:49 BST (UK) »
Another little tip I've picked up for linked but private trees

Look at images that have been shared on a tree
Photo of church where children baptised . Snip of a certificate.
Etc and check in gallery  details which other members have copied it especially who submitted it . Then look at their trees ...the answer could be in a well researched but non DNA tested member .
So for example a photo of Mary Poole's descendant shared by 6 people one of them may have  correct details for Mary JOHNSON
You can add a comment to the image

My grandfather came from a recomposed marriage both had children with previous spouses and descendants from both sides have tested so I'm used to DNA for half cousins 3 xremoved etc . Grandads elder half siblings were 20 years older than him ..

I find thru line suggestions can be useful too
Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline brigidmac

  • RootsChat Marquessate
  • *******
  • Posts: 6,490
  • Computer incompetent but stiil trying
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #7 on: Sunday 19 May 24 09:14 BST (UK) »
Regarding the 1871 census it's interesting that youngest of John's children Emily is listed as 19 months but baby Kate is listed as 1 year
Although all children are listed as Horslock I think it was convention to put step children after the others so Martha age 4 and Kate 1 . Were likely to be hers .even if they had had descendants they wouldn't match you

Why is Tom recorded as John on 1871

Very nice tree by the way . I like the way you have included the wrong Mary with explanation so relatives can see if they compare trees
Roberts,Fellman.Macdermid smith jones,Bloch,Irvine,Hallis Stevenson

Offline ali.aitch

  • RootsChat Extra
  • **
  • Posts: 18
    • View Profile
Re: My g-g-grandfather had kids with three different women - DNA strategy?
« Reply #8 on: Sunday 19 May 24 11:56 BST (UK) »
Thanks again to CAL for that further explanation, I get what you mean now. That would work like a dream if only I had Tom's DNA and he subsequently had his own list of matches. As it is, I only have my own matches (or of my sister/cousin/niece.) Where I can build trees for those matches who are descended somehow from John Horlock, there is only one (Alf) who descends through Emily Jane (so I know they have Mary Johnson DNA) several who descend through Mary Poole (wrong woman) and several who descend though Mary Johnson but they all have no clue who she was. Catch 22. You are almost certainly right, it looks like it could just be a waiting game. Trouble is, I'm in my 70s now and waiting is not the option I want to accept. Upside, though - it's a huge incentive to keep me going till I become the oldest woman in the world, and the perfect excuse not to do any housework.
Horlock Gloucester/London
Applegate Norfolk/London
Sparrow London
Jaye London
Spillman London
Wood London