The DNA-Secret that went to the grave too early


DNA testing for family history is reveling many secrets. Many of these relate to people still living such as adoptees, but some are reveling secrets that the people involved either didn’t know or took to the grave with them. This post is about an event almost 150 year ago that I don’t believe anyone now alive knew about.

The second aspect I want to show in this post is how combining DNA testing information with traditional genealogical research (and a little bit of common sense) can get us close to the truth.

Finally this post is also about the joy of sharing research work. As many genealogists will know, our hobby can be a lonely event, only brightened by the occasional family-history question at Christmas gatherings. For this work I was lucky to share the research load with Christine, a second cousin on my Stewart side.

The Unknown Match

I’ve always assumed my ancestors were relatively “un-remarkable” people. Almost all of them were Ag-Labs (Agricultural Labourers) who came from small rural communities, where most people knew most things. It was only the turbulence of the Industrial Revolution with its change in employment options and increased travel opportunities that made them leave these villages.

But things are always a little more complicated. A couple of years ago I discovered a new second cousin, whose grand-father had been born outside of a marriage and was quickly passed on for adoption.

Even before this, in 2018, I had another unusual match. It was a lady, we’ll call her Sally, who I matched with 174 Centimorgans of my DNA. The Ancestry website suggests that this means we are somewhere between second cousins or equivalent (60% probability) and third cousins (3% probability), with there being a less than 1% change of a more distant relationship. The thing is, I had no idea who Sally was and which ancestors we shared. It was time to do some research.

Evaluating a DNA Match

As with any DNA match there are three important steps to do:

Evaluate Shared matches

This is one of my first “go-to” approaches of any DNA-related genealogical work. Your “Shared Matches” can, in most cases, immediately tell you which side of the family you are connected through. In this case it provided a wealth of information. Sally was a shared match with Christine, a cousin through my Ulster-born Stewart family and my Durham-based Armstrong family. On top of that Sally matched a whole clan of more distant relatives on the Stewart side, implying that she was related through the Stewarts.

I’ve written quite a bit about the Stewarts. They are plantation-Scots hailing from the village of Lissan in County Tyrone. The earliest known Stewarts I can trace are John Stewart (b.1811) and his wife Jane Bell (b.1817). Their story is however a little more complicated. By the time they married on the 7th November 1848 they were already widow and widower. It’s certain that John Stewart had children from his first marriage and it’s unknown if Jane Bell also had children. They came across to industrial North-East England probably sometime after the birth of their last child, Robert Stewart (my g-grandfather), in 1859. One of children of John Stewart’s first marriage was married in Shildon in 1860, but it’s unknown when exactly the whole family moved to Teesside. They moved first to the Shildon area and then to Darlington. They had certainly all arrived in the North-East by the 1871 census.

For those people unfamiliar with the name Shildon, it was the “Cradle of the Railway”. It is here on the 25th September 1825 that the Locomotive set-off as the first passenger-carrying steam-train on its trip to Stockton.

Contact your Relative

I wrote to Sally using Ancestry’s in-built messaging system. Her story is quite a complicated one. In her words:

“I have no idea my parents as far as I know my parents were divorced when I was 2 years old and I was known as [redacted] my mother’s Maiden name and was brought up by my Grandmother it was not until I was almost 18years old when I had to produce my birth certificate when I was taking a nurses exam and had to send down to Somerset house for my birth certificate that I was registered as [redacted] Natrass so really I have no idea of any of my ancestors”

Review their Family-Tree(s) and/or build your own

I’m not sure this needs any explanation. If you have a non-trivial amount of DNA that matches someone, you have some shared ancestor(s). I was expecting to find some of my Stewart ancestors in Sally’s tree. There were none, in part because the tree was a relatively small one. More importantly, since Sally knew very little about her immediate family, she had, unfortunately, built her family-tree around a similar, but unconnected, family.

Out of curiosity, I also built my own version of her tree as a private-tree on Ancestry and waited for the website to help build out the tree based on its hints algorithms. Again this generated no useful information that might connect us.

The Blind Alley

I was really hoping I could help Sally with her family history, but sadly my research at the time, couldn’t find a connection with my own ancestors. Clearly there was a Non-Paternity Event (NPE) somewhere in the family. That is to say the father was not the one recorded in any documents or was unknown/un-named.

Looking back at my research at the time there were a couple of problems that hampered my research.

  1. Sally’s ancestors came from a working-class background in North-East England, as a result they were not so well connected with the civil registrations processes. Combine this a surname like Nattrass that has multiple spellings (single/double T and S plus a middle vowel that could be a or e) and much of the traditional genealogical paper work is fairly easily obscured.
  2. The correct Civil Registration marriage record was partially illegible. Over at the FreeBMD website where there should have been an a in one surname the transcriber had used standard notation for an unclear letter [ao]. This resulted in the search failing to find any marriage records connecting Sally’s parent’s Surnames. The same record on Ancestry had mixed up the registration district “Houghton”, which is another name for Houghton-Le-Spring in County Durham with Houghton in Sussex.

A New Hope

Despite my frustration, the challenge remained unsolved until this year. Recently I spotted that a new DNA match also shared the name Nattrass which allowed me to re-start the research. Thanks to the work of my cousin we were finally able to find the correct record for Sally’s parent’s marriage and build a connection between these two DNA matches. They connected through a couple Richard Nattrass (b.1873 in Dinsdale near Darlington, Co. Durham) and Annie Smithson (b.1878 in Eldon, Co. Durham). Sally and this new match were descended from two different children of this couple, with Sally being theoretically the granddaughter of this couple. Tracing these two family-lines further back still did not reveal any Stewart ancestors.

At least this family grouping was an important stepping stone, for a couple of reasons:

The DNA Connection

There are in theory three possible generations where this unexplained relationship could have come from:

  1. Two of the children of Richard and Sarah could have been Non-Paternity Events (NPE) i.e. the father of both children was not Richard Nattrass. or:
  2. Either Richard Nattrass or Annie Smithson was the child of someone within the Stewart family. or:
  3. One of the grandparents of Richard Nattrass or Annie Smithson was a Stewart.

Ruling out the Unlikely

I mentioned earlier that the Stewart family moved from Ulster to Co. Durham some time before 1871 and possibly as early as 1860. This implies that an NPE cannot have occurred before the early 1860’s.

Given that Richard Nattrass is born in 1872 it is very unlikely that an un-named Stewart would be his grandfather. Annie Smithson was born in 1877, so it is also very unlikely that one of her parents was the child of a Stewart. (It’s actually an “edge case” – A child born in 1860 would have been 16 or 17 when they became a parent AND it would have to be a child of John Stewart’s first marriage)

The other obvious statement is that it is significantly more likely that there was one Non Paterity Event rather than two. That is to say, it is unlikely that Richard Nattrass was NOT the father of two children born after his marriage to Annie Smithson.

What Are the Odds

One of my favourite genetic genealogy tools is the “What Are the Odds” tool over at . This tool uses the concept of cumulative probabilities to highlight the most likely relationships. For example Ancestry gives me a range of possible relationships, based on the amount of DNA I share with Sally, if I add in the amount of shared DNA my cousin Christine shares with Sally then that will refine the probability of some relationships over others.

In this case I created an “intelligent” probability tree, based on the following rules:

  • Limit the hypotheses to only 2nd and 3rd Cousins. In this case this meant going back to my gg-grandfather (John Stewart).
  • Restrict the option that Sally was a 3rd cousin once removed to my great-grandfather’s half-sibling, as his full siblings were too young.
  • Ignored the scenario where one of Robert Stewart’s children was the father of both of Annie Smithson’s children. (for clarity it is shown in the tree below as the path male Stewart -> Nattrass -> child.). Robert Stewart could also NOT have been the father of both Nattrass children as his whereabouts at the time are known.
  • Use the shared-DNA data only from myself and my cousin. I had DNA information from one other, more distant, Stewart. It does not significantly change the results, so I’ve excluded this.
What are the Odds probability tree
The relative-probability tree for “Sally” generated by the “What Are The Odds” tool at

As the WATO site states the probability that Hypothesis 1 is correct is “About 13 times more likely than the next hypothesis”. In other words the most probable explanation is that Robert Stewart was the father of either Richard Nattrass or Annie Smithson. This is good DNA evidence of who is my shared ancestor with Sally AND where in the family-tree to find his child. This is not, however, overwhelming evidence. There is still a reasonable chance that one of the other Hypotheses is correct.

Back to the Paper Trail

With the information from the WATO website it’s possible to review the “classic” genealogical paper-trail to look for further information that might clarift things. The 1901 census shows Richard Nattrass as born in Darlington together with his wife Annie born in Eldon (a small village next to Shildon). This information backs up the record of their marriage in 1897. Richard and Annie can be traced back through the previous censuses of England and Wales. The earliest reference to Richard Nattrass is in the 1881 census where he is living with his parents, William Nattrass and Hannah (née Maynard), in Dinsdale (a small village just out Darlington). Annie similarly can be traced back to the 1881 census. She is recorded as living with her grandparents William Smithson and Caroline (née Leng). William and Caroline have four of their children living with them, three sons and only one daughter, Jane.

The 1881 census for William Smithson and family, including his daughter Jane (aged 20) and Grand-daughter Annie (aged 3)

The birth certificate for Annie Smithson confirms what we already suspected. Annie’s mother is Jane and there is no father recorded. Jane is recorded as being born in Q3 1860, which suggests she was probably had just turned 17 when she gave birth to Annie.

The birth certificate for Annie Smithson on 17th Oct 1877. I love the fact her address is recorded as the “Old Bit” of Eldon.

Robert Stewart married my great-grandmother Maria Singleton Armstrong on the 24th December 18877 in the Registry Office in Darlington. This was less than two months after the birth of Annie Smithson.

Jane Smithson was married twice after the birth of her daughter but never had any other children.


The most probable explanation of my relationship with Sally is that we are connected through my great-grandfather Robert Stewart and Jane Smithson via their daughter Annie Smithson. This is not an absolute proof, but it is highly probable. It’s also a reasonable simple and eloquent explanation of all the known information.

What this research cannot tell us is the circumstances that led to Annie’s birth, the relationship between Jane Smithson and my great-grandfather or even if Robert knew about his child.

I assume that the circumstances of Annie’s birth were finally lost when Jane Smithson died.

Sadly it appears that in the four years since I was first in contact with Sally she has died. At a human level this is quite frustrating. It would have been nice to pass on this information to the person most connected with this story, even more so as Sally knew so little of her family history.

My cousin has been in contact with our other DNA match through Annie and at least her part of the family now know this story. One interesting snippet of information from this part of the family, Jane Smithson was referred to by her grand-daughter as “Aunt Jane” rather than Granny. This is further anecdotal evidence to support the research. The term Aunt/Auntie is within many families a handy way to obscure many different relationships.

Final Thoughts

It’s hard not to feel a certain amount of anger at my great-grandfather. Whilst we don’t have a full understanding of the situation with Annie Smithson, I would have expected him to marry Jane Smithson. There are plenty others within my family who married shortly before or after the birth of their first child.

Writing a story like this involves a certain trade-off between telling the events and respecting people’s privacy. This is after all an event that happened that happened almost 150 years ago. Jane Smithson dies in 1936, her daughter Annie in 1956 and my great-grandfather in 1926. In addition all of Annie’s children, born around the start of the 20th century are unlikely to still be alive. Nevertheless it still feels a little awkward writing this research.  I have tried to obscure most of the people involved in this story, even though the records pertaining to their lives are easily available on-line at a number of genealogy websites. I hope I have provided a reasonable balance.

Having finished this research it acts as a good reminder that as time passes there will be more and more “Orphan” DNA results. The person tested has passed away and no one involved in managing the estate of the deceased is even aware of the issue.

The genealogy companies have no interest in reducing their DNA databases, but are also left in a difficult legal position. They cannot change the terms and conditions of how the DNA samples and results are used without consent, but there is no one to consent to or reject any changes.

For living genealogists it’s important that such samples and results must be clearly recorded as well as your wishes for the use/storage of your DNA information on your death.

Finally, I want to repeat how enjoyable it has been to share this “quest” with my cousin.  It’s rare to be able to share the enjoyment of each step in the discovery process with someone that understands the significance of each step.

This entry was posted in Autosomal DNA, Genealogy, Genetic Genealogy, Ulster and tagged , , , , , . Bookmark the permalink.

2 Responses to The DNA-Secret that went to the grave too early

  1. adinarinat says:

    Hi! I believe we may share ancestors. My grandpa also tested posted for ZZ52_1 and is of Blackhall and Ardgowan. I believe we also have an Ulster connection. My 5th Great grandfather is Charles Thomas Stuart possibly born in Ireland in the late 1700s of Scottish ancestry. I’d love to be in touch.

    • Phil Grass says:


      many thanks for getting in contact. I’ve sent an email to the address associated with your wordpress account. Hopefully you have already received that, otherwise please leave a comment here and we’ll work out a way to connect as I’d love to hear more about your Stuarts.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.