A Quick(ish??) Post about the Ancestry DNA Match change

A lot of people have written very well about this. It impacts every researcher who uses DNA matching and everyone has a different opinion. Which interests me as much as the event itself.

I’m no kind of expert. This post is about what the change means to me and the way I do my research.

So for context – if you do a DNA test with Ancestry, and you have a family tree on their site, and you link your DNA test to yourself in your tree, then Ancestry will compare your tree with the trees of your matches to see if it finds any common ancestors. It will then highlight these common ancestors so that you can compare the trees and see if you agree. If it’s correct then yes, you’ve found a cousin.

It looks like this:

The above match is female, has a linked tree with 79 people in it, shares 7cM with me which is a reasonably small match, and apparently has a common ancestor. I can then click on the tree link and check it out.

The suggestions go all the way down to myself and my match. I’ve cut off the lower ones for the sake of privacy.

I have several matches to these ancestors, one of whom is through an ancestral daughter who I’d never managed to trace. The common ancestors are definitely the same people. The paper trail verifies their movements. These are genuine matches. Two share 7 centiMorgans with me, one shares 6 centiMorgans.

Is it actually a DNA Match? I don’t know, because Ancestry doesn’t give us access to a chromosome browser. To me, it doesn’t necessarily matter. Well – it does and it doesn’t.

And this, I think, is where all the confusion and different opinions and complexity comes in.

On Family Tree DNA you do get a chromosome browser.

I picked two matches who also match each other and here’s the place that we all match. Same place on the same chromosome, that’s what makes it a real match. Of course, chromosomes are paired – there are two ‘number nine’ chromosomes, one that we got from our father, one from our mother. It takes a bit of comparing with known matches to determine which of that pair these matches are actually matching with. If they match in the same place on different chromosomes they are not a match, but if it’s the exact same one they are. Working it out can be done, and many of us enjoy that game.

If you can figure out which ancestral line a specific segment was inherited from, that’s useful. For the above ancestor John Peard and his wife Bridget Woodley, I could look for that precise segment match among the unidentified hordes who also match me, but haven’t got a tree – and I could say positively ‘This person also descends from John Peard and Bridget Woodley’. Or maybe I could say, ‘this person descends from a sibling of John Peard and Bridget Woodley’.

My example match is so far back, it takes a solid paper trail and a lot of time and a lot of matches with trees to even begin. I’ve been lucky with this one.

This is Ancestry’s ‘thrulines’ facility. It has pulled together every DNA match with those same ancestors – or with enough ancestors that some tree in the system can bridge the gap – and it shows them to me. I have twelve matches on Ancestry who also descend from John Peard and Bridget Woodley. It’s quite amazing, because I rarely stumble across Peard researchers. I’d never have found these guys were it not for DNA matching. I have another four on FtDNA who haven’t tested with Ancestry at all, including one who descends from John Peard’s brother. That’s sixteen matches back to my 3rd/4th/5th great grandfather in this family line.

The one on the far right is a genuine match but their tree is slightly wrong, so they don’t show up in the correct place.

Onto the harder ones:

The magic number in Genealogical DNA is 7 centiMorgans. Some scientists determined several years ago that a match of this length (or longer) is more likely to be genuine. If you go smaller, some will be genuine, some might not. And then you get what is called a ‘false match’.

I hate that term. It’s misleading. But that’s what everyone calls it.

A false match isn’t actually a false match. It’s not like a ‘false positive’ in a medical test. It’s not saying there’s a match when there isn’t. There really is a match.

Someone else’s chromosomes could have accidentally come together in the same sequence as yours – that can happen, and since they don’t share your ancestors at all it’s useless to spend time looking for a connection. This is the most genuine form of ‘false match’. It’s like the two boys in ‘The Prince and the Pauper – they looked identical, but they were not actually related at all. In a perfect world, there’d be some way of spotting these and filtering them out so what we could see were just the real ones.

But there are other sorts of ‘false matches’. You might match in much smaller segments from different ancestors that have ended up next to each other so it looks like a bigger match than it really is. A 4 centiMorgan segment from your father’s side next to a 5 centiMorgan segment from your mother’s looks at face value like a 9 centiMorgan match, until you check the matches in common. With the massive endogamy coming out of little villages across the world where cousins married second cousins and their children did the same, century after century, you end up with a lot of third/fourth cousins emigrating to the Americas or Australia or Canada who actually share enough DNA to look like first cousins. So we can look closely related when we aren’t.

That is, if I have an ancestor from West Limerick and you do too, we’re likely to have a very small DNA match whoever our ancestors were. And if I have an ancestor from Isle of Harris in Scotland and so do you, we’ll share DNA from there too. Just think how many people must have one ancestor from West Limerick and another from Isle of Harris. If those two segments are next to each other it looks like a different match to the one it really is.

It is still a genuine match though, really. Just way too hard to identify, and there’s always the chance that one portion is an accidental resemblance even if part is by descent. That one takes some untangling.

The next ‘false match’ is where the match has come down unchanged for so many centuries that the common ancestor is way before written records. This happens more often than anyone first imagined, and those segments are known as ‘sticky segments’. Since they have come down unchanged to so many people, you end up matching thousands of people who seem like fifth cousins but are really twenty fifth cousins. You’ll hear this multi-match chromosome region referred to as a ‘pile-up’ region.

Once again it’s a genuine match. But at this point we can’t use them. Nobody has the trees. I think in time we’ll get there, with enough people testing, enough records coming on line, enough chromosome painting. But not yet, and the process is fraught with error. So for now it’s a ‘false match’.

That’s the background.

DNA testing at Ancestry has really taken off. The site is groaning under the strain of all those testers, stuck at home now due to lockdowns, isolated and diligently exploring their match. Site response times are very slow. We often get error messages saying their backend servers are overloaded. It can’t go on this way.

So they’ve decided to remove all those 6 and 7 centiMorgan matches, because most are ‘false’. And the remaining 8+ matches will be far clearer.

So, what do I think about that?

A lot of people are utterly appalled. Even more people, I think, are happy about it. My knee jerk reaction was horror, because I hate to let anything go before I’ve had a chance to look at it for myself.

Now, I’m lucky. I come from Tasmania, a small island settled by convicts and soldiers. Every step my immediate ancestors took was recorded. Not only that but the records were deemed to be part of our heritage and were digitized and made available to us all for free. Civil registrations, convict files and newspapers. And being an island, all entry and exit was recorded too, in shipping records. We have excellent paper trails back to England, to those 7th/8th/9th great grandparents. So we – like most others – can really make use of small matches. But most people can’t. So to most people, those tiny impossible hints really are ‘false matches’.

I would prefer that it didn’t happen so soon, but they just might be right. In some respects.

I have a total of 38,047 matches on Ancestry. 410 are close matches (fourth cousin or closer). I know just how I connect to every close match that has a tree with the exception of four, and for those I at least know the region/ancestral branch. I can take a guess at many close matches without trees, but a few of them are a complete mystery.

What would I have lost had I never looked at my 6-7cM matches?

Actually, a lot!

I have identified twenty cousins through 6-7cM matches. Some were the Peard matches mentioned above. Another is the sister of my Irish Catholic ancestor Mary Woulfe of West Limerick, born in Sugar Hill at the cusp of parish records. Others are descendants of the Brown family, subjects of my last blog post.

Yes, some of my most exciting discoveries have been made through matches at this level. No, I’m not sure if they are DNA matches, but DNA matching has brought me paper-verifiable matches that I’d never have found because they moved to districts I could not have predicted.

So if you want DNA purity, then yes, this is probably the best idea. And Ancestry’s common ancestor feature only works back to about eight(?) generations, if you match at the ninth you have to put in the legwork yourself and look at trees, search for names. It’s a good idea for Ancestry, they can reduce their server overloads and keep us genies focused on more accurate data, not screwing up our trees making fake cousins fit when they don’t. I get it.

All the same, I wish I could have looked through my 30,000 or so 6-7cM matches before it happened. But if I’ve missed some amazing breakthrough, I’ll never know. So I’ll let it go.

It’s a pity, but it’s not the end.

An image showing ‘The Last Mile of England’ (Shropshire). It seemed appropriate.

A Family in Cambridgeshire in the late 1700s

This is the story of a family named Brown.

There’s a struggle most family historians will know. You spend years poring over records, identifying the ancestor, working on that jump back to the previous generation. Hoping that the unknown parents will turn out to be Winterfink or Albercoyle, something searchable that will stand out in lists and newspaper articles and interment records. Names like Brown or Smith can doom an ancestor to permanent obscurity.

My ancestors were John and Mary Brown, married 1760 in Fordham, Cambridgeshire.

As it happens they were more visible than expected, partly due to their grandson Robert born 1818. He was transported to Australia as a convict so we know a lot about him. Not only that, he lived a very long life among family equally blessed with longevity so the oral history is good. Robert Brown died in 1911, his son died in the 1950s.

Tracing the family back from Robert to his father Benjamin was simple. Benjamin (born 1785) married Susan (Susannah) Sargent, the names are unique enough, we have them in every British census and could build their family with ease. They spent their entire lives in the same region, which helped a lot. They were labourers, farmers, ploughmen. Agricultural pursuits right through. Our Robert seems to have been the only black sheep, the others appear fleetingly through births, marriages, death and census. A quiet and excessively numerous family.

The complication was Suffolk. Fordham is right there by the border and while Cambridgeshire records are available online, Suffolk is a great challenge for anyone not on the ground. Benjamin came from Fordham, his wife Susan from Exning. Susan was a brick wall for many years because of this.

Focusing on Fordham, Benjamin was the son of John and Mary Brown nee Boon.

Here is the whole family:

This may be incomplete. Eleven children seems like quite enough but there is a suspicious gap between Sarah No. 2 and John. There’s room for two more children there.

It’s complicated from here on in. For clarity, John Brown’s parents are referred to as ‘Granddad John’ and ‘Grandma Sarah’. John and Mary’s same name children are John Jr and Mary Jr.

We can only guess at the size of their house. Granddad John lived for some years in Landwade but was buried 1772 in Fordham, so he might have lived with them too.

A few generations earlier the Browns owned land and had wills, so our John Brown might have lived on his own land too. But his children clearly lived in rental places and there’s no indication land was passed on to any of them.

It was probably lucky that the eldest children in the family were girls. By the time their mother reached her less energetic middle years when pregnancy is harder and childbirth even more risky, she had a group of capable youngsters able to take the load on their own shoulders.

The first Sarah died at the age of six and was buried on 1st December 1767 in Fordham. She was the only child they lost.

John Junior was baptised on 27 November 1774, not long after the death of Granddad John. There was probably some rejoicing or at least relief that the head of this household would finally have someone to help with heavy work and take over the household if need be. Maybe they held their breath until he survived that dangerous first year. But the baby was hale and hearty, it seems, and grew up just fine.

Baby John had a bevy of personal carers in Mary (12), Elizabeth (10), Ann (8) and Sarah (6). That’s if Mary was still around; girls in this region were commonly out at work by the age of 11.

John was the only son for some time. His birth was followed by Hannah and Martha, and by now the eldest daughters were young adults, probably doing farm work or in service around the district.

Child number nine was Jane, baptised on 30th June 1782. She was six months old when her eldest sister was married.

The marriage was straightforward like everything else in the life of this family. Mary Jr was about to turn twenty. No doubt a capable home manager, experienced in childcare, she knew how to cook and sew and milk cows and nurse children through fevers and all those things required of a farmer’s wife in 1782. So when local man John Munns looked for a wife she was going to be a strong contender.

The Munns family, like the Browns and the Sargents, jumped across the county border between Fordham, Landwade and Exning. I’m not sure which side of the border they began on. I actually think we’ll find an earlier Munns link, maybe they were connected to Mary Boon. There’s a good chance young Mary Brown and John Munns knew each other their entire lives.

I imagine they announced their engagement and made plans. The marriage took place on the 18th November 1782. Everything respectable and straightforward.

I learned about Mary’s marriage through DNA matches with three of her descendants.

Two years later, Ann Brown formalised her union with farmer John Burling of Swaffham Prior, a village six miles from Fordham, three miles from Exning. It wasn’t a big move, but a good one socially and economically. John was a landowner in a modest way and his farm was quite successful.

I discovered Ann through another DNA match.

From this point on, the generations are slightly muddled. John and Mary Brown’s tenth child – their second son Benjamin – was baptised in the same week as Mary Jr’s apparent first child Henry Munns, and also John and Ann Burling’s first child Susannah.

At a calculated guess based on ages and proximity through the decades, I would say Benjamin Brown grew up playing with his nephew Henry Munns and his cousin William Brown. And the Sergeants/Sargents were there too, not yet married into the family but at all the social events. As were the Burlings.

This is the time to introduce William and Mary Sergeant of Exning.

William Sergeant and Mary Levett were married in 1778 and their eldest son John was born the same year. George, Susannah, William and Mary were born over the next few years, followed finally by Elizabeth, James and Eleanor. They lived in Exning and mostly stayed there, but they attended Brown weddings and the families knew each other well.

This, as far as I can see, is what English village life really was. Not individual families as we do it today, the whole population personally connected again and again. I’ve kept it simple so far.

In 1787, Susan Brown’s birth completed the family of John and Mary Brown. In this same year, little Susannah Burling was buried and a new Burling daughter was given the same name.

By 1788, Elizabeth Brown must have felt like an old maid. There she was aged 24, nieces and nephew popping up everywhere (Mary Munns and Isaac Burling came into the world this year). But finally it was time for her own wedding. She married John Frost in Isleham, Cambridgeshire.

I have DNA matches with Elizabeth’s descendants too.

Sarah may have married in Isleham too, there is a plausible marriage with John Peachey on 24th October, 1787. But I cannot be sure of it.

So having introduced all the characters now, we can pull the whole thing together. Mary Munns, Elizabeth Frost, Ann Burling and Sarah Peachey(?) are settled. There’s a gap now, which matches that six year child gap between Sarah and John. The four married daughters increased their families significantly. The Sergeant children are likewise growing up

On 27th September 1798, John Brown Junior was married to Ann Shinn in Fordham. He may have felt some pressure to do so, his father was sixty years old and probably starting to ail.

It was another time of rapid change. Hannah married Thomas Bishop in 1799 (DNA matches brought this to light), then came the death of John Brown, father and grandfather to this giant family. He was buried in the Fordham cemetery on 10th April 1801.

A year later was the first union with the Sergeant family through the marriage of Martha Brown and John, the eldest Sergeant boy. They spent a couple of years in Suffolk before moving to Fordham.

I discovered Martha through DNA matches too.

To conclude this part of their history, Benjamin Brown married Susan(nah) Sergeant in 1810, while Jane married William Durrant four years later. I have not located a marriage for Susan.

There’s just one more marriage to add.

The fifth Sergeant child was Mary, born a year later than my Benjamin, baptised 09th October 1786 in Exning.

Mary Sargent married Henry Munns, the eldest son of John Munns and Mary Brown.

It is confusing. John Sergeant married Martha Brown. Susan Sergeant married Benjamin Brown. Mary Sergeant married Henry Munns. You can see how I might have strong matches with other descendants of this family. But I also have matches with descendants of Eleanor Sergeant who married an Andrews, and with Burlings and Frosts and Bishops so it’s not just endogamy from that particular grouping.

In conclusion, this is the family of John and Mary Brown as I now know it, a long way ahead from my first post about the family. I wrote this first to help others researching this family, and second to show what DNA matching can achieve.

Now I’ll get on with more research.

Image captions:

1 Image by Bob Jones : Track near Fordham, from Wikimedia CC BY-SA 2.0 .No changes made.

2 Robert Brown born 1818 with grandchildren Lucy, Amy and Mary in Australia.

3 George Stubbs: Haymakers 1785 (public domain)

4 John Constable’s Wivenhoe Park, Essex showing rural features. There are very few paintings featuring Cambridgeshire.

5 Old man Family Hour Magazine 1854

6. I chose this image for the mud. Very much a part of British farming.

Clues and Theories

Lachlan near New Norfolk 1992

This is a complicated story. I warn everyone now. It’s a search for a DNA connection. However, this is the DNA cousin matching experience most of us are having. If you are forewarned, you will probably come through it. Names of living people have been changed to protect privacy.

Readers of my blog might know that we have a mystery relative, Ted, discovered by a Family Finder DNA test. That relative has been accepted into the family quite warmly, but we still do not know how he fits. He is a cousin or uncle to my tested family member John.

Ted and John are the same age and grew up in the same small town. Research has been complicated by the small gene pool in this region. Many living residents share two or three sets of ancestors in different configurations.

John has a relative called Ronald. The two of them share great great grandparents William Sargent and Mary Ann Kingshott on John’s father’s side. William Sargent came from Hawkhurst in Kent. Mary Ann came from Hampshire. The pair married in 1840 in New Norfolk, Van Diemen’s Land. William and Mary Ann had daughters Elizabeth and Eliza, among other children. Elizabeth is John’s great grandmother. Eliza is Ronald’s great grandmother. Through this connection they are third cousins.

John and his relative Ronald are also connected on John’s mother’s side. They share great great grandparents Robert Briers and Lydia Jelley. Both Robert and Lydia came from Leicestershire but married in New Norfolk, Van Diemen’s Land in 1845. Robert and Lydia had two sons Robert and William, among other children. Robert is John’s great grandfather, William is Ronald’s great grandfather. They are third cousins all over again.

Here’s my rough and ready visual with Sargent in black lines, Briers in Green. John and Ronald are both Sargent and Briers descent. I won’t add in more names to maintain simplicity and protect the privacy of living people.

At first I thought maybe we could use Ronald to close in on the connecting relative. But no! John and Ronald, as it turns out, have a large X-Chromosome match which should not be there. John and Ronald’s sister Anne have an even larger X- match. It’s a new puzzle to be solved.

On the above chart, an X chromosome from Sargent-Kingshott could pass through Elizabeth to her daughter (John’s grandmother) and from there to John’s father. But there it stops. It will pass from John’s father to one of his daughters (John’s sisters), but it cannot pass to John himself.

An X-chromosome from Lydia Jelley will pass to her two sons Robert and William, but will not pass from her son Robert to his son (John’s grandfather).

On the other hand, William Sargent, Mary Ann Kingshott and Lydia Jelley are all on Ronald’s X lines. The likelihood is that the discrepancy exists in John’s tree not in Ronald’s tree. We already had two NPEs showing in John’s tree (one being Ted, the other three generations back in John’s direct paternal line). This said, both families lived in this same complicated region so I could not be sure where the discrepancy lay.

New Norfolk 1992 from the lookout

Learning that the tree is still wrong does not surprise me, disappointing though it is. This branch came from a rural location where life was very rustic and human relations meant everything. Couples either married when both of them were about sixteen years old, or the age gap was anything up to sixty years. The modern traditional family structure cannot be taken for granted. Extramarital affairs were not uncommon and were rarely seen as cause for divorce. Desertion was cause for divorce and this right was exercised when needed. An affair on the part of one spouse was just as likely to be seen as ‘permission’ for the other spouse to do the same. Our family knows and understands that this is how it was. These recent ancestors were good people who loved their children and grandchildren, and successfully brought their families through some tough times.

It does make for a very complicated task when identifying DNA matches. John has five hundred and ninety eight Family Finder DNA matches and the only ones we can place with confidence are his own son and grandchildren. Even his known third cousin might be more closely related than expected! After a year of working at this, I think I have ruled out most of the incorrect family, but have not managed to replace with anyone.

Lately, I have been looking at the ancestor lists of his matches and searching their names in the local registers. If I found a neighbour of my family, related to an ancestor of one of our DNA matches, I’d be onto something.

Upper Swamp Road Lachlan circa 1993

Then came a new FtDNA family finder match, showing as predicted 3rd-5th cousin to John, and as predicted 2nd-4th cousin with Ted. I’ll call him Arthur. He shares a 26cM segment with Ted and 17cM of that same segment with John. There are a couple of other segments around 7-11cM shared with one or the other. There is a common ancestor somewhere. Arthur, however, does not show as ‘in common with’ Ronald. To be quite sure they were not a match, I emailed Ronald who told me Arthur is not on his match list.

This was rather exciting, but I’d been to this point so many times that I was not yet hopeful about it. I sent an email to Arthur, a somewhat subdued if not outright dejected email, saying I was trying to resolve an anomaly in the tree and asking if among Arthur’s ancestors, did any have siblings or descendants who settled in Tasmania? I wasn’t even sure what country Arthur lived in.

I received a response not many days later, with an offer to share trees and a few initial suggestions. Three of them are surnames from the correct region of Tasmania, but one name stood out to me. ‘Gleeson from Nenagh, Tipperary’, the email said. Catherine Gleeson born 1834 in Nenagh arrived it seems as a free settler and married Phillip Sallabank in Bendigo 1857. I looked in my tree and found a very exciting prospect.

In my family chart above, Robert Briers and Lydia Jelley had two sons Robert and William. To keep the family lines clear I simply labelled their spouses as ‘wife’.

William Briers’ wife was named Catherine Whelan. She was born in South Australia to free settlers William Whelan and his wife Winifred Moloughny, both from Tipperary in Ireland. They ended up at the goldfields in Victoria where William Whelan died, leaving Winifred with two young children. Winifred married again in 1855, as women generally did in that circumstance. Her second husband was Thomas Gleeson from Knocknagoogh in Nenagh, Tipperary. Then they moved to Tasmania, bringing Catherine, and settled in Lachlan. Four children were born to them in Lachlan before Thomas Gleeson died in a horseriding accident while drunk.

So, I thought to myself, what would it mean if John was a descendant of Thomas Gleeson and Winifred Moloughny? Was it physically possible? Did the years match?

The Gleesons in our tree

Somehow, I’d never looked properly at the Gleesons. Thomas and Winifred married in 1855 and the four children were born in 1856, 1858, 1859 and 1861. Thomas Gleeson senior died in 1862. I’ve blogged in the past about our Daley branch, Evelyn’s supposed paternal line. There is a question mark over every single generation as each time the mother was already pregnant or the child already born at the time of marriage. Evelyn was born in 1894 and her mother, Fannie Rawlinson, was a week past her sixteenth birthday. No official registration has been found for Evelyn’s birth. By the time Fannie married Sydney Daley in 1899 she was a single mother of three girls. Evelyn was the eldest of those three.

John received his X-chromosome from his mother who was Evelyn’s daughter, so the chances are strong that at least some of John’s X came from Evelyn. I have therefore looked closely at her as the pathway of this mysterious X-match portion, but of course her true father might have been anyone.

The dates do not match for the Gleeson girls to be Evelyn’s paternal grandmother. If one of them was her mother, then that girl gave a baby up for adoption to a just-turned-sixteen year old single girl? That doesn’t sound likely. The most likely scenario here – still a scenario I point out, merely a working theory – is that Thomas Joseph Gleeson born 1856 might be Evelyn’s father. Thomas Joseph Gleeson had received an X-chromosome from Winifred Moloughny. Catherine Whelan also received an X chromosome from Winifred Moloughny. Catherine’s X Chromosome might have passed down – daughter to daughter – to Ronald. If Thomas J Gleeson was Evelyn’s father, his X chromosome would have passed to her, intact as he had it from Winifred Moloughny. Evelyn passed it to her daughter who passed it to John. Ronald, while related to Winifred Moloughny, is not related by blood to Thomas Gleeson so would not match Arthur at all.

It’s a theory which works. Not an easy theory since Thomas J Gleeson married Harriet Brooks in 1876, and was still married in 1894. The legitimate Gleeson descendants might be surprised and concerned. Also, Fannie Rawlinson was in Ouse while Thomas J Gleeson was in Lachlan. One of them must have travelled to the other for some reason. Work maybe?

Over the next few months in my spare time, I will be attempting to prove or disprove this theory.