Elizabeth: A DNA Match from the Cincinnati Butlers; Lisa’s Butler Connections

I recently discovered that Elizabeth was a match to my late father-in-law Richard. I found her by doing a Tier 1 Cluster Analysis at Gedmatch. Richard’s clusters look like this:

All the gray squares connected to the colored squares probably indicate Richard’s French Canadian side. The last two small clusters likely represent Richard’s Irish side as they are not connected to the other clusters. The last two people in the last cluster are Elizabeth and Nathan. I wrote to Elizabeth and she confirmed that she was from the Cincinnati Branch of Butlers. I probably would not have known about Elizabeth if she had not uploaded her DNA to Gedmatch as Richard did not test at Ancestry. His two sisters tested there, but they don’t match Elizabeth by DNA.

Here are Nathan and Elizabeth on a tree:

The green boxes are for people who have detailed information on their chromosomes. They have tested at FTDNA, MyHeritage or 23andMe or uploaded their AncestryDNA results to Gedmatch.  The people in the bottom white boxes are connected by DNA and genealogy at Ancestry, but don’t have detailed information on their DNA.  This is what I call the Cincinnati Branch of Butlers. Nathan descends from the first wife of George Butler and Elizabeth descends from a second wife. Nathan’s ancestors moved to Nova Scotia where my wife’s ancestor married. Because of the genealogy, any DNA that Nathan and Elizabeth share have to come from George Butler but not either of his wives.

Here is the whole tree:

Because there are so many DNA connections between these two lines, I have proposed a Butler father to Michael Butler the earliest known ancestor of Richard and Henry Butler, the earliest known Butler ancestor of Elizabeth.  There are other possibilities.

Lisa with Possible Butler Ancestry

I wrote to Lisa about a year ago, because she had a large match with my wife’s two Aunts. Lisa would like to know how she fits in. Here is my wife’s side of the tree:

If I put up some DNA match numers for Lisa, it should tell where Lisa fits in on this tree:

That means that Lisa most likely descends from Marguerite Butler born in 1912. Here are some possible relationships between Amanda and Lisa:

My guess would be that Amanda and Lisa could be 1/2 first cousins or first cousins once removed. When I look at shared matches between Lisa and Amanda, I see a Herman in there which means that Lisa may have Herman and Butler ancestry somehow. This should give Lisa something to work on.

Lisa and Elizabeth

I bring Lisa up here because she also has a small match with Elizabeth:

That means that Lisa also has ancient Butler DNA links going back to Ireland.

Summary and Conclusions

  • I was fortunate to find Elizabeth through Gedmatch’s Cluster Program
  • Elizabeth is from the Cincinnati Butlers which link my wife’s Butlers back to Ireland
  • Lisa, who I had corresponded with previously also has a small match with Elizabeth
  • Lisa was unsure of her ancestry
  • Lisa gave me access to her DNA results at Ancestry. I compared her results to my wife’s Butler lines and she matched best with the Marguerite Butler Line.
  • Lisa and Marguerite’s granddaughter Amanda also have a shared match with a Herman which means that Lisa and Amanda could be as close as first cousins once removed.




Butler Visual Phasing: Part 4

In my previous Blog, I worked on updating the visual phasing for my late father-in-law and his two sisters. While doing this, I updated their DNA Painter Maps. I made an interesting discovery based on an old Cincinnati Butler descendant match that I had found back in 2015. The old spreadsheet that I had on Richard was helpful, so I updated that for him and one of his sisters. I will next update Virginia’s spreadsheet from Gedmatch.

Chromosome 17

My note says that this is done but  the paternal side needs to be idendified:

On Chromosome 17, I was trying to figure out what DNA my wife inherited from her father based on the Chromosome Mapping.

The key to Chromosome 17 is finding paternal matches. Not unlike the key to all the chromosomes. I’ll run Virginia against Jack, who is a top Kerivan match. I’ll look for the matches in common at Gedmatch and then run a Matching Segment Search. I come up with one match on Chromosome 17:

It looks like Virginia’s paternal crossover is around 6. That would mean that paternal GP1 would be Kerivan. I see that Richard has a match with Jane also:

I don’t see a DNA match between Jane and Lorraine.

I’m ready to declare victory over Chromosome 17 and move on to the next battle.

Chromosome 18

Assuming my previous work was correct, I first need to identify the easier maternal side:

Next, I’ll add DNA Painter information:

This works well, except for Richard’s purple match with Fortin/Tremblay. Based on my visual phasing, there should be no crossovers in that purple area. My guess is that orange is LeFevre and that the purple match that Richard has with Gerry is not right.

Here is Gerry’s tree at MyHeritage:

MyHeritage was not able to figure out our common ancestor. I’m not sure how I came up with what I did. I would say now that it is not right or that there is a closer common ancestor. I decided to delete Gerry from Richard’s map until I can figure out how he belongs.

The remaining question is the location of Virgina’s maternal crossover. That can be found at the Gedmatch full resolution image of the match between Richard and Virginia:


That happens here at about 42.5. That fits in with Virginia’s Lefevre match on Chromosome 18:

Next, I need to find a paternal match to identify the paternal side of Chromosome 18.

It appears that the salmon colored paternal side is predominately all Kerivan or all Butler. An in common match between Jack (who represents Kerivan) and Virginia didn’t show any matches on Chromosome 18. That means that I should run an in-common match for one of the Butler matches. Lorraine has a pretty good match with Kim at Gedmatch:

I haven’t connected these two families by genealogy but they are connected by DNA and many circumstantial incidences. I had no luck finding Chromosome 18 matches with this comparison either.

Chromosome 19

I’ll add the DNA Painter map to the bottom of the work I had done previously:

It looks like Lorraine should have a maternal crossover where her match ends, but Richard’s Lefevre green is actually two matches:

The problem is that I have not mapped Jane to Lorraine’s map. Mapping Jane onto Virginia’s map helps show Virginia’s Pouliot segment between the two Lefevre segments:

It looks like there should be a crossover for Lorraine on the right side of her match with Jane. However, it appears that the match between Jane and Richard was larger previously. Now it is in line with the matches between Lorraine and Virginia:

I guess the algorhythm changed at Gedmatch. I don’t feel like changing Richard’s results right now.

A Paternal Match for Chromosome 19?

Using my previous in-common analyses, I have this Kerivan-type match for Lorraine:

I have this Butler-type match – also for Lorraine:

Unfortunately, that does not fit in with the way I have Lorraine mapped, unless I change the visual phasing. Due to uncertainties, I’ll leave Chromosome 19 for now.

Chromosome 20

Here is what I had done previously for Chromosome 20:

I’ll convert this to Excel from Powerpoint. I also want to look at the DNA Painter maps for these three siblings. Here is Lorraine:

She has LeFevre DNA at the beginning and Kerivan DNA at the right end of Chromosome 20. Lorraine’s matches appear to define Paternal GP1 as Kerivan.

Virginia shows Pouliot up to position 60:

Here is what Richard has:

When I re-do this Chromosome, I’ll start from the righ hand side as the left-hand first crossover is not as important. I did this but got the same results:

However, now I know the paternal side. At some time, I need to look at Virginia’s match list to see where her matches are falling out after position 60.

Chromosome 21

Here is what I had done previously in Powerpoint:

I’ll start by looking at the three DNA Painter maps. Only Richard has one match so far. That is for Pouliot between 10 and 22:

When I bring this into Excel, I need to see the left-hand side:

With the left side completed, I see the only color that Richard has uniquely there is blue, so that has to be Pouliot. That defines the maternal side of the three siblings’ Chromosome 21 and defines the maternal grandparents. However, it leaves a gap in Lorraine’s Chromosome 21.

Chromosome 22

Here is the work I did previously:

All l need to do is identify the two paternal grandparents. I checked the three DNA Painter maps and saw no paternal matches that would help me.

Summary and Conclusions

My list of completed chromosomes is getting larger:

I now have only 8 chromosomes that are not completed. If I do some more work on these, I may be able to complete one or two more.

  • It was helpful referring to the DNA Painter maps for these three siblings.
  • Looking at in common matches was helpful at Gedmatch
  • I will likely continue working on these unfinished chromosomes and follow up on some of the paternal in common DNA matches – especially on the Butler side.

Visual Phasing My Father-in-law’s Butler Side: Part 3

Here is where I am with visual phasing of my father-in-law’s Butler side:

I started about four or five years ago and recently have been updating their visual phasing. Visual Phasing maps out in what way three siblings got their DNA from their four grandparents.

I’ll look at the chromosomes that I haven’t completed and see if I can complete or improve on any of them.

Chromosome 1

I’ll update this to Excel and start over. The crossovers seem straightforward until we get to the right end. Then there are three in a row:

Here is my attempt at visual phasing:

When there are two crossovers in a row for the same person, it means that there is usually a Half Identical Region (HIR) which goes unidentified.

If I add some more Pouliot to Virginia’s DNAPainter map, it will help. I will also add more LeFevre:

This brings Virginia up to 18% painted.

Next, I’ll add the LeFevre matches to Lorraine’s map:

This doesn’t add any new LeFevre information for Lorraine, but may be helpful for other chromosomes. Next, I put the DNAPainter maps below the Visual Phasing Chart. This will tell if the theoretical and actual matches agree:

The scales may not match here. For Lorraine, it is clear that there is a crossover from LeFevre (blues) to Pouliot at 205:


I checked Richard vs. Virginia and that crossover is actually 204, not 202.

Virginia has a crossover from Pouliot to LeFevre at the same location:

That likely means that Richard does not have a crossover at 204. That is good to know.

Now Chromosome 1 is starting to come into it’s own. I see that Richard has a crossover here:

At 217, the crossover goes from Pouliot to LeFevre. I also see that Richard matches a LeFevre here:

I now have Richard and Virginia’s maternal sides finished and identified:

I think that this is as far as I can get, unless I find more matches between 0 and 70:

Chromosome 5

I had that this Chromosome was not started.

I recopied these comparisons from Gedmatch as he previous ones didn’t seem right. This is what I get without looking at Half Identical Regions (HIRs):

When I add my one allowed HIR, I get this:

Next I added in matches from DNAPainter:

Virginia has a large match that Lorraine and Richard do not have. This match is from 75 to 118. Michelle represents LeFevre, so green must represent Lefevre. Dark blue on Lorraines bar is also Lefevre, so that would support salmon being Lefevre.  Then there are the three Michelle matches near the beginning of Chromosome 5.

  • Lorraine: 18-31
  • Virginia: 28-34
  • Richard: 18-34

This suggests that Lorrain has Lefevre to Pouliot crossover at 31 and that Virginia has a Pouliot to Lefevre crossover at 28. That is good, but it means that I have to adjust my crossovers. Instead of two Richard crossovers, there is a Virginia crossover and then a Lorraine crossover. I think the problem is that I didn’t add enough detail to the comparisons. For that, I will need to lower the match level to 3 cM:

This shows another segment from 35 to 38. I may have to come back to this Chromosome as it seems quite complicated – especially in the region around the 30’s.

Chromosome 8

Here is the work I had done in 2016:

I redid Chromosome 8 in Excel:

This time I saw an R and L crossover where I saw a V crossover previously on the right side of the Chromosome. This is what I get this time:

The small blue segment on Lorraine’s top bar looks ridiculous, so I will take that out. Next, I add information from DNAPainter. However, these matches are Lefevre only:

This tells me that Lefevre is salmon and Pouliot is green.  This completes the mapping for Lorraine and Richard, but it does not identify and paternal segments. The arrows point out where the crossovers were for Richard and Lorraine, that I had previously identified as a Virgininia crossover:

It may look like Virginia should have a maternal crossover, but that is only because I did not map one of her segments. I just did that and here is here extra segment:

Joseph Methot was the father-in-law of Edmund LeFevre, so on the LeFevre side.

Chromosome 13

Here is what I had done with Chromosome 13 in 2017:

An identifying match for Virginia would help between 39 and 59.


It turns out that Virginia’s match with Philip that I just added is helpful. The match was between 40 and 60:

It appears that there was a match to a Rooney descendant which is on the Kerivan side. That helped identify the maternal side of Chromosome 13. While I’m at it, I’ll add Philip to Richard’s DNA Painter Map:

Richard’s match with Philip starts at 36, so it defines Virginia’s crossover from Pouliot to Lefevre at 39 or 40.

Chromosome 16

I had origingally worked on this Chromosome in Word. I redid the work in Excel and came up with the same basic results:

Next, I look at DNA Painter. Here is what I have for Richard:

He shows Pouliot at the start and near the end. then in the middle, there is some really old Lefevre DNA going back to 1715. This is from a match from Christine, that I found at Gedmatch. She had uploaded from Ancestry, but I can’t find her at Ancestry right now. However, I assume that I had the common ancestors correct. I see that Richard also has a match to Doris at MyHeritage. This common ancestor is on the Pouliot side:

This should define a Pouliot to Lefevre crossover at about 50. The problem is that I haven’t identified a crossover for Richard or any crossover at the 50 location on Chromosome 16.

One thing that I do know is that all three Butler siblings match on the Pouliot side at the beginning of the Chromosome:


Virginia and Lorraine do not match on the Pouliot side near the end of the Chromosome, so I came up with the above map. I think that French Canadian intermarriage may have messed with the middle of the Chromosome, so I may have to change this at some point.

Looking for Paternal Matches

I went to Gedmatch to download 1,000 of Richard’s newest matches. Most of these should be from Ancestry. As Richard tested at FTDNA, I would be better off downloading VIrginia and Lorraine’s results. They tested at Ancestry. However, I did find something interesting on Richard’s old spreadsheet:

Here is a match James who I have on Richard’s maternal side. My note says he has Cincinnati roots. Here is  the maternal side of his tree at FTDNA:

The least documented side is Branch where I think the connection is. Alma’s mother was Rebecca Butler based on Alma’s marriage record:

Here is Rebecca in the 1870 Ward 17 Cincinnati Census:

My working theory is that my wife’s Butlers are related to these Butlers. That helps me solve Chromosme 16:

Here is how Richard and James match:

This is how I have connected the two trees, but I don’t have the genealogy to do it at the top level.

Lorraine Matches James Also

Lorraine matches James here on Chromosomes 4 and 16:

Virginia  matches James on Chromosome 4 only.

Summary and Conclusions

  • 9 out of the Butler sibling’s chromosomes are believed to be completed
  • While looking for paternal matches for Visual Phasing, I found one that I had discovered in 2015. It turns out that this match was a descendant of the Cincinnati Butlers who are related to my wife’s Butlers by DNA.
  • DNA Painter has been helpful in identifying matches also and I have been adding to these three Butler siblings’ maps
  • The only cousins on the Butler grandparent side used to check the visual phasing are those that have not been proved by genealogy, but are certainly cousins by DNA.
  • I will continue on with Chromosomes 17 through 22 in a subsequent Blog




Updating the Visual Phasing of My Father-in-law’s Butlers: Part 2

In my previous Blog, I was successful in updating the visual phasing for the four grandparents of my father-in-law. I was able to complete several of the 23 chromosomes. This was my previous summary:

Chromosome 11

It appaears that I should be able to find some maternal matches to complete Chromosome 11. This is what I had come up with in 2016:

Here is what I have for Richard on his Chromosome 11 in DNAPainter:

I found this match for my wife from a previous Blog:

This indicates that Kbou would be on the Pouliot side for my father-in-law. From the key for Richard’s DNAPainter map, it appeared that Gagne/Girard was improperly put in a group with LeFevres, so I will move that down:

That means that Richard’s crossover at position 117 is his change from Pouliot to LeFevre:

Here is the finished Chromosome 11:

Kbou is also an X Chromosome match.

Marie Girard was my father-in-law’s mother’s mother’s mother’s mother’s mother’s mother. If Richard or one of his sisters were to have the mitochondrial DNA test, that would pass to the mother’s only side also.

Chromosome 17

This was some of my early work in Visual Phasing as it is in Word:

As paternal matches are very scarce, it is difficult to identify whether blue or purple is Butler or Kerivan.

Chromosome 22

Unfortunately, I could find no paternal matches on Chromosome 22.

Chromosome 5

The crossovers were complicated in this Chromosome which is perhaps why I did not pursue this Chromosome previously:

I’ll give this a shot now. I’ll start in the middle:

Things seem clearer there. The confusing par is around 30 to 40. I’ll work on solving the right side and hope for good matches on the left side. Here is a partial solution:

Lorraine and Alan have a small match on Chromosome 5:

Lorraine has a better match with Yvonne:

This would be a good match to compare with Richard and Virginia. Here is Richard’s match with Yvonne:

I don’t see a match for Virginia.

Unfortunately, Richard and Lorraine have a Fully Identical Region from 77 to 116:

That means that this hint does not help.

Back to Alan

Virginia and Richard show no match with Alan on Chromosome 5. If Lorraine’s match is valid, I show this:

That match identifies the top bar as maternal and the bottom as paternal. That would also meanj that Lorraine and Richard would have LeFevre DNA between 77 and 116 which corroborates with the Yvonne match. I’ll leave Chromosome like this for now:

Chromosome 1

I worked on this Chromosome in 2016:

On Richard’s DNAPainter map, he has this second green match with KK:

Thjat match represents LeFevre which I have a purple:

Chromosome 2

This Chromosome is interesting as it has some Butler DNA from an unidentified Butler ancestor. I will re-do this Chromosome in Excel. Richard’s DNAPainter map shows that he has Kerivan DNA on the right side of his paternal copy of Chromosome 2:

Light blue represents Kerivan and dark blue is unknown Butler DNA.

It would be helpful to ‘paint’ Lorraine’s paternal Chromosome 2 as well as Virginia’s.

Here is Lorraine:

As a bonus, Lorraine’s match with Kim shows a bonusk segment of unknown Butler in Chromosome 4.

I don’t have anything mapped for Virginia yet. I’m not showing a match between Virginia and Kim. Here is Virginia’s matches with Jack who has Kerivan ancestry:

This mapping is not necessary, but it is helpful to see where Virginia’s Kerivan matches are. At the right side of Chromosome 2, I also see the possibility of a double crossover:

Instead of two Virginia crossovers, the last Virginia may be a Richard and Lorraine crossover. Here is what I get for the right side of Chromosome 2:

Next, I need some more maternal side for Lorraine. I unlocked something at FTDNA so I could get a good match. This is with Michelle – on Lorraine’s LeFevre side:

Next I unlock Virginia’s FTNDA results. Here is how she matches Michelle:

I have not been consistent with the colors between siblings. From just cousin matches, the Chromosome 2 Map looks like this:

I can expand some of the segments like this:

That leaves the left side of the chromosome undone. I can pretend I don’t know the results for the right side, solve the left side and then add in the right side.

The bottom three pairs of bars were visually phased.

Michelle’s DNA Matches with Richard, Lorraine and Virginia

I am hoping that key to solving Chromosome 2 lies in Michelle. Michelle matches:

  • Lorraine 31-69
  • Virginia 37-115
  • Richard 38-75

The key appears to be in the beginning of the matches. My guess is that Michelle who represents LeFevre is the salmon color:

That also means that the map would be the same except that the colors for LeFevre and Pouliot are switched. Virginia will still have the blue Kerivan on her paternal side. Green will be Butler and orange will be Pouliot.

And the answer is:

The key to solving this Chromosome:

  • Matches with Michelle and Jack
  • Solving the left side as if I didn’t know about the other DNA matches and then adding them in.
  • A double crossover for Richard and Lorraine at position 227.6.

The Importance of Chromosome 2 for Butler Genealogy

The green Butler segments of Chromosome 2 are where Richard and Lorraine match Kim and Nathan. This is my best guess of how the Butler families tie together:

Kim and Nathan descend from George Butler who lived in Cincinnati . My wife’s 2nd great-grandfather Edward Butler also lived in Cincinnati for a while – not too far from George Butler. Here is how Nathan and Kim match each other:

This is the DNA that Kim and Nathan share that came down from George Butler who was born in 1826 in Ireland.

Summary and Conclusions

  • 12 of the chromosomes have been completed.
  • It takes time to get matches to confirm and improve the work
  • As Butler matches are scarce, it would be better to try to get some Butler matches from AncestryDNA to upload to Gedmatch, MyHeritage or FTDNA.
  • It takes a long time to figure out who matches you on a specific chromosome. If you have a profile on DNAPainter, that makes the job easier.



Updating My Wife’s Father’s Side Butler Visual Phasing

Several years ago, I worked on the visual phasing of my wife’s paternal side. I administer the DNA for my wife’s father and two aunts. That gives me enough information to do visual phasing. I downloaded the Stephen Fox spreadsheet, but this does not work well now. That means that I can just update my old excel individual Chromosome maps.

New Butler Matches

Part of the problem I had with visual phasing of the Butler family is that there were a lot of matches on their maternal French Canadian side, but not as many matches on their Irish side. In there past several years, there should be new matches that could help in identifying the paternal grandparents.

Chromosome 2

This is what I had for Chromosome 2 – last worked on in 2018:

Here Gaby is not very helpful as she is a first cousin. Here is another match on Chromosome 2:

Lorraine and John are 2nd cousins, once removed on the Kerivan side. Here is how John and Lorraine match:

John will be even more helpful on Chromosomes 7, 15 and 20.

The position of 227M is significant as it occurs at a crossover. Here is John’s match with Virginia:

Here is John and Richard:

This shows that my former Chromosome 2 map was wrong on the right. All the DNA on the right should be Kerivan and not Butler. I’m a bit out of practice with these, so I’ll move on to Chromosome 7 which I hope will be easier.

Chromosome 7

Here is what I had for Chromosome 7 back in 2017:

It looks like I was having trouble with this one also as I have two different tries. I can now see by Richard’s match with John, that his crossover at 83 is on the paternal side:

Following John’s matches, it appears that this could be the answer:

This is a case where one good match can map both the paternal and maternal sides of a whole Chromosome. While that is good, you can see that there is quite a bit of Butler [green] DNA missing between these three siblings. It would be a good idea to verify the maternal side. I checked MyHeritage and Fred matches Virginia there on Chromosome 7 where I show Fred’s sister matching between about position 70 and 80. That confirms my earlier work. It is satisfying to have this Chromosome finished.

Summary of Butler Chromosomes

It would make sense to summarize the condition of the Butler 23 chromosomes in a spreadsheet.

It took a while to go through all my files. This shows that six out of 23 chromosomes are phased by grandparent. At this point, I will assume that the green highlighted chromosomes are correct. Next I can leither look at the yellow highlighted chromosomes or revisit the Butler matches with John on the Kerivan side.

Chromosome 4

This Chromosome was analyzed in PowerPoint which is not ideal:

These larger chromosomes can have a lot of crossovers. Fortunately, we now have John’s DNA match at Gedmatch. From above, I see:

  • John doesn’t match Lorraine
  • John matches Virginia in three places
  • John matches Richard in one place near the right end of Chromosome 4 (at 183 to 185)

From this it seems obvious that Butler is green and Kerivan is blue on the map above. This is how the paternal side comes out:

Richard’s match to John ends at 185, so that describes his last paternal crossover at 185. VIrginia’s match to John is from 183 to 189. The question then is why doesn’t Lorraine match John on the right side of the Chromosome if she shows blue Kerivan there? I can show more detail on the match between Lorraine and John:

I’m really showing nothing – especially on the right hand side. That means that Lorraine should be all Butler maternally and other adjustments have to be made. Here is my correction for the right side of Chromosome 4 which appears to be closer to the truth:

I moved Richard’s crossover to 185. I ignored the two crossovers for Lorraine.

Back to Chromosome 2

I will stay with the John matches and see if I am closer to tackling Chromosome 2. Here I added John’s DNA matches at the bottom:

Because Virginia matches John sooner than Richard and Lorraine do on the right, this could indicate multiple crossovers. I’ll take out the last Virginia crossover and add in two for Richard and Lorraine.

I’m not sure how I would have figured out that there was a double crossover at 227.5 were it not for the match the Bulter family has with John.

Chromosome 6

Here John matches only Virginia:

That small match identifies green as Kerivan:

However, it raises the question as to why Lorraine does not match John between positions 103 and 108. When I lower thre threshold, I see that she does:

That tells me that this was a valid match between Lorraine and John. It just got clipped on both ends. This also confirms Lorraine’s paternal crossover at about 107.5.

John’s DNA Match and Chromosome 8

John matches Virginia only on Chromosome 8:

Here is what I had done previously on Chromosome 8:

John’s match with Virginia is to the far right of Chromosome 8. That means that blue is Kerivan and Green is Butler.

John’s match with Virginia did not define any new crossovers but it did make the work that I did previously more useful. For example, if Richard comes across a paternal match on the right side of Chromosome 8 at 75 or higher, it will be on the Butler side.

Chromosome 9

John matches Virginia and Richard (but not Lorraine) on the same area:

I had already ‘solved’ Chromosome 9 previously, but let’s see if John’s two matches fit in:

I see already that I had a labeling error as I have Butler in the orange and green segments. Also I have Lorraine and Virginia as the same paternal color which is wrong. Problems. Here is the link to the Blog where I made the mistakes. This was how I had Chromosome 9 before I went wrong:

It seems like this is a better rendering of Chromosome 9:

I changed Virginia’s first crossover to her  maternal side. This is because she matches John who has Kerivan ancestry at position 8-14. I notice again, if I mapped this right, that the Butler DNA is skimpy. If there is a big Butler match waiting to be found in the middle of Chromosome 9, it will not be found by these three siblings.

Chromosome 10

Virginia and John match on Chromosome 10:

Right now I have no names on Chromosome 10:

I would need more matches to figure out if that is maternal or paternal.

At MyHeritage:

Here is the DNA match:

Lorraine’s match to Philip looks the same:

Philip and Richard do not match on Chromosome 10:

Unfortunately, even that does not help. Thinking a bit more, the first match between John and Virginia does help:

That match with John representing Kerivan must be on the salmon color. That is because no one else matches Virginia’s salmon color in that area of the match. If the match was in the green, then Lorraine would also have to match John there. This appears to be the answer:

From looking at first cousin matches previously, I seem to have figured out the maternal and paternal sides of this Chromosome.

Chromosome 12

John matches Virginia and Richard here:

I had already figured out Chromosome 12:

The match with John supports the mapping on the right as Lorraine shows Butler in yellow.

Chromosome 15

Just two chromosomes to go. John matches Lorraine here:

Some of my early visual phasing was done in Word:

Virginia matches known 2nd cousin Fred at MyHeritage:

Here is the DNA match:

Here is Lorraine and Fred:

This suggests a maternal crossover for Virginia at about position 28

I’m going to try to start Chromosome 15 again. This time in Excel:

I put in who owns the crossovers and most of the positiongs. I started at the right where Lorraine and Richard have no matches, so opposite colors. I then moved Lorraine’s colors to her next crossover. Lorraine and Virginia have a Fully Identical Region (FIR) in green so I added that in. Next, I’ll do a random half identical region between Lorraine and Virginia and see where that goes.

That looks better. It is all done except for the left side. I now see that John’s match with Lorraine must be on the salmon color. That is because John matches only Lorraine and not Virginia nor Richard. That means that salmon represents the Kerivan quarter and blue the Butler quarter.

Now I just need to look at the two Pouliot matches from MyHeritage. This should be the finished Chromosome 15 for Richard, Lorraine and Virginia:

That match between Fred, Lorraine and Virginia helped define two maternal crossovers. One was at 28 and one at 33. These were a little off from where I drew the original crossover lines at 27 and 31. On this Chromosome, some Pouliot DNA was lost between the three siblings from 60 to 95. I put Richard’s crossover at35 on his paternal side as he should have had a match with Fred (Pouliot side) if his crossover was on the maternal side.

Chromosome 20

Chromosome 20 looks fairly simple. I used a first cousin match somehow to come up with the map. Lorraine and Richard have this match with John (Kerivan side):

The way I mapped it, Virginia and Richard have the same Paternal grandparent. That means that I made a mistake or that the match is wrong. The one place I can go to for matches by Chromosome is DNAPainter. I have painte some of Richard’s DNA there:

This shows no paternal matches for Richard, but matches with two people on the maternal side. Michelle is at FTDNA here grandfather was Joseph Martin LeFevre. Richard descends from a first wife and Michelle from a second wife. That means that there is only one common ancestor between Michelle and Richard. From DNAPainter, I see that the match is from 0-8. That means that LeFevre is right for the left part of Richard’s

Line is at Gedmatch and goes back to Delisle who is on the LeFevre line. This would also be correct for the right side of Richard’s Chromosome 20. Here is Line and Lorraine:

That means that the right side of Lorraine’s Chromosome 20 is right. I don’t see Line matching Virginia at Gedmatch, so that would support the right side of Virginia’s Chromosome 20 also.

At this point, I’m at an impasse. It could be that Virginia has two extra crossovers here:

I’ll just leave this Chromosome as is for now and try to solve it later.

Summary and Conclusions

Here is my spreadsheet:

  • I went from 5 to 10 chromosomes completed.
  • I corrected one Chromosome I thought had had completed previously (Chromosome 9).
  • I improved some chromosomes without solving them, making them more useful
  • The chromosomes were helped by a match to second cousin once removed  John. He has his DNA at Ancestry and at Gedmatch.
  • MyHeritage was also helpful as they have ‘Theories of Relativity’ which give likely common ancestors and have detailed chromosome matching information.
  • Finally, DNAPainter is helpful in looking at specific chromosomes to see who the matches are there.
  • I will need to follow up on ‘painting’ Virginia and Lorraine.
  • I will also need to follow up on working on completing more of the visual phasing for the DNA of siblings Richard, Lorraine and Virginia.

More Painting of My Wife’s DNA

I call this painting my wife’s DNA because I use a program called DNA Painter. But really, it is a DNA map of my wife’s identified DNA matches. Here is what I have so far for Marie:

The top row of every one of Marie’s chromosomes is what she got from her father and the bottom row is what she got from her mother. Likewise on the key to the bottom right of the image, the top part shows the shared paternal ancestors represented in different colors on Marie’s map. The bottom part of the key has Marie’s maternal ancestors represented as colors that she got from her DNA matches.

Marie is About One Third Painted

Actually a little less:

Marie is actually doing the best as far as DNA matches go, on her paternal side:

Now that I think of it, that makes sense as there a lot of French Canadian matches on that side. That must mean that Marie is less painted on her maternal side:

Marie only has 57 DNA segments painted on her maternal side. My goal in this Blog is to get Marie’s numbers up somehow. It is not easy as there has to be a link to a common ancestor and they have to have had their DNA tested at a place that gives detailed DNA match information. AncestryDNA is the biggest tester and they don’t give detailed information on their DNA matches.

Searching for More DNA Matches for Marie

The easiest place to search would be at MyHeritage. They have an option called Theories of Relativity where they match up DNA matches and common ancestors. One of the first matches I see at MyHeritage that I haven’t mapped yet is with Ivan:

Ivan has two paths to Marie and Ivan’s proposed common ancestors of Peter Upshall and Margaret Burton.

Here is the DNA that Marie and Ivan share:

By default, I don’t think that DNA Painter will map Chromosome 11 as the match is 6.9 cM and DNA Painter has a 7.0 cutoff. That is sort of too bad because here is Marie’s Chromosome 11:

Marie’s matches with Esther, Anne and Edward are with ancestors on the Upshall side. But I don’t want to change the default either. Here is what the data looks like that will go into DNA Painter:

Unfortunately, Ivan’s areas of matches were already taken up with others for the most part. There was a small exception on Chromosome 1:

Here Ivan’s match with Marie in green was a little longer than Marie’s match with her half great Aunt Esther.

Marie’s Theory with James

Here is what MyHeritage shows:

James Ellis and Marion (with variations) MacArthur (or McArthur) are Marie’s third great-grandparents. Marion MacArthur was James Ellis’ second wife:

I show Thomas Birch Ellis as James Heny Ellis’ brother so that is good.

I’ll paint this match onto Marie’s map:

The parts of James’ DNA matches that were over 7 cM were on Chromosomes 1 and 2. They are in Ellis regions. There is no overlap, so they constitute new unmapped areas for Marie. This gets Marie up a percent overall from 31% to 32% mapped. On Marie’s Maternal side, she went from 24% to 25% mapped. Progress. I see that Marie also matches Stephen who may be James’ brother. However, Stephen does not appear to add any more DNA than James added.

Marie and Glenys

Glenys shows as a third cousin to Marie:

MyHeritage has her descended from Alexander Upshall. The record I have for him says he died at a lunatic assylum:

Let’s map this match out using DNA Painter. DNA Painter gives a hint:

I’m not sure if this is helpful or not, but it seems like the bottom hint is right. Glenys’ matches with Marie on Chromosomes 9 and 11 are already taken up:

The match on Chromosome 22 is Marie’s first on her maternal side:

This small match is not enough to increase Marie’s percantages, but it helps.

Keep Going with Marie’s DNA Match Caroline?

Caroline is supposed to match Marie on the LeFevre/LeFebre Line:

The connection looks legit. LeFebvre is the older way of spelling LeFevre. Someone left out the b along the way. Hopefully the 1891 for the Charles in Caroline’s line isn’t for his birth as that would to make sense. 1841 would make more sense. At this point, I can either assume that Caroline has the right tree or check on it myself. As I’m feeling lazy today, I’ll just check the DNA to see if it is in the right vicinity.

Turns out I already have Caroline mapped. Here is her bigger connection to Marie on Chromosome 12:

That match is solidly in LeFevre territory, so the connection is likely right.

Marie and Alan

This time I’ll go right to the DNA:

Alan overlaps other DNA Matches except for Chromosome 13:

Alan also adds a new French Canadian common ancestor to Marie’s DNA match who are Joseph Methot and Marguerite Anger.

Marie and Edna

This could be a long Blog. I see that Marie has at least five pages of Theories of Relativity. Edna is Marie’s first Theory on Page 2.

Edna adds some old Christopher Dicks DNA from 1784 on Chromosomes 1, 9 and 12. Here is the connection:

I’m not sure if Planter is the right last name for Margaret. One problem is a DNA overlap on the Ellis side at the end of Chromosome 1:

It could be that Edna’s small match with Marie there is a false positive.

Marie and Marilyn

I think that I have been in touch with Marilyn before as I have done some work on the Dick’s Line:

Marilyn doesn’t add new DNA, but shows that the match Marie has with Esther and Anne goes further back than the 1870’s or 1812. I’m still trying to get Marie’s painted DNA up to 33% but I’m not there yet.

Marie and Andre

Now we are up the 5th cousin level for Marie. As you might expect, the match levels are smaller:

Andre adds a little DNA to Chromosome 15 at the end and gives older DNA bac to 1773 on the LeFevre Line. I’m still not up to 33% painted for Marie:

However, we have 200 segments painted.

Marie and David

This match goes back to Gorrill:

As I don’t see this couple in my DNA Painter Key, this must be a non-painted match.

David provided Marie’s first maternal mapped DNA match on Chromosome 14:

This match got Marie up to 1/3 painted or 33% overall:

That got Marie from 25% to 27% painted on her maternal side.

Reorganizing Marie’s Key

At this point, it would make sense to reorganize the Key in Marie’s DNA Painter Map. Right now I have a line separating Marie’s paternal and maternal matches. Marie’s four grandparents are Butler, LeFevre, Ellis and Upshall. I can organize Marie’s key that way. Unfortunately, I only see one Butler match right now.

The LeFevre matches I just organized by date.

Marie’s Ellis and Upshall matches split out more evenly:

I see a mistake as I left Richard Gorrill up with the LeFevre’s.

Marie’s match with Robert on the Ellis Line


Robert adds some new DNA on Chromosome 3 and gives corroborating evidence on Chromosome 4.

Nathalie: A Fifth Cousin Match on the LeFevre Line

I’ll paint in Nathalie’s DNA as I don’t want to take the time to check every tree:

Nathalie provides new DNA mapping for Marie on her Chromosomes 14 and 19. Now Marie is 40% mapped paternally:

Marie’s Progress Update

Marie has DNA mapped on every chromosome. However, she is missing paternal DNA on Chromosome 5 and 17 and she is missing maternal DNA mapped on Chromosome 16.

Marie and Charolette

I am a little suspicious of this match as there may be a connection on Charolette’s MacArthur side also.

So whether Marie and Charolette connect on the William Ellis Line, MacArthur line or both is a fine point. This could be resolved possibly by further looking into Charolette’s ancestry. Marie is now 34% mapped or painted overall.

Marie and Agnes

Marie and Agnes have several potential connections:

This is the closest theory, so I would go with this one. The other theories are further out:

I don’t show an Edward Rayner in my tree, but I do show that the John Rayner at the top of the tree was the son of Edward John Rayner, so it would make sense if he named a son after his father. Here is where Agnes sorts out on Marie’s map:


Marie and Barbara – Still at MyHeritage

I made a mistake with Barbara at DNA Painter. I put down that she descended from Ellis/MacArthur:

She actually descends from Ellis/Ramsay. That should be easy to fix in DNA Painter. I can edit Barbara in DNA Painter:

I thought that I had an entry for just James Ellis 1801, but I don’t. So I’ll create a new group.

I’m getting there, but it says that James Ellis is not a known ancestor. I need to change that. I go back to the edit section for Barbara and check the known ancestor box:

Barbara fills in some new areas on three chromosomes:

This is turniing out to be a long Blog, but I am getting a lot of mapping done.

Marie and Gilles

As expected, Gilles matches Marie on her French Canadian side. This is on her Fortin/Tremblay side where Nathalie matched earlier in the Blog:

Under Nathalie, the common ancestors were Eustache Jacquies and Marie Judith, but these must be the same ancestors. Gilles adds some more DNA on Chromosome 1. His matches on other segments were below the 7 cM threshold.

I changed the Fortin/Tremblay mapped color to light blue as the previous color was too close to the pink LeFevre color. This match with Gilles maps some older DNA from the latter part of the 1700’s.

Two Possibilities for Jason

I am leaning away from Theory 1:

My tree shows that Marie descends from John Rayner, not James Rayner. However, Jason and Agnes triangulate on Chromosome 4:

I had mapped Agnes to Rayner. I think I’ll skip Jason for now.

Let’s Try Richard

Richard is on Marie’s MacArthur/MacDougall Line.

Richard added DNA to Marie’s map in previously unmapped areas on Chromosome 1, 9 andd 17:

This finally gets Marie mapped overall up to 35%. If I mapped this right, Richard’s match to Marie on Chromosome 17 points out a crossover. The green and light blue are from Marie’s Upshall side and the dark blue is from Marie’s Ellis side. This shows about where Marie’s maternal DNA went from Upshall, then to Ellis and Back to Upshall.

Marie’s X Chromosome

I don’t think that MyHeritage covers the X Chromosome. That is covered at Gedmatch and FTDNA. Here are Marie’s X Chromosome matches at Gedmatch:

Marie’s top two matches are her Aunts. They would be too closely related to map. The third match has a large X Chromosome match, but no autosomal match, so I would not be able to locate that match at Ancestry.

Marie and Lise

The fourth match is with Lise at Ancestry. Perhaps I can find her. I found someone who looks like Lise at Ancestry, but she does not have a maternal line given.

Marie and Muriel

After some dead ends, I came across Muriel. She is at Ancestry and Gedmatch and shows a common ancestor:

Here is where Gedmatch shows the X Chromosome (or 23) match:

This mapping was in a previously mapped area with Marie’s 2nd cousin. It does show that this is old DNA going back to the early 1700’s if Ancestry is right:

I also added Muriel’s matches on Chromosome 14:

Although I didn’t add much new information, I did show that I could add X Chromosome information.

Marie’s New Chromosome Mapped Update


Summary and Conclusions

  • My primary goal for this Blog was to update Marie’s Chromosome Map which I did.
  • The secondary goal was to get Marie at least 33% mapped and I did that also. I was able to get Marie mapped overall from 31% up to 35%
  • I also looked into Marie’s X Chromosome but that was more difficult as MyHeritage does not include information on the X Chromosome.
  • I also improved Marie’s Key and changed some of the colors on Marie’s map

Butler DNA and Genealogy Update

Butler researchers Peter and Neil in England have been working hard to tie together the George and Edward Butler families. Both these families lived in Cincinnati for a while. There are genetic ties between the two families as well as circumstantial ties. Neil believes that he descends from (I think) either George’s father or grandfather. Below is my attempt to connect the two families:

I have Henry Butler’s family on the left and Michael Butler’s family on the right. I don’t know the father of either, so this is speculation based on DNA matches between the two families. The DNA matches between the two sides seem to support the tree above. The main family under Henry is George Butler who moved to Cincinnati from Wexford, Ireland. The main family under Michael is Edward (Henry) Butler. He lived in Cincinnati for a while but moved around a bit.

Circumstantial Connections Between the George Butler and Edward Henry Butler Families

Mary A Butler Born 1858 in Cincinnati

My wife’s ancestor Edward Henry Butler married Mary E Crowley in 1855 in St. John, New Brunswick. The family moved after that to Cincinnati. George Butler had a daughter Mary A Butler born in Cincinnati in 1858. She moved to St. John and married Thomas Joseph Murphy in 1878.

Murphy descendants match my wife’s Edward Henry Butler side of the family

Edward Butler, Son of Henry Born 1839 Wexford

Peter found this announcement:

Peter’s research shows Edward as the 7th child of Henry Butler:

Above, the newspaper funeral notice mentioned that the funeral was at 220 California Street, Newton. Here is the Newton Directory for 1993:

I had rejected that my wife’s ancestor Edward could have been a clerk as I thought that he could not read or write. However, he must have picked up reading and writing as the 1910 census says that he could do both.

The Veteran’s Census has Edward H in Newtonville in 1890:

There must have been a connection between Edward H Butler and Edward Butler of Wexford, in order for Edward H to host Edward’s funeral.

A New Holman Connection

Peter has been bringing the Cincinnati George Butler family forward:

Peter asked Neil and  me to check for DNA matches to Grogan, Holman and Middendorf. I was able to find a Holman match:

He matched with my wife’s Aunt Lorraine at AncestryDNA. He also matched my wife’s Aunt Virginia. This couple had three boys. The Holman above must descend from one of these three. Here is the new branch on the right:

More on the Holman/Butler Match

The match I will call Holman has shared matches with my wife’s two Aunts at Ancestry DNA. Here are the results:

The two in yellow are also matches to Neil’s nephew who tested at AncestryDNA. In the last column above EHB stands for Edward H Butler and GB stands for George Butler. These are the two lines that are connected by DNA.

It would make sense to do this same exercise with at least Patty and Michael.

I don’t think that gave me any more information. I looked at Michael’s shared matches and didn’t see anything helpful there either.

One Side Benefit

I did put a few close Butler DNA matches on the Butler DNA tree here:

I added in Deborah and Chester on the Alice Mary Butler line.

Summary and Conclusion

  • I like to summarize and conclude because while I’m blogging I sometimes get off the subject.
  • The main point here is to secure the connection between the George Butler and Edward H Butler Lines.
  • The other point is to secure Neil from England by DNA as he appears to be connected to the George Butler Line.
  • We were able to make DNA connections between Neil’s nephew who tested at AncestryDNA and one match confirmed to be on the George Butler Line. Neil’s nephew also matches another common match who has no listed tree.
  • Through Peter’s research a new George Butler descendant line has been found. A person from the Holman family was found to be a shared match across with Neil’s nephew and my wife’s Aunts. This further solidified the Butler family connections.



My Wife’s Theories of Relativity and DNA Painting

My wife Marie has Butler, Ellis, Lefevre and  Upshall. Butler is originally from Ireland, Ellis from PEI, LeFevre from Quebec and Upshall from Newfoundland. I have uplaoded Marie’s DNA results to MyHeritage. They have a utility called Theories of Relativity. This matches DNA with family trees. Once I get those connections, I can map Marie’s DNA using DNA Painter an online utility.

DNA Painter

I have already mapped quite a bit of Marie’s DNA here:

This shows that Marie is 30% painted or mapped. I’d like to improve this by looking at MyHeritage’s Theories.

Marie’s Theories of Relativity (TOR)

Marie’s top TOR is already mapped. That is Fred. Marie’s second TOR is Jo-Ann. Their common ancestors are Hopgood and Watson:

Marie and Jo-Ann match here:

I downloaded the details of this DNA match and entered them in at DNA Painter. I didn’t have these ancestors at DNA Painter, so I added them along with a new suggested color:

When I do this, I notice a potential problem:

This indicates that Jo-Ann’s match is bumping into Sarah’s match. That makes me suspect that I have mapped Sarah wrong. Sarah may have Hopgood/Watson ancestors also that I didn’t notice.

Another Look at Sarah

Sarah’s results are at AncestryDNA and Gedmatch. This is how I have Sarah at Ancestry:

That means that there is a mix-up somewhere. The reason I suspect it is on Sarah’s side is because the DNA match for Marie and Sarah is high for a 4th cousin once removed. I don’t want to try to fix this at this time, so I’ll just note the discrepancy. The problem is that one shared segment should represent one shared common ancestor. In this case it represents two.

Even with the overlaps, Jo-Ann brings up Marie’s mapped DNA to 31%.

Caroline and the LeFevre/Boure Line

Next is Wallace who I already mapped. Then Caroline. TOR shows a common ancestor with this couple:

This tree is also not without its problems. How could Charles Lefebvre be born in 1891 and have a daughter born 1870? This Ancestry Tree from Marie’s cousin has a Charles:

When I checked details on Caroline’s tree, it said that Charles was born before 1891. So I’ll say Caroline’s tree is OK. Caroline doesn’t add much new DNA, but doesn’t conflict with other DNA

Caroline overlaps with orange and pink but those are also LeFevre matches from more recent generations:

Pierre -Luc and an Older Pouliot Ancestor

Here is how MyHeritage shows the connection:

My suspicious side says that there could be other ancestral connections, but my lazy side says, put this in as is. Pierre-Luc’s DNA doesn’t bump into anyone that it shouldn’t bump into.

Pierre-Luc bumps into Joe and Patricia but they have common ancestors with Marie of LeFevre and Pouliot. That means that Joe and Patricia’s pink segments above Pierre are most likely Pouliot DNA. That means that if I wanted to get fancy, I could re-assign those two Joe and Patricia segments to Emma Pouliot. But I won’t.

The Problem with Daniel: Too Many Ancestors

Here is how MyHeritage shows Daniel:

But also like this:

DNA Painter may help figure out which DNA goes where. First, I’ll put Daniel in ambiguously:

Here are the hairs we are trying to split:

On Chromosome 2, we still can’t tell where Daniel belongs:


First I had to change Daniel’s color to green so he would show up better. In order to tell where Daniel belongs, we need an older match. The pink, orange and blue matches are too recent. That means that I entered Daniel correctly as Methot or LeFevre. For brevity, I left out the spouses. Sorry, spouses.

Daniel’s DNA matches with Marie were just under the limit of 7 cM, so they didn’t get painted:

Irma with PEI Ancestry

Matches on Chromosomes 2 and 3 will be too small to paint:

Painting this brings up more problems:

Here we have some bad overlaps between Ellis, Hopgood and MacArthur. One may be explained in that Irma has a different path to Ellis:

The Hopgood segment was one we just mapped from Jo-Ann – but with reservations.

Here is another path for Jo-Ann:

Here is a more likely, but slightly more distant relationship:


The Problem with Marie’s DNA Matches

Marie has four grandparents as do we all:

  • Ellis from PEI – Island genealogy and intermarriage, but the records are pretty good
  • Upshall from Newfoundland – More intermarriage like in PEI, but the records are not as good or missing
  • Butler from Ireland – No known intermarriage but very few relatives who have tested or posted genealogies
  • LeFevre from Quebec – Very good genealogies but a lot of intermarriage

Summary and Observations

  • Marie has confusing intermarriage issues on three out of four sides of her tree. This makes analyzing her genetic genealogy difficult
  • The further back the match is, the more possibility there is that the DNA could represent multiple sets of common ancestors
  • DNA Painter points out some of these issues. However, it is possible that DNA Painter could also sort out from which ancestors these DNA matches come from where there is more than one possibility.
  • I may come back to this later and try to sort this out.



Edward H Butler May Not Be the Son of Michael Butler and Margaret Croke?

In my previous Blog, I wrote an update on Butler DNA. In that Blog, I discussed a match between my wife’s Aunt Lorraine and Brian:

Brian shows up at AncestryDNA as a potential third cousin to Lorraine. That means that unless there is an unusual circumstance, my proposed DNA/genealogy chart cannot be right.

That chart shows Brian and Lorraine as 4th cousins once removed. However, reported data indicates that sharing 147 cM of DNA is outside the range of 4C1R possibilities, but is within the realm of likelihood (albeit on the high end) for a 3C1R:

As we have a birth record for George Butler, but not for Edward H Butler, that suggests that Edward H (likely Henry) Butler could have been the son of Henry Butler and Ann Russel.

Playing With the Butler Family Tree

There, I just made Lorraine and Brian third cousins, once removed. However, Edward H just lost MIchael Butler as his father. Here is the family tree that Butler researcher Peter has:

HENRY1 BUTLER was born in 1800 in Wexford. He married Ann Russel on 02 Jul 1824 in Wexford.

She was born in 1800 in Wexford.

Henry Butler and Ann Russel had the following children:

i. GEORGE2 BUTLER was born on 03 Oct 1826 in Wexford, Ireland. He died on 23 Dec 1890 in Hamilton County, Ohio. He married (1) MARY WHITTY, daughter of Richard Whitty and Margaret, on 12 Jun 1849 in Mountain Gate Rathangan Wexford. She was born on 03 Dec 1824 in Rathangan Wexford, Ireland. She died on 11 Jan 1865 in Hamilton County, Ohio. He married (2) MARGARET SINNOTT on 11 Nov 1868 in All Saints Catholic Church Cincinatti. She was born in 1845 in wexford Ireland. She died in 1887 in Hamilton County, Ohio.

ii. NICHOLAS BUTLER was born on 23 Apr 1828 in Wexford. He married Christina Lambert on 06 Oct 1848 in Wexford Ireland. She was born in Wexford.

iii. ELIZA BUTLER was born on 06 Mar 1830 in Wexford.

HENRY BUTLER was born on 29 Mar 1832 in Wexford.

MARY ANNE BUTLER was born on 23 Mar 1833 in Wexford.

BRIDGET BUTLER was born on 15 Jul 1836 in Wexford.

vii. ADAM BUTLER was born on 24 Jul 1839 in Wexford. I think this is a Baptism date

Peter also found another daughter for this family born in 1842:

Peter notes that George’s other name also appeared as Adam on his baptismal record and that the Adam born in 1839 went by Edward. There are a few reasons why Edward H may have been part of the Henry Butler/Ann Russell family:

  • The large DNA match between Lorraine and Brian
  • The fact that Edward did not name any of his children after Michael nor after Margaret
  • Edward did name his first son George and lived in Cincinnati for several years where the (presumably) elder George Butler lived.
  • The middle name of Henry which was carried down could have been for the Henry Butler born about 1800 in the genealogy above.
  • There has been no birth record found for Edward as son of Michael (nor as son of Henry).
  • Most records for Edward’s sons list his father born in Kilkenny. However, one record lists his father as being from Wexford.

Under my scenario, Edward is born to Henry in Wexford but is adopted by Michael Butler of Poulrone, Killkenny for some reason.

James Butler Born About 1823 Kilkenny, IRE

I have noted before that it appears that Edward was a brother to a James Butler. This was inferred from the US Census of 1910:

Here Mary Mellie is actually Mary Butler, the daughter of James Butler and Mary Quinn. She was born in 1858, though she would like us to think that she was born in 1880. That is quite a difference. Joanna, born in 1860 would have liked to have us think she was born in 1877. Edward Butler was clearly not Mary’s father. That lead me to believe that Edward and James could have been brothers and Edward was acting as Mary’s father. If Edward gave his correct age, he would have been born about 1832 (see below).

However, I note that this James Butler’s eldest son was Michael Butler:

Perhaps this James actually was a son of Michael Butler. James’ son Matthew could have been named for his brother.

Here is Mathias or Matthew Butler who is recorded as a son of Michael Butler and Margaret Croak:

So under this scenario (likely or not), Edward goes to live with and Uncle and considers the Uncle and Aunt his parents. He also considers James and Matthew his brothers.

So Where Does Edward H Butler Fit In?

Peter recently discovered  Naturalization papers for an Edward Butler from Poulrone:

In those papers, he gives his birth year as 1825. However, if he is the same as Henry born in Wexford in 1832, that would fit in better with other recorded birth years for Edward:

If Henry was named for his father, it would make sense that he might go by a different name to avoid confusion. One problem with the timing is that if Edward came to the US in 1843, he would have been only 11. That leaves a few possibilities:

  1. Edward was actually born 17 August 1825. To do that, he would have been squeezed in between his parents’ marriage and the birth of George in 1826.
  2. Edward was born in 1832, came to the US, say in 1848 when he was about 16 and was naturalized at about age 18. That would results in several fibs as his witnesses were supposed to have known him for 5 years before his naturalization.

More DNA Analysis

As I mentioned above, the DNA match between Lorraine and Brian has lead me to the above scenario. Here is an AutoCluster I performed for Lorraine earlier in the year:

These clusters are difficult to see. Clusters are where you match people in a group and that group of people match each other for the most part. Lorraine is part French Canadian and part Irish. The French Canadian took over for the most part. Lorraine’s Irish clusters are in the bottom right. Brian is in Cluster 35 which has 7 members.

Here is a closeup of the Clusters 33-36:


Brian is the second person in Cluster 35. He matches Barbara, Donna, Patty and Kim:

That means that in Cluster 35, there are three descendants of Edward H and three descendants of George Butler. I don’t know if the fact that there are three people on each side make the proposed tree more likely or not. To me, it suggests that it could be more likely.

Cluster 34

Brian also matches two people in Cluster 34. That is a Kerivan Cluster, but Edward H’s son Edward Henry born 1875 married a Kerivan, so that explains the connection. Brian is matching on the Butler side of the Kerivan Cluster. Brian matches Amanda and Dawn in Cluster 34. Unfortunately, Amanda and Dawn either have no tree or not enough to figure out their ancestors.

Analyzing the Size of DNA Matches

There are two ways to do this. One is by the Chart I showed at the top of the Blog. It is possible to analyze the possibilities at Ancestry also. Here is how AncestryDNA shows Brian and Lorraine:

Their most likely relationship given the match would be 2nd cousin once removed. However, they have a 1% chance of being 3C1R. I’ll put that into a chart:

I’ll be looking at my wife’s Aunt Lorraine and her sister Virginia as I manage their DNA results. Here is how the results look:

Were it not for the match between Lorraine and Brian, I would say that the results are inconclusive. However, there appears to be no chance that Brian and Lorraine could be fourth cousins once removed.

Looking at Gedmatch

I can also look at Gedmatch, however, there is not a one-to-one correlation between AncestryDNA and Gedmatch. Here is how the match between Lorraine and Kim looks at Gedmatch:

Gedmatch expresses it’s guesses in generations. So a 4.5 means generations to a common ancestor. That would be equivalent to a third cousin once removed. The green above represents the new chart that I have with George Butler and Edward H Butler as brothers. The yellow represents George Butler and Edward H Butler as first cousins.

Here is some more data, though, again, I don’t see any clear conclusions:

Summary and Conclusions

  • A high DNA match between Lorraine and Brian in conjunction with lack of some key birth records has lead me to believe that George Butler born 1826 and my wife’s ancestor Edward H Butler could be brothers.
  • That would mean that Edward H Butler’s parents would be Henry Butler and Ann Russell.
  • The why would tradition and records on the Edward H Butler line have his parents as Michael Butler and Margaret Croke? One possibility is that Michael and Margaret were Edward’s Uncle and Aunt and that he went to live with them and considered them to be his parents. Perhaps MIchael needed help on his farm or Edward had some falling out with his family.
  • This should lead to a renewed interest to find out more about the Henry Butler and Michael Butler families.
  • I would be open to other interpretations, but due to a high DNA match which indicates that George Butler and Henry H Butlers should be brothers, no other clear possibilities come to mind right now.
  • I still surmise that Henry Butler and Michael Butler could be brothers. I have a birth record for Matthew, son of Michael and my assumption is that Michael Butler had a son named James. James had several children. One of his daughters was Mary who married a Mellie and housed Edward H Butler in Newton, Massachusetts according to the 1910 Census.





AutoClustering My Wife’s Aunt Lorraine’s AncestryDNA Results

AutoClustering is working well. I have previously run an autocluster report for Lorraine’s sister Virginia:

Here are some comparisons:

Virginia’s number of 4th cousins or closer and her SAHs are as of today and I did her autocluster about a month ago. I changed the upper limit for Lorraine to 600 cM because I was having trouble identifying some of the clusters. I had set the lower limit down to 12 because I was looking for distant Butler relatives.

Lorraine’s AutoCluster

Since the time I ran Virginia’s autocluster, the clusters have been arranged differently to show connections between the clusters. This has been a very helpful innovation.

Adding Names to Lorraine’s Clusters

I’ll start with a table:

This table starts with each of Lorraine’s clusters. That is followed by the top match name in the cluster and the amount that top match has in cMs. I just need to fill in which grandparent side each cluster belongs to and which common ancestors the cluster seems to point to,

Lorraine’s Ancestors

These are some of the ancestors that I will pick from:

I am interested mostly in the top part of the tree. The bottom part is where most of the matches will be. The bottom represents the maternal French Canadian side.

Name Those Clusters

To get the ball rolling, I’ll start with Fred. I have have been in touch with Fred who a second cousin on Lorraine’s Pouliot maternal grandparent side:

Turns out that is Lorraine’s largest Cluster:

That’s a lot of Pouliot’s. These could be all descended from a certain common ancestor along the Pouliot or Fortin Lines.

The Second Largest Cluster: LeFevre

Sandra shows up a lot in my analyses. Here she is:

Sandra is also in Lorraine’s Cluster 1:

Skipping Down to Clusters 34 and 35: Kerivan and Butler

These are the Clusters I am more interested in.

Clusters 34 and 35 are the purple and tan Clusters. They show a lot of connections between those two Clusters.

Cluster 34 – Kerivan

Amanda is the first person in Cluster 34, but she has no tree. Donna is the third match in Cluster 34. Here is the paternal side of her tree:

Turns out Donna is a second cousin to Lorraine also:

Cluster 35 – Butler

The top match for Lorraine in her Cluster 35 is Barbara. Barbara has a short tree:

Here is Barbara in a tree with other Butlers:



She shows up as Lorraine’s 2nd cousin. What is interesting about Cluster 35 is that it includes Butlers from Cincinnati. My guess is that they are related this way:

There is a branch on the left of Cincinnati Butlers headed by a George Butler born about 1826. My wife’s ancestor Edward Butler was also living in Cincinnati for a while. His first son was named George – perhaps after the Cincinnati Georg Butler. I haven’t worked out all the details yet, but the DNA is showing a definite connection.

Lorraine’s Cluster Summary

Here are the bones of Lorraine’s clusters:

It is possible that there are 33 French Canadian Clusters and 3 Irish Clusters. I would have to look at all the clusters to be sure. However, as I scan the clusters, it looks like that could be the case. Here is my best guess:

That means that finding the 1/2 Irish side among the French Canadian half, is like looking for a needle in a haystack.

Comparing Lorraine’s Clusters to Virginia’s Clusters

Here is a comparison of the two sisters’ clusters:

This shows that Virginia split in two both of Lorraine’s Clusters 34 and 35. Here are some of the clusters that I tried to identify for Virginia:

So with that comparison and looking at some of Lorraine’s Shared Ancestor HInts at AncestryNDA give me this cluster chart for Lorraine:

It is possible that Cluster 16 is wrong based on the placement within Pouliot’s.

Summary and Conclusions

  • Lorraine’s AutoCluster Chart looked like a mess at first but seemed to sort out between her four grandparents.
  • I didn’t look at why there were so many matches between the Kerivan and Butler Lines.
  • I compared Lorraine’s Clusters to her sister Virginia’s Clusters
  • The new ordering of clusters makes a lot of sense and makes the identification and organization of clusters much clearer.