Author: joel@jmhartley.com

Posted on

Comparing Grant, Stuart and Frazer by YDNA

In my previous Blog, I looked at Frazer STRs and came up with a STR Tree. It is a bit of a mess, but it tries to show where the STR mutations are:

I like one explanation I read about STRs. STRs are short for Short Tandem Repeats. This excerpt is from an FTDNA article called Understanding YDNA Matches:

Our bodies work as copy machines when it comes to the Y-DNA. You can have a copy machine doing 1,000 copies without a problem, and then, the 1,001 copy may have an “o” that looks more like an “e”. And when we use this copy to make additional ones, all the new ones will now have an “e” instead of an “o”. This is a simple way to explain how mutations occur in our Y-DNA when it’s transferred (copied) from father to son. Mutations don’t happen frequently, on the contrary, very seldom, but they can happen randomly in time, which means that I could be one mutation off of my father. That is why all those matches or close matches on 12 markers will in most of the cases go away when they happen between different surnames, and we increased the numbered of markers that are compared: more mutations showing up, which means way back in time when the common ancestor lived.

Ancestral and Derived – What’s the Big Deal?

STRs are either ancestral or derived. That means that they are older or newer. Older or newer is important if you are trying to figure out timelines. It is also important in creating trees and figuring out who belongs in which branch of the tree.

Grant YDNA

In my previous Blog, I noted that the Frazer marker of DYS710 with a value of 34 was probably ancestral:

Assuming that to be the case, that meant that DYS710 = 33 would define the James Line of the Frazers above.

I wrote to Grant to get the STR values from his Grant BigY test. However, it appears that the Grant STR value for DYS710 is missing along with the STRs from 68-111. I don’t know if this is a mistake by FTDNA or not:

Grant Matches Paul by BigY but Not by STRs

Another surprise is that Grant matches one Frazer by the BigY and not by STR matching. Here is my cousin Paul’s matches by BigY:

Here are Paul’s 67 STR matches:

Here Paul matches a Grant, but not the same one who took the BigY test. Note that the Grant above doesn’t show that he had Big Y tested.

How Does Grant Match Jonathan?

Jonathan had fewer STR mutations, so perhaps he matches Grant. Below, we see Jonathan matches many Grants, including the one who took the BigY test:

There is one interesting thing to note here. Look at Jonathan’s match with Stuart who took the big Y. At 67 markers, there is a Genetic Difference (GD) of 3. With Grant the GD is twice as much at 6. However, when the BigY markers are added in, the differences between Jonathan and Stuart are 12 and only 11 with our BigY Grant.

Comparing Grants Extra 435 STRs with Jonathan

Thanks to the Grant family sending me the Grant BigY STRs, I can now compare them. In my previous Blog, I noted these differences between the Frazer Archibald Line and the James Line:

Paul is from the Archibald Line and Jonathan and Rodney are both from the James Line. Grant could be the tie-breaker to tell which values are older for these markers.

FTY299

What do you think? Based on the above, and knowing nothing else, I would say that 12 is ancestral and 13 is derived. That puts FTY299 as a marker for the Archibald Line. I’ll add that to my STR tree after I look at the other two markers.

DYS523 and Good Old FTY269

 

Here we go again. The mutations seem to be skewed more toward the Archibald Line. However, recall that these are mutations that Paul has, so they could have taken place any time between Archibald in 1715 and Paul:

If Rick had taken the BigY test, we could have narrowed that down a bit.

A New BigY 500 STR Tree

With this new information, I can build a New BigY STR Tree:

Here I wrote Paul’s results a little differently as I didn’t have room at the top. The first number is ancestral and should apply to Archibald Frazer born around 1690. The second is Paul’s value. This mutation could have happened between about 1715 and when Paul was born.

Any Other Differences?

It seems that there must be. Jonathan had 5 additional differences in the BigY STRs. However, these may be differences between Grant and Frazer in general.

Here in DYS514, we see a difference between Frazer and Grant. We can’t tell which one is older, because we need a tie-breaker.

Parallel Mutation or Back-Mutation

Let’s consider DYS516. I had said that the value of 18 for STR DYS516 represented the overall Frazer line because it was shared by Paul and Jonathan whose common ancestor was the original known Irish Frazer from about 1690. However, Grant has a value of 17 shared with Rodney. In other cases, we had considered the value to be shared by Grant and Frazer to be the older one. What happened? This is a case of back mutation or parallel mutation. It could be that the old value shared by Grant and Frazer was 17. Then sometime before 1690 it went up to 18 for Frazer. Then Rodney’s branch went back down to 17. That would be a back mutation. For a parallel mutation, the original value that Grant and Frazer shared would have been 18. Then at some time Grant mutated down to 17. In an unrelated (or parallel) way, Rodney’s branch also went down to 17. By getting more information on other people’s BigY results, it might be possible to figure out which happened.

There is a less likely scenario where 17 would be ancestral. That would have to mean that Jonathan and Paul had independent or parallel mutations. As this would have happened over a shorter period of time, it is less likely that this happened.

DYS542

Here is another of Jonathan’s 5 of 435 STR mismatches with Grant:

Note that Grant and Frazer have a difference of two. Again, we don’t know if 12, 13 or 14 was the ancestral value. It could be 13. In that case Frazer would have mutated up and Grant mutated down.

FTY289

This should be the 4th out of 5 differences between Jonathan and Grant.

FTY510

This should be the 5th difference or GD.

Grant Vs Rodney

I would suspect that Grant and Rodney’s results should be similar to Grant and Jonathan’s results.

They are the same except the number of BigY STR results tested are slightly lower.

FTDNA Block Tree Vs YFull’s YTree

Both FTDNA and YFull have SNP trees. Sometime one gets ahead of the other. Here is the FTDNA Block tree from Jonathan’s perspective:

Jonathan matches two Frazers at YP6489. Further out he matchs Grant at YP6488. At at more distant level he matches Hayes at YP6479. That doesn’t mean that Frazer descends from Grant and Stuart who descends from Hayes. It just means that these families descend from a common ancestor.

Another point to note is that the Frazer SNP of R-YP6489 is probably a family SNP. That means that this SNP probably applies to just our Frazer branch. YP3189 is one step below YP6488. That means that it is more recent. YP6488 is older and represents Frazer, Grant and Stuart. Because there is more than one family that shares this SNP, it would not be considered a family or surname SNP.

Here is the YTree version:

R-BY26344 Is for Grant and Stuart

This tree goes back one more level and includes dates. Then there is one important detail I missed previously. It appears that Grant and Stuart have a new SNP called R-BY26344. YFull uses IDs, but it appears that those IDs are for Grant and Stuart. This is the part where YFull’s YTree goes ahead of FTDNA.

My interpretation of the YFull YTree above:

Again, this does not show that Stuart, Grant and Frazer descend from Hayes, but that all four surnames descend from a common ancestor born around 900 AD. This shows that the tested Stuart, Grant and Frazer had a common ancestor from around 1300 AD. I assume at that time, surnames may not have been settled or commonly used in Scotland. According to scottish-at-heart.com:

The use of  ‘fixed’ (or recognized) Scottish surnames appeared occasionally as early as the 10th or 12th centuries, but didn’t begin to be used with any sort of consistency until the 16th century.

Even this, this practice was slow to ‘catch on’, and it took until the late 18th and early 19th century to spread to the Highlands and northern isles.

A New Terminal SNP for Grant and Stuart and a New Mystery

In my previous Blog, I had missed that Grant and Stuart have a new Terminal SNP. They are now BY26344. However, YFull gives a common ancestor for Grant and Stuart at 225 years before present. I roughly called that 1775. The problem is that if Stuart and Grant had a common ancestor in 1775, was it Grant or Stuart? Also, it appears that the Grant and Stuart genealogies don’t match up as to where these families lived at that time.

Grant and Stuart Genealogies

This is an area where I have very little knowledge. I have that the Grant BigY tester’s earliest verifiable ancestor was James GRANT “of Carron”, 1728 – 1790. From a quick Google search, this appears to be Carron:

The Stuart tester has this information:

Charles Stewart/Stuart b. abt 1695 d. 1753 Virginia

This is the problem. If Stuart and Grant had a common ancestor around 1775, then how could Stewart have been in Virginia in the early 1700’s and Grant been in Scotland in the early 1700’s? This suggests that one of the genealogies is wrong or that the common ancestor dating is wrong. The closest reconciliation that I could make up is that Charles Stewart was actually Charles Grant. He had James Grant in Scotland, then came to Virginia and changed his name. I feel uncomfortable making wild guesses for others’ genealogies that I know little about, so I will not go further in this direction.

TMRCA By YFull and FTDNA

What if YFull is wrong with their date of most recent common ancestor? Here is the 6488 Branch:

I don’t totally understand YFull’s dating. However, the reasoning is that the more SNPs in your branch, the older your branch. That makes sense to me. For example, the Frazer R-YP6489 includes 6 other SNPs. The Grant/Stuart branch of R-BY26344 has only two other SNPs. It stands to reason that BY26344 would have a more recent common ancestor than YP6489. The Frazer date checks out well, but three people tested. We think our common ancestor was born in the vicinity of 1690 which is about 329 years ago. YFull gives 375 years as a date. That seems pretty close to me.

A scroll-over of the 375 years before present for the Frazer common ancestor shows this:

This gives a pretty wide margin of error.

For BY26344, there is a larger margin of error:

I assume that the reason is that only two people tested for BY26344.

FTDNA is not as helpful with dates. In addition, FTDNA does not have Stuart and Grant as BY26344. Perhaps if they update their tree, they will and there may be a way to estimate a common ancestor then. However, having said that, the a YDNA project administrator has made this prediction for Grant and Stuart in an email to Stuart:

You had 19 Unnamed variants and now you have 19 – 2 = 17 left over. 

Out of these 17 SNP’s there was 1 SNP which you share with Grant.

So both of you create a new Subclade BY26344 downstream YP6488

(btw. FTDNA has not identified this new subclade yet)

Now my best estimate for your MRCA with Grant is about 1200 AD. 

This is interesting because this is an earlier date than what YFull has. The discrepancy may be due to the fact that Stuart did the Big Y 700 and other testers have only done the Big Y 500. Also, I don’t think that Stuart has uploaded his Big Y 700 results to YFull. It will be interesting to see if that makes a difference with YFull’s common ancestor calculations.

The FTDNA administrator further writes to Stuart:

Up to now you are the only one in subclade YP432 with a BigY-700, thus we will only know where these SNP’s are exactly located when they are willing to upgrade from BigY-500 to BigY-700.

Summary and Conclusions

  • Grant and Stuart are the closest YDNA matches to the Frazer family
  • I used some of the Grant and Stuart STR results to find out which of the STR vaules were older or newer for the two Frazer Lines.
  • Grant and Stuart are in a new Branch of R-BY26344
  • There are discrepancies for the date of the common ancestor between Grant and Stuart. These calculations were done by YFull and an FTDNA administrator. These dates may be fine tuned by Stuart adding his Big Y 700 results to YFull and/or by others in the YP432 Group doing Big Y 700 testing.

 

 

Posted on

Frazer YDNA STRs Compared

Blaine Bettinger recently has started a project comparing STR differences to known relationships. It turns out in our Frazers of Roscommon DNA Project, we have four people who have tested where we know their relationships. Here are those relationships:

Rick has tested 67 STRs and Paul, Rodney and Jonathan have tested for the BigY 500. Here is how the numbers look:

Generally, what we see is that going from top to bottom, the further out the relationship is, the more differences there are in the STRs. The numbers between 0-5 above are the genetic differences. Then as the test gets higher there should be more differences going from left to right as more STRs are being tested.

Paul’s Odd Results

However, Paul’s number goes down when he is compared to Jonathan between the 111 STR test and the BigY 500 test. Why is that? My assumption is that the Big Y test missed one of the STRs tested in the 111 STR test. It should be possible to find this STR looking at the results.

DYS710

Checking Paul’s result, he has a different result from Rodney and Jonathan for marker DYS710. DYS710 is the first marker tested in the 111 STR test. Paul has 34 and Rodney and Jonathan have 33.

Here is Paul’s BigY STR result for DYS710:

Here is Jonathan’s result:

So that is not the answer. I can’t explain how FTDNA came up with a lower genetic distance for the BigY 500 STR results.

A Frazer STR Tree

Over a year ago, I came up with this tree:

This shows the genetic differences. It basically shows that there is a difference of three between the two Frazer lines of Archibald and James. In general, the two lines differ over DYS391 and CDY. At some point between Archibald Frazer of 1720 and James Frazer of 1804, there were two mutations that defined the Archibald Line from the original ancestral Frazer DNA. Then under the James branch of 1804, there are two branches. Rick’s branch is defined by DYS444 and Paul’s branch is defined by DYS576 being 19. However, we don’t know when these mutations occured. For Rick’s branch it could be anyone starting at Richard Patterson Frazer down to Rick. For Paul’s branch it could be anyone between Paul and George Frazer.

An Updated Frazer STR Tree with DYS710

However, the above Frazer STR tree doesn’t explain all the differences. Paul and Rodney have a difference of 5 STRs at 111 markers. I need to add in DYS710. Recall that Paul had a value of 34 for DYS710 and Rodney and Jonatham had 33. But which is ancestral? Or which came first?

The last time I looked for ancestral STRs, I looked at the Frazers more distant relatives: the Grants and Stuarts:

Note that their SNP is R-YP6488. FTDNA now has a block tree:

This block tree was taken from Jonathan’s perspective. However, it shows that R-YP6488 represented by matches with Grant and Stuart is an older SNP. Hayes is from an even older group represented by SNP R-YP6479. I had originally thought that 33 was the older STR for DYS710 but Stuart has a value of 34. Hayes would make a good tie-breaker but he only tested out to 67 STRs.

Checking YFull

Perhaps Hayes has uploaded his results to YFull.

Under Paul’s distant STR matches I see someone who shows as R-YP6479. This is likely Hayes. However, it doesn’t tell me what his value was for DYS710.

Back to the Updated Frazer STR Tree

Assuming that there were no parallel mutations, I’ll try this:

This shows a value of 34 for DYS710 as ancestral. Then on the James line, somewhere between James and Thomas Henry Frazer DYS710 changed from 34 to 33. That means that at least the Thomas Henry Frazer branch is characterized by a value of 33 for for DYS710.

In order to check my tree, I look at Paul and Rodney:

They have a Genetic DIstance (GD) of 5 at 111 STRs. Looking at the tree, we see that the STR differences add up to 5.

  • The Archibald Line has 381 = 11 and CDY = 35-40
  • The George McMaster branch has 576 = 19
  • The James Line has DYS710 = 33
  • The William Frazer branch has 552 = 24

The Mystery of Paul and Jonathan Solved

I posted my question to the Genetic Genealogy – Tips & Techniques Facebook Page. Skip tells me that the 3 of 425 differences are additional differences. That adds up to 7 of 536 STRs. That leads to this observation. Rodney and Jonathan have a GD of 2 of 427. Paul has a GD o f 3 of 425 with Jonathan and 5 of 415 with Rodney. That seems to indicate, if my logic is right, that the extra two mutations are on Rodney’s side.

A BIgY 500 Frazer STR Tree

This leads to another change in the Frazer STR tree. In order to find the GDs between Rodney and Jonathan, I downloaded all the results. These appear to be the two extra differences:

These were on Lines 425 and 475 of my Excel Spreadsheet. This shows that the mutations belong to Rodney. The mismatch column was just to point out any differences between Jonathan’s and Rodney’s STR results.

Here I ran out of room to describe Rodney’s branch at the top of his branch, so the extra descriptors went on the bottom. Keep in mind that STRs can mutate up or down in number. These last two mutations that Rodney had went down in number. In fact, it appears that all the STR mutations on the James Line are going down.

Any Other STR Differences?

Yes. There are a few more differences between the Archibald and James Branches of the Frazers:

Here, Paul represents the Archibald Line and Jonathan and Rodney represent the James Line. Unfortunately, as the newer STR results are not posted on-line, I would not know how to figure out which values are ancestral and which are the newer values.

The Big Picture: SNPs and Haplogroups

The Big Picture is that our Frazers have the haplogroup of R1a:

Most Frazers in general are R1b and not even closely related by DNA. However, before Frazers became Frazers, some of our Scandinavian R1a’s made their way to Scotland and became Frazers when the R1b Frazers were also becoming Frazers.

The R1a Tree

Here is an outdated R1a Tree. It is still nice as it has images on it:

Here is another shot of our Frazer Block Tree:

The trick is connecting the two trees. The Block tree listing at the top mentions R-M198. The “picture’ tree has that at about 6,000 BC, so that is a start.

R-M417 shows as 4800 BC on the picture tree. Next is R-CTS4385:

This shows that before our Frazers were Scots, they were Germanic. Next were the L664 Group:

Our L664 ancestors entered the scene about around 3,000 BC. They still have plenty of time to make it to Scotland. They could have hopped over to England around then, but likely made their way up to Western Scandinavia first.

From there, our ancestors kept branching as families do:

That brings us down to the bottom of the picture chart. S2880. S2880 is right above R-YP432 where the Frazer Block Chart starts:

This shows flags. My interpretation is that during the time of R-YP432 our ancestors were living in Scandinavia. The flag on the right that connects to YP432 is the Swedish flag.

The YFull YTree dates YP432 at 3100 years before present.

Between looking at the Block Tree, the YTree and the R1a Administrator’s Tree, it could be that our ancestors could have made their way to Scotland around the time of Christ. At the time the Frazer Clan was formed, our ancestors were in the area of what is now Inverness, Scotland.

Summary and Conclusions

  • Getting data to a project which compares YDNA STR matches to known relationships gave me a chance to look at our Frazer YDNA STR matches.
  • This lead to a better understanding of what the BigY 500 STR results mean.
  • For the four people who have taken the YDNA test and especially the three Frazers who have taken the Big Y test, I was able to refine the YDNA differences between the different lines and branches.
  • After that I gave a rough overview of how our Frazer ancestors made their way to Scotland from Scandinavia perhaps around the time of Christ.
  • Based on the YTree and other DNA sources, our own Frazer branch of R-YP6489 began around the early 1300’s, but our common ancestor in that branch was from the early 1600’s.

 

 

Posted on

Updates to Butler YDNA

I see my last Blog on Butler YDNA was about a year ago.  A little or a lot can happen in a year. Turns out some things have happened.

Three Big Y Testers: 2 Butlers and a Batt

There are three Big Y testers that show that they have a common ancestor in the past 700 years. Here is the simplest way to show how these three are related:

Note that while Batt’s SNP is I-Y238315, I didn’t put Batt in that box. That is because Batt and Butler both descend from I-128315. In other words, Butler does not descend from Batt, but both names descend from a common ancestor. This shows that the three testers have a common ancestor who was born 700 years ago. This also shows that the two Butlers who have tested for the Big Y have a common (likely Butler) ancestor from about 225 years ago. Here is how the YTree shows it:

These are the dates that I didn’t have as of my previous Blog. Previously, we were just up to 700 years ago. This new information thanks to the new Big Y Butler tester brings us 475 years closer to the present date.

However, there is a catch to these dates.

SNPs Vs. STRs

The above trees are based on SNPs. FTDNA also tests for STRs. The STRs show that Batt is more closely related to my late father in law Richard Butler than to the other Butler. So which is right? Whenever there is a conflict between SNPs and STRs, SNPs are always right.

Here is how my father in law Richard matches Batt and the other Butler by STRs:

Based on STRs, it looks like Batt is a closer match than Butler. However, by SNPs my father in law matches Batt 700 years ago and Butler 225 years ago. This is due to the variability of STRs or back mutations. Also the Butler above could have had more STR mutations than usual and/or the 225 year estimate for common Butler ancestors could be on the low side

A New Butler STR Match

I was notified recently that there was a new Bulter 37 STR match. That is good news. Also this Butler has ancestors from County Wexford. Peter, who has been researching Butlers feels that Wexford is a likely origin for the Butlers. Here is my father in law’s match list at the 37 STR level:

The new Butler match at the top is a perfect match with my father in law at the 37 level. However, that could change as more STRs are tested. My guess is that the three Butlers on the list are I-Y128364 and the non-Butlers on the list are not. Also notice that the first two Butlers matching my father in law have Wexford roots.

Here is a partial map of Wexford:

I believe that my father in law’s family came from County Kilkenny just North of Waterford. However the distance from that area to Wexford is not far. At any rate, this new YDNA Butler match seems to suggest that a common Butler ancestor could be from Wexford.

What’s Next?

The new Butler match may want to take an Autosomal DNA test. This could be done at FTDNA or at AncestryDNA. The advantage of AncestryDNA is that they have more people who have taken that test. That means a bigger chance of finding matches. The DNA matches will be looking for genealogical connections. The new match could do further YDNA testing, but that is a bit more expensive than the autosomal DNA test.

 

 

Posted on

Penny’s Frazer ThruLines

My last Blog on Penny is here. That was prior to ThruLines. I also wrote a Blog on Penny’s sister Toni here. Toni’s DNA is not linked to her tree, so she does not have ThruLines.

Penny’s Genealogy

Penny is in red in the lower left. She descends for William Fitzgerald Frazer. This tree is based on research by Frazer researcher, Joanna.

William Fitzgerald Frazer Born 1821 Ireland

I am interested in William Fitzgerald Frazer:

Penny mentions a biography in her Ancestry Tree:

This was from An illustrated history of Sacramento County, California. Containing a history of Sacramento County from the earliest period of its occupancy to the present time published in 1890. Penny further gives this chronology for William:

This raises questions as to the mother of the first three children. Penny also mentions another daughter Angelina G Frazer as mentioned in William’s will. She is named as Angeline G Law. That gives a hint:

In 1911, Angelina is listed as Gracella. I assume that was her middle name:

This puts William Fitzgerald in Leitrim in the 1840’s. Here is another interesting record:

In this record Angelina “Gracenda” is listed as Augustine. Here is Angelina’s wedding publication from 1872 – not long after her father’s divorce:

This more specifically puts William in Drumkeerin:

That means that William Fitzgerald Frazer lived in Drumkeeran. He had Angelina Gracenda around 1843. His second daughter appears to have been Mary Ann Frazer. Penny has her born in 1844.

Mary Ann Frazer

Mary Ann married in 1864:

About a year later, these two were living in Poughkeepsie, New York:

Thomas Quinton may be William’s father. Margaret Frazier is Mary A’s sister. Poughkeepsie is on the Hudson River:

How Can the DNA Help Us?

Right now on the James side of the Frazer Tree, descendants have been placed into two main branches:

There is an Archibald Branch on the left where Penny is. There is a Michael Branch on the right. There is an Elizabeth branch in the middle also. However, her daughter Catherine Knott married Archibald believed to be the son of Michael on the right side. We can use the DNA matches to test the genealogies above. First, I try to see if the DNA supports that the people within the box on the left match each other. Then I see if they match the people in the box on the right at the appropriate levels.

Penny’s ThruLines

Penny’s first cousin level ThruLines are good:

89% of the time Penelope’s match with Pamela should be a 1st cousin once removed or similar relationship:

Penny’s ThruLines with William Fitzgerald Frazer

At Penny’s Frazer great grandparent level there are no new ThruLines. There are some at the 2nd great-grandparent level:

These matches are from the descendants of Mary Ann Frazer. I’ll start a chart for Penn y:

Penny and Frazer Third Great-grandfather ThruLines

This part goes beyond Penny’s Ancestry Tree to work that Frazer reasearcher Joanna has done:

Here there are 7 matches to Penny on the Thomas Henry Frazer Line. Joanna has that Thomas Henry Frazer was born in Drumkeerin, so there is that connection. Joanna is on the Edward F Frazer Branch. I should point out that Penny has no matches on the Edward Wynn Branch:

Edward Wynn is the brother in the middle group above.

Penny’s Distant ThruLines

When I put the common ancestor up two more generations to James Frazer born about 1720, I pick up two additional matches for Penny:

These two matches were recently added under Michael to explain DNA matches. It would make sense that Penny could match these two, as the theory is that Mary married William Frazer:

Adding Penny’s ThruLines to Madeline’s and Charlotte’s

Although Penny doesn’t share any ThruLine matches with Madeline and Charlotte, they both match some of the same matches as shown in blue below:

I think that Charlotte’s match to JS above is really Janet. The question is: if Penny, Charlotte and Madeline have many of the same ThruLines, then why doesn’t Penny match Charlotte and Madeline? I suppose there could be many explanations:

  • The common ancestor is Archibald born 1792. Charlotte and Madeline may be related on the Archibald side and Penny may be related on Archibald’s wife’s side. That would result in Penny not matching Charlotte and Madeline.
  • Another variation of that is that Archibald may have had more than one wife. Penny may descend from one wife and Charlotte and Madeline from another.
  • It may be that we have the genealogy wrong.
  • There may be other connections on collateral lines.

Adding In Joanna

If I add in Joanna to the mix, I’ll have representatives from the three brothers:

 

These are William Fitzgerald, Edward Wynn Frazer and Thomas Henry Frazer (Joanna’s great-grandfather). Here is a partial list of Joanna’s ThruLines:

I’m more interested in the matches at the third cousin level. Here we have the two brothers plus Ann. The third brother doesn’t show here as he shows at the 2nd cousin level for Joanna. For some reason, the ThruLines seemed to skip a generationbetween William Fitzgerald 1821 and Minnie Jane born 1895. I have written quite a bit about the McPartland Branch. They have the Ancestor Ann Frazer who has been difficult to place. Joanna has made a guess that she was the daughter of her 2nd great-grandfather Archibald born 1792. I have been going back and forth as to whether Ann should be under Archibald or Michael Frazer.

Another interesting point is that Edward Archibald born 1867 ended up in Poughkeepsie, NY where Mary A Frazer Quinton and Margaret Jane Frazer lived in 1855. I don’t know if this is a coincidence or not.

I see that Joanna matches WG. WG is on the Edward Wynn Branch where Madeline and Charlotte are. Joanna has a good match with WG but Madeline and Charlotte do not. This may be a coincidence, but I would keep my eye on this situation. For Charlotte and Madeline the level that they match WG indicates a second cousin once removed only <1% of the time:

I added WG into the tree here:

Joanna and James of Clerragh

I didn’t see any additional ThruLines for Joanna at her Archibald/Peyton level. However, there are some new ones at the level of James of Clerragh:

The odd thing with this ThruLine is that Archibald born 1690 who is usally the father of James of Clerragh is shown as James’ son. However, the relationship shown is correct at 2nd cousin once removed.

When I add the results in with the others, I get this:

Those are a lot of numbers to look at. Basically:

  • Penny had no ThruLine matches at her 2nd cousin level.
  • Joanna shows good match levels at the 2nd cousin level. That is, except with Chris
  • Madeline and Charlotte show poor 2nd cousin match levels. However, the matches are just with one person. Perhaps there are not many descendants on this line who have had their DNA tested.
  • At the third cousin level out of all the matches, there are no perfect scores. I am not sure why that is. If I add up all the percentages for the relationships and divide by the top percentages in the third cousin range, I get 42%. As these ThruLines are somewhat new, I don’t know how usual or unusual that is. The results seem anemic to me. It is as if Archibald Frazer had each of his sons by different wives.

Here are the rest of the results:

At the fourth cousin level, Penny has matches with the Thomas Henry Branch, but not the Edward Wynn Branch.

Summary and Conclusions

  • Looking back at my Blog on Penny’s sister Toni, it appears that Toni had better Frazer matches. However, Toni has not linked her DNA to her Ancestry Tree and therefor does not show up on ThruLines.
  • If Toni links her DNA to her Ancestry Tree, we should get a fuller picture of the situation.
  • I had fun getting a better understanding of Penny and Toni’s genealogy.
  • Perhaps as a result of Penny’s less than average matches on the Frazer side, the DNA matches looked a bit anemic.
  • However, having said that, at the third cousin level there are a lot of matches under Archibald Frazer born in 1792. However, they appear to be all less than perfect matches, with only one match that I see that is above average. That is the match between Charlotte and Gail at 97 cM.
Posted on

Bonnie’s Frazer ThruLines and An Update On Beverly’s

In this Blog I would like to look at Bonnie’s ThruLInes. Here is Bonnie on my Frazer DNA and genealogy chart:

I circled Beverly as I already looked at her ThruLines here. It took two Blogs to write about Beverly’s ThruLines. The connection of Beverly and Bonnie to Archibald Frazer and Catherine Knott above is quite certain. However, one other Frazer researcher has the connection after that going up to a different Frazer couple:

Bonnie’s Frazer ThruLines

Bonnie’s ThruLines are based on her genealogy going up to Michael and Margaret Frazer. That could cause some bias toward that outcome, but we will see.

Bonnie’s ThruLines with her second cousins are not in question, but it is always good to go from the known to the unknown in genealogy:

Bonnie’s Third Cousin ThruLInes

At this level three additional people are added:

I mentioned Beverly above. Then there are two more people from the Catherine Frazer Line. Bonnie’s tree has Catherine as Catherine Peyton Frazer, but Frazer researcher Joanna points out that the Peyton is not correct as that is from a different Catherine Frazer. Someone added a photo of Catherine Frazer’s daughter Rachel Patton from NO’s ancestry:

LO and NO are from two different lines as Catherine Frazer married a Pattison (aka Patterson) who died. She then married a Patton.

Here are Bonnie’s ThruLines so far in a chart:

This means that Bonnie’s match with Gary and Judith indicate a second cousin match 62% of the time. Bonnie’s match with Beverly indicates that her 39 cM match indicates a third cousin 13% of the time. It would have been more likely for the two of them to be fourth cousins. 22% of the time a 39 cM match indicates a 4th cousin relationship. That means that they don’t have the most likely match level for third cousins, but still a likely match.

Bonnie’s Third Great-Grandparents

Here is something I had not considered:

At this point, Bonnie has another likely Frazer ancestor in Elizabeth Frazer. This makes life confusing. According to Frazer researcher Joanna, Catherine Knott was either a daughter or niece of William Knott and Elizabeth Frazer. The scenario showing in ThruLines has Catherine as a daughter. Another point is that Michael and Elizabeth Frazer were siblings. Looking at ThruLines by way of William Knott and Elizabeth Frazer do not add any more matches.

Here Bonnie has six additional matches.

Bonnie appears to have strong ties with the Patrick Henry Line of Frazers. This is good based on the fact that there are records for a both a Patrick and a Fitzgerald born to a Michael and Margaret Frazer:

14 Dec1803  b.         Patrick s. of Michael and Margaret                        Eastersnow   

16 Dec1803  bpt.      Patrick s. of Michael and Margaret                        Eastersnow                                                                    

2 Apr 1804    b.         Fitzgerald s. of Michael and Margaret                  Eastersnow   

5 Apr 1804    bpt.      Fitzgerald s. of Michael and Margaret                  Eastersnow   

Frazer researcher Michael was fortunate to find such early birth and baptismal records from Eastersnow Parish in North County Roscommon. However, it is curious as to how these two children were born less than four months apart from each other.

Mary Frazer

Mary is a new addition to the Frazer DNA/Genealogy Chart. Joanna added her in to explain DNA matches to descendants of Catherine Matilda Fraser. Frazer researcher Joanna made an educated guess that Mary Frazer married William Frazer shown below:

That translates to these two ThruLine matches for Bonnie:

Bonnie’s ThruLines with James Frazer and a Surprising Match

Here are Bonnie’s ThruLines up to her 4th great-grandfather. Bonnie has five matches. The surprising one is with Brittney:

Brittney and Bonnie have a 60 cM match which is very high for a 5th cousin once removed. AncestryDNA only gives statistics up to 5th cousin. According to the ISOGG web page, 60 cM is not unheard of for fifth cousins once removed:

Here are the matches charted out:

What if Bonnie Was In the Tree of Archibald Frazer and Catherine Peyton?

I mentioned early in this Blog that at least one researcher has Archibald Frazer and Catherine Peyton as the parents of Bonnie’s ancestors Archibald Frazer and Catherine Knott. It would be possible for me to test out what would happen if we put Bonnie there.

That scenario would replace Archibald born about 1792 above with Archibald Frazer and Catherine Knott. The effect is that Bonnie would be more closely related to those on the Frazer/Peyton Line and further related to those on the Michael and Margaret Frazer Line. There is no need to change Bonnie’s ancestry to see what ThruLines will do. I think that this would be the result. The original analysis is first and the new one second:

It appears that Bonnie’s original genealogy has a slight edge based on the ThruLines. In the analysis above, she has four rows of green where she has the highest probability for the relationship. In the analysis below she has two. I added up the probabilities for each scenario. In the original scenario, the percentages added up to 215. In the alternate scenario, Bonnie gets a 211. There is not much difference there. This analysis quantifies my previous observations that the matches between the Archibald Line and the Michael Lines are pretty even.

Comparing Bonnie’s and Beverly’s ThruLines

Beverly and Bonnie are third cousins. I had written two Blogs on Beverly’s ThruLines, but they have since changed. Also, it appears that Beverly’s DNA may not have been attached properly. I have her mother as Edith Frazer. In the ThruLines below, Edith shows incorrectly as Beverly’s grandmother:

That means that Beverly would be a second great Aunt to Krista and Jordan rather than a 1st cousin twice removed. Also, I show Beverly’s maternal grandfather as Fitzgerald – not Alexander Frazer:

Apparently there was an Alexander with a daughter Edith but not Beverly’s Edith. Beverly’s Frazer side was from the Parry Sound area of Ontario:

For further out ThruLines, the tree corrects itself somehow. Here Fitzgerald Frazer is correctly shown as Beverly’s grandfather:

Here is my new summary for Beverly’s Frazer ThruLines:

Mollie is in a pink color. Her ThruLine has gone away since my last Blog on Beverly. This is good because I couldn’t explain how her Robert McMaster ancestor fit in. The yellow matches are new since my last Blog on Beverly’s ThruLines.

Combining Beverly’s and Bonnie’s ThruLines

A few observations:

  • Beverly has 21 Frazer ThruLines compared to Bonnie’s 14.
  • Bonnie and Beverly don’t share many ThruLine matches. I am not sure why that is or whether they should. They both share Brittney who is the most distant relative.
  • The numbers seem good for the relationships. This tells me that the DNA evidence supports the trees we have.
  • Under the Michael and Margaret Frazer common ancestors, there is a good mix of matches from their children: 8 for Patrick (who is the same as Henry Patrick); 6 for Margaret Frazer; 2 for Mary (recently added in as a hypothesis) and one for Fitzgerald.
  • I didn’t do the reverse analysis for Beverly that I did for Bonnie. That was the using the assumption that the 4th cousin matches descended from Archibald of Tullynure rather than Michael Frazer. One of the problems I find with that assumption is that it does not account well genealogically for the many DNA matches that have been considered. In other words, all these Frazer descendants came from somewhere and under the alternate assumption, it is unclear how these people would fit in.

Summary and Conclusions

  • In this Blog, I looked at Bonnie’s ThruLines.
  • Bonnie’s ThruLines support the tree we have for her descending from Michael and Margaret Frazer.
  • I did an analysis of what the DNA matches would look like if Bonnie descended from a different Frazer Line. The matches were close but not as good as for her descending from the Michael Frazer Line. Also I didn’t like the genealogy for Bonnie being in a different Frazer Line as I couldn’t see how everyone else could fit in with her being there.
  • Next, I did and update for Beverly’s ThruLines. Beverly and Bonnie are third cousins.
  • Beverly had more matches than Bonnie. Her matches were consistent with Bonnie’s.
  • It helped to combine the two results. This method should work in other lines.

 

Posted on

My Scottish DNA and Matchbox Results

Someone signed me up for the Scottish DNA Facebook Page. Probably because I administer a Frazer DNA Project. The Scottish DNA Facebook Page has a Matchbox utility which I had never used until now. The administrators checked the people who hadn’t used it and put out a request for Scottish DNA members to run matchbox. Their interest got me to act.

My Scottish Matchbox Matches

My grandmother was a Frazer. That is Scottish, but her ancestors since the 1700’s at least lived in Ireland. That means that I am potentially up to 25% Scottish. However, I would think it possible that I could match other members of the Scottish DNA Facebook Page on non-Scottish Lines. I ended up with 21 matches.

 

I had run Matchbox at the Newfoundland Gedmatch Facebook Page before, so I was a little familiar with it. Still, it took a little while to figure out how to use it. The Facebook Name on the right is to identify Scottish DNA members so they can be tagged. Then they will know that someone matches them or a DNA kit they administer.

Jane at Scottish DNA Matchbox

I match with Jane at the above Matchbox match list about halfway down the list. I have looked at Jane’s DNA quite a bit as she is part of the Frazer DNA Project that I administer. In fact, due to intermarriage, we match on two different Frazer Lines:

I’m not in the next Frazer Branch shown:

Jane is in this Archibald Frazer/Stinson Branch. The place where we match is one level up. Our common ancestor is another Archibald Frazer.

Here is my DNA match with Jane:

I have mapped out my DNA. Here is my Chromosome 12:

I am the third sibling down (Joel). My match with Jane is to the right side of the Chromosome where I am clearly Frazer on my paternal side.

I also triangulate with Jane and other people from the Richard Frazer Branch on Chromosome 12:

Green in the box indicates a triangulating DNA match. This indicates that our common ancestor is Richard Frazer and his unknown wife. As expected, the DNA match is not with the more distant (to me) Archibald Frazer/Stinson Branch.

Top Matchbox Match Sharon

Sharon is at the top of my list above. When I run the one to one for Sharon, the match is not as close as it looks on the overall list:

Sharon and I have two smaller matches on Chromosomes 9 and 10.

At the start of my Chromosome 9, I have mapped myself as (orange) Hartley:

That means a non-Scottish match as my Hartley ancestors were from Lancashire, England.

However, at the right end of my Chromosome 10, I have Frazer DNA:

Hartley Matches: Babe, Neil, Cindy and Teckie

To make sure, I run my match against my paternal side phased kit at Gedmatch and get this match with Babe:

Unfortunately, on this part of my Chromosome 10, I have Hartley DNA also:

Neil on my match list also shows as under my Hartley DNA, but on Chromosome 1. Cindy maps to my Hartley side on Chromosome 15. Teckie is a Hartley match on Chromosome 16.

Matchbox Amy, Donna, Greene, WTD, Joan and John: Neither Here Nor There

I show a match with Any, but when I look for her on my phased paternal and maternal DNA match lists at Gedmatch, I can’t find her. That must mean that her match got phased out. So far, other than Jane, nothing is panning out.

I get the same thing for Donna further down the list. Here is what her non-phased match looked like:

I got similar results for Greene, WTD, Joan and John.

Gary and Elizabeth: Non-Scottish Match

These two appear to be related. Here is my match with Gary:

Here is a broken match. I have Chromosome 9 mapped as Hartley for myself, so I’ll rule this out also as a Scottish match. Elizabeth is about the same match, but only the second part of the match above.

Elizabeth #2: Genuine Scottish

The second Elizabeth on my list has DNA in the area of the Jane I mentioned above who shares Frazer ancestry with me:

I take that as a good sign. If Jane and Elizabeth match each other, that means that we have triangulation. Here is how Jane and Elizabeth match each other:

My guess is that Elizabeth’s ancestry goes back to the Frazers in Roscommon, Ireland.

Sallee, MPS, Jane and Shannon Match on Chromosome 20

Sallee, MPS, Jane and Shannon match on an area of Chromosome 20 that is mapped to my Frazer side. Unfortunately, I have what I consider to be an overmatch area on Chromosome 20 with hundreds of matches. Here is my match with Sallee:

These matches may be significant, but due to the number of matches I have, it would take a lot of work to analyze these matches. Plus the number of matches that I have make me skeptical as to the quality of the matches.

The Jane I mention above is the 2nd Jane on my list.

Jean: Genuine Scottish DNA

Here is my match with Jean:

I have all Frazer (blue) DNA on my Chromosome 22:

My Match with Bear on Chromosome 6

I have all Frazer DNA on Chromosome 6. Here is the match:

This match is in an area I have mapped to my Clarke and Spratt ancestors who lived in County Sligo:

Let’s see if Bear matches Stephen:

As expected, they match and triangulate.

Here is how Stephen matches me:

Statistics

 

All the Matchbox matches were either on my paternal side or false DNA matches.

  • Paternal – 71%
  • Unphased (false matches) 29%

Of the paternal matches more came out on my Frazer Scottish side than my Hartley non-Scottish side: 9 Frazer and 7 Hartley. My match with Sharon was on both sides.

Two of the matches triangulated with known matches that have known genealogical connections to me. These were with Elizabeth and Bear and would be worth following up on.

Two of the Frazer side DNA matches were in chromosome areas where I had no identified Frazer matches.

Four matches were from Chromosome 20 where I have hundreds of matches that I cannot identify.

Summary and Conclusions

  • Use your phased results if they are available. When I did this, I found that 29% of my matches fell out as false matches.
  • If you have mapped your DNA, use that information. That will further eliminate matches that you are not looking for.
  • 9 out of 21 or 43% of my Matchbox matches were along my Frazer (Scottish) grandparent side
  • If you have existing identified matches with Scottish background, check to see if there is triangulation with those matches to determine common ancestors.
  • These results are based on my own genealogy where only one of my grandparents has Scottish background.

 

 

 

 

Posted on

Tracy’s Upshall Ancestry and DNA Matches

As I go through my wife’s Aunt Esther’s DNA matches at Ancestry, I came upon Tracy.

According to AncestryDNA, Tracy and Esther could be second cousins or closer.

Tracy’s Genealogy

Here is Tracy’s Ancestry Tree:

I am interested in Tracy’s Halfyard maternal side. I built this tree:

I am interested in Margaret Jane Upshall. Perhaps she will fit into Esther’s tree. The records are not good for Harbour Buffett where many of the Upshalls lived.

Henry Upshall Born About 1841 and Catherine Dicks

Esther descends from Henry Upshall and Catherine Dicks:

Esther is a half-sister to my wife’s grandmother Florence:

This is an Upshall DNA tree. These people have had their DNA tested and match each other.

Jessie Upshall Born 1886

A while back, I came across Karen who descends from Jessie Upshall. By DNA and name it seemed like Jessie would have to be another daughter of Henry and Catherine Upshall:

Karen’s DNA matches to Esther, Joan and Elaine were perfect for her being a 1st cousin twice removed to Esther. I could see no other explanation other than Jessie Kate Upshall was a daughter of Henry Upshall. I added her to the tree even though the records were missing from Harbour Buffett due to a church fire.

Now Tracy has an Uphall great-grandmother born in 1874. Perhaps Margaret Jane Upshall could be another daughter of Henry and Catherine Upshall.

Adding Tracy to the Henry Upshall Tree

I’ll add Tracy in to see how she fits:

If my guess is right, Tracy is:

  • 1st cousin twice removed to Esther
  • 2nd cousin once removed to Joan and Elaine
  • 3rd cousin to Marie, Joanne and Karen
  • 3rd cousin, once removed to Tina

I administer the DNA for Esther, Joan, Elaine and Marie, so I can check those DNA matches.

Esther gets a low score for being a first cousin twice removed to Tracy because she matches by more DNA than expected. Based on their high DNA match, Ancestry would rather see them as 1st cousins once removed. That means that by chance Esther and Tracy share more DNA than average or it could mean that Esther and Joan share ancestries from other branches.

Karen’s Mom Shirley

Karen’s mom Shirley has also tested at Ancestry:

Shirley has perfect textbook matches to Esther, Joan, Elaine and Marie. They get the highest possible scores for these three different levels of relationships.

Summary and Conclusions

  • I think that it would be safe to add Tracy to the Henry Upshall Branch of Upshalls based on close DNA matches to Esther and her family.
  • The DNA matches between Tracy and Esther’s family are a little closer than expected. It is possible that there are other family connections further out.
  • If Tracy uploads her DNA to Gedmatch, it would be possible to do more detailed DNA analysis.
  • If Karen shares her mother’s DNA results with me, I would be able to see how closely Shirley matches Tracy. That would further solidify the Upshall Family Tree.

 

 

Posted on

Esther’s Newfoundland ThruLines at AncestryDNA

Esther is my wife’s 1/2 great Aunt and has Newfoundland Ancestry on both sides. AncestryDNA’s ThruLines takes DNA matches and matches those up with Ancestry Tree matches. Let’s take a look at Esther’s ThruLines to see what they show.

Esther’s Parents

Here are Esther’s Parents:

Here are the ThruLines for Fred Upshall:

These people are not related on the Shave side, so the relationships show correctly as half relationships. These people are my mother in law, Joan, her sister Elaine, my wife Marie and Marie’s niece Tina. These four should show up in Esther’s more distant Upshall side ThruLines.

Esther’s Grandparents

 

Very quickly, Esther is back to 1841 with her Upshall grandfather. I am more interested in the Upshall and Dicks side as that is the side where my wife and her mother are related.

Here are the Henry Upshall ThruLines:

On the right is Karen. I have written Blogs about her family. I put her great-grandmother Jessie as the daughter of Henry Upshall based on DNA evidence even though there was no genealogical evidence. This was a bit dicey, but based on the amount of DNA shared between Esther and Karen, it seems like the right thing to do. According to Ancestry DNA, that amount of shared DNA is most likely to be a 1st cousin, twice removed relationship:

Nicholas and Esther

How does Nicholas fit in? For one thing, 1st cousin four times removed goes right off the chart:

I would think that a 1st cousin four times removed would be similar to the 2nd cousin once removed. Either way, the chances of this being right are <1% according to AncestryDNA.

Nicholas’ Genealogy

Here is the expanded view of ThruLines:

One problem is that Nicholas’ parents and grandparents are shown as living, so I don’t have information on them. Nicholas’ tree is managed by Kara. She has Gladys Uppshall Knight as Nicholas’ maternal great-grandmother. From there, Stephanie’s tree kicks in with Theodore the father of Gladys.

Here is the 1935 Census for St John’s West:

The assumption is that this is the same Gladys in 1945:

Here is a replacement birth record for Malcolm:

This shows that Malcolm’s mother was a Shave. That makes it more surprising that Esther and Nicholas don’t have more matching DNA as Esther’s mother was a Shave.

Here is another after the fact record for Malcolm’s brother:

This could be William in the 1945 Census living next to his brother:

Was Alexander the Son of Henry Upshall?

I have it that way on my Ancestry Tree, but not on my Upshall Web Page. I don’t see any genealogical evidence that would put Alexander as the son of Henry Upshall. It is that way in some on-line Trees. Based on the low amount of DNA match and the shaky genealogical evidence, I would call into question this particular ThruLine at this time.

Esther’s Upshall Great-Grandparents

Here, things get more interesting as there is no known genealogical evidence for Esther’s great-grandparents. These ancestors are a best guess scenario. Here are the ThruLines for best-guess Peter Upshall:

At first look, it seems as though there are a lot of different DNA matches for these likely children of Peter Upshall.

George Upshall Branch 1829

Here, note that Pat, like Nicholas has Shave and Upshall in her ancestry. I note she also has Burton which Eshter has, so perhaps other names also. 280 cM is a huge match for a 2nd cousin twice removed:

Ancestry says that Pat and Esther look like 2nd cousins. However, that is assuming that they are not related on other lines which they are.  For comparison, I had mentioned above that Nicholas’ match to Esther should have been equivalent to a 2nd cousin once removed. Another thing I’d like to do is to see if these people have shared matches with each other. I note that Nicholas does not have a shared match with Pat. That means that he shares less than 20 cM with Pat.

Checking with Joan and Elaine

One way to check these matches is by Joan and Elaine. They are Esther’s half nieces. That means that they match on the Upshall side and not the Shave side. Here is what Joan shows:

Now the DNA matches are quite a bit different. When I look at the numbers for Joan and her sister Elaine, I get this for the proposed George Upshall Branch:

Sarah Upshall Branch 1831

Here Esther has 10 DNA matches on the Sarah Upshall Collett Branch:

Here are the results for Eshter, Joan and Elaine:

  • I added in a Maximum % column, so we could see what the highest likelihood is for the DNA match
  • Green indicates that the DNA matches got that highest mark
  • This seems to indicate that Esther does not match this Sarah Upshall Collett on other lines.
  • None of the numbers seem unreasonable. The <1% can be explained by matching on many other lines and getting too high of a match.
  • Joan and Elaine match Sarah and Esther doesn’t. That means that Florence (who is Joan and Elaine’s mother) got DNA that her sister Esther did not.
  • Esther went off the chart with the 2nd cousin 4 times removed category.
  • To do this right, I would need to check all the genealogy. I didn’t. The genealogy is likely good up to the children of Peter and Margaret Upshall and then missing.
  • All these numbers seem to confirm that the trees are likely trees.

Jane Upshall Branch

Jane appears to be well documented by DNA. Esther has 14 matches to Jane’s descendants along the Tulk Line. I have noted close matches to Tulk in the past, so this may explain those DNA matches.

Jane also has a rare Upshall birth record:

However, I find it odd that she was born in November and baptized in September. Perhaps she was born in 1838. I assume that there was only one Peter and Margaret Upshall at the time. Burin was about 90 miles away from Harbour Buffett, so I assume there was a travelling minister who performed these baptisms on 29 September 1839. I have that Jane was born in Burin which is not accurate. She was born in Garnish Gut according to this record. I’m not sure where Garnish Gut is, but I assume that it is part of or near Harbour Buffett.

An additional confusing point is that Peter was the father of Christopher Upshall also in 1833:

This means that Peter may have had a first wife Mary or the transcription may have been off. Christopher doesn’t show up on the ThruLines. I don’t see many trees for Christopher. Here is one:

This person shows Margaret as the mother and the birth at Famish Gut which perhaps makes more sense. Perhaps there were few descendants on this line.

So far the matches for the descendants of the three potential children of Peter and Margaret Upshall looks like this:

The evidence looks pretty good. I added Karen from the more recent Henry Upshall Line to show how good her numbers were.

Susan Upshall Born 1848

This Line is a little more confusing:

Roy’s grandmother is Margaret Collett. That begs the question of Elizabeth Webber. Shouldn’t she be a Collett also? Here is some information from http://www.collettfamilyhistory.net/Part-32-The-Newfoundland-Line-Rev.12.htm:

The same web page has this on Elizabeth:

According to this information, Elizabeth Webber is really Elizabeth Collett.

Here is the last of Esther’s ThruLines at this level:

This was a bit of an annoying exercise, but it does seem to verify the trees for the descendants. However, I have not tried to disprove the theory. That would take a bit of work. I do recognize interfering influences of the DNA matches through intermarriage.  I showed some of those effects by comparing Esther’s matches to her half nieces Joan and Elaine. The main interferences could be other relationships with the Dicks, Shave and Burton families. The other thing I didn’t do was to check the genealogy of each line. This would take a while to do. The last confirmation would be to check to see if each of these matches match each other. The best way to do that is with Genetic Affairs’ AutoClusters.

Esther and AutoClusters

Here is a ‘basic’ AutoCluster I ran for Esther.

Esther has all Newfoundland ancestry, so almost all her clusters match other clusters. This AutoCluster ran January 9, 2019 between the match levels of 50 and 250 cM. I checked my Chart above to see which clusters the matches were in:

It turns out that the Jane Upshall Tulk Line were all in Cluster 14. In the other Lines, there was not enough information or the matches were outside the range. However, the other two Clusters found were 8 and 11.

This means that while Esther is matching Cluster 14 by Peter and Margaret Upshall, all the 10 in my Chart are matching each other by Jane Upshall and her Tulk husband.

Esther’s ThruLines With Christopher Dicks Born 1812

Christopher was married to Elizabeth. My best guess is that she was Elizabeth Crann. Esther has 12 new DNA matches on the Christopher Dicks Line. These matches could have gotten their DNA from Christopher Dicks or from his wife Elizabeth or both.

The matches under Catherine Dicks have already been looked at. Compare this with the Gedmatch Tree I have been working on for the Dicks Family:

ThruLines has a David Branch that I don’t have and I have a Christopher Branch that Thrulines doesn’t have.

The David Dicks Branch

Here is a partial opening of that Branch:

One interesting thing here is that Frances is also related to Esther through the Jane Upshall Tulk Line:

That is because Caroline Dicks married Edwin Kingwell Tulk according to the trees at Ancestry.

Cathy in the David Dicks Branch

Here is how ThruLines shows Cathy:

I had Cathy in a different Branch. However, if the ThruLines are right and I am right, she could be in two Dicks Branches of Frances Dicks Burton and Christopher Dicks:

Here is a summary of the David Dicks Branch:

It seems like everything checks out OK. Esther was matching too high to Newf due to matches on Esther’s maternal side. Joan isn’t related to Esther on Esther’s maternal side, so that took the extra DNA out of the equation. I skipped checking Joan’s sister Elaine’s results to save time. Joan matched one person David, who didn’t match Esther.

Going Up One Generation to Christopher Dicks Born about 1784

The Dicks family was huge and then spread out from there.

Here are nine children of Christopher who have matching DNA with Esther:

  1. John Michael – before 1808 – I haven’t been following this Branch. Has some Joseph’s in this Branch
  2. Elizabeth – 1809 – Married Thomas Adams
  3. Joseph – born 1810
  4. Frances born 1811 – Married Charles Burton
  5. Christopher Born 1812 – covered in previous section
  6. Rachel born 1817 – married James Joyce
  7. James Dicks born 1830 – also has a Joseph Dicks son. I haven’t been following this Branch either.
  8. George born 1832
  9. Robert born 1824 – so should be above

I trace six children in my Dicks Gedmatch DNA project. I didn’t have John Michael, James or Joseph. Altogether, there are 77 DNA matches represented in these 8 proposed children. That would be too many matches to check for this Blog.

Joseph Dicks

Although I don’t have Joseph Dicks on my Dicks DNA Chart, I have that Esther descends from Joseph Dicks on her maternal side:

I have Joseph Dicks born at Famish Gut in 1810 to Christopher and Margaret. However, this does not appear on Esther’s ThruLines or on my own Dicks DNA Chart.

I’ll look at the genealogy behind the John Michael and James DIcks Branches as I am unfamiliar with them.

John Michael Dicks Born 1808

The ThruLines Profile for John Michael shows this:

This information is a bit sparse. It shows that John Michael would have been over 105 years old when he died.

In a previous Blog I note this Joseph:

My thought was that this was a more likely ancestor for Esther.

Joseph Dicks and Violet

After changing Esther’s Jane Ann Line, I noticed that ThruLines did have a Joseph Dicks Branch with one large match:

However, this follows from the way I had Esther’s tree. Violet has a huge match with Esther at 359 cM. Here is a much closer relationship between Violet and Esther at 2nd cousin:

My mother-in-law, Joan also has a good match with Violet:

Violet’s Tree

Now I’m suspicious that Tulk may be related to Upshall.

Upshall of Dorset ThruLines

Newfoundland is too complicated with the intermarriages and lost records. Lets go back to Dorset, England:

My current thinking is that Peter Upshall born in 1800 in Dorset was the daughter of Sarah Upshall. Sarah is shown here with three of her siblings. This Dorset, England ThruLines appears to confirm that Sarah was the sister of Joseph, George and Charles who shows a Private above. This is a chance to find out what my wife’s Upshall relatives have been up to since 1737. Judy above is probably in Australia. Margaret’s paternal grandmother was born in England and died in Rhode Island.

The Charles Upshall Branch

Ross and Peter’s parents were both from Dorset, England.

Here is one of the trees I had been working with:

For some reason I am missing the Jane Upshall/Tulk Branch. I also need to go up two levels to John Upshall.

I had previously added an additional sibling Ann Upshall due to a match with my wife’s Aunt Elaine. Due to the fact that none of Peter’s Aunts or Uncles did not live in Newfoundland, my guess is that the DNA matches represent John and Elisabeth Upshall.

Here are some potential siblings to Sarah Upshall that I took from an online Ancestry Upshall Tree:

After quite a bit of typing, I get this:

This is a small version of the Upshall DNA match tree. I didn’t add in Ann Upshall or many of the Newfoundland Upshalls.

Shared DNA Matches

When I check how Judy on the bottom left of the chart above matches Esther, I get these shared DNA matches:

Many of these shared matches are with descendants of Newfoundland Upshalls.

Elaine and Joan’s Hazelbury Bryan Thrulines

One of Elaine’s Hazelbury Bryan Thrulines is with Peter on the Charles Upshall Branch shown above. The other is with Hazel on the Joseph Branch.

It looks like Hazel’s Branch stayed in England while Judy’s branch went to Australia:

Joan also has a ThruLines match with Hazel.

More of Esther’s English Upshall Shared DNA Matches

Elaine and Margaret from the George Upshall Branch have a shared DNA match named Trenton. Trenton has a good sized tree:

Trenton’s father’s mother’s father was Peter Upshall Boutcher. This sounds suspicious. There must be an Upshall in Peter’s ancestry. The Collett Genealogy that I referred to above is very helpful:

Here is one relationship that ThruLines didn’t figure out, but it helped in conjunction with Shared DNA Matches:

Putting the Upshall’s Together

Here I added a representative person from the Upshall/Tulk Branch:

Based on the ThruLines there are over 10 matches each on the Sarah Upshall Collett and Jane Upshall Tulk Branches. So the above tree is just a skeleton.

One Last Shared Match Between Esther and Judy

I don’t want to leave too much low hanging fruit. AU is a shared match with Dorset descendant Judy and Esther. Here is his three person tree and his DNA match:

AU and Esther have a DNA match of 69 cM. Here is Jacob in 1935 in Harbour Buffett:

Edward and Martha married in 1916:

Unfortunately, I soon got stuck at the level of Edward Upshall:

Summary and Conclusions

  • Going Through many of Esther’s ThruLines was a large undertaking. I may have been better off just looking at the Upshall ThruLines.
  • It was helpful have two of Esther’s 1/2 Nieces to compare the results. This is because Esther matches on many intermarried lines and Joan and Elaine only match on Esther’s paternal side.
  • Esther’s ThruLines complemented and expanded upon the previous DNA work I have done on the Upshall and Dicks Families.
  • I compared Esther’s Upshall ThruLines with her AutoCluster results. The Jane Upshall/Tulk Line showed clearly in one of Esther’s Clusters.
  • I brought Esther’s ThruLines back to John Upshall born 1737 in Dorset, England and his wife Elisabeth Ellis. To me, the results clearly show that Esther descends through John Upshall and his daughter Sarah Upshall to Peter Upshall who was an early settler in Newfoundland in the Harbour Buffett area.
  • I was a little surprised that Esther had four DNA matches going back to a common ancestor who was born in 1737
  • ThruLines work well with Shared DNA Matches. I was able to find at least one new Upshall Line using Shared Matches to ThruLine Matches.
  • In the future, I would like to concentrate more on the Upshall Branches and build them out in my Excel spreadsheet.
  • It may be helpful to also check to see how many Upshall descendants have their DNA at Gedmatch, FTDNA and MyHeritage. This would allow for more detailed DNA analysis.
  • ThruLines are able to put together many trees and DNA matches in order to see a possible big picture solution to some genealogical problems.

 

 

Posted on

Some Upshall and Hollett Genealogy

I recently had this request from Shirley:

My Mother was Lillian Florence May Hollett who was born in Montreal, but her father was Samuel Hollett from Spensers Cove Placentia Bay NL.
My father was Cecil Llewelyn Upshall, born in Harbour Buffett NL.
Can you give me any information on either of those backgrounds?

This could be the couple in 1945:

Cecil was born in Placentia and Lillie in Canada according to this Census. Here is Grand Falls where the couple were in 1945:

The 1921 Census

This gives some more information. Cecil’s parents were Isaac and Rebecca. Cecil’s grandmother was Lizzie Hann from Harbour Buffett. Here is the marriage record for Isaac and Rebecca:

Isaac was a fisherman. Both Isaac and Rebecca were living in Harbour Buffet at the time of their marriage. Before this, the genealogy gets a little shaky:

Martha shows that Isaac’s parents were Peter and Hannah. Martha has this further information:

Martha has Peter dying in Kingwell:

Here is the tree I have for my wife:

My wife’s ancestor Henry Upshall was born around 1841. Perhaps Peter was a brother or other relative to Henry.

Shirley’s Hollett Side

So far Shirley’s tree looks like this:

Here is the marriage record for Samuel and May:

This couple married in a Methodist Church in Montreal. Samuel was a carpenter and his parents were John and Susan Butcher. It appears that Samuel was born at Sound Island in 1860.

This is not too far from Spencers Cove.

 

Records have his death at Botwood:

Botwood is to the Northeast of Grand Falls. Samuel is listed as 84 years old in 1944. The number next to the cause of death is for something else. He shows as being born in Burin. This record for another Hollett birth appears to tie things together:

This shows that Sound Island was in the District and Parish of Burin. I was not aware of this connection previously:

Sound Island and Burin appear to be about 100 miles away from each other.  It is a good thing that the Holletts were Methodists as those records appear to have survived better than some of the other records.

Here is the tree that I have for Shirley:

Due to missing records, it may be helpful for Shirley to take a DNA test. This could help to solidify existing connections and perhaps suggest some new connections.

 

 

 

Posted on

Finding More Frazers: Matthew

I match Matthew at AncestryDNA. In fact, AncestryDNA calls it a very high DNA match. Ancestry further says that they think I have a 99% chance of finding shared ancestors with Matthew. I took that as a challenge. I messaged Matthew and he gave me enough information to get started on a tree:

Matthew’s Frazer Genealogy

I was interested in Matthew’s Frazer side as that is where we must match:

 

 

Matthew gave me the name of his father and of his grandmother. That was enough to get going. From there I found his Frazer grandfather who was born in Manitoba, Canada. From there others have created Frazer trees that I can use:

Now I am getting into familiar territory with George and Susan Frazer:

Here is George Frazer and Susannah with four boys. Here is my list of children for George and Susannah:

It looks as though Ellen and Margaret died young. My guess is that the four youngest boys could be the ones in the photo above.

Next, I’ll put Matthew on my own Frazer DNA/Genealogy Chart:

This shows that Matthew is my 4th cousin twice removed. He is also Gladys’ 1st cousin three times removed.

One Tree Leads To Another

Speaking of Gladys, she has shared DNA matches with Matthew. One of their shared matches is Sandra. Here is Sandra and her paternal tree:

I’ll add Sandra also to my DNA/genealogy tree:

Sandra is Gladys’ 1st cousin once removed.

ND Helps Fill Out the George Frazer Branch

Gladys and Sandra have a shared DNA match with ND:

One interesting thing is that I match Matthew by more DNA than I match Sandra or ND. I match Matthew by 49 cM across three segments. If I match Sandra or ND it would have to be by less than 20 cM as that is the limit for shared matches.

Summary and Conclusions

  • By using Matthew’s hints, I was able to convert our DNA match into a family tree.
  • Matthew’s DNA match adds to the existing DNA matches from the Richard Price Frazer and George Harvey Frazer Branches
  • By looking at shared matches between Matthew and Gladys, I was able to add Sandra and ND onto the James Clarence Frazer Branch.