I'm finally ‘English’! (MyHeritage ethnicity updates)

 It seems that, for many people, when considering DNA testing, it is the 'ethnicity' that they are mainly interested in - how "English" they are, or whether they are a "Viking", or a 'native' to whichever country they live in.  I remember when I first received my Ancestry results, and contacted my closest match - they hadn't even realised that they would be able to find other people who were genetically related to them.  

Whereas, for me, it has always been the DNA matching that was important, in order to confirm (or disprove!) my family history research.  I think that's one advantage of being in a family history organisation like the Guild of One-Name Studies - there have usually been 'pioneers' in any aspect of the research, people who have already gone on ahead and have fed back some of the pros and cons they've discovered.

And so I was aware that, for the ethnicity estimates, the clue is in the name - these are 'estimates', and they depend very much on which reference populations the companies are using.  That didn't mean I took no notice of them - I just took them 'with a pinch of salt', especially when one company, MyHeritage, indicated that I had no English DNA.  

Several of the companies included the UK as part of a general "Northern Europe" grouping, so I hadn't actually noticed at first glance - the following is from a screenshot I took in 2017:

I think one often doesn't notice when something is 'missing', only when it's there but doesn't 'fit'.  So it was only when I explored the maps and figures in more detail that I realised the estimate was showing me as having no English:

  

This was rather strange, considering that, with the exception of one 5xgreat grandfather (a German, who was in England by 1802), every one of my identified ancestral lines is either in England, or in the South Wales border area!

Since these results seemed so far 'off', I haven't paid much attention to the ethnicity estimates at MyHeritage over the years.  

But the companies are always refining these reference populations and MyHeritage has recently released a new update, version 2.5. You can see their blog post about it at https://blog.myheritage.com/2025/02/introducing-ethnicity-estimate-v2-5-improved-dna-ethnicity-model/

I hadn't actually noticed that my initial results, which were from when the ethnicity estimates were still described as "Beta", had been refined slightly, at some stage, by version 0.95:

That had brought in some English, at 3%, with hints of several genetic groups from specific areas within England.

But now, with this latest update, the estimate has changed considerably and is much more in accordance with my known ancestry:

Even the "Germanic" is now showing up!

The above results all come from the same DNA test that I originally transferred to MyHeritage, so it is not that my DNA has changed in any way.  What has changed is the method by which MyHeritage are analysing it, along with their updated reference populations.

As Roberta Estes states in her blog about the update, "the whole purpose of updating ethnicity results is to obtain either more granular results, or more accurate results, or both." ¹

In my case, the update definitely seems to have produced both!

Notes and Sources

1. Roberta Estes blog post about the MyHeritage update - https://dna-explained.com/2025/02/06/myheritage-introduces-ethnicity-v2-5/

DNA progress - Ancestry Pro Tools and my NAYLOR/NAYLER family

 I have made a start on reading about some of the other bloggers' experiences with Pro Tools and found that the main feature they appreciate is the one that I think will also be the most useful to me - the ability to see how much DNA is shared between a specific match and those other matches that the specific match and I have in common. 

I was intending to illustrate this with some data from a few of my first and second cousins.  However, that post will have to wait a while, since a couple of recent new matches on Ancestry have sent me off on a sidetrack.  Since they are also good examples of how Pro Tools can help, I'm going to use the data from them instead.

So how does Pro Tools help?

The two matches happen to be a mother and son. How do I know that?  Because Pro Tools tells me so:

 

The mother matches me by 47cM across two segments (unweighted shared DNA 52cM, longest segment 45cM).  She has a family tree - but there's only one person on it.  The son matches me by 26 cM across one segment (unweighted shared DNA and longest segment both 31cM). He has an unlinked tree, with about fifty people on it.  

Previously, both of the matches would have appeared on my "4th cousin or closer" match list, since they both share more than 20cM with me.  But, when looking at the shared matches, although they would each appear when I viewed the other's list, I would not have been able to confirm the relationship between them, because I only had the family trees to work with. 

Whereas now, with the Pro Tools, Ancestry shows me the quantity of DNA they share between them, as well as telling me the predicted relationship.

A parent/child predicted relationship is the only one that (as far as I am aware) will always be correct and, as relationships become more distant, the predictions by the DNA companies become less reliable, since they are based on a range of possibilities for the quantity of DNA shared. 

But this information is still a major benefit whenever relatively close members of a family have all tested their DNA.

For example, in this case, having seen that the son is the home person on his family tree, I can immediately identify which side of the tree is the relevant one to research, in order to look for our shared ancestry, because I know the connection is through his mother.

If a match happens to have first or second cousins tested, and it is possible to identify where their common ancestry with the match is, then each of those generations back to their shared ancestor also narrows down the relevant portion of their family tree that I would need to focus on.

Without Pro Tools, I might not be able to identify such cousins - in fact, they might not even show up on the shared match list, if the DNA they share with me has fallen below 20cM.  But the fact that Pro Tools seems to show the shared matches where just one of us shares at least 20cM with them, means there are matches on the lists, which I would not have previously seen.

To illustrate this - based on the old 20cM threshold, only twenty-four of my matches would have shown up as shared matches to the mother.    Eleven of these share between 21cM - 25cM with me, five are in the 30cM - 46cM range, and five between 54cM - 59cM. Then the closest three share 78cM, 146cM and 250cM respectively with me.

The last two are my half first cousin, and a half 1st cousin 1 removed, so I recognise them and know where they fit in my family. The next largest, at 78cM, has a family tree with thirty-six people on it, including a Frederick NAYLOR in Hawaii (supposedly b 1870, no birthplace, and no death details given.)  Now, NAYLOR is one of my ancestral names, and it is relevant to the two higher matches, as well.  The unattached family tree on the son's profile also shows a descent from the same Frederick NAYLOR, in Hawaii, as the 78cM match's tree does.

But, other than identifying that these matches are 'potentially' connected to my NAYLOR line, and that the two new matches will probably connect more closely to the 78cM match, based on their tree, I don't think that I'd have been able to identify much more about them.

However, with Pro Tools, there are fifty-three shared matches shown between the mother and I, rather than just the twenty-four.  As well as her son, these include a predicted half-brother or nephew, and seventeen matches with a predicted relationship involving the term "1st cousin".  Nine of these, including the half-brother/nephew, would not have even shown up as shared matches to me, without the Pro Tools, since they share less than 20cM with me.

But they are all close enough to the new match that I should be able to work out how most of them connect to each other.  

A downside to pro tools?

Yes, there is a downside to all this additional information (at least, for me, and the way I work.)  

Previously, before taking out the Pro Tools, I would check on my new matches at Ancestry most days, in order to keep track of the number in the "4th cousin or closer" category.  If that total had increased, I'd view those matches first, check for any shared matches between us and, if there were any, and I'd already made some progress in identifying our connection, I'd add a note to that effect to the new match's profile.  Once any close matches were dealt with, I'd check through the other, more distant, new matches, looking for any that did show shared matches with me and, again, add a note to their profile. Since, without Pro Tools, the only shared matches had to be ones sharing over 20cM with me, I frequently found these, more distant, new matches did not show any shared matches with me.

But, of course, now that Pro Tools means I can see any shared matches that share greater than 20cM with the new match, even if they only share down to 9cM with me, just about everybody shows some shared matches (in a couple of cases seen so far, there's been nine pages of them!)

So, this makes the task of viewing new matches so much more time consuming, and I am going to need to modify my routine - perhaps not even checking the shared matches unless I have some clear indication that there's a 'potentially findable' link to them.

Returning to the two recent new matches…

As indicated above, based on both family trees and other shared matches, it seems the family share ancestry with me, at some level, through the NAYLOR family.  The NAYLOR line is one that quite a few of my matches seem to connect to. (I mentioned the NAYLOR cluster, "Group 1", in my post on 9 August 2017 at https://notjusttheparrys.blogspot.com/2017/08/ancestry-shared-matches-and-new.html.)

Some years ago, because of the number of matches in this group, I constructed a 'rough' family tree, on paper, predominantly derived from other people's family trees (with a little bit of 'fact checking'. :-) ) 

It has remained on paper ever since - mainly because, once I discovered a Herald at the College of Arms in the early 19c was a possible sibling to my line, sifting through the information to distinguish fact from fiction became much more difficult, since there is so much of it!

But now, with Pro Tools showing how my matches relate to each other, I think I will stand more chance of being able to fit my matches into the NAYLOR line, and actually confirm the links, than I was able to do before (bearing in mind that many of them either have no tree, a partial tree, or even an incorrect tree.)

So that has been my 'sidetrack.'  This week I have been entering all of the rough information into FamilyTreeMaker, the program I use for my own personal family history. I am now beginning to check the 'facts' more thoroughly, as best I can, before making the information publicly available on my Ancestry tree.

I don't know whether I shall be able to resolve who the parents of the Fred NAYLOR in Hawaii were - although the son's tree has his birth as England, I do know that other records indicate it was in Australia (and I think there's one record that suggests the USA instead).  There is a potential Fred born in Australia - and at least one researcher on Ancestry has placed the Hawaii Fred into that family - but there is an issue in that Fred's marriage in the US indicates his father was also called Frederick, whereas the father in the Australian birth was a Charles.

It is a common frustration, when an emigration causes such a break in a family line.  I am hoping that, by placing many of my DNA matches onto the tree, I will be able to develop, and test, theories as to where Fred fits.

But it is still a 'work in progress' - and there will be some caveats to the predicted relationships (which I hope to explain further, when I finally get that "1st and 2nd cousins" post written.)

In closing, I'll include the details of two monuments to the family, reported to be in the church of St John the Baptist, Gloucester.¹:

Epitaphs in St. John the Baptist's Church, Gloucester. 

On a large mural tablet in the south aisle : 

Sacred to the Memory of Captain Joshua NAYLER, 

who departed this life 14th Decr. 1750, aged 67 years. 

Also of GEORGE NAYLER, Son of GEORGE NAYLER, 

of this city, Surgeon. who died 19th March, 1750, aged 6 weeks. 

Also of the above GEORGE NAYLER, Esqr. 

only Son of the said Captain JOSHUA NAYLER, 

who died 12th Septr. 1780, aged 58 years. 

He married Sarah, only Child of John Park of Chitherow [sic], 

in the County Palatine of Lancaster, Esqr. by Frances his Wife, 

Daughter and sole Heir of William Osman, Esqr. and grand-daughr. of John Park 

Of Little Urswick, 

in the same county, Esqr. by Margaret Senhouse, his Wife, 

and by the said Sarah had issue six Sons and three Daughters. 

Also of JOSHUA NAYLER, youngest Son of the said George and Sarah Nayler, 

who died 12th Decr. 1787, aged 20 years. 

Also of EDWARD HENRY NAYLER, only Child of Richard Nayler, Esqr. 

(fourth Son of the above George and Sarah Nayler) by Harriot Howe, 

his First Wife, who died 6 Decr. 1792, aged 4 years. 

Also of CHARLOTTE MARY NAYLER, eldest Daughter of George Nayler, Esqr. 

York Herald (fifth Son of the above George and Sarah Nayler,) 

who died 4th Augst. 1794, aged 

Also of the above-named SARAH NAYLER, Widow, 

who died 31st Jany. 1802, aged 78 years. 

Also of FRANCES NAYLER, Second Wife of the above 

Richard Nayler, Esqr. Eldest Daughter and Coheir of Thomas Blunt, 

of Huntley, in this county, Esqr. she died 19th Decr. 1805, aged 35 years. 

Also of the said RICHARD NAYLER, Esqr. 

who departed this Life 6th Decr. 1816, aged 56 years. 

And of MARIA NAYLER, Second Daughter of the above George 

and Sarah Nayler, who died 28th March, 1821, aged 58 years. 

Below the inscription, on a sort of foliaged corbel, is a shield bearing the arms of 

Nayler, and on an escocheon of pretence those of Park and Osman quarterly. 

On another mural monument placed on the same wall— 

Sacred to the Memory of MARY, Wife of THOMAS NAYLER, 

Lieutenant in his Majesty's Marine Forces, and Daughter of 

Thomas Grimshaw of Preston, in the County Palatine of Lancaster, Esq. 

who ended her course of mortality on the 25th day of September, 1790, 

after having sustained with singular Fortitude and Resignation the tedious progress 

of a lingering Disease. 

Reader! if Devotion without pretence, and Charity void of Ostentation, if filial 

Piety and Conjugal Fidelity be Virtues which thy Justice would commend and Zeal 

would emulate: know here was an Example which might have claimed Applause and 

commanded Imitation. 

This is on a white marble tablet with an urn upon it: on a blue marble back-ground, 

of pyramidal shape, is suspended a small shield, Quarterly 1st and 4th Nayler, 2. Park, 

3. Osman; impaling, Or, a griffin segreant sable, for Grimshaw. 

If anyone can confirm that such monuments actually do exist, I'd be very grateful!

Notes and Sources

1. The epitaphs are given in "The Herald and Genealogist" Volume 7, pages 79/80, as part of an article relating to Sir George NAYLER, pages 72 - 80, which is available at https://archive.org/details/heraldgenealogis07nich/page/72/mode/2up?q=nayler 

DNA progress - first steps

 At the end of 2023, Ancestry released their "pro-tools" in the UK.  This is an additional set of tools for family history, and for more advanced DNA research, than are available through their normal subscriptions. But it does require both a current subscription, and additional payments.  Although I was 'tempted' when it was first released, I left it for a while because that was a busy period and I knew I wouldn't have time for research. But I was then disappointed to discover, when I returned to it later, that the monthly cost had already increased from £4.99 to £7.99.  

That put paid to that!

However, a recent post on FB alerted me to the fact there was an offer on (until 20th January), and I have now been able to take out a cheaper option for six months.  I'll see how I get on with it, and how useful it proves to be, as to whether I continue to subscribe, or not.

Of course, the additional tools and information should be of help - for example, it is now possible for me to see how much DNA is shared, and the suggested relationship, between one of my matches and the matches we share.  The thresholds at which the shared matches are shown is also less restricted than it is with the standard tools.  

This will be very useful in cases where several members of a family have tested but perhaps only one or two of them share 20cM or more with me, so the more distant ones didn't previously feature in the shared match list.  This should  make it easier for grouping matches and allocating some of the more distant ones to potential ancestral lines. 

My main hesitation is how to get to grips with recording all of the additional detail.  So my next step will be to read up on some of the blog posts by other researchers, to find out how they are managing the data.

DNA Update

In my last post, I mentioned the need to focus on my own family history again.  One aspect of that is making the most of the opportunities that DNA provides in tracing more distant or 'lost' relatives.  It's been a while since I did any serious work with my DNA results so, as a start, I've updated the graph I initially posted in April 2020¹, showing the numbers of my matches who are predicted to be my "4th cousin or closer" at Ancestry:

I'm currently up to 345 matches in that category.  As can be seen, the rate of increase has slowed down since early 2020, but new matches are still coming in relatively frequently.  I check Ancestry most days and, whenever there are any new matches, the first thing I do is look to see if they have any 'shared matches' with me, since those can help with placing the new match in the correct area of my family tree.  Although the more distant new matches often show no shared matches, most of those in the "4th cousin or closer" category will match 'somebody' and so I can add a note about this to the profile I see for them.   

That's about as far as I've been going over the last few years.  

Back in 2017, I'd worked out how matches tended to group together and what that indicated.²  But everything DNA related seems to have become much more 'complicated' over recent years, what with increasing numbers of matches, changes to the company websites and the information that's now available, and also, consequently, changes to some of the tools used for managing the data.  

It might take me a while to catch up with the best methods for dealing with all these matches now, but at least the "basic principles" about DNA transmission haven't changed, so that the task doesn't feel impossible.

Updates will follow as I make progress!

Notes and Sources
1. Matches graphed in April 2020: https://notjusttheparrys.blogspot.com/2020/04/ancestry-dna-matches-passing-200-4th.html
2. The grouping of shared matches: https://notjusttheparrys.blogspot.com/2017/08/ancestry-shared-matches-and-new.html

Family Connections

I recently had the opportunity to meet up with some of my cousins, and their children, for lunch, which was lovely - so often, as we get older, it seems as if the only time we get together with our more distant relatives is at funerals! 

One of them has also been researching his family history, so it was good to chat about his progress, DNA successes, and experiences with some of the older documents we delve into.  (Although most of my own lines are only traced back into the 1700s, the research I have been doing into a local historic building has involved some documents that are written in Latin - a frustration for me, since I don't understand it!)

Although most of my cousins are not specifically interested in researching, they are now of an age where they have given up work, and have a bit more free time (theoretically!) and, for those who have members of the previous generation still alive, they are also realizing the importance of asking questions 'now', before it's too late.  

This, of course, has prompted discussions about things my parents, who are no longer with us, told me, as opposed to the versions their siblings remember. 

All of which has reminded me of the need to focus on my own family history again - the unwritten stories, the wider research possible now that record availability is so much better than it was when my parents began their research in the 1980s, the opportunities that DNA provides in tracing more distant or 'lost' relatives....  

And the need for getting all of my own research and paperwork organized, so that it is in a good state to pass on to my next generations!

 

   

Elsie May THOMAS - another look at the 1921 census

Back in January 2022, I wrote about the release of the UK 1921 census, which is currently only available through FindMyPast, and about those ancestors I had expected to find in it. [That post, "First look at the 1921 census" can be seen here ]

At the time, I was unable to find my grandmother, Elsie May THOMAS, who was not at home with her parents, George and Rose THOMAS in Collington.  Six of her siblings were with them, including Elsie's older brother, Ernest:

As can be seen, the four youngest children are all aged under fourteen, and still in full time education, but Matilda, at 15, is already working as a servant for someone else in nearby Wolferlow.

Since the ages of the 'missing' older girls ranged from seventeen to twenty four, I suspected they were all living away from home, probably employed in some form of domestic service.  This was soon confirmed to be the case for the eldest girl, Edith, aged 24, who I found with her grandparents, John and Priscilla THOMAS, in Stoke Bliss, recorded as working for John, dealing with the poultry and dairy work.

The other two girls were not difficult to find: 

Hilda Mary THOMAS, aged 22, was in Mamble, Cleobury Mortimer, in Shropshire, employed as a "Help" to a farmer and his wife, who had an eight month old baby.

And Ada Annie, at 17, was a domestic servant for another family in Collington. 

But there was no sign of Elsie.  I suspected that she was in the database somewhere, probably local to the rest of the family, but just mistranscribed to such an extent that it made her difficult to find.  I did try some possible alternative spellings, but without success.

This is always a problem when using computerised databases - if you don't search using the spellings that match what is recorded in the database, even with the use of wildcards, it can be impossible to find people.

Fortunately, in October last year, FindMyPast introduced an additional subscription covering the 1921 census.  So, for a relatively small payment (compared to what I could have previously spent trying out "possible" entries), I would be able to view all of the 1921 census images.  This meant that I could go back to using the "old" method for finding someone in a census, ie ploughing through the images page by page.

And it didn't take long.  

As I had suspected, Elsie was still local, in Upper Sapey, just a few miles from Collington, working as a general domestic servant.  She had been transcribed under the name of "Elsie Mary NAMAS"

 

After obtaining confirmation from some of the other Guild members, who agreed with me that it was "May," rather than "Mary," and also that the surname was more likely to be THOMAS, than NAMAS, I contacted FindMyPast, who also agreed and soon made the correction.

With hindsight, it becomes easy to ask, "Why didn't I try.... combination of search terms?" In my defence, all I can say is that I did my best, especially given that it was past midnight, and many of the combinations I did try were producing hundreds, if not thousands, of results. 

So I am just happy that Elsie had a job fairly locally and that the 'old' method therefore worked quite quickly.

The ongoing search for my ALLEN ancestry

I recently returned to looking at my mother's ALLEN ancestry.  The research on this line hit a 'brick wall' some years ago, with an Edward ALLEN, born about 1786 who moved into London from Hertfordshire, probably sometime around 1840.  He appears in Lambeth in the 1841 census, with his wife, Sophia, and four of their eight children, all of them recorded as born outside of the county.* 

The connection back to Hertfordshire had been made (after many years of searching!) following the discovery of the baptism of Edward's son, John Prosser ALLEN, my 2x great grandfather, in the parish of Thundridge, on one of the LDS Vital Records cds.    

In the later censuses, once the family are in London, they never seem to put their parish of birth, merely the county, and, in several of the census entries, it reads more like "Herefordshire" than "Hertfordshire".  Hence the long search for where they had come from.

Having finally identified the relevant parish, I'd viewed the parish registers on microfilm at my local LDS Family History Centre. The register for John's baptism gave the family's address as High Cross, Standon - which is actually a hamlet in the adjoining parish to Thundridge.  Working through the register, I found the baptisms of the other children who appear in the censuses, as well as discovering the burial of Edward and Sophia's eldest daughter, Louisa, in April 1841, with the address recorded as Lambeth. So that helped to confirm the connection.

But attempting to take the family line any further back stalled with the discovery that there were two Edward ALLENs born around the same time - one Edward born 1785, son of a John and Jane, and the other Edward born 1787, son of an Edward and Ann - and no obvious way to confirm which one was 'mine'.  

ALLEN is quite a 'popular' surname and so, with other activities to deal with at the time, it was easiest to pause the research there.  Returning to it again recently, I first refreshed my memory by grouping the various entries, which I'd previously collected in a spreadsheet, into families.  John and Jane had four children between 1779-1785: William, Ann, Sarah, and Edward, with a possible fifth, an Elizabeth, christened with parents, John and "Jenny", in 1789.  Edward and Ann had six children between 1781-1791: George, Christian, Hariott, Edward, Richard and John Canfield. [Note: two spellings of this name, "Canfield" and "Camfield," appear in the records so, for any references below, to searching for the name, I will have checked both versions.]  

Since the two couples were having children in the same place, over similar time periods, it seemed possible that Edward and John might be brothers.

The John Canfield ALLEN later used the middle name of "Canfield" for his first two sons, and then the middle name of "Kilham" for two others, but at least three other sons had no middle name.  In view of the fact that my Edward also only gave one of his seven sons a middle name, that seemed intriguing.  But, since I knew that this John Canfield ALLEN was not my ancestor, I had not paid a great deal of attention to him before.  

However, this time, I decided to research him, since he was the brother of one of the Edwards who 'might' be my ancestor.  Following him through the censuses - I promptly realised that, in 1851, he has his brother Edward living with him!*  Maybe I saw this and didn't realise the significance before, or maybe it wasn't available then (we forget how many records the genealogy companies have added over the years).  But, since I know that 'my' Edward had died in Lambeth in 1849, this obviously means that I can now discount the baptism entry for Edward, the son of Edward and Ann.  

So, potentially the Edward, son of John and Jane, could be the correct one for my line.  

But how could I be sure? 

As well as searching for John Canfield ALLEN on the genealogy sites, I'd looked in the National Archives catalogue, since the combination of his three names was likely to be easy to identify.  There I found a couple of relevant references, for example "Henry Allen of Wadesmill, Thundridge, son of John Canfield Allen, farmer..." and a conveyance to him of land in High Cross, Standon.

Then I tried just "Canfield AND Allen" - that resulted in a reference to a Court Roll in April 1744, which was described as "Admission of Mary Allen, widow, under will and surrender of Andrew Camfield."

Since this item is held by Hertfordshire Archives and Local Studies, I switched to their catalogue, where I found another interesting and very informative entry:

Title: Waggon and Horses

Description: (Formerly Gun and Catherine Wheel, formerly Gun), blacksmiths shop and yard; messuage abutting on its north side; 2 other messuages all in High Cross, copy will, Andrew Canfield of Highcross blacksmith proved 15 Sep 1741; copy will and codicil, Edward Allen the elder of Highcross, blacksmith, 29 Dec 1791; probate John Forster of High Cross, labourer, 1 May 1746; copy will, Edward Mason of Stanstead Abbott, cordwainer, proved 16 May 1788 (74)

Aha! Relevant names in the relevant place.  And, what was particularly interesting to me, they were both blacksmiths.  Because, although John Canfield ALLEN's occupation is generally recorded as farmer, or corn factor, the occupation of my ancestor, Edward ALLEN, was that of a blacksmith.  

Back on FindMyPast, I found a burial for an Andrew Canfield, blacksmith, in Great Hormead on 20 August 1741, which seemed to be the only one that fitted the Will reference.  Another entry among the results caught my eye - one, dated 1740, for an Andrew Canfield in the "Essex Wills Beneficiaries Index, 1675-1858"  Checking the details indicated that he was the father-in-law to the testator, a "W Allin", of Great Hormead.

The Essex Wills Beneficiaries Index is, as one might expect, an index of the beneficiaries in Wills - but there are a few 'unexpected' features.  It just covers Wills held in the Essex Record Office and primarily only lists beneficiaries who have a different surname from the testator.  The place given might be for the testator, and not the beneficiary.  Also the date given is the date of the Will, ie when it was written, not the date when it was proved, which could be some years later.  I found helpful information about the Index on the "Essex and Sussex Surnames" site at, https://essexandsuffolksurnames.co.uk/essex-wills-beneficiaries-index/ 

Having found the entry referring to Andrew CANFIELD, I then searched for ALLENs in the Beneficiaries Index and found the following, potentially relevant, entries:

BR22 383, dated 1731, testator: A Canfield [Standon] listing ALLEN beneficiaries of Mary (dau), William (son-in-law), and then William, John and Edward, all grandsons.

MR12 50, dated 1752, testator M BANGS [Standon] listing ALLEN beneficiaries of Jane (daughter) and John (son-in-law).

MR12 50, dated 1755, testator W BANGS [Standon] listing ALLEN beneficiaries of Jane (daughter) and John (son-in-law).

So, based on the first entry, Andrew CANFIELD had a daughter, Mary, who married a William ALLEN.  They had children, William, John and Edward.  Could the John and Edward be the two men in the "Edward and Ann" and "John and Jane" couples above?

At this point, several issues dawned on me:

• The marriage of Jane BANGS to John ALLEN (of Braughing), which, according to FindMyPast, took place in Standon in 1749, was not in my spreadsheet. Therefore my spreadsheet must be incomplete.
• If that's the John and Jane ALLEN having children between 1779-1785, then that's a long gap between their marriage and having children.
• Likewise, since Andrew Canfield listed his grandsons, William, John and Edward, when he wrote his Will in 1731, that John and Edward would have to be aged around fifty before having children, if they were the fathers in the two families I listed above. 

So clearly, despite the occurrence of the "right" names and occupations, the entries are unlikely to fit together in such a straightforward way as it initially might appear - which is why this is an "ongoing" search!

Next steps:

• Extract more of the parish register entries into my spreadsheet, widening the search to other parishes beyond my initial selection.
• Obtain copies of the relevant Wills, transcribe them, and confirm the relationships between the  individuals mentioned, as well as look for any further identifying features.
• Check out the tithe records relating to land and properties owned, and occupied, by the ALLENs to establish exactly who was living where.
• Explore, in more detail, the entries in the various archives' catalogues relating to land and property ownership.

The following is just a screenshot of a quick map, drawn in Google Maps and then opened in Google Earth, in order to show the relationships between the various places mentioned above.  The linear pattern is interesting, since I remember visiting an historic "Smithy" once, where it was mentioned that blacksmiths' workshops would often be along a main route. Had the family established themselves along a main route from London to Cambridge? [or is it just an artifact of how I produced the map! 😊]

* Additional information about sources:

1841 Lambeth census for Edward, Sophia and 4 children: HO107/1058/9/13/20/2387 (Piece/Book/Folio/Page/Schedule) Union Street, St Mary Lambeth, London.

1851 census for John Canfield ALLEN and his family: HO107/1705/570/1/3 (Piece/Folio/Page/Schedule) Plashes Farm, Standon, Ware, Hertfordshire, England

TNA Entry relating to the Court Roll: https://discovery.nationalarchives.gov.uk/details/r/2e9b8cb6-8ad3-48ac-98bc-c9b0d1636a79

Hertfordshire Archives and Local Studies entry about the Waggon and Horses: https://www.hertfordshirearchives.org.uk/collections/getrecord/GB46_CDESb_1_1_40_2