अब Google Groups, यूज़नेट फ़ोरम में नई पोस्ट करने या उसकी सदस्यता लेने की सुविधा नहीं देता है. हालांकि, यूज़नेट फ़ोरम में मौजूद पुराना कॉन्टेंट देखा जा सकता है.

Zenobia Monke 1555-1620

337 बार देखा गया
नहीं पढ़े गए पहले मैसेज पर जाएं

chris wertz

नहीं पढ़ी गई,
3 सित॰ 2018, 2:17:06 pm3/9/18
ईमेल पाने वाला
Hello, I am researching the Monkes and I am interested in Zenobia Monke who is shown in the "The Visitations of the County of Devon" on page 569 as the daughter of John Monke by a woman named "Bond".

That would place her in the Plantagenet line because John is the son of Frances Plantagenet.

Also, she is connected to many Americans because her daughter Elizabeth Morrison came to America.

taf

नहीं पढ़ी गई,
3 सित॰ 2018, 4:00:18 pm3/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 11:17:06 AM UTC-7, chris wertz wrote:
> Hello, I am researching the Monkes and I am interested in Zenobia Monke who
> is shown in the "The Visitations of the County of Devon" on page 569 as the
> daughter of John Monke by a woman named "Bond".

For starters, let's take it to the primary record. The 1564 visitation of Devon reports that Thomas Monke and Frances Plantagenet had 6 children, none of whom were yet married. The second given son is John. While the published edition of this visitation was taken from copies in the British Library, usually with additions, the material in question would be expected to appear in this visitation, and there are no obvious red flags. of copies, with additions, etc., this is information you would expect to be given, contemporary.

The published 1620 visitation is likewise not from the original, and other pedigrees in this edition have obvious late additions (events after 1620 appear), errors, and fraudulent material (unclear if these are in the original in the College of Arms). It again shows Thomas Monk marrying Francis (sic) Plantagenet, and by her having second son John, who married "d. of Bond" and had a daughter Zenobia. Note that the details are identical to those that Vivian repeated in Visitations of Devon, so this is his source. No further information appears concerning her, notably no marriage. This might mean that Zenobia was as yet unmarried, or simply that the information was not provided when the pedigree was entered in the record. The pedigree was not signed, but the information centers on the family of her first cousin, Thomas Monk, son of her father's older brother, whose oldest surviving child (his third son) was aged 14.

> That would place her in the Plantagenet line because John is the son of
> Frances Plantagenet.

This would indeed seem to be the case.

> Also, she is connected to many Americans because her daughter Elizabeth
> Morrison came to America.

Here is where you need to be very careful. I see there are pedigrees all over the internet claiming this relationship, but there are red flags. Many of them call her 'Zenobia le Moyne Monke'. This may seem insignificant but is a huge cause for concern. This is not the name of a person, it is a chimera. You see something like this when someone found some pedigrees that had her as le Moyne, and others had her as Monke, and rather than investigating to see which was right this person simply concluded they were different representations of the same person and combined the the. That is one option - the other is worse, that the documentation said she was le Moyne, but nothing could be found on her. However, someone found Zenobia Monke in the visitation with a glorious pedigree, and with nothing more in its favor than that Monke and Moyne both begin with 'Mo...', they concluded that Monke was what the document reporting Moyne really meant to say. Either of these would mean that Elizabeth Morrison was not daughter of Zenobia Monke (or at least not demonstrably so).

The second red flag is with the dates. Most of pedigrees I see show Elizabeth Morrison born in the neighborhood of 1580. Were this truly the case, it pushes the limits of credulity to make her granddaughter of John Monke - not only was he not yet married in 1564, his elder brother had yet to wed. Geni.com puts Zenobia's birth in 1555, which does not match with the visitation. One pedigree out there instead puts Zenobia's birth in 1580, but ridiculously places it in Charleston, South Carolina. This pedigree shows Elizabeth born 1590, just ten years after her mother. While it is possible that the relationships are real, but based on documents without dates, or the dating information was unknown to the people constructing these pedigrees, when you see a pedigree with imprecise, grossly-conflicting dates and places, it suggests that all of it is made up.

It seems likely to me that there is no actual historical reality underlying this supposed connection. Were I you I would not put too much effort into investigating the ancestry of Zenobia Monke until you shore up the connection between Elizabeth Morrison and Zenobia, and confirm that the mother wasn't actually Zenobia le Moyne (as atypical as Zenobia might seem, it was actually used commonly in 16th century Devon, so there is every reason to expect there to have been two or more women named 'Zenobia Mo...'.

So, let's start with that - what in 'known' about Elizabeth? Not what can be found in the internet genealogy echo chamber, but what do the actual primary documents say?

taf

chris wertz

नहीं पढ़ी गई,
3 सित॰ 2018, 4:22:44 pm3/9/18
ईमेल पाने वाला
Terrific. Thank you for the thorough reply. I just sent out emails looking into the sourcing of the Morrison connection. So we're on the same page there for sure.

I'll post updates on Morrison as I get them.


I'm not as worried about the "Le Moyne Monke" confusion as you are, but maybe I should be. The Monke family is generally believed to stretch back some ways to a time when they were called Le Moyne. I assumed that people who've put both names in their tree are just trying to keep the surname the same through the line. Does that makes sense? Here's a Wikipedia snippet on them:

Monk


According to the Devon antiquarian Sir William Pole (died 1635), Potheridge was the residence of the family of Monk (alias Monke, Monck, etc.) since at the latest 1287.[2]The family was recorded in ancient Norman-French charters as le Moigne[10] (modern French le moine, "the monk") or de Moigne[11] and was Latinized as Monachus,[11]from ancient Greek μοναχός (monachos), "single, solitary"[12] and Anglicised as "Monk", or "Monck". According to Tristram Risdon (died 1640) in about 1216 Roger le Moyney held one fee in the Devon parish of West Anstey and was succeeded by William le Moyney.[13] As recorded in the Book of Fees William le Moigne and Roger le Moyne held land in West Anstey from Ralph de Champeus[14] who held from the feudal barony of Barnstaple.[15] The family is memorialised by today's "Money Common" in that parish. Another part of Anstey was held by the feudal barony of Okehampton.[15] The descent of the family of "Monk of Potheridge" is given as follows in the Heraldic Visitation of Devon:[16]

rodc...@gmail.com

नहीं पढ़ी गई,
3 सित॰ 2018, 9:53:23 pm3/9/18
ईमेल पाने वाला
I don't see multiple variations on names as a huge red flag. I come across multiple spellings routinely in cases where literacy is not prevalent and also when people translate names into English.

Even in the preface of Vivian's Visitations of 1531, 1564 and 1620 he states he was getting numerous complaints about variations in names. See the second paragraph, which is completely taken up with this issue. Since this was an issue already by 1620 I think we can at most say such practice in Internet times is a continuation, not that it is the source of all such name variation.

Dates are an issue I agree, if John Monke had not yet married until after 1564, contradicting a birth date of 1555 for Zenobia. Where does the 'married after 1564' claim come from?

I've seen gross errors like Zenobia being born in South Carolina instead of Devon for many people. I've even seen ones which would be the equivalent to Devon, South Carolina, mixing two locations together. I have encountered many examples of reuse of a name by European settlers in North America but when I see such I check the plausibility of such mixes. In cases where I determine it is indeed an error I agree the source's reliability is downgraded for everything it attests to but I tend to not allow that one source to taint all others which might agree with it on other details.

Excellent detail on the name Zenobia.

taf

नहीं पढ़ी गई,
3 सित॰ 2018, 11:18:50 pm3/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 1:22:44 PM UTC-7, chris wertz wrote:
> According to the Devon antiquarian Sir William Pole (died 1635),
> Potheridge was the residence of the family of Monk (alias Monke,
> Monck, etc.) since at the latest 1287.[2]The family was recorded
> in ancient Norman-French charters as le Moigne[10] (modern French
> le moine, "the monk") or de Moigne[11]

Well, that was pretty obvious. It still tells me something to the type of genealogist who have put the line together.

Looking at the online pedigrees, they show Elizabeth Morrison married to James Willmarth, but this is matched with a marriage of "James Wilymott gentl and Elizabeth Morison", married 21 May 1618 at St. Mildred Poultry, London.

The records of St. Mildred Poultry were published in 1874, and referring to this marriage are two footnotes:

* James Willymot, of Kellshull co. Hertford, baptized 19 Novr. 1581, died 12 and buried 15 Sept., 1662, at Kellshull; father of James Willymot, esquire, of Kellshull, J.P., Deputy-Lieutenant and High Sheriff of co. Hertford; and of Thomas Willymot, of Royston co. Hertford.

+ Second daughter of Thomas Morison, esquire, buried at Kellshull, 6 June, 1634, age 34."

Where might this come from? The burial information is clearly from a church record, but for a Hertfordshire gentryman during this period, the visitation is a good place to look, and in the 1635 visitation of Herts, we find a pedigree of Willimot of Kelshall. This pedigree shows James Willimot of Kelshall, living in 1634, married to Elizabeth Morrison of Sandon, Herts, with children James, aged 13, Thomas, Elizabeth, Anne, Mary & Hellyn.

The Herts visitation also shows a pedigree for Morrison of Sandon. This does not show an Elizabeth married to Willimot, but the pedigree is not fully elaborated - it gives no indication who was alive at the time of the visitation, nor the age of the heiress. However, I think we are on the right track - Charles Morrison of Sandon married at St Mildred Poultry in 1627 to Anne Hill. The visitation pedigree runs as follow:

Thomas of Cadeby, Lincs married ".... da. of .... Moyne"

Thomas of Sandon, 2d son (with brother Edward remaining in Cadeby) married Ellin dau, of Edw Pulter of Bradfield

Charles of Sandon, m1 Elizabeth da Francis Montfort, m2 Anne da. of Francis Allen of London.

Elizabeth

My best bet is that the Anne Allen of London is the Anne of the St Mildred Poultry marriage, that just as he was a widower, she was a widow. Elizabeth, wife of Willimot was most likely the daughter of Thomas of Sandon who married Ellen, although if Charles was a lot older, she could be the daughter of Charles at the base of the pedigree, mistakenly assigned to a Thomas in the Willimot pedigree. (I think this unlikely - there is a Pulter pedigree, and though it does not name Ellen, it has an Edward where you would expect him, as grandfather of the informant,which would slot him two generations before Elizabeth Morrison (though it doesn't always work that way).

Note, though that at the top of this pedigree we have a Thomas Morrison married to a Moyne. That is clearly the origin of the Moyne name assigned to Zenobia. How someone decided to translate Moyne to Monk so that they could then assign this person as the Zenobia Monk of the Devon visitation, who can tell, but we are clearly not on the same page. Zenobia was of the generation of the person providing the information in 1620, while this Moyne, wife of Thomas Morrison, falls two generations before the person apparently supplying the information in 1634 - they clearly belonged to different generations.

Well, it has worked so far, so let's look for Morrison in Lincs. Sure enough, the Lincolnshire visitation of 1592 has a Morison pedigree, and it shows Thomas Morysine of Sandon Herts, husband of Ellen daughter of Edward Powlter, as second son of Thomas Morysone of Cadby, and his wife Elizabeth daughter and co-heiress of Thomas Moyne by Bridget, daughter of Sir William Hansard of North Kelsey, Lincs. There is no Moyne in the visitation, nor does Hansard appear there, but there is a Hansard pedigree in the 1562 visitation that named Bridget but marries her to Vincent Grantham. The solution to this is found in a published collection of Lincolnshire pedigrees, where Bridget is shown marrying fist Thomas Moigne of North Willingham, attainted and executed 1536, second to Vincent Grantham, and third Thomas Taylor of Lincs. Elsewhere in the same volume is a pedigree for Moigne of North Willingham, which shows several generation above Thomas, husband of Bridget and father of Elizabeth, wife of Thomas Moryson of Cadeby. (I can't help notice that Thomas Moigne's grandmother was Joan, daughter of Robert Sheffield of Butterwick. This means he was likely a relative of the Belwood family that produced Anne, wife of mayor John Browne of London, and matriarch of a cluster of mayors and wives in following generations, that we have discussed here periodically. Thomas has a Wikipedia page, and is in History of Parliament.)

So, whoever put together seems to have only gone as far as the Herts visitation, where they found a _____ Moyne (in the process apparently skipping a generation between Elizabeth Morrison Willimot and the Thomas Morrison who married the Moyne. They presumably looked for a Moyne available, and knowing that it translated as Monk, they found a Zenobia Monk who was granddaughter of a Plantagenet just too good to pass up, so they merged the two and gave her the dual name, Zenobia le Moyne Monk to reaffirm the connection. They were obviously unaware that the Lincs visitation gave the full name of this ____ Moyne as Elizabeth, and named her father Thomas, also giving her mother and maternal grandfather, and that this would have enabled them to trace the family back many many generations (but would not have involved anyone with the surname Plantagenet).

So, Zenobia is out, but the other end of this needs a closer look. Before spending too much time on following the ancestry of James Willimot and Elizabeth Morrison, you had best make sure the connections that lead back to them have any validity. It was all too common for someone with an immigrant to just look for someone with a similar name in a visitation and decide it must be them, even if the information found in the visitation itself seems contradictory to the immigrant's history. Here we have seen that someone did that to link Zenobia, but you started this thread saying that Elizabeth came to America - this appears not to have been the case, so what is up with that?

Best I can tell, the immigrant appears as Thomas Willmarth. Somebody has decided this person was the same as Thomas, second surviving son of James and Elizabeth Willimot. Not impossible, but the surname shift seems forced to me. You really need to consider the possibility that this too is based on a connection that was nothing but a groundless guess that amounts to nothing but wishful thinking.

taf

taf

नहीं पढ़ी गई,
3 सित॰ 2018, 11:40:01 pm3/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 6:53:23 PM UTC-7, Rod Carty wrote:
> I don't see multiple variations on names as a huge red flag. I come across
> multiple spellings routinely in cases where literacy is not prevalent and
> also when people translate names into English.

It is not multiple spelling variants that concerns me, it is when someone enters information in their database that way - it suggests they don't really appreciate what it is they are doing. In particular, I have seen way too many cases of chimera like I describe, and a public database entering two alternative names like this as if the person had multiple names almost always belongs to someone who doesn't really know what they are doing, or who blindly copied their database without evaluating it, which amounts to the same thing.

In this case, repeating the chain of research from scratch does indeed show that Moyne was attached as a way of establishing a bogus connection between Zenobia Monke and ____ Moyne, wife of Thomas Morrison, just as I suspected.


> Even in the preface of Vivian's Visitations of 1531, 1564 and 1620 he states
> he was getting numerous complaints about variations in names.

And yet he only shows one in each generation.

> See the second paragraph, which is completely taken up with this issue.
> Since this was an issue already by 1620 I think we can at most say such
> practice in Internet times is a continuation, not that it is the source
> of all such name variation.

And yet it isn;t in this case. In this case it is due to someone with a Hertfordshire dead end trying to force a connection to a Devonshire family.

> Dates are an issue I agree, if John Monke had not yet married until after
> 1564, contradicting a birth date of 1555 for Zenobia. Where does the
> 'married after 1564' claim come from?

The 1564 visitation shows them without indication of marriage, which would have been very unusual given that their father was the apparent informant. Since Elizabeth turned out to be born about 1600, the chronology of the Zenobia connection turns out to have been sound - it is just the descent itself that wasn't, made, as I suspected, by splicing together two distinct women, one named Moyne, the other named Monk.

> I've seen gross errors like Zenobia being born in South Carolina instead
> of Devon for many people. I've even seen ones which would be the equivalent
> to Devon, South Carolina, mixing two locations together. I have encountered
> many examples of reuse of a name by European settlers in North America but
> when I see such I check the plausibility of such mixes. In cases where I
> determine it is indeed an error I agree the source's reliability is
> downgraded for everything it attests to but I tend to not allow that one
> source to taint all others which might agree with it on other details.

When you see a range of dates and a range of places, it usually means that people are playing fill in the blank, and often are doing so because there are blanks - that the information may not have been compiled from primary records but slapped together through guesswork. It is a red flag. It doesn't mean the material is wrong, but you had best turn your skepticism dial up to 11.

taf

Andrew Lancaster

नहीं पढ़ी गई,
4 सित॰ 2018, 5:31:30 am4/9/18
ईमेल पाने वाला
On Tuesday, September 4, 2018 at 3:53:23 AM UTC+2, Rod Carty wrote:
> I don't see multiple variations on names as a huge red flag. I come across multiple spellings routinely in cases where literacy is not prevalent and also when people translate names into English.

Just adding to the replies already given, someone may be thinking that this particular type of "spelling variant" (which is clearly more than just a spelling variant, because the pronunciation had to be different) could indeed make sense in medieval contexts where surnames referred to a real description, and could thus be translated.

...However, in this early modern period I think that would be extraordinary.

That leaves transcription errors and other simple mistakes but these are always "last resort" explanations because essentially arbitrary. Anything can be a mistake, but in this case there is no known reason for thinking there was one.


chris wertz

नहीं पढ़ी गई,
4 सित॰ 2018, 8:38:42 am4/9/18
ईमेल पाने वाला
This is really impressive. Thank you. Great work.

Since I found this line I was worried about the lack of sourcing for these two pairings. But I thought the uniquity of that name Zenobia together with the well established Plantagenet line might be enough to tie this together. I never considered that the mistake was (one of them anyway!) that the person in fact was not Zenobia.

Richard Smith

नहीं पढ़ी गई,
4 सित॰ 2018, 9:05:58 am4/9/18
ईमेल पाने वाला
On 04/09/18 10:31, Andrew Lancaster wrote:

> Just adding to the replies already given, someone may be thinking
> that this particular type of "spelling variant" (which is clearly
> more than just a spelling variant, because the pronunciation had to
> be different) could indeed make sense in medieval contexts where
> surnames referred to a real description, and could thus be
> translated.
>
> ...However, in this early modern period I think that would be
> extraordinary.

I'm only familiar with one such example, and even there I'm a little
sceptical that the generally accepted version is true.

In about 1536 in Brabant, John Rogers, the Marian martyr, married a
woman who is variously given the surname Pratt or de Weyden. The
difference is surnames is commonly explained by them being translated
forms of the same word: "weiden" is the Dutch word for meadows, and
"prata" is the Latin word, which, so the theory goes, has been
Anglicised to give Pratt. I must try to investigating this further.
Given how much nonsense has been written about John Rogers over the
years, I would be particularly surprised to discover two people had been
conflated, though if so, it is not a modern error and dates at least to
the 17th century.

Richard

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 9:31:58 am4/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 8:18:50 PM UTC-7, taf wrote:

> The visitation pedigree runs as follow:
>
> Thomas of Cadeby, Lincs married ".... da. of .... Moyne"
>
> Thomas of Sandon, 2d son (with brother Edward remaining in Cadeby) married Ellin dau, of Edw Pulter of Bradfield
>
> Charles of Sandon, m1 Elizabeth da Francis Montfort, m2 Anne da. of Francis Allen of London.
>
> Elizabeth
>
> My best bet is that the Anne Allen of London is the Anne of the St Mildred Poultry marriage, that just as he was a widower, she was a widow. Elizabeth, wife of Willimot was most likely the daughter of Thomas of Sandon who married Ellen, although if Charles was a lot older, she could be the daughter of Charles at the base of the pedigree, mistakenly assigned to a Thomas in the Willimot pedigree. (I think this unlikely - there is a Pulter pedigree, and though it does not name Ellen, it has an Edward where you would expect him, as grandfather of the informant,which would slot him two generations before Elizabeth Morrison (though it doesn't always work that way).
>

OK, I can resolve this now. Elizabeth, daughter of Charles Morrison of Sandon
and his first wife, was bapt. at Sandon, 1626. (He had a second daughter, Margaret, bap. 1630, who must have died by 1634.) Meanwhile, Charles' grandfather, Thomas Morrison of Cadeby, husband of Elizabeth Moyne, is in History of Parliament, where it places their marriage ca. 1560. Since Elizabeth, wife of James Willimott, was born ca. 1600, she was clearly a sister of Charles, and daughter of Thomas of Sandon and Ellen (Helen, Eleanor) Poulter and granddaughter of Thomas and Elizabeth Moyne.

taf.

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 10:35:16 am4/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 8:18:50 PM UTC-7, taf wrote:

> Best I can tell, the immigrant appears as Thomas Willmarth. Somebody has
> decided this person was the same as Thomas, second surviving son of James
> and Elizabeth Willimot. Not impossible, but the surname shift seems forced
> to me. You really need to consider the possibility that this too is based
> on a connection that was nothing but a groundless guess that amounts to
> nothing but wishful thinking.

I see that James Willymott of Kelshall left a PCC will (1662). Unfortunately, due to a poor-quality scan where I am looking, it is hard to read what exactly he is leaving Thomas, but it looks like some property rights, and later 'tenne pounds', but I see nothing suggesting Thomas was in America. at the time.

and to add some colour, I find a snippet reporting a Hertfordshire suit claiming "That Richard Totnam of Kelshall, weaver, did maliciously defame James Willymott, esquire, JP, at Tharfield, by speaking the following words, "I doe not care a turd for James Willymott and although he was a justice of the peace yett I would be a justice of the quorum." This is from 1666, so would refer to James 'the younger', the eldest son of James and Elizabeth. (for whom see his brief entry in Alumni Cantabrigienses)

taf

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 11:21:08 am4/9/18
ईमेल पाने वाला
On Tuesday, September 4, 2018 at 7:35:16 AM UTC-7, taf wrote:
>
> I see that James Willymott of Kelshall left a PCC will (1662). Unfortunately,
> due to a poor-quality scan where I am looking, it is hard to read what exactly
> he is leaving Thomas, but it looks like some property rights, and later 'tenne
> pounds', but I see nothing suggesting Thomas was in America. at the time.
>

Continuing on this thread, I find in A History of Royston, Hertfordshire, p. 152, the following: "This James Wilmot, or Willymot, lived at Kelshall and was a Justice of the Peace of some note. He was High Sheriff for Herts. in 1683, and was also a Justice of the Peace for the Isle of Ely. He was the son of James Willymot of Kelshall, who, says Chauncey, "was much valued for his skill and care and honesty and fidelity in his profession, by which means he improved his estate, kept a free and generous table, and lived hospitably amongst his neighbours." Another son was Thomas Willymot, of Royston, whose son, William Willymot, the scholar, will be found referred to under the head of " Royston Worthies."

VCH Herts says of the manor of Woodhall alias Philpots, "Henry Philpott was holding in 1650, (fn. 49) but apparently he also released his right to the Willymots, and the manor was given by James Willymot to his younger son Thomas. (fn. 50) He married Rachel the daughter of Dr. Pindar, and the manor was settled on her. (fn. 51) "

Note the chronology is long here - Thomas and Rachel start having children in Royston in 1666. That is 45 years after the birth of James Willymott the younger. This is important, because it raises the possibility that Thomas of Royston was younger son of James the younger, rather than of James the elder (leaving the forlorn hope that the sources that say otherwise are mistaken and Thomas, son of James the elder, is still available to be the immigrant). That being said, the PCC will of James the younger names sons James & Ralph and daughters Deborah, Anne and Dorothy, along with their husbands - no Thomas, so though the second son in 1634, I think the daughters likely fell in between and Thomas was born, in the 1630-1634 range, marrying Rachel when he was in his early-to-mid 30s, which was not all that unusual.

Bottom line - barring some extraordinary evidence in favor of the identification, Thomas Willmarth of Rehoboth, Massachusetts, was not the same as Thomas Willymott, second son of James Willymott the elder of Kelshall. The whole thing was a wild goose chase. This is, unfortunately, all too common with identifications of the English origins of American immigrants.

taf

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 4:25:30 pm4/9/18
ईमेल पाने वाला
On Monday, September 3, 2018 at 8:18:50 PM UTC-7, taf wrote:
> Sure enough, the Lincolnshire visitation of 1592 has a Morison pedigree,
> and it shows Thomas Morysine of Sandon Herts, husband of Ellen daughter
> of Edward Powlter, as second son of Thomas Morysone of Cadby, and his wife
> Elizabeth daughter and co-heiress of Thomas Moyne by Bridget, daughter of
> Sir William Hansard of North Kelsey, Lincs.

I suspect a lot of people have lost interest given the decapitation, but I note that Thomas Morison of 'Caudebie' (or ? Candebey) Lincs and St Botolph without Aldersgate, London, left a PCC will prob 1597, naming sons Fines/Fynes, Richard, Henrie/Hairie, Edward (and his children Thomas, Francis and Elizabeth Morison), Thomas (and Helen his wife, left his belongings at Sandon, and their daughter Ann); son(-in-law) George Alwington (and sons Charles and Hughe Alwington) and son(-in-law Frances Mussenden); daughters Jane Alwington and Faithe Mussenden; 'brother William Morison his children'; nephew Thomas Morison, son of brother Henrie. (I think that is all of them, it is a rambling will that jumps back and forth among family members so I may have missed someone in my quick read.)

taf

chris wertz

नहीं पढ़ी गई,
4 सित॰ 2018, 4:52:37 pm4/9/18
ईमेल पाने वाला
I haven't lost interest at all. Many of the people I invited to read this have access to the records of these people on some of the most used sites. So I know what you have found is very helpful.

I can't help but wonder what ever happened to Zenobia. She just lost her whole American clan.


When I get time later, I'm going to start digging into the Wilmarths. I think I'll start with
Elizabeth Marie Willmarth.
1647–1690
BIRTH 4 APR 1647 • Rehoboth, Bristol, Massachusetts, USA
DEATH 4 OCTOBER 1690 • Rehoboth Township, Bristol County, MA

She used to be the great-granddaughter of Zenobia.

Though there's a problem with her too. In most trees she's sourced under this name and under Elizabeth Willmott. One or both of these women married Jonathan Fuller around 1664.

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 5:31:28 pm4/9/18
ईमेल पाने वाला
And while we are addressing slip-shod genealogy, here is a piece that has found its way all the way into the ODNB:

The 1634 visitation Morrison pedigree shows Thomas Morrison of Sandon and his wife Helen/Ellen to have had two sons, Charles and Richard. The same volume includes an appendix of pedigrees of county families. This shows a pedigree for Morison of Cashiobury that begins with Sir Richard, husband of Bridget Hussey, father of Sir Charles and grandfather of another Sir Charles. It is important to note that in no place in either pedigree is there the slightest indication that these might be the same person.

Nonetheless, Christina Hallowell Garrett, in her 1938 work, The Marian Exiles, decided the entries referred to the same person, knowingly doing so in the face of people got it in their heads that they must be the same person. This is done in spite of the fact that Richard of Sandon in the visitation was the brother of the apparent 1634 informant, while Richard of Cassiobury of the non-visitation pedigree was a man who died in 1556, and likewise in spite of the fact that both Venn and Foster called the Cassiobury man son of a different Thomas Morrison of Chardwell, Yorks.

ODNB is apparently following Garrett, showing Robert Morrison as child of Thomas Morrison of Sandon, by the daughter of Thomas Merry of Hatfield. This wife belongs to the other Thomas Morrison - in 1874, in an article on The Stately Homes of England, in The Art Journal, S. C. Hall writes an account of the Morrisons that begins with William Morrison of Chardwell. His grandson (son of his won William) was Thomas of Chardwell, who married the daughter of Thomas Merry of Hatfield, and by her was father of Richard of Cassiobury.

So, ODNB is replacing the husband with a chronologically impossible alternative (the Morrisons didn't even hold Sandon when Richard was born), yet retaining the wife that belonged to the other guy.

HOP calls the parents Thomas Morrison of Herts and the daughter of Thomas Merry of Hatfield.

taf

chris wertz

नहीं पढ़ी गई,
4 सित॰ 2018, 5:46:25 pm4/9/18
ईमेल पाने वाला
At least I wasn't the only one confused by this line.

taf

नहीं पढ़ी गई,
4 सित॰ 2018, 8:33:26 pm4/9/18
ईमेल पाने वाला
On Tuesday, September 4, 2018 at 2:46:25 PM UTC-7, chris wertz wrote:

> At least I wasn't the only one confused by this line.

There is a long history of this type of thing - name's-the-same 'identifications' that ignore chronology and other factors. It is particularly pervasive with regard to the visitations. Until very recently, they were the one source for names from this period that received broad distribution in America, so people would go and find someone of the same name, and conclude the connection based on nothing more than this coincidence. As we have seen, the match didn't have to be very close. This 'solution' then gets into circulation and is accepted as legit because nobody bothers to go back to the original records and reconstruct where it came from. That is the true lesson here. Don't trust anything unless you see what it is based on.

taf

chris wertz

नहीं पढ़ी गई,
4 सित॰ 2018, 9:09:03 pm4/9/18
ईमेल पाने वाला
this has been quite a lesson. I consider myself very careful, but this has taught me that I have to be even more careful. I have to dig into why this woman has two different last names and two different birth dates.

Andrew Lancaster

नहीं पढ़ी गई,
5 सित॰ 2018, 10:47:31 am5/9/18
ईमेल पाने वाला
I have seen another example from the 1500s in Belgium, just doing local genealogy. A Boon (meaning Bean, or at least it was interpreted that way) family was sometimes translated. So I think it was still possible to happen in Belgium.

Andrew

taf

नहीं पढ़ी गई,
5 सित॰ 2018, 12:41:34 pm5/9/18
ईमेल पाने वाला
Lest there be any question of Thomas of Royston being the Thomas, younger son of James of Kellshall, there is a pedigree of the family in Chauncy's Historical Antiquities of Hertfordshire includes a pedigree that is said to be taken from the 1634 and 1669 visitations together. This is not really accurate, as it contains many details from after 1669, so it clearly has additions from an uncredited source or sources. This is often a cause for concern, but in this case, it is indicating contemporary recordkeeping - Chauncy began writing his work in 1680, and originally published it in 1700, with the children of James Willymott and Elizabeth Morrison still living.

The 1634 visitation shows James Willimott marrying Elizabeth, daughter of Thomas Morrison of Sandon and having James, eldest son, aged 13 in 1634, Thomas, Elizabeth, Anne, Mary and Heleyn. Chauncy's pedigree shows James and Elizabeth having the same six children (in the order he gives them, the numbers representing birth order): Thomas (2) of Royston m. dau. Dr. Pindar of Springfield, Essex; James (1) m. Elizabeth dau. Alex Dorrington of the Inner Temple; Elizabeth (1) m. William Turner of Throcking; Anne (2) m. John Willymot; Mary (3) m. Richard Fordham; Helen (4) m. Richard Ball, D.D.

The next generation shows as children of Thomas: Thomas (1) a proctor in Civil Law; William (2); Jane (1), o.sp.; Rachel (2); Jane (3). These match the children baptized to Thomas and Rachel Willymott at Royston, except the last: Jane 1666, Rachel 1668, Thomas 1670, William 1672, and Elizabeth, 1674. Perhaps Elizabeth came to be called Jane after her sister died, or it could be an editorial slip. Then we have Ralph (3) b. 21 Dec. 1571; James (2) b. 10 Aug. 1570; James (1) b. 26 Dec. 1662 died young; Mary (1) died young; Mary (2) died unmarried; Elizabeth (3) died unmarried; Mary (4) b. 15 May 1666; Anne (5) b. 1 April 1668; Judith (6) died young; Debora (7) 20 July 1673; Martha (8) b. 23 March 1676 as children of James and Elizabeth Dorrington.

So again, Thomas did not go to Rehoboth, he went to Royston, and we have this from someone writing during his own life. As an aside, he also explicitly makes Elizabeth (Morrison) Willymott the sister of Charles Morrison of Sandon.

taf

chris wertz

नहीं पढ़ी गई,
8 सित॰ 2018, 2:04:04 pm8/9/18
ईमेल पाने वाला
First of all, thank you again for taking this on. I think this was an important link to explore and ultimately disprove. I invited the overseers of the WikiTrees for these all of these people to follow this thread so we can have a public source of the right information.

I have been too busy lately to look into the American end I was looking for to help try and straighten out how it went wrong on the trip back to England. But I'll get to that soon.

To recap:
You showed

that Zenobia Monke was indeed in the Plantagenet line, but her spouse and descendents are yet unknown.

that Thomas Morrison married a Elizabeth Moyne, not Zenobia.

that da. of Thomas, Elizabeth Morrison married James Wilmott

that this Wilmott is unlikely to have traveled to America and is probably not the James Wilmarth who died in Massachusetts.

So the current trees out there that link the Wilmarths to Zenobia are wrong in that they not only do not link to Zenobia, but they also do not link to the Morrisons that are cited.

I am still curious about the person Elizabeth Marie Willmarth.
1647–1690
BIRTH 4 APR 1647 • Rehoboth, Bristol, Massachusetts, USA
DEATH 4 OCTOBER 1690 • Rehoboth Township, Bristol County, MA


I have found her as Willmarth and Willmott in sources such as Massachusetts, Town and Vital Records, 1620-1988 and U.S. and International Marriage Records, 1560-1900. It does make for an odd coincidence.




taf

नहीं पढ़ी गई,
8 सित॰ 2018, 3:57:15 pm8/9/18
ईमेल पाने वाला
On Saturday, September 8, 2018 at 11:04:04 AM UTC-7, chris wertz wrote:
> To recap:
> You showed
>
> that Zenobia Monke was indeed in the Plantagenet line, but her spouse
> and descendents are yet unknown.

. . . if any . . . I don't think it can be taken for granted that she ever married. Given where I would ballpark her birth to have been, say 1475, she should have been married by the time of the visitation that names her, but no spouse is given - might she have died while unmarried? I note that some online pedigrees give her father John Monck another daughter, Elizabeth, and I have not taken the time to track this claimed connection down, but I would be highly skeptical. While visitations often leave out some or all daughters when there are sons, I would not expect one going to the trouble of giving the father a sole heiress without a spouse (and hence it was not selected to give a genealogical connection) to the exclusion of a coheiress (plus the Plantagenet link is just too fine a prize, and creative genealogical inventions to connect to it would be entirely expected, as we have seen with the Zenobia case.

> that Thomas Morrison married a Elizabeth Moyne, not Zenobia.

Yes, though the form Moigne seems more common.

> that da. of Thomas, Elizabeth Morrison married James Wilmott

Just to be clear, this Elizabeth is daughter of Thomas Morrison (the younger), and granddaughter of Thomas Morrison (the elder) and Elizabeth Moigne

> that this Wilmott is unlikely to have traveled to America and is probably
> not the James Wilmarth who died in Massachusetts.

This Willimott certainly did not travel to America. He is found repeatedly in the same parish from his baptism to those of his children to his wife's burial to his own, as well as appearing in documents with his adult son in between the last two named events. Equally importantly, his son James was a man of prominence in the area, a Justice of the Peace who like his father leaves a continuous record in England. A Rehoboth Thomas Wilmarth has been attached as a child of James and Elizabeth, and they did have a son Thomas, but he likewise is shown by a contemporary account of the family to have resided on Royston and to have had a family there born in the 1660s and 1670s, completely inconsistent with the Massachusetts man.

> So the current trees out there that link the Wilmarths to Zenobia are wrong
> in that they not only do not link to Zenobia, but they also do not link to
> the Morrisons that are cited.

Or even to these specific Willimotts.

> I am still curious about the person Elizabeth Marie Willmarth.
> 1647–1690
> BIRTH 4 APR 1647 • Rehoboth, Bristol, Massachusetts, USA
> DEATH 4 OCTOBER 1690 • Rehoboth Township, Bristol County, MA
>
>
> I have found her as Willmarth and Willmott in sources such as Massachusetts,
> Town and Vital Records, 1620-1988 and U.S. and International Marriage
> Records, 1560-1900. It does make for an odd coincidence.

Not really. The Willmarths may well have been distantly related to the Willimotts we have been discussing, but they did not derive from the specific couple of James and Elizabeth (Morrison). However, there is no reason to think this relationship was a close one, vs a distant one, vs the two just happening to end up with similar surnames. Elizabeth was a relatively common name, so there is nothing of note there. 'Elizabeth Maria' would be very unusual though. Multiple names were extremely rare at this time. Thus I suspect that somehow some apocryphal information has slipped in (and along these lines, if you have not already done so you should confirm the forms of the surname found in the original records, that Willmott really does appear in the records and does not appear in secondary sources because it was written by someone who already assumed there was a connection with the Hertfordshire family and wrote it down accordingly, or was extrapolated back from a form used by descendants).

taf

taf

नहीं पढ़ी गई,
9 सित॰ 2018, 6:14:22 pm9/9/18
ईमेल पाने वाला
On Saturday, September 8, 2018 at 12:57:15 PM UTC-7, taf wrote:

> I note that some online pedigrees give her father John Monck another
> daughter, Elizabeth, and I have not taken the time to track this claimed
> connection down, but I would be highly skeptical. While visitations
> often leave out some or all daughters when there are sons, I would not
> expect one going to the trouble of giving the father a sole heiress
> without a spouse (and hence it was not selected to give a genealogical
> connection) to the exclusion of a coheiress (plus the Plantagenet link
> is just too fine a prize, and creative genealogical inventions to connect
> to it would be entirely expected, as we have seen with the Zenobia case.

I have dug into this some more. The claim is that immigrant Ralph Farnham of Andover was son of Robert Farnham and Elizabeth Monk, with Robert being son of Robert Farnham of Quorndon and Mary Langham, while Elizabeth is given as sister of Zenobia. I can't find the origins of this - it may derive from a marriage record I have yet to turn up. What I can say is that the elder Robert had three sons by Mary Langham, none of them named Robert. THis is shown by a 1563 visitation pedigree, taken when he was dead (so there does not seem to have been a second marriage that could have produced such a son. As to Elizabeth Monk, it is pretty clear that, like with her husband, she has just been attached to the best available father of the same surname (the one who was great-grandson of a king), in the complete absence of evidence.

taf

JBrand

नहीं पढ़ी गई,
9 सित॰ 2018, 10:18:42 pm9/9/18
ईमेल पाने वाला
The English origin of Ralph Farnham of Andover has been published, possibly in TAG in the 1990s. He was from a yeoman family in co. Kent, I believe.

John Higgins

नहीं पढ़ी गई,
10 सित॰ 2018, 12:05:37 am10/9/18
ईमेल पाने वाला
Perhaps this is the article?

Russell C. Farnham, The English Origin of Ralph Farnham of Ipswich, MA
TAG vol. 69 (1994) pp. 32-36

taf

नहीं पढ़ी गई,
10 सित॰ 2018, 1:44:43 am10/9/18
ईमेल पाने वाला
On Wednesday, September 5, 2018 at 9:41:34 AM UTC-7, taf wrote:

> The next generation shows as children of Thomas: Thomas (1) a proctor
> in Civil Law; William (2); Jane (1), o.sp.; Rachel (2); Jane (3).
> These match the children baptized to Thomas and Rachel Willymott at
> Royston, except the last: Jane 1666, Rachel 1668, Thomas 1670, William
> 1672, and Elizabeth, 1674. Perhaps Elizabeth came to be called Jane
> after her sister died, or it could be an editorial slip.

Or an error of a different kind. There was a Jane Willymott, daughter of Thomas and Rachel, baptized in London in 1677. The likely scenario is that the total list of daughters was Jane, Rachel, Elizabeth and Jane, and Chauncy omitted Elizabeth, who perhaps d. inf.

taf

ravinma...@yahoo.com

नहीं पढ़ी गई,
10 सित॰ 2018, 11:28:08 am10/9/18
ईमेल पाने वाला
Yes, that sounds right. This is one of my ancestral families.

I'm thinking that only Ralph and his wife Alice (and some children) were found, with no statement about his parentage.

wjhonson

नहीं पढ़ी गई,
12 सित॰ 2018, 3:27:27 pm12/9/18
ईमेल पाने वाला

taf

नहीं पढ़ी गई,
12 सित॰ 2018, 4:44:47 pm12/9/18
ईमेल पाने वाला
On Wednesday, September 12, 2018 at 12:27:27 PM UTC-7, wjhonson wrote:
> Elizabeth (Moigne) Morrison
> wife of Thomas Morrison of Cadeby, co Linc; Clerk of the Pipe
>
>
> https://wc.rootsweb.com/cgi-bin/igm.cgi?op=GET&db=wjhonson&id=I167745

Except the line via her son Thomas is rather incomplete. See:
https://archive.org/stream/visitationshert00philgoog#page/n91/mode/2up

I think it likely the son John shown in the Lincolnshire Pedigrees is authentic, dying in infancy. You will note, though that your pedigree is at odds with this one regarding the wives of Charles, plus there are all those other siblings. Also the second wife shown, Anne Allen, is almost certainly identical to the Anne Hill (presumably widow) who married Charles Morrison in 1627 at St Mildred Poultry (where the Willimott/Morrison marriage took place a few years earlier). See also Chauncy's writings on Moryson and Willymot (p. 164 and following):

https://archive.org/stream/historicalantiq03chaugoog#page/n203

Also, p. 52:
https://dcms.lds.org/delivery/DeliveryManagerServlet?dps_pid=IE3739027

(One of your links, the Internet Archive, is giving me an error message. From the link I take it this is Lincolnshire Pedigrees, vol. 1, but what page did you intend?)

taf

taf

नहीं पढ़ी गई,
12 सित॰ 2018, 9:11:38 pm12/9/18
ईमेल पाने वाला
On Wednesday, September 12, 2018 at 12:27:27 PM UTC-7, wjhonson wrote:
> Elizabeth (Moigne) Morrison
> wife of Thomas Morrison of Cadeby, co Linc; Clerk of the Pipe

As long as we are sharing links:

Thomas Moryson:
https://en.wikipedia.org/wiki/Thomas_Moryson

his son Fynes Moryson:
https://en.wikipedia.org/wiki/Fynes_Moryson
https://en.wikisource.org/wiki/Moryson,_Fynes_(DNB00)
(note in particular his is credited with the first use, in pring, of the phrase "Merry Christmas".

another son Richard Moryson:
https://en.wikipedia.org/wiki/Richard_Moryson

and Robert's son, Thomas' grandson Francis Moryson, Governor of Virginia:
https://en.wikipedia.org/wiki/Francis_Moryson

and a fine involving a lot of the families involved in the generation of Thomas' children: Sir Thomas Cecil, Robert Wingfield, Hugh Alyngton, Francis Mussenden, and Thomas Taylor, plaintiffs, and Thomas Willymott, deforciant.
http://discovery.nationalarchives.gov.uk/details/r/1eec7d62-51e9-448a-ada6-518be08922fd

taf
0 नया मैसेज