Gurney Genealogy Library Updated 4 August 2026 by Allen Gurney

These research notes are provided as-is and contain supplementary working research.

Identity — Ackworth, Yorkshire

John Gurnoe married Mary Burton at Ackworth St Cuthbert, Yorkshire, on 6 June 1636, and a child was baptised at the same parish seven months later, in January 1637. Those two records are the parish’s entire Gurney presence, and the register has now been read page by page to prove it.[1][2]

It is the only unaccounted-for John × Mary marriage in England before 1660, on five independent instruments, and the only candidate in the field scoring a father named John and a mother named Mary together. It stands at 8%, second among named candidates.

And it is no longer an origin candidate. The groom left no baptism, no family and no prior record in the parish where he married, which makes him an incomer; seventeenth-century couples married in the bride’s parish. Ackworth is Mary Burton’s parish, not John Gurnoe’s — so every leaf pulled there tests the wrong parish for the question of where this man came from. What Ackworth can settle is whether the household existed and whether it left, and that is now nearly settled in the affirmative. The row should be read as a possible departure point, and the origin question runs through two other doors: Mary Burton’s own family, and wherever the surname came from.[3]

Why the row is as high as it is

The marriage is the hardest requirement in the colonial signature, and this is the only one left standing. Every John Gurney-cluster groom in England 1600–1660 is now named with his bride, and not one marries a Mary — a negative sharper than “none found,” because nine Gurney men married Marys in the same holdings, so it is specifically the Johns who have none. Exactly two John × Mary marriages exist in England before 1660: Eythorne, Kent, 6 November 1632, eliminated on two burials, and Ackworth. Ackworth is what is left.[4]

The departure signature is the strongest in the file, and it is now measured two ways. The project has twice manufactured a false departure gap from an uncontrolled index query, at Berkhamsted and at Stewkley, so the standard here is deliberately high. This one is neither an index query nor uncontrolled.

  1. At image level. No Gurney-family form appears in the parish register across 1617–1628 or 1632–1644, read leaf by leaf on two parallel witnesses, with every visual near-match individually adjudicated and excluded — Burneley, Greenfield, Garner, Guntinger, Turner, Gaven, Bradly. Twenty-four of the twenty-seven years between 1617 and 1644 are covered. No John Gurnoe senior, no earlier marriage, no burial of either spouse, no other child, nothing before and nothing after.[2][5]
  2. At national surname level. No person of the surname Gurnoe appears anywhere in England between 1637 and 1762, on either FamilySearch or Ancestry. Both parties vanish from the record completely.[6]

The index-level control that first opened this line still holds and now sits underneath the image work: the burial index returns 170 Smith burials at Ackworth and zero Gurney burials, so a missing Gurney burial there is informative rather than an artefact of coverage.[7]

Gurnoe is isolated at Ackworth, and the local false-friend risk is cleared. A live worry was that Gurnoe might be a Yorkshire clerk’s rendering of a different local surname, which would dissolve the candidate. The county’s own register corpus says otherwise. Across West Yorkshire 1575–1796 the local forms are Gurnet, Gurnett, Gurner, Gurnhall, Gurnthias, Gurneshaw, Gurnitt, Gurnel, Gurnhill and Gurney, written correctly dozens of times — and the nearest Gurnell family, at Pontefract three miles away, begins only in 1686, fifty years after the Ackworth entries. Gurnoe occurs nowhere in the county except Ackworth. The clerk demonstrably knew the local names; Gurnoe is what he wrote for one he had not met, which is what an incomer’s unfamiliar surname looks like in a register. This is the opposite of Upton upon Severn, where a continuous local surname presence was the ground for downgrading: at Ackworth the surname arrives, produces two records, and stops.[8]

What the register cannot resolve

The marriage is real and it says almost nothing. The 6 June 1636 entry reads, in full, “John Gurnoe & Mary Burton married — June 6.” Both names read at 99%, and the entry carries no marital status for either party at 98% confidence: it is two names, an ampersand, and the word married. That bare form is the parish’s normal one, so the silence is not evidence that this was a first marriage; it simply leaves the question where it was.[2]

That matters because the objection capping this candidate has always been conditional. The colonial John’s daughter Mary married Daniel Shed at Braintree in 1647, so at a minimum marriage age of sixteen she was born by 1631 and her parents married by about 1630 — a primary constraint, not a compiled estimate. A June 1636 marriage cannot produce her. But that is decisive only if 1636 was John’s first marriage, and a widower remarrying in 1636 with children already born fits the colonial household exactly. The register does not say. So the image closes a route rather than moving the number. Widower and widow designations are recorded only sporadically in registers of this period, and this parish’s marriage entries do not carry them at all, so the hoped-for test was asking the document for something it was never likely to supply. The second-marriage escape now has no local support either, because he did not marry a first wife at Ackworth.[9]

The 1637 baptism is less resolved than it was, not more. Three witnesses disagree.

Witness Reading Date
Original register face John Thomas Gurnoe (98%) 28 January
Facing alternate-hand leaf John Gurnoe junior (~90%) 19 January
Provider index, all three providers John Thomas Gurnoe 19 January

Neither manuscript witness names a parent, and a re-check specifically for a mother’s forename or a maternal surname found none.[2]

The consequence is sharper than a discrepancy. The standing question was whether “John Thomas” is a compound forename or the collapsed indexing artefact “John son of Thomas Gurnoe” — the same defect already caught at Toddington. A reading of “junior” is neither. Junior distinguishes a child from a living father of the same name, and the 1636 groom himself supplies that father: a son baptised seven months after his father’s marriage in the same parish is naturally written junior to separate him from the man the clerk had just recorded. The junior reading needs no earlier generation, is undamaged by the empty 1617–1628 run, and does the work that matters — it attaches the baptism to the marriage, which is precisely the objection that the two records may not belong together. If instead the original face’s Thomas is right, the household dissolves and only the marriage survives. The two readings point in opposite directions and the dates differ by nine days, so an entry-alignment fault between the leaves is live too. This must not be collapsed either way without a third witness.[10]

The compiled tradition does not support the row

Older compiled Gurney histories are said to give the colonial John’s wife as “Mary Barton” or “Mary Burton,” and that was carried as a third support: a tradition pointing at Ackworth independently of the index. It cannot be located.

  • The compiler Rigler does not carry it in either edition held here. The c.1980 family-group worksheets never reach the colonial John’s wife, and the 1994 published edition opens its Richard-2 entry with “probably the son of John Gurney-1” — it does not firmly assert even the parentage, and nowhere gives John’s wife a maiden name.[11]
  • Open-web compiled genealogies give “Mary, maiden name unknown.” The Lysander Franklin Gurney sketch, the standard compiled tree databases and the general literature all render her that way; none supplies Barton or Burton.[12]

The test set for the tradition was chronological — a tradition predating modern indexing would be independent evidence and the row should rise, one first appearing in index-derived trees would be circular and it should not. That test turns out to have been the wrong instrument. The Ackworth register was never printed for these years: only the 1558–1600 portion was published, in Yorkshire Notes and Queries, volume 1. So there was no route by which any compiler, at any date before the modern indexes, could have reached a 1636 Ackworth marriage in print.[13]

That closes the question in the unhelpful direction. The only document supplying the pairing “John Gurney” and “Mary Burton” is the Ackworth entry itself, so a tradition with no traceable source, whose only possible origin is the record it is offered to corroborate, is not independent evidence — it is the record coming back around. The claim is preserved here with its sourcing gap stated rather than deleted, because the search was bounded: both Rigler extracts are scoped to G9–G12, so Rigler’s own John Gurney-1 entry has not been read in full. This is “not found where looked,” not “does not exist.” Evidence

The surname trail runs the wrong way, at low weight

Set against the departure signature, the distribution of the surname Gurnoe points away from the Great Migration corridor. The complete pre-1700 English Gurnoe corpus, on two independent instruments, is four records in three households:

Record Date Place
Alice Gurnoe, marriage 24 April 1575 Cheshire
Edward Gurnoe, baptism; father Andrew Gurnoe, mother Ann Lightollers 2 November 1626 St James, Clerkenwell, London
John Gurnoe × Mary Burton, marriage 6 June 1636 Ackworth St Cuthbert, Yorkshire
John Thomas Gurnoe, baptism January 1637 Ackworth, Yorkshire

There is no Gurnoe in Norfolk, Suffolk or Essex at any date. The mother’s surname in the only household that names parents — Lightollers — is a Lancashire name, clustered at Stretford, Withington, Brindle, Blackburn and Rochdale from the 1580s and reaching into the West Riding at Huddersfield, twenty-two miles from Ackworth. The axis is Cheshire–Lancashire–London, not East Anglia.[14]

This is carried as a low-weight consideration, not as a counterweight. The project’s standing doctrine is that surname forms are weights, not binaries, and on that doctrine the weight sits toward the Gurney reading: no Gurnoe family continues at Ackworth, where a resident family would recur; the clerk had the local forms and used them; and the emigrant is never Gurnoe in New England, always Gurney, Gurny or Garny — so if the Ackworth man is he, the form is the clerk’s and the north-western distribution says nothing about him. The Clerkenwell household carries one indexed child, Edward, and no John, so it supplies no bridge either way. Establishing whether Gurnoe is a real family or a clerk’s rendering is the single question that would move this row more than a point in either direction.[15]

The same tolerance runs the other way and applies to the exclusions above. Garner, Burneley, Gaven and Bradly were read confidently and those readings stand as readings — but a confident reading of a different surname is not a zero probability that the entry belongs to a Gurney, because a clerk who knows a local Garner family can write Garner for an incomer’s unfamiliar Gurney by simple familiarity bias. Elizabeth Garner, 9 November 1628 is the one excluded entry worth a residual weight: it is the only G-initial reading inside the window where a first marriage or an early child would fall.[5]

Mary Burton’s own absence is not established

The register masters were re-swept for Burton and its variants, scoring each reading on closeness to the name rather than in or out. Three readings were retained: Susanna Barton, baptised 11 August 1622, a listed variant on a bare entry naming no parent; a marriage of 29 July 1634 whose bride reads Elizabeth Boulton at 96%, retained only as a phonetic near-match; and the 1636 marriage itself. Roger Burneley of 16 November 1627, which the degraded original can invite as a false Burneton, was rejected on the parallel witness.[16]

No Mary Burton baptism, no Burton household, no post-1636 Burton continuation and no Burton burial appears on the readable leaves. But the sweep’s boundary matters more than its result: it does not reach before 1617, and a bride marrying in June 1636 would most naturally have been baptised between about 1610 and 1618. The larger part of her likely baptism window has never been read. Her absence is therefore not established, and the “Ackworth is the bride’s parish” reading rests on the groom’s demonstrated absence — separately evidenced across 1617–1628 and 1632–1644 — rather than on her presence.

Ancestry’s West Yorkshire collection cannot answer it either: its Ackworth baptism indexing does not begin until the 1660s, so the two Burton baptisms it returns, Francis 1667 and Elizabeth 1670, are all it has, and the absence of a Mary Burton baptism around 1610–1620 is a coverage artefact rather than evidence.[17]

Why it is still not higher

The date block stands, on the terms set out above, and the register cannot break it.

Yorkshire is outside the corridor, and that objection is now stronger rather than weaker: it sits well beyond the Norfolk–Suffolk–Essex–London belt that produced the Great Migration, no primary record bridges an Ackworth Gurnoe household to any Massachusetts arrival or to any of the colonial John’s identified associates, and the surname form has no corridor presence either.

The naming and birth order do not line up. A first child in January 1637, seven months after the marriage, is not a Sarah or a Mary; and a 1637 son named John collides with the colonial John Jr., placed about 1633 by compiled estimate.

The one point that came off came off the failed tradition leg, and nothing else. Of unit 35’s three original supports, one fails outright, one — the departure gap — strengthens materially, and the marriage negative is untouched. That is close to a wash, and manufacturing a larger move would misrepresent the evidence. The point goes to no record survives, since a candidate weakening with none rising is what that row describes.

Open work

The remaining value is not in more register leaves. The 1629–1631 gap — film 32355_249046 images 01573–01577, companion 32355_249045 images 00062–00065 — is worth closing for completeness but predicts nothing: a first marriage in those years would still postdate the 1631 bound on the colonial Mary’s birth. Two doors remain.

  1. Mary Burton’s family at Ackworth. She is the local party, and her kin — and any Burton with a New England connection — are untested. The route is free: the register images for 1617–1628 and 1632–1644 are already held, and the pre-1617 leaves are the ones that would carry her own baptism. This is the one remaining question at Ackworth that a held or cheap image can answer.
  2. The Gurnoe surname trail — Cheshire 1575, Clerkenwell 1626, and the Lightollers connection into Lancashire and the West Riding. It decides whether Ackworth belongs in the corridor at all. Evidence

  1. FindMyPast England Marriages 1538–1973, transcript R_855220028: John Gurnoe and Mary Burton, marriage 6 June 1636, Ackworth, Yorkshire (Source ID: findmypast-ackworth-gurnoe-burton-marriage-1636). FindMyPast England Births & Baptisms 1538–1975, transcript R_948023155: John Thomas Gurnoe, baptism 19 January 1637, Ackworth, Yorkshire; father and mother not indexed (Source ID: findmypast-ackworth-gurnoe-baptism-1637-john-thomas). FamilySearch’s record index carries the same two entries with the same gap, and both providers’ entries are derived from the International Genealogical Index, so this is one transcription seen twice rather than two witnesses. Ancestry returns the marriage twice — once from the same IGI-derived England, Select Marriages, 1538–1973, and once from the West Yorkshire Archive Service collection cited below, which is the independent transcription and the one that supplies the parish church, St Cuthbert. Source IDs: fs-england-marriages-1538-1973; fs-england-births-christenings; ancestry-england-select-marriages-9852. ↩︎

  2. Ackworth St Cuthbert parish register, read at image level in Ancestry collection 2256, West Yorkshire, England, Church of England Baptisms, Marriages and Burials, 1512–1812 (West Yorkshire Archive Service partnership), 2 August 2026. The register survives as two parallel films of the same run — 32355_249046, the original register face, and 32355_249045, a cleaner copy in a different hand. The marriage is at 32355_249046 image 01583; the 1637 baptism at image 01584, manuscript page 69, with the alternate-hand reading on the facing leaf. Diplomatic readings, confidence figures, the whole-page sweep and every adjudicated near-match are at sources/corpus_supplement/ackworth-st-cuthbert-register-1617-1644-gurnoe-burton-readings.md. Validation note: sources/validations/ancestry-west-yorkshire-1512-1812.md. Source ID: ancestry-west-yorkshire-1512-1812. ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  3. The bride’s-parish reasoning is the project’s own, applied to Buckinghamshire, where the county’s marriage negative is held to “tighten rather than eliminate, since grooms married in the bride’s parish” — 31-candidate-a-aylesbury.md, the county-wide marriage negative. Cross-reference, not a source. The image evidence that the groom left no other trace in the parish is the sweep at [2] and [5]. ↩︎

  4. The marriage arm is closed on five independent instruments: Ancestry’s England, Select Marriages 1538–1973 run England-wide with both parties indexed (Source ID: ancestry-england-select-marriages-9852); the Hertfordshire county set (Source ID: findmypast-hertfordshire-banns-marriages); the Buckinghamshire county set tested from both sides (Source ID: findmypast-bucks-marriage-index); FindMyPast’s national sets (Source ID: findmypast-uk-parish-baptisms); and FreeREG, whose record URLs carry both parties’ names in the slug, so its 178 marriage rows yield 97 unique Gurney-cluster events for 1615–1645 without opening a record (Source ID: freebmd-freereg; working dataset at sources/intake/archive/g13-july-2026-sweeps/freereg-g13-sweep-2026-07-28/). In those holdings Gurney men married nine Marys — Robert × Mary Lame (Norwich 1622), Edward × Mary Bedward (Stepney 1641), Walter × Mary Hammond (Westminster 1642), Henry × Mary Bodiman (Harefield 1647), Edward × Mary Hancke (Lowestoft 1656), James × Mary Randall (Rickmansworth 1656), Andrew × Mary Minns (Norwich 1657), George × Mary Pupplet (Bishopsgate 1658), Thomas × Mary Cockrell (Norwich 1660) — and no John did. The only John × Mary in the FreeREG run is John Garne × Mary White at Brigstock in 1659, thirty years too late and a Garne. Full working at 73-refactor-the-funnel-closed-and-the-unassembled-row.md. ↩︎

  5. The image sweep covers 1617–1622, 1623–1628 and 1632–1644 on both films, leaf by leaf, with full-page grids and page-side band ladders; the readings and their reproducibility material are staged at sources/intake/paleography-staging/packet-56-ackworth-1636-1640.report.md, packet-57-ackworth-1617-1622.report.md and packet-58-ackworth-1623-1628.report.md, with grids and crop ladders under each packet’s images/packet-NN/_local/. 1629–1631 is not held, and nothing before 1617 was examined; the 1642–1644 original-register faces are severely stained and rest on the parallel witness alone. Excluded near-matches, with image-level readings and confidences: Guntinger and Turner 1620, Garner 1621, Greenfield 1625, Burneley 1627 and 1633, Garner 1628, Gaven 1634, Bradly 1636, Garner 1642 and the three-entry Garner burial cluster of July 1643. Source ID: ancestry-west-yorkshire-1512-1812. ↩︎ ↩︎ ↩︎

  6. FamilySearch record search and Ancestry, 2 August 2026, surname Gurnoe run England-wide with no date bound: no person of that surname appears anywhere in England between the 1637 Ackworth baptism and 1762. Bounded by what both providers index rather than by what survives. Source IDs: fs-england-births-christenings; ancestry-england-select-births-christenings-9841; ancestry-west-yorkshire-1512-1812. ↩︎

  7. FindMyPast burial coverage for Ackworth, authenticated session, 29 July 2026: the common-surname control returns 170 Smith burials in the parish against zero Gurney burials. The departure attribute and its mandatory coverage control are set out at 38-the-shape-and-the-two-families.md, correction 5; the two false gaps it exists to prevent are the Berkhamsted and Stewkley negatives. A separate limit bounds the parish’s baptism index rather than its burials: FindMyPast’s Yorkshire Baptisms returns zero for a Gurney query at Ackworth while the Smith control returns 127 results with none earlier than 1689, and because a year bound and a place keyword cannot be combined in that query mode, that set’s pre-1689 Ackworth coverage is unproven and its Gurney zero is uninterpretable. The register images are what make the parish testable at all. Source IDs: findmypast-uk-parish-baptisms; findmypast-county-baptism-sets-2026-07-sweep. ↩︎

  8. Ancestry collection 2256, Gurn* swept across the whole West Yorkshire register corpus, 2 August 2026. The collection’s early West Yorkshire rows are Marye Gurnet at Halifax 1575, Joseph Gurney at Halifax 1588, Marie Gurnhall at Leeds 1597, Thomas Gurney at Dewsbury 1603 to a father Richard, Rosamund Gurnet at Elland 1607, Isabell Gurnthias at Leeds 1608, Anna Gurnett at Wakefield 1630 and Isabell Gurnett at Bradford 1647; across 1575–1796 the county’s forms are Gurnet, Gurnett, Gurner, Gurnhall, Gurnthias, Gurneshaw, Gurnitt, Gurnel, Gurnhill and Gurney. The 1636 Ackworth marriage is the collection’s only Ackworth Gurney-variant row of any date. The nearest Gurnell family, at Pontefract three miles away, begins in 1686. The Upton upon Severn comparison — a continuous local surname presence used as a ground for downgrading — is at 36-other-eliminations.md. Source ID: ancestry-west-yorkshire-1512-1812. ↩︎

  9. Mary Gurney’s marriage to Daniel Shed at Braintree in 1647 is in Samuel A. Bates, ed., Records of the Town of Braintree, 1640 to 1793 (1886), and in Clarence Almon Torrey, New England Marriages Prior to 1700, p. 331. The compiled estimate placing the colonial Sarah about 1628 is Waldo Chamberlain Sprague, Genealogies of the Families of Braintree, Mass., 1640–1850 (Boston: NEHGS, 2001), p. 695, and the History of Weymouth, Massachusetts (1923), vol. 3, p. 251. The 1636 Ackworth marriage record carries no marital status for either party, confirmed at image level. Source IDs: braintree-records-1640-1793-1886; torrey-new-england-marriages-prior-1700; sprague-braintree; history-of-weymouth; ancestry-west-yorkshire-1512-1812. ↩︎

  10. The Toddington artefact — an index rendering “X son of Y” as a compound forename — is worked at 71-refactor-burials-as-the-elimination-instrument.md. The Ackworth index transcript carries no father and no mother field, and neither manuscript witness names a parent, so nothing in the record distinguishes the three readings. Source IDs: ancestry-west-yorkshire-1512-1812; findmypast-ackworth-gurnoe-baptism-1637-john-thomas. ↩︎

  11. Robert G. Rigler, Gurney family group record worksheets, c. 1980 (Source ID: rigler-genealogy-notes-c1980), which never reach the colonial John’s wife; and Jean Gurney Rigler, The Gurney Family from Aaron to Zuinglius, rev. ed. (Honolulu, 1994), whose Richard-2 entry opens “probably the son of John Gurney-1” and nowhere gives John’s wife a maiden name (Source ID: rigler-gurney-family-aaron-zuinglius-1994). Both extracts held here carry corpusStatus: partial and are scoped to G9–G12, so Rigler’s own John Gurney-1 entry has not been read in full. Tested 2 August 2026. ↩︎

  12. The Lysander Franklin Gurney sketch and the standard compiled tree databases render the colonial John’s wife “Mary, maiden name unknown”; none supplies Barton or Burton. Checked 2 August 2026. Compiled-pedigree sites are collectors rather than authorities and are cited here only for what they do not contain. Source ID: accessgenealogy-lysander-franklin-gurney. ↩︎

  13. Yorkshire Notes and Queries, vol. 1, which prints the Ackworth St Cuthbert register for 1558–1600 only; no later portion of the register was ever printed. Validation note: sources/validations/yorkshire-notes-and-queries-v1.md. Source ID: yorkshire-notes-and-queries-v1. ↩︎

  14. FamilySearch record search and Ancestry, 2 August 2026, exact surname Gurnoe across England with no date bound: Alice Gurnoe, marriage 24 April 1575, Cheshire; Edward Gurnoe, baptised 2 November 1626, St James Clerkenwell, London, father Andrew Gurnoe, mother Ann Lightollers; and the two Ackworth entries. No Gurnoe appears in Norfolk, Suffolk or Essex at any date. The Lightollers distribution — Stretford, Withington, Brindle, Blackburn, Rochdale from the 1580s, and Huddersfield in the West Riding — is from the same instruments. Bounded by provider indexing rather than by survival. Source IDs: fs-england-births-christenings; fs-england-marriages-1538-1973; ancestry-england-select-births-christenings-9841. ↩︎

  15. The variant scale and the rule that a spelling is a weight on an individual record rather than a class of surnames to exclude are at 36-other-eliminations.md and 38-the-shape-and-the-two-families.md, correction 7 — cross-references, not sources. The colonial forms of the name are Gurney, Gurny and Garny throughout John’s New England career; see Colonial Braintree community, cited cross-unit. ↩︎

  16. The Burton sweep re-read the masters already held for the 1617–1628 and 1632–1644 windows, scoring Burton, Barton, Berton, Burdon, Bourton, Burten and Barten by closeness to the name rather than in or out; staged at sources/intake/paleography-staging/packet-59-ackworth-burton.report.md with readings and confidences carried into the corpus extract cited at [2]. It does not reach before 1617 and does not cover 1629–1631. Source ID: ancestry-west-yorkshire-1512-1812. ↩︎

  17. Ancestry collection 2256, coverage control at Ackworth, 2 August 2026: a Smith search at the parish returns marriages and burials only, no baptisms at all, the earliest events a burial of 1652 and a marriage of 1661. The collection’s Ackworth baptism indexing begins in the 1660s, which is also why its only Burton baptisms there are Francis 1667 and Elizabeth 1670. Source ID: ancestry-west-yorkshire-1512-1812. ↩︎