Record#

New to this archive? Start Here · Press and Archive Leads · Master Change Log

Page ID: 8 · Title: Great House Farm (Tŷ Mawr): The Williams–Buckler Dispute · Evidence status: Not yet labelled — see the Editing Protocol · Version: 1.1.0 · Editing Protocol

Current high-level working theory: Cedwyn Estate and repeated parcel capture#

The family’s hypothesis is that Great House Farm / Tŷ Mawr may have belonged to a wider historic Cedwyn/Cydfin estate connected with Llandough and Leckwith, and that successive land transactions, possession events and registrations may have fragmented or obscured older parcel interests. The proposed forty-parcel extent and any Wales-wide colonial/imperial mechanism remain unproved hypotheses. See the Cedwyn Estate Working Theory for the supporting leads, competing explanations and decisive document tests.

Great House Farm (Tŷ Mawr), beside St… .. read more

Great House Farm (Tŷ Mawr), beside St Dochdwy’s Church at Llandough-juxta-Penarth, was the Williams family home. Mary (Nancy) Williams’s witness statement c.1974 (Mary (Nancy) Williams testimony) traces its occupation to at least 1667. The central dispute is whether the farmhouse plot remained a distinct parcel and was later improperly folded into another title.

Original Parcel A and Parcel B — then three later Parcel-A strands#

The working reconstruction is that the 1877… .. read more

The working reconstruction is that the 1877 Bute–Thomas transaction marks the separation between Parcel A—the farmhouse, garden, buildings and associated land—and Parcel B, the eastern agricultural fields. The family’s working title theory treats the end of quarrying and last rent in 1928 as the point at which Williams became the “virtual freeholder” of Parcel A; the original title documents remain unresolved. The 1877 Bute–Thomas agreement, the 1938 Bute–WGR conveyance and the 1939 tenancy and plan are key records for testing that reconstruction [PENDING: EV-0019].

Three proposed Parcel-A successor routes: Forest, BP centre and Council/VG41#

BP’s documented title chain concerned Parcel… .. read more

BP’s documented title chain concerned Parcel B. The family now tests a three-strand Parcel-A model: a northern woodland route associated with a 1961 deed and later Forest of Cardiff title history; a south-eastern route associated with VG41’s 1970 village-green registration; and the occupied central house-and-garden route associated with BP’s 1982–83 registration and the 1987 amalgamation with Parcel B. In this reconstruction, the BP synthetic Parcel A that this Wiki identifies concerns the middle component, not the entirety of the historic Parcel A. The 1961 deed and 1970 register remain separate records whose mapped scope and title effect must be checked; neither by itself proves a freehold transfer or coordinated action. The family identifies a sequence that must be kept distinct: the 1949–1953 unwritten agricultural tenancy (which the Court of Appeal later treated as a protected periodic tenancy, while the family disputes any extension to Parcel A); a separate unsuccessful attempt to make Frederick a tenant in 1954; and later tenancy offers and proceedings in 1959, 1962, 1965 and 1974. This remains subject to source-by-source analysis and is not a finding of deliberate wrongdoing. The allegation of a synthetic title must be tested against the original instruments, plans, court records and registration files—not treated as a court finding. See the permissive-occupation claim, the Parcel A/B title claim and the Land Registry title history.

[SOURCE PENDING] Mary is described as… .. read more

[SOURCE PENDING] Mary is described as disabled, having undergone a leg amputation. The family’s position is also that her water supply was cut off and that family title deeds were removed or stolen, and that Parcel A was registered behind her back, without effective notice to her, during the period of her final illness and death. The disputed registration and notification issue is examined in the 1983 registration and reported late-notice record and Mary (Nancy) Williams’s statement.

ITV News footage — 1988#

The ITV Cymru Wales report records the 1988… .. read more

The ITV Cymru Wales report records the 1988 dispute at Great House Farm. Compare it with the South Wales Echo report, the event record and the demolition chronology.

Institutional decisions and records handling#

The current dispute involves the Land… .. read more

The current dispute involves the Land Registry registration history, the Cadw ATISN 27021 response, and surviving records held by archives and public bodies. Correspondence and decisions are linked in the FOI and Public-Authority Reply Register, Source Conflicts Register and Missing Records Register.

The family’s outstanding legal and… .. read more

The family’s outstanding legal and procedural concerns are recorded in the Master Scrutiny Issue Register and Legal Review, with institutional analyses in the Cadw forensic critique and GGAT forensic review.

The two parcels#

Parcel A — House Plot: the farmhouse… .. read more

Parcel A — House Plot: the farmhouse, garden, buildings and associated land, as defined in the Parcel A/B title claim. The exact legal extent depends on the deeds and plans linked there.

Parcel B — Eastern agricultural land: the… .. read more

Parcel B — Eastern agricultural land: the fields east of the House Plot, identified in the 1938–39 tenancy record and its linked plan; the legal extent must be tested against the original instrument.

The family’s case is that Parcel A never… .. read more

The family’s case is that Parcel A never passed into the Western Ground Rents → BP title chain. The relationship between the 1939 Parcel B tenancy, later possession proceedings and the 1982–87 registration/amalgamation remains disputed and is tested in the Parcel A/B title claim, the 1987 amalgamation review and the Land Registry titles. The Two Parcels →

Parcel map: Bute → A/B → three proposed Parcel-A successor routes#

How to read it: follow the lines vertically. Each elbow shows only a year and parcel label; select it to expand the evidence, source links and limits. The faded grey line is the historic Williams Parcel-A estate as claimed in the working reconstruction. The proposed later routes are Forest of Cardiff (upper/north), BP’s house-and-garden registration (middle), and the Council/VG41 green-record route (lower/south-east), alongside Parcel B. This is a hypothesis that the old Parcel A was later represented by three distinct successor or administrative strands; only BP’s middle registration is described here as the synthetic registration (this Wiki’s case, not a court finding). The exact polygons and title effects remain to be proved against the original plans.

BUTE · ROOT

Starting point: Bute Estate. The tube-map model starts with a single Bute freeholder line before the reported 1877 division. The precise instruments, conveyed rights and boundaries must be read from the original deeds and plans; the graphic records the present working model, not an established legal finding.

See Bute Estate and the 1877 Agreement.

1877 · A

Reported split in Mary (Nancy) Williams’s account. Mary (Nancy) Williams’s witness statement c.1974 (Mary (Nancy) Williams testimony) says that a Bute-to-Daniel Thomas & Son arrangement concerned the house parcel for quarrying, with a route by which the freehold was to pass to Williams when quarrying ended. The original 1877 Great House Farm instrument has not been located; the related EV-0006 entry remains pending. A separate 1877 99-year lease identified for a Clyde Street house plot in Roath is not Great House Farm title evidence and is excluded.

Sources: EV-0006 · Daniel Thomas and Alfred Thomas · Mary (Nancy) Williams’s statement.

1877 · B

Parcel B track. On the working model, the eastern agricultural fields remain on the Bute side of the division until the 1938 transfer to Western Ground Rents. This is the separate-fields route, not evidence that the same instrument conveyed the whole farmhouse plot.

Compare the 1938 Bute–WGR conveyance and plan (EV-0019) and The Two Parcels.

DATE UNKNOWN · BARN

Undated local-history marker — not a title event. A secondary summary says Jim Driscoll (“Peerless Jim”) had his first fight in a barn at Great House Farm, Llandough, but the year and the underlying National Monuments Record citation have not yet been located in the Wiki. Until the cited record is identified, treat the location as a reported historical association, not as a dated fact or evidence of ownership.

Lead: secondary biography summary. This is a research lead, not a substitute for the cited record.

1897 · LOCAL

Regional context only. A press cutting catalogued as May 1897 reports Marconi wireless experiments at Lavernock Point, with signals tested across the water toward Flat Holm. This was not an event at Great House Farm and does not bear on Parcel-A title. It is included as a local-history milestone only.

Wiki source: EV-0011 — “Wireless Telegraphy. Marconi Invention Tested at Lavernock”.

1916 · A+B

A+B in one reported agricultural tenancy. In February 1916 the Marquess of Bute granted John Williams a yearly agricultural tenancy described in the judicial history as concerning the “whole farm”. On Mary (Nancy) Williams’s witness statement c.1974 (Mary (Nancy) Williams testimony), the house parcel had already been separated by the reported 1877 arrangement; that makes the extent of the 1916 tenancy a central scope conflict, not something this line resolves. The 1939 plan later depicts Parcel B only.

Sources: 1916 tenancy · EV-0014 · 1938–39 revised tenancy.

1928 · A

Family’s “virtual freeholder” position. Mary (Nancy) Williams, formerly Buckler, recalled that quarrying ended and the last rent to Alfred Thomas was paid in 1928. The family treats this as the point at which its Parcel-A interest became the Williams family’s own freehold/beneficial estate. The receipt and original 1877 instrument are not presently held, so the exact legal effect remains a title proposition to prove rather than a completed registration finding.

Sources: EV-0016 — 1928 rent receipt (not held) · 1877/1928 chronology · Mary (Nancy) Williams’s statement.

1938 · B

Parcel B → Western Ground Rents (WGR). The present reconstruction places the eastern fields on the WGR route after the 1938 Bute–WGR conveyance of the reversion. The deed and plan must be compared with the reported earlier Parcel-A split; do not infer that the WGR chain necessarily included all three later Parcel-A components.

Source: EV-0019 — Bute–WGR conveyance and plan · Western Ground Rents.

1939 · B

WGR, not Bute, grants the revised tenancy. The May 1939 WGR–John Williams agreement and plan have been examined; the displayed plan depicts the eastern fields (Parcel B). This is a distinct tenancy from the 1916 Bute grant. The plan does not prove Parcel A’s root title by itself.

Source: EV-0023 — DBDT123/7, 1939 agreement and plan.

1949–53 · A?

Do not conflate the two occupancy stories. The Court of Appeal accepted that Frederick had a protected periodic agricultural tenancy, despite there being no written agreement; that tenancy is associated with the agricultural holding/Parcel B. Separately, an attempt was reportedly made [SOURCE PENDING] to treat Mary’s Parcel-A occupation as permissive in this period. The exact offer and reply have not been located, and the available record does not establish that any offer covered the north woodland or south-east green.

Sources: Frederick Buckler — tenancy · Permissive-occupation claim.

1949–53 · B

Agricultural tenancy strand. The real periodic agricultural tenancy should not be described as fictional or failed. The unresolved issue is its mapped extent and whether the description “whole farm” was later used more broadly than the original tenancy evidence supports.

Sources: Frederick Buckler · 1939 plan (EV-0023).

1954 · B?

Separate reported further tenant-making attempt. The Wiki records a reported 1954 attempt to treat Frederick as a tenant, but the original source has not been attached. The proposed land, exact terms and reason it failed remain unconfirmed. Do not apply it automatically to all of Parcel A or to the forest/green strands.

Source: EV-0031 — pending source.

1955 · A?

Reported physical separation, exact legal scope pending. The working reconstruction (SOURCE PENDING until EV-0032 and EV-0034 are obtained) is that the 4 July 1955 execution took Parcel B together with the north forest and south-east ground of Parcel A, while Mary remained in physical occupation of the central farmhouse/buildings/garden group. The sealed order, writ/return and any plan are needed to establish exactly what land and parties were covered. This event is the claimed physical split; the 1961 and 1970 records are later and separate strands.

Sources: EV-0032 · EV-0034 · 1955 possession order.

1955 · B

Parcel B possession route. The 1955 process is reported as enforcement against the farm holding, but the original order and execution return have not been obtained. Keep the order’s stated area separate from the land physically entered or later registered.

Sources: EV-0032 · EV-0034.

1959 · A-C

Central occupied house and garden. The Court of Appeal’s history describes WGR’s tenancy offer to Mary in 1959 and her refusal. The original offer and reply remain unlocated. Present record: a dispute about the occupied centre; no map ties the offer to the north woodland or lower green.

Source: EV-0039 — tenancy offers (1959/1965).

1961 · A-N

Northern woodland / Forest of Cardiff route — possible mapped overlap. The catalogue identifies a 29 March 1961 deed involving WGR, South Western Brick & Tile and “Cardiff City”; the deed and plan have not yet been inspected. The wiki catalogue says the land is associated with the later woodland title, but until the original plan is plotted it is a high-value lead, not proof that the deed transferred this Parcel-A polygon or that Cardiff/Forest of Cardiff then held the whole upper third.

Source: EV-0040 — 1961 deed (pending original/plan) · Three-part Parcel A assessment.

1962 · A-C

Central house-and-garden proceeding. The Wiki records a Cardiff County Court possession order dated 11 December 1962, but the original order and plan remain pending. Its extent must be plotted; the separate forest and VG41 routes should not be assumed to have been within it.

Source: EV-0041 — 1962 order (not yet held).

1965 · A-C

Further offer concerning the centre. The Court of Appeal records a further weekly tenancy offer concerning the farmhouse and garden, which was not signed or returned and for which no rent was paid. The original offer/reply remain unlocated. This is not evidence by itself about A-North or A-South-East.

Source: EV-0039 — tenancy offers.

1969 · B

WGR → BP Pension Trust on the B route. The Wiki records a 31 December 1969 conveyance to BP Pension Trust. The instrument and plan should establish which WGR land entered this chain. Under the working model this is Parcel B; it should not be treated as proof that the north woodland or VG41 ground transferred with it.

Source: EV-0043 — 1969 conveyance (pending copy) · BP Properties and BP Pension Trust.

1970 · A-SE

South-east ground / VG41 register route. The Wiki records VG41 as registered on 16 June 1970. The original application, map and ownership file remain outstanding. This is the later administrative/green-register strand associated with the lower Parcel-A segment. Village-green registration does not, on its own, prove that Llandough Council acquired the freehold or that the exact polygon matches the asserted Parcel-A boundary.

Source: EV-0044 — VG41 registration (file/map pending) · Woodland + VG41 loss dossier.

1974 · A-C

BP licence letters: central house-and-garden only. Letters dated 31 October 1974 were later treated by the Court of Appeal as placing Mary’s continued possession under a licence. The legal effect of those letters was material to the appeal; their described subject is the farmhouse and garden. They must not simply be projected onto the northern woodland or lower VG41 ground.

Sources: EV-0050 — licence letters · EV-0080 — 1987 Court of Appeal judgment.

1982–83 · A-C

BP’s synthetic Parcel A (our case, not a court finding): middle / occupied centre. The WA231076 and WA240304 registration applications belong on this centre line in the working model. The allegation under investigation is that the registered plot mimicked the historic Parcel A centre and was later combined with Parcel B. The original registration applications, supporting deeds, plans and notifications must decide that proposition. This must not be shown as if it automatically covered the forest or VG41 area.

Sources: Land Registry titles · EV-0066 — WA240304 record · EV-0078 — 1987 amalgamation plan/application.

1987 · A-C+B

1987 endpoint: the middle A line joins Parcel B. This is the family’s three-strand reconstruction of the recorded 2 February 1987 amalgamation: the house-and-garden/middle Parcel-A route associated with BP is combined with Parcel B. The north/woodland route and south-east/VG41 route remain separate lines in this model because their earlier records and mapped extents are distinct. That is a research proposition to test against the registration plan and original source titles, not a finding that the court adjudicated all three parcels or that fraud occurred.

Sources: EV-0078 — amalgamation plan/application · EV-0080 — Court of Appeal judgment · Three-part Parcel-A assessment.

1987 · B

Parcel B endpoint. On the working model, Parcel B and the registered middle/house-and-garden Parcel-A strand join in the 1987 registration/amalgamation record. The northern woodland and south-east green are not assumed to follow the same path without polygon-level proof.

Source: EV-0078.

A-N · FOREST

Upper / northern strand. The working model associates A-North with the later Forest of Cardiff woodland title route. The 1961 deed-plan match is still unconfirmed and a later 2005 transfer/excision is recorded in the Wiki, so present ownership must be checked against current title plans before being asserted.

Sources: EV-0040 · GHF-E150 — 2005 transfer/excision.

A-C · BP + B

Middle strand. The BP synthetic Parcel A that this Wiki identifies is confined in this model to the central houses/buildings/garden, with the 1987 amalgamation joining that registered route to Parcel B. The precise boundary and root-title basis remain under test.

Source: Three-part Parcel A assessment.

A-SE · LCC/VG41

Lower / south-east strand. The line is the VG41/council green-registration path, not proof of council freehold. The register map and ownership file must be compared with the asserted Parcel-A polygon.

Source: EV-0044.

B · BP

Parcel B. The eastern fields follow the Bute → WGR → BP Pension Trust route on the working model. The 1939 plan depicts Parcel B only. The 1987 registration event is shown as joining Parcel B with the middle house-and-garden title route, not with every area historically claimed as Parcel A.

Sources: EV-0019 · EV-0023.

Reading rule: a coloured line identifies a proposed parcel pathway; it does not itself prove freehold ownership, a conveyance, a fraudulent act or coordination between institutions. The decisive comparison is deed/order/registration plan against the same three Parcel-A polygons. For the complete working assessment see Three Parts of Parcel A.

Case records and current action#

Questions still at issue#

Page version & change log#

Current version: 1.1.0 · details in the Master Change Log

Version

Date (UTC)

Editor (human or agent name)

Change

Source / evidence

1.0.0

2026-10-18T15:14Z

Sion’s Tab agent

Footer standardised; no content change

Editing Protocol, 18 Oct 2026

Agents and editors: follow the Editing Protocol.