© Copyright John Palmer.
This is work in progress, last updated 2009-06-04.
Please use this link
to send me comments and suggestions.
In Purbeck-
- name Square and Compass roofing
- number 1
- site St Aldhelm's Head
(if site is as given by RCHM 1970)
- grid SY 9633 7597 (if site is as given by RCHM)
- source JP's personal observation 2003-03-17
- source pers. comm.
Charlie Newman, Square and Compass
- publ RCHM 1970 Dorset SE pt.3 p.474
(if this is indeed the site at which Mr Newman found them)
- desc 6 Purbeck limestone tiles on display, arranged as they would
be on a roof (except that they are on a vertical wall).
In shape they are like the illustration,
somewhat pentagonal or necktie-shaped, about 450mm long and 280mm wide
(estimated by eye); the two at the top (the only ones fully visible)
used to have one nailhole each, but the top of each has broken off through
the nailhole and new holes have been drilled for the present exhibition.
The substance is a shelly limestone; one at least of the slabs
has a number of visible bivalve shells about 10mm across.
- loc On display at the Square and Compass public house, Worth Matravers
- subst PL. Mr Newman says Downs Vein, except for the one with bivalves
- date Labelled as Roman and evidently of Roman style
- comment Mr Newman says he has some smaller tiles from
Compact Farm
- comment Photographs by JP
2003-01:4.17
2003-01:4.16
-
- name Worth Purbeck Marble roofing
- number 2
- site Compact Farm, Worth Matravers, c.2000
- grid SY 975 780
- source Graham, Hinton and Peacock 2002 p.42
- desc `At the Worth site, [PM] was used for roofing slates, three pieces being
found'
- loc `All worked stone from the site was kept and recorded' p.41
- subst PM
- comment Unusual material for roofing
-
- name Worth Purbeck building-stone roofing
- number 3
- site Compact Farm, Worth Matravers, c.2000
- grid SY 975 780
- source Graham, Hinton and Peacock 2002 p.42
- desc `There were pieces of [PL] roofing slate, some with peg-holes but
none complete'
- loc `All worked stone from the site was kept and recorded' p.41
- subst PL
- comment PL is the local rock at this site
-
- name Encombe roofing
- number 4
- site Encombe, Corfe Castle
- grid SY 9427 7898
- publ Farrar RAH 1954 p.80
- publ Farrar RAH 1966 p.120
- desc RB building and shale armlet mfg. near the Obelisk, Encombe
- desc 1954: Remains of 3 sides of a rectang bldg aligned N-S and measuring
51ft by 20-25ft. Foundation about 2ft wide and of selected and roughly
trimmed blocks of PL with rubble infilling the centre, app. without mortar.
Floored with limestone slabs. Against the inner side of the N wall
a rectangular pit ... had been sunk from floor level 12in deep into
the natural brash capping the limestone rock, floored carefully with
flags and lined on all sides except the wall side with reused
hexagonal roofing slates of Purbeck stone.
- subst PL
- comment Reused
-
- name Gallows Gore roof slabs
- number 5
- site Gallows Gore West (Lander's Quarry)
- grid SY 9782 7901
- source RCHM 1970 Dorset SE pt.3
- publ Calkin and Piggott 1938
- publ Calkin 1947b
- publ Calkin 1948,
p.40-42, 46-50, 53-54
- publ Calkin 1953 p.52, 69
- desc RCHM: Occupation continuous from late IA to late Roman ...
a few short lengths of drystone wall thought to be C2 or later, near which
were floor-slabs and stone roof-tiles; also 2 rectangular storage cists,
one of reused roof-tiles
- desc Calkin 1947b: a rectangular chamber about 4ft6in by 3ft consisting
of Purbeck flagstones set on edge ... inside, 3 jugs ... sides of chamber
appear to be reused roof tiles. "I have seen 2 similar chambers in the
vicinity"; one 3ft by 2ft6in with a large stone slab as its floor; nearby
various C1--C3 relics; the second in "another part of the settlement" some
300yd further E, 4ft by 3ft, of slabs set on edge; 6ft away and parallel
to its shorter side a length of wall 11ft by 2ft. [Pottery suggests
date after Hadrian.] Calkin supposes the stone cists were for storage.
[Jugs made prob. before 150AD]
- loc Objects in Poole Mus, Christchurch Mus, DCM and British Museum
- subst PL
- date maybe C2 or later
- comment see entry
under Quarries or RB Occupation.
Many of the publications listed here may not refer to
the roofing slabs.
- comment Some reused
-
- name Langton Museum (Bucknowle) rooftile
- number 6
- site Bucknowle villa
- grid SY 9545 8154 (Bucknowle villa)
- publ Putnam 1984
- desc Putnam: Diamond pattern stone tiles on Roman sites in Dorset
- desc Roman roof tile or `slat', 49cm by 29.5cm by 1.5cm.
The first time I saw this it appeared to me diamond-shaped, with one obtuse
angle cut off to form an irregular pentagon. One nail hole near one
acute angle. (I thought it might be from the gable-edge;
for usual shape see Dorchester
Prison roofing slabs and illustration.)
However visiting on 2003-06-23 I found a hexagonal tile, with the
left side leaning in somewhat towards the top. There is one nail
hole, slightly to one side of the true top centre of gravity.
The slat was given to the Museum by
Frank and Mary Baxter
who had worked on the excavation of the Bucknowle villa;
diggers were allowed to take some surplus finds
(Reg Saville, pers. comm. 2004-07-12).
Nails were found with the slats (Reg Saville).
- loc Langton Matravers Museum
- subst Langton Museum: PL, probably
Downs Vein
- date given as 1st cent., but site was occupied C5 BC to C4 AD
- comment Langton Museum ticket says that PL slats were formerly bedded in
moss with no mortar. This presumably does not refer to Roman practice,
but something of the kind may have been done in Roman times. This may
be relevant when considering the pitch of roofs
(see Colliton Park and Dewlish
and discussion there)
- cont Reg Saville
-
- name Ower roofing
- number 7
- site Ower
- grid SY 99 85
- publ Keen 1978 p.112-3
- publ Sunter and Woodward 1987 p.105
- publ Cox and Hearne 1991 p.77, 175
- desc Woodward, table 7 on p.100: 17 complete limestone roofing tiles,
of which 15 in 3rd-4th cent. occupation and structures, 2 in post-Roman
agricultural disturbance
- desc Woodward, p.105: Limestone, shale and ceramic roofing tiles occurred
in small quantities ... generally in association with the 3rd-4th cent. AD
building complex ... paucity of material, and lack of fixing holes ...
suggests none ... had actually been used for roofing. ... the limestone
tiles possessed a larger variation in dimnensions than has been used on
villa sites in Dorset, where consistency in size enables [fitting]
(Cook 1980) ... not likely to
have been manufactured here for export "because [far] from the source"
... [some] limestone tiles ...
used for post-packing and cist construction
- desc Cox and Hearne p,77: "nature of collapse entirely consistent with
that recorded in 1978 (Woodward) including the presence of unperforated
limestone tiles".
- desc Cox and Hearne p.175 (J.M.Mills): 5 Purbeck limestone roof tile
fragments, from Phase 2 and 3 contexts (1st-4th cent.) Like those
recorded by Woodward, no nail holes and may not have been used as roofing
material. Possibly in course of manufacture.
- loc DCM ?
- subst PL
- date AD 200-400
- comment see the
padstones from this site
-
- name Cleavel Point roofing
- number 8
- cat RELATED
- site Cleavel Point, Ower
- grid SZ 002 860
- source Keen L 1978 p.112-3
- desc Excav. by P Woodward. RB site of at least 27000sq.m.
Site spreads 100m SE of Ower Quay and to E of Ower Quay Road, over the
whole peninsula as far as the low land W of Cleavel Pt. Extends to N
and S shores where it has been eroded since RB times. Also spreads
around Newton Bay to the S.
Production centre for BB1 pottery; also shale and salt.
Purbeck limestone and shale times found.
- subst PL
- date AD 200-400
- comment Related, because the same site as Ower
Dorchester-
- name Charles Street roofing (seven slabs)
- number 10
- site Charles Street, Dorchester (Acland St carpark) 1989-90
- grid SY 6935 9045
- source K. Knowles, pers.comm.
- publ Adam J 1990
- publ
Davies SM, Farwell D 1989
- desc Knowles:
9) Complete Purbeck roof tile of Lower Building Stone.
Large heavy oval stone slab with two uneven rough shelly surfaces. It
contains a hole which took an iron peg that secured the tile to the roof
( this is located towards the centre towards its widest end). Dimensions
370 x 270mm and 30mm thick, 4550g.
Medieval soil accumulation 76/00/01. [context 2003]
10) Complete Purbeck roof tile of Lower Building Stone.
Large heavy oval stone slab with two uneven shelly surfaces and a hole
for an iron peg towards its widest end. Dimensions 370 x 280mm and 25mm
thick, 3600g.
Med soil accumulation. 76/00/01. [context 2003]
11) Complete Purbeck roof tile of Lower Building Stone.
Large heavy oval stone slab with an uneven shelly surface and hole in
its widest end. Dimensions 370 x 250mm and 20mm thick, c3100g.
Medieval soil accumulation 76/00/01. [context 2062]
12) Almost complete Purbeck roof tile of Lower Building Stone. Large
heavy rough oval stone slab, (similar to 11 above) with a hole at its
widest end. Dimensions 370 x 250mm and 20mm thick, c2300g.
Medieval soil accumulation 76/00/01. [context 2062]
13) Partial Purbeck roof tile of Lower Building Stone, a large heavy
rough stone slab with an iron peg still to be found in its hole.
Dimensions 200 x 190mm and 28mm thick, c1450g.
Medieval soil accumulation 76/00/01. [context 2003]
14) Partial Purbeck roof tile of Lower Building Stone, a large heavy
rough slab with a hole in its widest end. Dimensions 210 x 105mm and
20mm thick, 921g.
Medieval soil accumulation 76/00/01. [context 2003]
Unnumbered (not in catalogue of Worked stone)
1 Lower Building Stone roof tile c122x130x20mm 576g
[in late Roman layers]
- loc DCM, 1996.31 (W310):
from pers. obs. in DCM reserve coll. bay 158 shelf B, 2003-04-22,
1 bag of Roman pattern rooftile fragments, 2 pieces, context 176;
also 2 bags of similar fragments, 5 pieces, context 2003;
shelly limestone with irregular surface.
See archive boxes 591--611 and 615, finds bays 145--149
- subst K Knowles: Purbeck Lower Building Stone
- date In mediaeval layers but probably residual Roman
- comment These items are not specifically mentioned in the articles
cited above, but appear in a petrological report (1991) by K Knowles on
items from this dig
- comment see also
fragments of
(probably architectural) PM,
and other fragments
(early Roman),
(late Roman) and
(doubtful)
from the same excavation
- cont
Kathryn Knowles, U. Southampton
-
- name Greyhound Yard roofing
- number 11
- site Dorchester, Greyhound Yard 1981-4
- grid SY 693 906
- publ Bellamy PS, in
Woodward PJ 1993 p.168-172
- desc Stone roof tiles. Mostly very fragmentary. Only 10
complete enough to describe. All lozenge-shaped. Length 320-515mm
breadth 225-320mm, either 1 or 2 holes, single hole either central or offset.
- loc JP: DCM, reserve coll, site W67, bay 158 shelf A, 6 Roman pattern
rooftiles in limestone, rough-faced. acc.no.1985.31.40
- subst Paul Ensom: PL, all of Middle Purbeck limestone except
one which may be lower Lias.
- comment This item also in
Architectural section
-
- name Colliton Park roof
- number 12
- site Dorchester, Colliton Park, Roman Town House, 1936 and later
- grid SY 690 909
- source
Williams 1971 p. 178
- publ
Drew and Collingwood Selby 1937 p.6
- desc Purbeck limestone roofing slates
- desc Roofed with stone slabs excluding the 2 stoke-holes which
seem to have been covered with tiles. (The two stokeholes were not
contemporary with each other; one seems to have been built when the other
was built-over to provide a major extension to the building.)
- desc There are places where roof-slabs have been built into walls in
a herringbone pattern: see the inside of the lararium recess,
and the underpinning of the wall opposite that recess. (Pointed out
by the guide during a tour of the site.) The walls would have been
rendered, so the pattern is not merely decoration.
- desc The cover building
erected over the site in the 1990s has a Purbeck slabbed roof intended to
imitate the original; but whereas
Roman slabs
have irregular and slightly
curved sides, the modern ones have straight parallel sides, and 2 nails
per slab instead of one; see the lower part of the
illustration. The
pitch of this reconstructed roof is about 45 deg, but there would have
been no excavation evidence for this;
Dewlish villa has
a much less steep pitch. There is however
a preserved Roman gable at
Meonstoke, Hants with a slope of about 45 deg.
In both Roman and modern imitation roofing
there is a kind of `double-overlap'. The modern tiles are much more
regularly shaped than the Roman, being approx 370 by 280 by 25mm,
rough-dressed on top surfaces and oblique edges at the bottom, otherwise
sawn (including a very flat back).
- loc DCM, arch. coll. Bay 158, shelf C has
`1 Roman roof slab from Colliton Park Townhouse'
- loc DCM, arch. coll. Bay 157 shelf E, site W136 (TWA 1986),
`2 no. rooftiles, Garden pit 1, 26' level, Colliton Park'
- subst PL
- date 200-400 AD
- comment ``The only tiled roofing at Dorchester was over
the stokehole at Colliton Park I'' (Williams)
- comment Building I is the presently displayed Town-house.
-
- name Dorchester Prison roofing tiles
- number 13
- site Dorchester Prison, 1858
- grid SY 693 909
- source Notes by Oliver VL
- publ Bingham CW 1859
(reporting Lawrance JV c.1858)
- desc Bingham: House ... with
pentagonal stone roofing tiles with
upper edge convex
- desc Oliver: hexagonal stone roofing tiles. (The sample
in the display at DCM is more like pentagonal)
- desc Size (measured by JP) averages around 360 x 240 x 25mm;
the obtuse side angle is approx. 1/3 of the length up from the bottom point;
the tiles are rough-dressed on all sides,
and quite irregular in shape as well as size
- loc Bingham, Lawrance ap. Hutchins: DCM
- loc DCM, see comment below
- loc DCM, reserve coll. bay 158 shelf B: one box misc. rooftiles and
stonework, includes 1 `portion of the roofing of a Roman house exhumed
in the grounds of Dorchester castle August 1858'
- subst unrecorded
- comment See
Dorchester Prison pavement
- comment See also Colliton Park house
and description of cover-building there
- comment One such tile at least is included in
display under stairs in Victorian hall of DCM;
inscription on the tile (in C19 paint)
says that it is part of the roofing of a Roman house
found Aug 10, 1858 at Dorchester Prison
and presented to the Museum by order of the Visiting Justices;
(less clearly legible) a nail was in the (single) hole when found.
-
-
- name Poundbury roofing
- number 15
- site Poundbury, Dorchester, 1966-1980
- grid SY 68 91 approx
- source pers. obs. in DCM, reserve coll
- publ Farwell and Mollison 1993
- desc Farwell and Mollison p.61-2: Stone linings and cists ... occurred
[in the cemetery] where limestone, flint and chalk rubble was used with or in
place of a plain nailed wooden coffing. The limestone and rubble was
frequently reused building material ... [mostly] complete and fragmentary
roof tiles. Limestone tiles occurred below, above and alongside wooden
coffins, in some cases completely covering the coffin (e.g. [grave] 1102)
and in others using only one or two tiles over head or feet (e.g. 141 and 626).
Same degree of variation was evident in graves with no evidence for wooden
coffins (e.g. 110 had a full tile cist and 1172 a single tile over the skull).
p.21 plate 13: Grave 1114, stone tile cist covering child burial.
Eastern peripheral cemetery. Plate shows 10 roof slabs, no piercings
evident in the picture. Size of slabs est. 350 by 280mm (scale non-metric
and not quite clear). Also Fig.21 p.29. 6 year old congenitally deaf
child, prone position. The ref. to Pl.65 p.29 doesn't seem to relate
to this grave. See also p.187, "Congenital anomalies".
p.23--4: Burial in the eastern peripheral cemetery continued through the
3rd and into the 4th cent.
- desc 6 Roman pattern roof slabs marked `Poundbury'
- loc DCM, reserve coll, bay 162 one shelf above bottom,
still there 2008-08-19
- subst JP: creamy limestone, probably PL
- date 3rd-4th cent. AD
-
- name South Street roofing
- cat DOUBTFUL
- site South Street, presumed Dorchester
- grid SY 692 906
- source pers. obs. in DCM reserve collection 2003-04-22
- desc from shelf list: ``1 [pieces] South St''
- loc DCM, reserve coll, bay 158 shelf B,
1 box misc. rooftiles and stonework
- subst unknown, may even be ceramic
-
- name Dorchester Hospital roofing
- cat DOUBTFUL
- site Dorset County Hospital site, Dorchester
- grid SY 690 904 probably
- source pers. obs. in DCM reserve coll. 2003-04-22
- desc from shelf label: 2 boxes unidentified rooftiles;
rooftile from DCH 70/1 context 156;
part rooftile? from DCH DHF 70 2 context 9
- loc DCM, reserve coll, bay 157 shelf F
- subst unknown, may even be ceramic
Dorset (except Dorchester, Purbeck and NE)-
- name Puncknowle rooftile
- number 9
- site 'Walls', Puncknowle (Puncknoll), c.1965-1969
- grid SY 5399 8730
- source pers. obs. in DCM reserve collection
- publ Bailey CJ 1986
- desc One large PL rooftile,
which however is not mentioned in the above excavation report
- loc DCM, reserve coll, bay 157 shelf B, acc.no.1988.13.3
- subst looks like PL
- comment see also mortaria from
the same site
-
- name Dewlish villa roofing
- number 16
- site Dewlish villa, c.1969-84
- grid SY 77 97
- source WG Putnam pers.comm.
- publ Putnam 1969,
1970,
1971,
Putnam and Rainey 1972,
Putnam 1973,
Putnam and Rainey 1974,
Putnam 1975,
Keen 1976 p.54-55,
Keen 1977 p.120,
Keen 1978 p.113-114
- desc Putnam 1971: Although imbrices and tegulae were
found SE of the possible verandah in 1969, the main roof of the villa was
of Purbeck or Portland stone tiles, with dressed Ham stone blocks at the
ridge. Substantial numbers of these tiles were found as they had fallen
in several parts of the villa, particularly in room 10 and outside the
apse of room 4 [ a trapezoidal apse, which would have suited Roman-type
clay tiles: JP. ] [ Rm 11 is the triclinium. Rm 4 may be the winter
reception room. ]
[ Purbeck stone, not Portland, from appearance in WGP's slides. JP ]
Ham stone finial now displayed in DCM.
Plaster has been found which clearly shows the angle of the rafters of
the main building to the horizontal, namely about 20 degrees; contrast
Colliton Park
where a PL stone roof has been reconstructed at a much steeper slope,
and Meonstoke, Hants
for clear evidence of a much steeper pitch.
PL slabs include specially
shaped ones for edge of gable and lower edge of roof. Nail holes in
some regular slabs are to one side of the centre-line (cf.
Langton Museum roof-tile.
- loc stored near site, Bill Putnam pers.comm., not yet seen
- subst PL
- date Putnam and Rainey 1972:
Final phase of building c.350 (coin evidence)
-
- name Halstock villa roof tiles
- cat DOUBTFUL
- site Halstock, Dorset, 1967--1985
- grid ST 53 07
- source pers. obs. JP in DCM reserve collection
- publ Lucas RN 1993 p. 109, 111
- desc Lucas: 85 pieces of stone rooftile, of which 59 are complete tiles.
Many more found but not kept. Mostly Lower Lias (Street, Som.),
also Old Red Sandstone (Mendips) and one flaggy limestone; geol. ident.
by Paul Ensom. Sizes various, 200--500mm long, 150--350 wide, 11--35 thick.
63 pointed at bottom, 18 with bottom edge cut square, 4 uncertain.
- desc JP: Many tiles of even-bedded, thin-bedded limestone, micritic,
macrofossils scarce, some calcite veins; shape not unlike
the diagram. Several are
cut square at the bottom, i.e. the lower point is missing, as
if they had been designed for use at the eaves.
- loc DCM, reserve coll, bay 157 shelf D, 158 shelf D, some at 158 shelf C.
Accession number may be 1988.20.12, some doubt
- subst Lias with some ORS and one doubtful (Ensom)
-
- name Maiden Castle temple roof
- cat DOUBTFUL
- site Maiden Castle, RB temple, c.1936
- grid SY 671 884
- source pers. obs. JP, in collection
- publ ? Wheeler REM 1943 ?
- desc collection shelf label: `roof tiles from RB temple'
- loc DCM, arch. collection bay 158 shelf B, `1 box containing roof tile from
Maiden Castle temple (29.9.36), also 6 large rooftiles, loose'
- subst not stated on label, not yet seen
-
- site Unknown, presumed Dorset
- source pers. obs. in DCM, reserve coll. 2003-04-22, 2008-08-19
- desc 1 partial Roman pattern roof slab with irregular surface
- loc DCM, reserve coll, bay 162 shelf ?B (one down from top)
- subst gastropod limestone, probably PL
-
- name Thornford roofing
- cat REJECT
- site Thornford, NW Dorset
- grid ST 60 13
- source Buckman J 1877
- desc Roman roof slates: typical kipper-tie shape
- subst Buckman: local Lias
- date Roman
NE Dorset/Wiltshire-
- name Badbury Temple fragments
- number 17
- site Badbury Rings
- grid ST 96 03
- publ
Papworth M 1996 p.134
- publ Papworth M 2001
- desc 1996: Resistivity survey of temple made 1995.
Finds from the site of the square building [include]
fragments of PL probably once used as roofing tiles.
- desc 2001 p.28: a single posthole ... filled with fragments of PL
roof slates. Posthole square, 0.4m wide, 0.15m deep.
Courtyard ... a compacted chalk floor dating to C4 lay above a deposit of
PL chippings which covered a floor of flint, limestone, chalk and clay tile
fragments with a C2 coin embedded in it.
p.29: may have been roofed with PL in the late Roman period but with clay
tiles earlier in the Roman period.
- subst PL
- date 1996: Pottery evidence suggests use from Iron Age to C5
- date 2001: PL roofing probably late Roman (as desc above)
-
- name Hod Hill roofing
- number 18
- site Hod Hill, 1860
- grid ST 86 10
- source Notes by Oliver VL
- publ Shipp W, Hodson JW 1861
1 p.309
- desc on N (so Shipp and Hodson, in Oliver corr. to SE)
side of the camp, large quantities of stone roofing tiles,
many with large flat-headed iron nails still in the holes,
in a field called Great Bones. Coins of Constantinian period.
-
- name Tarrant Hinton roofing
- number 19
- site Tarrant Hinton
- grid ST 927 118
- source Roman Britain in 1973
- publ Tanner RM 1973
- publ Graham A 2006
- desc RB: Purbeck stone rooftiles over a workshop/store,
possibly used for lead working (charcoal and cerusite found)
- desc Graham p.55: `Building 1 ... timber roof tiled with
roughly diamond-shaped tiles of Purbeck stone'
- desc Graham p.58: `Building 2 was separated from Building 1 by a narrow
footpath less than 1.5ft wide, paved with broken roof tiles of both diamond type
and earthenware'
- desc A. G. Giles in Graham p.149-151:
Plate 27, photograph, stone roof tiles
... a group of these was found in the infilling of the well in the SE part
of the courtyard.
Fig.61, drawing:
`2. Standard roof tile of Purbeck limestone.
By far the greater majority [sic] of roof tiles used at Tarrant Hinton were of
this material and this size. In most of the roof tiles the nail hole was
offset, generally to the right of centre and often very near to the edge of the
tile. The most logical explanation of this is that the tiler started work at
the lower left hand side of the roof and that the tiles were offered up to the
roof timbers before the positions of the nail holes were marked. This would
seem to indicate that the tiles were fixed to battens, or the like, as had the
roofs been close boarded the position of the nails would be immaterial. However
the fact that some of the tiles exhibit nail holes that were obviously pierced
from the under to the upper surface suggests that some at least were pierced
before being offered up. 3. `Standard' roof tile of fine grained limestone
(possibly Lias from south Somerset). Although fragments of this material
were found on all parts of the site they were far less commonly used than
those of Purbeck stone. However, the majority ... recovered from the well on
Site 1 were of this material and those that remained intact were very well
preserved, some exhibiting lichen stains on their upper surfaces and thus
showing the exact way in which they overlapped ... 4. Non-standard roof tile,
smaller than usual. 5. Non-standard roof tile, much larger than usual.
Both ... 4 and 5 are of PL. It would seem difficult, if not impossible, to
incorporate tiles of these dimensions into roofs that were constructed mainly
of `standard' tiles. The non-standard tiles were not, however, found in
any quantity on any one part of the site and so cannot be allocated to any
particular roof. 6. Roof tile of PL cut in half longitudinally, pierced
with two nail holes, and most probably used to finish the tiles in a straight
line at the gable end ... 7. Roof tile of PL with the lower edge trimmed off
to give a straight line at the eaves. 8. Triangular roof tile of PL. Used
to finish the roof at the ridge. Only two of these tiles were found complete
and both were in the same non-standard size as ... no.5.
- desc Tanner: `The [Northern] wing proved to measure 91' by 36' externally,
with mortared flint walls 1'9" thick supporting a roof tiled with
Purbeck stone. Features and finds suggest that its function was for
working and storage ... Room IV contained a small storage bay for
locally made tesserae, and several large worked stone blocks were found
against one of the walls. In Room VI a small hearth was found with
material including a mixture of cerusite (lead carbonate) and charcoal'
- subst PL
- comment see also
mortars,
tesserae,
architectural stone,
structural stone,
inscription,
from this site
-
-
- name Iwerne Roofing
- number 22
- site Iwerne, 1897-1900
- grid ST 86 13
- source Salisbury Museum, Pitt-Rivers collection display
- publ Hawkes 1948 48--62
- desc Part of a limestone roof tile with iron nail;
displayed in museum, not mentioned by Hawkes
- loc Salisbury Museum, Iwerne display
-
-
- name Oakley Down tile
- number 25
- site Oakley Down, Wimborne St Giles, Dorset, Trench IV
- grid SU 016 177
- publ
Brown L, Corney M, Woodward PJ 1995
- desc 1 Purbeck limestone rooftile with a pierced fixing hole
- subst PL
- date prob. C3-C4 for filling of Trench IV
- comment Oakley Down is nearer Sixpenny Handley
than Wimborne St Giles !
-
- name Woodcuts Tiles
- number 27
- site Woodcuts
- grid ST 963 181
- publ Pitt-Rivers 1887
p.137 and Fig.13; see also p.135 and pl.XLVII no.47
- desc ``Roofing tile of Purbeck shale, highly fossiliferous.
These tiles vary from 0.4in to 1.1in in [thickness] and are perforated
by a round hole from 0.22 to 0.6in diam. in one of which an iron nail
was found fixed by corrosion. 648 frags were found in all. 86 had holes
or portions of holes. [81 of the 86 were] in surface trenching in the NW
quarter [the prosperous quarter]. Nearest stone of this kind is
between Wardour and Tisbury,
6 miles N.''
Footnote: ``Roofing tiles, in some instances having the
nails still in them, have been found with Roman remains at:
Baydon near Marlborough, Wilts (WAM X, 105); Holbury near Dean, Wilts
(WAM XIII, 40) and North Wraxall (WAM VII, 66). See also Wright,
Uriconium, 212''
- loc Salisbury Museum has 2 roof slab fragments on show; 1
looks like PL with gastropods, 1 has complex current?-bedding
- subst PL
- comment Pitt-Rivers implies that houses with tiled roofs were
mainly in the NW quarter, as terracotta tiles and stone tiles were both
mainly in that part of the village
- comment The Index (vol.5) to Excavations in Cranborne Chase
claims that one at least of the entries on
p.179 of vol.1
refers to a PL rooftile
-
- name Downton roofing
- number 28
- site Downton, Wilts, 1954;
"some 200 yards from Downton Moot" (Anon. 1954)
- grid SU 18 21
- source
Williams 1971 app.IIa
- publ Anon. 1954
- publ Rahtz PA 1963 p305
- publ Taylor and Wright 1956
p.142-3
- desc Purbeck limestone roofing
- desc Anon. 1953: Excavation directed by Morley-Hewitt of Fordingbridge.
Roof had been of Purbeck tiles, and some still retained nails.
- desc Rahtz p.305: ``A few heavy nails were all the evidence to
represent the roof timbers, on which were nailed hexagonal [JP's
emphasis) slates of PL ... capped by a ridge coping of
oolite''; p.326:
hand specimen examination by Helen AM Macdonald of Geol Survey: ``Roof
slates: sandy shelly limestone, similar to specimens of Purbeck age from
Swanage ... MOW ref. no. 570176''
- desc JRS: House of late C3 to early C4, not long occupied (only one
structural alteration), roof of Purbeck slates
- desc Macdonald in Rahtz p.326, and Williams,
say that the oolite is possibly from the
Portland beds of Tisbury
- subst PL by Geological Survey opinion
- date 4th cent.
- comment Beavis 1970 p.185, records
that these tiles, and similar ones at another villa S of Salisbury
(probably East Grimstead, JP),
were mentioned by CN Moore
at a CBA Gp 12 conference in Salisbury, Oct 1968 as possibly of
Wardour Vale origin
- comment the PL is likely to be from
Vale of Wardour also
- comment Illustrated in Rahtz' fig.16: get a copy
-
- name East Grimstead Tiles
- number 30
- cat roofing
- site East Grimstead, Wilts
- grid SU 22 27
- source
Williams 1971 app.IIa
- publ
Sumner H 1924
p.19, 30, 37 (but p.26 acc. to Williams)
- desc p.19: ``Bath-house I: bridging of hypocaust flue made of 3
layers of slab-stone roofing tiles: see also
PHFCAS 9 (2) 218
for a photograph of such construction''
- desc p.30: Bath-house II with tiled roof, each tile 14lb weight
- loc Sumner says: finds in Salisbury Museum
- subst p.37: PL ``Slab-stone roofing tiles from Purbeck''
- date Williams: 3rd-4th cent
- comment probably those described as of probable Wardour Vale origin by
Moore as quoted by Beavis
-
-
- name Stratford-sub-Castle (Stratford Road) tiles
- number 36
- site Salisbury, Stratford-sub-Castle, west side of Stratford Road, 1965
- grid SU 1340 3175 (west side of Road, James)
- source James 2002 p.5 (no.32), p.8
- publ Stratton 1966 p.106 tiles not mentioned
- desc James: Stone roofing tile (possibly PL by comparison with nearby sites)
- subst not stated
- date C1-4
-
- name Bishopdown (Salisbury) tiles
- number 38
- site Salisbury, Bishopdown, Paul's Dene Estate,
gardens between 14 and 22 Juniper Drive, 1953-55
- grid SU 1435 3225 (James)
- source Williams 1971 app.IIa
- source James 2002 p.5 (no.25), p.7
- publ - Stone and Algar 1955
- desc Williams p.117: from 14-22 Juniper Drive, fragments of perforated roofing
tiles of shelly limestone, one ovate specimen complete.
- desc James: Perforated (baked clay and) stone roofing tiles
- subst likely to be PL
- date C3-4
- comment Williams says WAM 1956 but it appears in the 1955
fascicule of vol.56
- comment site identical with Old Sarum Juniper Drive
-
- name Old Sarum NE sector tiles
- number 39
- site Old Sarum, NE sector, 1957
- grid SU 143 328 (James says 193 328, corrected from map)
- source James 2002 p.3, p.5 (no.26)
- publ Rahtz and Musty 1960 tiles not mentioned
- desc James: Pieces of Purbeck or Chilmark roofing tiles, with part of a
block of dressed Chilmark stone; not mentioned in Rahtz-Musty.
- subst PL, see interp below
- date C1-3
- interp I take "Purbeck or Chilmark" to mean a doubt about the quarry-site,
not about the origin of the tiles from the Purbeck Limestone Group.
-
-
- name Old Sarum Juniper Drive roofing
- number 40
- cat REJECT
- site Old Sarum
- grid SU 14 34
- source
Williams 1971 app.IIa
- publ - Stone and Algar 1955
- desc ?Purbeck limestone? roofing material
- desc Williams p.117: from 14--22 Juniper Drive, fragments of perforated roofing
tiles of shelly limestone, one ovate specimen complete.
- subst PL? shelly limestone
- comment Williams says WAM 1956 but it appears in the 1955
fascicule of vol.56
- comment reject because site identical with Bishopdown
-
- name Netheravon villa tiles
- number 41
- site Netheravon, Wilts, Air Force site, 1996
- grid SU 148 482
- source Rawlings 2001
- desc p.151: Large quantity of stone roof tiles, almost 150kg, mostly
small fragments, a few almost complete, some with surviving nail holes,
one with nail in situ. 3 main stone types: cream/white shelly limestone,
greenish-grey sandstone, red micaceous (Pennant) sandstone. 2 pieces of
oolitic limestone ... appear to have formed parts of larger architectural
fragments
- date C3-4, some activity C1-2
- comment the shelly limestone is the possible PL but "cream/white" is odd
-
- name Draycot Foliat roofing
- cat DUBIUM
- site Draycot Foliat, Wilts
- grid SU 191 770
- source RB in 1996 p.445
- desc `fallen roof slates' (could be Purbeck slabs, if `slate'
is loosely used here)
- subst unrecorded
- cont Bryn Walters, Association for Roman Archaeology
-
- name Gussage All Saints roofing
- cat DOUBTFUL
- site Gussage All Saints ?
- grid SU 00 10 approx
- source pers. obs. in DCM reserve collection 2003-04-22
- desc from shelf list: ``2 [pieces] Gussage A S ?''
- loc DCM, reserve coll, bay 158 shelf B,
1 box misc. rooftiles and stonework
- subst unknown, may even be ceramic
-
- name Okus building
- cat DOUBTFUL
- site Okus Quarry, Swindon, c.1899
- grid SU 145 835
- source Passmore AD 1899
- desc Roman building,
also mentioned under Quarries.
Fragments of walls, possibly a farmhouse. Lower layers of stone (in walls)
placed on edge in hard mortar. Many flue-tiles. Coin of Constans.
Much (fragmentary) wall-painting. Burial of young person, extended, feet E.
Stone drain 6ft deep. Much black (?BB1?) ware, Samian, etc.
Stone roofing tiles (material not stated).
Fragments of a cornice. Cellar 24 by 4 by 7ft, top 6ft below present surface.
Contained lumps of `hard blue Swindon rock found at a lower level'.
Skeleton of young female near the building,
surrounded with an oval of small stones
the size of cocoa-nuts, from the higher Portland beds, locally procurable;
date of burial unclear.
Description
- loc Site destroyed for road material
- subst Unknown, PS present, PL maybe
West Hampshire-
- name Rockbourne PL Roof
- number 24
- site Rockbourne, Hants (West Park)
- grid SU 120 170
- source
Williams 1971 app.IIa
- source inf. from Morley-Hewitt to Williams
- source Scott HA79
- source Bowen HC in
Rivet ALF 1969 p.47
- source Beavis 1970 p.185
- publ - Morley-Hewitt AT 1960,
Roman Villa, West Park, Rockbourne, Interim Report ns
- publ
Morley-Hewitt AT 1965
- publ Roman Britain in 1961 p.195
Tetricus stone
- publ - 1943, JRS 33 75 ns
- desc Morley-Hewitt: ``Room XXII [a peristyle or corridor with
a sloping roof around a courtyard] ... an almost complete collabsed roof of
Purbeck tiles, enabling one to measure and compare tilers' practices of
the Roman period and that [sic] of today.''
- desc Bowen: ``1 1/2 miles to the N [ of the Rockbourne Down villa ]
is the carefully built enclosure ... with
3 T-shaped corn-drying
ovens called hypocausts by the excavator.
... the flues of the ... ovens included ...
Purbeck stone roofing slabs''
- loc Several such slabs on display on site; one (pink-?stained)
at Salisbury museum
- subst PL
- date probably late 3rd cent (Morley-Hewitt says the building
was partly destroyed in late 3rd cent but reconstructed and fully
occupied till [367-8] and pertly occupied till after [410])
- date 3rd-4th cent
- comment RB in 1964, JRS 55,
RB in 1965, JRS 56
describe Rockbourne but do not mention PL
- comment RB in 1979, Britannia 11 p.394 describes
Rockbourne villa but does not mention PL
- comment probably from
Wardour Vale (Tisbury) outcrop
- comment hints from Williams
or Rahtz that there may be Purbeck
walling
at Rockbourne
-
- name Rockbourne Down Corn-driers
- number 26
- site Rockbourne Down, Hants
- grid SU 11 18
- publ Sumner H 1914 p.42
- publ Bowen HC in
Rivet ALF 1969 p.47
- desc Roman roofing tile slabs of Purbeck stone; 10 found of
larger size (15-18in by 12in) and 4 of smaller (12in by 8in). Mostly
used to bridge flues in hypocaust
- desc Bowen: ``1 1/2 miles to the N [ of
the Rockbourne villa ]
is the carefully built enclosure ...
with 3 T-shaped corn-drying ovens called hypocausts by the
excavator. ... the flues of the ... ovens included ... Purbeck stone roofing
slabs''
- subst Sumner: Purbeck stone (PL)
- comment ``the excavator'' is surely Sumner !
-
- name Martin roofing fragments
- number 45
- site Martin, Hampshire
- grid SU 07 19
- source Pers. comm. Iain Giles, June 2009
- desc `my garden and the ditch on the other side of the road,
that carries the winterbourne,
have yielded pieces of Purbeck limestone similar in thickness
to those used as roof tiles from the Roman Villa at Rockbourne.
One piece is about half of one tile
and has been identified at Winchester (Hampshire Museums Service)
as likely to have been a roof tile.'
- subst Giles: PL
- comment number 45, the first item added after writing
Palmer 2009
- cont Iain Giles
-
- name Bitterne
- number 21
- site Bitterne (Clausentum)
- site Russel: site SOU 857
- grid SU 44 13
- source Williams 1971, 178
- desc Williams:
`Purbeck limestone [roofing] may have been used at Clausentum'
- desc Pers. comm. Andrew Russel, 2001:
`You have a reference to possible Purbeck roof slates from Bitterne,
Southampton. I can confirm that we do have them from Site SOU 857 on the
Bitterne peninsula. Earlier sites threw out the stone unrecorded so we
do not know how widespread its use might have been. They seem to turn up
in the same area of the Roman town as grey and white loose tesserae.
David Peacock suggested the grey ones might be Dorset stone.'
- cont
Andy Russel, Southampton Archaeology Unit
-
-
- name Winchester, The Brooks
- number 32
- site Winchester, The Brooks
- grid SU 484 296
- source Roman Britain in 1989 p.318
- publ Zant JM 1993 p.107
- desc RB in 1988: S of the E--W street, and overlying traces of 2 earlier
periods of occupation, two parallel E--W timber buildings, separated by
a metalled alley 3.5m wide, were built in the NE angle of the insula in
the Flavian period ... Both demolished c.125--140 and replaced by a larger
masonry- or part-masonry house of double corridor plan ... In the late C3
the building was replaced by an L-shaped house which, probably in the
earlier C4, was converted to a courtyard building by adding a narrow third
wing on the S side. The house was extensively modified with tesselated
floors, 3 mosaics and a hypocaust ... The roof had been of Purbeck limestone
tiles.
- desc Zant: 'most of the building débris associated with XXIII.3
was removed by machine, fragments of stone roofing slates were seen to be
present (Period IV [c.400AD] Phase 56) though none of this material was
recovered'
- loc LOST
- subst PL
-
- name Winchester St George St
- number 33
- site Winchester, building N side of St George's St
in future central carpark (i.e. Brooks) area, 1953
- grid SU 484 296
- source
Williams 1971 app.II [and] p.178
- publ
Bennett-Clark MA 1954, p.320
- publ Roman Britain in 1953 p.101
- desc Excav in advance of widening St George St. Building
has a mosaic in red (tile) and white (chalk)
- desc PHFCAS:
``Plaster and roofing slabs [of PL from Dorset] lying on
the floors to a depth of 20in''
- desc JRS:
``a substantial building with a roof of Purbeck slabs''
- subst PL
- date ``From an occupation layer above the débris
came late Roman pottery, frags of glass imported from
Belgium and the Rhineland, and coins of Gratian (367-383) and
Arcadius (395-408)'' so could be early to mid 4th cent.
- date Williams: 3rd cent
-
- name Sparsholt roofing
- number 34
- site Sparsholt, Hants
- grid SU 415 301
- source Williams 1971 app.II and p.178
- source pers. comm. DE Johnston to Williams
- publ `Roman Britain in 1965' p.214 (no mention of PL)
- publ Johnston DE 1968 (no mention of PL)
- desc `the body of the house was roofed with Purbeck slates
apart from one room at either end'
- loc Winchester City Museum ?
- subst PL
- date 3rd-4th cent
-
- name Kings Somborne tiles
- number 37
- site Kings Somborne, Hants
- grid SU 36 31
- source Scott, E. p. 139
- publ Geographic J. (1923) p. 342 (map)
- desc Purbeck stone roof tiles.
- subst PL
-
- name Over Wallop slab
- number 44
- site Scott: Over Wallop, Hants
- site Piggott: Middle Wallop, top of chalk downs W of village
a few yards E of main road and approx halfway along the boundary fence
of the airfield
- grid SU 28 38 (Over Wallop)
- source Scott HA71
- publ Piggott S 1952,
Proc. Hants Field Club 17 60-63 but no mention of PL there
- desc Scott: A Purbeck stone roofing slab
- desc Piggott: At end of May 1939 [during construction of
Middle Wallop airfield] 3 graves oriented N-S were cut through ...
nos. 1 and 2 destroyed by trenching, 3 relatively intact, all Roman;
PL not mentioned.
- subst Scott: PL
East Hampshire-
- name North Warnborough Purbeck roofing
- number 42
- site Lodge Farm, North Warnborough, Hants, 1929-30
- grid SU 73 51
- source Williams 1971 app.IIa
- publ Liddell DM 1931 p.226
- desc Williams: Purbeck limestone roofing
- desc Liddell: `The Baths ... The roof was chiefly of red, flat, flanged
tiles, the joins of the flanges being covered with ... imbrices,
and partly of small (JP's emphasis) slates of Purbeck stone.
Most of these stone fragments came from room E and the adjacent part of G.'
... `Room G ... the lavatorium or washing-room.'
Fig 4 shows room E as a small chamber c.2m square.
E and G are both in the middle of one long side of the building.
(The building was originally a residence and was converted
to baths later, when a new house was built beside it).
- subst PL
- date 4th century. 71 coins, 67 of which are 308-381AD
- comment There is no PL roofing at Chilworth Ring (Liddell p.224)
-
- name Silchester
- number 43
- site Silchester, 1895
- grid SU 63 62
- publ
Hope and Fox 1896 p.218
- desc `The one house that stood within Ins.XIII .. from the broken
pieces .. evident that it was roofed with the usual thin hexagonal stone
slabs or slates' (nature of stone unspecified)
Elsewhere-
- name Gloucester, St Michael's
- cat DOUBTFUL
- site Gloucester
- grid SO 83 18
- publ Cra'ster MD 1961
- desc A few stone slates, diamond-shaped, one with the iron nail
in its hole; A3534, 3568. (Formation of stone unstated.)
- loc Gloucester Museum, ref. as above
-
- name St Martin de Brux church
- cat RELATED
- site Church of 12th-century build, Brux (Vienne),
40km SSW of Poitiers, France
- source http://www.ouvroir.com/brux/exter.htm (2004-11-07)
- desc Roof of stone slabs, not Roman ("Romane" means "Romanesque").
Described and illustrated at above website,
with French terms for stone roof slabs
To top of page and Search
To title page of Database
To front page
