_jl I I_

CHAPTER 11

REGIONAL GEOLOGY by C. Roger Bristow and Bernard C. Wo rssam

The geology ofsouth-west England is indicated in oudine STONE TYPES USED FOR TIIE SCULPTURES in Fig. 9 below;the succession ofstrata isshown in Table 1. In the south-west of the district is the high ground The various types of stone used for sculptures are fo rmed by theCarboniferous Dartmoor Granite intruded described in stratigraphical order. Geological fo rmations into older Carboniferous deposits. may contain other beds than those of building (or The Jurassic limestones, which provide most of the sculptural) stone quality: for example,not allthe Chalfield stones used in Anglo-Saxonsculptures, fo rm prominent, Oolite Formation is made up ofBath stone. mostly west-facing, escarpments.The Inferior Oolite is Bothsandstones and limestones were used, and most in a persistent unit that extends in an approximate north­ each category were arenites, i.e. having sand-sized south line across the whole district. Inthe northern part particles.Arenites lie within the total size-range 0.06- of the district are the Cotswold Hills underlain by the 2 mm diameter. Fine-grained arenites have particles up limestones of the Great Oolite Group. These limestones to 0.2 mm diameter, medium-grained 0.2--0.6 mm, and only extend to just south of Bath, and are replaced coarse-grained from 0.6-2 mm. For identification, the southwards by clays. The outcrops of the Corallian and stone of each sculpture was examined by hand lens wit h Porcland groups are intermittent as they are overlain, in a built-in measuring grid, allowing the diameter of the part, by the unconformable CretaceousGault and Upper grains above 0.1 mm to be measured withconfidence. Greensand fo rmations. The main type of stone used was Bath stone from the The Upper Greensand gives rise to a prominent Chalfield Oolite Formation,which waswidely exploited escarpment in the southernpart ofthe districtand forms in the Bath area. In addition, some ten types of stone the high ground of the Black.down and Haldon Hills, from localities in, and one (Taynton stone) from just respectively north-east and south-west of Exeter. The outside, south-west England were used. A fe w carvings Chalk, which hasa wide outcrop inthe east of the district, are ofstone of uncertain provenance. See Fig. 8. gives rise to tracts ofhigh ground with a generally open aspect. DARTMOOR GRANITB by R. C. Scrivener The Palaeogene sands and clays occur in a downfold in the Cretaceous strata in the south-east of the districtand Dartmoor is the largest of the granite bosses that trend occupy the western part of the Hampshire Basin. west-south-west from Exeter to Land's End and covers In the centre of the district are the alluvial and peaty an area of some 250 square kilometres. Characteristically, deposits fo rming the low-lying Somerset Levels. the granite gives rise to rugged tors overlookingareas of Table 1 shows the relation between chrono­ moorland or rough pasture. Ingeneral, the granite consists stratigraphical divisions based on geological age, and of quartz, perthitic orthoclase crystals up to 17 cm long, lithostratigraphical divisions (as in Fig. 9), represented in plagiodase and dark brown mica (biotite). Common south-west England. accessoryminerals are tourmaline, zircon and apatite.

13

11 11- _j l IL

14 CHAPTER I!

0 50km I

p

Pg ss su sx sz

Alluvium and peat Fuller's Earth to Oxford Clay Palaeogene strata Bradford stone and Great Oolite limestones (including Bath stone) Chalk Fuller's Earth (north of Bath) Upper Greensand and Gault Inferior Oolite (including Dundry and Doulting stone) Portland, Purbeck and We alden groups Lias Group (including Ham Hill stone) Kimmeridge Clay Palaeozoic strata Corallian limestones (including Osmington Oolite and To dber Freestone)

FIGURE 9

Geology of south-west England

11 11 _jl I I_

REGIONAL GEOLOGY 15

Chalk Group Boyne Hollow Chert Upper Greensand 0� SbiftaburySancutone tl Formation � Cann Sand � Gault u Lower Greensand We alden Group

Pllltleck Made Purbeck Group

Stoncl l'ordlad,_ I"' - a Wocldey Member 1, PortlandChert Portland Group -TiiliiiiY.MmaJo.. _ Ji':] Portland Sand Kimmeridge Clay

OamiDpm Oolite Corallian Group u 1bclber fleeltxlae ."'.... "' l Oxford Clay & Kella�Fonnation Cornbrash Forest Marble

<""' IIIDDe•AIIdilFOalile Great Cbalfield Bath Oolite Oolite Oolite Frome Formation Tw inhoe Beds Group Clay (Bathstone) Combe Down Oolite Tavnt ou Limestone Fm= Lowerer's Full Earth DoultiDg/_.,. stone Inferior Oolite

Ham stone Hill - Lias Group BlueLW

TABLE 1

Gc:nerilised vertical section (not toscale) ofthe Mesozoic stnta insouth-west England. Stone used for Anglo-Su:on sculptures highlighted

11 11- _jl I I_

16 CHAPTERII

In general, the granite can be divided into an upper (outer) one of'giant' granite which fo rms most of the tors and is characterised by large rectangular fe ldspar crystals, and a lower (inner) one of'blue' that contains fe wer and smaller fe ldspar megacrysts. Muscovite (white mica) occurs as a secondary mineral resulting from post­ crystallisation pneumatolytic or hydrothermal activity, and its presence commonly renders the granite more amenable to being carved or shaped (cf. the Copplestone 1 and Exeter 1 crosses). The typical granite of most of northern Dartmoor, the nearest part of the outcrop fo r the fo ur granite carvings in the present area, isa coarse biotite granite with variable abundance ofK-feldspar megacrysts and does not appear to be the source of the stone for these carvings. The closest match for Copplestone 1 would be the granite of the Merrivale area near Princetown, where coarse­ grained biotite-muscovite granite was worked at Merrivale Quarry until recendy. The stone ofExeter 1

could be a 'contaminated' granite of the type fo rmerly StrauYQUDger thm worked at Swcltor in southern Dartmoor. The Plymstock Ham HillStoue 1 cross is typical of granite on the south side of the Ham HillStoue § Str.ata older thau batholith. The very fresh and hard biotite granite with HamHill Stone sparse K-feldspar megacrysts ofthe Chulmleigh 2 stone (Appendix B) is verysimilar tothat quarried in the Haytor ,) quarry 'blue' granite quarries in the nineteenth century, and extensively used fo r building and facing stone, also by monumental masons. Because of its relative intractabilty FIGURE 10 to the techniques used by the early masons, Chulmleigh Outc:rop ofthe Ham Hill Stone 2 may be a late fa ke.

LlAS GROUP,BLUE LlAS FORMATION LlAS GROUP, BRIDPORT SANDFORMATION, HAM HILL STONE MEMBER. The Blue Lias Formation consists dominandy of an alternating sequence of hard, bluish grey, fine-grained, The Ham Hill stone, named from Hamdon Hill, south sparsely shelly limestones up to 0.25 m thick, alternating of Stoke-sub-Hamdon, Somerset, is an orange-brown, with grey mudstones of similar thickness. The fo rmation, shelly, strongly cross-bedded (not obvious in small blocks) which is thick (up to 144m) and widespread in Somerset, limestone developed at the top of the Bridport Sand thins over the Mendips and is only about 5 m thick Formation over a limited area in southern Somerset north of the Mendips in Gloucestershire. (Fig. 1 0). Because ofsuperficial lithological similarity, the In Somerset, the Blue Lias has been widely, and still is Ham Hill stone has sometimes been included as part of locally, quarried as a building and walling stone, but, the Inferior Oolite. However, in lithological detail and because ofits hardness, hasrarely been used fo r sculptures. age (as determined by ammonites) it can be distinguished It has also been widely used for gravestones, but is not fro m the Inferior Oolite. The stone on Chiselborough particularly long-lasting as weathering causes it to split Hill about 2 km south of Hamdon Hill has also been and flake. called North Perrott stone (Green 1998). Although the One carved stone (Muchelney 4) from Muchelney Ham Hill stone is up to 27 m thick, only about 15 m Abbey (much of which isbuilt ofBlue Lias limestone) is of it is suitable fo r working as ahigh-<:lass freestone. The a pale grey, sparsely shelly micrite and resembles the stone has been worked from Roman times (Wilson et al. limestones of the Blue Lias, except that it is fa irly soft 1958, 63) and can be fo und in buildings over a wide area (it easily marks with a soft wire brush) and presumably is (Thomas 1990, 297; Robinson 1995, 163-4);Montacute fa irly easy to carve. House and Sherborne Abbey are good examples.

11 11- _jl I I_

REGIONAL GEOLOGY 17

It continues to be worked as a building stone and stone INFERIOR OOLITB GROUP, UPPER INFERIOR OOLITB fo r renovation at both the northern and southern end of 'FORMATION', DOULTING STONE MEMBER Hamdon Hill. The Inferior Oolite Group is a widespread and persistent Sculptured Ham Hill stone has only been ide tified at ": limestone unit across the present district. North of the two localities (Muchelney, Somerset and Stmsford, Mendips, the Inferior Oolite is up to 100 m thick and Dorset) . At Muchelney,two (Muchelney 2 and 3) of the can be divided into mappable fo rmations. From the fo ur sculptured stones are of Ham Hill stone. One Mendips southwards to the Dorset coast, the Inferior (Muchelney 2) is a very pale orange (10YR 8/2), very Oolite is generally a thinner (1-25 m), but nevertheless shelly limestone. Most shells {bivalves) are elongate and persistent, unit. Within this latter tract, however, �e vary from 0.5 to 6.0 mm across and vary from sub­ Inferior Oolite cannot be subdivided into mappable un1ts angular to sub-rounded and are clast supported. A crude and is regarded as a fo rmation (see Bristow et al. 1995; alignment can be seen with a hand lens. The second id. 1999, 30). However, to conform with earlier Corpus stone (Muchelney 3) is a pale yellowish orange (10YR volumes, in this account it is afforded group status. It 8/6) shelly limestone. The shell clasts are in the range appears thatstone from the Mendip area, and also fa rther 0.5 to 5.0 mm, but most are about 2 mm across. The south in southern Dorset, hasbeen used Anglo-Saxon clasts are sub-rounded to elongate {most common) and for sculptures. clast-supported. The yellow,fe rruginous, matrix is f.Urly Where well developed, the Inferior Oolite can be soft. The Stinsford 1 angel is a greyish orange (10YR 7/ divided into three 'formations' (Lower, Middle and 4), coarsely shelly, matrix-supported limestone, wi� a Upper) . For the most part, south of the Mendips, the matrix of crystalline calcite. A fe w scattered quartz grams Lower and Middle Inferior Oolite is thin or absent and and ooliths {or verywell-rounded shell fragments) occur, it is the Upper Inferior Oolite which has been widely togetherwith a few, small scattered dark (?siderite) grains quarried for building stone and used fo r sculpting. In and with some larger (up to 3 mm) dark fragments. Some Dorset alone, Thomas (1990) records 138 quarries in 20 of the well-rounded clasts and larger patches are parishes; the last working quarry, near Beam�ter, closed limonitised.Shell fragments (thinbivalv es),up to 10 mm in about 1999 .Innorth Dorset, a grey,fine-gramed, sandy across, not aligned and generally pale grey in contrast to limestone (see Bristow et al. 1995, pls.1C and 10), up to the matrix, occur. 15 m thick, the Sherborne Building stone, has been extensively quarried as a local building stone. It has been used in the exterior walls of Sherborne Abbey. Farther (I Ovenil extent ofDoulting north in south Somerset, the equivalent of the upper and probable Doulting stone '---/ part o the Sherborne Building stone consists ofmassively used inAnglo-Saxon sculp f bedded, somewhat nodular, fe rruginous, brown, s arry, / Outcrop of the Inferior � Oolite peloidal detrital,very fo ssiliferous limestone (see Bnstow et al. 1999, pi. 2) and is known as Hadspen stone. It was used in the construction ofHadspen House. The best­ known and most widely used Inferior Oolite stone in south-west England is Doulting stone from Doulting, about 4 kmeast ofShepton Mallet, Somerset. The stone has been worked from Roman times up to the present day. Doultingstone, up to 9 m thick, is a warm, yellowish orange (10YR 7/6) where freshly exposed (see Bristow et al. 1999, pl. 1), but weathers to a yellowish grey (5Y 7I 2 or SY 8/1), cross-bedded, medium- to coarse-grained, • Anglo-Saxon locality using Doulting stone bioclastic limestone fo rmed largely of the fragmented "'InfCrior Oolite limestone,probably Doulting stone remains of crinoids. It had been widely used fo r local buildings (e.g. We lls Cathedral and Glastonbury Ab��) and fa rther afield in Oxford, London and other c1tles. Similar, but paler, bioclastic limestones (with coral FIGURE 11 fragments) on Dundry Hill to the west are known as Distribution ofAnglo-Saxon sculptures using Doulting stone Dundry freestone.

11 11- _jl I I_

18 CHAPTER II

Doulting stone is the only unit of the Inferior Oolite colour, while that from other localities, e.g. Sherborne, to be identified with certainty that has been used in Windrush and Burford, is a paler, yellowish grey. Anglo-Saxon sculptures in south-west England. It has The fo rmation is overlain by the clayey Hampen been recognised at Aller 1, Frome 2, Glastonbury 8, Formation, and that in turn by the White Limestone Holcombe 1, Keynsham 13, Shepton Mallet 1 and 2, and the Forest Marble. It is underlain by a thin Fuller's and Wells 5 (all in Somerset) (Fig. 11). Although at most Earth day (Worssam and Bisson 1961). The work of places yellowish grey (SY 7 /2), at Aller it is a pale "Wyatt (1996) and others has shown that southwards along yellowish brown (1 OYR 6/2), clast-supported, bioclastic the Cotswolds the Taynton Limestone and overlying beds, limestone. Clasts vary from sub-angular to rounded and up to near the top of the White Limestone, pass laterally from 0.2 to 5.0 mm across, but most are in the range into the upper part of the Fuller's Earth of the Bath 0.3 to 2.0 mm. district. Stones used fo r Puddletown 1 and 2, Wa reham 1-9, The Anglo-Saxon sculptures at three localities, Whitchurch Canonicorum 1, Whitcombe 1 and 2, and Inglesham, Cricklade and Eysey (all in Wiltshire), are Winterboume Steepleton 1 and 2 (all in Dorset) differ identified as of Taynton stone. At Inglesham is a large from typical Doulting stone by their better grading and carved slab of pale brownish yellow (10YR 8/3-4) the higher proportion of well-rounded clasts, some of unevenly graded shelly oolite. Only one surface, probably which closely resemble ooliths. Stones from these last a bedding plane, is exposed. However, the columns of localities consist of yellowish grey (5Y 7/2) or pinkish the early thirteenth-century south nave arcade of grey (SYR 8/1), fairly well-graded, medium-grained, Inglesham church match the lithology of the slab, and bioclastic limestone, with bioclasts sub- to well-rounded also, in cross-section,show the hard shelly streaks or 'bars' and clast supported; clasts range from 0.2 to 1.0 mm that typify Taynton stone. The locality is on the river across, but most are in the range 0.3 to 0.6 mm. The Thames and less than 20 km from Taynton. source of these stones has not been determined. They Both Cricklade 1 and 2, built into a wall in the north are presumed to be Inferior Oolite as their lithology is porch of St Sampson's church, also show only one fa ce typical of that fo rmation and possibly these are the 'Top of the stone, which however in texture and colour Beds' or 'Top Limestones' fo rmerly exposed in many of resembles the stone at Inglesham. Like lnglesham and the quarries in west Dorset (Wilson et al. 1958, 76) (one Eysey, the town has a Thames-side location. of the lithologies present in the last-worked pit near Beaminster is a medium-grained biodastic oolite). As GREAT OOLITE GROUP, CHALFIBLD OOLITE the Inferior Oolite has been so widely exploited in FORMATION (BATH STONE) Dorset, a Dorset origin seems likely. Bath stone is a yellowish grey (10YR 8/2-3) cross­ bedded oolitic limestone, consisting predominantly of GREAT OOLITE GROUP, TAYNTON LIMESTONE eoliths of0.3 to 0.6 mm diameter, together wit h ovoid FORMATION (TAYNTON STONE) pellets up to 0.9 mm in length and scattered shell The Taynton Limestone Formation provides a good­ fragments of 5 mm or so diameter, in a crystalline calcite quality building stone, Taynton stone, at its outcrop along matrix. On weathering,eoliths at the surface of the stone, the valley of the river Windrush, a tributary of the being relatively soft,are eroded awayto leave a hard pitted Thames, fo r some 10 km upstream from Burford, calcite surface. A fe ature of much Bath stone is the Oxfordshire (see Arkell 1947b; Wo rssam and Bisson existence of calcite veinlets of about 2 mm width, cutting 1961). Quarries at Taynton, 2 km west ofBurford, were across the bedding; the veinlets are known to quarrymen mentioned in the Domesday Book (Morris 1978, 13, 1). as 'watermarks'. The fo rmation, up to about 12 m thick, consists of Bath stone was the principal stone in use fo r Anglo­ massive, cross-bedded shelly oolitic limestones, that Saxon sculpture in Wiltshire, in Somerset and in the originated as underwater dunes in a shallow sea. On northern part ofDorset. It had been used by the Romans lithification, pore spaces between the constituent eoliths both in Bath and for large monumentalworks in London. and shell fragments became infilled with calcite, a Perkins et al. (1979) tentatively identified two areas of crystalline fo rm of calcium carbonate. The calcite cement Roman quarrying:just south ofBath near the Fosse Way predominates in streaks with more abundant shell (on Odd Down); and east ofBath,near Box. SomeAngle­ fragments that weather out as ribs or 'bars'. Stone from Saxon sculpture may be of re-used Roman stone from the Taynton quarries has typically a brownish yellow Bath, but Biddle andKj 0lbye-Biddle (1995, 102-5) fo und

11 11- _jl I I_

REGIONAL GEOLOGY 19

use of Great Oolite as a fo rmation name, as well as for • 0-..p CIWfiddaflbo Oalite Fo=ation the larger unit the Great Oolite Group, have proposed • �locolity...,,. wiagBuh � the term Chalfield Oolite Formation fo r Green and <) ����m���== Donovan's Great Oolite exclusive of the Upper Rags �;E��� {see also Sumbler etal. 2003) . Notall of the Chalfi.eld Oolite Formation is ofbuilding­ stone quality. The principal worked freestones occur in the upper part of the Combe Down Oolite and in the Bath Oolite in the Bath vicinity. On approaching an east-west line about 8 km south of Bath the whole Chalfield Formation passes laterally into a clay formation, the Frame Clay. This area is shown in Fig. 13 (which is based on fig. 3a of Green and Donovan {1969), with isopachytes from Penn and Wyatt (1979)). The Chalfi.eld Oolite extends for some 15 km northwards from the area included in Fig. 13 (see Wyattand Cave 2002, fig.1). There is litde evidence offreestone having beenworked in that northern tract. FIGURE 12 The resistance ofCombe Down Oolite to weathering Distribution ofAnglo-Saxon stone types usingBath stone tends to be better than that of stone from theBath Oolite, compelling evidence from their Winchester excavations the matrix of which may include micrite (fine-grained that Bath stone had been freshly quarried at all Anglo­ calcite).While mostAnglo-Saxon Bath stone shows good Saxon periods from the seventh century onwards. resistance to weathering and maywell be from the Corn be At least from the twelfth to the sixteenth centuries Down Oolite, the distinction between this division and quarrying seems to have been concentrated at Hazelbury, the Bath Oolite at Hazelbury and at nearby Box Hill near Box. Old workings there are indicated by 'hills and {Green and Donovan 1969, 22-3) is not clearly marked, hollows' in a wood (ST 832688) just below the crest of since the Tw inhoe Beds that separate them are there a steep escarpment on the east side of the Avon valley. thin, approaching their northern limit. If, therefore, Records have survived of grants of stone from Hazel bury Hazelbury was a main source ofAnglo-Saxon, as well as to religious houses from c. 1189 to 1306, and in the oflater medieval stone, the possibility of some of it being fifteenth and sixteenth centuries fo r Great Chalfield Manor House and Longleat House {Saunders 1959); ChalliddOolite while the fa bric accounts of Salisbury Cathedral record D Formation outcrop freestone'brought from Haselbury' in 1479-81 attoo­(T Chalfield Oolite Formation Brown 1998). Early quarries at Hazelbury would have D been rather better placed fo r the transport of the stone eastwards towards its main markets, by cart across a gendy sloping terrain, than quarries at Odd Down or Combe Down, west of the deeply cut Avon valley. In the early nineteenth century, the term Great Oolite, based on the Bath area, was introduced fo r oolitic limestones that occur as a group between the Fuller's Earth {mainly clays), below, and the Forest Marble (shelly limestones and clays), above. Green and Donovan (1969) defined the Great Oolite as a lithostratigraphical unit comprising fo ur subdivisions: the lowest being the Combe Down Oolite; separated by rubbly and marly �en limestones named the Tw inhoe Beds from the Bath ----�ruo -- emlimit ofChalfield oolite Oolite, above; and that in turn overlain by a highest or 'Upper �·divi sion.Penn and Wyatt(1979) transfer red FIGURE 13 the Upper Rags to the Forest Marble. Wyatt and Cave lsopachytes inmetres of the Chal.fi.eld Oolite Fonnation inthe (2002), in view of potential confusion arising from the vicinity ofBath

11 11- _jl I I_

20 CHAPTER II

fragments of polyzoa (or bryozoa) are common. Ovenllextent of.Anglo-Saxon 1\ Mr G. W. Green (pers. comm.) has kindly identified L.::::. &CUI.ptures usingBradford stone the stone used in the constructionofSt Laurence's Chapel in Bradford-on-Avon, and fo r the decorated slab (Bradford-on-Avon 5) in the chapel, as Bradford stone. Quarryingof the stone therefore undoubtedly took place in the Anglo-Saxon period. The AndiH" Oolite crops out on the slopes of the Avon valley at Bradford, dipping eastwards to river level within 1 km east of the town, and stone could have been transported downstream along the Avon from quarry workings to sites such as Bath Abbey (Bath 1 and 2) and Bristol. Additionally, Bradford stone has been recognised at Glastonbury, Nunney, Lullington and Hanging Langford (Fig. 14).

CORALLIAN GROUP,OSMINGTON OOLITE FORMATION/TODBER FREESTONB and CLAVEILATA FORMATION, ECCLIFFE MEMBER

FIGURE 14. There are two principal outcrops of limestones of the Distribution of Anglo-Saxon carftd stones utilisingBndfonl stone Corallian Group in south-west England:in south Dorset from Redcliff Point north-east of We ymouth to Bath Oolite cannot be ruled out. Calcite veinlets are to Abbotsbury, and from near Buckland Newton in mid­ be seen inmany Anglo-Saxon Bathstone sculptures.They Dorset to just north-east of Wincanton in south are known from the Combe Down Oolite, but whether Somerset. In Wiltshire, there are smaller outcrops near they occur in the Bath Oolite also is uncertain. In this Tr owbridge and Calne. Within these tracts, it is only the volume, therefore, the term Bath stone rather than upper part of the Group which contains limestones and Combe Down Oolite is used in description of Anglo­ it is only some of thehigher limestones (Calne Freestone, Saxon sculptures. Osmington Oolite and To dber Freestone) which have been worked as freestones.A locally developed freestone unit, the Eccliffe Member, at the top of the Clavellata GREAT OOLITE GROUP, FOREST MARBLE Beds, present between East Stour and Bourton in north FORMATION,ANCLIFF OOLITE MEMBER is Dorset. In hand specimen, it is almost identical to the (BRADFORD STONE) To dber Freestone. Other, more massive or flaggy, The Upper Rags division extends northwards from Bath limestones (Cucklington Oolite, Clavellata Beds) have as far as Cirencester and then eastwards to near Fairford been used as building or wallingstones. (Perm and Wyatt 1979, 46; Wyatt 1996, 314). Over most Thomas (1990, 298-9) notes thatAbbotsbury Abbeyis of this tract it consists mainly of variably oolitic, coarsely built of Osmington Oolite which probably came from shell-fragmental limestones. Inthe Bath area, Green and large quarries on Linton Hill to the east of the Abbey. Donovan (1969, 19--22) recognised three subdivisions These quarries fe ll into disuse in the nineteenth century. of the Upper Rags, comprising basal shelly beds 1 to In the northern tract, both the Todber Freestone and 2 m thick and including the Corsham Coral Bed; Clavellata Beds were extensively worked fo r building overlain by the AncliffOolite, up to 8 m thick; capped stone (much of the villages ofHinton St Mary,Marnhull, by the Bradford Coral Bed up to 2.5 m thick.The Andur To dber and Stour Provost are built of these stones). All Oolite, named after Andiff or Avo ncli.ff, is described as are now disused, although the last one did not cease consisting of detrital-shelly oolitic limestones becoming operations until about 1999. more purely oolitic southwards, so that south and east of No sample of Calne Freestone has been examined. We stwood (Fig. 13) theyhave been worked as a freestone. Green (1998, 12) describes it as a 'white, cross-bedded, In thisaccount, stone from the Ancliff Oolite isref erred pellet limestone with much well-rolled fo ssil debris'. to as Bradford stone. Its shelly nature, often with �rs Samples ofthe To dber Frees tone fromthe type locality of shell fr agments along planes of crossbedding, consist ofyellowish grey (SY 8/1 ), medium-grained,well­ distinguishes it from Bath stone. Among the shell debris, graded, matrix supported, variably shelly, cross-bedded

11 11- _jl I I_

REGIONAL GEOLOGY 21

oolite. The eoliths, which vary from 0.3 to 0.6 mm the unit is known as the Portesham Stone (Green 1998). diameter, mostly weather proud in hand specimen, but a Thin sections of Portland Stone {Base Bed freestone, fe w breakacross or weather out to leave small depressions. Whit Bed freestone, Fancy Beach Bed stone) from the The eoliths are closely packed, but nevertheless occur Isle of Portland examined by Dr G. Lott of the British 'floating' in a crystalline matrix (see Bristow et al. 1995, Geological Survey, are fm e- to medium-grained, with pi. 6B). Sparse, comminuted, thin, shell fragments the medium-grained eoliths mostly in the range 0.25 to (?bivalves), up to 5 mm across occur. The only whole 0.35 mm, but with a fe w larger ones. Fine-grained fo ssil is the occasional burrowing echinoid Nucleolites 'eoliths'appear to be peloidal, ratherthan oolitic. All the scutatus. thin sections show good intergranular porosity with little Samples of Osmington Oolite from Linton Hill, or no spar cement; the ooliths are held together at point Abbots bury,vary from yellowish grey (5Y 7/2) togreyish of contact. The bioclastic debris, which can be coarse­ orange (1 OYR 7I 4), medium-grained, well-graded, grained,varies from sparse to abundant. matrix-supported,patchily shelly, oolite. The ooliths are In the Vale of Wardour, the Portland Stone has mostly between 0.4 and 0.6 mm in diameter, although traditionally been divided into three units:in ascending a fe w are up to 0. 9 mm. The density of the ooliths varies sequence, the Lower Building Stones, a median from almost touching, to much more obviously 'floating' unproductive unit, and the Upper Building Stones (Blake in a crystalline matrix. Thin platy (?bivalve) fragments, 1880; Wo odward 1895). Wimbledon (1976) formalised up to 6 mm across have abraded to well-rounded the stratigraphical nomenclature and recognised fo ur margins. A fe w larger shell fragments (up to 1.5 mm members. More recently (Bristow 1995; Bristow et al. long by 2 mm thick) also occur. 1999), modified Wimbledon's scheme to include only It is the matrix-supported nature ofthe eoliths which three members: in ascending sequence, the Tisbury distinguish the Corallian Group oolitic limestones from (15-25 m thick),Wo ckley (D-15 m thick) and Chilmark those in the Portland Group (see below) . However, this (G-?10 m thick) members. distinction is not always clear cut in some of the poorly The Lower Building Stones are generally referred to as exposed and/ or badly weatheredAnglo-Saxon sculptures. Tisbury stone or Tisbury freestone in the Tisbury area, Sculptures at Banwell, Dolton, Colyton and Ye tminster and Chilmark stone in the Chilmark area. It is unfortunate are attributedto the Corallian Group.The Ta ller Frattum that there is potential confusion between the terms angel (Appendix B) is too heavily encrusted fo r certain Chilmark Member (Wimbledon 1976) and Chilmark identification and is only tentatively regarded as from stone. To avoid anyambiguity , in this account, we revert the Corallian Group (it could equally well be from the to the old name of Upper Building Stones fo r stone Portland Group). from the Chilmark Member The Tisbury Member has a wide outcrop on either side of Tisbury, and occurs as a narrow inlier in the PORTLAND GROUP, PORTLAND STONE FORMATION, Chilmark valley (Fig. 15). It has been extensively used as TISBURY and UPPER BUILDING STONE MEMBERS a building stone (e.g. in Salisbury Cathedral (Tattoo­ (Vale ofW:udour) and PORTLAND FREESTONE MEMBER Brown 1998), Wilton House, Old Wa rdour Castle).The (Isle ofPurbeck) medieval quarries just east of Tisbury that may have The Portland Group in south-west England crops out in supplied the large amount of'Chilmark stone' needed fo ur main areas: in Dorset on the Isle ofPurbeck, on the fo r the construction of Salisbury Cathedral (Tattoo­ Isle ofPortland,and from near Osmington to Portesham Brown 1998, 41), may have also been in existence in in the west, and in Wiltshire in the Vale ofWardour. It Anglo-Saxon times. Some thirty-three quarries are has been worked, extensively in places, as a building stone known in the Tisbury and Chilmark areas; two wherever it occurs. (Chicksgrove and the Lower Quarry, Chilmark) are The Portland Group is divided into two fo rmations: worked at the present day (Fig. 15). the Portland Sand below and the Portland Stone above. The Tisbury Member has commonly been referred to In Dorset, the Portland Stone is divided into two as a calcareous sandstone, but in fa ct, thinsections of50 members: the Portland Chert below and the Portland samples studied under the microscope by Dr G. Lott Freestone above. It is stone from thelatt er member which fromacross the whole outcrop and from boreholes show has been widely used forbuilding (e.g. St Paul's Cathedral) that it is really a fine-grained, sandy, locally very sandy, and, locally, fo r Anglo-Saxon sculptures. In Dorset, the bioclasticlimestone. The quartz sand content varies from Portland Freestone thins westwards from about 15 m at 4 to 42 per cent, but averages about 30 per cent {Bristow Portland to 4.5 m near Portesham. In this latter area, et al. 1999, 64) . The sub-angular to sub-rounded quartz

11 11- _jl I I_

22 CHAPTER II

grains vary from 0.06 to 0.18 mm across; the bioclastic medium-grained,clast-supported oolite with a micritised debris varies from 0.25 to 0.35 mm. Tattoo-Brown's matrix; most ooliths are in the range 0.1 to 0.3 mm (1998, 41) statement that Tis bury stone is less green (i.e. diameter (see Bristow et al. 1999, pi. 3b). less glauconitic) than Chilmark stone appears to be true Tisbury and Chilmark stone could have been only fo r some of the currently worked stone. Across the transported eastwards along the river Nadder, a tributary whole outcrop theof Tisbury Member, the glauconite of the Salisbury Avo n. content in different parts of the sequence can varyfrom Three examples of stone from the Tisbury Member 0 to 6 per cent. have been identified (Amesbury 2, Shafi:esbury 7, and The Upper Building Stone, fo rmerly thought to be the Knook 2 tympanum) and several examples of the restricted to the Chilmark Va lley, is now known to extend Upper Building Stones (Knook 1 , Britford, and possibly westwards as far as just north of Tisbury (see 1 :50,000 East Stour and Gillingham).Allother sculptures attributed Wincanton Geological map, 1996). However, it appears to the Portland Group (Batcombe Down, Buckland to have been exploited only in the Chilmark Va lley. Newton, Cattistock?,Cranborne, Henstridge, Melcombe The one example of the Upper Building Stones from Horsey and Melbury Bubb) appear to be Portland the Upper 'Teffont Evias' Quarry in the Chilmark Va lley Freestone. thatDr Lott examined consisted of well-graded, fm e- to

9

Pb

Purbeck and younger strata

0 Medieval quarries Upper Building Stones Portland * Fifteenth-century quarry Stone Fm. Mainly 19thand 1 • 20th-century quarries � Fault

FIGURE 15

Geology of thePortland Formation inthe Tisbury and Chilmaik areas

11 11- _jl I I_

REGIONAL GEOLOGY 23 PURBECK GROUP, DURLSTON FORMATION, PEVERIL sandstone indurated and knownas the'Rag' or 'R.agstone', POINTMEMBER, UNIO BEDS capped by the Boyne Hollow Chert which consists of glauconitic sands with common chert nodules.In south­ The Purbeck Group has a similar distribution to the �st Dorset (south and eastwards from Crewkerne), a Portland Group, wit h the exception that on the Isle of higher,fo urth member, the Eggardon Grit, consisting of Portland, only the lower, non-marble-bearing, part of well-cemented, fine- to coarse-grained, bioclastic, the group, the Lulworth Formation, is present. calcareously cemented, fo ssiliferous sandstone, occurs. Throu out itsout rop,the limestones ofthe group have � � The Shafi:esbury Sandstone has been extensively used been wtdely quarned fo r building and walling stones as a building stone - much of the old towns of Mere and tiles -Purbeck stone. H er, only locally on the owev and Shaftesbury and Shafi:esbury Abbey is built of this Isle of Purbeck are the marble-bearing beds in the material - and as a roadstone (particularly the Rag). uppermostmember (Peveril Point Member) of thegroup There are numerous quarries in and around Shaftesbury present (see Haysom 1998,£g.12). Inthe Va le ofWardour, and also to the north and south-west.The grain size of Towl son (1991, 43) refers to 'Fovant Marble', named in the weakly cemented sandstones which occur between 1862, and which has been speculated to have been used the better-cemented stones, variesfrom 0.14 to 0.26 mm in Salisbury Cathedral, although To wlson believed it to (average of six samples, 0.2 mm). H r, the harder, have been a later working.To wlson also refers to a buff­ oweve better -cemented,stones appear to be slightlycoarser and coloured 'Cyrena' or 'Neomiodon Marble' that was with more bioclastic debris. Hurdcott stone named from exposed ab e the adit in the Upper (Teffont Evias) ' � the village about 1 0 km east of Tisbury, is a local name Quarry,Ch1lmark, but states that it was unlikely to have fo r the Shaftesbury Sandstone; it wasmuch used in south been of economic value. Wiltshire in the tweltth century (for example at Old On the Isle ofPurbeck, the Marble occurs as thinunits Sarum) (Tatton-Brown 1998, 39). The Boyne Hollow interbedded with shale. Different units have been Chert has been used in walling and as roadstone. The recognised,named in ascending sequence 'Blue' 'Green' EggardonGrit has been used as a building stone -Forde and 'Grey Marble (see Haysom 1998, £gs. 11 d 13). � Abbey is built out of this stone (Thomas 1990). Around The coffin ofKing JE.thelberht(?) in Sherbome Abbey Potterne, where the Upper Greensand is not subdivided (Appendix B) is made up of several slabs of Purbeck into mappable units, a soft, fine-grained malmstone Marble. MrPaul Enson of the Natural History Museum known as the 'Potterne stone' has been worked. and Trel n Haysom suggest (pers. commns.) that . � � Several Anglo-Saxon sculptures at Shaftesbury hm estone 1s from the Upper Cypris Clays and Shales this (Shafi:esbury 2 and 3) are ofShaftesbury Sandstone. The at the top of the Peveril Point Member and more highly patinated sculptures at Batcombe are most specifically, possibly from the Green Marbl . two : likely Shaftesbury Sandstone, as is the large slab with cross at West Parley. UPPER GREENSAND FORMATION, SHAFTESBURY SANDSTONE MEMBER STONES OF UNCERTA IN PROVENANCE The Upper Greensand has an extensive outcrop across Devon, the roundel with a £gure with outstretched south-west England. It fo rms a prominent escarpment In h ds (Chulmleigh 1) is very heavily coated wit h orange from of the Chalk escarpment from the north-east of � pamtlplaster. There is a very small area on the south­ the area, just to the west of Wa ntage, to just west of west quadrant relatively free of coating which exposes a Dorchester in the centre of the district. In the south­ non-calcareous,fine-grained sandstone of unknown stone east, there is narrow, steeply dipping, partially faulted type. outcrop that extends from just south of Dorchester to Somerset, the worn sundialon the south side of the just north of Swanage. There are large outliers on the In tower ofSt Nicholas old church, Uphill, is a calcareous Blackdown Hills and the Haldon Hill respectivelyto the stone too lichen encrusted fo r identi£cation.The Barton north-east and south-west ofExeter. St David stone is a dark ydlowish brown (10YR 4/2), Over mu h fs u - stEngland (south ofWestbury), � � � � � weakly calcareous, fine-grained, finely laminated thefo rmaoon 1S dw1ded mtothree members, in ascending sandstone; the clasts are up to 0.2 mm across; the stone sequence: the Cann Sand consisting of unconsolidated type has not determined, but is probably of Palaeozoic glau nitic sa ds, a median Shaftesbury Sandstone �� � age. conslStm� �fvar�ably cemented,£ne-to medium-grained, glaucomt1c sandstones with the highest, bioclastic,

11 11- _jl I I_

24

N A

Reconstructed Roman Roads .,•\ . /

HeightAbove Sea Level (metre:o�) c::::J Omto SOm c::::J 50mto 100m c::::J lOOm to 150m D 150m to 300m - 300m to SOOm - SOOm and Above

FIGURE 16 To pography of south-wet England

11 11-