Hard.
Black.
The main inclusions are fossiliferous limestone fragments, c.0.01mm to 2.0mm. Less common are inclusions of rounded quartz, c.0.1mm to 0.7mm and scattered fragments of oolitic limestone, angular chert, up to c.0.6mm across, fine and coarse-grained sandstones, including ferruginous fragments with white mica and quartz inclusions, flint and rounded pellets of silty clay with a high iron content.
Anisotropic.
Bath; M356, M360, M363.
Unknown. The petrology is very similar to that of Glos. TF41A (Gloucester Late Saxon ware) which is tempered with a Severn Valley sand derived from the Worcestershire basin and the Cotswold scarp. A similar source would not fit the distribution of the ware as it is now known.
Handmade globular vessels with sagging bases. The rims are all everted but vary from small types in which the rims curve out from the body to larger types with a sharp neck angle. The first type varies from 100mm to 220mm diameter and the second type from 140mm to 440mm diameter.
Forms very similar to those of the larger cooking pots but with rectangular or rod-sectioned handles with stamped decoration.
One sherd from Bath, Citizen House, may be from a straight-sided spouted bowl (No.255).
Found at Bath, Citizen House in the earliest medieval levels, associated with Cheddar fabric E. Therefore, likely to have a 10th century origin. Some of the forms, the larger cooking pots , the spouted pitchers and bowl should be 11th or 12th century, and the ware is common in 11th and 12th century pits at that site. By the early 13th century the ware accounted for a minute proportion of the pottery found. Also found at Saltford, Barrow Mead and (possibly) Chewton Keynsham.
Published description in Vince, (1979 b). Illustrated examples in Greene (1979) and Cunliffe (1979).
Hard.
Dark grey or black (10YR 4/0) with oxidized dark brown to yellowish red surfaces (7.5YR 4/2 to 5YR 4/6). Refires to red (2.5YR 4/6).
Well rounded and sorted inclusions of quartz and quartzite (some with possible mica inclusions), fine-grained limestone with angular quartz and brown amorphous inclusions, poorly sorted sandstone with some mica and a brown-stained cement and chert (subangular with brown inclusions except for one fragment of black angular chert with numerous pores filled with quartz). Single fragments of mudstone (a broken rounded grain with quartz and brown inclusions) and fine-grained sandstone with silica cement and brown veins were seen in thin-section. The distinction between fabrics A and B is one of grain size, A being coarser. There is no evidence for any difference in date, distribution or typology between the two fabrics, which are therefore treated as one ware here.
finely divided limestone, quartz and a little white mica except for one sample from Chepstow with a coarser quartz and white mica up to 0.1mm long.
Bristol; M292, M293, M294, M417. Chepstow; M846-8, M850. Dublin; M367-8.
A few crinoid fragments were seen, suggesting that the sand tempering was partially derived from Carboniferous rocks. The roundness of the inclusions and the variety of rock types found suggests a river or beach sand. Similar sands (without carboniferous rock fragments) were found at Severn Beach (Thin-sections M672, M673). The distribution evidence suggests a source close to Bristol.
All handmade, squat vessels with sagging bases. The rims show evidence for rotary finishing. There is little variety in rim forms, which are either vertical and rounded or everted stubby rounded, angular or sharply everted (rare). The Chepstow examples have rim diameters ranging from 150mm to 320mm with a mean of 230mm.
Vessel forms similar to those of the cooking pots, stubby everted rims and taller everted rims. Grooved wavy lines on the shoulder and tubular spouts. Handle form unknown
Single shell types with a flat pedestal base.
Found below the Castle Mound at Bristol associated with 3 phases of timber buildings. Therefore probably mid-late 11th century. Earlier evidence in Bristol is elusive, in keeping with the historical evidence for the main growth of Bristol in the early 11th century. Found in early-mid-11th century contexts at Dublin and in post-conquest contexts in Chepstow. Although unseen by the author, it it likely from the published description that some of the pottery from Castle Tower, Penmaen, is of this fabric or Bristol A/B (Talbot, 1966). Possibly in use throughout 12th century at Chepstow but in reduced quantities. The spouted pitchers have been found at Bristol and Dublin and the lamps only at Bristol.
Ponsford (1974) illustrates the Bristol Castle material.
Soft to hard.
Dark grey (7.5YR 4/0) with oxidized light yellowish brown surfaces (10YR 6/4). Refires red (2.5YR 4/6)
Angular fragments of limestone and sandstone. The limestone fragments contain both crystalline and brown-stained, fine-grained calcite. Several fragments are brown-stained around the edges and in veins. One fragment contains pellets of calcite in a brown-stained, fine-grained matrix and another contains muddy limestone intergrown with silica and cut by a quartz vein (?). Red iron replaced crinoid fragments were visible in some samples under the binocular microscope. Rounded quartz up to c.0.5mm was present together with clay pellets with the same texture as the clay matrix up to c.0,6mm across. The sandstone contains interlocking grains of quartz, mica and amorphous brown material. A few voids are present in the sandstone fragments, surrounded by brown staining. A distinct variant of this fabric contains little or no limestone and a higher proportion of rounded quartz. Firing and typology are identical with the limestone-tempered ware.
A high quantity of white mica and some angular quartz up to c.0.1mm
Chepstow; M842-4, M853. Bristol; M298, M419, M420. Dublin; M370, M375.
The limestone is matched closely by Carboniferous limestone, which however outcrops to the west of Bristol, around Chepstow as well as further afield in the Forest of Dean and the Mendip Hills. A source around Bristol is most likely. The petrology suggests that the clay and tempering might be derived from the weathering of carboniferous limestone with little transportation of the material by water.
All handmade, squat vessels with sagging bases. There is little variety in rim forms, which are either everted with external thickening (rarely having a triangular section), everted with thickening inside and out or vertical. Most of the vertical rims have an internal groove and external thickening. In some cases this becomes a distinct bead which is sometimes thumbed.
The same form as the cooking pots with 'wheel' or 'grid' stamps on the shoulder, sometimes combined with incised wavy, zig-zag or cross-hatched lines. Free-standing tubular spouts on the shoulder of the vessel, rod-handles, sometimes stamped.
One example from Bristol, stamped on the top of the socket with a 'grid' stamp.
Examples from Bristol, rims only (so unclassifiable).
One vessel with a 'wheel' stamp from Gloucester.
Found at Bristol, Chepstow, Dublin and rarely at Gloucester and Bath ( the latter two sites not examined in thin-section). Cooking pots were found at all sites but spouted pitchers only at Chepstow (one vessel) and Bristol. The only stratigraphic information available to the author is from Chepstow, where the cooking pots had a similar occurence to Bristol A and B vessels, ie. late 11th century and 12th century but probably residual by the early 13th century. A similar fabric is present at Cheddar (fabric C) but given the lack of Bristol A/B vessels at Cheddar it is likely that Cheddar C ware is tempered with Mendip limestone and Bristol C ware with Clifton limestone.
Vince (forthcoming h) describes the petrology of the ware, as does Vince (forthcoming e). Both reports include illustrations of the cooking pots, whilst the spouted pitchers and bowl are illustrated by Watts (forthcoming b). The lid? is illustrated in Vince (forthcoming f).
Hard.
Either completely oxidized pale yellow (2.5YR 8/4) or oxidized pink (7.5YR 7 to 8/4) with a light grey core (7.5YR 7/0). Refires yellow (10YR 8/6) to reddish yellow (5YR 7/6 to 7.5YR 7/6).
The quantity of inclusions varies considerably from vessel to vessel but the following inclusion types can usually be seen if enough of the broken sherd edges are examined. Rounded quartz and quartzite, mainly less than c.0.3mm but up to 1.2mm, rounded light-coloured clay pellets up to 1.0mm, rarer sandstone with a dark brown matrix, silicious sandstone (quartzite with brown inclusions), fragments of fine-grained sandstone with a silica matrix (grains up to c.0.2mm) up to 0.7mm. Small iron ore fragments c.0.2mm across were found, surrounded by a halo of brown staining. Rounded, decomposed limestone up to c.0.3mm and possible fragments of wood (present only in reduced areas of the fabric) are much rarer. Two roof tile wasters from Bristol Redcliffe were examined in thin-section. One has a temper similar to that of the hollow wares but much more abundant and with more limestone (T-S154) whilst the other contained numerous large clay pellets and sandstone fragments. Some of these inclusions were very coarse, the sandstones sometimes reaching 3.0mm across and the clay pellets 10.0mm. Little rounded quartz or quartzite is present and the clay matrix contains abundant angular quartz up to 0.04mm across. This variant is not found at comsumer sites.
Remarkably pure, contains a few white mica fragments and sparse fine angular quartz.
Bristol Redcliffe wasters; M153, M155, M156, M159, Chepstow M162, M827-9. Caerleon; M207. Grosmont; M215, M256. White Castle; M240. Llantwit Major; M250. Laugharne (Dyfed); M268. Hereford; M570.
Bristol Redcliffe wasters; 152, 154. Woodspring Priory; 160. Cardiff; 254. Chepstow; 758-760.
Bristol ware is a typical coal-measure light-firing fabric and shares many characteristics with post-medieval Staffordshire light-firing wares and other medieval light-firing wares, eg. fine angular quartz in a relatively clean matrix, light-coloured clay pellets, quartz, quartzite and sandstone inclusions and some iron ore or iron staining. All of the sectioned samples had these characteristics, but so do samples of wares which on typological or distribution evidence could not be Bristol Products. Thus, Ham Green ware is reliably characterised but Bristol ware is not. Both Ham Green and Bristol wares utilise a coal-measure clay. Neither site is situated on the coal-measures, although outcrops are not far away, to the north and north-east of Bristol, the clay must therefore have been transported to the kiln sites.
All wheelthrown vessels varying in size, shape and decoration. Most are large jugs similar to Ham Green B jugs, either with flat rims with a cordon or moulding just below the rim or plain flat topped rims (measured diameters between 100mm and 140mm). Distinctive types have applied heads around the rim or an applied thumbed strip. Both sagging, frilled bases and plain flat bases are found with diameters between 140mm and 160mm. All examples have strap handles, which are either plain or decorated with knife slashes. A distinctive type found at Bristol St. Peters has diagonal slashing between two vertical slashed lines. Bridge spouts and pulled spouts are found. A distinctive type of bridge spout is moulded into a face shape and decorated with applied red-firing and plain clay. Decoration varies considerably. Some vessels are plain but applied clay strips, of plain or red-firing clay are common. These can form curvilinear patterns, often horse-shoes, or ocassionally figurative scenes. The face-jugs are sometimes made more anthropomorphic by the addition of hands and elements of dress. Turned grooves are found, usually not combined with applied decoration (unlike Ham Green ware). Square-toothed roller-stamping is found but is rare. Clear external lead glaze is the norm but copper-flecked glazes are found as are internally glazed jugs.
Wheelthrown globular vessels with flat-topped rims, sagging bases and a patchy internal glaze are found but are not common.
Small internally glazed, flat based vessels.
Some sherds from Chepstow were thought to be from a bowl with curving walls and an internal glaze.
A vessel from Bristol in the form of a knight on horseback.
A possible inkwell from St. Peters, Bristol is part of a double vessel, decorated with stabbing and incised lines.
Bristol ware occurs in mid-late 13th century contexts and later at Bristol,Chepstow and Gloucester. The industry survived into the 15th century but by the 16th century was extinct (since Bristol was supplied solely by Malvern Chase). A sequence of jug forms has been obtained by M. Ponsford which enables two groups to be distinguished 'early' and 'late' Bristol jugs. The early jugs (late 13th to early 14th century) are often highly decorated, have thumbed bases and have a cordon or moulding below the rim. Bridge spouts, including the face-spouts, are early. Late jugs (later 14th to 15th century) are often plain with flat topped rims, sometimes with applied thumbed strips below the rim and have flat bases. At Chepstow and Gloucester both early and late jugs are found and there is unlikely to be any major change in distribution between the two periods.
Dawson et al. (1972) describe a collection of wasters from the foundations of St. Peters church, Bristol. It is now thought that these were imported to the site as make-up and could come from the Redcliffe potteries. They date from the later 14th century. The Redcliffe site itself is not yet published. Bristol vessels published by Barton (1959) and the face jugs are published by Ponsford (1979).
N.B. Not fully described.
Soft to hard.
Large fragments of white sandstone
Unknown, but somewhere in South Avon on distribution evidence.
Handmade, squat vessels with curving walls, sagging bases and everted rims.
Glazed, handmade vessels.
Dated by Rahtz to the 11th to 12th centuries. An eleventh century date would be unusual for the tripod pitchers and indeed in the absence of relative stratigraphy on the Chew Valley Lake sites this ware could be of late 12th to 13th century date, contemporary with the Ham Green and Minety wares found. A compromise 12th century date is most likely.
Rahtz and Greenfield (1977) illustrate the vessels but do not define them as a ware, nor give useful fabric descriptions.
Hard to very hard.
Oxidized dusky red (2.5YR 3/2) to dark grayish brown (10YR 4/2). Glaze mottled dusky red to reddish black (10R 3/2 to 10r 2/1).
Moderate to sparse inclusions of rounded quartz and quartzite up to 1.0mm and rounded iron ore up to 0.4mm.
Sparse specks of angular quartz up to c.0.06mm.
Chepstow; M957. Gloucester; M317-322a. Hereford; M605-6. Newbury; M1157. Worcester; M698, M737-740.
The quartz and quartzite sand is similar to that found in the Severn Valley terrace sands and in pottery tempered with this sand (for example Worcester-type ware, Glos. TF43 and Droitwich and Canynges-type floor tiles). The ulitmate origin of this sand is the Triassic of the West Midlands. The texture of the clay matrix is more similar to lias clay than to Keuper marl or alluvial clay samples from the Severn Valley. Thin-sections of samples from Worcester, Hereford, Gloucester and Chepstow share the same characteristics and could all be from the same source. Waste from a kiln producing this type of ware has been found at Falfield, Avon and it is suggested by Bennett that the kilns may have been at Tortworth. However, the Falfield ware has not yet been examined petrologically.
Two or three handled cups are the only products known in this fabric. Three distinct forms are recognised:
All forms have a thick glaze inside and out, sometimes omiting the base, which is invariably of foot-ring type and wire-drawn. Some vessels, of the globular forms, have white clay pads around the girth.
Found first in the early 16th century. The Falfield waste is dated by Bennett to the mid-16th century by comparison with Bristol. At Chepstow one vessel was found in a pre-dissolution context (site XI phase 4) and at Gloucester examples were found in the construction levels of the Horsepool at the Eastgate, dated by documents to pre-1540. At Hereford this ware is most common in the late 16th century and may be replaced by Herefordshire cups in the 17th century. At Gloucester however the ware is certainly present in a group of the 1st quarter of the 17th century and is regularly associated with early to mid-17th century clay pipes.
The Falfield kiln waste is published by Bennett (Fowler and Bennett,1974, 123). Examples from Hereford and Gloucester are to be published in Vince (forthcoming a and f). The Worcester vessels are published by Morris (1980) together with a petrological report by the author (Vince, 1978b).
Three wares are included under this heading; 'proto-Ham Green ware', Ham Green cooking pot ware and Ham Green ware (jugs).
Hard.
Black to grey (5YR 2/1 to 10YR 4/1) sometimes with an oxidized yellowish red skin (5YR 4/6). Refires dark red (2.5YR 3/6).
Abundant medium to coarse sand, mainly up to 0.7mm across but with sandstone fragments up to 1.0mm across. Subangular and a little rounded quartz, often cloudy with brown veins, some plagioclase felspar, fragments of coarse-grained sandstone with overgrown quartz or quartzite grains (up to 0.7mm across), brown chert, often varying in colour within one fragment and crossed with quartz veins. Silicious sandstone with brown inclusions, opaque iron ore (smaller than the other inclusions, up to 0.3mm), and dark brown inclusionless clay pellets.
Porous, with sparse angular quartz.
Chepstow; M838-841, M877-8.
Unknown. In appearance and typology there is a strong similarity to Ham Green cooking pot ware but the fabric differs in three ways. Firstly, in grain-size and sorting (Ham Green vessels are better sorted and have a finer sand), secondly the grains are more angular and thirdly there is less limestone. Ponsford has suggested a source at Long Ashton on the basis of possible wasters found there. Petrology suggests that the tempering may be derived from a medium to coarse sandstone.
handmade, squat vessels with sagging bases, curving walls and everted or vertical rims, often with an external thickening. Some vessels have thumbing along the top of the rim. Horizontal or wavy combing is sometimes found on the shoulders and the exterior of the rim. Rim diameters at Chepstow vary from 120mm to 290mm with a mean of 230mm.
One tubular spout was found at Chepstow, probably from a spouted pitcher but possibly from a socketed bowl.
Two sherds from Chepstow may come from shallow bowls. Both have wavy combing on the exterior and one has stabbed combing on the top of the rim.
Found at Chepstow in 12th century contexts and the most common ware there at that time (between 45% and 83% of all sherds) but may have been superceded by Chepstow fabric HA in the early 13th century, although still found in contexts of that date. Found at Bristol. Similar vessels found all along the south Welsh coast but have not been petrologically examined, for example Ware A at Castle Tower, Penmaen (Talbot, 1966).
Talbot (1966), Vince (forthcoming e) describes the ware at Chepstow.
Hard.
Oxidized yellowish red (5YR 4/6) sometimes with a dark grey core (5YR 4/1). Refires dark red (2.5YR 3/6).
Abundant inclusions of subangular and rounded quartz and sparse sandstone fragments (poorly sorted grains of quartz in a brown-stained silica cement), chert (some brown-stained), fine-grained micaceous sandstone (with a silica cement) and limestone. The sand is well sorted and rarely larger than 0.3mm across, except for the rare sandstone fragments up to 0.5mm.
Sparse angular quartz and white mica up to 0.1mm. A curious feature of this fabric is the presence of lenses of vitrified clay, always reduced light grey, highly vesicular and optically isotropic. Each lens contains a void at its centre, probably where some inclusion which has acted as a fluxing agent existed.
Chepstow; M836-7, M845, M875-6.
The Ham Green kiln site. Many vessels from settlement sites have a thin glaze, but this is probably an ash glaze rather than accidental lead glaze. A comparison with the jug fabric shows that different clay bodies were used but the same sand tempering.
Handmade, squat vessels with curving walls, sagging bases and everted or rolled out rims. Horizontal or wavy coming is often found on the shoulder or exterior of the rim and vertical thumbed strips are found on the body. Rim diameters at Chepstow vary from 180mm to 250mm with a mean of 220mm.
At Chepstow Ham Green cooking pots do not occur until the arrival of the glazed wares. In the earliest contexts they are as common as the jugs (1% and 4%) but in later contexts are outnumbered (5% against 15%, 1% against 16%, and 1% against 6%). This may be a reflection of output or possibly a reflection of the marketing strategy. The overall distribution of Ham Green cooking pots is much more restricted than that of the jugs. In any quantity the vessels are limited to the Bristol Avon valley as far inland as Keynsham and sites along the South Welsh coast, such as Chepstow.
Barton (1963) illustrates the range of vessels found on the Ham Green kiln site, including some decorated with roller-stamping, and therefore contemporary with the type A jugs.
Hard.
Reduced light grey (7.5YR 5/0 to 7/0) usually with an oxidized light yellowish brown , very pale brown or sometimes pink interior surface (10YR 6/4 to 7/4 & 7.5YR 7/4). Refires pink to reddish yellow (7.4YR 7/4 to 5YR 6/6 to 6/8). One sample has a thin light grey slip on the interior which refires red (2.5YR 5/6).
Varying proportions of a well-sorted, predominantly quartz and limestone sand and angular to rounded clay pellets. This is presumably the basis of a division into A and B fabrics by Barton (1963,96-7), A fabrics having high proportions of clay pellets and limestone and B fabrics having high proportions of quartz. In practice there appear to be more indeterminate sherds than ones which fit into these two groups. Few of the sand inclusions are larger than 0.3mm but the clay pellets are often several millimetres across. Subangular and rounded quartz, angular to rounded clay pellets (the same colour as the clay matrix but sometimes denser and almost opaque in thin-section, some may be indurated), rounded limestone (often decomposed), rare rounded iron ore and colourless chert (or possibly flint or silicious sandstone).
very sparse angular quartz and white mica up to 0.1mm across.
Aberystwyth; M277-8. Barry Island; M236. Box; M77. Caerleon; M206. Chepstow; M162, M830-2, M834, M893. Dublin; M371, M374. Gloucester; M78-80, M82-3, M174, M329. Hereford; M569. Kidwelly; M274. Rumney Castle; M1132. St. Justinians; M203.
Ham Green, near Pill. Comparison with the cooking pot fabric shows that the quartz and limestone sand is common to both but that the light-coloured clay and clay pellets are found only in the jug fabric. A similar clay is used for the Bristol jugs and is probably a coal-measure clay from the north or north-east of Bristol.
Four sherds of possible tripod pitcher were found at Chepstow (site XI, phase 2) ; one is part of a cylindrical necked vessel with a flat topped rim 150mm diameter and decorated with roller-stamping. Two further sherds come from a wide sagging base with a patchy external glaze. Other sherds of the same type are ocassionally found, for example a combed vessel from Lydney with a rod-handle, a rod handle from Gloucester and a tripod foot from Bristol. These are however extremely rare.
The majority of vessels fit into Bartons scheme of 'A' and 'B' styles. 'A' style jugs usually have wide sagging bases without a thumbed frill and wide, rolled out rims. The necks are often rilled and the vessel can be decorated with combing or diamond roller-stamping. Handles are rectangular and sometimes stabbed with a circular tool.
'B' style jugs are taller and have narrower bases, decorated with a thumbed frill. This is sometimes an added feature. The rims are flat topped with a moulding below the rim. Bridge spouts are the norm and rectangular handles, either stabbed or slashed. The vessels are usually decorated, firstly with rotary applied combing or grooving, which can be combined with vertical combed or grooved lines. Over this in some cases applied decoration is found. This is usually figurative and is best represented by a complete jug from St. Peters, Bristol which is decorated with a hunting scene of a man with bow and arrow, a deer and a dog. An almost identical deer was found at the kiln site.
At Chepstow the four sherds of possible tripod pitcher described above were found in the construction levels of the priory buildings on site XI and formed 1% of the assemblage. 'A' style jugs were found alone in the earliest occupation levels (phase 3a) and formed 4% of the pottery found and subsequent occupation in the Priory buildings contained both A and B style vessels, as did the construction debris at site 1 (the Priory Barn) and later 13th century pits at site 2 (15%, 16%, and 6% respectively). This is the only stratigraphic evidence available to the author of a chronological sequence of styles in Ham Green ware. At Gloucester, Ham Green A and B jugs are contemporary. There is also no evidence that the highly decorated jugs are later than the simpler, combed and grooved jugs, although such a sequence is likely. The dating of Ham Green ware is difficult but in chapter 6 it is suggested that the tripod pitcher phase is late 12th century, the A style only phase is late 12th to early 13th century and the A and B style phase is mid-to-late 13th century. Ponsford claims that Ham Green ware is no longer found in Bristol by c.1300 but that the kilns site may have still been distributing wares coastally after that. The distribution of Ham Green jugs is extensive but in quantitative terms the ware is unimportant outside of the Severn Valley, the Bristol Avon valley, the Wye valley as far north as Monmouth and the South Welsh coast and South-East Ireland. All of these areas are directly accessable by boat. Outlying finds spots can however be well inland, for example Avebury and Highworth in east and north-east Wiltshire respectively.
Barton (1963 and 1967a) describes the discovery of the kiln site and gives a very thorough description of the vessels found there.
Not fully described.
hard.
Includes coarse quartz and limestone fragments.
Not sectioned.
Unknown, but must be in North Avon on distribution evidence.
(Not illustrated)
Squat, handhmade vessels with sagging bases and vertical or everted rims, wich have a slight groove on the inside.
(Distribution not plotted) Found at Hillesley, a Manor House complex, possibly fortified in the 11th or 12th centuries. The ware is not stratified (?) and is dated to the 11th or 12th centuries by typological comparison with Bristol A/B, and C wares (B. Williams, pers. comm.).
Ponsford and Williams (forthcoming).
Soft to hard.
Oxidized with a light grey core.
Sparse red iron ore or clay pellets and limestone fragments up to 1.0mm across. The limestone is heat-altered.
Abundant fine angular quartz and some microcline and plagioclase felspar and rare white mica up to 0.1mm across.
Kiln waste; M1072-83.
Wanstrow, Near Frome.
Both slipwares and kitchen wares (bowls and jars) were found in fieldwork at the kiln site but the only definite wasters seen were of kitchen ware. It is from these that the thin-section samples were taken.
probably 16th and 17th centuries. The distribution is unknown since the ware is visually identical in fabric to South Somerset wares. Thus 'South Somerset ware' found in the Bristol Avon valley, for example at Bath and Newton St. Loe might be Wanstrow ware, carried overland from the kiln site, South Somerset ware carried by ship to Bristol and redistributed from there or South Somerset ware carried overland from the kiln site up the Fosse way.