Feeding the Moat Excavations Near the Site of Edenbridge Manor House Brady

127 FEEDING THE MOAT: EXCAVATIONS NEAR THE SITE OF EDENBRIDGE MANOR HOUSE kate brady and edward biddulph Archaeological works undertaken by Oxford Archaeology (OA) between June and August 2004 and by Archaeology South East (ASE) in February and May 2004 along the route of the Edenbridge’s western relief road (Fig. 1) revealed evidence of a substantial watercourse infilled with material dating to the late twelfth to mid thirteenth century. This large man-made channel carried water to the moat surrounding Edenbridge manor house and provided dating evidence for the digging of the moat and the early use of the manor. Evidence of metalworking activity was also found, indicating that a smithy existed nearby in the twelfth to thirteenth centuries. The new road by-passes Edenbridge High Street (the London to Lewes Roman road) on a curving route that runs north to south from Station Road to Mont St Aignan Way (TQ 4427 4660 and TQ 431 4595). The site is situated approximately 200m north of the River Eden where it crosses the High Street (Fig. 2). The river was named after the settlement, called Eadhelmesbregge (Eadhelme’s Bridge), first documented as such in around 1100. A church in Edenbridge was documented in the Textus Roffensis in 1120, although the present church dates to the twelfth to thirteenth centuries (Irwin 1964). The Historic Environment Record (HER, formerly the Sites and Monuments Record), held at Kent County Council, documents the site of Stangrave (originally Edenbridge) manor house and moat, situated approximately 100m to the west of the relief road. The HER states that although none of the building remains, the western and southern arms of the moat can be seen as a slight depression in the lawns of the present manor house and are shown as water-filled on the 1870 OS map. The HER suggests that a stretch of water to the north of the present house may have formed the northern arm of the moat. Therefore, until now, the only remaining part of the moat to be identified was the eastern arm. The Kent and Sussex Weald was a major centre of iron production in the Roman and early post-medieval periods; this is well documented, and the archaeological evidence has been closely studied. Edenbridge is situated on the northern edge of the Weald Clay in the area of the central KATE BRADY AND EDWARD BIDDULPH 128 MAIDSTONESevenoaksROYALTUNBRIDGEWELLSRyeRiverArunRiverRotherLONDONRiverMedwayTenterdenChathamGravesendTilbury020 km(Site location)TheWealdRiverEdenMarlpit HillEdenbridgeFour ElmsTroyTownMoatMoatDevils DenBroxhamManorMoatSite LocationNFigure 1: Site locationEdenbridge70683371696734321:500000 2000 m Fig. 1. Site location. FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 129 High Weald where all of the medieval iron working sites identified by Straker (1931) and Cleere and Crossley (1995) were located. Of these dozen or so sites, five appear to have been slightly earlier than the rest, belonging to the twelfth to thirteenth century. These sites are clustered mainly between Crawley and Tonbridge. A mine was situated at East Grinstead, which was documented in 1263 and iron working at Crawley was referred to in court rolls of 1265-6 (Cleere and Crossley 1995). Many of the sites were situated near to Ashdown forest and thirteenth-century documentary sources record woodland management for fuel at this time (ibid.). Slightly later, sites such as Tudeley (Hodgkinson and Whittick 1998) have provided evidence that the industry had grown to a scale able to provide for substantial purchases by the crown in the periods between 1250 and 1370 (Cleere and Crossley 1995). In the early part of the fifteenth century, water-powered blast furnaces began to be used, heralding the substantial Wealden iron industry that survived until the nineteenth century (ibid.), when two forges were situated in Edenbridge. Both were to the south of the current archaeological site and shown on the OS map of 1870. The River Eden was crucial, too, for the functioning of a number of watermills that existed in the medieval town. Domesday records a mill belonging to the manors of Westerham and Edenbridge, and by 1291 there were three (Somers-Cocks and Boyson 1912, 268). One watermill existed on the site of an eighteenth-century mill presently occupying on the east side of the High Street close to the river. Medieval features The earliest medieval feature was identified during OA’s watching brief (Fig. 2). Cut 41/54, measured approximately 10m in width, and extended beyond the eastern and western limits of ASE evaluation trench 21, which was the southernmost excavated. The shape of the cut is not known, as only a small part of it was revealed in the trench. At a depth of 0.4m a step in the sides of the cut was encountered, and the cut then continued beyond this to at least 0.57m in depth, although it was not bottomed and may have been much deeper. Its upper fill was a layer of mottled yellow-blue clay, suggesting that water had previously been contained within the cut. The feature was not dated, but was stratigraphically earlier than the main twelfth-/thirteenth-century features on the site. Diverted water course The main feature identified was a large channel that extended for well over 100m n-s across the site and beyond the southern limit of investigation. It was identified by the ASE evaluation to the south of Lingfield Road KATE BRADY AND EDWARD BIDDULPH 130 Graphics Studio - Illustration Template for Archaeologia Cantiana publication LINGFIELD ROADB2026 STATION ROADSTANGROVE ROADKeyRelief road corridorPondLimit of excavation1:50000 200 mArchaeology South East trenchesOA excavation areaArea of watching briefFigure 2: Plan of the routeHIGHSTREETN222120161514131112109874321Location of Stangrave Manor House (HER)ExtantPond19 Fig. 2. Plan of the route. FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 131 where it measured 8m in width, and terminated in the southern part of the area excavated by OA, to the north of Lingfield Road. In the OA excavation, it was a feature made up of two related parts. The main body of the cut (143/136) was wide, with steep sides and a generally flat base, though was more irregular in the north (Fig. 3). It measured 3.5m in width and 0.65m in depth. At the base of this feature, at a slightly oblique angle to it, was a smaller channel (167/168), the eastern edge of which was lined with stone (Fig. 4, section 14). This channel measured 1.4m in width and 0.28m in depth with vertical sides and a concave base. Four sherds of late thirteenth-century pottery were recovered from the fill. To the eastern side of this channel were waterlain silts suggesting overflow of water from this channel into the main feature. The course and design of the main channel suggests that, together, they were used to divert water from the River Eden, which runs e-w approximately 200m to the south of the southernmost extent of the channel. The exact function of the small stone-lined channel at the base is not known, but it may have been a feeder gully to control the amount of water flowing into the main cut. Because the supposed junction of the channel and the river has not been investigated it is impossible to know how water flow into the channel was controlled, but it is possible that there was a sluice. The channel terminated within the excavated area of the site, and water may have been contained here in the form of a pond. An extant pond can still be seen approximately 50m to the north, which was almost certainly part of the eastern arm of the Edenbridge (later Stangrave) manor house moat. It presently measures approximately 30m in length and 10m in width. That the pond and the water features identified by the excavation are contemporaneous is suggested by the presence of nine sherds of a decorated twelfth-century bowl recovered from alluvial deposits encountered just to the east of the extant pond. The waterlogged state of the channel fills upon excavation indicated that the water table was reached near to the base of the cut. Other water channels Further features recorded on the site give a clue to the intricacies of the water management system. From the terminus of channel 136/143 extended a smaller channel (120/142) that was fairly shallow at 0.22m in depth at its junction with the larger channel, but deepened to 0.74m as it extended north. This deepening served to level the base of the cut to compensate for the gradient of the slope of the land from the river. It measured 1m in width. The cut, with waterlogged material within its fill, was probably another watercourse related to the function of the main channel. The channel extended beneath the trench edge, in the direction of the extant pond/moat to the north, and may have been used to transfer water from the diverted watercourse to the moat. The upper of the two KATE BRADY AND EDWARD BIDDULPH 132 Graphics Studio - Illustration Template for Archaeologia Cantiana publication 1:5000 20 m124126128Section16170Location of repostitioned ASE Trench 19Figure 3: Plan of excavation areaNExtent of subsoilSection 13Modernservices (pipes)Electricity cableStoneSection 12167144136165132134ModernintrusionModernSection 14143168142120118180182136Section 19Plot boundaryServices TrenchStone122KEYDisturbed ground184186 Fig. 3. Plan of the excavation area. FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 133 fills of this channel contained two sherds of twelfth-/ thirteenth-century pottery and small pieces of slag. Further south, in ASE trench 19, a similar gully/channel was identified. It joined the main channel from the west and measured 0.15m in depth. Its full width was not revealed within the trench. The fills of this feature suggest early episodes of waterborne silting, indicated by gleyey clays followed by more rapid backfill (probably deliberate) of material containing slag, stone, pottery (thirteenth-/early fourteenth-century), bone and burnt clay. Other features The remaining features identified during the investigations were not obviously linked to water management (Fig. 3). A shallow depression (132/134) was truncated by the eastern side of the main channel/pond (Fig. 4, section 12). It was not dated, and may have been a tree throw hole or similar natural feature. A V-shaped ditch (170) was situated in the northern part of the OA excavation area and was roughly perpendicular to gully/channel 120/142 and aligned ene-wsw. It measured 0.9m in width and 0.3m in depth and contained two fills, one of which was extremely rich in charcoal and slag (Fig. 4, section 16). Pottery recovered from this upper fill numbered 34 sherds dating to the late twelfth to early thirteenth century. The function of this ditch is not clear, but the large amount of charcoal and slag suggests that metalworking activity was taking place nearby in the late twelfth to early thirteenth century. Inter-cutting pits 180 and 182 were situated at the north eastern limit of excavation and were identified only in section (Fig. 4, section 19). Neither was dated, but the earlier (182) contained 3,410g of undiagnostic slag and a dumped smithing hearth bottom. Analysis of the slag suggests that it was consistent with a medieval date. This, combined with the significant amount of slag recovered from the fills of some of the other features on the site indicates that a small smithy was located nearby. This activity superseded the water management, as the waste from the metalworking industry was discarded in the channels. A few small pits or postholes were located adjacent to the main channel (Fig. 3; Fig. 4, section 13). One of these (128) was situated on the eastern edge of the smaller gully or channel, but as this eastern side of the gully had been truncated at this point by undated pit 126, its relationship with the gully is not known; it was probably a posthole. One sherd of twelfth-/thirteenth-century pottery recovered from the fill suggests that it was broadly contemporary with the channels. Another similar feature (165) was situated on the eastern side of the main channel. This was probably also a posthole. It did not contain any datable material. KATE BRADY AND EDWARD BIDDULPH 134 Graphics Studio - Illustration Template for Archaeologia Cantiana publication (portrait) (July 2001)Figure 4: Section 12, 13, 14, 16 and 19SN184133132135134136144145137138139141140137167123131125146127129130122124126147128148149150142151152153154156157158164159163160161162155143171172Slag170Sectioncollapse180181179178Topsoil192Stone136Stone192168168SubsoilSubsoilSubsoilTopsoil130182186Section 12Section 13Section 14Section 16Section 1940.75 m OD40.99m OD41.76 m OD41.39 m ODWEWEWENS1:1000 5 m Fig. 4. Sections 12, 13, 14, 16 and 19. FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 135 Post-medieval features A small number of features were dated to the post-medieval phase. A ditch and a subsequent recut were aligned ene-wsw across the excavated area. The earliest cut (186) was probably originally around 1.8m in width and measured 0.8m in depth (Fig. 3). The recut was slightly narrower and shallower, measuring 1.04m in width and 0.64m in depth. Some nineteenth-century pottery and glass was recovered from the fills. The recut probably represents the cleaning out and maintenance of a property boundary defined by the earlier ditch. This boundary is aligned with the properties fronting onto Lingfield Road and almost certainly belongs to one of these, or an earlier property on the same alignment. A few unphased features were not dated by artefacts and were not related stratigraphically to those that were. In trench 21, south of Lingfield Road, gully 34 terminated to the east of the main channel. The only find recovered from the fill was a piece of burnt flint. In the far north of the evaluation area, trench 9 revealed a wide shallow depression. No finds were recovered from the fill. A small shallow oval-shaped feature (118) was situated a little distance to the east of gully/channel 120/142. It was not dated and its function is not known. Discussion Although the evidence is patchy, the presence at Edenbridge of a large, man-made channel filled by the mid-thirteenth century is intriguing. The feature appears to be part of a water management system. The identification of this channel in both the ASE evaluation trenches south of Lingfield Road and its terminus in the OA excavation to the north suggests that it was used to divert water from the River Eden, possibly to fill the moat around Edenbridge manor house. Alternatively, the channel formed part of a mill race, although the location of the watermill serving Edenbridge Manor is unknown. The possibility in any case seems less likely, since the water channel extended some distance from the river; a location closer to the river, like that of the known medieval watermill east of the High Street, would have offered greater economy and efficiency. If the channel did indeed feed the moat, then water may have entered the larger upper part of the cut via the smaller stone-lined channel at the base suggested by the waterlain deposits either side of it. This would have made for easier control of the water flow. The southern extent of the large cut is not known, or whether it extended the whole 200m or so to the river. It may be that a smaller channel carried the water from the river to this large feature. The extant pond to the west of the relief road was almost certainly part of the moat of Stangrave manor house, known in the mid-thirteenth KATE BRADY AND EDWARD BIDDULPH 136 century as Edenbridge Manor. At this time, land value was increasing due to the transformation of woodland into farmland. As a result some former dens, which were connected to manors in north Kent, became manors in their own right; the den of Westerham became Edenbridge manor (Irwin 1964). A royal grant from Henry III to John de Camvill records the existence of the manor house in 1263 (Somers-Cocks and Boyson 1912, 33), and a mid-thirteenth-century chronology is supported by the pottery recovered from the archaeological investigations, which dates the infilling of the channels from c.1220 and no later than c.1250. In 1293 the Stangrave family took possession of the manor and it became known as Stangrave Manor House (ibid.). This documentary evidence, combined with the archaeological results, suggests that the channel represents a feeder for the moat or a pond related to the original moat construction, as it is probable that the moat would have been maintained well into and beyond the Stangrave occupation of the manor. Further, the extant pond (the probable eastern arm of the moat) to the west of the relief road indicates that the moat itself was not backfilled deliberately but naturally silted up over time. The OS map shows that the western and southern arms of the moat were water-filled as recently as 1870. The thirteenth to fifteenth centuries were a time when many moats were being constructed around manor houses as an expression of status, and in imitation of higher-status castle sites. This expression of status was obviously very important, as diverting water from the river approximately 200m away would have been a labour intensive process, but was clearly justified for a manor house situated in the centre of the town. Other medieval moated sites just outside Edenbridge such as Broxham and Devil’s Den would have been constructed by digging a simple loop from the nearby streamlet and around the house (Fig. 1). Unfortunately, a defined date for these other moated sites is not known and therefore an inter-site chronology cannot be established. The ceramic assemblage, consisting of mainly jars and cooking pots, indicates domestic activity in the area, and these broken pots were most likely discarded from nearby houses with other domestic waste, and used to backfill the water channels; several medieval houses still exist in Edenbridge, fronting the High Street to the west of the site (cf. Irwin 1964). The backfill of the water channels also indicated that ironworking took place in the vicinity in the late twelfth to early thirteenth century. The significant amount of tap slag and small amount of charcoal recovered, along with a smithing hearth bottom, is important evidence for ironworking in the Weald in this period. At this time the ironworking industry of the Weald was developing again, after an apparent hiatus following the well established Roman industry. Mines and bloomeries around Ashdown Forest were producing ore and blooms in the twelfth and thirteenth centuries (Cleere and Crossley 1995) that were probably used FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 137 in village smithies at a small scale, mainly serving the local population. It is highly likely that a village such as Edenbridge would have had such a smithy, which would have benefited from a location by the main London to Lewes road, to the east of the site. Here, the smith would be able to exploit passing trade, serving travellers perhaps needing horseshoes and other objects. A roadside smithy of this kind has been identified at Godmanchester in Cambridgeshire (Webster and Cherry 1975, 259-60). The small amount of tap slag and charcoal suggests that blooms made elsewhere were reheated on site prior to forging. More centralised production and wider distribution at sites such as Tudeley (Hodgkinson and Whittick 1998) did not take place until the fourteenth century. There is no evidence that the water channels discovered on the site were used in the manufacture of blooms or iron artefacts. The archaeological investigation of the route of the Edenbridge Western Relief road has revealed valuable evidence of water management probably related to moat construction and has by implication provided previously lacking dating evidence for the establishment of the moated site. The evidence places the construction of the moat in the period preceding the late thirteenth century and more specifically to between 1220 and 1250, when the property was still named Edenbridge Manor. The results also add valuable information about medieval iron working in the High Weald during the early stages of an industry that is still little understood. acknowledgements Oxford Archaeology would like to thank Ringway Highway Services for funding the project. The authors would like to thank Archaeology South East for allowing extensive use of their evaluation report, written by Neil Griffin. Thanks are also extended to Bryan Matthews (OA field supervisor), Alan Ford, Rebecca Peacock, James Vessey (site archaeologists) and to Leigh Allen, Paul Blinkhorn, John Cotter, Emma-Jayne Evans, Lynne Keys, Ruth Shaffrey, Laila Sikking and Fay Worley for their specialist contributions. The assistance of Neil Guiden of English Heritage for assistance with research on moated manor houses, and Adam Single, archaeological officer at Kent County Council, is gratefully acknowledged. This report was edited by Edward Biddulph, who also managed the post-excavation project. Illustrations were prepared by Julia Moxham. Full stratigraphic descriptions and artefactual reports are available in the site archive currently held at Oxford Archaeology (accession code: EDWRR 04); the archive will in due course be deposited with Eden Valley Museum, Edenbridge (accession code EDWRR04). KATE BRADY AND EDWARD BIDDULPH 138 APPENDIX 1 pottery POTTERY CATALOGUE by John Cotter The assemblage recovered by OA comprises a total of 70 sherds of pottery weighing 619g. Apart from three small sherds of nineteenth-century date all the pottery dates to the later twelfth to thirteenth centuries and consists of types common to Surrey (particularly the Limpsfield kilns) and north-west Kent. A basic catalogue of the pottery has been compiled (details in archive). For each context and fabric the total sherd count and weight were recorded, followed by the context spot-date which is the date-bracket during which the latest pottery types in the context are estimated to have been produced or were in general circulation. Comments on the presence of recognisable vessel forms and decoration and the like were also recorded. Fabric codes used here are those of the Canterbury Archaeological Trust (Cotter 2001 and forthcoming; Jarrett 2005). None of the pottery, which is generally quite fragmentary, has been illustrated although published parallels are cited where appropriate. The pottery was initially assessed by Paul Blinkhorn, some of whose comments are incorporated below. Pottery fabrics and forms are as follows: EM2 Early medieval shelly ware c.1050-1250 (14 sherds, 30g). A relatively sand-free fabric tempered with coarse shell. Sherds possibly from a single jar/cooking pot. Associated in same context (171) with a roof tile fragment and thus likely to post-date c.1150. M44A Coarse Limpsfield-type grey ware c.1150-1300 (29 sherds, 334g). A coarse fabric with an abrasive feel. Rounded iron-tinted, clear and milky quartz up to 2mm across. Sparse rounded flint to 4mm and sparse grey sandstone. The commonest ware on the site. Rims from a minimum of three fairly large jars/cooking pots (diameters 230, 260 and 280mm) and a thick body sherd possibly from a steep shouldered jug. All handmade, possibly with rims finished-off on a turntable. The rims present are of simple squared or thickened flat-topped form similar to examples from nearby Tonbridge (Jarrett 2005, fig. 8.13, fig. 9.16-17). Several sherds show external sooting. M44C Coarse Limpsfield-type shelly-sandy ware c.1150-1300 (2 sherds, 14g). Superficially similar to north or west Kent shelly-sandy ware (EM36) but with coarser quartz inclusions like M44A. Shell often dissolved-out. Sherds from a single vessel. M38A North or west Kent sandy ware c.1150-1400 (12 sherds, 122g). Generally a reduced grey ware related to and possibly including Limpsfield products. Production at sites in Kent also seems very FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 139 likely. Sherds from at least two vessels represented including a sagging base from a jar/cooking pot and a broad recessed base probably from a jug possibly imitating recessed jug bases in London-type ware (Pearce et al. 1985, fig. 20.39, fig. 27.62 passim). M38B North or west Kent fine-medium sandy ware c.1225-1400 (4 sherds, 89g). A finer variant of M38A often used for jugs but not exclusively. Sherds from a minimum of two vessels represented both probably jugs of rounded or squat form. These include a sagging base and the lower wall of a jug with traces of a horizontal combed band of decoration at the girth. Combed decoration such as this is typical of the north-west Kent grey wares. M49 Earlswood-type ware c.1150-1250 (6 sherds, 20g). Produced at Earlswood, Surrey. Few definite examples of Earlswood-type glazed jugs have been identified from Kent although this appears to be one such example. Confusion with Ashford/Wealden-type oxidised wares (M40A, B, C, LM32) is possible because of the use of similar-firing Wealden clays. The six small joining sherds come from the shoulder area of a jug of probable early rounded form. The fabric has a pale orange-buff colour with abundant well-sorted rounded quartz mostly under 0.5mm with occasional coarser grains to 1mm. The quartz grains have an orange-pink tint. The matrix is smooth with moderate fine red iron oxide or iron-rich clay pellets up to 2mm, sparse white clay pellets and sparse very coarse grey clay pellets up to 5mm. Abundant very fine mica barely visible with the naked eye. The exterior of the sherd is covered with a white slip through which decoration has been combed sgraffito-style and the whole then covered with a pitted clear glaze showing dark brown in the combed areas. This sgraffito technique of decoration is paralleled at the Earlswood kiln (Turner 1975, figs. 4 and 5) although the exact scheme on the present sherd is not closely matched. The decorative scheme here seems to comprise a single horizontal band of combing on the shoulder of the jug and below this probably a combed lattice or trellis formed of intersecting diagonal bands of combing on the body. This scheme of decoration is paralleled in white slip (but not sgraffito) decoration at Earlswood (ibid., fig. 4.9) and in sgraffito on twelfth-century early rounded jugs in London-type ware (Pearce et al. 1985, fig. 18.29). LPM14 Staffordshire-type refined white earthenwares c.1825-1900+ (3 sherds, 10g). Blue transfer-printed wares etc. The assemblage comprises at least ten vessels of medieval date. The fabrics and the typology of the vessel forms present suggests a late KATE BRADY AND EDWARD BIDDULPH 140 twelfth- to mid thirteenth-century dating. Most of the vessels appear to have been handmade. Products of the Surrey Limpsfield area kilns (Jones 1997), located only 5 miles north-west of Edenbridge, appear to predominate although other unlocated sources in north-west Kent may also be represented. The absence of north or west Kent shelly-sandy ware (Fabric EM36) is surprising in a group of this date as the ware is fairly common at Tonbridge about 10 miles to the east (Jarrett 2005). Jars/cooking pots are the main form represented plus a few jugs of early character including a highly decorated jug from Earlswood – a more westerly Surrey source. The absence of Surrey white wares (c.1230+) may also have chronological implications. APPENDIX 2 metalworking METALWORKING SLAG by Lynne Keys A small assemblage (almost 6.4kg) of iron slag was recovered by ASE and OA. The slag was very fragmentary; some had obviously been broken during excavation and post-excavation handling, and so had to be described as undiagnostic since it could not be assigned to either smelting or smithing. Some pieces were, however, still intact or could be pieced together and these had been produced by smithing activity. No slag diagnostic of smelting was present. The assemblage appears to represent a short period of smithing activity somewhere nearby in the medieval period. The slag was then thrown into the nearest cut features or found its way there in later backfilling. bibliography Cleere, H. and Crossley, D., 1995, The Wealden Iron Industry, 2 edn. Cotter, J.P., 2001, ‘The Pottery’, in Hicks, M. and Hicks, A. (eds), St Gregory’s Priory, Northgate, Canterbury Excavations 1988-1991, The Archaeology of Canterbury New Series II, 231-266. Cotter, J., 2006, ‘The Pottery’, in Parfitt, K., Corke, B. and Cotter, J., Excavations at Townwall Street, Dover, 1996, CAT Occasional Papers (HMP). Griffin, N., 2004, ‘An Archaeological Evaluation on the Route of Edenbridge Western Relief Road’, ASE evaluation report. Hodgkinson, J.S. and Whittick, C.H.C., 1998, ‘The Tudeley ironworks accounts’, Wealden Iron, 18, 7-38. Irwin, J., 1964, Place-names of Edenbridge, Edenbridge and District Historical Society. Jarrett, C., 2005, ‘Pottery Analysis’, in Wragg, E., Jarrett C. and Haslam J., ‘The Development of Medieval Tonbridge Reviewed in the Light of Recent Excavations at Lyons, East Street’, Archaeologia Cantiana, cxxv, 141-9. FEEDING THE MOAT: EXCAVATIONS NEAR EDENBRIDGE MANOR HOUSE 141 Jones, P., 1997, ‘The Pottery’, in Hayman, G., ‘The excavation of two medieval pottery kiln sites and two sections through the London-Lewes Roman road at Clacket Lane, near Titsey, 1992’, Surrey Archaeological Collections, 84, 32-76. OA, 2004, ‘Edenbridge Western Relief Road, Post-excavation Assessment and Updated Project Design’. Pearce, J.E., Vince, A.G. and Jenner, M.A., 1985, ‘A Dated type-series of London Medieval pottery, Part 2: London-type ware’, Trans London Middx Archaeol Soc., Special Paper 6. Somers-Cocks, H.L. and Boyson, V.F., 1912, Edenbridge, Edenbridge Chronicle Office, Edenbridge. Straker, E., 1931, Wealden Iron, London. Turner, D.J., 1975, ‘Medieval pottery kiln at Bushfield Shaw, Earlswood: interim report’, Surrey Archaeological Collections, 70, 47-55. Webster, L.E. and Cherry, J., 1975, ‘Medieval Britain in 1974’, Medieval Archaeology, 19, 259-60.