The Medieval ‘Upmarket' Ward of Dover Archaeological Evidence from Laureston Place Castle Hill

THE MEDIEVAL ‘UPMARKET’ WARD OF DOVER: ARCHAEOLOGICAL EVIDENCE FROM LAURESTON PLACE, CASTLE HILL

keith parfitt

In April 2001 members of Canterbury Archaeological Trust maintained a watching-brief during the construction of an extension to St Mary’s Primary School, off Laureston Place, Dover.¹ Reflecting its hill-side location, much of the site had previously been terraced to allow construction of the late twentieth-century school building and its playground but the area examined had been little affected by these earlier earthmoving operations.

The modern school is situated on the north-eastern side of the Dour valley, on the lower slopes of Castle Hill, directly below Dover Castle some 250m from the outer bailey wall (Fig. 1). This area lies within the historic parish of St James, about 150m north-west of the now ruined Norman parish church, close to the boundary with the former parish of Charlton. The site falls within a triangular block of ground bounded by Laureston Place on the east, Ashen Tree Lane on the south-west (downhill) side and Harold Passage on the north-west side. Through a series of historic maps, the overall outline of this block of ground, bounded by these same thoroughfares and sub-divided into three principal plots, can be traced back largely unchanged to the early seventeenth century. Throughout this time the region seems to have been mostly open land, mainly used as gardens.

The investigated area lay on the south-eastern side of the existing school building at NGR TR 3220 4170, some 20m to the south-west of Laureston Place, between the 20 and 17m od contours. Here, a rectangular area of former lawn and garden, with maximum dimensions of 15 by 11m was cleared down to the surface of the natural chalk by the builders. During the course of this work undisturbed soil deposits and seven features of archaeological interest were exposed and rapidly recorded (Fig. 2). The available dating evidence indicates that several of the features belong to the medieval period, implying the presence of a previously unknown occupation site in the area. This seems to have been located well beyond the principal habitation areas of medieval Dover.

Based on some reasonable foundation, historical tradition asserts that the hillside between Laureston Place and Ashen Tree Lane was once known as ‘Upmarket’ and a document of 1304 lists this as one of the wards of Dover (see below).

Two factors may have been significant in the location of any market in this area: reasonable access from both the town and castle and a not too steep area of hillside upon which to erect market stalls and booths. Although the present area is situated well outside the limits of the medieval town, the routeway represented by present-day Laureston Place was formerly of some considerable importance as it constituted the main medieval road leading up from the town towards the castle and the Downland beyond (Fig. 1). It was subsequently replaced by a new military road (Castle Hill Road) further east in 1797;² until this time Laureston Place would have formed part of an important thoroughfare, regularly used by traffic going to and from the castle.

The Medieval History of Upmarket Ward, Dover by Sheila Sweetinburgh

Here, it is said, was the site of an out-of-town market at the foot of Castle Hill used by the inhabitants of the castle and St James’s parish as an alternative to the town-centre market, adjacent to the church of St Martin le Grand.³ Tradition also claims that the open ground in this vicinity was one of the areas where the French Dauphin’s forces were encamped during the great siege of Dover Castle in 1216. Though within the liberty of the town and port of Dover, this was a marginal area to the north-east of the town centre, occupying a position between the town and the castle. As such, the site may have been especially important during the development of Dover Castle in the thirteenth century when large numbers of masons, carpenters and others from the building trades were drawn to the town, a situation which would be repeated in Dover in the sixteenth century when the harbour was rebuilt and extended. This itinerant, seasonal workforce presumably settled close to the castle, on the margins of the town, where it was more convenient to seek provisions from the surrounding countryside rather than exclusively from Dover itself. Whether this would have resulted in a formal market is unclear, but it seems likely the civic authorities would have sought to control any marketing in the area, both to provide the town with income from tolls and to regulate the supply of basic foodstuffs to the townspeople. It was not unusual for a town to have several, often specialist but occasionally ‘common’ or general, markets, a situation found at Dover in the later Middle Ages, which suggests it would be reasonable to assume there could have been a market here during the thirteenth century. This would seem the likely derivation of the name Upmarket, the name which was also used to describe the local ward.

As part of the liberty the inhabitants of Upmarket were required to help the town provide twenty ships for Ship Service to the crown, a requirement recorded in the Domesday Survey of 1086.⁴ At some point in the Middle Ages it was decided to subdivide the liberty into twenty wards as a way of allocating responsibility for the town’s ships, and one of the wards was called Upmarket warde (most of the ward names remained in use throughout the Middle Ages and into the sixteenth century).⁵ This administrative sub-division of the liberty appears to have taken place in the early thirteenth century, and some of the wards were apparently named after particular local inhabitants, like ‘Manekynes warde’ and ‘Wolues warde’ (possibly Manekyn of Dover and Joseph’s father, Wolf).⁶ The first documentary reference to these wards may be a charter witnessed by William de Albrincis, constable of Dover Castle, concerning a mill called ‘Handsexisbregge’ in ‘Manekynes warde’, which though undated must be 1226.⁷ Interestingly, the other charters of Henry III’s reign used street names, places or features like mills to locate the property under discussion; for example, a charter of 1271 referred to a place called ‘atte Boor’, ‘le Brok’ and the king’s highway.⁸ By Edward I’s time, however, the wards were recorded more frequently, like the references to Biggin ward in a Dover Priory charter dated 1279, and a civic charter dated 1286; and Wolves ward in a St Bartholomew’s hospital charter of 1295.⁹

The first reference to ‘Upmarkatte warde’ occurs in a charter of 1304 recorded in the Dover Priory register.¹⁰ The transaction, between Thomas de Ackolte and John Wolnoth of Dover, lists four tenements in Upmarket ward and the holders of the adjoining properties. From this it appears two of the tenements were next to each other and all four may have been alongside the same road. The size of the rents, two properties each at 6d. per annum and two at 32d., would seem to imply a range of properties in the area, the more expensive ones possibly providing substantial timber-framed houses. As one of the tenants of the 32d. per annum properties, the Lord of Chilham’s presence might indicate a buoyant property market, especially in terms of the rented sector, because he was presumably sub-renting the property at a profit. Over-crowding may also have been a factor in Upmarket. The recording of shared tenements by tenants who were the heirs of particular townsmen seems to suggest this, though it is possible they too were sub-renting. Yet even as tenant-landlords the returns per person would have been meagre because of the need to divide the proceeds among all the heirs. Together, such factors suggest a strong demand for housing on the urban fringe in the late thirteenth and early fourteenth centuries, which is not surprising when considered against the national backdrop of rising population, land shortage and expanding urban development. However, in Dover urban expansion at Upmarket in particular may have been even more significant in the previous hundred years as a consequence of the extensive castle-building programme.

Yet, like any marginal urban area the domestic/craft properties were principally alongside the main thoroughfares, whereas the agricultural-type buildings were sited in the gaps and possibly off the side lanes. Upmarket fits this pattern because in addition to the tenements there were barns and pieces of land, probably used for food production. Interestingly, Dover Priory held two hemp barns in the area, they leased both in 1359, which suggests hemp was grown locally for use in the port’s ship-building and repairing industries, and also possibly in the production of linen and canvas.¹¹ There is considerable evidence for peasant production of hemp in the Romney Marsh during the early fourteenth century and it seems likely the crop was grown elsewhere in the Kent coastal strip wherever conditions were suitable (A. Butcher, private correspondence).

Upmarket’s location on the fringes of Dover may explain the apparent lack of civic holdings in the area. Similarly the ecclesiastical institutions were very minor landlords; neither St Bartholomew’s nor St Mary’s hospitals held property there, Dover Priory had the two barns and a few other holdings, some with buildings, the parson at the castle church held a piece of land called le Ferthyng and St James’s church had ‘one land’.¹² It is possible the crown was a major land holder during the Norman and Angevin periods, but it may have been colonized by the local townsmen by the thirteenth century, who held the land as freemen or barons of Dover. Although using surnames does have certain problems, men like Robert de Boclande and Luke de Tylmanstone were presumably from Buckland and Tilmanstone respectively, while Walter de Denne and John de Mari were known Dover townsmen, which suggests that many of the Upmarket land holders were from Dover and its hinterland.¹³ Upmarket in the High Middle Ages, therefore, presented an opportunity for migrants and more established citizens, either as a place to reside, or at times as a means of generating income through rents. Such opportunities probably declined considerably after the mid-fourteenth century as a consequence of high mortality, socio-economic problems and the need for high civic expenditure to combat the threat of foreign invasion. Even though it is difficult to find direct evidence of the impact of these factors on areas like Upmarket, the fact that Dover Priory was employing long leases for its property there would suggest they were trying to minimise the problems of falling demand in the local land market. For example, the barns were each leased for 60 years, and the pieces of land for 59 and 99 years respectively.¹⁴

For the property holders there in the early sixteenth century, however, the area had certain advantages compared to the wards to the south which were suffering from encroachment by the sea – ‘lands’ had been lost in Ballys, Oxes, Wolfys and Derman wards.¹⁵ Yet it seems likely that like the other eastern wards the inhabitants may have suffered from the increasing focus of the town’s trading community on the new market area near St Martin’s church and the new harbour to the west.

As a result, trade/travellers bypassed Upmarket and its southern neighbours as the main axis of the town was from Biggin Gate and St Mary’s ward in the north to Snargate and the Piers in the south. The late medieval testamentary evidence may substantiate this view, though as a group the Dover testators apparently recorded far less property details compared to their fellow portsmen from Sandwich and Hythe. Of the property said to be in Upmarket, three men listed gardens and William Foche also had a barn.¹⁶ Presumably for these men their gardens were an investment, sources of rent while they resided near the town centre, or in William Foche’s case in St James’s parish (he was a fisherman); and assets which could be passed on to the next generation. Unfortunately, the early sixteenth-century ward lists rarely record more than the tax payer so it is impossible to know the balance of property-types in Upmarket ward, though it might be surmised that between the thirteenth and early sixteenth centuries the population density had fallen due to a shift from a relatively packed urban environment to a more rural situation.¹⁷

The Recorded Remains

In the light of the historical information outlined above, the medieval dis-coveries made during the present watching-brief are of particular interest, being located close to the main road to the castle at a point almost half way between St James’s Church and the Constable’s Gate of the castle (Fig. 1).

Observation of the area cleared by the builders revealed a series of archaeological features (Fig. 2) consisting of a ditch [F. 1], four circular pits [Fs 3, 5, 11 and 29] and two shallow terraces [Fs 7 and 9]. These were all cut into the natural chalk and were sealed by a deposit of hillwash, under a thick layer of recent garden soil. The hillwash layer consisted of a cream-grey clay loam with chalk grits and produced a few sherds of medieval and early post-medieval pottery, together with two sections of clay pipe stem, several fragments of peg-tile, part of a Caen stone cresset lamp (see below, SF 31), a heavily abraded piece of Roman tegula and a prehistoric struck flint. From the latest datable finds, it would seem that the hillwash was accumulating between c.1675 and 1750 and perhaps later. Its presence implies that significant soil disturbance was occurring further up-slope, presumably through cultivation but just conceivably connected with re-modelling of the outer defences of the castle.

Boundary Ditch, F. 1

This was located running across the central part of the site and was traced for a minimum distance of 14.45m. At the north-west end it was cut away by the existing school structure and at the south-east end had been removed by a later pit [F. 11] which produced a small quantity of medieval material. The ditch itself failed to produce any finds in the 5.30m length excavated and must remain undated, although its relationship to F. 11 indicates that it cannot be later than the medieval period. Most probably it served as a boundary marker between different properties. Where undamaged at the south-east, it was 1.10m wide and 0.55m deep with sloping sides and a dished base. The filling consisted of a series of orange clays, with varying amounts of weathered chalk and some flints. Occasional flecks of charcoal occurred throughout these deposits but there were no other finds.

The Pits, Fs 3, 5, 11 and 29

Pits Fs 3 and 5 were located on the north-eastern side of the cleared area, some 0.50m apart. Although they had been truncated by between 0.50m and 0.80m during the initial site machining, as surviving each was more than 1.50m in depth.

Pit, F. 3 lay very close to the north-eastern edge of Ditch, F. 1. It was circular in plan with a diameter of 1.25m. Only the upper 0.50m of the surviving filling could be excavated archaeologically but subsequent machine excavation by the builders indicated an overall original depth of 1.90m. The excavated filling consisted of a light brown clay loam with much chalk rubble, some flints and charcoal specks. This produced nine pot-sherds dated to the period c.1175-1200/25, together with fifteen pieces of animal bone and a burnt oyster shell.

Pit, F. 5 lay immediately to the north-east of F. 3 and slightly less than half was exposed within the cleared area. From the portion revealed it would seem to have been larger than F. 3, with a diameter of about 1.90 m. Again, only the upper 0.50m of the surviving filling could be excavated but probing revealed that it continued for at least another 1.20m below this, indicating an overall minimum depth of 2.50m from the top of the original chalk surface. The filling of the pit consisted of a series of brown and grey-brown clay loams, with chalk and flint lumps and carbon specks. These yielded a combined total of twenty-eight pot-sherds, mostly datable to the period c.1175-1225, but also included a residual sherd of Roman samian ware. There was also a quantity of animal bone and marine shell (limpet, winkle, mussel, whelk, oyster), fragments of burnt daub, metal-working waste, an iron key and several nails. Wet sieving of a bulk sample yielded fragments of slag, flake and spheroidal hammerscale, a small fish assemblage including bones of herring and shad, eggshell and some well-preserved charred cereal grains (analysis by Enid Allison).

Pit, F. 11 was located on the south-eastern side of the cleared area, cutting through the earlier boundary ditch [F. 1]. Only the northern half could be partially excavated; the southern half lay outside the development limits. The pit was probably circular or oval in shape and was at least 2.20m across. It was about 0.50m deep, with very steep sides and a flat base. The filling consisted of a cream-brown clay loam with much chalk, flint and greensand rubble. Although no mortar was observed, the general impression gained was that much of this material was building debris – a number of the chalk lumps had been roughly squared. Two medieval pot-sherds were also recovered from the filling of the pit and these may be dated to the period c.1225-1350. There was also part of a Caen stone cresset lamp (see below, SF 29) and a rough circular rubber of greensand (see below, SF 30).

Pit, F. 29 was located during trenching immediately to the west of F. 11. It was circular in shape with a diameter of 1.45m. It was 1.30m deep with vertical sides and a flat base. The filling consisted of a light brown clay loam with moderate amounts of chalk rubble, some flints, occasional greensand lumps and carbon specks. A total of twelve pot-sherds dated to the period c.1175-1200/25 was recovered, together with three pieces of animal bone, a few oyster, limpet and mussel shells and a derived prehistoric struck flint.

The Terraces, Fs 7 and 9

Two shallow, parallel terrace cuts [Fs 7 and 9] were located on the down-hill, south-western, side of the ditch [F. 1], running on a slightly different axis. Their upper edges were some 2.60m apart. The upper terrace cut [F. 7] was traced for a distance of 9.40m. It was 0.23m deep and was filled by a light orange-brown clay loam with chalk, flint and charcoal specks [8], devoid of finds.

The lower terrace cut [F. 9] was traced for a distance of 7.45m. It was 0.05m deep and was again filled with a light orange-brown clay loam with chalk, flint and charcoal specks [10]. This produced a single, unworked prehistoric struck flint.

the finds

Medieval Pottery (not illustrated) by John Cotter

A total of fifty-five medieval pot-sherds was recovered from the site. Most of these came from the filling of the pits. The material consists of readily recognisable fabrics and forms that can be paralleled in the assemblages recovered from the more extensively excavated medieval occupation site off Townwall Street, some 300m to the south.¹⁸

Medieval Iron Key by Ian Riddler

A complete iron key was recovered from the filling of pit F. 5, alongside a small assemblage of ceramics dating to the period between c.1175 and 1225 (Fig. 3). The size of the key (length: 109mm) suggests that it was used on a door lock, or possibly with a chest. The wards of the key bit are cut broadly in a stepped pattern. The circular shape of the loop-bow and the type of bit allow the key to be assigned to Ward Perkins type II, which was current between the late eleventh and the thirteenth century.¹⁹ A late twelfth- or early thirteenth-century date for the key would therefore be entirely appropriate. A series of keys of medieval date have come from excavations in Canterbury, but most of those are of later types.²⁰ Keys of this date are also absent from excavations elsewhere within Dover, as at Townwall Street, for example.

Blacksmithing Waste by Keith Parfitt

A quantity of iron slag representing blacksmithing waste was recovered from the filling of pit, F. 5. A soil sample from this feature yielded both flake and spheroidal hammerscale providing further evidence for medieval blacksmithing in the immediate area.

Cresset Lamps (not illustrated) by Ian Riddler

Two fragments of Caen stone stem from separate examples of cresset lamps were recovered:

1) A fragment of a cresset lamp, made from Caen stone. The oil res-ervoir is square in shape, with an inner lip on the upper edge. The corners below the rolled upper lip and the lower edge are chamfered. Traces of a residue can be seen within the reservoir area, extending to roughly half the height of the inner surround. Height: 85mm. From filling of medieval pit, F. 11 (SF 29; Context 12).

2) A small fragment of Caen stone with a flat base and a curved circumference, extending originally to c.60mm. The middle section of the stone is burnt and has a pink colouration. From post-medieval hillwash layer sealing the excavated features (SF 31; Context 14).

Lamp 1 is readily identifiable, with a square or rectangular reservoir that still retains a stain from its contents. The stain extends to roughly half of the level of the inner wall of the lamp. The lamp was probably square in shape originally (or possibly rectangular), with chamfered corners and lower edge. Lamp 2 is merely a small fragment of Caen stone with a flat base and curved stem, and its precise shape can no longer be identified.

Biddle has distinguished three types of stone lamp: ²¹

A block lamps, either rectangular or cylindrical;

B pedestal lamps;

C stands or sticks.

Lamps 1 and 2 fit into the first category. The depth of the reservoir on the square or rectangular example (1) distinguishes it from lamp bases, which are often square, but have only a shallow receptacle.²² The squared shape of the reservoir is unusual, but the exterior of the lamp is chamfered, thereby providing an octagonal section. An elaborate cresset lamp from Winchester has an octagonal stem above a square base, and a second example, closer perhaps to the Dover lamp, is octagonal throughout.²³ It is similar to examples from Canterbury, London and Norwich.²⁴ A limestone object thought not to be a lamp, with a rectangular reservoir, came from Westbury.²⁵ Square lamps are known also from Cheddar.²⁶ The upper rounded moulding of the Dover lamp projects beyond the lower surface on the surviving corner and this would have allowed the lamp to be seated in an iron surround.

The majority of cresset lamps recovered to date from east Kent are circular in form. They include three examples from Townwall Street, Dover, each of which is of a different type of plain form common in the late eleventh and twelfth centuries, as well as several examples from Canterbury.²⁷ Multiple cressets, which have come from St Augustine’s Abbey and near Burgate Street in Canterbury, may have been largely confined to ecclesiastical establishments, as Evans suggests.²⁸

Smoothing Stone (not illustrated) by Ian Riddler

A section of Lower Greensand, almost certainly derived from a quern, probably of Roman date, which has been trimmed to a roughly oval shape. The lower face is flat and smooth and may have represented the original grinding face of the quern, worn smooth. The edges are rounded and the top surface is coarsely finished, with irregular chisel marks. Length: 110mm; width: 102mm; height: 35mm; weight: 610g. From filling of pit, F. 11 (SF 30: Context 12).

Lower Greensand querns were popular during the Roman period and were widely dispersed throughout east Kent. A centre for their production has been identified at Folkestone.²⁹ The shape and size of this stone allow it to be placed, however, in a group of medieval objects, other examples of which are known from Beverley, Cheddar, Exeter, Launceston Castle, Westbury, Whithorn and Winchester. It has been suggested that sections of basalt lava from Canterbury and Southampton were also re-used in this manner.³⁰ These stones are too large and coarse to have served as slick-stones used to smooth textiles, although other stone objects, mostly of an earlier date, have been identified for this purpose.³¹ They all share a single, flat, smoothed surface and a coarse texture to the remainder of the stone, and they fit well in the hand. Nicholson noted the association of the Whithorn stones with metalworking debris and suggested that they could have been used as abrasives, for polishing and for descaling iron objects during forging.³²

dating and discussion

The features excavated clearly form part of a larger archaeological com-plex, the limits of which were not determined. The primary ditch running across the area investigated [F. 1], probably marked a boundary but remains undated. It is earlier than pit [F. 11], which was perhaps infilled during the late thirteenth or early fourteenth century but this need not preclude a medieval date for the ditch itself. Also undated are the two shallow terrace-cuts recorded [Fs 7 and 9]. These are most likely to be roughly contemporary but since they follow a different axis to the ditch, it seems probable that they are not associated with it.

More definite dating evidence for the four pits recorded is available, as each produced sherds of medieval pottery. On the evidence of their similar form and substantial depth, pits Fs 3, 5 and 29 are likely to be roughly contemporary and the available pottery dating suggests that they were filled during the late twelfth or early thirteenth century. Their occurrence on either side of the boundary ditch, F. 1, shows that the area used for pit digging was not delimited by this feature and most probably the ditch is completely unrelated. The broad, shallow form of pit F. 11 implies that it had a different function to the other pits and the probable building rubble recovered from it suggests that it might be some sort of demolition pit. The available pottery dating also indicates that its filling is slightly later than the other pits.

The discovery of evidence for medieval activity in this little known part of Dover is of some considerable interest, given the historical background (see above). Located well outside the principal medieval occupation areas – the castle, the town west of the Dour and the eastern suburbs around St James’s Church – any settlement here clearly stood somewhat detached (Fig. 1). Nevertheless, its location by the side of what must have been the busy medieval road up to the castle would have meant that it was by no means isolated.

The features assigned to the medieval period and the finds they produced seem more consistent with permanent habitation in this region, rather than any temporary market place, as might have been anticipated. Some evidence for the former existence of buildings in the area is perhaps provided by the (un-mortared) squared chalk blocks and Greensand lumps found in pit, F. 11, and by a few burnt daub fragments, including one bearing wattle marks, found in pit, F. 5. The pits also seem to have been used for dumping (limited) amounts of domestic rubbish, in the form of pottery, animal bone, fish bone and marine shell, together with some blacksmithing waste and discarded iron-work. The documentary evidence certainly indicates the presence of a number of tenements in Upmarket Ward during the early fourteenth century (see above). Ribbon development along the principal route-way between the castle and the town seems highly likely. From the date-range of the pottery in the pits it may be suggested that occupation in the area began during the late twelfth century and continued up to the mid-fourteenth. Most of the excavated material falls within the period c.1175-1225, which closely corresponds with the main construction period of Dover Castle.

Once abandoned, the present area was sealed by a thick layer of hillwash and seems to have remained largely unoccupied throughout the late medieval and most of the post-medieval period. It seems likely that further evidence for medieval occupation, including traces of buildings, still remains to be discovered in this part of Dover.

acknowledgements

Thanks are due first and foremost to the building contractors, W.W. Martin Ltd of Thanet, who financed the entire operation. In particular, the assistance of Mr M.J. Loveday and Mr P. Hill is gratefully recorded. Dr Sheila Sweetinburgh kindly prepared an historical survey of medieval Upmarket. Mr Barry Corke assisted the writer on site and prepared the plans for publication. Miss Beverley Leader drew the iron key.

endnotes

¹ Parfitt, K., 2001, ‘Report on an Archaeological Watching-brief at St Mary’s Primary School, Dover, 2001’, CAT archive report.

² Bavington Jones, J., 1907, Dover: a Perambulation of the Town, Port and Fortress, Dover, 258.

³ Ibid., 256; Welby, D., 1976, The Tidy Ruin; History of the Parish Church of St James the Apostle Dover, Kent (Dover Archaeological Group), 73.

⁴ Morgan, P. (ed.), 1983, Domesday Book, Kent, Chichester, 1 a.

⁵ Statham, S., 1899, The History of the Castle, Town and Port of Dover, London, 64-5.

⁶ Lambeth Palace Library [LPL]: MS 241, fols 79v, 80v.

⁷ LPL: MS 241, fol. 80v.

⁸ LPL: MS 241, fol. 60v.

9 LPL: MS 241, fol. 66; Statham, S. (ed.), 1902, Dover Charters and Other Documents in the Possession of the Corporation of Dover, London, 29; Bodleian: Rawlinson MS B 335, fol. 25v-26.

¹⁰ LPL: MS 241, fol. 87.

11 LPL: MS 241, fols 84v, 85.

12 LPL: MS 241, fols 84v-87; British Library [BL]: Add. MS 29810, fol. 5v; BL: Egerton MS 2093, fol. 15.

13 LPL: MS 241, fols 84v, 87.

14 LPL: MS 241, fols 84v-85v.

15 BL: Egerton MS 2093, fols 12-14.

¹⁶ CKS: PRC 32/2/614 Thomas Tok, 32/3/74 William Foche, 32/7/86 Henry at Wod.

¹⁷ Bavington Jones, J., 1920, Records of Dover, Dover, 186; BL: Egerton MS 2093, fol. 15.

¹⁸ Parfitt, K., Corke, B. and Cotter, J., 2006 Townwall Street, Dover, Excavations 1996, Archaeology of Canterbury, New Series vol. III.

¹⁹ Ward-Perkins, J.B., 1940, Medieval Catalogue, London Museum Catalogues 7, London, 136.

20 Riddler, I.D., 2001, ‘The Small Finds’, in M. Hicks and A. Hicks, St. Gregory’s Priory, Northgate, Canterbury. Excavations 1988-1991, Archaeology of Canterbury, New Series vol. II, 276.

²¹ Biddle, M., 1990, Object and Economy in Medieval Winchester, Winchester Studies 7ii, Oxford, p. 985.

22 Knight, J., 1972, ‘A 12th century stone lamp from Llangwm Uchaf, Monmouthshire’, Medieval Archaeology, 16, 132; Pritchard, F., 1991, ‘Small Finds’, in A.G. Vince (ed.), Finds and Environmental Evidence, Aspects of Saxo-Norman London: II, London, 161 and fig. 3.44, 149; Mynard, D.C., Zeepvat, R.J. and Williams, R.J., 1992, Excavations at Great Linford, 1974-80 (Bucks. Arch. Soc. Monograph 3, Aylesbury, fig. 74, 179; Blockley, K., Blockley, M., Blockley, P., Frere, S. S. and. Stow, S., 1995, Excavations in the Marlowe Car Park and Surrounding Areas, The Archaeology of Canterbury 5, Whitstable, 1218, nos. 1437-8; Walsh, C., 1997, Archaeological Excavations at Patrick, Nicholas and Winetavern Streets, Dublin, Dublin, fig. 72.4.

23 Cunliffe, B., 1964, Winchester Excavations 1949-1960, Winchester, fig. 51, 4-5.

24 Frere, S.S. and Stow, S., 1983, Excavations in the St. Georges Street and Burgate Street Areas, The Archaeology of Canterbury 7, Maidstone, 183 and fig. 72.7; Ward-Perkins 1940, 175 and fig. 54.4; Ayers, B., 1987, Excavations at St. Martin-at-Palace Plain, Norwich, 1981, East Anglian Archaeology, 37, Gressenhall, 73, fig. 62.5.

25 Ivens, R., Busby, P. and Shepherd, N., 1995, Tattenhoe and Westbury. Two Deserted Medieval Settlements in Milton Keynes, Bucks. Arch. Soc. Monograph, Series 8, Aylesbury, 389 and fig. 178).

²⁶ Rahtz, P.A., 1979, The Saxon and Medieval Palaces at Cheddar, BAR British Series 65, Oxford, fig. 78.

²⁷ Riddler, I.D., 2006, ‘Cresset lamps’, in Parfitt, Corke and Cotter, 315; Blockley et al. 1995, 1218 and fig. 539, 1436.

²⁸ Sherlock, D. and Woods, H., 1988, St Augustine’s Abbey: Report on Excavations, 1960-78, KAS Monograph 4, Maidstone, 199, no. 80 and fig. 61.80; Frere and Stow 1983, 186 and fig. 72.8; Evans, J., 1969, ‘A Discovery of two unusual objects in New Shoreham’, Sussex Archaeological Collections, 107, 79-86, 83.

²⁹ Keller, P.T., 1989, ‘Quern Production at Folkestone, South-East Kent: an Interim Note’, Britannia, 20, 193-200.

³⁰ Foreman, M., 1991, ‘The Objects of Stone and Fired Clay’, in P. Armstrong, D. Tomlinson and D.H. Evans (eds), Excavations at Lurk Lane, Beverley, 1979-82, Sheffield Excavation Reports 1, Sheffield, fig. 90.61; Rahtz 1979, fig. 78.69; Allan, J.P., 1984, Medieval and Post-Medieval Finds from Exeter, 1971-1980, Exeter Archaeological Report 3, Exeter, 300 no. S33; Riddler, I.D., 2006, ‘Objects of Stone, Bone and Antler’, in A.D. Saunders (ed.), Excavations at Launceston Castle, Cornwall (Society for Medieval Archaeology Mono. 24; Ivens et al. 1995, 389 no. 1758; Nicholson, A., 1997, ‘The Stone Artefacts’, in P. Hill (ed.), Whithorn and St. Ninian. The Excavation of a Monastic Town, 1984-91, 447-464 (Stroud), 459; Biddle 1990, 466, no. 1006; Blockley et al. 1995, 1218; Platt, C. and Coleman-Smith, R., 1975, Excavations in medieval Southampton, 1953-1969. Volume 2: The Finds, Leicester, fig. 270.2232.

31 Walton Rogers, P., 1997, Textile Production at 16-22 Coppergate, The Archaeology of York. The Small Finds 17/11, London, 1779 and fig. 827.

³² Nicholson 1997, 459.

Fig. 1 Map showing location of site, in relation to the medieval town and castle.

Fig. 2 Overall site plan showing excavated features.

Fig. 3 Iron key from the filling of medieval pit F. 5: scale two-thirds.

(Drawn by Beverley Leader.)