St Peters House, Risbygate Street Street BSE 367

St Peters House, Risbygate Street Street BSE 367 Archaeological Excavation Report SCCAS Report No. 2014/037 Client: County Care Homes Author: Andrew Tester April 2014 Suffolk County Council Archaeological Service

St Peters House Archaeological Excavation Report SCCAS Report No. 2012/180 Author: Andrew Tester Contributions By: Andy Fawcett Sue Anderson, Julie Curl Richenda Goffin, Anna West and David Gill Illustrator: Ellie Cook and Gemma Adams Editor: Richenda Goffin Report Date: April 2014

HER Information Site Code: BSE 367 Site Name: St Peters House, Out Risbygate Street Report Number SCCAS Report no. 2014/036 Planning Application No: SE/11/0188 Date of Fieldwork: 2011 Grid Reference: TL 842 657 Oasis Reference: Curatorial Officer: Project Officer: Client/Funding Body: suffolkc1-137734 Abby Antrobus Andrew Tester County Care Homes Digital report submitted to Archaeological Data Service: http://ads.ahds.ac.uk/catalogue/library/greylit Disclaimer Any opinions expressed in this report about the need for further archaeological work are those of the Field Projects Team alone. Ultimately the need for further work will be determined by the Local Planning Authority and its Archaeological Advisors when a planning application is registered. Suffolk County Council s archaeological contracting services cannot accept responsibility for inconvenience caused to the clients should the Planning Authority take a different view to that expressed in the report. Prepared By: Andrew Tester Date: April 2014 Approved By: Position: Date: Signed:

Contents Summary Drawing Conventions 1. Introduction 4 2. The Excavation 9 2.1 Site location geology and topography 9 2.2 Archaeological and historical background 9 The location plan (Fig.1) includes selected entries from the Historic Environment Record: 9 3. Methodology 10 4. Results 10 4.1 Descriptions 10 Introduction 10 Phase 1 Medieval to 17 th century 11 Phase 2 Post 1549 to 17th century 17 5. The finds evidence 21 5.1 Introduction 21 5.2 Pottery 21 Introduction 21 Methodology 21 Pottery by period 22 Pottery of the 11th-14th centuries 23 Late and post-medieval 24 Modern 26 Unidentified 26

Pottery by context 27 Discussion 28 5.3 Ceramic building material 28 Introduction 28 Methodology 29 The assemblage 29 CBM by context 34 Discussion 34 5.4 Fired clay 35 5.5 Plaster/mortar 36 5.6 Clay tobacco pipe 36 5.7 Worked flint 36 5.8 Burnt flint/stone 37 5.9 Worked stone 37 5.11 Iron nails 37 5.10 Glass 38 5.12 Slag 38 5.13 Small finds 38 6. The environmental evidence 40 6.1 Faunal remains 40 Methodology 40 The assemblage provenance and preservation 40 Pit 0269 42 Conclusions 52 6.2 Shell 53 6.3 Conclusions 53 7. General Discussion 55

8. Recommendations for Further work 58 9. Archive deposition 58 10. Acknowledgements 59 List of Figures Figure 1. Location plan, showing excavation area(red) evaluation trenches(blue) features (black) and HER entries mentioned in text. 5 Figure 2. 1880 OS Map with excavation area outlined in red. 6 Figure 3. Excavation plan 7 Figure 4. Phase plan 12th to 17th centuries 8 Figure 5. Phase plan 17th to 19th centuries including details of chimney foundations 15 Figure 6. Selected sections 16 List of Tables Table 1. Finds quantities 21 Table 2. Pottery quantification by fabric. 22 Table 3. Pottery by feature. 27 Table 4. CBM quantities by form 29 Table 5. Roofing tiles by fabric (fragment count). 30 Table 6. Bricks by fabric (fragment count) 31 Table 7. Late brick forms 31 Table 8. Floor tiles by fabric (fragment count) 32 Table 9. CBM forms by feature (fragment count) 34 Table 10. Flint descriptions 36 List of Plates Plate 1 Sections exposing the coarse fill of ditch 0282 Section facing west, scale 1m 11 Plate Detail of burnt medieval surface cut by chimney foundation 0273, section facing west. 12 Plate 3 Probable cess pit 0269 cut by concrete stanchion, section facing south, scale 1m 12 Plate 4 Structure 0252 number board and scale over stones facing west, 1m scale. 14 Plate 5 chimney foundation showing perpendicular tracery, looking west, scale 1m 18 Plate 6 Lower foundation layer made up from segments of a stone chimney, photo looking west, scale 1m. 18 Plate 7 Chalk floor, partially removed by, with postholes indicating a wall line with flints and retained chalk surface beyond. Photo looking southeast, 2m and 1m scales 19 Plate 8 Detail of wall line with postholes and raised flint surface to the right, photo facing east, 2m scale. 20 Plate 9 Jug with incised wavy lines 24 Plate 10 Chafing dish 25 Plate 11 sgrafifito decoration 26 Plate 12. Painted window glass 38 Plate 13. Mandibles of sheep and goat from 1 8years 45 Plate 14 Natural shed antler of 2 3year old buck 47 Plate 15 Arthritis in a cattle proximal phalange 49 Plate 16. pathological bone growth 51

List of Appendices Appendix 1. Appendix 2. Appendix 3. Appendix 4. Appendix 5. Appendix 6. Brief and specification Context list Dressed Stone catalogue Pottery catalogue CBM catalogue Small finds catalogue

Summary Excavation in advance of building work on land adjacent to St Peters House Risbygate Street identified several pits containing rubbish likely to have come from the medieval hospital of St Peters, one of the five hospitals associated with the Benedictine Abbey located on the main roads outside of the town gates of Bury. The contents of the pits included the usual range of domestic livestock with both wild and domestic birds; of particular significance was a varied collection of fish bones including: eels, salmon, Trout, Cod, skate, and herring, alongside oyster shells, food remains which are consistent with the diet of a religious house. The pottery included some fine tableware s dating from the 12th to the 17th centuries (when the hospital went out of use). A burnt clay surface and an early medieval ditch were the only insitu remains likely to directly relate to the hospital but the displaced remains of the buildings included green window glass, with some painted fragments, and limestone tracery with segments of a stone chimney and at least two perpendicular windows one of which was blind, all reused in the foundations of a later chimney. 2

Drawing Conventions To be provided by graphics team 3

1. Introduction An archaeological excavation was carried out on land adjoining St Peters House Care Home in advance of the construction of a large extension. The work took place during 2011 and followed on from two phases of evaluation trenching over the development area. It was carried out by the field team of the Suffolk County Council Archaeological Service. The site is located on one of the major medieval routes into the town of Bury St Edmunds and is on the site of a Benedictine Hospital associated with the Abbey of St Edmund 2. The Excavation 2.1 Site location geology and topography The site lies at TL 842 657 on the south side of Out Risbygate Street, Bury St Edmunds. It is situated on a northwest facing slope over the floodplain of the River Lark at C. 36-37m OD. Surface geology consists of drift deposits of sand and gravel over Cretaceous chalk bedrock (Lewes Nodular Chalk Formation, Seaford Chalk Formation, Newhaven Chalk Formation and Culver Chalk Formation, British Geological Society.). Site conditions have revealed truncation with orange silt/sand present over much of the site with an irregular profile of chalk bedrock protruding to the surface in places. The previous site use was as a car park with tarmac over an orange gravel sub-base. 2.2 Archaeological and historical background A documentary search has been carried out by Archaeological Project Services (Report No. 114/10). The search identified cartographic, documentary and archaeological evidence which identified the site as probably part of the medieval hospital of St Peter s. The hospital was associated with the Benedictine Abbey that was at the centre of the medieval town. Despite the dissolution of the monasteries in the 16th century the site was allowed to continue in use as a hospital at least into the middle of the 17th century and graves have been found, particularly on the site of St Peter s House and on the properties to the east. The cartographic evidence suggests that the area of the excavation was once a part of the grounds associated with the hospital. 4

A King's Lynn King's Lynn Norwich Norwich 20 Norfolk 28 Norfolk 51.5m Lowestoft Thetford Thetford 16 20 GROVE RO AD 53 Suffolk SUFFOLK Bury St. Edmunds A Bury St. Edmunds Cambridge CHALLICE 18 Ipswich B Ipswich 11 52.0m 58 El Sub Sta Felixstowe West Suffolk College El Sub Sta Colchester Colchester Essex 62 63 64 Hertford Chelmsford 0 25 km 25 km 77 78 2 km 6a 1 0 6 0 4 Chelmsford 69 70 GRO VE P ARK 6 Harlow 63 Essex ROAD 4 584500 584400 B 65 Cambridgeshire N Sunways 2 BSE 171 1 Playground Suffolk Regimental Museum 70 FS Shelter Stone Shelter 50.2m 38 37 1 to 9 28 26 22 17 29 S LE Garage The Beeches 23 24 23 to 33 16 15 21 22 CLO SE 6a 10 to 21 22 1 5 Ward Bdy CR 31 32 RACE YORK TER BSE 392 38 4 37 31 264400 6 32 18 17 WA El Sub Sta 16 33 OF 12 49a CE 44.5m 8 49 41 IN OUT RISBYGATE 1 PR 54 BSE 048 34 to 45 to9 12 11 to 14 House gate 7 8 Grovebury Out Risby 46 1 E 7 264500 2 Stone OUT RI SBYGAT 30 51.4m 149 7 147 146 141 47.9m 132 28 129 7a 137 123 d a York Ro 6 12 98 33 23 31 0 100m 9 10 36 42 Crown Copyright. All rights reserved. Suffolk County Council Licence No. 100023395 2014 8 YORK CLOSE TL 35 26 56.5m 264300 34 36 37 4 3 85 47 49 51 Figure 1. Location area, showing excavation area (red), evaluation trenches (blue), features (black) and HER entries mentioned in the text 30 51a 13 14 PRINCE 15 14 12 51b 11 17 19 18 26 21 20 24 25 Posts El Sub Sta 11 Garden House 15 16 Zulu Lane 300 17 1 16 00 1 2 93 89 10 52.6m 103 1 118 109

N 0 50m Plan Scale 1:1000 Figure 2. 1886 OS map with excavation area (red) 1

N Chalk Clay Modern Bucket treads 0215 0217 0219 0223 0221 Brick Gravel 0201 S.38 S.16 0247 S.23 0202 0214 0212 S.17 1 0280 S.39 0252 S.37 0271 0244 0269 0269 0256 0233 0282 0273 0250 0237 0235 0248 S.41 0 10m Plan Scale 1:150 S.42 S.23 0264 S.24 0267 0285 Figure X. Figure Caption

N Evaluation trench S.23 0202 0214 0212 S.17 0252 0280 S.37 0271 S.39 0269 0282 S.41 Burnt surface 0 10m Plan Scale 1:150 Figure X. Figure Caption 1

2. The Excavation 2.1 Site location geology and topography The site lies at TL 842 657 on the south side of Out Risbygate Street, Bury St Edmunds. It is situated on a northwest facing slope over the floodplain of the River Lark at C. 36-37m OD. Surface geology consists of drift deposits of sand and gravel over Cretaceous chalk bedrock (Lewes Nodular Chalk Formation, Seaford Chalk Formation, Newhaven Chalk Formation and Culver Chalk Formation, British Geological Society.). Site conditions have revealed truncation with orange silt/sand present over much of the site with an irregular profile of chalk bedrock protruding to the surface in places. The previous site use was as a car park with tarmac over an orange gravel sub-base. 2.2 Archaeological and historical background A documentary search has been carried out by Archaeological Project Services (Report No. 114/10). The search identified cartographic, documentary and archaeological evidence which identified the site as probably part of the medieval hospital of St Peter s. The hospital was associated with the Benedictine Abbey that was at the centre of the medieval town. Despite the dissolution of the monasteries in the 16th century the site was allowed to continue in use as a hospital at least into the middle of the 17th century and graves have been found, particularly on the site of St Peter s House and on the properties to the east. The cartographic evidence suggests that the area of the excavation was once a part of the grounds associated with the hospital. The location plan (Fig.1) includes selected entries from the Historic Environment Record: BSE 048 OS location of St Peters Hospital from the OS. BSE 171 Site of chalk pit and possible location of c.17 th century plague pits BSE 392 Monitoring of a kitchen extension exposed medieval burials relating to the medieval hospital (Brooks 2013). 9

3. Methodology The area of the excavation was stripped of concrete, sub-base and a mixed layer of redeposited dark silt with clay flints and brick rubble, by a large back-acting machine. The strip exposed the natural subsoil at the level where it had been truncated by previous phases of building work. The site was hand drawn and located with a Total Station Theodolite. Drawings were at scales of 1:50 for the site plan with selected details at 1:20 with all sections at 1:20. At least 50% of all pit fills were excavated and selective sampling was carried out..a single sequence numbering system was used for all contexts and digital colour photographs were taken of all stages of the fieldwork, and these are included in the archive. Site data has been input onto an MS Access database and recorded using the County Sites and Monuments Code BSE 367. Drawings have been scanned and digitised. An OASIS form has been completed for the project (reference suffolkc1-137734) and a digital copy of the report submitted for inclusion on the Archaeology Data Service database. 4. Results The results of the excavation are presented by phase. These are illustrated in Figures 1-4. Selected plates are included in the rear. Full feature descriptions are included in the context list; all those of archaeological significance are referred to in the text. 4.1 Descriptions Introduction Phase 1 covers an extended period when the site is suggested to have been a part of the hospital of St Peters, one of the five recorded medieval hospitals situated on the roads out of Bury St Edmunds that were held by the Great Benedictine Abbey of Bury. The features are divided between those likely to be contemporary with the hospital until its demolition in the 17 th century (Phase 1) and those later, which were probably associated with the farmhouse that occupied the site till the late 19 th century (Phase 2). 10

Phase 1 Medieval to 17 th century Ditch 0282 surface 0247 Identified parallel with and partially beneath the northern baulk of the site was ditch 0282. (Fig. 3 and sections). It was exposed in a small extension where a six metre length was sectioned lengthways against the baulk. It was c. 0.9m deep and, based on projecting the width at the deepest point of the section, about 2.8m wide. The main fill and was made up of course flint with nodules up to c.0.12m wide - containing little organic material. Most of the ditch was removed and from 48 pottery sherds it can be dated to the 12 th century. The top fill of the ditch was partially sealed by layers of hard fired clay, including fired cracked flints, yellow clay and mortar, 0247. These deposits are interpreted as the remains of a floor surface: It was cut by the foundations for a chimney, 0273 (Phase 2) and several later postholes. The surface had been truncated, probably during the 20th century, surviving only in the top of the ditch. Plate 1 Sections exposing the coarse fill of ditch 0282 Section facing west, scale 1m 11

Plate Detail of burnt medieval surface cut by chimney foundation 0273, section facing west. Plate 3 Probable cess pit 0269 cut by concrete stanchion, section facing south, scale 1m 12

Pits 0269, 0205 and 0212 The earliest of the pits was 0269, which was approximately circular, 2.5m wide and 1.4m deep (the upper fill was partially removed by a concrete stanchion base). Five layers of fill were identified and the most significant of these, 0262, was composed of general waste of pottery and animal bone with dark silt, which also containing ash and charcoal. The pit produced 153 sherds of pottery dating from the 11th to 14th centuries. Pit 0205 was oblong measuring c.4m in length by up to 1.5m wide although the upper edges appear to have fallen in concealing an almost rectangular trench measuring 3.5m x 1m. There were six separate fills (this entire feature was excavated and duplicate numbers were issued either side of the pit for the same layer). There was a high proportion of domestic waste throughout and the lowest fill contained a high proportion of oyster shells. Soil samples were taken from context 0232, which was close to the base of the pit and contained charcoal, for fine sieving. The feature produced 144 sherds of pottery, dating the feature to the 16 th century but included a substantial quantity of floor tile fragments, of likely medieval date (tiles of a similar fabric were reused as hard-core in fire place foundation trench 0275 (Phase 2). A single decorated fragment of floor tile was also recovered from context 0206, which is likely to have originated in the medieval hospital. Pit 0212 was at least 3m in length (running beneath the northern site baulk) and up to 1m in depth with steep sides and a flat base. It was cut by a shallow pit. 0214. Soil samples were taken and twenty six sherds of pottery recovered proving a date of the 16th to 17th centuries; a probable 19 th century brick from a middle upper fill is suggested to be a late intrusion, impressed into a backfilled feature. Structure 0252 This feature was partially dug during the earlier evaluation but more fully exposed during the excavation. It was approximately 2m wide and at least 5m in length and 0.4m deep (Plt. 4).It had been damaged by a concrete stanchion to the south, where it exited the site, and by a contractors bucket during the removal of the overlying concrete. Only the base of the feature survived. The sequence of construction would appear to have been the excavation of a pit that was levelled out on a surface of hard chalk. Within the pit a rectangular structure was built; five roughly shaped large limestone boulders 13

(possibly glacial? erratic s) were used as the base for the north side and on the east side flat facing stones, mostly quite large and made of flint, with some smaller flints were arranged to create a flat face to the inside of the pit, the wall was not mortared in place. The fill behind the walls was made up of fairly sterile silt and gravel; inside the structure was the base of a layer of very dark silt, which contained twenty-six sherds of pottery. The pottery ranged in date from early medieval period to the 13th to 14th centuries; it also contained a small amount of tile and animal bone. This feature is thought to be a well-constructed cess pit dating from the 13th or 14th centuries. Cess pits were often cleaned out to provide manure or simply to make space in cramped, built up, areas such as towns. This well-built feature would was built for ease of cleaning being relatively shallow making it accessible, and long. The stones used on the construction were mostly rough limestone and large flint blocks (some measuring c.0.4m x 0.3m); elsewhere in the town building stone was robbed from ecclesiastical buildings following the dissolution of the monastery s in this case however the limestone was quite rough and the flint, simply cut and very large and may simply have been offcuts and various odd pieces not required for the standing buildings. The presence of a probable cess deposit in the base may be evidence that the feature was not wholly cleaned out before abandonment. Plate 4 Structure 0252 number board and scale over stones facing west, 1m scale. 14

N Brick Chalk surface Tile Chalk surface continuation Flint 0273 0273 0273 0 10m Plan Scale 1:200 0 1.00m 2.50m Plan Scale 1:50 (0273) Figure X. Figure Caption 1

S.17 WNW ESE N S S.23 W E 0214 0213 0202 0203 0204 0209 0211 0278 0279 0231 0208 0207 0230 0205 0206 0232 0213 0214 0231 0210 0212 S N W E S.38 0240 0241 0242 0253 0243 1 0254 S.41 S.42 W E N S E W 0259 0247 0290 0258 0289 0229 0291 0256 0292 0238 0293 0273 0287 0282 0284 Chalk Stone Tile 0 1.00m 2.00m Brick Flint Limestone Section Scale 1:40 Figure X. Figure Caption

Phase 2 Post 1549 to 17th century Introduction Documentary research suggests the hospital went out of use in the 17 th century and was replaced by a farm. Evidence of this was recorded intermittently across the site with floor surfaces and some structural remains, most conspicuously the foundations of a large chimney. Chimney foundation 0273 Located against the eastern edge of the site was the probable foundation for a chimney. It projected approximately 2m into the site running east-west with a projection at right angles of 1.45m to the north; this western projection would have been matched by one at the eastern end, located beyond the site. The main body was 1.1m wide and the projection was slightly narrower at 0.9m. Positioned over ditch 0282 (Phase 1) the foundations were up to c.1m deep taking them to the base of the ditch but slightly shallower beneath the western arm, which was dug into firmer ground beyond the ditch. The lowest fill was predominantly of flint which was then packed with dressed stone. These were in several layers with large sections of a limestone chimney in the middle (these consisted of semi-circles of stones which combined in pairs and stacked formed the chimney; the inside face of chimney was blackened with soot and burnt red in places) structure. The middle and upper fills included large pieces of tracery from perpendicular style windows; represented were at least two windows one of which was blind. On the surface of the western projection was a single large stone with mouldings measuring c.0.3m wide x 0.3m deep and 0.6m long. The stone was all from Barnack in Northamptonshire and likely to date from after the 13th century a catalogue of the dressed stone is included as Appendix 3. The use of Late medieval dressed stone in a footing is typical of post-medieval buildings in Bury St Edmunds where the Abbey and other monastic buildings including the hospitals were quarried for building stone for use in general building and in footings. The shape of the foundation would be typical of a large single sided fireplace. 17

Plate 5 chimney foundation showing perpendicular tracery, looking west, scale 1m Plate 6 Lower foundation layer made up from segments of a stone chimney, photo looking west, scale 1m. 18

The fireplace was cut through a probable medieval clay floor surface that was burnt hard in places. It is possible that the new chimney replaced an earlier one on - or close to - the same site, although this is unprovable. Floor surface. Surviving in the northwest corner of the excavation was the continuation of a cobble, clay and chalk surface first encountered in the evaluation (Stirk. 2010). It was up to 5.5m wide at the northwest edge of the excavation extending 3.75m into the site, where it became less substantial. From the evaluation trench we can confirm that a surface, in several layers with clay, chalk and cobbles and more clay extended a further 10m towards Out Risbygate where it was up to c.0.8m deep in places. The surface was bounded by a line of six small postholes on the south edge in a narrow trench, possibly the location of a ground beam. A neat row of flints immediately to the south retain a more solid layer of re-deposited chalk This layer was truncated c.1.25m to the south exposing a gravel sub-layer that extended a further 2.5m close to the western baulk. Over the rest of the site natural orange sand with patches of chalk were directly beneath disturbed deposits indicating that 20 th century building work had severely truncated the site. The relatively good preservation in the northwest corner was probably due to less truncation, beneath the site of a previous building, and to a deeper disturbance beneath the chalk surfaces that contained silt and gravel (sec.38). Plate 7 Chalk floor, partially removed by, with postholes indicating a wall line with flints and retained chalk surface beyond. Photo looking southeast, 2m and 1m scales 19

Plate 8 Detail of wall line with postholes and raised flint surface to the right, photo facing east, 2m scale. Across the truncated surface of the chalk were patches of hard blue-grey clay with impressed brick fragments. These were removed where possibly representing the remains of destruction debris smeared across the site, probably during the construction of the recently demolished car showroom. In south west corner of the site were several postholes (Fig. 3), which cut through surface 0247. These included 0256, 0289, 0292, 0233, 0264, and 0285. Several of these included brick fragments and blue clay, indicative of later features and are interpreted as features relating to the farm and probably 19 th century in date. 20

5. The finds evidence Andy Fawcett 5.1 Introduction Table 1 shows the quantities of finds collected in each context from the excavation. The finds were retrieved from forty-five contexts which include twenty-three pit fills, twelve layers, four post-holes, two ditch, one building, one gutter and two unstratified contexts. A full breakdown by context of the finds can be seen in Appendix *1. Also present are nineteen small finds which have been recorded separately. 5.2 Pottery Sue Anderson Introduction Find type No Wt/g Pottery 494 12975 CBM 344 50770 Fired clay 6 35 Mortar/plaster 3 56 Clay pipe 1 6 Worked flint 4 90 Burnt flint/stone 3 100 Stone 5 2069 Glass 14 109 Iron nails 23 321 Slag 6 1033 Animal bone 3434 8784 Shell 137 1202 Totals 4474 77550 Table 1. Finds quantities A total of 494 sherds of pottery weighing 12975g was collected from 38 contexts. The assemblage was generally in good condition with only minor abrasion of some residual material. Methodology Quantification was carried out using sherd count, weight and estimated vessel equivalent (eve). The minimum number of vessels (MNV) within each context was also recorded, but cross-fitting was not attempted unless particularly distinctive vessels were observed in more than one context. A full quantification by fabric, context and feature is available in archive. All fabric codes were assigned from the author s post-roman fabric series, which includes East Anglian and Midlands fabrics, as well as imported 21

wares; Essex wares are as described by Cotter (2000). Form terminology for medieval pottery is based on MPRG (1998). Recording uses a system of letters for fabric codes together with number codes for ease of sorting in database format. The results were input directly onto an Access database. Pottery by period Table 2 shows the quantification by fabric; a summary catalogue by context is included as Appendix *2. Description Fabric Code No Wt/g Eve MNV Thetford Ware (Grimston) THETG 2.57 3 47 1 St. Neot's Ware STNE 2.70 1 3 1 Early medieval ware EMW 3.10 51 311 0.35 28 Essex-type EMW EMWE 3.102 2 20 0.10 2 Early medieval ware gritty EMWG 3.11 1 2 1 Early medieval ware chalky EMWC 3.12 10 135 4 EMW micaceous EMWM 3.16 1 4 1 Early medieval sparse shelly ware EMWSS 3.19 2 9 2 Early medieval gritty with shell EMWSG 3.191 1 12 1 EMW shell-dusted EMWSD 3.192 32 239 0.14 6 Total 11th-12th c. 104 782 0.59 47 Medieval coarseware MCW 3.20 29 342 0.42 20 Grimston coarseware GRCW 3.22 4 27 1 Bury sandy ware BSW 3.30 2 10 2 Bury sandy fine ware BSFW 3.31 32 349 0.29 28 Bury medieval coarseware BMCW 3.33 77 1067 0.66 45 Bury medieval coarseware gritty BMCWG 3.34 4 52 2 Hedingham coarseware HCW 3.43 1 16 0.05 1 Medieval shelly wares MSHW 3.50 1 3 1 Bury medieval shelly ware BMSW 3.53 24 486 0.05 9 Medieval shell-dusted ware MSDW 3.55 1 10 1 Stamford Ware Fabric B STAMB 3.71 1 24 1 Mill Green Ware MGW 4.22 2 29 1 Total 12th-14th c. 178 2415 1.47 112 Late medieval and transitional LMT 5.10 70 3112 2.16 22 Cistercian type Ware CTW 5.20 1 17 0.25 1 Late Grimston-type ware GRIL 5.30 26 2069 1 Late Essex-type Wares LMTE 5.60 9 432 0.18 7 Raeran/Aachen Stoneware GSW3 7.13 26 1057 0.98 10 Cologne/Frechen Stoneware GSW4 7.14 9 247 0.36 6 Dutch-type redwares DUTR 7.21 12 703 0.62 2 Total L.14th-16th c. 153 7637 4.55 49 Post-medieval redwares PMRW 6.10 1 5 0.10 1 Iron-glazed blackwares IGBW 6.11 1 20 1 Glazed red earthenware GRE 6.12 24 1185 0.96 17 Border Wares BORD 6.22 1 6 0.17 1 Post-medieval slipwares PMSW 6.40 3 76 0.08 2 Total 16th-18th c. 30 1292 1.31 22 Late post-medieval unglazed earthenwares LPME 8.01 2 41 0.06 1 Refined white earthenwares REFW 8.03 4 13 3 English Stoneware ESW 8.20 19 467 0.11 8 Late slipped redware LSRW 8.51 3 138 3 Total L.18th-20th c. 28 659 0.17 15 Unidentified UNID 0.001 1 190 1 Total 494 12975 8.09 246 Table 2. Pottery quantification by fabric. 22

Pottery of the 11th-14th centuries Almost half of this assemblage by sherd count is made up of pottery of 11th- to 14thcentury date. This includes wheelmade Late Saxon wares, the handmade wares (some of which had wheel-finished rims) classified as early medieval and the wheel-made greywares classified as medieval. The earliest pottery comprises three sherds of a Grimston-type Thetford ware vessel (0174) and one small body sherd of St Neot s ware (0262). Both are likely to be of 11thcentury date and broadly contemporary with the earliest early medieval wares. The early medieval wares are dominated by the fine to medium sandy fabric, generally dark grey-black and thin walled, but sometimes oxidised, which is typically found in Norfolk and north Suffolk. A few coarser Essex-type wares are present and shelltempered wares make up a small proportion of the group. Conversely there is a high proportion of shell-dusted wares, although the large number of sherds represents only six vessels. Chalk-tempered wares are also relatively frequent in this group, and this is a fabric which occurs rarely elsewhere in the region. Thetford-type wares with occasional chalk inclusions are often found in Bury and less frequently elsewhere, so perhaps this is a local ware. However as chalk is a common component of the regional geology, further work is required to ascertain the origins of these fabrics. Most sherds are probably from jars, but only eight rims are present. Seven of these are simple everted or flaring forms, and one is beaded, all from jars. No bowls were identified and the jar rims vary in diameter between 90 220mm. Fabrics in the high medieval greywares are generally typical of the town, being dominated by the medium sandy BMCW. Fine and coarser wares are also present, as well as some sherds which are similar to Colchester wares but without the coarse inclusions. Sixteen rims are present, thirteen of which are jars, one jug, one bowl and one?spouted pitcher or handled jar. All medieval coarseware rim forms are early types of probable 12th/13th-century date, with everted bead types being the most frequent, followed by upright beaded/everted/thickened types. There are no square-beaded forms which are typical of the later Bury wares. Decoration is present on eighteen vessels and includes applied thumbed strips (1 example), combed wavy lines (1 23

example), finger-tip impressions (1 example), girth-grooving (2 examples), thumbed bases (2 examples), thumbed rims (1 example), shell dusting (10 examples), and a jug with incised wavy lines on the neck and comb stabbing on the rim Plate 9 Jug with incised wavy lines Only two glazed wares are present. Unusually for Bury there is no Hedingham ware or high medieval Grimston ware in this group. There is a strap handle in yellow-glazed Stamford Ware fabric B, and there are two sherds of a globular-bodied jug of Mill Green type with white slip decoration and clear glaze. Late and post-medieval By weight, the late medieval and early post-medieval group is the largest in the assemblage, but several vessels are largely complete and the group contains many large sherds. By count and MNV it is therefore smaller than the medieval group. In terms of vessel numbers, the group is dominated by local redwares (LMT, GRE) and there is also a relatively high proportion of German stonewares. Smaller quantities of non-local wares include Cistercian-type ware, late Grimston ware, Border ware, unidentified slipware and Dutch redware. 24

Late medieval vessel foms include four bowls or pancheons, a jar/pipkin, a skillet, a cauldron, a dish, a?cistern, two jugs and three mugs in earthenware fabrics, and at least seven mugs and a jug in stonewares. There is also a near-complete example of a chafing dish in LMT, which has a thick pedestal base, a flaring bowl with flat-topped everted rim decorated with finger-nail impressions, two horizontal rod handles, and small domed knobs. Most of the glaze within the bowl is orange, although it appears green in some areas, particularly on the rim. Plate 10 Chafing dish The stoneware mugs include some small globular types which were made in Cologne in the 16th century, one of which has applied decoration in the form of oak leaves and acorns. The possible cistern is a late Grimston product with a slightly rounded base and?everted rim, and two wide strap handles; it is decorated with green glaze inside and out, and has five-toothed stabmarks randomly distributed across one area of the body. Post-medieval forms include two bowls or pancheons, a dish, a jug and three pipkins, a small Border ware jar (cf Pearce 1992, no. 413), and a slipware mug and dish. The 25

slipwares are unusual types, having fairly crude sgraffito decoration cut through the white slip on the rims, and their origin is uncertain. No parallels have been found amongst the Norwich slipwares, Metropolitan slipwares from Harlow, or any of the sgraffito wares from Colchester, and the vessels do not appear to be of continental origin. They may perhaps represent a short-lived local producer. Plate 11 sgrafifito decoration Modern The small quantity of modern pottery is dominated by fragments of English stoneware vessels. White table wares are relatively infrequent, and there are a few sherds of coarser red earthenwares (LPME, LSRW). The only identifiable vessels in this group are a brown stoneware inkpot, a brown stoneware?bowl with a large beaded rim, an unglazed redware bowl, and a transfer-printed whiteware cup. Unidentified One sherd is unidentified. This is a complete, crudely made, pedestal base in a fine sandy fabric, slightly oxidised and abraded on the broken edges. It appears handmade and may be of Early Saxon date. 26

Pottery by context The majority of the assemblage was from stratified fills and layers, with only sixteen sherds recovered as unstratified finds. A summary of the pottery by context is provided in Appendix *2. Table *3 shows the distribution of pottery by feature. Context Identifier Fabrics Spotdate Features 0164 Gutter BMCW 12th-14th c. 0166 Building? EMW, BSW, BMCW 12th-14th c. 0168 Pit BSW, MCW, MGW L.13th-E.14th c. 0201 Pit BMCW 12th-14th c. 0205 Pit BMCW, CTW, DUTR, GRIL, GSW3, GSW4, LMT, LMTE 16th c. 0212 Pit LMT, LMTE, GRE, IGBW, PMRW, PMSW, LSRW 19th c. 0214 Pit GSW3 L.15th-16th c. 0244 Pit BMCW, LMT 15th-16th c. 0252 Feature BMCW, BSFW L.12th-14th c. 0258 Posthole EMW, EMWC 11th-12th c. 0269 Pit STNE, STAMB, EMW, EMWC, EMWE, EMWG, EMWM, EMWSD, 12th-13th c. EMWSG, EMWSS, MCW, BSFW, MSDW, BMSW, BMCW, UNID 0271 Ditch EMW, BMSW, MSHW, BSFW 12th c.+ 0280 Ditch EMW, EMWSD, GRCW, BSFW, BMCW 12th-13th c. 0282 Pit EMW, EMWE, EMWSD, HCW, BMCW 12th c.+ Layers 0153 Layer ESW 17th-19th c. 0155 Layer GRE, GSW4 16th-17th c. 0157 Layer REFW L.18th-20th c. 0239 Layer EMW, EMWC, BMCW, LMT, GRE, ESW, LSRW, REFW L.18th-19th c. 0240 Layer ESW, LSRW, REFW L.18th-19th c. 0241 Layer ESW 19th c. 0243 Layer GRE, GSW4, LPME, ESW 19th c. 0254 Layer BSFW, LMT, GRE, PMSW, BORD 17th c.? Unstratified 0174 U/S THETG 10th-11th c. 0294 U/S EMWSS, MCW, BMCW, BMCWG, BMSW, BSFW L.12th-14th c. Table 3. Pottery by feature. Three pits produced over 1kg of pottery each. Pit 0269 contained 153 sherds (2371g; MNV = 90) of early and high medieval pottery, including the rims of nine jars, one jug, one bowl and a spouted pitcher, and is likely to date to the 12th/13th century. Pit 0205 produced 144 sherds (7405g; MNV = 41), of which only a small quantity was residual medieval, the majority being of probable 16th-century date and including the rims of three large bowls or pancheons, a handled bowl, a cauldron, a chafing dish, a possible cistern, a dish, a jar/pipkin, three jugs, nine mugs and a skillet. Pit 0212 contained 26 sherds (1246g; MNV = 19) of which the majority belongs to the 16th/17th century, but the presence of 19th-century pottery and brick in middle fill 0204 suggests that the backfill, if not the feature, may be later. One other pit, 0282, contained a large group (48 sherds, 305g, MNV = 17), including the rims of four jars, and is likely to be of broadly 12th-century date. Two pieces of later CBM were also collected from the upper fill but are probably intrusive. All other features 27

contained less than twenty sherds each; most of these are medieval. The layers appear to be mainly of post-medieval date. Discussion Pottery of the early and high medieval periods forms a large proportion of this assemblage. The presence of 11th/12th-century wares on the site suggests that there was limited activity here prior to the foundation of the hospital, although it seems that this was not particularly intensive in the Late Saxon period. Much of the high medieval assemblage appears to date to the 12th and 13th centuries, perhaps suggesting that the area was not used for rubbish disposal in the later years of the hospital. The range of wares and forms in use is not unusual for the town, but the lack of glazed wares is surprising. Glazed wares can be less frequent on sites away from the town centre, but excavations at another Bury medieval hospital on the periphery, St Saviour s, produced a much higher proportion of these wares, particularly the products of Hedingham and Grimston (Blinkhorn 1997). Pottery associated with the late medieval and early post-medieval use of the site forms a small but interesting assemblage. Most of the wares and forms are again typical of the period, but a few vessels are significant or unusual types. These include the nearcomplete Late Medieval and Transitional ware chafing dish, the Grimston cistern, and the small decorated Cologne mug. These tablewares suggest a degree of status during the 16th century. Amongst the 17th-century wares, the two items of slipware stand out as being unusual for both the town and the region, although their manufacture is not of a particularly high standard. The few sherds of modern pottery suggest that rubbish was being deposited elsewhere during this period and the fragments probably represent vessels broken in the vicinity and accidentally incorporated into garden soil. The number of stoneware vessels is unusual, perhaps suggesting that some of these vessels were reused as horticultural pots. 5.3 Ceramic building material Sue Anderson Introduction 28

A total of 344 fragments of CBM (50,770g) was collected from 31 contexts. Table *4 presents the count and weight quantification by form. A catalogue by context is included in the Appendix *3. Methodology Type Form Code No Wt (g) Roman Roman tile RBT 2 512 Roman tile? RBT? 1 268 Roof Plain roof tile RT 106 5646 Plain roof tile? RT? 2 10 Ridge tile RID 1 967 Ridge tile? RID? 1 60 Pantile PAN 2 194 Wall Early brick EB 3 3816 Early brick? EB? 3 402 Late brick LB 83 14413 Late brick? LB? 2 67 Flooring Medieval floor tile FT 121 19800 Flemish floor tile FFT 14 4416 Floor tile? FT? 1 16 Miscellaneous Drainpipe DP 1 176 Unidentified UN 1 7 Total 344 50770 Table 4. CBM quantities by form The assemblage was quantified (count and weight) by fabric and form. Fabrics were identified on the basis of macroscopic appearance and main inclusions. The width, length and thickness of bricks and floor tiles were measured, but roof tile thicknesses were only measured when another dimension was available. Forms were identified from work in Norwich (Drury 1993), based on measurements. Other form terminology follows Brunskill s glossary (1990). The assemblage Roman tile Three fragments of possible Roman tiles were recovered. They are in fine to medium sandy fabrics with occasional flint inclusions and two pieces are abraded. They are 31 33mm thick and two fragments have reduced surfaces. One of these pieces has a rubbed edge, suggesting that it may have been reused in a later period. Roofing Roof tiles form one of the largest components of this assemblage. They comprise plain tiles (108 fragments), ridge tiles (2 fragments) and pantiles (2 fragments). Table 2 shows the quantities by fabric. 29

fabric code RT RT? RID RID? PAN estuarine clays est 1 fine sandy with few other inclusions fs 19 1 2 medium sandy with few other inclusions ms 8 fine sandy with red clay pellets fscp 4 medium sandy with red clay pellets mscp 2 medium sandy with chalk msc 4 fine sandy with ferrous inclusions fsfe 18 medium sandy with ferrous inclusions msfe 39 1 1 medium sandy with flint msf 5 medium sandy with flint and ferrous msffe 2 fine sandy with grog fsg 3 fine sandy micaceous fsm 1 1 Table 5. Roofing tiles by fabric (fragment count). The majority of plain tiles are in red-firing fine and medium sandy fabrics, with ferrous fragments being the most frequent additional inclusions. Five fragments have peg holes (4 round, 1 square). Where it is possible to tell, most peg tiles had two holes. Only one fragment in the estuarine fabric of medieval date was found. Red-firing fragments were assigned a medieval date on a variety of attributes including coarseness of fabric and form, presence of a reduced core and/or surfaces, and presence of glaze. On this basis, there are only seven medieval fragments, a further two which appear transitional and may be late medieval, and ninety-seven post-medieval fragments. Two fragments show signs of glaze. No fragments are complete enough for measurement. A few plain tiles in 0254 show signs of sooting or burning and may have been used in a hearth or chimney. Many fragments have thin deposits of lime mortar on the surfaces and breaks and had probably been re-used in walling, although thick mortar on the underside of one, in 0207, may represent plastering of an open roof space. Two fragments of ridge tiles, both of post-medieval date, were identified, one in a medium sandy fabric and 20mm thick, the other fine micaceous and 11mm thick. Two pantile fragments are both in fine fabrics, and at least one appears to have been machine-pressed, suggesting a 19th-century or later date. 30

Bricks Ninety-one fragments of brick were recovered, in fabric groups shown in Table *6. fabric code EB EB? LB LB? estuarine clays est 3 3 fine sandy fs 4 medium sandy ms 2 fine sandy with flint fsf 1 medium sandy with flint msf 24 fine sandy with ferrous inclusions fsfe 8 1 medium sandy with ferrous inclusions msfe 17 medium sandy streaked with white clay, and ferrous inclusions msxfe 2 medium sandy with flint and ferrous inclusions msffe 8 fine sandy with grog fsg 1 medium sandy with grog msg 1 medium sandy with grog and ferrous msgfe 1 fine sandy micaceous fsm 5 1 fine sandy with common voids fsv 4 medium sandy with chalk msc 1 white-firing fine sandy wfs 4 Table 6. Bricks by fabric (fragment count) Early bricks in estuarine clay fabrics, as described by Drury (1993) and dated to the 13th-15th centuries, comprised six fragments. One or more dimensions were recorded for five bricks, and of these three could be assigned to Drury s forms (two EB4?, one EB9). Drury has suggested that bricks with sanded bases are earlier than those which have straw impressions. In this assemblage, both types occurred in the same context, but this is probably due to redeposition. Late bricks comprised eighty-five fragments (14,480g). Table *6 shows that a very broad range of fabric types was present, as is typical of post-medieval assemblages. Twenty-five bricks could be measured in one or more dimensions, and forms were assigned where possible, as shown in Table *7. Thicknesses varied between 45 66mm, with the majority below 53mm, and nine widths varied between 97 124mm. There was no obvious clustering of brick forms by fabric, all red brick forms occurring in fabrics msf and msfe most frequently. Colour Form No. Red LB1 1 LB1? 1 LB1/2 15 LB3/6 3 LB4? 1 LB5? 1 LB6 1 White LB9 1 LB10 1 Table 7. Late brick forms 31

Bricks within the size ranges for LB1/LB2, assigned to the mid 16th 18th centuries and 16th century or later respectively were the most common in this assemblage. Several fragments show signs of burning, partial vitrification or reduction of the surfaces and/or core, much of which is likely be due to firing conditions and suggests a premodern date. It is also possible that some of the burnt fragments were parts of hearths or chimneys, however. One white-firing fragment from layer 0241 showed signs of wear and may have been used as a paviour, although at 50mm it was thicker than the white bricks normally made for this purpose. A variety of mortar types is present on the bricks. For example, a particularly coarse buff-coloured mortar containing small flint fragments is present on a brick of probable 18th/19th-century date in 0204, and a pink medium sandy mortar containing chalk is found on a Tudor brick in 0274. Most of the other mortars in the assemblage are white or cream-coloured with medium sandy aggregates. A white brick in layer 0241 measures 97mm wide and 53mm thick, but appears to have been rubbed on one stretcher face. A red brick fragment in pit fill 0213 has a concave surface, but appears to have been moulded in this shape rather than rubbed. A very shallow frog is present on a half-bick from layer 0241, measuring 110 x 59mm, and is probably of 19th-century or later date. Flooring Floor tiles are represented by 136 fragments in a variety of fabrics, as shown in Table *8. Fabric code FT FT? FFT estuarine with coarse sand est(cs) 117 fine sandy fs 2 fine sandy with red clay pellets fscp 1 fine sandy ferrous fsfe 2 11 fine sandy micaceous fsm 1 1 medium sandy with grog msg 1 Table 8. Floor tiles by fabric (fragment count) 32

A fragment of a relief floor tile in fabric fsm was recovered from pit fill 0206. The fragment is not worn and the design shows clearly under the orange glaze. It appears to be part of an animal (most likely a lion or griffin) with the head end to the right and a long tail to the left, and foliage in the centre (Fig. xx). No parallels have been identified amongst the Suffolk relief tile series (Sherlock n.d.). This type of tile was made between the 12th and 14th centuries, although the best-known local production site, at Bawsey in Norfolk, is dated to the second half of the 14th century (Lemmen 2000, 8 9). Fourteen fragments in fine sandy fabrics are pieces of Flemish floor tile, many of which were recovered from foundation trench fill 0275. These were either contemporary with or slightly later than the relief tile. All except one are worn, but traces of glaze and/or slip are present on the sides, showing that they were originally coloured brown, or yellow with white slip. One fragment had been cut to a triangle. Thicknesses vary between 26 45mm (most less than 30mm), and three larger fragments were over 115mm, 155mm and 167mm square. It is likely that all the fragments were from the larger size range of this type of tile. Four fragments of floor tile in fine or medium sandy fabrics are of uncertain date, but could be pieces of Flemish or relief tiles and appear more likely to be medieval or late medieval, rather than pieces of quarry tile of post-medieval date. Two are measurable and are within the thickness range of the Flemish tiles in this group at 28 29mm. The most unusual find in this group is a large quantity (117 fragments, 19,224g) of floor tile in a fabric similar to the estuarine bricks but with abundant coarse sand inclusions. The fabric is very friable and the number of fragments reflects this, but it is unclear exactly how many tiles are present as attempts to fit the fragments together were largely unsuccessful; based on the weight there is a minimum of ten, but the variation in thickness and appearance suggests that the total is probably greater. Some large pieces are present, and one full width measures 217mm. Thicknesses vary between 32 40mm, with the majority around 35mm thick. Several tiles appeared to be decorated with three parallel incised lines running from each corner across the tile to form a cross, and one fragment has traces of dark brown glaze. All fragments were collected from fills of pit 0205 and foundation trench 0273, the majority from the latter. 33

Miscellaneous One fragment of a compressed fabric drainpipe with light brown glaze was collected from 0241. This is of 19th-century or later date. One flake in a medium sandy micaecous fabric with a reduced base was of uncertain form. CBM by context Table *9 shows the distribution of CBM forms by feature, based on fragment count. Two of the largest groups of CBM were recovered from pits 0205 and 0212, features which also contained significant quantities of post-medieval pottery. Several layers also contained relatively large quantities, most notably layer 0243, although the pieces from these contexts were generally smaller and more abraded, suggesting accidental incorporation of material which was already present on the site before the layers formed. The largest quantity overall was from a structural feature, foundation trench 0273, which also produced a large quantity of worked stone assumed to be demolition rubble from the medieval hospital. Context Identifier RBT RT RID PAN DP EB LB FFT FT? Features 0168 Pit 6 1 0201 Pit 2 1 0205 Pit 1 12 1 4 2 6 0212 Pit 1 40 1 13 1 1 0214 Pit 4 0217 Posthole 1 0221 Posthole 1 0244 Pit 4 1 0258 Posthole 1 3 0269 Pit 1 0273 Foundation trench 33 11 114 0282 Pit 2 Layers 0155 Layer 5 2 0157 Layer 2 1 0159 Layer 8 5 1 0239 Layer 1 2 0240 Layer 1 1 0241 Layer 1 7 0243 Layer 22 1 1 11 0254 Layer 3 Discussion Table 9. CBM forms by feature (fragment count) The majority of stratified CBM from this site was collected from a single foundation trench (158 fragments) and pit fills (105 fragments). General layers contained 75 34

fragments, with smaller quantities from post-holes (6 fragments). No CBM was recovered from walls or floors. Much of this assemblage therefore represents hardcore, whether intentionally or unintentionally used, and demolition rubble. It can, however, provide some clues to buildings and structures which once stood on or near the site. The few fragments of possible Roman tile were redeposited in medieval and postmedieval pits but were probably associated with Late Saxon activity on the site. The medieval pit (0269) also contained residual Late Saxon pottery. Although only small quantities of early brick and medieval roof tile were recovered, a high proportion of the assemblage was of medieval date. This is due to the large quantities of coarse medieval floor tile recovered from two features, as noted above. This unusual group may represent a one-off attempt to produce floor tiles by manufacturers who were more used to making roof tiles or bricks. Certainly the tiles are very crudely made and the fabric is comparable with roof tiles from Thetford Grammar School, where they appear to have been introduced in the early 13th century (Anderson 2008). A few fragments of finer medieval tiles, most notably a relief-decorated fragment, were also recovered, suggesting that the remains came from at least two areas of tiled flooring within the hospital. CBM representing the late medieval period comprised a couple of roof tile fragments, Tudor late bricks, and fragments of Flemish-type floor tiles. Most of them were recovered from features which contained later CBM or pottery, where they were probably deposited following either demolition of the building or replacement of a floor. The post-medieval assemblage comprises plain roof tiles and some red-firing late bricks. Whilst some of the material could represent construction debris, the majority of it had been used and probably indicates demolition of post-medieval structures. Modern 18th and 19th-century structural ceramics included red and white-firing bricks, pantiles and a drainpipe. 5.4 Fired clay Two medieval contexts contain small fragments of fired clay, building fill 0165 and pit fill 0169, although most of the pieces are in context 0165 (5 fragments @ 33g). The pieces are generally only slightly abraded. All of the examples are oxidised and are in a 35

medium sandy fabric with abundant ill-sorted chalk (msch). One fragment in fill 0165 exhibits a buff irregular-flat surface. No impressions such as wattle rods or finger marks are present on any of the pieces. The fragments are too small and few to suggest being part of an oven or wall for instance. Both of the contexts also contain medieval pottery. 5.5 Plaster/mortar Pit fills 0170, 0206 and the unstratified 0246 contain single pieces of mortar (56g). With the exception of the fragment in context 0206 all of the mortar is small and abraded. The pieces are predominantly buff in a medium sandy fabric with abundant ill sorted lime (msl). Early post-medieval pottery is present in fill 0206. 5.6 Clay tobacco pipe A single clay tobacco pipe stem (6g) with part of the foot attached is present in layer 0243. The piece is dated to the post-medieval period and is accompanied by pottery of the same date. 5.7 Worked flint Justine Biddle Four pieces of struck flint was recovered from four contexts. It was recorded by type and other descriptive comments about appearance, condition and technology were noted and a date has been suggested. Descriptions are included in the table below. Context Type No Patinated Notes Date 0171 Flake 1 No The broken distal end of a thin flake with approximately 90% cortex remaining on the dorsal surface. There is no evidence of use-wear or retouch. 0200 Flake 1 No Large, flat, sub-triangular flake with pronounced ripples. There is a negative flake scar on the dorsal surface and there is approximately 10% cotex remaining on two edges. There is no evidence of retouch or use-wear. 0263 Flake 1 No Thin flake with approximately 10% cortex remaining on the distal end. There is no evidence of retouch or usewear. 0281 Flake 1 Yes Patinated flake which is broken at the distal end and has a hinge fracture. There is no evidence of retouch or usewear. Table 10. Flint descriptions Later Prehistoric Later Prehistoric Later Prehistoric Later Prehistoric 36

The assemblage consists of four objects, none of which are definitively diagnostic of any period and therefore a general later prehistoric (Neolithic to Iron Age) date has been assigned to them. 5.8 Burnt flint/stone Three fragments of burnt flint are present, one each in pit fills 0245, 0260 and ditch fill 0281. All of the fragments are coloured grey/white and may represent pieces used in the preparation and cooking of food however, all are residual in medieval contexts. 5.9 Worked stone Five stone fragments (2069g) are divided between three layers 0241, 0274 and 0275. The piece in fill 0241 is a small thin and abraded fragment of marble (3g), the context also contains post-medieval pottery. A single shattered fragment of sandstone with one smoothed surface (32g) is present in fill 0274. Context 0275 contains three different stone fragments. Two are abraded pieces of clunch (673g) both with partial remains of surfaces. One example displays fine tooling marks on it surface. The fragments are quite fine and are more likely to have been used on the interior of a structure. The final piece in this context is large coarse limestone fragment. It has a depth of 65mm and has the remains of one tapered and rounded worked edge (110mm). Contexts 0274 and 0275 both contain CBM dated from the 15th/16th century. Dressed stone retrieved from the Chimney foundation, 0252, has been individually photographed and is included separately as Appendix 3 with this report. 5.11 Iron nails Corroded iron nail fragments are present in eleven contexts. The shafts, in section, are a mixture of square and rectangular, whereas most of the heads are square. With the exception of pit fills 0262 and 0279 (dated to the medieval period) the remaining contexts with iron nail fragments all contain post-medieval pottery. 37

Plate 12. Painted window glass 5.10 Glass Three contexts contain pieces of post-medieval glass layers 0240, 0241 and pit fill 0245. With the exception of a single window glass fragment in 0241 the remainder of the assemblage is made up of clear, green and brown bottle glass. All of the contexts also contain post-medieval pottery. 5.12 Slag Two contexts contain fragments of slag, pit fill 0260 and post-hole fill 0268 (6 pieces @ 1033g). Five of the pieces (968g) are present in context 0268 and all of these are magnetic, some of which also display copper streaks and spots. The large fragment in fill 0260 is only slightly magnetic. It is a conglomerate of small materials, such as pottery, chalk, charcoal as well as slag. The context also contains 13th century pottery. 5.13 Small finds Identified by Ruth Beveridge In total nineteen small finds are present within the finds assemblage, four of which are copper alloy seven are iron, two stone, one lead and five glass. The small finds were recorded in eight pit fills and one layer. Eleven of the small finds are located in the late medieval/16th century pit 0205, three in post-medieval layer 0254 and three in the 38

medieval pit 0269. A chronological summary of the small finds follows and a full description can be seen in Appendix *4. Medieval 1. A complete wire wound copper alloy pin. It has a spherical head and similar types can be seen in the Norwich Households corpus (Margeson 1993, 12; fig 5). SF1007 (Pit fill 0208). 2. Two fragments of dark green window glass (12g). Both are slightly dirt encrusted and abraded. One has a straight edge and the other two. SF1015 (Pit fill 0206). 3. Eight pieces of irregular shaped dark green window glass (53g). They are of a variable size and mostly dirt encrusted. However of note is one decorated fragment (Plate 1). It is dark green and displays a very light coloured orange/yellow line arrangement. The lines represent the corner of a border in which the outer one is thin, followed by a thicker line and two diagonal ones running from the left hand corner. The fragment is slightly bent and a clipped upper edge can be observed. SF1016 (Pit fill 0208). 4. A single triangular shaped dark green window glass fragment (3g). SF1017 (Layer 0254). 5. An abraded sub-triangular shaped dark green window glass fragment (1g). SF1018 (Pit fill 0262). 6. Four irregular shaped and dirt encrusted fragments of dark green window glass (8g). SF1019 (Pit fill 0277). Post-medieval 7. A worn copper alloy Nuremberg token (2g) dated from AD1550-1600. It is in the rose and orb style, however very little of its decoration can be seen and none of the legend is legible. SF1001 (Pit fill 0210). 8. A worn copper alloy Dutch duit dated to AD1606. The denomination is unknown, however halves, ones and twos are the most frequent. The obverse reads FRI/SA-1606 and the legend is within a wreath. The reverse has a shield containing two lions, and a possible crown above it. The legend is only partially legible NIS-A.N. SF1005 (Layer fill 0254). Unknown 9. A corroded copper alloy object which is possibly part of a handle or a mount. It is oval in section and curved at one terminal. SF1002 (Pit fill 0206). 10. A fragment of a rectangular whetstone. The piece is significantly worn all over, however the centre of the object is thinner. SF1003 (Pit fill 0262). 11. Two fragments of lava quernstone (1506g). The working surface can still be seen as well as the remains of the central perforation. The fragments are possibly Rhenish, a type of stone imported to East Anglia in the Roman period and then from the Middle Saxon through to the post-medieval period. SF1004 (Pit fill 0262). 12. A worn and broken possible lead window came fragment (13g). It consists of two prongs of?sheet lead and is H in section. SF1006 (Pit fill 0278). 13. A heavily corroded flat and roughly rectangular fragment of iron which is possibly part of a blade (7g). SF1008 (Pit fill 0208). 14. An incomplete and heavily corroded iron spur fragment (46g). The sides are D in section whereas the neck is square and low set. SF1009 (Pit fill 0209). 15. Two small and thin shaped fragments of iron (42g). They are rectangular in section and are possibly part of a blade. SF1010 (Layer 0254). 16. A long bar like piece of iron that is rectangular in section (134g). It is heavily corroded and broken at one terminal. The piece narrows in width at one terminal. It is possibly a door hinge. SF1011 (Pit fill 0207). 17. An L shaped iron bracket/fitting which is heavily corroded (128g). The shaft is square in section and overall it is similar to a hinge pivot. A good example can be seen in the Norwich Households corpus (Goodall 1993, 149; fig 109). SF1012 (Pit fill 0206). 18. Two corroded fragments of large iron square headed nails (280g). The shafts are rectangular in section. Their size suggests the nails had some sort of structural use. SF1013 (Pit fill 0206). 19. Six corroded fragments of large iron square headed nails (408g). The heads have a width of 55cm and the shafts are rectangular in section. The nails are like to have had some form of structural use. SF1014 (Pit fill 0278). 39

6. The environmental evidence 6.1 Faunal remains Julie Curl Methodology The bone in this assemblage consisted largely of hand-collected material, three sieved samples were also examined. All of the bone was identified to species wherever possible using a variety of comparative reference material when required. The mammal bones were recorded using a modified version of guidelines described in Davis (1992). Measurements (listed in the appendix) were taken where appropriate, generally following Von Den Dreisch (1976). Humerus BT and HTC and metapodial a and b are recorded as suggested (Davis 1992). Tooth wear was recorded following Hillson (1986). Any butchering was recorded, noting the type of butchering, such as cut, chopped or sawn and location of butchering. A note was also made of any burnt bone. Pathologies were also recorded with the type of injury or disease, the element affected and the location on the bone. Other modifications were also recorded, such as any possible working, working waste or animal gnawing. Weights and total number of pieces counts were also taken for each context, along with the number of pieces for each individual species present (NISP) and these appear in the appendix. All information was recorded directly into an Excel database for analysis. A catalogue is provided in the appendix giving a summary of all of the faunal remains by context with all other quantifications and measurements. The full faunal data record is available in the digital archive and has additional counts for species groups and element counts. The assemblage provenance and preservation A total of 8784g of faunal remains, consisting of 3434 elements, was recovered from excavations at this site. Remains were retrieved from a total of thirty-two contexts, which included ditch, foundation and layers. The bulk of the faunal remains were derived from pit fills, the material from which amounted to 81% of the assemblage in terms of weight. The bone was recovered with a variety of ceramics, the majority of 40

which were of a medieval date range, lesser amounts of bone was recovered with ceramics of a late to post-medieval date. Quantification of the assemblage by weight, date and feature type is presented in Table *11. Type Finds spot date Building? Ditch Feature Foundation Layer Pit fill Post-hole 11th - 12th 4g 4g Total by date 12th - 13th 184g 4660g 4844g 12th - 14th 54g 57g 111g 12th + 5g 235g 240g 13th 203g 203g 15th - 16th 882g 882g 16th - 17th 14g 14g 17th (19th intr?) 621g 621g 17th? 23g 63g 86g L 18th/19th 261g 261g Post - medieval 10g 1505g 1515g PM/18th/19th 3g 3g Feature Total 203g 189g 54g 882g 308g 7144g 4g 8784g Table *11. Quantification of the faunal assemblage by date, feature type and weight The majority (3202 and 93% by element count) of the 3434 individual bones were recovered from a variety of pit fills, with 2987 of the pieces (86% of the assemblage by count) recovered from four fills in the pit 0269. Quantification of the faunal assemblage by bone count, feature and date can be seen in Table *12. Type Finds spot date Building? Ditch Feature 11th - 12th 5 5 Foundation Layer Pit fill Post-hole Total by date 12th - 13th 25 2987 3012 12th - 14th 35 16 51 12th + 4 79 83 13th 14 14 15th - 16th 88 88 16th - 17th 2 2 17th (19th intr?) 26 26 41

17th? 2 19 21 L 18th/19th 56 56 Post -Medieval 1 73 74 PM/18th/19th 2 2 Feature Total 14 29 35 88 61 3202 5 3434 Table *12. Quantification of the faunal assemblage by date, feature type and count of fragments A total of ninety-seven burnt fragments were seen, of which eighty-one were from the food waste in fills in pit 0269. Fifteen burnt pieces were seen in the foundation cut 0273 (fill 0274) and a single burnt fragment was recovered from the layer 0239. Some low level burning had occurred with a few pieces, but most elements, particularly smaller pieces of bird and fish bone, were fully oxidised and of a white colouring. These burnt fragments are probably from domestic food and cooking waste disposed of with other household rubbish. Gnawing was observed on six pieces of mammal and bird bone. Many of the gnaw marks suggested quite small teeth and may be from either cat or dog, with scavenging by fox or feral cats also a possibility, although the remains are likely to be from meat bones given to domestic animals. Pit 0269 Feature 0269, a 12th to 13th century pit, produced the largest group of faunal remains amounting to 4660g, with a total of 2987 pieces from a combination of four fills; the remains from the hand collected and sieved material from this pit amounted to just over 53% of the bone assemblage by weight. Relatively few bones of cattle were seen, but the remains of sheep/goat and bird were abundant. The sheep/goat consisted largely of a range of good quality meat-bearing bones such as the upper limbs, scapulas and pelvic bones, along with mandibles, one incomplete skull, ribs and vertebrae. Elements from at least four individual sheep were recorded from this pit fill, along with a minimum of two pigs, six individual domestic birds (chicken and geese), along with partridge, duck, cattle and a range of marine and freshwater fish. Bones of Common Frog and Blackbird were also seen in fill 0262, which are possibly natural intrusions, but may well have been included from food waste. All of these remains were in good condition, although many are fragmented from butchering and wear on the smaller, more fragile bones. Many elements, including small fish bones, had undergone a range of burning with some singed black and others fully oxidised and white in appearance. Species, modifications and discussion 42

At least eighteen individual species were identified in this assemblage, five of which were mammal, six of bird, six species of fish and one species of herpetofauna. Heavy butchering and fragmentation of many of the smaller, more fragile remains, resulted around 80% of the assemblage remaining unidentifiable to species, but, where possible, this was assigned to groups such as mammal or bird. The greatest congregation of species were recovered from the pit 0269, which has fourteen species within the four fills, with ten of these species only represented in this pit (see pit 0269). Quantification of the species by NISP and feature type is in Table *13. Species Type Building? Ditch Feature Bird 1 104 105 Bird - Blackbird 1 1 Bird - Duck sp. 1 1 Bird - Fowl 1 1 2 63 67 Bird - Goose 2 17 1 20 Bird - Partridge 3 3 Bird - Woodcock 1 1 Cattle 6 1 6 3 74 90 Deer - Fallow 2 1 3 Deer - Red 3 3 Mammal 7 21 31 60 44 2583 4 2750 Pig 3 35 38 Sheep/goat 4 1 4 9 135 153 Fish 12 68 80 Fish - Cod 6 6 Fish - Eel 8 8 Fish - Herring 87 87 Fish - Salmon 5 5 Fish - Skate 7 7 Fish - Trout 2 2 Herpetofauna 4 4 Feature Total 14 29 35 88 61 3202 5 3434 Foundation Layer Pit fill Post-hole Species NISP Total Table *13. Quantification of the faunal assemblage by species, NISP count and feature type The main domestic species and ages 43

Of the bone identifiable to species, sheep/goat were the most commonly recorded, representing 22% of the bone assigned to specific species. Cattle accounted for 13% of these identifiable remains and pig amounted to 4% of these bones. The domestic fowl amounted to at least 10% of the identifiable remains and goose around 3%, although with the geese there is the possibility that some bone may have been from wild birds, although the metrical data suggests birds within the range for the larger domestic birds. The ages of the sheep/goat in particular varied considerably, with the remains from four individuals in pit 0269 (fill 0262) showing adults with tooth wear suggesting animals of at least 6 to 8 years old along with mandibles from neonatal of less than a month at death (Plate 12). More adult remains were seen in later periods, suggesting a wide range of uses, not least for the provision of fleeces for the increasing wool trade in East Anglia. Juveniles of cattle were only seen in later (PM), which may have been due to the increase in the demands on this species for milk (Albarella, 1997) and the need to cull the young to facilitate this. With the pig, more juveniles were seen throughout, which is typical of most assemblages with this species, which has little use other than for meat, which is at its best when young. With the domestic birds, both the fowl and geese were represented by both adults and juveniles, with a greater number of adult birds. 44

Plate 13. Mandibles of sheep and goat from 1 8years 45

Wild species Fallow Deer antler was recovered from two fills, with a broken palm in the post-medieval pit fill 0206 and a natural shed antler of a two to three year old buck (Plate 13), comprising of the burr with the brow and bez tine, but the palm missing, in the 12th to 13th century ditch fill 0281. Three Red Deer antler pieces (burr, body fragment and tine) were seen in the Sample <3>, from the 15th to 16th century foundation fill 0274. All of the antler in this assemblage had been naturally shed and no obvious working evidence was seen, although the possibility that they were collected for working cannot be ruled out. Given the lack of modification, it is possible that they could have been collected for decorative purposes. Three deer parks belonging to the abbey of Bury Saint Edmunds were known in the area in 1130 (Meeres, 2002) and would have contained and supported the imported Fallow in the early medieval period and possibly Red Deer, so the antlers that are shed in spring would have been easily available. Red Deer, native to Britain, would have been readily available in local woodland. For the first few decades after the introduction, Fallow Deer would have been a rarity, to be admired rather than eaten (Sykes, 2010) but by the 12th century the increase in hunting and consumption saw large numbers being imported and farmed. Wild birds were represented by small numbers of partridge and duck in the12th to 13th century pit 0269 and Woodcock was recorded from the 15th to 16th century foundation fill 0274. Six species of fish were identified, three of marine origin and three freshwater. The marine fish consisted of Skate, Cod and Herring; the Skate and Herring commonly fished off all the UK coastlines at any time and the Cod more commonly caught close to beaches all round Britain during the migration in the autumn. The freshwater species are Salmon, Trout and Eel, all of which may have been caught in local waters. 46

Plate 14 Natural shed antler of 2 3year old buck There are records of large numbers of Eels given by Canute s wife as a gift to the Bury abbey from her estate in Lakenheath and there are records in the Domesday book of 25,000 Herrings a year being sent to the abbey from Southwold (Meeres, 2002). The fish (and bird) would have not only provided a variation to the everyday diet, but would have been an important food to all in Bury for non-meat fasting days and throughout Lent. The 12th to 13th century pit 0269 (fill 0262) also produced four bones from the Common Frog. What led to these frog bones appearing in this pit fill is uncertain; the pit may have acted as a pitfall trap for frogs or it is possible that the frog could have formed part of the stomach contents of the eel in the same pit as frog is a regular food for eels. It is possible that the frog was part of the menu for people on fasting days when almost any species of aquatic origin could be classed as not meat, with beaver and frog counted unequivocally as fish (Tannahill, 2007). Body parts Overall, there is a lower number of primary butchering waste elements compared to secondary meat-production and consumption waste bones, suggesting that at least the 47

bulk of the remains were probably brought to the site as prepared cuts and most of the assemblage represents food waste. The main exception is with the bird and fish remains, which are likely to have been bought whole for preparation in the kitchen. The deer in this assemblage are only represented by antler, with no post-cranial elements seen; these antlers were naturally shed and may well have been collected for decorative purposes alone. Butchering Heavy cleaver chops were seen on the main meat-bearing bones such as the upper limbs, pelvis and scapula from the dismemberment of the carcass and the division into cuts of meat. Vertebrae were frequently seen with lengthways chops from the division of the carcass into left and right sides. Finer knife cuts were seen from the removal of the meat and some filleting was noted on one cattle scapula. Some finer cuts from skinning were noted on the few foot bones in the assemblage which would have occurred from the skinning process and these feet have subsequently been used for food. Relatively little evidence for butchering was seen with the bird and fish bone. However, both are frequently cooked whole or with little butchering required as preparation for cooking and once cooked, meat is easily removed from the bone with little cutting required. With the fish remains, one dermal denticle appears to have been cut, removing the rounded underside, which is likely to have occurred when the skate was prepared and the thorns along the back were cut away. No evidence was found for waste from any bone, antler or other faunal working activities. Pathologies Several pathologies were seen with the domestic mammals and birds that give some indication of the state of the health and husbandry of the domestic stock. 48

Plate 15 Arthritis in a cattle proximal phalange 49

The 12th to 13th century pit 0269 (fill 0262) produced a cattle proximal phalange (Plt. 14) and a distal phalange (both probably from the same animal) that show arthritis. A proximal cattle metatarsal from the post-medieval pit fill 0206 also showed arthritis. Arthritic problems are commonly seen in older animals and in those that are used for traction. A right side sheep horncore from the 12th to 13th century pit 0269 (fill 0262) showed a thumbprint depression (Albarella, 1995) on the dorsal side, close to the base. This depression on the horncore is likely to have been caused by the reabsorbtion of calcium rather than mechanical pressure or injury. The reasons for this reabsorbtion are not certain as there are many factors that are thought to cause this condition, with physical stress, over-milking and breeding thought to be the main causes. It is quite possible that it was caused by malnutrition or over breeding, particularly if the animals in Bury St Edmunds during the medieval period were kept in overcrowded or poor conditions. It is probable that the sheep more intensively reared at this time to provide fleeces for the increasing wool trade. Similar depressions have been relatively frequently seen on sheep horncores from other Late Saxon to Medieval urban assemblages, such as at Oak Street in Norwich (Curl, 1999). Pigs were generally healthy as most were culled when young. A mandible from the 12th to 13th century pit 0269 (fill 0262) showed a well worn third molar (indicating a more mature pig) and a worn and broken second molar. Such a break in a tooth might occur with an older animal or one eating hard food. There appears to be some decay in this broken tooth, although it is not certain if the decay had begun prior to (and was the cause of) the break. Given the very varied omnivore diet, which, with the domestic animal, can include meat waste and bone, it is possible that hard elements or sweet food waste in the diet was the cause. 50

Plate 16. pathological bone growth Chicken pathologies A few spurred tarsometatarsi (TMTs) from male domestic fowl were recorded. One TMT from pit 0269 (fill (0260) showed pathological bone growth around the spur and upwards along the shaft (Plate ; it is probable that this additional growth continued below the spur, but the distal part of the bone is missing. The growth appears to be as a result of infection and may be some form of viral or bacterial arthritis that has resulted in the proliferation of new bone on this specimen. Similar pathologies have been found on two chicken TMTs from Dudley Castle in the West Midlands. One of the Dudley Castle TMTs is on the proximal end of the bone and thought to be osteopetrosis, caused by the avian leucosis group, the other is on the distal end and possibly associated with a fifth toe (Thomas, pers. comm). Two far more severe cases have been recorded on complete TMTs at the Millennium Library Site in Norwich (Curl, forthcoming). It was noted that pathology on both the Forum site and Dudley Castle chicken bones is only present on spurred males, as with the example from Bury St Edmunds. It is possible to speculate that this pathology may be associated with males involved in cock-fighting and maybe these pathologies found have arisen from infections from fighting. It is possible that the nature of cock-fighting would have resulted in excessive 51