Excavation at Prior's Gate House, Rochester 1976-77

EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER 1976-77 A. C. HARRISON, B.A., F.S.A. AND D. WILLIAMS, Dip. A.O. INTRODUCTION The site, which comprises the front garden of Prior's Gate House, 1 lies immediately to the east and north of the only part of the medieval palace of the Bishops of Rochester still surviving above ground (Fig. 1 ). It has generally been assumed that this building is only one side of a larger edifice, having wings projecting from its eastern and western extremities and enclosing a quadrangular courtyard, and this is how the building is shown in the drawing published by Harris in 1719. 2 The opportunity, therefore, of deciding the existence or otherwise of the eastern wing by excavation was welcomed and special thanks are due to the Dean and Chapter, for giving pennission, and to Canon P. A. Welsby, M.A., Ph.D., and to the late Lt.-Col. E. P. Ball, M.B.E., T.D., B.Sc., F.R.I.C., for their sustained interest and support. THE EXCAVATION The results are reported chronologically according to the period to which each feature has been assigned. 1. Romano-British The results obtained for this period tended to confinn rather than amplify those obtained from previous excavations in Rochester. As the south wall of the , city was known to run through the site the opportunity was taken to cut a long section through the earthworks inside it (Fig. 1, a-b). The sequence of construction was the same as that noted further to the east in the Deanery Garden, 3 that is to say 1 Arch. Cam., xxi (1895), 44; xxiv (1900), 65. 2 John Harris, History of Kent (1719). 3 Arch. Cant., lxxxiii (1968), 57-62 and Figs. 4-6. 19 Iv 0 PRIORS GATE HOUSE, ROCHESTER. lon,\ion P1&n I M3 I .,M I I . ... . ... l"Trenchl -, garden wall• . . '. - J Bishops' Palace --7 I ·􀀌-􀀍wM_- ..- -_- ---·1􀀍1-:.i .: Trench 8 hea.rtlf􀁮 ·, · ... 􀀃::;' ;Trench E .mortar floor ,-'i ) e L 1'renchH - -! .. .. Prlors Gate House (lath. cent.) Fig. 1. Site Plan 􀁭 Roman CltyWall = llth• Century 12 th . Century (robbed) I 15th• Century C.l4S0s1 Post-Medieval -----=:-.-- metres O.EW. 1977 PRIOR'S GATE HOUSE, ROCHESTER. metre s - ,,. - ... /􀀎JU:%'.:;/;,.;· -, ;.o;:oll . 2 Rubbl e EtEN􀂟---- . 5 Gr ey She ll Soil . 3C halky Rubble. 4 E 8 Ora n geC I􀂠 e • 6 Charco al. 7 Brow n G􀂡 ey _soi l (Gul ly), 11 Dark Br1>w!; 􀂢 r avel . 9 Brown Sol l. 1 /􀂣 1l(D1splacedWall-Bank) J '􀀈:·""• co,., .:.􀀣: ",;􀀤 ·:·- '·"" :::::;;-:"· " ,...;,,.,.. 18 rown S h ell y S . ar Sh ell y Soll 170 c earth. 2 1D ark Brow n S oil. 91 Ch arcoal. 2 0 Ora . ran ge Cl ayey Grave 24 Ligh t Brow n S ol.􀂤 2 D2 ark Shel ly Sol l 2 n 3 g H e C layey Gra ve l I. 2 011 25 D' · ard LI · . 7Dlrty Brlcke a rth .2 8 irty Brlckearth. 26 A•h • ght Brown Sol l. Trench) 30 R . Undisturbed Brl k , Bones &Sh ells S . ampart.31 R b c earth.29 M . ol l. 34 Clay 35 B u ble.32 Dark B ortar(Robbe 38 Light Brow􀂥 S ol l rown Soil. 36 Medieva l r ;_ wn Soll. 33D ark Bro::- . ,c. 37 Green-Grey Soi l. • , and A). Fig. 2. Sections (Trenches F E D TRENCH D Foundati􀂦n􀂧 Trench EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 that a rampart had been constructed upon the Romano-British topsoil made up of subsoil laminated with layers of clay. This had been cut back in front to allow the wall to be constructed and was subsequently enveloped in the bank of earth piled up against the inside of the wall. At this point the rampart survived to a height of 4 ft. 3 in. close to the wall and extended for 23 ft. 6 in. ·to the north. One point not previously observed was the mound of brown soil at its northern extremity which is probably to be interpreted as a marking-out bank used in its construction. The wall itself, which survived to a height of 6 ft., was again built of coursed ragstone upon a foundation trench filled with rammed flints. I'􀁲 was however noteworthy that its lower courses were very rough indeed suggesting that they were in effect trench-built against the face of the cut-back rampart. It should be noted that the north wall of Prior's Gate House-and therefore the surviving part of the Palace which is in alignment-does not stand on the !Roman wall, as shown in Canon Livett's Map of Medieval Rochester,4 but is immediately to the south of it suggesting that it was no longer visible when the Palace was built. The significance of this is discussed below (p. 25). The wall-bank had been mutilated but enough survived to show that it originally extended approximately 40 ft. to the north of the wall which agrees closely with previous estimates. None of the pottery recovered from either the rampart or the wall-bank was of value for dating purposes, consisting as it did of scraps of first- and second-century pottery of a rubbish-survival nature, but a silver-plated forgery of a denarius of Julia Domna, of which the prototype was issued in A.01 204 (see p. 26), supports an early third-century date for the construction of the wall. 6 The large pit (Fig. 2, Trench F), approximately 6 ft. 6 in. wide and 3 ft. deep, had been partially filled with oyster shells and ash from the north before being levelled with clean brown soil from the opposite direction, presumably to facilitate the building of the wall. It contained no datable material. A section (Trench N, Fig. 3) was cut across the whole width of the city-wall in the south-eastern corner of the site. The wall survived to a height of 6 ft., the width was 8 ft. 6 in. and a stone plinth projected from its base for 1 ft. 6 in. on the south side. As noted elsewhere the wall stood upon a concrete raft resting upon a foundation-trench of rammed flints. Two courses only survived of the original external facing. Inside:the wall all stratification had been destroyed by a stonefilled pit, interpreted as a soak-away, and by the foundations of a later wall. 4 Ibid .. xxi (1895), 17. 5 Ibid., lxxxvii (1972), 124, and Fig. 3. 6 Ibid., lxxxiii (1968), 76. 21 A. C. HARRISON AND D. WILLIAMS 2. Twelfth Century If the results of the excavation in the Romano-British period were largely predictable, the discovery of a very substantial Norman building was quite unexpected. The Roman wall-bank had been levelled to produce a flat surface about 6 ft. above the original ground level. Upon this a building had been constructed of which the west, north and part of the east walls survived to as much as 1 ft. 6 in. above foundation level although the south-east corner and all of the south wall were represented by robber-trenches only, the latter being further obscured by a large medieval pit (M3). The building was long and narrow, externally 39 ft. 9 in. from east to west and is ft. from north to south, excluding buttresses, having an internal width of only 11 ft. 6 in. The foundations contained a high proportion of chalk lumps and on the north were carried down through the made soil of the levelled wall-bank to firmer ground below. The surviving walls were made of ragstone, chalk and flint rubble with a facing of roughly squared ragstone set in a very shelly soft yellow mortar with wide gaps between the courses (Plate I). The quoins and door jambs were of fine Caenstone ashlar with diagonal tooling. A wide splayed doorway was set centrally in the north wall (Plate II) and a small gully at the north-east corner seemed to be contemporary with the building and may have served to carry off eaves drips. The ragstone cobbling, which occurred at a depth of 5 ft., was probably to facilitate the movement of wagons during the process of levelling the wall-bank. 3. Later Medieval At right-angles to the Roman wall and the surviving part of the Palace the east wing projected externally for 38 ft. 6 in. including a l ft. 3 in. buttress and had a width of 31 ft. (Plate III). The walls 4 ft. 6 in. wide on the east and 3 ft. 6 in. wide on the north, were constructed of ragstone and flint rubble with thin courses of flat stones and peg-tiles set in hard creamy mortar. The only ashlar surviving in situ was a block of Reigate stone on the east wall, possibly part of a door jamb. At a point 27 ft. along the east wall the foundations were raised in an arch 4 ft. wide and rising to within 1 ft. 6 in. of the top of the foundation. As below this arch the soil of the Roman wall-bank was undisturbed, it cannot ever have been functional and may have been intended to accommodate a drain. While the east and north walls were of one build, from the north-west corner the foundations of the Norman building were utilised for the first 18 ft. of the west wall and a wall of similar construction to the east wall bridged the gap back to the 22 N w PRIOR'S GATE HOUSE, ROCHESTER. , , , . / _, - . Clay. and Flint _ Foundation ,- ROMAN CITY WALL ----- -- ----- -- ·-- - - Trench N : unexcavated ' ' ' ' Fig. 3. Section (Trench N): J. Topsoil; 2. Gravel and Chalk; 3. Chalk; 4. Mortar and Rubble; 5. Dark Soil; 6. Eighteenth-century Brick Drain; 7. Eighteenth-century Clay Floor; 8. Dark Brown Soil; 9. Peg-tiles (End of Drain); 10. Clay; 11. Clay; 12. Brown Soil; 13. Dark Soil with Stones {Soakaway7); 14. Black Soil; 15. Brown Soil; 16. Ash and Charcoal; 17. Undisturbed Brickearth. A. C. HARRISON AND D. WILLIAMS Roman wall. The floor level was represented by a layer of chalk and mortar which had probably been tiled as numerous fragments of floortiles were found. At the south-east end of the wing was a hearth measuring 4 ft. by at least 6 ft. and constructed of peg-tiles set on edge in clay. This was set into the thickness of the east wall of the wing and directly on top of the Roman wall (Plate IV). The position of the hearth shows that the east wing at least overlapped the existing block, contrary to what appears to have been the situation of the west wing in Harris' drawing. It was not possible to excavate the large medieval pit sealed beneath the floor and its nature was not determined. 4. Later Features (a) Blocking Wall. This ran diagonally from the north-east corner of the Norman building. It was of crude construction with much chalk set in a brown mortar. It crossed an earlier wall, 2 ft. 6 in. wide, associated with a chalk floor containing fourteenth-century pottery. (b) Foundation. At some period after the demolition of the Norman building and partially overlapping its northern side, a foundation 3 ft.thick was constructed of ragstone, chalk and some pieces of firereddened Caen stone set in a poor brown mortar similar to that of the blocking wall. Its extent, date and purpose are unknown. (c) A wall of ragstone and flint observed in Trench P and the foundation observed in Trench N seem to have been boundary walls of a relatively modern date. (d) The stone wall, 2 ft. 6 in. wide and containing eighteenthcentury bricks, built on top of the city wall in Trench N was possibly part of the Grammar School (cf. plan of Episcopal Precinct opposite p. 45, Arch. Cant., xxi (1895) and map in Dennes, History of Rochester). (e) The strip of flint cobbling (Trench P), 7 ft.-wide and extending for at least 15 ft. was bounded by a well constructed strip of peg-tiles 3 ft.-wide set on edge in clay. As this is too wide for a hearth this feature is interpreted as a yard bounded by an open drain or path, terminating in the soak-away noted in Trench N. CONCLUSION I. The discovery of the Norman building, which from the style of the masonry seems likely to be earlier than c. 1150 and may well be connected with the re-building of much of the Priory by Ernulf (I 114-1124), raises several questions. Firstly, what was the purpose of the building itself and of the building to which it was attached? The 24 EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 most likely purpose for a narrow building with substantial walls of this kind would be a porch or stairway leading to a first-floor hall and in that case a parallel is afforded by the Norman stairway still existing in the Cathedral precincts at Canterbury which afforded access to the Aula Nova. The dimensions of that stairway are comparable in size, and the wide doorway would be suitable for a building of this type. The date of the Canterbury staircase is known to be earlier than 1165 as it is shown in the celebrated drawing of Christchurch Priory, prepared in connection with the new water-supply and ascribed to approximately that date, 7 so the suggested date of the first half of the twelfth century for our building corresponds well enough. The purpose of the Au/a Nova is not certainly known, though it is usu􀁦lly considered to have afforded accommodation for the less exalted visitors to the Priory, 'the house of pilgrims and paupers' as it has been described, 8 and it is possible that our feature was attached to a similar building. The possibility is strengthened by its position close to the main gateway9 into the Priory precinct, which is also the position of the Au/a Nova. An alternative possibility is that it was part of the original Bishops' Palace, which could well have been in this position and this point is further discussed below (p. 26). It does not seem possible to reach any more definite conclusion unless and until an opportunity occurs of excavating further to the west. 2. Neither the dimensions nor the position of the east wing of the episcopal palace presented any surprises, and we are here mostly concerned with its dating. It has been argued elsewhere10 that the Roman wall continued to form the southern defence of the city until the latter half of the fourteenth century. It is quite evident that it had ceased to do so when the east wing was constructed and its hearth placed actually on top of its remains.· Furthermore, it would seem likely that a considerable interval had elapsed since the city wall was razed, as otherwise its stump would surely have been used as the foundation of the existing block of the palace (the same argument suggests that the Norman building had been demolished at least in part some time prior to the building of the wing as only on the west do their walls coincide). If, therefore, the Roman city wall was demolished in the middle of the fourteenth century in accordance with Edward III's charter of 1344, it is not unreasonable to suppose that this palace was not built until the following century and two pieces of evidence support this: firstly, the 7 Ibid., vii (1868), 4, and Pl. I, 196. 8 lbid., 149. 9 This was attached to and extended westwards from the southern turret of the west front of the Cathedral. It is shown in Russell's map of the Rochester bridge property dated 1717 (Arch. Cant., xviii (1889), 200) and seems to have been taken down in 1744. 10 Arch. Cant., lxxxiii (1968), 78-9. 25 A. C. HARRISON AND D. WILLIAMS actual construction with roofing-tiles used as levelling courses is not normally found much before the end of the thirteen hundreds but is frequent in the fifteenth century and, secondly, there is a piece of documentary evidence. Writing in 1459, Bishop Lowe headed an agreement 'in our new palace at Rochester'. This has usually been interpreted to mean that he had rebuilt the existing palace, but in view of our recent findings it could well be that he meant what he said and that he had constructed an entirely new building to the south of the previous site. Certainly, the windows of the surviving block suggest work of his date and the numerous fragments of tufa and Caen-stone incorporated in it could well have come from our Norman building. THE FINDS 1. BUILDING MATERIALS Medieval period. Stone. The quoins and door-jambs of the earlier building were of Caen-stone with diagonal tooling. No ashlar survived in situ in the Palace wing except for one piece of greensand, which may have been re-used. Flint, ragstone and chalk were used in both periods for rubble walling. A rounded fragment of Bethersden marble found in the demolition debris of the Palace was probably part of a string course. It had been heavily burnt. Floor tiles. Fragments of plain tiles with brown and green lead glaze were numerous in the demolition rubble of the Palace. Two were examined by Mr. Peter Lorimer who reported as follows: (a) Tile with chamfered corners 7 x 4 in. to chamfer. Brown manganese lead glaze. Fired upright thus causing slight running of glaze. Firing temperature approximately 960-1000° in a clean atmosphere with slight reduction later on. (b) Tile with bevelled sides 4½ in. sq. made of low-fired clay covered with an unvitrified colloidal cream-coloured slip. 2. COINS AND TOKENS (a) As of Vespasian, sealed below Phase I rampart in Trench D. Obv. IMP. CAES.-VESP. AVG. P.M.T.P. Cos IIII CBNS. Laureate head r. Rev. S.C. Aequitas August. stg. 1. RIC 557a. (b) Forged denarius of Julia Domna in a wall-bank of the city wall. Obv. IVLIA AVGVSTA. Rev. PIETAS AVGG. RIC 574. Dr. J. P. C. Kent, B.A., Ph.D., F.S.A., reports: It is a contemporary plated copy of a denarius of Julia Domna, and was probably made not long after the issue of the prototype in A.D. 204. Genuine Severan denarii circulated in quantity down to the middle of the third century, but forgeries tend not 26 EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 to have so long a life; the fact that this one has so much of the plating surviving suggests that you should probably think of it as going to ground by about A.D. 220. Of course, as residual (and undesirable) material, it could have found its way into a wall-bank long after. (c) Ae. 20 of Claudius II, unstratified. Obv.----AVG ---Radiate head r. Rev. DIA(NAE CONSERVATOR!). Stg. I. (d) Ae. 20 of Tetricus II. Unstratified. Obv. ---(TETR)ICVS --Radiate head r. Rev. PIETAS AVG. Sacrificial implements. (e) Siliqua of Valentinianus, in mortar of Norman building. Obs. D N VALENTINIANVS PP A VG. Diademed head r., draped shoulders. Rev. VIRTVSROMANORVM. AQPS (Aquileia) RIC. 41A. (f) Mr. S. E. Rigold; M.A., F.S.A., F.R.Hist.S., reports: This is an Anglo-Saxon 'sceat', or small-flan penny, in design a typical 'Secondary' sceat (Plate V). Dies well centred on the relatively thick flan (both dies anti flan about I· I mm. diameter). Weight l · 17 gm. (18·1 gr.), high for a secondary sceat but below the best standards of the 'Primary' and 'Intermediate' series. Not submitted to quantitative analysis, but superficial appearance suggests that the coin is somewhat (not excessively) debased, with some corroded superficial enrichment of copper, and perhaps plated. Obv. diademed, draped bust r. considerably devolved from the lateRoman archetype (the ear appears to be reversed on the cheek; the hair marked by cross-hatching); in front of a face spreading, lis-like floral ornament. Rev. conventionally called a 'shield'-central boss surrounded by circles, forming a typical Anglo-Saxon cross with pellets on each arm. This type probably originated, and survived on some series, as an obverse. In terms of the 'types' enumerated in the British Museum Catalogue (an unsatisfactory means of classification) this combination could be described as Type 14, variety without LVNDONIA legend (rather than as any sort of 'Mule'). Recently (Brit. Numismatic Journal, forthcoming) a classification by 'Series' (using, but not exclusively, several types) has been proposed, but this belongs to the most difficult series, called L, best represented by a find of light and base coins from the Thames at London, and presumably belongs to an earlier phase of the (rapidly declining?) series, of which some carry a version of the De Lundonia legend, and which take their designs from some of the best and earliest Secondary sceattas. On the hypothesis (tentative, but not more than a decade or so out) that these begin c. 715-20 and that debasement was complete by the 740's, it may be provisionally dated c. 725-30. From redeposited wall bank material (Fig. 2, layer 7). (g) Forged groat of Edward III. Unstratified. Miss M. Archibald, M.A., F.S.A., F.M.A., reports: The coin is a medieval forgery purporting to be a groat of Edward 27 A. C. HARRISON AND D. WILLIAMS III of the Pre-Treaty Series C, 1351-2, struck at London. It is identified as this series by the annulet stops, transversely-barred Ns and the wedge-tailed R. The initial cross, which is also diagnostic, is scarcely legible here on either side. Comparison with the official coins reveals certain unacceptable deviations from the norm. Although closely resembling the official fount, the letters of the outer legend are abnormally large and all the letters are somewhat crudely formed-the letters T and G are particularly noticeable. The crown is narrower and more upright than on the official issues and the shape of the hair is abnormal. The form of the shoulders in which the neck has almost disappeared is also unacceptable on an official coin. The tressures on the obverse are rather flat and irregular and the fleurs and other details do not conform to the official norm. The weight of the coin is 2·62 gm. = 40·4 gr. which is too low for a coin which purports to have been issued at a period when the standard weight was 72 gr. Clipping would not account for a loss of between a third and a half of the issueweight. Although the dies which struck this coin were made from a set of punches as were the official issues, it is none the less a forgery. The manufacture of forgeries in this way was a regular practice. An interesting group of similar forgeries struck c. 1465 was included in the Wyre Piddle hoard (NC 1970, 133-62). The date of production of this particular forgery is more difficult to determine since it could have been made at any time during the currency of the prototype which extended into the mid-fifteenth century. The narrowing of the crown is indeed characteristic of the Lancastrian period but the letter-fount used on this coin follows the Edward III issues so closely that the prototype must have been found in good enough condition for the form of the letters to be clearly visible. A date of manufacture before the end of the fourteenth century therefore appears to be most likely. (h) Nuremburg Jetton. Unstratified. Obv. HANNS KRAUWINCKEL IN NUR Flower in centre, three fieur-de-lys and three crowns surround. Rev. GOTTES GABEN SOL MANLOB Orb in trefoil. 3. SMALL FINDS (Fig. 4) l . Spindle Whorl. Made from a sherd of Romano-British black coarse-ware. From Roman wall-bank. 2. Fragment of first-century brooch, cf. Collingwood Group F, no. 7, the so-called "Colchester" brooch. From rampart-bank. 3. Fragment of iron spear-head. From rampart-bank. 4. A fragment of opus sectile found in the redeposited wall-bank material has been identified as follows. The bulk of the specimen 28 EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 0 '• • .. ,t ... • 􀀆 0 􀀂 @ '16 14 IS 17 I 4 18 - 19 Fig. 4. Small Finds (Scale:¼) 29 20 A. C. HARRISON AND D. WILLIAMS consists of a sliced calcite vein, but there is a small corner of fossiliferous limestone conceivably carboniferous limestone. 5. Bone Comb. The two outer plates decorated with incised lines and attached to centre-piece with iron rivets. Probably late Roman cf. Richborough IV, Pl. LVI, no. 266, but it could possibly be Saxon. From brick-earth seal of pit (Fig. 2, layer 20). 6. Iron key. Early medieval or even pre-Norman type. Sealed beneath floor of Palace wing. Cf. London Museum Medieval Catalogue, 135, Fig. 42, no. IA, and Pl. XXIX, no. l. 7. Bone Comb. Probably medieval cf. London Museum Catalogue, 291, Pl. IXXXVIII, no. 1. From demolition rubble of Palace wing. 8. A fragment of architectural ornament of Reigate stone. It was probably part of a finial. The hole drilled for the metal connecting rod contained traces of lead. Unstratified. 9. Bronze book-clasp. Decorated with five Tudor roses and nine acorns and a zig-zag border. Trace of iron plate on back and fragment of leather. From demolition-rubble of the Palace wing and datable to the Tudor period. 10. Bronze fleur-de-lys. Decorated with a zig-zag pattern with a central boss and three pin-holes for attachment and thought to be intended to decorate and strengthen the corner of a book cover. Late medieval from the demolition-rubble of Palace wing. 11. Bronze spur. Mr. M. I. Moad reports: This rowel spur, is of brass and generally well preserved, Jacking only the rowel and suspension buckles. The spur has a short facetted neck with a down-turned rowel box, the total length of both neck and rowel box being 1 i in. The slot in the rowel box would indicate that the missing rowel was approximately I½ in. diameter. This rowel would almost certainly have rotated on a steel pin, and it is the corrosion of these ferrous pins which accounts for almost all the losses of rowels in excavated spurs such as the example under discussion here. The sides of the spur curve gracefully round from the well-formed heel plate and finish in flattened terminals, each pierced with a pair of holes to which the buckles and straps were fastened which served to attach the spur to the riding boot. Clear parallels for spurs of identical form to the Prior's Gate House specimen may be found in two examples in the R. L. Scott Collection now at Glasgow City Museum (Scott Catalogue, nos. 110 and 114);11 although both of these examples are of iron with silver decoration. The form is attributed to the mid-seventeenth century. Unstratified. 12 and 13. Fragments of two brass candle-wick trimmers. One 11 Catalogue of the Co/lec1ion of European Arms and Armour formed at Greenock by R. L. Scott. Described by Felix Joubert, F.S.A.Scot., 3 vols., 1924. 30 EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 stamped with jleur-de-lys and initials C.H. Probably eighteenth century. Unstratified. 14 and 15. Two lead tokens. Possibly connected with hop-picking and of eighteenth-century date. Below floor in Trench N. 16-19 Hones 16. Micaceous mudstone. 17. Micaceous grit or sub-greywacke. 18. Greywacke grit, with two distinct grain sizes (sand and silt) and abundant micaceous aggregates. 19. Dark fine-grained biotite bearing phyllite or low grade schist. All these are found in the glacial drift of Essex and East Anglia, but the ultimate sources are the palaeozoic geosynclinal areas of the north and west of which south-eastern Scotland is perhaps the most probable. 4. MEDIEVAL JUGS. By P. J. Tester, F.S.A. {Figs. 5-6) Group I. Five glazed jugs sufficiently similar in ware and general form to be considered as probably contemporary. A very close parallel for no. 1 is illustrated in B. Rackham's Medieval English Pottery (2nd. Ed., 1972), no. 25 considered by Mr. J. G. Hurst to be fourteenth century. Similarly, a jug from London of this shape and rim-form is figured in the London Archaeologist, i, no. 16 (1972), 373, no. 1, and there dated to the fourteenth century on the basis of its heraldic decoration and other evidence. This applied decoration on nos. 2 and 3 is likewise typical of that period. 1. West Kent conical jug of hard red fabric; buff slip covers the exterior over which has been applied an uneven speckled green glaze with much of the yellowish glazed slip showing through. Thumbimpressed base and strap handle. Decoration of combed lines on body in attempt to achieve graffito effect. 2. Red ware, similar to the last, coated externally with slight slip over which is an even speckled green glaze. Round-sectioned handle and decoration of applied moulded strips with centrally indented pellets. There are signs of trimming round the base while the vessel was in a leather-hard state. 3. General characteristics including slip and glaze similar to the last. Decoration of narrow applied strips. 4. Red ware; rilling on neck and body; strap handle and pinched lip; green glaze over light slip. 31 w N ' ' ' ' ' \ I l I I ( I J I I I I I I I : I I l Fig. 5. Medieval Jugs (Scale: ¼) • / ' ' n:i"-,L"'_""_=~-,,' ,,,,. .... ',,\ '\\ ' \ 3 \ \ \ I I I I I I \ I w w "" .. 4 􀀊,/: ..,,: :.=\ •... ' ,Th ! • I - .. .,,. .. , . -· .:.--- ,, /,' ,;,,',, , ,,,,., ,::: ( i' ,, ,, \:". . ,,,, ,,.,. , •' 7 6 . F1. g . 6. Med1eva I Jugs (Scale: ¼) ' '•􀁀·,- ...- ,,,:.:;-- --; .􀀿􀀿 ::-, :::. · ...·./ \ --- A. C. HARRISON AND D. WILLIAMS 5. Red fabric reduced to grey on interior surface; pinched lip. Patchy dark-green glaze over light slip. Decoration of combed lines similar to no. 1. Group II. Two grey unglazed jugs, probably to be considered as contemporary with the finer wares 1-5 above. 6. Hard grey gritty ware with rilled neck and stabbed handle. Cf. a glazed jug from London figured in the London Archaeologist, ii, no. 14 (1976), 361, no. 2, and described as possibly a west Kent product of the fourteenth century. 7. Part of a grey ware jug with globular body, less gritty than the last. Thumbed base and bands of combed decoration around body. 1- - - -- - - - - - - - -, 1------ \ I ------ \ I --- -·------- .,_...._.... 1 in Fig. 7. Fragment of Anglo-Saxon Sculpture (Reconstruction) 5. A SECOND ANGLO-SAXON SCULPTURED FRAGMENT FROM ROCHESTER. By Dr. M. J. Swanton, B.A., Ph.D., F.S.A., F.R.Hist.S. The small fragment of sculptured stone from the Prior's Gate House site (Fig. 7, Pl. VI) is only the second known piece attributable to the 34 EXCAVATION AT PRIOR'S GATE HOUSE, ROCHESTER, 1976-77 pre-Conquest church at Rochester. 12 It measures approximately 6 x 8 x 5 in. One broad surface bears ornament; two adjacent faces are roughly dressed, the others broken. The ornament consists of one corner of an abstract panel: a border of five parallel bands, and the terminal section of a single-strand interlace of the kind characteristically found decorating eighth- and ninth-century southern English manuscripts and which has affinities in contemporary Anglian sculptures in the North of England. The design seems competently handled, although insufficient remains to comment further on the nature of the plait. Ornament of this type is found on both cross-sculpture and architectural sculpture at this time, and the Prior's Gate House piece could well represent either. The outer, i.e. decorated, face is reddened as though the monument of which it formed a part had been subject to fire at some time. There is reason to believe that the church at Rochester became progressively dilapidated at least in Anglo-Saxon times. And like so much AngloSaxon sculpture, the monument was no doubt considered demode and reduced to rubble in post-Conquest times, to be used in the foundation of the Norman building on the site of which it was discovered. Like the earlier-known fragment found in the foundations of the Norman cathedral, the material is not local, but Jurassic limestone: a fine oolite with a sparry (i.e. calcite) matrix. It could well have come from the Barnack region of Northamptonshire, although grains of iron salts present suggest that it should perhaps be assigned to a quarry in the Cotswolds. ACKNOWLEDGEMENTS The hard work of the excavation was carried out by many volunteers, of whom the following call for special acknowledgement by reason of their sustained efforts: Mrs. D. Williams, Miss D. Gange, and Messrs. R. A. Baldwin, P. Bradford, R. Green, J. Henley, T. Ithell, P. R. Payne, R. Pink, R. Stephenson and H. F. L. Williams. We are most grateful also to Mr. R. J. Cruse and Dr. P. Hayes, for their work on the pottery, to Mr. J. D. Brand, for advice on the coins, to Messrs. S. E. Ellis, D. Moore, B.Sc., and A. Sutton, for identifying the hones, and to Miss M. Archibald, M.A., F.S.A., F.M.A., Dr. J.P. C. Kent, B.A.,P h.D., F.S.A., Mr.P . Lorimer, Mr. M. Moad, Mr. S. E. Rigold, M.A., F.S.A., F.R.Hist.S. and Dr. M. J. Swanton, B.A., Ph.D., F.S.A., F.R.Hist.S., for the specialist contributions that bear 12 M. J. Swanton, 'A pre-Conquest sculptural Fragment from Rochester Cathedral', Arch. Cant., lxxxviii (1973), 201-3. 35 A. C. HARRISON AND D. WILLIAMS their names. Above all our thanks are due to Mr. P. J. Tester, F.S.A., who, besides supplying the report on the medieval pottery, gave us invaluable advice and support throughout the project. 36 PLATE I Trench F: North-east Comer of the Norman Building: Comer of the Fifteenth-Century Palace Wing in the Angle of the Walls. PLATE II Trench I: Splayed north Doorway of Norman Building. PLATE Ill Trench A: Junction of Roman City Wall (left) and east Wall of Palace Wing. PLATE IV Trench A: Tile Hearth set into the east Wall of the Palace Wing. PLATE V 􀀂\ • Anglo-Saxon sceat (Scale: t) PLATE VJ Fragment of Anglo-Sa􀀕on Sculpture.