Excavations at Eccles, 1970

EXCAVATIONS AT ECCLES, 19701 NINTH INTERIM REPORT By A. P. DETSICAS, M.A., E.S.A. INTRODUCTION THE ninth season of excavations at the site of the large Romano- British villa at Rowe Place Farm, Eccles, in the parish of Aylesford (N.G.R. TQ 722605; O.S. 6-inch Sheet TQ 76 SW), was undertaken by the Eccles Excavation Committee, under my direction, at weekends from early April till October, 1970. A continuous fortnight's work was also carried out in August during which period a training course in Romano-British archseology, jointly sponsored by the Committee and the Kent Archaeological Society, was based on the site. Once again, I am indebted to the landowners, Messrs. Associated Portland Cement Manufacturers Limited, for readily granting their permission for this work to be continued. My thanks are also due to their tenant farmers, Messrs. A. A. and A. C. Southwell, who again made us very welcome on their land. This work was financiaUy supported by grants from the Kent Archseological Society, the British Academy, the Society of Antiquaries of London, the Haverfield Bequest of the University of Oxford and private donors to whom I am very grateful. Considerations of space alone prevent me from mentioning individuaUy the many volunteers who made this excavation possible; however, I must make an exception in the case of the foUowing for their sustained support throughout a long season's work: Misses H. Balsdon, B.A., and S. M. Emerton, B.A., and Messrs. R. W. Chapman, B.A., 0. K. Hales, A. C. Harrison, B.A., E.S.A., T. Hetherington, T. ItheU, B.Eng., W. A. Knowles, R. Lowson, B.A., C. E. J. Martin, and P. Thornhffl, B.A. I am also grateful to Mr. I. J. Bissett, for drawing the pottery and the smaU finds; Miss D. Charlesworth, M.A., E.S.A., for reporting on the glass; Miss S. M. Emerton, B.A., for much help with field drawing and processing the pottery; Mr. R. G. Eoord, for the monochrome sitephotography and the prints illustrating this report; Mrs. K. E. Hartley, 1 Arch. Cant., Ixxviii (1963), 125-41; Ixxix (1964), 121-35; Ixxx (1965), 69-91; xxxi (1966), 44-52; Ixxxiii (1967), 162-78; Ixxxiii (1968), 39-48; Ixxxiv (1969), 93-106, and lxxxv (1970), 55-60.1 am glad to record my personal appreciation to Professor S. S. Frere, M.A., F.B.A., F.S.A., for his interest in and support of this work over the years, and for reading this report in draft form and suggesting a number of improvements which I have incorporated. 25 A. P. DETSICAS B.A., for reporting on the mortaria; Dr. J. P. C. Kent, B.A., Ph.D., P.S.A., for identifying the coins; and Mr. R. P. Wright, M.A., E.S.A., for reading and reporting on the graffiti. FinaUy, I must record my appreciation to my wife and to my pupU, Miss D. E. Rooke, for shouldering the laborious task of initiaUy processing the bulk of the pottery. THE EXCAVATION The main objectives of this season's work were to continue the examination of features aheady partly explored in the previous years' excavations and examine the area to the north-east of the viUa's Uving accommodation. Period IV, c. A.D. 65-120: The Living Quarters Eour new trenches were cut across the ahgnment of Boom 94 and confirmed the evidence secured in earher trenches; aU that had survived of this feature were clear indications of the construction trenches cut for the loose rubble footings of this room. As noted in previous years, the whole area to the south-east of the viUa's facade had been disturbed at great depth by an extensive excavation the fiUing of which yielded a few fragments of medieval pottery and tile; this excavation had removed practicaUy aU Romano-British stratification, though enough survived to underline that this feature (Room 94) was of rather flimsy construction. In the south-easternmost of these new trenches clear indications were found that the north-east waU of this room was to be located immediately beyond the excavation as the loose rubble laid down for the foundation of the floor was seen to be gradually rising. Further examination of this structure showed that it was constructed on an ahgnment parallel to that of the main block of the viUa in this period; this reinforced in turn the evidence of the stratified pottery which indicated that the main block and Room 94 belong to the same building period. Though the purpose of this long structure is not yet clearly understood, it is now undoubted that, in the original plan at least, the villa faced to north-east,2 with this long room occupying the rear of the viUa; a probable interpretation of its use, which can find some support in the flimsiness of its construction, is as housing for domestic staff and farm labourers as weU as for storage-rooms and workshops. DemoUtion ofthis structure seems to have been completed by about A.D. 180 during the period of construction which saw the buUding of the southeast wing of the viUa and the re-organization of the area formerly occupied by Room 94 as an internal courtyard facing the main block and flanked on one side by the baths and on the other by the south-east 2 As already proposed in the previous report; Arch. Cant., lxxxv (1970), 57. 26 EXCAVATIONS AT ECCLES, 1970 wing which stiU awaits excavation. This is certainly borne out by various layers of domestic rubbish used as make-up below the flooring of the fronting corridor, Room 93; a prehminary examination of the pottery found in this deposit and in the debris resulting from the demolition of Room 94 strengthens this conclusion. Further work in the area of the main block has estabhshed the full size of Boom 114 at 22 ft. by 24 ft. 6 in. (6 • 71 m. by 7 • 33 m.),3 and added one final room, Boom 116, to the central range; this last room measured 19 ft. by 24 ft. 6 in. (5-79 m. by 7 -33 m.). No floors survived below the shght depth of topsoU, but their make-up layer consisted of a fairly thick deposit of yeUow clay which suggests that the floors would have consisted of yeUow mortar and tUes as known in other parts of the villa.* The north-east waU of Room 116 is clearly not part of the original build; apart from the fact that it was constructed of flint set in yeUow mortar as opposed to the ragstone construction of aU other known walls of the villa's hving quarters, it is clearly abutted, at both its north-east and south-west extremities (Plate IA), on to the main waUs of the central range of rooms which terminated in this area; the south-west waU extended a httle more than 1 ft. (0-30 m.) further to south-east than its counterpart. Both ends of these walls are regularly finished and show very clearly that the original house ended at this point; this is further supported by the absence of any construction or robber trenches in the area immediately beyond the ends of these walls. This, however, poses the question of what sort of construction may have existed on the line of the later flint waU, and further examination is needed here to provide a reasonable explanation: it could be that the flint waU superseded a timber construction of sleeper-beams and wooden posts forming an entrance to the central range; alternatively, Room 116 may have been a smaU open courtyard leading into the central range. Period V, c. A.D. 120-180: The Living Quarters Renewed examination along the Une of the villa's later fronting corridor (Boom 93) has been carried out and its north-eastern limit has now been reached, demonstrating a total length of 217 ft. 6 in. (66-33 m.), or a frontage of 285 ft. 6 in. (77 m.) if Room 37, which continues the fronting corridor, is included. Very shght traces of the floor survived below the ploughsoU and consisted of yeUow mortar on a make-up deposit of yeUow sandy clay; it is Ukely that tUes were set on this mortar as surfacing for the corridor.4 3 The possibility of a partition-wall across this room sub-dividing it into two smaller ones, as in the case of Room 108, will be explored in 1971. 4 Arch. Oant., lxxxv (1970), 59, and PL IIA. 27 A. P. DETSICAS Evidence was also found, in the extreme north-east area of this room, for raising the floor-level and re-flooring in a manner similar to the original scheme. The south-west waU of this room ends before its counterpart to north-east; its construction trench had not been cut beyond the point shown on the main plan (Fig. 1), whereas the main range of rooms projects further to north-east by the distance required by the width of a corridor turning south-west. This extension did in fact take place, but only when the corridor was re-floored, as can be seen by the fact that the north-west and south-east (Plate IA) waUs of the unexcavated south-east wing are abutted on to pre-existing walls.5 Period VI, c. A.D. 180-290: The Living Quarters To south-west of the viUa re-flooring of the fronting corridor (Room 93) and its extension to south-west to form the new south-east wing were completed during this building period. Both walls have partiaUy survived; the north-west waU was rather shghter (1 ft. 10 in. (0-55) m.) than the south-east one and was rendered with painted waUplaster internally,6 i.e. on the side facing into the courtyard, and presumably on the internal face of the corridor as well, though no wallplaster survived in situ; the south-east wall survived only below offset level which accounts for its greater width. Both walls were built of ragstone set in yellow mortar. Continued examination to the rear of the villa has shown that the rear corridor (Boom 104; Fig. 2) extended, as was to be expeoted, along the full length of the villa, giving it a total length of 244 ft. 6 in. (74-57 m.). A small wing, demonstrably of the same buUding period as the rear corridor, projected to north-east beyond the north-east waU of the corridor; it consisted of three rooms: Boom 118 measured 17 ft. 6 in. by 20 ft. 6 in. (5-33 m. by 6-25 m.) and was constructed, as was the rest of this wing, of ragstone and yellow mortar, its walls being 2 ft. (0-61 m.) wide above offset level—nothing survived of its floor, except a few undisturbed patches of yellow mortar to suggest a tiled floor; Boom 119 (Plates IB and IIA) measured 17 ft. 6 in. by 14 ft. 6 in. (5-33 m. by 4-42 m.), and the smaUer Boom 120 (Plate IIB), which measured only 4 ft. 6 in. by 10 ft. (1 • 37 m. by 3 • 05 m.), opened originally off the larger room (Fig. 1, inset). It is clear that the original intention was to provide these two rooms with a floor suspended over a pillared 6 The discrepancy in the length of these two walls is probably due to the faot that the south-western wall was originally oonstruoted of sleeper-beams whose construction-trench clearly did not penetrate deep enough into the subsoil to survive later rebuilding. 0 For an identical situation at the exaotly opposite part of the courtyard, see Arch. Oant., Ixxxiii (1968), 45. 28 PLATE IA at **/ . OSMF • df'" ft- Photo: It. (1. Foard The Eastern Corner of the Villa, showing Walls abutted on to the Northeastern Wall of the original House (partly robbed). PLATK IM Photo: R. d. Foord Rooms 119 and 120: General View from the North-east. [face p. 2S PLATE IIA , -*... Photo: R. Q. Foord Rooms 119 and 120: General View from the South-east with, in the Foreground, the Stokehole-flue into the channelled Hypocaust, and in the Background, the eastern End of Ditch VII. PLATE IIB Hi£ttyM| Photo: R. G. Foord Room 120, from the South-west, with Flues, in the Foreground. \UiLIi Ditch Vlil Pit &j$\ JrBwIal Burial [Burial Graft/ Bufiai -.•W*. ! V Diteh W. Oitch III •iVZm. WTmwmm, ^ f ^ ^ p IsfiMiit ^w^^^psssEir:zz» 5 ^ ^ ^DiKtl 3 UGDD''*?^ \ aaff'i w ••' i S «M fcivifi"'"^-11-! Burial ^Cobbling f:-Lj Ttsstllation inftrrtd. \_\Opus signinum bidding Ptriod V Robbtd or inftrrtd. \^}ptriod Vlt Ptriod VI- EZ3'**"o r f VtbRobbtd or inftrrtd Ptriod VI.Robbtd or inftrrtd. $£feriod VII SU Ptriod IV: fiobbtd or inftrrtd WtP'riod Via ^ * ' * * ™ R°tb*d °r " " " " " Ptriod V. utfAPtriod Vld.Rabbtdor intarrtd. [~iPtriod Vila CZM*A Pv.t-hoit. o-j: Iron collars in woodtn piptlint O Ptnod Vila Robbtd or inftrrtd. [j;;i;;j Ttsstllation in situ g g Ptriod III Ptriod III Robbtd or inftrrtd Ptriod IV yttzmzvm WyWfMiMfmzm!Zffwmwy//'Z;.iZi>m!mmsmmmm'/j0i ?^//jwmcv/',. VKW%:W^. /wz/jmrmffi^m^mz^m. FIG. 1. [face p. 28 EXCAVATIONS AT ECCLES, 1970 hypocaust; indeed, Room 120 (Plate IIB) had been excavated to a greater depth than Room 119 and a partition wall had been constructed between these two rooms and provided with two flues, each 1 ft. 6 in. (0-45 m.) wide. However, before the construction was completed, this plan seems to have been abandoned in favour of a channeUed hypocaust. The deeper excavation for Room 120 was fiUed in with ragstone rubble and building debris, containing decayed painted waU-plaster, and an opus signinum floor, some 2 in. (0-05 m.) in thickness, was laid over the whole area occupied by these two rooms to form Boom 121; in the original Room 119 this floor was laid directly upon the Romano- British ploughsoil which had not been removed to anything hke the depth in Room 120. On top of this floor was bunt a channeUed hypocaust: it consisted of two main flues (1 ft. 6 in. (0 -45 m.) wide) at right angles to each other, and several secondary flues (1 ft. (0-30 m.) wide), disposed more or less symmetrically over the whole area. The walls of the channels were of clay-bonded bonding- and pila-tiha and showed signs of some burning. The main north-east flue projected into the area of the origmal Room 120 and opened out into a Y-shape; at the opposite end, it formed an inverted T-junction with another wide flue. Presumably, openings through the channel waUs were provided to allow heat to penetrate into the spaces between the two main southwest channels but none of these survived demohtion; in fact, very httle of the channel walls survived but their position could be accurately estabhshed as only where channels had been aUowed for was the opus signinum smooth whereas the areas upon which the tiles for the channel waUs would have been set were roughened. A regular opening through the south-east waU of the hypocaust had been provided for the heat generated in a rather small stokehole located immediately outside this wall; this stokehole must have been in the open air as no enclosing walls were found to suggest a more substantial furnace-room. The stokehole-flue had been lined with chalk and floored with fire-reddened and cracked tiles. A small deposit of wood ash had accumulated in this flue and some soot at the end of the channel immediately opposite the stokehole-flue, but the general impression, gained from the condition of the surviving tiles used to build the channel walls, is that the hypocaust could not have been in use for any great length of time before the end of the villa's hfe and, in consequence, must be one of the latest structural additions. Whatever the intention behind the original construction of this wing when the rear corridor was added to the plan of the house (and the depth below Room 120 does seem to suggest an intended plungebath), a likely interpretation of the channeUed hypocaust modification is as a corn-drier; it may be also suspected that this use was resorted to in the final period of building when the rear corridor was sub-divided 29 A. P. DETSICAS into smaUer rooms and another channeUed hypocaust was inserted into Room 95.7 However, the absence of any sealed finds below the floor makes this impossible to substantiate. Period VII, c. A.D. 290-400: The Living Quarters Continued examination in the area of the viUa's rear corridor has estabhshed the dimensions of Boom 115 at 53 ft. 3 in. by 11 ft. 6 in. (16-24 m. by 3 -37 m.). A further room, Boom 117, had been made by a partition-waU across the corridor, and it measured 18 ft. by 11 ft. 6 in. (5 -40 m. by 3 -37 m.). No floors survived, but it is very hkely that they would have been of mortared tiles as in rooms further to north-west.8 The total length of the ditch (Ditch VII) to east of the rear corridor was cleared; it measured a total of 51 ft. (15-55 m.) and stopped just short of the north corner of Room 121. I t had been fiUed with domestic refuse in diminishing quantities the furthest it ran away from the rear corridor; the eastern.third of its filling consisted almost exclusively of demolition debris. The pottery and coins contained in this fiUing make it clear that, whatever the purpose for which this ditch had originally been cut, it was later used as a rubbish-tip, certainly during this period, and the fact that the ditch stops short of Room 121 must mean that it could not have been cut before c. A.D. 180 (Period V) at the earhest. Two burials had been inserted into this ditch after it had been fiUed; no grave-goods were found in undoubted association with them. They may be contemporary with the further burials described on pp. 31-2. Periods V-VII, c. A.D. 120-400: The North-eastern Area A large area to the north-east of the viUa's later rear corridor was mechamcaUy stripped of its topsoil and then sectioned. The whole area was founi to be badly disturbed by several pits cut for the deposition of domestic refuse, but some evidence also survived to show that, originaUy at any rate, the Romano-British surface here had been laid with yeUow mortar, varying in thickness but on average 2 in. (0-05 m.) thick; this mortar flooring extended for the entire width of the excavated area. The north-westernmost of three narrow trenches cut in this area provided evidence also for gravelling but, owing to the subsequent disturbance, it was not possible to determine whether this layer of gravel had been laid on top of the mortar flooring further to north-east than shown on the main plan (Fig. 1); on the other hand, the mortar flooring was not present below the gravel further to south-west than the junction of these two deposits. No gravel was present in either of the two 11bid., Ixxxiv (1969), 100-1. 8 Cf. note 4, above. 30 EXCAVATIONS AT ECCLES, 1970 trenches cut further to south-east but, as this area had been cut into by pits of fairly recent date (used for the burial of ragstone, clearly brought up by the plough, and dead farm animals), it was impossible to determine whether this graveUed area had in fact extended much to south-east or was a narrow strip used as a path. Immediately to north-east of the mortared area, another trench cut across the line of a ditch, Ditch VIII, which, hke Ditch I I I further to north-west of this area, may first have been cut as a boundary ditch and later used for the deposition of domestic rubbish and buUding debris; its filling contained pottery which is predominantly of secondcentury date. Whatever the purpose for which the yeUow mortar had been laid in this area, at a later stage, clearly coinciding with the re-orientation of the villa (cf. pp. 26, 33), the whole area was spread over with a layer of dark grey soil, containing much organic material and domestic refuse but Uttle buUding debris, which suggests dehberate spreading on what must have been turned into a kitchen garden. A fairly large pit, cut later through this dark grey soU, produced some late pottery which would indicate that the use of this part of the site as a vegetable garden lasted probably until the end of the Romano-British occupation of the site. The Cemetery Though a few scattered burials had been found at other parts of the site in earher seasons of excavation9 and pointed to the well-known practice of burial in the unproductive ground of a ruined building, evidence was secured this year for what appears to be a fairly large cemetery. Apart from the two interments in the hne of Ditch VII (cf. p. 30), an indeterminate number of other burials has been exposed immediately to east of Room 121. One of these burials had been inserted just to east of the stokehole, a second had necessitated the removal of part of the waU immediately beyond the south-west cheek of the stokehole-flue into the channeUed hypocaust. The greatest number lay, however, in the area further to east10 where the ground had been excavated to some depth for the interment of several burials: the first few had been interred at subsoil depth and covered with a shght layer of dark soU and a further two layers of burials had been laid in similar fashion and above each other to within about 1 ft. (0-30 m.) of the modern surface. Although these burials had been interred in a regular manner, no 0 Arch. Oant, Ixxviii (1963), 140, and PI. VIII; Ixxix (1964), 130; and lxxxv (1970), 60. 10 A trenoh in this area is not shown on the main plan (Fig. 1) for reasons of economy of spaoe. 31 A. P. DETSICAS grave-goods were found anywhere in unmistakable association with these inhumations, and the few Romano-British sherds present in the soil covering them are obviously rubbish survivals and of httle chronological significance; moreover, some of these burials were not intact, in spite of their being inserted deeper than the plough can reach, though this is probably due to their being disturbed by later interments: hence, the total number of burials in the excavated area is at present not known. The dating of this cemetery is also indeterminable: their insertion on top of the demohshed Room 121 and in area so close to the villa demonstrates that they may be sub-Roman,11 the laying out of most of these burials in an approximate east-west ahgnment is an indication of Christian practice and the depth of soil between subsoU and modern surface points to some considerable lapse of time between the end of the villa's occupation and the interment of these burials; beyond this, however, it would be unsafe to conjecture. Dating No new evidence has been recovered this season to cause any reconsideration of the dating sequence provisionally proposed in earlier reports; quite to the contrary, all evidence secured in this year's work, whether structural or stratified, has supported this chronology. Summary and Discussion (Fig. 2) With the completion of the excavation of the villa's main range of rooms, it can now be seen that the original structure was a strip house, some 248 ft. (75-64 m.) long, consisting of twelve rooms with a corridor to their rear affording access from one end of the range to the other and facing to north-east, i.e. in the approximate direction of the Roman road from Rochester to the reputed settlement at Maidstone and beyond, on the high ground of Blue Bell Hill above the villa; whether this points to a connection between the original foundation and the Roman town of Rochester is a matter for conjecture. The house seems to have been entered from the south-east, through Room 116 which is provisionally interpreted as an entrance-hall, and its best furnished rooms, laid at least with tessellated pavements,12 were concentrated at the opposite end of the house. 11 Professor S. S. Frere, M.A., F.B.A., F.S.A., has referred me to the very late cemetery at Cirencester. 12 I t is not impossible that the remnants of tessellation in Booms 89 and 90 are the surviving parts of the borders surrounding mosaic panels, that the opus signmum floor in Boom 91 had also a mosaio pavement laid on it; many lumps of mosaic, used as hard-core in later construction trenches (Arch. Oant., Ixxxiv (1969), 100), establish beyond doubt the existence of mosaic floors whioh were removed in the later stages of the villa's life. 32 E C C L E S R O M A N V I L LA !!!!!H!!!!!!:l!!:!!!l;n!i!!ii;!i!!i!:!n::i :..::::-:™:i;:H;;H;;!;s:;dil;;;;:!::;:::i:ii: . : : J M : : : : : : : : : : ; : : : : : : : : : : : : - : : " :: ;j:;i;:;:;sj:;;|njTjn;n:;;i;::i!;:!! II' C.A.I). 12U-ISH ::;;:;!;;i;;ii;;;;i:!i:i;;H!;: illlllli III c.A.I). ISO-29(1 \ :c A l ) . 29U-41H) Fio. 2. Ifacep. 32 EXCAVATIONS AT ECCLES, 1970 Behind the villa and less than 30 ft. (9 • 15 m.) to its south-west was erected another building (Room 94); it was less weU buUt than the main house and could have housed domestic staff and farm workers, provided accommodation for domestic and/or farm animals and herds as well as storage-rooms, perhaps even farm workshops; the length of this structure, even though shorter than that of the main house, is far too long to have been used for housing only. Definite evidence, however, has been irretrievably lost not only through the subsequent demolition of this building in advance of the viUa's re-orientation but also as a result of the deep disturbance of this area in early medieval times.13 In the following building phase (Period V), the original corridor was reconstructed: its sleeper-beam and wooden posts supporting a penthouse roof were replaced by a solid ragstone waU and short pillars supporting the roof, its floor relaid with tihng instead of the original planking and a smaU wing, projecting to south-west, was added; whether to house domestic staff, as previously suggested,14 is open to reconsideration now that it has been estabhshed that Room 94 was not demohshed before the end ofthis period. Period VI sees the re-orientation of the villa to face now to southwest. A new rear corridor is added to the house and the original rear corridor now becomes the fronting one as well as carried further to south-west and south-east at its two ends in order to form two projecting wings and enclose a courtyard laid with cobbhng; the north-west wing contains the baths, an entirely new structure, its south-eastern counterpart remains to be excavated. As if to emphasize the new orientation of the villa, a new wing, Rooms 118-121, projects beyond the rear corridor, and the presumed entrance-hall, Room 116, is closed bythe construction of a wall across its open end. The house now faces towards the Medway, with open views across its vaUey, possibly betraying a connection with shipping; however, this is undoubtedly a period of great prosperity as underhned by the extensive new building undertaken. The fourth century (Period VII) is marked by modifications to existing structures rather than new building and by the partitioning of the rear corridor into smaller rooms; the use of some of these rooms is not yet quite estabhshed, but the insertion of a channeUed hypocaust below Room 95 and the re-laying of the tessellated floor above it15 point to the continued use ofthis part of the house as hving accommodation during at least the initial part of this period. Later, however, the use of the north-west end of the house as living quarters was clearly 13 The slight amount of early medieval material, recovered in various robbertrenohes, suggests a date not discordant with the foundation of Aylesford Priory at a short distance from the site of the villa. 11 Arch. Cant., Ixxix (1964), 130. 15 Cf. note 6, above. 33 A. P. DETSICAS abandoned; for not only were its floors removed but also a hearth was sited within Room 89.16 Though the pottery recovered in the soot and ash deposits associated with this hearth as weU as an accumulation of pottery on the tessellation in Room 103 make it clear that this is a very late development, its dating is not yet definitely estabhshed. It would be premature to suggest that the rest of the house had been abandoned by that time; for, alternatively, the hearth and associated pottery deposits could point only to a change in the function ofthis part of the house and removal of the hving rooms to another. It is clear, however, that during this final period the hfe of the viUa continued on a rather reduced scale; it is not impossible that this is a direct consequence of the troubled conditions prevaihng after A.D. 367. The questions of the anomalous early foundation of the viUa and original ownership still remain unresolved. In the latter case, it can now be tentatively suggested that the change in the orientation of the house and its contemporary extensive additions may be the only surviving evidence not only of its increased prosperity but also of a change in ownership. The final question is posed by the cemetery: it could be associated with the estate at a time of dechne, a poor, Hi-organized sub-Roman cemetery as at Cirencester, the surviving evidence of people stiU clinging to the decaying viUa;17 on the other hand, it could be argued that these burials, with httle attention to the customary ritual of burial in individual graves, with or without grave furniture, clearly denote the occurrence of many simultaneous deaths and the consequent necessity for a rather rapid interment. The ages and sexes of these inhumations as weU as the total absence of any items of mihtary equipment preclude any warlike incident, and the only other alternative that wUl fit the excavated evidence is that these burials, allowing for the fact that the full extent of the cemetery is stiU to be ascertained, are the result of a catastrophic event such as an epidemic disease. 10 Arch. Oant., Ixxxiii (1968), 47. 17 I am indebted to Professor S. S. Frere, M.A., F.B.A., F.S.A., for this reference. 34