Excavations at Eccles 1965

EXCAVATIONS AT ECCLES, 1965 FOURTH INTERIM REPORT By A. P. DETSIOA.S, M.A., F.S.A. INTRODUCTION Excavations were continued} every weekend from late March to the beginning of November and for a fortnight in August 1965, at the site of the 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). For permission to continue with this work, I am again indebted to the landowners, Messrs. Portland Cement Manufacturers Limited, to Mr. N. M. Adams, their Regional Estates Surveyor, and to the Reed Paper Group Limited. I am also greatly in the debt of the tenant farmers, Messrs. A. A. and A. C. Southwell, whose generous assistance over the years has made the tasks of the excavation considerably lighter. The excavation was :financed by generous grants from the Kent Archaeological Society, the Society of Antiquaries of London, the Craven Fund and the Haverfield Trust of the University of Oxford, and by public donations. My thanks are once more due to the members of the Lower Medway Archaeological Research Group and the many other volunteers who ma.de this excavation possible, and in particular to the following for their sustained support throughout the season: Mrs. E. V. Jones, Misses M. Robinson, S. Vine, B.Sc., and M. B. V. Webster, and Messrs. I. J. Bissett, R. W. Chapman, A. C. Harrison, B.A., T. Hetherington, T. Ithell, D. T. Jones, W. A. Knowles and C. E. J. Martin. I am also especially grateful to Messrs. I. J. Bissett and A. C. Harrison, B.A., for much help with field drawing, Mr. R. G. Foord for undertaking the monochrome photography and supplying the prints illustrating this report, Mrs. K. F. Hartley, B.A., for reporting on the mortaria, Dr. J.P. C. Kent, F.S.A., for identifying the coins, and Mr. E. R. Swain for taking charge of the drawing of small finds. Lastly, I must record my appreciation to my wife who has coped with the onerous task of processing the bulk of the pottery. 1 Arch. Oant., 1xxvili (1963), 125-41; lxxix (1964), 121-35; lxxx (1965), 69-91. I a.m grateful to Professor S.S. Frere, F.S.A., for kindly reading this report in draft and contributing several valuable suggestiona. 44 EXCAVATIONS AT ECCLES, 1965 THE EXOAVATION Work this year was aimed at the completion of the plan of the earliest bath building and the investigation of the area to the north of this structure where it was hoped to explore the living quarters of the villa belonging to this period. To this effect, excavation was concentrated mainly to north and south of the areas excavated in 1962-64. Period I, to c. A.D. 55: The Ditches Some 183 ft. of the ditch first exposed in 19642 have now been cleared, but neither its true width nor depth are lmown; for, nowhere along its excavated alignment, has it been possible to ascertain these dimensions on account of supervening features which had removed the upper part of this ditch. However, it is lmown that the profile of the ditch was a rounded V-shape, and that its width and depth cannot have been much more than 6 ft. in either case; it had been cut into the Gault clay subsoil and filled with similar material and domestic refuse which yielded much pottery and two coins of Cunobelinus as well as several fragments of embossed bronze, which appear very similar to bronze fittings for a wooden casket found at Richborough.8 A total of 11 post-holes have been recorded so far, almost all of them north of the first baths, though this location can only be due to the fact that later structures have both removed the evidence and prevented excavation to the bottom of the ditch; there can be no doubt that, as already suggested in the preceding report, 4 a fence had been erected over the ditch when it had ceased to fulfil the purpose for which it was originally cut and been filled in with rubbish and clay. The majority of these post-holes were observed where they had just penetrated into the subsoil (Plate I), though the southernmost was seen to a depth of some 18 in. in the ditch filling and packed all round with stones; they were approximately positioned at the centre of the ditch. There is a slight change in the alignment of this ditch immediately beyond the north corner of Room 69, but both sections seem to be quite rectilinear. As the area to west of this feature is not yet adequately explored, little can be said about the purpose of this long ditch; whether it encloses a site occupied in pre-Roman times or was cut for some other reason at that time cannot yet be shown. North of the site of .the earliest bath building, a few sections were cut across another ditch, which traverses that part of the site on a south-west to north-east course. This ditch was found to be almost completely removed by later features and about 2 ft. 6 in. in width; 1 Arch. Oant., 1xxx (1966}, 70. 8 J. P. Bushe-Fox, Excavations at the Roman Fort at Richboro'U{Jh, Kent, Fourth Report, Oxford, 1949, p. 142 and pls. XLVII-XLVIII. ' Arch. Cant., lxxx (1966}, 70. 45 EXCAVATIONS AT ECCLES, 1965 very little stratified material has been recovered in its filling of clay, and this suggests that this ditch was cut some time before the much wider one, a fact which could be conclusively shown at the intersection of the two ditches. However, as it would seem from a preliminary examination of the stratified material that little time elapsed between the filling of these two ditches, it was thought advisable to consider them under the same heading and await more evidence as to their respective purposes. Period II, c. A.D. 55-65 Several sections were cut to the south-west of the first bath building in order to expose its south-western wall beyond the general area contained between Rooms 52 and 72, and it was found that the first baths were built inside the west corner of a boundary wall enclosing a large part of the site. This wall, which was some 1 ft. 6 in. wide, had been built of ragstone and yellow mortar on a foundation of gravel and cobbles along its south-western alignment and for most part of its north-western course as well, though it was very noticeable that in this stretch its foundations went much deeper into the subsoil and were well bedded in construction trenches,6 with ragstone chippings used as foundation material instead of the gravel and cobbles found elsewhere; and, as it can be shown that these chippings derived from dressing ragstone for the building of the granary, it seems reasonable to suppose that an effort was made in this area to rebuild this boundary wall on a more stable foundation, particularly over the alignment of the filled-in ditch. Further to the south-west of the present area, this boundary wall is undoubtedly part of a section of wall found in 1962 below the piscina (Room 17) of the third baths and its surrounding corridor (Room 18)6 and extending to south-west. The picture that emerges so far in this area is of a wall enclosing a site which initially contained some structures to south-east of the area excavated in the last three seasons of work (i.e. Rooms 53 and 54, and further to south-east), with a granary constructed at a later stage just within the enclosed area, and a final clearance of these structures at the start of Period III for the construction of the first baths, except where such structures could be incorporated into the baths. Period III, o. A.D. 65-120: The Bath Building The complete plan of the earliest baths has now been reoovered with the excavation of its north-west and south-west range of rooms, and it is now known that, during the five phases of construction which 6 ibid., fig. 3, Section G-H, Layer 68. 6 Arch. Gant., lxxvili (1963), 128. 46 PLATES I AND II Plwto: R. G. Foorcl Poriod I. Post-holes in Ditch. Period I. Section through Ditch Filling. l'holo: R. G. }'oord [ face p. 46 PLATES nr AND IV Periods I LI . J v. If earth. l'hOlo: R. (,'. Foord i• .. ,·iod n·. By-pass Drain. PhOlo: R. U. Foore/ PLATE V P/,o(O: R. G. Foord Period IV. By-pass Drain. PLATE VI Photo: R. G. FQOrtl Period lV. Room 51. EXCAVATIONS AT ECCLES, 1965 have been recognized, the total accommodation of the baths suite varied from a minimum of 19 to a maximum in the closing phases of 22 rooms (Fig. 2). Excavation was continued below the levels reached in 1964 inside the furnace-room in Phase A, Room 46, which subsequently ceased to perform this function; as a result, it has now been established that the testudo described in the previous report7 was a secondary feature of this room. Removal of the tiled cheeks of the testudo flue has shown that the original testudo flue was shorter than in its final phase, and there was no evidence to show whether there was a boiler attached to it; this could, of course, be due to the fact that, had there been such a boiler, which would be normal, its emplacement was destroyed in the subsequent lengthening of the flue. Excavation below the shorter tesfludo flue established that, when the prrefurnium was first constructed, its heating system had contained a cylindrical boiler whose emplacement had an internal diameter of 3 ft. 10 in. and was built of tiles set in grey clay to a total external diameter of 4 ft. 8 in.; the whole structure was encased in mortared ragstone which was bonded to the north-east, south-east and south-west walls of the room, and was placed upon a floor consisting of opus signinum some 2 in. in thickness laid directly upon the subsoil. The boiler would, presumably, have been contained within this emplacement and must have been about 3 ft. 6 in. in diameter, or smaller. Room 70, which contained Room 46, measured 24 ft. 6 in. by 40 ft. There was no sign of a floor, but it is not unlikely that it served as an open courtyard, a function which the whole area of Rooms 70-72 certainly fulfilled in the later baths of Period IV; on the other hand, it may well have been floored with tiles as was Room 72. Room 72 comprised an irregular area extending from the northwest wall of the baths to the extreme south-east limit of the building and ending, to the south-west, upon the re-used boundary wall of Period II. It would seem that originally the whole of this area served as an open courtyard with a tiled floor, and that a portion of this area was divided off and laid with a new mortar floor towards the end of the baths' life. But, until this took place in Phase E, there is no evidence of a partition wall beyond the south corner of Room 73 and projecting towards the south-west boundary wall; for, although this whole area to the south-west of the bath building had been systematically robbed to considerable depth, the construction trenches of the north-west and south-west walls of Room 73 were plainly visible where they had penetrated well into the subsoil-yet neither projected to south-west. In Phase E, however, as part of the general reconstruction of the middle range of rooms, the north-west wall of Rooms 57 and 73 was demolished 7 Arch. Gant., Ixxx (1965), 77-9. 4:7 EXCAVATIONS AT ECCLES, 1965 and a new one was built further to south-east thus reducing slightly the size of these two rooms. Room 71, measuring 25 by 17 ft. 6 in., was the result of a new partition wall built in Phase E to divide off the north-east part of the earlier Room 72; as already mentioned, a few tiles found in situ at the east corner of this room support the suggestion that it had a tiled floor. Room 74 (approximately 25 by 20 ft.) was another result of this partitioning. The new north-east wall of this room was constructed of the standard materials (ragstone and bright yellow mortar) to a width of 2 ft. and was abutted on to the north-west wall of Rooms 70-72; to the south-east, it ended abruptly some 2 ft. 3 in. from the east corner of Room 7 4. The end of the wall at this point had a regular, finished face, and below it there was a narrow (3 ft. wide) strip of opus signinum; together these two factors make it quite clear that a doorway must have existed here connecting Rooms 73 and 74. However, no trace remained of the conjectural section of wall needed beyond the doorway and up to the south-west boundary wall as neither its ragstone courses nor its construction trench, which would not have reached to the same depth as those for the north-west and south-west walls of the original Room 73, survived the subsequent wholesale demolition at depth. Nevertheless, the suggestion of a doorway at this point is further supported by the floor of Room 74, which consisted of a 2-in. thickness of bright yellow mortar and extended, though in places badly worn, over the whole area of this room, but not beyond the line of the conjectural south-east wall. Room 73 originally measured 15 by 9 ft., but was finally reduced to 14 by 9 ft.; it was a small room at the south-west end of the middle range of rooms whose function is not clear. No traces of a floor remained, except for some opus signinum adjacent to the conjectural doorway through its north-west wall in Phase E, and its walls were thoroughly robbed down to their lowest course of ragstone. It is probable, however, that this room may have been the latrine of the baths suite on the grounds that no other room could have served this purpose so well; furthermore, a drain, constructed of ragstone wa.lls with a tiled bottom and beginning near the north corner of this room (Fig. 1), would strongly support such a function. Room 62, which contained the furnaces, was completely cleared; its total area measured 11 ft. 6 in. by 15 ft. 9 in., quite commensurate with the number and size of the rooms (Rooms 38 and 39 and, proba.bly, 28 and 29) it had to serve. Its walls had suffered less in the subsequent robbing than those of rooms further to north-west, but were nonetheless reduced to their bottom courses; and of these walls only the northwest wall was the standard 2-ft. thickness, the other two being slighter at l ft. 6 in., all three built in construction trenches out into the subsoil. 48 . > I I ! -- -- \ ... - I I • I -- \ - -1111111111t11􀀖11􀀗lr1111111111111T11111rr1 : I \ j 1111111111111/11111,1,1 l l l l l l llfl.l Hl􀀦J l 11 l􀀧 I I ...·--\ ,,...-,.. \!.- \ ..- .,.,, e· C C L £ S I 9 6. S 􀀍I 􀀨, 􀀌 􀀍 ¾I 'l. 11/. 'l. 11/. J ___ - .. -, I I I ,- -·- --: i _____ , ______ - --·- I I I ---- ... H , Hypocaust. ! i M, Mosaic T , Tessellation. , • Post-Hole. '· = Wooden Pipeline. Ii 􀂇 Period Ila. 11 ,1 . Foundations or ' - - - QP"r1od Ila · Construction Trench. 'I 􀂈Period /lb. 􀂉 Period Ill o-e. 􀂊Period lllo-e,Robbed fflm P"riod I/lo: D(?molished. Ill Period Ill b-e. 􀂋 P(lrtod 1/lc-d. 􀂌Per1od llld-e. fflffl Period IV. ---- 0 Flo. L ' ------·- - - ·- -·-·-·- \ I I ·􀀘------- s M[fAtS ' I 10 [faC6p. 48 EXCAVATIONS AT ECCLES, 1965 The south-west wall of the furnace-room was buttressed by a short length of wall, 2 ft. 6 in. wide, which was abutted on to the southwest boundary wall. The floor of the furnace-room consisted of two layers of bonding-tiles set in clay and mostly covered by a thick deposit of soot and ashes. There was no definite evidence for a fuel store, but this is likely to have been the narrow space immediately to the south-west of Rooms 73 and 52. There is little doubt that the south-west boundary wall of Period II containing this bath building projects further to north-west beyond the exoavated area in order to connect with the foundations forming the north-west limit of this baths suite.8 Room 69 was the fuel store of Room 50, the furnace-room of the laconicum, Room 32. It mea.sured 15 ft. 3 in. by 9 ft. 9 in. and had walls consisting of coursed bonding-tiles set in clay and 1 ft. 6 in. thick; its floor consisted of white mortar and had been partly destroyed by the later insertion of a tiled drain. This room would seem to have been added to the plan of the baths some time after the construction of Room 50, for the north-east wall of Room 69 is abutted on to the ragstone north-west ,vall of Room 50. Periods III-IV, c. 65-180: The? Workshops The area to north and west of the first baths was partly excavated in an effort to locate the living quarters of the earliest villa, but it is now clear that this accommodation does not lie in this area. Instead, a succession of floors and other features has come to light and it seems probable that this area was occupied mainly by various workshops. The earliest evidence for any activity in this area appears to be a construction trench, filled with burnt wattle and daub, which delimits the north-east part of the excavated area. There was also found evidence for a system of water conduits consisting of wooden pipes connected together by iron collars. Sealed by the gradual build-up of mortar floors were debris layers which appear to be contemporary, at one chronological extremity, with the construction of the first bath building and its two mosaics and, at the other, with the salvaging of the mosaic floors from the burnt first baths at the start of Period IV. Part of this area was floored with a wooden floor with the traces of its planking clearly visible on the yellow mortar upon which it had been lajd, another part was floored with gravel which ended, to north-west, upon another floor of yello,v mortar beneath which ,vas found a small hearth built of bonding-tiles (Plate IlI). Muoh more work, however, is needed in this part of the site before 8 Confirmed in 1966. 49 EXCAVATIONS AT ECCLES, 1965 it can be fully interpreted; this will be carried out in the following excavation season and the whole area described in the next report. Period IV, c. A.D. 120-180: The Bath Building Further work this year established that the small room found immediately north-west of the courtyard wall of the baths in Period IV and which was thought as probably belonging to the first baths9 is in fact the latrine of the second baths. Room 51 measured 10 ft. 3 in. by 8 ft. 6 in.; its walls were 1 ft. 6 in. wide and were built of the standard materials (ragstone and off-white mortar) of the Period IV structures. It had been floored with a tessellated pavement, consisting of red and buff tesserae, about 1 in. square and roughly cut from tiles; these tesserae were set in o:pus signinum which also formed the bedding of the floor, with a thickness varying from 3 to 6 in. The tessellation extended over the whole area of this room, except for a space at the north corner where a hole, with tiles set in opus signinum to form an open channel into the drain passing below the room, denotes the actual siting of the latrine (Plate VI). For the drainage of this latrine, use was made of existing drains from the earlier baths. As it was first noted in 1963,10 the drain used to dispose of water from Room 28, which was the main sewer of the first bath building, was kept functioning (Fig. l); it would seem, however, that it was either blocked or in bad repair about the north corner of the earlier Room 71 which now formed part of this period's open courtyard. It was found necessary, therefore, to construct another length of drain by-passing the blocked area, though it is not immediately apparent why this by-pass drain had to be built rather than a repair effected in the blocked area-the reason may, of course, he that that was the lesser of two evils. In the course of work carried out in this area in 1964, it was observed at the extreme south-west section of a trench that the subsoil had been removed to some depth and that bonding-tiles had been inserted into the courtyard wall of the second baths in Period IV from the level of the undisturbed subsoil downward into the excavation; though a reason for this obvious rebuilding could not be reasonably suggested at that time, it is now clear that the courtyard wall had been breached at this point to allow for the construction of the new by-pass drain, which was carried through the wall under a tiled archway (Plate IV). This new drain was built with ragstone walls and a tiled bottom, and its channel was 1 ft. wide; its beginning at its junction with the main sewer could not be seen because that area had been thoroughly disturbed in subsequent robbing, but -- from that point it followed a northerly course, cutting through the 0 Arch. Oant., lxxx-(1966), 86, 10 Arch. Oant., lxxix (1964),128. 50 EXCAVATIONS AT ECCLES, 1965 disused drain of the earlier Room 58 (Plate V) and avoiding any area of particular difficulty, such as corners of walls, until it adopted the course of the earlier drain of Room 31 but at a higher level than the latter. Once it reached below Room 51, the new drain assumed the additional function of draining the latrine into the main sewer. An exploratory trench intended to establish the course of this sewer beyond the north-west wall of Room 51 failed to expose it, except for some evidence of cutting into the subsoil at the south corner of the trench; it is now clear that, at this point, there was a sharp change in the alignment of the main sewer which would now seem to follow a westerly course down the gently sloping ground towards an outflow or a soak.away further to west. Perioda IV- V, o. A.D. 120-290; The Ditch Further sections were cut across the line of the ditch which seems at present to form the north-east boundary of the site. Its width and depth remained more or less constant at the south-east end of the sections, but the ditch soon began to become less deep and its sides less steep the further it continued to the north-west; this was already noticed in 1964.11 At the extreme north-west point reached in the course of this featw·e, the filling no longer consists of domestic refuse mixed with building debris, but almost exclusively of the latter only, and this consisted in the main of burnt clay deriving, no doubt, from the clearance of the destroyed debris of the first baths; but the ditch as such had virtually ceased to exist and it was practically impossible to ensure that it ever continued this far to the north-west. DATING No evidence has come to light to cause any revision or reconsideration of the dating sequence suggested in the previous report,12 except to establish that the beginning of Phase E in Period III must be placed some time after A.D. 95 on the basis of a worn as of Domitian which was found stratified in the loose ragstone foundations at the north corner of Room 69. S"CJM:MARY .AND DISCUSSION After a further season's work at this extensive site, the questions which have remained unanswered since the beginning of the excavation must· still a.wait much additional investigation before a reasoned and reasonable answer can be suggested, and the hypotheses which were 11 Arch. Oant., Ixxx (1965), 88. 12 ibid., 88-9. 61 EXCAVATIONS AT ECCLES, 1965 proposed in lieu of definitive answers remain still valid. Evidence is steadily accumulating of an occupation before the Roman conquest-­ the amount of early pottery is too large and the stratified early coins are too many to be mere chance survivals, and the early ditches must have formed part of some sort of settlement, but excavation has not yet provided more than a few pointers as to ,vhat this occupation must have consisted of, which are still insufficient to suggest even a working hypothesis. It is hoped, however, that the following season's work may furnish evidence which is at present lacking. 52 £ C C L E S I 47 70 PERIOD I/IA 0 10 Feet 9 6 3 JO M 20 JO JI M 40 6 5 69 PERIOD 11/B·C 8 A r H 70 72 Disvsecl 8 . U I FIG. 2. 7J • i I) I 69 M: MOSAIC H : HYPOCAUS r PERIOD D·E N G s /0 •lliiiiiiiiiiiiiiiiiiiiiiiiiiiii􀀐!!!!!!!!!'!!!!!!!!!!!!!!!!'!!!!!!!!!IMetres 0 s (face p. 52