An Iron Age and Romano-British Site at Stone Castle Quarry, Greenhithe

.AN IRON AGE .AND ROMANO-BRITISH SITE AT STONE CASTLE QUARRY, GREENHITHE* By A. P. DETSIOAS, M.A., F.S.A. !.NTRODUOTION For O.S. A brief trial excavation was carried out late in the autumn of 1960 following the discovery of a rubbish pit containing Romano-British pottery at Stone Castle Quarry, Greenhithe; and, as further evidence of occupation came to light in later quarrying operations, a full-scale rescue excavation was undertaken the following year on behalf of the then Ministry of Works, by permission of the landowners, Messrs. Associated Portland Cement Manufacturers Limited, and the kind assistance of their local manager, Mr. A. J. Thoma.s.1 My many thanks are due to the many volunteers who helped in this excavation and, in particular, to Messrs. R. G. Foord, L. A. Griffith, T. Hetherington, G. K. Horner and J. Vickers for their sustained support throughout the course of the work. I am also greatly indebted to Mr. D. F. Allen, B.A., F.B.A., F.S.A., for identifying the coins; Mr. I. J. Bissett, for drawing some of the pottery; Miss D. Charlesworth, M.A., F.S.A., for reporting on the glass; Mr. R. G. Foord, for photographing some of the finds; Mrs. K. F. Hartley, B.A., for reporting on the mortaria; my former colleague, Mr. A. J. Hewitt, B.A., for carrying out a preliminary survey of the site in very adverse weather conditions; Miss J. E. King, for identifying the bones; my brother-in-law, Dr. R. P. S. Jefferies, B.A., Ph.D., F.G.S., for the original information of the site and his notes on its geology; and Mr. E. R. Swain, for his drawings of the small finds. TRE SrTE The site lies approximately one mile south-east of the hamlet of Stone and some 300 yards north of Watling Street (A2); it is situated, at about 100 feet above O.D., directly on Upper Chalk, but within 150 yards of a Coombe Rock deposit2 (N.G.R. TQ/583732; O.S. 6-inch Sheet TQ 57 SE). "'The Ministry of Public Building and Works contributed to the cost of printing this paper. 1 Arch. Oant., lxxvii (1062), 200-1. I have greatly benefited by the advice of Professor S.S. Frere, F.S.A., who has kindly read this report in draft and con• tributed several valuable suggestions. 2 Information from Dr. R, P. S. Jefferies, F,G.S. 136 GREENHITHE REPORT In the course of quarrying, the mechanical excavator exposed a deposit contafaing the fossilized remains of cold-phase Pleistocene mammals, including mammoth;2 this discovery was eventually notified to the British Museum (Natural History) and, on inspection of the site, it was observed that the mechanical excavator had also cut through what appeared to be a pit filled with domestic refuse and pottery (Pit 1). Some of this pottery was recovered and, with the assistance of Mr. L. A. Griffith, an attempt was made to cut a regular section through this deposit exposed in the quarry face. At the same time as this work was taking place, the landowners began the mechanical stripping of the topsoil from the area immediately to the south of this pit in order to extend quarrying in that direction. Watch was kept on that area, and it soon became clear that a much longer season of excavation would have to be carried out so as to secure as much of the evidence as possible before it was finally destroyed by quarrying. THE Exe.A. VATION The main area of the excavation was bounded to north and west by the quarry face, to south by a small copse and to east by a slight drop in ground level, probably the result of the silting of a small stream (Fig. 1, inset map). The topsoil, which was nowhere deeper than 6 in. at most, consisted mainly of rough scrub, roots and bushes, and was stripped by means of a bulldozer whose primary concern was to clear the area for further quarrying rather than to prepare the ground for an archreological investigation; as a result,· some of the underlying chalk was removed as well and with it all evidence of post-holes that might have survived within the enclosures. In fact, it is likely that most of the evidence that was secured would have been quite lost in the course of bulldozing but for the fact that the pits, gullies, etc., had been cut deeper into the chalk than the bulldozer penetration; however, enough evidence was salvaged to establish that this part of the site-for it is practically certain that quarrying has destroyed much other evidence of settlement to north and west of the excavated area-was occupied during the Iron Age, and then from the late-first century .A..D. into the third century, and probably later. In brief, a number of rectilinear ditches have been traced as far as possible and of these two. can be shown to be pre-Roman, two ditched enclosures with associated storage pits and an isolated hearth which afforded ample evidence of lron Age date have been investigated, and a fairly large quarry pit with three associated ovens were exposed and fully recorded. These features are described below within their respective periods. 137 GREENHITHE REPORT (A) IRON AGE Ditch 3 was situated almost on the very edge of the slope to the east of the site and has suffered very badly in the course of bulldozing. It was traced for some 60 ft. on a south to north-east direction; its surviving width was about 2 ft. and its depth 1 ft. This ditch was found filled with rammed chalk and clay; there was neither any silt at the bottom nor any stratified material. It was impossible to trace its course further than shown on the site plan (Fig. 1), but it must date to a time soon after the abandonment of Enclosure B as the ditch cuts through it; furthermore, the slight deviation from its rectilinear course at the south end of the site must have been necessary in order to avoid Pit 4 which, still in existence if not in use, it would have otherwise bisected. Its purpose, like that of the other ditch of this period, can only be conjectured; it may have served as a boundary ditch. Ditch 2 antedates both Ditch 1 and the larger of the two enclosures as it is cut by both of these features; it is less wide and deep than Ditch 2, quite rectilinear and would seem to have formed the boundary of a near-rectangular enclosure. It was found to be 1 ft. wide and had survived to a depth of about 4-6 in.; it was traced for some 40 ft. to west from its south-east corner and 156 ft. to north ·where it was lost in a pipe of soft clay near the edge of the quarry face. A gap of some 11 ft. in this longer alignment may have been an entrance into the enclosed area; if this entrance were placed centrally, then the north-east corner of the enclosure would have been some 11 ft. further north of its last known point and a little beyond the edge of the quarry face, making a probable total length of some 167 ft. The length of the south side of this enclosure cannot be computed. No dating material was found anywhere in the compact clay and chalk filling of this ditch, though it is certain that it constitutes one of the earliest features on the site. Hearth. Among several areas of discoloration, showing in the subsoil after the mechanical removal of the topsoil, was a roughly circular hole in the chalk; this was first thought to be a natural clay pipe, but a section cut across it from north to south showed that it was not a natural deposit. The chalk had been removed to a depth of about 3 ft. and back-filled with a layer of red loam similar to such material found re-deposited inside the quarry pit at a later date; whether this hole had originally been used as a storage pit could not be ascertained, though it might be thought that its presence outside either of the two enclosures would preclude this use. Whatever its original function, it had a small hearth (about 3 ft. in diameter) placed on the red loam and dating to the same period as the smaller of the two enclosures; animal bones, much soot and charcoal as well as some pottery were recovered in such small quantities that, unless much more of this pottery was lost 138 GREENHITHE REPORT in the bulldozing, it seems likely that this hearth was not in use over a very long period of time. Enclosure A. This is the smaller of two enclosures found at the site and the one most likely to have contained a hut. It is very nearly circular, with a diameter of about 54 ft., and was enclosed by a shallow gully which was rarely more than 6 in. deep and cut into the subsoil; no post-holes were found anywhere in the perimeter of this gully. The enclosure was entered through a gap (5 ft. 6 in. wide) in its gully facing towards the north-east; it is worth noting that this entrance corresponds to the gap in Ditch 2, which may suggest that the enclosure and the ditch were contemporary before the latter was superseded by Enclosure B. Two storage pits were found within Enclosure A, the larger and deeper of the two to the south-west of the entrance and the smaller to the south and immediately inside the gully. In spite of a systematic search of the whole enclosed area, no traces were found of any post-holes at all. If a small hut had been contained within the ditch. its location must have been to the south-west and behind Pit 2, but positive evidence for this was totally lacking; if, on the other hand, the whole of the enclosure was one hut, with the gully serving as an open drain collecting rain-water from the roof of the hut, then one would expect to find the post-holes for the timber framework of the hut either in the gully itself or not very far from its inner lip-once again such positive evidence was entirely wanting. The second alternative that this enclosure was one single hut seems preferable as, in this case, both storage pits would be under the cover of the hut's roof; alternatively, if a smaller hut had been sited beyond the pits, it could be argued that they were protected against the weather by means of independent roofs, which seems unnecessary. Conjectural though this must be, there was ample evidence to show how the wall and roof of the hut were built and weather-proofed. Posts would have been set at regular intervals and the spaces between each post filled with interwoven wattles and such light timber; finally, the whole structure would have been rendered waterproof by an overall application of cob, a thick coating of puddled challr and clay. The evidence for this is quite secure in that many fragments of cob, clearly showing the imprints of the wattles o n which it had been applied, survived the collapse of the hut {and the attention of the bulldozer!) inside Pit 3; none of this material {Plate II) was recovered elsewhere on the site. It could be argued that the cob fragments constitute the remains of the roof of the storage pit only, but there is no evidence to suggest that the hut itself was not lilcewise protected, and it seems reasonable to suspect it; if the cob afforded sufficient protection for the grain stored within the pit, why might it not have served to protect the inhabitants of the hut as wem 139 GREENHITHE REPORT Pit 2 (Fig. 1), the larger of the two storage pits, had been out through the chalk to a depth of a little more than 6 ft. and was slightly more than 7 ft. in diameter; it was found filled with an accumulation of debris layers, consisting mainly of chalk (Fig. 2, Section E-F), except for about 2 ft. from the surface which contained a filling of charcoal, ashes and domestic refuse. This deposit was the first surface indication leading to the discovery of the pit after the stripping of the topsoil; the pottery it contained was badly burnt after breakage, and it would seem beyond doubt that, after the pit had ceased to be used as a grain store, it was filled in and used as a rubbish pit for the disposal of material from the hearths on the site. Some of the pottery recovered in this deposit is of the same forms and fabrics as sherds recovered in the ashes of the isolated hearth found outside the enclosures. Pit 3 (Fig. 1) was much smaller and shallower than Pit 2; its diameter was only 5 ft., its surviving depth barely 2 ft. 6 in. Though most of its filling had been removed by the later cutting of the subsequent Ditch 1, it had been filled mainly with a deposit of fragments of cob, ashes, charcoal and rubbish which was practically identical with that found at the upper levels of Pit 2; the pottery found in Pit 3 is of the same forms and fabrics as that recovered from Pit 2, and it would follow that both these pits were filled in at the same time. The sides of this smaller pit had been made waterproof by the application of a coating of yellow sandy clay against the chalk sides of the pit; this would not only have been particularly desirable because of the proximity of the open gully outside it, but could also indicate that parched grain may have been stored loose in this pit rather than inside some sort of container or lining which is likely to have been the case within the unrendered sides of Pit 2. Enclosure B. This is the larger of the two enclosures exposed on the site, and the one least explored; the ground both to west, east and north of the area where its gully was trenched either slopes sharply and the bulldozer removed the evidence or, on more level ground, the shallowness of the gully and the stripping of the topsoil had the same result. However, the diameter of this enclosure can be computed at about 138 ft.; the enclosing gully was at 1 ft. 6 in. rather wider than that around Enclosure A, and generally deeper, with a maximum depth of about 1 ft. in parts. This gully was found filled in with a mixture of rammed chalk and clay and contained very little stratified material, except for most of one vessel (Fig. 5, no. 34) and a few other sherds. Pit 4 would have been contained inside Enclosure B and had a diameter of about 4 ft. and a depth of 1 ft. 6 in., though some of this depth had been lost in the bulldozing down the sharp slope of the ground; its filling consisted mainly of chalk and ashes as well as flints, and the pottery in this deposit, some of .:which conjoins with other 1.40 Mi'l8s o 10 .lO .JO 40 0,..10 -e- ,o'c"':JO􀀒.«>􀀓so Kilomdres - S.I.._Q NE. QUARRY, 0 /0 Ouurry Edg11, October !p6Q T T J 1r11i17771T7lTTT TT 7 I r 1 i .-------􀀌--·􀀍 .. , .. ·; Stone 2 Chorlton 3 Oldhury 4 Patch Grove s 􀀮lesford 6 Hvl611ry 7 /(estcn i-·---7 : i I I i_ ___ _! .CA_ ST.LE 20 s \ ' I 9 6 0-I .JO i '\ i 40 /0 I Metres I(,•• y IRON AG£ 8RlrAIN ,omet􀀔s o JO '° JOMt1es IO 􀀐 JO «J SO Inset Maps based on Ordnance Survey Maps and reproduced by permission of the Controller, H .M. Stationery Office. [/p. 140 Fig. 1. GREENHITHE REPORT sherds found in Pit 2, clearly shows that this pit, too, had been filled in at the same times as Pits 2 and 3. The purpose of this pit is less easy to determine and would depend on the interpretation of Enclosure B. ff it could be conclusively demonstrated that this enclosure had contained a hut or huts, then this pit could have been used for the storage of grain as is suggested for Pits 2 and 3; however, no evidence could be found for any huts, no post-holes were to be observed anywhere within the enclosed area in spite of careful search. On the other hand, if it can be accepted that the hut was situated within Enclosure A, then it is not unlikely that Enclosure B may have served for the herding of livestock; and, if this were so, then it is unlikely that Pit 4 was used as a grain store, though it could have had an open hearth placed within it. (B) ROM.ANO-BRITISH Except for the foundations of a small outbuilding constructed within the area of the quaITy pit, no evidence was found for a RomanoBritish building; but there are indications that such a structure may well have existed at some distance to the north-west of Ditch 1. The position of Pit 1 well clear of the area delimited by Ditch 1, the amount of building debris, containing many fragments of bonding-tiles, imbrices and tegulae, found in the filling of the quarry pit and in particular in the lowest layer (Fig. 2, Sections A-B and 0-D, Layer 17), the presence of three ovens and the considerable amounts of RomanoBritish pottery, all this makes it abundantly clear that occupation continued well into the third century A.D. The area to east and south of the excavation was searched for signs of any occupation and several exploratory trenches were cut, but nothing was found. Air photographs taken in May, 1961, were closely studied, again with negative results, and several traverses with a resistivity meter to the south of the line of trees proved inconclusive. The topsoil cover in the fields both to south and east of the excavated area is very slight, rarely deeper than about 6 inches; the chalk subsoil is often exposed in the course of normal ploughing and this could be very clearly seen on all air photographs; yet, there was none of the amount of debris to be expected in newly ploughed areas if a building lay buried there. It seems, therefore, reasonable to suppose that the Romano-British building suggested on the grounds mentioned above was situated to north-west of the site; if this were the case, it is liltely to have been removed completely in the course of the extensive quarrying operations carried out in that area during the last war. The ground level certainly rises to north-west, and that area would have been ideally suited for the location of a small building; small and unpretentious, it may well have been for, otherwise, it is difficult to see how it could have been so completely destroyed without any of the 141 GREENHITHE REPORT quarrymen noticing any signs of it-and on this point all the workmen questioned were quite emphatic that nothing was observed. It is not unlikely that this structure may have been built of chalk and timber and its construction trenches destroyed in the course of quarrying. The evidence for Romano-British occupation of the site consists of the original rubbish pit (Pit 1), a boundary ditch (Ditch 1), a quarry pit within which was placed a small oven, two other ovens cut out into the chalk subsoil (Ovens 1 and 2) and, finally, the foundations of a small outbuilding situated on top of the filled-in quarry pit. Ditch 1 constitutes the earliest evidence of occupation at the site during the Romano-British period; it had been cut through Enclosure A and its Pit 3 and traversed most of the excavated area on a southwest to north-east course. Some 162 ft. of its south-east alignment have been excavated, but it was not possible to trace its full course beyond the line of trees to the south of the site and into the field beyond them; it is quite certain, h􀄀wever, that the ditch continued further to south• west than its lam recorded point for it could be seen as a hollow amongst the trees. This ditch was 3 ft. wide, with a maximum surviving depth of about 2 ft., though undoubtedly bulldozing accounts for some loss of depth; its profile was a rounded V-shape, and it was filled with rammed chalk and clay, with little silt at the bottom. The south-east course of this ditch came to an end at a point where the ditch turned to north-west only to terminate some 23 ft. further from there. V\Thether this was an entrance through the ditch is impossible to say, but it does not seem likely; for, even allowing for bulldozing which in this area was more severe than elsewhere, some tra.ces of its north-east course should have been noticeable in the chalk, yet none were seen. It is difficult to believe that this ditch did in fact continue further to north-west without leaving such evidence when this was present, in the same area, along the course of the much slighter and shallower Ditch 2 of the earlier period. No apparent reason can be suggested for the end of the ditch at this point; but a regular entrance into this area across the ditch seems to have been allowed for at the south end of the site where a gap of 6 ft. is clearly intended for this purpose. This could mean that Ditch 1 was a boundary across open ground facing to south• east and terminating to north-west upon some geographical feature, such as a wood, which from that point onward would have served the same purpose. Immediately outside this ditch and barely 2 ft. 6 in. from its south• eastern lip, the chalk subsoil had been excavated to a known depth of very nearly 7 ft. and over an area of 18 by 19 ft. 6 in.; the striking feature of this excavated area was that its sides were almost smooth and practically vertical, and it was evident from the outset that efforts had been made here to excavate more than a mere rubbish pit. Various 142 s ..1.. -L .... .L. ..L. ...L. ..L - ..L .J... ..L ..L T .J... ..L ..L ..L .J.. • ..L ..L ..L ..L -'-' ..L -L-L.J....L. - ..L ..L ...L ..L ..L ..L..L..L.J.....L ..L • .J.. ..L ...L ..L ..L. ...L. ..l.. -L. .. L. ..L ..L - ..L ..L ...L ..1. -l.. • ..L ..L .J.. ..l.. ....L. ..L - .J... ..L J.. ...1. ..L. • ..L ...L. ..L .J.. ..L ..L • ..l.. ...L. -L ..L -l.. • ...L ..L .J.. ..L. ..L ..L • ...L. .J.. .J.. ..L ..L • ..1.. ..L -L ..L ...L ...L - ..L ..L .J.. ..J.. ..L • ..L...L...L...L..1.. ..L • .J...J....J..©..L..L...L...L. ..L 􀃱 • ..L ..L ...L ...L -1.... .J... ..L -L .I.. ..L ...L • ...L. ..L. ..L ..L ..L • ..L ..L ...L ..L ..L ..L - ,..1... ..L ..L ..L .J.. . ..L ..L -L .J... ..L ...1. ..L ..L ...L .., 􀃲 0 OJ]]Rubbish Filling 􀂇fine C/)(Jlk Rubble [:􀂈;,􀂉18a.wd Clay N E :t .6 15.:J Chalk-Subsoil a Clay Seal I== :jsrown Soll C A s T L E Q u A R R s e C i 0 n A 8 11111111I11111111 Ill 11111[11 ii I I I Ii li[1[ I I I lililililil ii ililil, lil􀀝:1:1:1: 1: 1: 1: l1lilil1lil: I ililili lil1l1l111 I 11111.111111 s e C i 0 n m GJJ]CompactChalk& Clay §Red Sandy Loam [I]]]]Chalk L umps IT[]]Grey F171/ng mmic􀀐;;d-ch􀀑lk. -· -- •Soot& Ash C 0 IIICI]Black Deposit' [Il]]]Bullo'ozer o,rt' §!§Clayd Flints ldLooseChalk - . © 􀂊Ash mJBrown Soi􀂋 Flints & Chalk Metres 􀂌 Fm.2. Y, s J 􀀄 I 9 6 0 - I. 􀀂 £. L .Q. .!l g \l1ll\1\l\l\l\l[l\l\lll\1l1\ll!llll\i􀁢l1\!lllllili\1\i\!\1ll\l\l\ili\1l1\il\1111 - - ·.1- ll l􀀌IE'􀀍I I 1 .. '•􀀎􀀏 ..t. -'- -'- .1. l:lp:􀀅 ·:t:±'·'::!::-',l::"-:,J,1:,,.1..,J.. •.l . ..; IL.LI ..!l l. l􀃳-.--..L---..L--..1.. -'- ...1... ..1... . ...L..L􀁁..L ..L..L􀁂􀁃 ..L ..L .J... ..L ..L ..L ..L ..L􀁄L .J... ..L.....L • ..L...L..l.. ..L..L-l..L.J....L ..L ..L ..L.. ..l.. .J...J..""-lJ..L ..1.. ...L..J... . 􀁅..L ..L.J.. ..L 􀁆 ..L ...L ..L ..L ..L ..L..L 􀁇--=:􀁈􀁉􀁊􀁆 􀁋􀁌 E I 2 J 4 s􀀃 · P 6 A(56D I Feet 􀀅 ' f (fac􀀊 p. 142 GREENHITHE REPORT possible interpretations, amongst which that of a deep room, were in turn discarded in that they fitted only some of the excavated evidence; the most likely explanation would now seem to be that the chalk from this area had been methodically removed for the manufacture of chalk bricks to be used in building construction, and that later this quarry pit was used as a rubbish pit and for other purposes described below. The basis for this interpretation is twofold: first, some of these chalk bricks were found (Plate I) and show unmistakable signs of tooling and, second, the lowest deposit in the back-filling of this quarry pit consisted of a layer of flints (Fig. 2, Section A-B, Layer 35) which represents unwanted material deriving from the chalk used for bricks. The use of chalk as building material is, of course, well known; and had the walls of the conjectural destroyed building further to northwest consisted mainly of chalk bricks with a few lacing courses of bonding-tiles and timber, it would be easier to accept that the destroyed building left such vestigial traces that they escaped notice in the course of quarrying. An excavation of this size and depth was an ideal site for the deposition of domestic refuse, and this is the purpose the quarry pit served soon after the deposition of the discarded flints, which were next covered by a substantial layer of rubbish (Fig. 2, Sections A-B and C-D, Layer 17); this contained, besides debris from the conjectural building, considerable amounts of pottery which afforded ample evidence of much activity in Antonine times. Howev<'lr, a change in the use of this area soon took place when a small oven was constructed upon the consolidated rubbish layers filling the quarry pit; several pieces of broken quern-stones, of volcanic lava either from Andernach or Niedermendig, were found at the bottom of this oven, and it is clear that they were placed there to form its bottom. All that remained of this oven was its collapsed roof of baked clay (Fig. 2, Section A-B, Layer 29) and associated layers of soot and ashes (Fig. 2, Section A-B, Layers 30 and 32). The diameter of this oven, assuming that it was more or less circular like the later ones, could not have been much more than 3 ft. Oven 1 is the earliest of two ovens cut through the chalk subsoil at a higher level than the original oven; it had been cut through the south-west side of the quarry pit, mea,mred 3 ft. 6 in. in diameter and was floored with a layer of clay, which had been burnt to the consistency and colour of brick (Fig. 2, Section C-D, Layer 41) and was about 3 in. thick. The oven was found filled with a deposit of challc lumps and flints (Fig. 2, Section C-D, Layer 40) beneath its baked-clay roof which had collapsed and spread out into the quarry pit (Fig. 2, Section C-D, Layer 43); that this oven was in use before the quarry pit had been finally filled in is made clear from the soot and ash 143 GREENHITHE REPORT (Fig. 2, Sections A-B, Layer 14 and C-D, Layer 47) which had been swept out of the oven and into the quarry pit below it. It is also clear, from a comparison not only of the amount of pottery present in their respective soot and ash layers, but also from the extent of the latter, that this oven had a longer life than its successor. Oven 2 was also cut out of the chalk subsoil beyond the south corner of the quarry pit; it was not quite circular as Oven 1, and its maximum dimensions were 4 by 3 ft. Its floor was also of baked clay, but rather thicker at 4 in. No trace remained of its roof which must have been completely cleared away after the destruction of this oven, but its soot and ash were plainly visible (Fig. 2, Section A-B, Layers 12 and 23). Stratigraphically, there is no doubt that Oven 2 is later than Oven 1; their soot and ash layers show this conclusively. Chronologically, the dillerence in time may be very slight; for the material stratified in these soot and ash layers is very difficult to separate, certainly with no more precision than Antonine and late-Antonine, or later. Eventually, the whole area of the quarry pit was filled in and levelled before a small building was constructed within it and following its outline; this structure measured 11 ft. 6 in. by 15 ft., with walls 1 ft. 6 in. wide. Of these walls only the foundations were preserved (Fig. 2, Sections A-B and C-D, Layers 3 and 11) and consisted of flints set in puddled chalk and clay, rather reminiscent of the flint footings at Cobham villa;3 a gap, 1 ft. 6 in. wide, at the north corner of the footings would seem too narrow to be a proper doorway, but it may have been wider at a higher level. Whatever the purpose of this outbuilding, it seems clear that it cannot have been constructed of stone and is more likely to have been built of timber; its dating, too, is based on the chronology of the ovens below it for little material was found securely associated with this outbuilding. The black deposit (Fig. 2, Section A-B, Layer 7), which filled the upper level of the quarry pit and betrayed its existence, is not of much help in dating this outbuilding in that it is a layer of back-filling and contains pottery earlier than material found at lower levels. Pit 1 w􀂹s the original section exposed in the face of the quarry and, though very little remained of it before a proper section could be cut across its filling, it is clear that it was opened for the deposit of domestic rubbish and building debris. It was situated about 102 ft. from the nearest lmown point of Ditch 1 and contained in its filling much burnt wattle and daub, fragments of roofing- and bonding-tiles as well as much pottery some of which was recovered; some of this material is not only of the same fabrics and types as those found in Layer 17 of the quarry pit, but also includes sherds which conjoin with others found in 3 Arch. Oant., lx:xvi (1961), 88 e.nd pl. IB. 144 .. ... Chnlk Brick􀀕. (Scnle in inches) l'horo: U. G. Poord Pi.ATE IT Fragments of Cob froin Hut. {Scale in inches) Pltoto: R. G. Foord Small Finds. (Scalo in inchos) PLATE III Pl1ow: R. 0. Foord GREENEITHE REPORT that layer. It is not unreasonable, therefore, to suppose that this pit wa6 the original rubbish deposit east or south-east of the postulated building until it became full and use was then made of the quarry pit. DATING Iron Age. There is ample evidence to suggest that the occupation of the site began during the first century B.O. or even earlier. The pottery {Figs. 3-5, nos. 1-44) recovered from those parts of the site which can be shown to be of pre-Roman date (Pits 2-4) is exclusively of the Wealden culture4 (Southern Second B) and its nearest parallels occur at Crayford;5 on the other hand, no Belgic pottery ha.s been found in spite of the proximity of the cemetery at Stone,6 which strongly suggests that this phase in the occupation of the site may have come to an abrupt end with the Belgic invasions. Romano-British. The earliest evidence for the re-occupation of the site is some of the pottery found in Pit I and in the filling of Ditch 1 (e.g. Fig. 18, no. 235); from this, it would seem likely that re-settlement took place in Vespasianic times, perhaps a little earlier, when Ditch 1 was cut to the north-east of the occupied area. Pit I may have been opened about that time, too, but it is impossible to be certain as most of the material in its filling wa6 lost when the mechanical excavator cut through it; it is not in doubt, however, that this pit was in use from c. A.D. 90-120 by which time it was becoming full. The.quarry pit, the building debris in its filling and in that of Pit I suggest that some re-building took place at the beginning of the second century A.D. Excavated in order to obtain chalk for this reconstruction, the quarry pit became the rubbish pit for the material deriving from this phase in the occupation of the site. The significant deposit is Layer 17 which can be dated toe. A.D.120-150, perhaps slightly earlier; though some of the pottery in this layer undoubtedly continued into use without much appreciable change beyond Antonina times, the total absence from this layer of the typical Castor ware forms with barbotine decoration would indicate that deposition ended about the middle of the second century A.D. The small oven placed inside the quarry clearly did not la6t in use for very long, but there was no material securely stratified with it to indicate when it was superseded by Oven 1; the latter need not be :much later than c. A.D. 160 and may not have had a very long life. 4 J.B. Wo.rd Pedtins, 'Excavations on the Iron Age Hill-fort of Oldblll'y, near Ightho.m, Kent', Arohreolog·ia, xo (1944), 144-5, figs. 5-6, and 160, fig.12. 􀁿 J, B. Wo.rd Perkins, 'An Early Iron Age Site a.t Cra.yford, Kent', PPS, iv (1088), 168, fig. o. G M.A. Cotton and K.M .R icha1•dson, •A Belgio Cremation Site at Stone, Kent', PPS, vii (1941), 184-41. 145 GREENHITHEREPORT Oven 2 can be dated to the last quarter of the second century A.D. and seems to have lasted in use until the last decade of the century. The outbuilding erected inside the filled-in quarry pit is likely to date from Severan times, but the evidence in this case is not very secure. DISCUSSION In spite of the fragmentary nature of the evidence salvaged in the course of this excavation, it is reasonable to suggest that the site at Stone Castle Quarry was during the Iron Age one of those which 'provided for smaller units of population than were accommodated in the large settlements'.7 Though it is not possible to say when this settlement first took place at this site, it is clear that it lasted a number of years before the arrival of the Belgae which seems to coincide with the end of the Iron Age occupation of the site. Whether this was the result of some danger, as suggested elsewhere for such sites of the Wealden culture,8 is a matter for conjecture; but, if this suggestion is acceptable for small farms and open settlements in the Ightham area, there is no reason to suppose that it did not obtain in the Dartford region. If there are grounds for supposing movements of population into fortified sites, Oldbury is barely ten miles from Greenhithe and Hulbury about five (see Fig. 1, inset map). Whatever the reasons for it, the Iron Age occupation came to a sudden end, and the site seems to have remained unoccupied until some time after the Roman conquest; and the total lack of any definite Belgic material would further reinforce the suggestion of abandonment in the face of the danger represented by the Belgic invasions. During the Iron Age occupation of the site, the hut which formed the centre of this open farm is rather larger than huts found elsewhere in southern Britain so far; other smaller huts may, of course, have been contained within the much larger Enclosure B .which could be evidence for a small settlement rather than a single farm unit, but this cannot be supported by the surviving evidence. The larger of the three huts (Hut A) at Draughton, Northants.,9 that at Colsterworth, Lincs.,10 and those at Heath Row, Middx.,11 to quote only a few, are all smaller than Enclosure A with its diameter of 54 ft.; this may suggest a larger family-unit accommodated within the hut than at the sites mentioned above. Again, if evidence had survived to establish the presence of 1 W. F. Grimes, 'SomeSma.llerSettlements:ASymposium', in (Ed.JS. S. Frere, Problems of the Iron Age in Southern Britain, Instit,ute of Arohreology, London (n.d.), 17. 8 J". B. Ward Perkins, Oldbwry, 149. o W. F. Grimes, Zoe. cit., 21•3. lO ibid., 23-5. 11 ibid., 25. 146 GREENHITHE REPORT other huts within Enclosure B, it might have been reasonable to think in terms of some sort of social pre-eminence for the members of the family occupying the single hut; as it is, it would be idle to attempt to draw any conclusions, however tentative, from such scant evidence on the social background of the occupation of the site. Likewise, though some tentative suggestions could be made about the population of the site on the basis of Bersu's postulated storage capacity of corn-pits and its recent re-appraisal,12 no evidence survived to indicate the economic background of the occupants of the site. All that can reasonably be suggested, o n the basis of the excavated evidence, is that a community had settled this site during the Iron Age and lived in a hut or huts, probably farming the gently rising land of this area, and that the onset of the Belgic invasions caused the site to be abandoned. It was, however, re-occupied about the third quarter of the first century A.D. Again, the archreological evidence is very fragmentary and allows only a few tentative suggestions and fewer firm conclusions. The same tantalizing situation exists at Stone Castle Quarry as at the site recently excavated at Joyden's Wood, near Bexley;13 here, too, there is strong circumstantial evidence for a building but no actual traces of it remained though, as suggested above, this is very probably due to quarrying. Even with this dearth of concrete evidence, it seems reasonable to postulate a small, unpretentious building, very likely of chalk and timber construction, occupying the area to the north-west of the ditch forming its boundary; the pottery and the building debris make this conclusion inescapable. Further support to the suggestion that this conjectural building could not have been of great standing is provided by the presence of ovens outside the enclosed area. Nothing can be said about the economic background of this occupation, which seems to have lasted without interruption for more than a ce.ntury; it would not be unreasonable to think of the occupation as a small farm14 so long as it is clearly understood that this is conjectural. No firm conclusions can be drawn, either, from the apparent end of the occupation in the third century A.D.; for the site may well have continued to be settled beyond Severan times, even developed into something approaching the status of a villa which would not be at all 12 G. Berau, 'Excavations at Little Woodbury, Wiltshire', PPS, vi (1940), 64 and 104-6; H. C. Bowen and P. J. Fowler, 'Romano-British Rural Settlements in Dorset and Wiltshire', in (Ed.) C. Thomas, Rurat Settlement in Roman Britain, C.B.A.. Research Report 7, London, 1966, 45 • . 1􀃂 P. J. Tester and J.E. L. Caiger, 'Excavations on the Site of a RomanoBr1t1Sh Settlement in J'oyden's Wood, near Bexley', Arch. Cant., lxviii (1954), 167-82. 14 B. J. Philp, 'Roma.no-British West Kent, A.D. 43-100', Arch. Cant., lxxviii (1963), 74-82. The tabulated references to Stone Castle Quarry, Greenhithe, need amendment. 147 15 GREENHITHE REPORT surprising in view of the proximity of the site to Watling Street and transport facilities. If this were so, then the evidence for it must have been lost in the course of quarrying. Oarwulodunum Oharlton OGP Cobham Colchester Ool,dham Crayford D. Dover Fawkham Gillnnn Hermet Hojheim Hovmes J.W. Leiceste1· o. Oldbury 0. eh P. Ospringe THE FINDS Abbreviations and References C. F. C. Hawkes and M. R. Hull, Camulodunum, Oxford, 1947. F. C. Elliston􀃸Erwood, 'The Earthworks at Charlton, London, S.E.', J.B.A.A., xxii (1916), 125-91. J. A. Stanfield and Grace Simpson, Central Gaulish Potters, London, 1958. P. J. Tester, 'The Roman Villa in Cobham Park, near Rochester', Arch. Oant., lxxvi (1961), 88-109. M. R. Hull, The Roman Potters' Kilns of Ookhester, Oxford, 1963. T. Potter, 'The Roman Pottery from Coldha.m Clamp and its Affinities', Proc. of the Cambridge Antiq. Soc., !viii (1965), 12-37. J.B. Ward Perkins, 'An Early Iron Age Site at Crayford, Kent ', PPS, iv (1938), 151-68. J. Dechelette, Les Va.sea ceramiques ornes ik la Gaule ·romaine, ii, Paris, 1913. L. Murray Threipland and K. A. Steer, 'Excavations in Dover, 1945-1947', Arch. Oamt., !xiv (1951), 130-49. B. J. Philp, 'The Romano-British Farmstead at Eastwood, Fawkham', Arch. Gant., lxxvili (1963), 55-73. J.P. Gillam, 'Types of Roman Coarse Pottery Vessels in Northern Britain', AA4, xxv (1957), 1-40. F. Hermet, La Graufesenque, Paris, 1934. E. Ritterling, 'Das fruhromische Lager bei Hofheim im Taunus', Annalen des Verein.s fur Nassauische Altertumskunde, xl (1912), Wiesbaden, 1913. J. Holmes, 'Romano-British Cemeteries of Haslemere and Charterhouse', Sy. A.O., Ii (1950), 1-28. P.J . Tester andJ.E.L. Caiger, 'Excavations on the Site of a Romano-British Settlement in Joyden's Wood, near Bexley', Arch. Oamt., lxviii (1954), 167-83. K. M. Kenyon, Excavation.s a-t the Jewry Wall Site, Leicester, Oxford, 1948. F. ·Oswald, Index of Figure-Types on Terra Sigillata, i-iv, Liverpool, 1936-7. J.B. Ward Perkins, 'Excavations on the Iron Age Hillfort of Oldbury, near Ightham, Kent', Archceologia, XC (1944), 127-76. F. Oswald and T. Davies·Pryce, An Introduction to the Stud,y of TerraSigiUata, London, 1920. W. Whiting, W. Hawley a.nd T. May, Report on the Excavation of the Roman Oemetery at Ospringe, Kent, Oxford, 1931. 148 Richborough I-IV Southwark Stone SW(JR'ling Upchurch Wela Winchester AA' Arch. Oant. J.B.A.A. JRS PPS Sx. A.O. Sy.A.O. l. Iron Age GREENHITHE REPORT J. P. Buahe-Fox, Jj):vca,vations of the Roman Fort at Richborough, Kent, Reports I-IV, Oxford, 1926-49. K. M. Kenyon, Excavations in Southwark, 1959. M. A. Cotton and K. M. Richardson, 'A Belgic Cremation Site at Stone, Kent', PPS, vii (1941), 134-41. J. P. Buahe-Fox, Excavation of the Late-Oeltic Urnfield at Swarling, Kent, Oxford, 1925. I. Noel Hume, 'Romano-British Potteries on the Upchurch Marshes', Arch. Oant., lxvili (1954), 72-90. P. Karnitsch, Die Reliefsigillata von Ovilava (Wela, Oberosterreich), Linz, 1959. B. Cunliffe, Winchester Excavatiom, 1949-1960, Winchester, 1964. Archreologia Aeliana, Fourth Series. A1·chreol,ogia Oantiana. Journal of the British Archreological Association. Journal of Roman Studies. Proceedings of the Prehistoric Society. Sussex Archooological Collections. Surrey Arclueological Collections. A. COARSE PO'J.'TERY16 (Figs. 3-18) Figs. 3 and 4 (nos. 1-33) describe a distinctive class of vessels of the Wealden culture which were found mostly in the upper back-filling layer in Pit 2, in Pit 3, in the gully of Enclosure A, and in some other back-filling layers elsewhere on the site; where not otherwise stated, Pit 2 is the find-spot. The vessels represented in this series are jars with mouths of varying width and a characteristic base or foot-ring; the inner surfaces of some of these vessels show unmistakable signs of hand-making whereas others with smooth inside surfaces were clearly wheel-made.16 Practically all this material had been heavily burnt after breakage and shows many changes from the original colour of the fabric but, as these changes did not penetrate beyond a few millimetres, it is clear that, originally, all these vessels were either light brown or brown, very rarely dark brown, with surfaces polished outside and to about half an inch inside the rim; their paste was invariably grey or dark grey and, without exception, very sandy. With the exception of nos. 1-3 which are large wide-mouthed ia In the following descriptions, paste is used to denote the core of the vessel, and fabric describes the finished vessel as for texture and colour. The finds have been deposited at Dartford Museum. 16 For a. similar mixture of band-made and wheel-made vessels and comments, see Oldbury, 148-4. 149 GREENHITHE REPORT storage jars, the remainder are jars of various sizes; except for no. 6, they all present the typical S-profile and foot-ring of the Oldbury 'foot-ring bowls' (Olabury, 143) and are strictly paralleled by similar material found at that site and associated with the construction of the primary defences (Olabury, fig. 12), and at Crayford (Crayford, 163, fig. 9, nos. 1-4). Nos. 25 and 26, in a finer, less sandy paste, are decorated with a curvilinear ornament, like a scroll, and recall some decorated vessels from Crayford.17 The foot-rings, nos. 28-31 and 33, clearly belong to the jars of this series, but it was not possible to reconstruct complete sections; no. 33, and another similar foot-ring from the area of the hearth, has a cross-like ornament made with a burnishing tool on the outside of the foot-ring (Ol,dbury, fig. 12, no. 11; also, at Caburn, Bx. A.G., lxviii (1927), pl. xvi) and shows the same stress marks as another Oldbury vessel (OldJYUry, fig. 12, no. 12). Nos. 21 and 24 were found in layers which were used as back-filling to make up the level in the deep excavation outside Ditch I, but nos. 12, 19 and 23 were recovered from the debris filling Pit 3 and the gully round Enclosure A. The vessels illustrated in Fig. 5 (nos. 34-48) form a collection of miscellaneous wares found in Iron Age contexts. Except for no. 34, which is much better made and close to the 'foot-ring bowls' of Oldbury, their main common characteristic is the coarseness both of the paste and the manufacture; they were all hand-made, pitted and heavily shell-gritted. However, as was the case at Oldbury,18 these vessels are strictly contemporary with the foot-ring jars with which they were found stratified. 34. Jar in brown fabric and grey paste, ·with a fairly smooth outer surface which shows traces of polishing, but is coarse and pitted inside; hand-made. This is probably of the same class as the foot-ring jars above (Oklbury, fig. 12, no. 8), but in a rather different fabric (Gully, Enclosure B). 35. Jar practically rimless, with a dark grey, matt fabric and heavily shell-gritted paste; hand-made (Gully, Enclosure A). 36. Jar, rimless, in reddish brown fabric and heavily shell-gritted paste; hand-made (Gully, Enclosure A). 37. Jar with a slightly everted, recessed rim, in red, matt fabric and heavily shell-gritted paste; hand-made (Gully, Enclosure A). 38. Jar, slightly beaded rim, with a dark grey, matt fabric and heavily shell-gritted paste; hand-made (Pit 3). 39. Jar, rimless, fabric and paste as for no. 38; hand-made (Oldbury, fig. 13, no. I) (Pit 3). 40. Jar, slightly recessed rim, with dark brown, matt fabric and heavily shell-gritted paste; hand-made (Hearth). 17 OrayfSs were found, in various fabrics; except for one sherd from Pit 1 (no. 136, below), they were all found in Layer 17. 133. Grey, smooth fabric, burnished inner surface; a shallow platter imitating samian Form Lu. Sb (Ookhester 310, Hadrianic). 134. Red fabric, imitating samian Curle 15. 135. Soft red fabric, imitation of samian Form 44 (Gilu,,m 200, A.D. 140-200). 136. Brown, smooth fabric, decorated with bands of small notches, probably applied with a roulette, imitating samian Form 37; several pieces of this vessel were found in Layer 17 and one sherd, conjoining with the remainder, in Pit 1 (Gillam 197, A.D. 140-200; Ool,cliester 330, Trajanic). (xii) Lids. 49. Dark grey fabric, sandy paste, grooved (Southwark, fig. 19, no. 13, A.D. 160-180; Leicester, fig. 31, no. 2, A.D. 150-200) (Layer 2). 50. Grey-black fabric and paste, heavily grooved body (Richborough III, 316, A.D. 80-120) (Layer49). 51. Light grey fabric, sandy paste (Leicester, fig. 31, no. 8) (Layer 33). 86. Dark brown-black fabric, reddish brown paste (Southwark, fig. 19, no. 1, A.D. 100-160) (Pit 1). 131. Dark grey fabric, coarse paste (Southwark, fig. 19, no. 13, A.D. 160-180; Leicester, fig. 31, no. 2, A.D.150-200) (Layer 17). 132. Grey fabric, gritty paste (Gilu,,m 340, A.D. 100-140) (Layer 17). (xiii) Miscell,aneous Vessels. Under this section are described severa,l vessels in a variety of wares, which do not belong to any of the previous sections on account either of their form or of their probable use. 67. A diminutive vessel in soft grey paste, now burnt pink, with upright rim and cordon at the base of the neck; to all intents and purposes, the form of this vessel is that of a 'poppy-head' beaker, and it is not unlikely that its present colour is due to much burning (Layer 47). 98. A dish with vertical wall, with a rudimentary rim which projects very slightly inside the vessel and is grooved at the top; the vessel is made in a buff paste and bears traces of soot inside-handmade (Pit 1). 142. A small jar or beaker in polished fabric, with a dark grey neck and rim and light grey body; the rim is rather hooked and the vessel is decorated with groups of dots applied en barbotine (Richborough, Ill, 178 GREENHITHE REPORT 􀀈􀀉􀀊􀀋--=·-·; i􀀋􀀌 -- i_ _n:1 211 [ I I: I l I z '1228 ,2JO l Fro. 17. (i) 179 17 -----· --- '" -'\, ·71--·------------------·- ---􀀥 ,.L --􀀐 --: ·--· -- -) i \ I - ' F------ -------\ -----)- -. GREENHITHE REPORT 278, A.D. 70-100; Leicester, fig. 27, no. 23, A.D. 80-120, for such vessels with barbotine decoration) (Pit 1). 207. A small beaker in pink, fairly soft fabric and paste, copying similar vessels in colour-coated wares ( cf. Colchester 411 ; Richborough I, 55, first-second century A.D.) (Layer 17). 209. A small beaker, in pink soft fabric and paste, with bands of faint rouletting round the body (cf. Ospringe 380, but in different fabric; Leicester, fig. 27, no. 38, A.D. 220-250, with different rim) (Layer 17). The following four vessels may have been in use as storage-jars, but this is not certain. 146. Jar in Patch Grove ware, reddish-brown leathery fabric and brown paste; the outer surface is decorated with burnished hoops, the inner is uneven and pitted-probably hand-made (Oldbury, fig. 14, no. 14; J. W., fig. 3, no. 7) (Layer 17). 203. Probably a storage vessel with neck, in Patch Grove ware, reddish-brown fabric and grey paste, uneven (Richborough III, 248; J. W., fig. 3, no. 5) (Layer 17). 211. Very similar in form to no. 146 above, but in grey-blue fabric and sandy paste, burnished around the neck and shoulder and above the base (Layer 17). 237. Jar in Patch Grove ware, light brown leathery fabric, dark brown paste (Oldbury, fig. 14, no. 14; J. W., fig. 3, no. 7) (Layer 11). (xiv) Mortaria. By lVIrs. K. F. Hartley, B.A. Made in Kent or Colchester. 97. In hard, slightly greyish cream fabric with flint grit. The nearest parallels are from Colchester (Colchester, fig. 63, no. 1) and Canterbury (Audrey Williams, Roman Canterbury, fig. 7, no. 8); c. A.D. 130-185 (Pitl). 212. Two pieces conjoining. The fabric has been heavily burnt and so appears greyish cream, or nearly black in places, but it was clearly cream originally; c. A.D. 140-190 (Layer 19). 213. Several fragments conjoining. Soft, off-white fabric with pinkish core and flint trituration grit. Stamps from the same die as this rather unusual herringbone type have, so far, only been found in Kent (Canterbury, Ham Saltings, near Upchurch, and Richborough), and manufacture there is highly probable, perhaps in the Canterbury area. In the second half of the second century, and in the third century, potters working at Colchester and at Canterbury (Colchester,· Arch. Oant., liii (1941), 118 ff.; lxxiv (1960), 158, fig. 5, no. 21; J RB, 1 (1960), 236), are known to have been malting mortaria, not only in identical fabrics, but with similar rim-forms and often using pattern-sta.mps of herringbone type. They were clearly working in the same general 48P ,. 00 ,. Fro. 18. (t) GREENHITHEREPORT tradition and their work cannot be distinguished except when stamps and their distribution patterns allow mortaria to be traced to one or other source (Bulletin of the Institute of Archa?,07,ogy, no. 5 (1965), 35-7). All of the mortaria in question (nos. 97, 212 and 213) are in this fabric (the occasional pink core is due to variations in the firing processes), and they can all be attributed either to Colchester, Canterbury or unknown sources in Kent. Although there is overwhelming evidence to show that production at Colchester was on much the larger scale, it would be reasonable to expect that most of these were from the more local source or sources. Dating evidence for this type of mortarium is limited. The most significant factor is the presence of large numbers at Antonina sites in Scotland (Colchester, 114-16). There is no conclusive evidence fixing the initial date for work in this tradition since more than one generation of potters· is involved, and typological evidence suggests that it could well be as early as A.D. 130 (e.g. no. 97). Ma.de in the V erulamium region 214. In sandy, grey fabric with flint trituration grit. Traces of internal scoring could indicate an earlier date than the fragment described below, but the mortar:ium was made in the same area. The surface is burnt. c. A.D. 85-115. (Not illustrated.) The sandy, orangy fabric with drab slip, and the form, similar to those used by Matugenus and Melus of Brockley Hill, Herts., point to manufacture in the Verulamium region. The piece is somewhat eroded. c. A.D. 90-130 (Pit 1). B. SAMIAN WARE (Figs. 19-21) The majority of the samian ware found was recovered in Pit 1 and Layer 17, but there were also several sherds in various layers of back-filling which, though of rather earlier date, are clearly rubbish survivals.· (i) Plain Forms. Form 27 was the predominant form of the various samian cups and is represented by fragments of eleven different vessels; the early Form 24 is entirely lacking and only a few pieces were found of Form 33. Plates are not at all well represented; three small fragments only were found of Form 15/17, and only three vessels of Forms 18 and 18/31 have been identified. The later Forms 31 and 31R are totally absent. Form 27 {not illustrated): (a) One sherd from a Central Gaulish cup which, on the evidence of its foot-ring section and encircling groove could ·be of Flavian date (0. &, P., pl. xlix, 15) (Pit I); (b) Three fragments belonging to rather small cups, probably of South 182 GREENHITHE REPORT 1 I. 􀀂 􀀃' 􀀄 183 GREENHITHE REPORT Gaulish origin, with rudimentary rims and internal groove below the rim, which could date to the Flavian period or later (0. & P., pl. xli.x, 7) (Layers 10 and 20); (c) Three other sherds from Central Gaulish vessels (0. & P., pl. xlix, 11), dating to c. A.D. 120-140; (d) One fragment from a small Central Gaulish cup, with inner groove below the rim (0. & P., pl. xlix, -17), datable to c. A.D. 95-120 (Layer 17); and (e) The foot-ring of a small South or Central Gaulish cup, with encircling groove and a spiral in lieu of the potter's stamp (Layer 17). 1 and 14. Central Gaulish. A large part of a very glossy Form 27 cup, stamped AM( ) ; the potter's name cannot be restored with certainty. The upper concavity of this vessel shows a rather flattened profile, and this cup could be one of the later examples of this form (Leicester, fig. 6, no. 12, Hadrianic) (Layer 17). 2. Form 27. South Gaulish. A fragment from the base of the cup, stamped (BI)RACILLVSF, by the well-known Banassac potter, c. A.D. 95-120 (B. Hofmann, Essai de Datation de la Ceramique sigillee de Banassac, RCRF Acta VII (1965), 39-65) (Pit 1). Form 33 (not illustrated): Two fragments, both Central Gaulish, both of Hadrianic-Antonine date (Layer 17 and unstratified). Form 15/17 (not illustrated): One fragment of South Gaulish provenance which could be Flavian or earlier (Layer 7); two other fragments conjoining, which are Vespasianic or later (0. & P., pl. xliii, 35) (Layer 10). Forms 35 and 36 (not illustrated): One sherd of the smaller Form 35 (0. & P., pl. liii, 3, Flavian) (Layer 50). Two vessels of Form 36: (a) One piece of a very shallow vessel, with a slight groove on the outer surface of the body, but no barbotine leaves (cf. 0. & P., pl. liii, 7, Trajanic) (Layer 11); (b) Two sherds from a very glossy Central Gaulish vessel, with internal groove below the rim (0. &; P., pl. liii, 10 and 15; Trajan-Hadrianic) (Layer 17). Form 18 (not illustrated): Central Gaulish. Most of one plate of this form was found, but the stamp is missing; there is a slight groove internally round the body of the vessel (0. & P., pl. xlv, 13, Flavian; Leicester, fig. 6, no. 6, Trajan-Hadrianic) (Layer 17). 3. Form 18/31. Central Gaulish. Very glossy. By the potter IVNIVS whose stamp IVNI · M is impressed across the central kick. Oswald's dating of this potter's activity to the Flavian period is too early, as it is amply demonstrated by the associated wares; moreover, this die was recently found at Lezoux28 stratified with the early work of SACER, ATTIANVS and DRVSVS II, which would give a range of activity from about A.D. 125-160, agreeing with the general date of this deposit at Greenhithe (Layer 17). 2s I owe this information to Mr. B. R. Hartley, F.S.A. 184 GREENHITHE REPORT 9 A/;o FIG. 20. (½) 185 GREENHITHE REPORT 12. Form Curle 11. Central Gaulish and in very good condition (very close to 0. &: P., pl. lxxi, 19, Trajan-Hadrianic). Another vessel of this form is represented by several small fragments of its flange only, which has small barbotine leaves in groups of three all round it; O. &: P., pl. lxri, 18, is the probable shape (both from Layer 17). 13. Form Ritterling 12 is represented by one sherd of a South Gaulish vessel of this form, with a groove about the middle of the external surface of the body and a small horizontal flange (0. &: P., pl. lxxi, 4) (Unstratified). (ii) Decorated Forms. Fragments of five decorated vessels were found; Form 30 is entirely lacking. 4. Form 37. Central Gaulish, in good condition. Several fragments of this bowl in the style of CINNAMVS were found. The decorative scheme consists of a winding scroll, with birds and vine leaves (Wels, Taf. 72/4), and is initiated by his ovolo no. 4 (OGP, p. 266, fig. 47) over an enclosing bead-row border; the vine leaves concerned are not his detail no. 38 (but cf. Wels, Taf. 72/4). The decoration contains four small birds (D.1019=0.2252, D.1038=0.2315, 0.2298 and 0.2239B) which are well attested in this potter's signed work, his large astragalus (detail 41, CGP, p. 266, fig. 47) used as a binding at the junctions of the scroll, and his eight-petalled rosette (detail 26). Date: c. A.D. 140- 195 (Layer 17). 5. Form 37. Central Gaulish. A small fragment from a bowl which could be in the style of IOENALIS who has used such horizontal wreaths (CGP, p. 37, fig. 10, no. 8, and pl. 40/462) and the floral detail (GGP, p. 37, fig. 10, no. 22). Date: c. A.D. 100-120 (Pit 1). 6 and 7. Two fragments from a bowl of South Gaulish origin, probably ·Form 37 rather than 29, with remnants of a horizontal wreath between wavy-line borders (Hermet, ii, pl. 35 bis, no. 9) and two figure-types: Hare to right (0.2072) and Dog to right (0.1924) used by several South Gaulish potters. Insufficient remains of the decorative scheme to allow a positive attribution, but this vessel could be ofVespasianic date (Layer 11). 8-10. Form 37. Central Gaulish. Six fragments are illustrated of several from a bowl in the style of DOCILIS. His ovolo no. 1 (CGP, p. 176, fig. 24) over a bead-row border and a decoration of panels; decorative details present are his rather elongated astragalus at the end or across the vertical 􀄊ead-row borders demarcating the panels (no. 2), his pillar (no. 16) and floral detail {no. 12). The figure-types are: Mars (D_.94=0.151) (OGP,pl. 91/1), Dancer {smalle;rthan D.372=0.365) (GGP, pl. 91/9) and Hercules (in size intermediate between D.449° =0.775 and D.449=0.774). Date: o. A.D. 130-150 (Layer 17). 11. Form 37. Central Gaulish. Three fragments from a bowl with very 186 /2 s &.S. Flo. 21. (½) GREENHITHE REPORT poor relief, due probably either to poor manufacture or a worn mould. The ovolo is mostly removed in the finishing process, but looks very similar to an ovolo used by the Potter of the Rosette, IOENALIS and DONNAVCVS; the panel decoration is divided by bead-row borders ending on seven-beaded rosettes except where an astragalus conceals the junction of border and half-medallion. The narrower panels are filled with the well-known floral ornament, favoured by Trajanic Central Gaulish potters, and the larger are sub-divided into two compartments. The two certain figure-types are Cock to left (D.1025b= 0.2361) and Apollo (D.52=O.83); the two animal figure-types may be Dog to left (0.20390) and Hare to left (O.2129A), but their restoration is by no means certain. This bowl could have been made by any one of the three Trajanic potters mentioned above, but on stylistic grounds this seems unlikely; the whole scheme of the decoration is less overcrowded than was usual with the Trajanic potters of Central Gaul, the panel decoration and the sub-division of these panels, the presence of a basal line instead of the characteristic Trajanic basal wreath, all these admittedly typological factors tend to suggest that this is a transitional decoration, which cannot be attributed with any certainty. Date: c. A.D. 100-150 (Layer 17). C. Corns By D. F. Allen, B.A., F.B.A., F.S.A. The three coins submitted belong to a type known as 'tin' or 'speculum' coins, Class I. One of these coins was in good condition, but the other two were variously corroded; they weighed 29·1, 21·0 and 13 · 9 grains respectively. Coins of this type were in circulation for a very long time; they may have been in use as early as 45 n.c., they were certainly in use in 43 A.D., and they still looked very much the same. The individual specimen is, therefore, hard to date. Their true home is undoubtedly Kent, but they were also found, especially in hoards, along the Thames. One suspects that they belong to one of the ma.ny constituent elements of pre-Roman Kent, and perhaps an Iron Age one. (D. F. Allen, 'The Origins of Coinage in Britain: A Rea.ppraisal' in (Ed.) S. S. Frere, Problems of the Iron Age in Southern Britain, Institute of Archreology Occasional Paper No. 11, London (n.d.), 97-308.) D. GLASS By Miss D. Charlesworth, M.A., F.S.A. Several fragments of glass were recovered in various layers, but were too fragmentary for detailed description, except for the following: 188 GREENHITHE REPORT I. Three fragments conjoining of a millefiori pillar moulded bowl, in emerald green with streaks of opaque yellow and a patch of opaque red, This type is found mainly in early Roman levels, e.g. at Camulodunum c. A.D. 40-60. In south-eastern Britain, it is fairly widespread, but further north in sites not occupied earlier than the Flavian period is very rare (Layer 7). 2. Three fragments of blue glass. The type of vessel cannot be identified. The base fragment could be from a bowl of the same type as 3 below or from a flagon or two-handled flask. First century A.D. 3. Several fragments of a globular ribbed jar in blue-green glass, with a rounded, outfolded rim, and flaking iridescence. This is a variation of a jar type common c. A.D. 70-150 which seems to have been made in the middle Rhineland and is widely distributed in that area, northern Gaul and Britain. The difference lies in the formation of the rim, which is generally folded outwards to form a hollow tube and stands up like a collar round the top of the vessel (for a complete example, see Richborough III, pl. XV, no. 57) (Layer 17). 4. Rim fragment of a jar in blue-green glass, with flaking iridescence; rim folded inwards (Layer 17). 5. Two fragments of angular bottles in blue-green glass. Firstsecond century A.D. (Layer 17). 6. Fragments from rim of .beaker in colourless glass, rim ground (Unstratified). 7. One fragment from the edge of a window -pane (Ditch 1). E. Sl\'IALL Fnms (Fig. 21) 1. Bronze ligula, with incised bands of decoration, of a type very common on Romano-British sites (cf. B.M. Guide to the Antiquitie,s of Roman Britain, London, 1958, fig. 5, no. 4, and numerous other examples from many sites (Layer 17). 2. Shed first-year antler probably of roe deer (capreolus capreolus)29 perforated for use as a bodkin. 3. Bone rat-tailed spoon, very common on Romano-British sites (cf. London in Roman Times, London, 1946, pl. xlv, nos. 2-3, and many other examples) (Layer 17). 4. Bone pin, with decorated head (two other incomplete specimens) (Layer 17), 5. Part of bone ligula. 6. (Not illustrated.) Bronze bracelet (cf. London in Roman Times, pl. xl, no. 4). 28 Kindly identified by Miss J.E. King. 189 GREENHITHE REPORT (Plate III) 7. To left, lead weight, probably from a bronze steelyard (B.M. Guide to the Antiquities of Roman Britain, fig. 40, no. 11) or a bronze scale-beam (London in Roman Times, fig. 22, no. 2); to right, remnant of iron rod with a lead weight, probably from a steelyard. Below, knife bone-handle (London in Roman Times, fig. 19, no. 1) (Layer 17). ANIMAL BONES By Miss J. E. King. Most of the bones submitted are those one would expect to find in a domestic rubbish-pit and are mostly those of oxen (bos sp.), but sheep (ovis aries) or goat (capra hircus), horse (equus caballus), pig (sus scrofa), roe deer (capreolus capreolus), red deer (cervus elaphus) or fallow deer (dama dama), and dog (canis familiaris) are represented by small numbers of bones; also, one fish bone (pleuronectes platessa). To get an idea of the size of the oxen, the measurements of the bones have been compared with those of the Chillingham and Chartley bulls in the collection of the British Museum (Natural History). These show that, in most instances, the bones came from a larger, rather than a smaller, type of animal. Two of the ox metatarsals showed signs of an arthritic condition of the joint. 190