A Prehistoric Site off Green Lane, Whitfield, near Dover

A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER KEITH P ARFlTT During Apri 1 1992 workmen removed (for re-use elsewhere) the plough-soil from an area measuring some 85 x 15m along the northwestern edge of a large uncultivated field adjacent to Green Lane, on the outskirts of Whitfield (Fig. 1). This process partially exposed the natural clay, sealed by a layer of sub-soil. In May the stripped area was inspected by members of the Dover Archaeological Group in search of Lower Palaeolithic artefacts, examples of which had been previously discovered in the same field (Hutchinson 1976). Further pieces of such early lithic material were recovered (Halliwell and Parfitt 1993) but examination of the stripped area also indicated the presence of later archaeological remains, including a significant number of cut features producing Belgic-style pottery. Clearly, part of a previously unknown prehistoric site of some significance had been revealed and a more detailed investigation of the area was undertaken over the next two years. A series of hand-dug trenches cut through the sub-soil layer to the top of the natural clay was employed to determine something of the nature, date and extent of the features located (Figs 2 & 3). No attempt at total excavation of the site was made. Situated above the north-east side of the Dour valley, on the socalled '400 Foot Plateau', within the parish of Whitfield, the field (O.S. Parcel No. 5804) stands on natural Clay-with-Flints over Upper Chalk at an elevation of about 128m above OD. (NGR TR 2942 4502.) In all, three separate periods of archaeological activity were identified in the area investigated - 1) Lower Palaeolithic; 2) NeolithicBronze Age; 3) Late Iron Age. Inspection of the prehistoric flint material showed that there were two distinct industries, of widely differing dates, present on the site. These constitute Periods 1 and 2; no definitely associated features were recognised. Period 3 was represented by a series of pits, post-holes, ditches, gullies, areas of metalling (Figs 2 & 3; Table 1), 373 / COLDRED / / / / J TEMPLE EWELL RIVER Old Park "'-. Road jDOVE􀀍 -􀀎. '· -.. Parish boundary _,,..\/. 'v \ 􀀄 I I I 􀀃 \ i _... fil \ 􀀉 GUSTON p \􀀂 \\ 1 Km Fig. 1 Location of Green Lane and other archaeological sites in the area: l) Knights Templar Preceptory; 2) Honeywood Road; 3) Whitfield Recreation Ground; 4) Church Whitfield crossroads. A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER pottery and other finds, all indicative of an occupation site, the full extent of which was not revealed. Period I: Lower Palaeolithic Activity The oldest, heavily patinated, prehistoric flint material recovered consists of some 320 individual pieces and includes more than twenty hand-axes of Lower Palaeolithic type, together with over sixty scrapers and other utilized flakes. This early flint assemblage is ultimately derived from the top of the natural Clay-with-Flints and a significant number of artefacts were discovered in situ, including five handaxes. Larger quantities of such flints, however, were recovered as residual material in later features and deposits, particularly the covering sub-soil and overlying plough-soil. Clearly indicating some sort of activity or occupation in the immediate area, the significance of these Lower Palaeolithic finds has been considered in a previous report (Halliwell and Parfitt 1993) and local research has continued (Parfitt and Halliwell 1996; Halliwell and Parfitt 1998). Period 2: Neolithic-Bronze Age Activity Recognisable by its essentially unpatinated state and relatively crude techniques of production, the later flint industry is represented by some 168 pieces. These largely consist of unworked waste flakes, with seven cores and core fragments. The flints were found in the subsoil and overlying plough-soil and as residual material in many of the late Iron Age features. Produced from locally collected Downland flint, there are few diagnostic tools within the assemblage but nine scrapers, three piercers and a notched flake may be identified, together with eight other flakes showing signs of utilisation. With the exception of one of the scrapers, all the worked pieces have been \ roughly fashioned, without any degree of precision. The overall character of these flints, with their general Jack of patination and fairly crude production method, suggests that the majority are broadly datable to the late Neolithic-Bronze Age period, 􀂂!though one or two earlier pieces might also be present. No associated cut features or deposits were positively identified within the area investigated but at least casual occupation in the region must be suggested. Lithic material of this general date can be found covering much of the Downland around Dover and significant quantities were recovered ahead of the construction of Dover's Eastern Bypass, Which passes close-by the present site (Gaunt, Parfitt and Halliwell 1977, 200). 375 GREEN LANE 10 c- --- - ¼J I / F􀀂 10 0 10 20 30 Metres Fig. 2 Overall site plan showing excavated trenches, metalled areas [Fs 12, 27 & 37) and late hollows [Fs 5 & 19). DI J ,-􀀆 :,= E 􀀇􀀈􀀉 􀀂-􀀃 \􀀅.35l- ) F.34 S21 F.31 _../ 10 0 10 Fig. 3 Overall site plan showing excavated ditches, gullies, pits and post-holes. 20 30 Metres V, -..J 00 SW : topsoil Section No. 22 SW 1' s-ubs-:oil - 􀀂- ........ .. . Metalling F .12 0 31 NE 2 Metres SW -J, topsoil subsoil NE Section No. 13 * NE 4 ,£ * 13 Section No. 3 NW SE 􀁒ubsoil - 2􀁓 - - - ; 􀀂7 Section No. 4 SW NE : s:b􀁔il- - - ;.; - - - - - - 􀁕 23 Ditch F.10 24 Section No. 8 Fig. 4 Sections across excavated ditches and metalling. A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER Period 3: Late Iron Age Occupation In addition to the prehistoric flints (see above), some thirty-six features were found cut into the top of the natural clay, sealed below the sub-soil deposit (Figs. 2 & 3; Table 1). These features produced a total of more than 1,300 sherds of late Iron Age to immediately post-Conquest 'Belgic' style pottery. About two-thirds of this material came from seven ditches and gullies found running across the north-eastern end of the cleared site (Fig. 3). These probably formed parts of a series of enclosures, but apparently represented at least three successive phases of development. The central part of the site was occupied by an extensive spread of deliberately laid flint metalling [F. 12], the full extent of which was not revealed (Fig. 2). Most of this deposit was left in situ but it was found to partially seal three inter-cutting ditches [Fs 2, 3 & 4; Fig. 4, Section Nos. 3 & 13). Although seemingly fairly late in the sequence, there is no reason to believe that this metalling is not broadly contemporary with the dated late Iron Age features. The south-western part of the site was occupied by a scatter of pits and hollows (Fig. 3), whose fillings were often found to be devoid of any datable material, although a number were found to contain small amounts of late Iron Age pottery. A few, isolated post-holes were also revealed. Ditches and Gullies [Fs 2, 3, 4, 7, 8, 10 & 11] Confined to the north-eastern part of the cleared area, a total of five ditches [Fs 2, 3, 4, 7 & I 0) and two gullies [Fs 8 & 11] was recorded (Figs. 3, 4 & 5). All but F. 10 were aligned on similar axes, either NNE-SSW or WNW-ESE. Of particular interest were three parallel, inter-cutting ditches [Fs 2, 3 & 4], partially sealed by the extensive metalled area, F. 12 (Fig. 4, Section Nos. 3 & 13). Although running upon similar axes, it was clear that these three ditches were not all of the same date. Ditch F. 2 had been replaced by F. 3; when in-filled, F. 3 had been cut into by F. 4 (Fig. 4, Section No. 22). Thus, they appear to represent successive replacements of a single boundary ditch, which had not been continued as far to the east in its later phases (Fig. 3). Much of the pottery recovered from the site was contained within these inter-cutting ditches [Assemblages 1-3), together with a significant amount of Chaff-tempered ware (see below), residual prehistoric flints and some animal bone. Gully F. 8 appeared to continue the line of the western edge of the metalled area F. 12, perhaps implying that the two were contemporary (Fig. 5, Section No. 12). However, the orientation of most of the 379 tJ.) 00 0 NE l topsoil 1 subsoil NW Section No. 19 SW N S ------- - 21 0 GullyF.11 Section No. 11 Section No. 9 NE SE 2 Metres Section No. 16 SW Fig. 5 Sections across excavated gullies and pits. SE NW 􀀎ullyF.8 Section No.12 SW NE Section No. 21 A PREHJSTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER ditches was similar to that of the metalling, implying that they had all been laid out following a common plan (Figs. 2 & 3). Cutting through an earlier ditch [F. 7], ditch F. 10 was aligned NW-SE and was one of the largest revealed (Fig. 4, Section No. 8). It produced just two pot-sherds, both of Belgic type and the date of its filling must remain uncertain. The size, axis and stratigraphic position (Fig. 3) would allow this ditch to be of a later date and unrelated to the main excavated complex but, in the absence of any clear evidence for subsequent activity on the site, it is regarded here as being broadly contemporary. Pits and Hollows A series of twenty-six pits and hollows of varying sizes were recorded, mainly across the south-western part of the site (Figs 3 & 5). These fall into two general types: small round and oval pits (total, 12) and large, often ill-defined, shallow pits and hollows (total, 14). Some of the larger pits may have been dug to extract clay, perhaps to be used to provide raw material for daub or pottery manufacture, or possibly flints for metalling. More than half the pits and hollows failed to yield any datable material and few contained significant quantities of pottery. Circular pit F. I, located between ditches towards the north-eastern end of the investigated area, yielded twenty-nine late Iron Age pot-sherds. At the southwest end, a large, deep pit [F. 32] produced seventy-five sherds also of late Iron Age date [Assemblage 4] and a quantity of Chaff-tempered ware (see below). It was cut by a smaller pit [F. 31 ], which produced two flint-tempered sherds (Fig. 5, Section No. 21 ). Nine other pits produced small amounts of pottery also broadly of this date. Large pit [F. 26] was cut by a smaller one [F. 25] (Fig. 5, Section No. 19) and was also partially sealed by a patch of metalling [F. 27). The undated pits are assumed to be of a similar age to the dated examples but this cannot be certain. Two hollows [Fs 5 and 19] are relatively late, being dug into the surface of metalled area [F. 12]. Hollow [F. 5] produced half-adozen late Iron Age sherds, which are likely to be residual. Most of the pits yielded varying amounts of calcined flint in their fillings, sometimes in association with flecks of carbon, indicating that the material was derived from cooking fires. No evidence for burning on the sides or bases of the various features was noted, however, so it seems unlikely that any represent actual hearth-pits. Post-holes { Fs I 3, 29 & 33 J Three small pits, seemingly representing post-holes relating to 381 I>) 00 N F. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Type Shape Pit Circ. Ditch Linear Ditch Linear Ditch Linear Hollow Oval Pit Circ. Ditch Linear Gully Linear Pit Circ/oval Ditch Li.near Gully Linear M'ling Sub-rect? P't-holc Circ. Pit Oval L'2e oit Irrel!'/oval L'gcoit Oval? Pit Circ. Pit Oval Hollow Oval? TABLE 1: PERIOD THREE FEATURES, POTTERY AND FINDS Length (m) Width(m) Ax.is Depth (m) Sides Base No./fabric sherds; {No.CTW frrumlents} Dia􀄃 0.43 - 0.10 Steep/slope Dished 29 IA2, B212.1; {3} 17.50 (min) 1.00 -1.30 WNW-ESE 0.22-0.60 Sloping Dished 117 Assemb. l· {11} 5.10 (min) 1.25 (min) WNW-ESE 0.83 Steep/slope Dished 353 Assemb. 2· {990} 8.70 (min) I.15 -1.75 WNW•ESE 0.40-0.68 Steep/slope Dished 280Assemb. 3· f44} 6.00 2.00(min) NE-SW 0.37 Steeo/slooe Dished 6 lAI B2 B8. Dia= 0.85 - 0.26 Sloping Dished No ootterv 19.00(min) 0.82 -1.25 NNE·SSW 0.44 Steep/slope Dished 53 lAl/2/4, B2/2.1/8; !2} 13.50(min) 0.70 NNE-SSW 0.35 Steep/slope Dished 82IA4,B2, B2.1 1.90 1.00(min) NE-SW 0.83 Steeo/slope I rreg. No ootterv 9.50 (min) 1.40 NW-SE 0.60-1.00 Sloping Round 2B2 11.50 (min) 0.56-0.68 WNW-ESE 0.20-0.30 Sloping Round 17 IA4,B2: {5} 32.00 13.00 (min) NNE·SSW 0.05-0.10 - - 1 IAl Dia = 0.32 - 0.10 Slooing Round No ootterv 0.92 0.76 E-W 0.20 Steep Dished 1 B8 8.00(min) 7.00 (min) NW-SE 035-0.50 Steeo/slooe Dished 14 Assemb. 5 2.75 0.65 (min) NE-SW 0.45 Steep/slooe Dished No potterv Dia = 0.80 - 0.10 Sloping Dished No ootterv 0.80 0.60 N-S 0.36 Steeo/vert. Flat l lAI 2.16 (min) 0.63 (min) NE-SW 0.25 Slooinll. Flat Nonotterv w 00 w F. No. 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Type Shape Pit Oval L'gepit Oval L'gepit Oval L'gepit Oval? L'ge oit Oval? Pit Oval L'gepit Oval? M'limr ? Pit Oval P't-hole Circ. Pit Oval? Pit Circloval L ' gepit Oval? P't-hole Oval Pit Oval? Pit Oval? Pit Oval? M'ling ? Pit Circloval Pit Ci rc/oval Length (m) Width (m) Axis 0.60 0.38 NW-SE 2.20 (min) 1.60 (min) N•S 3.05 0.60 (min) NE-SW 2.20 0.72 (min) NE-SW 1.70 (min) 0.45 (min) NE-SW 1.55 1.43 E·W 2.80 (min) l.90(min) NE-SW l.OO(min) 0.80(min) N-S 0.91 0.65 NE-SW Dia= 0.40 - 1.10 0.85 (min) NE-SW I.OS 0.69(min) NE-SW 2.65 0.85 (min) NE-SW 0.48 0.35 E-W 1.45 0.80(min) NE-SW 0.78 (min) 0.45 (min) E•W 0.80 0.40(min) E·W 5.00 (min) 0.80 (min) E·W 0.44 (min) 0.26 (min) NE-SW 0.66 0.30 (min) NE-SW Depth(m) Si des Base NoJfabric sherds; (No.CTW fral!Jnents} 0.12 Slooing Dished No ootterv 0.30-0.55 Slopimi Dished 2 IA2 B2.1 0.53-0.80 Steeo/slope Sloping S lAl, B8 0.61 Sloping Round 14 lAl, 1A4, B2 0.40 Sloping Round No ootterv 0.60 Steep Dished No pottery 0.50 Sloping Dished S 1A3, 1A4, B2 0.10 - - No ootterv 0.35 SIOl)ing Round Nol)Ottery 0.14 Slooing Round No ootterv 0.55 Steeo/vert. Round No ootterv 0.47 Steep Round 2 lAI 0.92 Sloping Round 75 Assemb. 4; 131} 0.14 Steeo/slooe Dished l 1A5 0.80 Steep/vert. Dished l 1A4 0.30 Steeo/vert. Dished No pottery 0.20 Steep Dished No ootterv 0.10 - - No oottery 0.11 Slooine. Flat No pottery 0.30 Steep/slope Dished No potterv KEITH PARFITT vanished timber structures, were recorded. These were widely scattered across the north-western side of the area investigated and no significant pattern can be discerned; other post-holes could have remained obscured below areas of uncleared sub-soil. F. 33 contained a single late Iron Age pot-sherd. Metalled Areas { Fs 12, 27 & 37] Partially sealing the three successive boundary ditches [Fs 2, 3 and 4] was an extensive area of laid flint metalling [F. 12]. The full extent of this was not revealed (Fig. 2) but it appeared to be sub-rectangular in shape and measured some 32m (NNE-SSW) by a minimum of 13m (ENE-WSW). It followed the same general orientation as the ditches located, indicating that it was broadly contemporary with the late Iron Age occupation of the site. Over much of its area the surface of the metalling was firm and well-defined but it was not level, dipping down to the south and east. Its purpose remains unclear - it could have provided the base for a timber building but more probably it was intended to give a firm out-door hard-standing or working surface, the great value of which becomes all too clear during periods of wet weather on this heavy clay site. Most of the metalling was left in situ but where excavated it was found to be 0.05-0. 1 Orn thick (Fig. 4, Section Nos 3 & 13). Making use of readily available raw material, it consisted of a mixture of natural flint fragments and nodules, locally collected small flint pebbles, calcined flints and some prehistoric flint-work, all set in brown clay. A single late Iron Age pot-sherd was recovered from its make-up. The soil sealing the metalling (GLW-92-13 & 21) produced forty-nine more pot-sherds, mostly of late Iron Age date but also including a fragment of early Roman necked-jar (see below). A piece of daub and a fragment of lava-stone quern was also recovered. Two smaller patches of metalling [Fs 27 and 3 7] were located at the south-western end of the cleared area (Fig. 2) but their full extent was not defined and it is not certain if they were contemporary with the main area. F. 27 partially sealed an earlier pit [F. 26]. 384 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER FINDS The archive for the project includes 80 recorded deposits, 39 recorded features (Table I), two site plans, 22 drawn sections and nine boxes of finds, mostly containing struck flints and pottery. No detailed analysis of the prehistoric flint-work has been included here (but see above). All of the finds from the site have been placed in store at Dover Museum, together with a copy of the field records. Pottery by Malcolm Lyne The site produced 1,349 sherds (I 0,603g) of late Iron Age to immediately post-Conquest Roman pottery from forty-one separate contexts. Most of the assemblages are very small and scrappy but the system of inter-cutting ditches at the north-eastern end of the site [Fs 2, 3 and 4) yielded an important sequence of somewhat more substantial assemblages spanning most, if not all, of the late Iron Age (Assemblages 1-3). Most of the pit assemblages are very small and limited to just one or two abraded sherds. Two of these features did, however, produce more substantial pottery assemblages (4-5). All of the assemblages were quantified by numbers of sherds and their weights per fabric. Fabrics were identified using a x8 magnification lens with inbuilt metric scale for determining the natures, sizes, shapes and frequencies of added filler inclusions. A numbered fabric series was drawn up for the calcined-flint and related mixed grit fabrics with the prefix IA: Belgic grog and sand-tempered, Gallo-Belgic imports and Romanised wares were given the codings created by the Canterbury Archaeological Trust for pottery in east Kent (Macpherson-Grant et al 1995). Fabrics IA] IA2 IA3 IA4 IAS IA6 BI 82 B2.1 B3 B8 Irregular handmade fabric with profuse ill-sorted up-to 5.00mm calcined-flint filler. Similar fabric but with sparse up-to 3.00mm calcined-flint filler. Polished black fabric with profuse silt-sized quartz and moderate to profuse up-to 1.00mm calcined flint filler. Similar fabric but with very sparse calcined-flint filler. Handmade fabric with profuse silt-sized quartz and very sparse up-to 0.50mm angular alluvial flint grit and limestone filler. Handmade fabric with profuse up-to 0.10mm quartz and larger grog, plus occasional leached-out chalk inclusions. Belgic fine-grog-tempered ware. Belgic coarse-grog-tempered ware. Belgic coarse-grog-tempered ware with pale siltstone grog. Belgic grog-tempered ware with additional sparse flint. Belgic fine-sanded handmade fabric. 385 KEITH PARFITT BI 0.ELGB Micaceous Terra-Ru bra from Central Gaul. c. AD I 0 • 50. BERS Gallo-Belgic White-ware. R5 Canterbury coarse-grey sand-tempered ware. c. AD 70 - 175. Assemblage 1: From the fills of Ditch F. 2 [GLW-92-7, 8, 15, 76 & 77] The various fills of this feature produced a total of 117 sherds ( 1,396g) of pottery, which were quantified by numbers of sherds and their weights per fabric:- TABLE 2: DETAILS OF POTTERY IN ASSEMBLAGE 1 (DITCH F. 2) Fabric No. sherds % Wei11ht (g) % IA! 91 77.7 1094 IA2 6 5.1 38 IA4 10 8.5 80 B2 I 0.9 128 B2.l I 0.9 1 0 B 3 I 0.9 2 B8 7 6.0 44 Total I 17 (100) 1396 78.4 2.7 5.7 9.2 0.7 0.1 3.2 (100) The breakdown of this assemblage shows it consists very largely of sherds with variants of calcined-flint filler (91 %): 'Belgic' grog-tempered and sand-tempered fabrics are very much in a minority. This suggests that the assemblage belongs to the earlier part of the late Iron Age (c. 50 - I BC). The following forms came from the lower fills of the ditch (Fig. 6): Straight-sided bead-rim jar of Thompson (1982) type C l -3 in handmade patchy black/buff/orange Fabric IA I with an irregular finish. Ext. rim diameter 180mm. GLW-92-8. 2 Storage-jar with flattened rim in similar fabric but fired brownblack with polished surfaces. Ext. rim diameter uncertain but more than 300mm. GLW-92-8. 3 ?Saucepan-pot or hole-mouthed vessel in grey-brown Fabric IA2 polished inside and outside. GLW-92-8. 4 Slack-profiled jar in similar fabric with polished exterior. GLW-92-8. 5 Slack-profiled jar in grey Fabric IA4 fired polished black with orange margins. Ext. rim diameter 180mm. GLW-92-77. 6 Fragment from carination on Thompson (I 982) ?class EI cup in sandy grey-black Fabric B8. GLW-92-8. The upper fills of the ditch yielded the following (Fig. 6):- 7 Bead-rim storage-jar in rough, black Fabric IA I. GLW-92- I 5. 386 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER . . . I ..... ; . .. . 1 ' Q -􀀅-: -􀀎􀀏·::?':'i .. 􀀐:-" .. - ' ... - _.... ..... ' -􀀩􀀪 . '1. • 2 I i. .·-.· · .. ·.r-- ...:..--􀀍- ; , o 10 . . . .. · . . · .. · .··. . . . . 6 -===-===- . . . . . ' 5 ,􀀇􀀋 ,. ., . J-;-.. .. ; . ' 7 8 Fi 17 15 .17: 19 I /'t 20 ,􀀂. ' 21·.J 18 22 ..- --i===;:::;::;;:􀀊 ro 􀀃-\; 16 5 􀀫- I \ 23 Fig. 6 Pottery excavated from the site. 387 cm. KEITH PARFITT 8 Jar with weak everted rim in buff-brown Fabric B8 variant with profuse up-to 0.20mm multi-coloured sub-angular quartz filler mixed with glauconitic sand. Fired rough grey-black with brown margins. GLW-92-15. Assemblage 2: From the fills of Ditch F. 3 cutting Ditch F. 2 (GLW-92-9, 10, 74 & 75). The 353 sherds (4,020g) of pottery from the various sections across this feature were quantified in the same way as Assemblage I. TABLE 3. DETAILS OF POTTERY IN ASSEMBLAGE 2 (DITCH F. 3) Fabric No. Sherds % Weight (g) % IA! 47 13.3 472 11.7 IA2 6 1.7 18 0.4 [A3 30 8.5 160 4.0 IA4 19 5.4 146 3.6 IA6 I 0.3 8 0.2 BI 10 2.8 JOO 2.5 82 168 47.6 1452 36.1 B2.1 38 10.7 1484 37.0 B3 6 1.7 54 1.3 B8 26 7.4 122 3.0 B9 I 0.3 2 0.1 BERS I 0.3 2 0.1 Total 353 (100) 4020 (100) The fabric breakdown of this assemblage differs markedly from that from the earlier ditch in that there is a marked fall off in the percentage of flinttempered wares (29 % ) and a corresponding increase in that of 'Belgic' grogtempered and sand-tempered wares. The following forms are present (Fig. 6):- 9 Greater part of jar with corrugated neck of Thompson (1982) type B2-4 in brown-black Fabric B2- 1 with body furrowing and rivet holes where it has been repaired. Ext. rim diameter 1 90mm. c. 50 BC - AD 50. GLW-92-74. I 0 Much smaller similar vessel in pale-grey wheel-turned Fabric BI fired darker grey with burnished chevrons on the body and rivet holes where it has been repaired. Ext. rim diameter 1 40mm. GLW-92-74. J J Jar of Thompson (1 982) type B 1 -1 in brown-black Fabric B2. 1 . Ext. rim diameter 1 40mm. c. 50 BC - AD 70. GLW-92-74. 12 Small bowl or crucible in similar fabric. Ext. rim diameter 90mm. There are plant impressions on the exterior surface, including that of a fern frond. GLW-92-74. 388 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER 13 Hole-mouthed vessel of Thompson ( 1982) type C3 in grey-black Fabric B3. Ext. rim diameter 180mm. c. 50 BC - AD 50. GLW-92-74. 14 Small handmade bead-rim jar in slightly vesicular grey-black fabric IA6 fired orange internally. Ext. rim diameter 120mm. GLW-92-75. 15 Rim sherd from small barrel-shaped butt-beaker of King Harry Lane variant I A I in white Fabric BERS (Stead and Rigby 1989, 137-141). Ext. rim diameter 80mm. c. 10 BC - AD 10. Other examples of this very early Gallo-Belgic butt-beaker type are known from the Mill Hill cemetery, Deal and the Elms Vale cemetery, Dover (Parfitt 1995, 41). GLW-92-74. The presence of the latter vessel suggests a date of c. AD 1 - 20 for this assemblage. Assemblage 3: From the fills of Ditch F. 4 cutting Ditch F. 3 (GLW-92-28, 29 & 72). The 280 sherds (2, 18 l g) of pottery making up this assemblage were quantified in the same manner as the two previous assemblages:- TABLE 4. DETAILS OF POTTERY IN ASSEMBLAGE 3 (DITCH F. 4) Fabric No. Sherds % Weight (g) % IA! 133 47.5 630 28.9 IA3 26 9.3 90 4.1 B2 79 28.2 372 17.1 B2.I 20 7.1 1020 46.7 B8 2 0.8 33 1.5 Misc 20 7.1 36 1.7 Total 280 (100) 2181 (I 00) This assemblage has a high percentage of sherds in fabrics with calcined-flint filler (57 % ) but this may be misleading in that 108 of the fragments in Fabric IA! come from the upper fill (GLW-92-28) and may all be from one vessel. Without them the calcined-flint tempered sherds would make up less than one third of the assemblage. Fragments from the following vessels are present (Fig. 6):- 16 Greater part of bead-rim jar of Thompson ( 1982) type C4 in brownblack Fabric B2.1 with body combing. Ext. rim diameter 180mm. c. AD 30 - 100. GLW-92-29. 17 Rim from necked-jar in polished black fabric IA4. GLW-92-29. 18 Bead-rim jar of Thompson ( 1982) type ?C 1-2 in patchy black/red Fabric B8. Ext. rim diameter 130mm. c. AD I - 100. 389 KEITH PARFITT These vessels suggest a date of c. AD 30 - 50+ for the assemblage. Assemblage 4: From the fills of pit F. 32 (GLW-92-64 & 66). The 75 sherds (346g) of pottery from this feature is too small an assemblage for meaningful quantification. Nevertheless, the predominance of wares with calcined-flint filler (76%) suggests a date during the earlier part of the late Iron Age for this feature (Fig. 6). 19 Rim sherd from bead-rim jar in lumpy black Fabric IA2 fired patchy orange/black. Lower filling, GLW-92-66. 20 Fragment from another, somewhat cruder, bead-rim jar in similar fabric. Lower filling, GLW-92-66. 21 Rim sherd from small jar with rudimentary rim in a vesicular Fabric B2 variant fired grey. Upper filling, GLW-92-64. 22 Jar with rippled shoulder, of Thompson ( 1982) type B2-3 in buff-grey Fabric B2 fired smooth grey. Ext. rim diameter 120mm. c. 50 - 1 BC. Upper filling, GLW-92-64. Assemblage 5: From the fills of pit F. 15 (GLW-92-32, 33, 37 & 46). The 14 sherds (53g) of pottery from this feature is also too small an assemblage for any meaningful quantification but includes a fragment from an amphora of indeterminate form and the following sherd (Fig. 6):- 23 Rim from ring-necked flagon in dull red micaceous Central Gaulish Terra Rubra fabric BIO.ELG B. c. AD 10 - 50. GLW-92-46. A few early Roman sherds came from the uppermost levels on the site: they comprise three sherds from an indeterminate vessel in sandy grey Canterbury Fabric R5 (c. AD 70 - 175) from the top-soil (GLW-92-1) and a fragment of a necked-jar in similar fabric from the sub-soil (GLW-92-13) over the metalled area, F. 12. Chaff-Tempered Ware (CTW) by Keith Parfitt and Geoff Halliwell Around 1, I 00 fragments (2,980g) derived from small, crude chaff-tempered ceramic vessels, of the type first described by Macpherson-Grant (1980) and Barford ( 1982) were recovered from the site. Some 990 pieces (2,830g) came from the filling of boundary ditch [F. 3] with about 700 of these in a single deposit (GLW-92-74). Smaller amounts of material were recovered from the adjacent ditches, F. 2 (11 pieces) and F. 4 (44 pieces). A further twenty-three fragments were contained within a soil layer sealing the top of these intercutting ditches (GLW-92-71; Fig. 4, Section No. 22). Four other features also produced small quantities of material [Fs I, 7, 11 & 32], most notably a large pit [F. 32] at the south-western end of the site, which yielded thirty-one fragments. The date-range of this quite distinctive ceramic-type seems to be entirely confined within the first century BC to first century AD and it is consistently 390 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER associated with 'Belgic' grog-tempered pottery (Macpherson-Grant 1980). The evidence from the present site thus conforms with this general dating (see above). However, since this ware was first identified, the overall size, shape and function of the vessels represented has occasioned much discussion. Study has been constantly hampered by the fragile nature of the material and the consequently small size of the pieces recovered (Barford 1982). During the Green Lane excavations it appeared as if much of the material in F. 3 was derived from a single vessel, smashed as it was thrown into the partially filled ditch. Examination of this material confirms that around half the fragments come from the thin-walled, simple-rimmed vessels as originally detailed by Macpherson-Grant ( I 980). A roughly equal number of fragments, however, are much thicker, sometimes showing one or more roughly finished surfaces. These have more than a passing resemblance to fragments of burnt daub but their distinctive fabric indicates that this is not so. A number of key fragments reveal the relationship of the thin-walled container sherds to these thicker pieces and show that they come from the same vessels, although it has not yet been possible to reconstruct a complete profile with certainty. Conjoining sherds from ditch F. 3 (Fig. 7) demonstrate that at least some Chaff-tempered ware vessels at Green Lane consisted of the already recognised small rounded containers, to which had been applied a very crude, thick, roughly shaped strip encasing the rim. The purpose of this strip seems to have been to provide an attachment for horizontal protrusions or lugs, projecting outwards (perhaps similar to Macpherson-Grant J 980, No. 23, there suggested as a base). The strips and their lugs were added before any heating or firing took place, although the technique of fixing to the main body of the vessel was very poor, so that the two elements remained quite distinct. Re-examination of previously excavated assemblages of Chaff-tempered ware from other east Kent sites has identified fragments of similar applied strips with lugs in material from Wood Hill at Kingsdown and Cherry Lane, Great Mongeham (see Barford 1982). Identical new finds come from the Broom Bungalows site at Sutton (Tony Redding pers comm.). In several examples rim sections of the main vessel have actually fallen out of their enclosing clay strip, demonstrating the poor standard of attachment during manufacture and serving to disguise the original composite arrangement. The exact number of protrusions provided on a vessel remains unclear - anywhere between one and four seems possible. The conjoining fragments from F. 3 (Fig. 7) include one certain projection and perhaps the start of another. If correct, assuming that such lugs were regularly spaced, it would seem that there might have originally been four projections around the circumference of this vessel. Like the associated strips, the protrusions are extremely crudely produced and seem to vary in length, width and crosssection. The method of crudely applying the lugged strip over the top of the rim is technically most inappropriate for a lifting action, so it seems highly unlikely that the lugs can represent handles of any sort. This suggests that the protrusions had some other function. It remains far from certain that these lugs were applied to a rim that was actually at the top of the vessel. Could they, in 391 KEITH PARFITT I I I I I I _, Fig. 7 Eight conjoining sherds of a Chaff-tempered ware vessel showing applied strip over the rim and one protruding 'lug'. There is the possible base of a second lug on the edge of the surviving portion but this has largely broken away. Max. external diameter, c. 95mm. (GLW-92-74.) fact, be basal supports for open bottomed vessels? In this context it is of interest to note that recognisable vessel base fragments seem to be consistently absent from excavated Chaff-tempered ware assemblages. Barford has previously attempted a reconstruction with a single central projection at the base of the main container (Barford 1982, 205). Such an arrangement cannot be confirmed by any of the material seen by the present writers. It seems more likely that Barford's central pedestal is actually another fragmentary example of a crude, horizontally projecting lug, as described above. Against any argument for basal projections is the observation that the applied strips are generally very unevenly finished, so that they may not have been very stable on a flat surface. On the reconstructed specimen (Fig. 7) the surviving projection does not make contact with the ground when the main vessel is placed on a level surface; however, this may not have been of particular importance on the complete vessel. Whether applied protrusions occurred on all Chaff-tempered vessels remains unclear - more that one type of vessel may have been produced. It has been suggested that these Chaff-tempered ware vessels represent salt 392 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER containers. Given the inland location of Green Lane and other sites, like Sutton and Canterbury, such material can only have arrived here from production centres on the coast. However, the reason for the suggested link with salt production is based solely on the superficial resemblance of the chafftempered fabric to the briquetage regularly found at salt production sites. It seems increasingly likely that this supposed linkage is clouding the issue. Perhaps more probably, these vessels were purely domestic in character, locally made and completely unrelated to salt production. Other Finds (not illustrated) by Keith Parfitt Finds other than prehistoric flints, pottery and Chaff-tempered ware were not numerous. Sixty-four pieces of animal bone (mostly teeth) came from the filling of the three inter-cutting ditches (Fs 2, 3 and 4). The generally acid nature of the soil meant that this material was in very poor condition and it seems certain that much other bone had not survived. Part of an imported quern of German lava-stone was found in the soil over the metalled area (F. 12) and several small fragments of burnt daub were recovered from the ditches. A thorough detector survey of the site failed to yield any metal-work. The main area of metalling [F. 12) and the filling of most features produced varying amounts of calcined flint; indeed in a number of features this was the only material recovered. Although it seems possible that some of this burnt flint is derived from the earlier prehistoric activity represented by the struck flints, it appears likely that most is contemporary with the late Iron Age settlement, where presumably the hot stones had been used in domestic cooking. DATING AND DISCUSSION Established in a place where there had been earlier, Stone Age activity, it seems clear that this previously unknown late Iron Age settlement site was considerably larger than the area investigated, with features continuing in all directions (Figs. 2 & 3). A fairly extensive occupation site, presumably a hamlet or farmstead, is thus implied, although nothing of significance was recorded during the construction of Dover's Eastern Bypass in 1975, which passes some 70m to the south-west of the southern end of the investigated area (Gaunt, Parfitt and Halliwell, 1977, 196). Evidence for the zoning of features within the area investigated seemed fairly clear with several phases of ditches and gullies occurring at the north-eastern end and a variety of pits, hollows and occasional post-holes occupying the remainder of the area (Fig. 3). Quite possibly, the ditches delimited adjacent fields or garden plots, whilst 393 KEITH PARFITT the pits may have occurred within the main habitation area. The re-cut boundary ditches, Features 2, 3 and 4, might have originally served to separate these two zones. One of the latest features located appeared to be the extensive spread of flint metalling [F. 12), most probably representing the surface of an open yard but perhaps the base for a building. The occurrence of evidence for Stone Age and late Iron Age occupation on this heavy clay land, traditionally thought of as largely uninhabited until the later medieval period, is of some interest. Other evidence for ancient occupation on these clay-lands is now accumulating. In fact, a series of late Iron Age-early Roman settlements have been recognised at intervals along the high plateau land overlooking the Dour valley. At Barham, about 6km to the north-west of Green Lane, a settlement site occupied during the second half of the first century AD was located during road-works in 1971 (Philp and Philp I 97 4 ). Evidence for Roman occupation, between the first and early third century AD, has also been revealed below the medieval Knights Templar Preceptory at Ewell, 1km north-west of the present site (Frere I 984, Fig. 1, 1 ). More recently, yet another early Roman settlement, perhaps with a pre-Conquest origin, has been located off Honeywood Road, about I km to the south􀂐east (Pratt I 998; Parfitt 1999, Fig. 1, 2). A late Iron Age enclosed farmstead, probably occupied from c. 150 to 50 BC, has been recorded just off the clay at Church Whitfield crossroads, some 2km to the north-east of Green Lane (Parfitt, Allen and Rady 1997, Fig. 1, 4). An early Iron Age occupation site has been previously recorded at Whitfield Recreation Ground, almost 1 km in the same direction (Parfitt 197 4; Parfitt 1975, Fig. 1, 3). It is thus becoming clear that these heavy clay-lands were occupied by ancient man, perhaps almost as intensively as the lighter soils on the lower parts of the east Kent Downs. The relationship of the excavated late Iron Age site to the adjacent lane is worthy of some consideration. Green Lane itself must represent an ancient routeway, linking the bottom of the Dour valley with the high Downs. It is clearly marked on the 18 I 9 O.S. map of the area. A brief survey of the distribution of tree and shrub species along its hedge-rows was undertaken in an attempt to further refine its age. An average of four separate species per 30 metre length was recorded. On this evidence, and following Hooper's Rule (Rackham 1986, 194 ), it may be tentatively suggested that the lane was first hedged in the sixteenth-seventeenth century. An earlier origin for the lane itself seems likely but it does not appear to have had any influence on the orientation of the Iron Age ditch system, suggesting that it is significantly later than this. 394 A PREHISTORIC SITE OFF GREEN LANE, WHITFIELD, NEAR DOVER ACKNOWLEDGEMENTS The investigations were undertaken with the ready consent of the landowner, Mr G. Laslett of Ash. Of the Group members who undertook the work thanks are due to Messrs Geoff Halliwell, Ben Stocker, Howard Jones, Pat Mungovan, David Holman, Peter Godden and David Woodcock, together with Mrs Pat Godden, Mrs Tina Parfitt and Mrs Jill Bowers for their hard work over many weekends. The site plans and sections have kindly been prepared for publication by Mr Barry Corke. Dr Malcolm Lyne produced the report on the pottery and Miss Jo Bacon drew the Chaff-tempered sherds. BIBLIOGRAPHY Harford, P. M., 1982, 'A New Type of Kentish Briquetage', Kent Arch. Rev., 69, 204-205. Frere, S.S., 1984, 'Roman Britain in 1983', Britannia, XV, 332. Gaunt, J., Parfitt, K. and Halliwell, G., 1977, 'Surveys Along the Dover By-pass', Kent Arch. Rev., 48, 196-200. Halliwell, G. and Parfitt, K., 1993, 'Non-River Gravel Lower and Middle Palaeolithic Discoveries in East Kent', Kent Arch. Rev. 114, 80-89. Halliwell, G. and Parfitt, K., I 998, 'Discussion', in R. Hoskins, 'Lower Palaeolithic Discoveries at Eythorne, near Dover', Kent Arch. Rev., 134, 77-79. Hutchinson, G., I 976, 'Two Palaeolithic Hand Axes from Whitfield', Kent Arch. Rev., 45, 130. Macpherson Grant, N., 1980, 'Chaff Tempered Ware', Kent Arch. Rev., 61, 2-4. Macpherson-Grant, N.,S avage, A., Cotter, J., Davey, M. and Riddler, I., 1995, Canterbury Ceramics 2. The Processing and Study of Excavated Pottery. Parfitt, K., 1974 • A Probable Iron Age Site at Whitfield, near Dover', Kent Arch. Rev., 38,243. Parfitt, K., 1975, 'Iron Age Site at Whitfield, near Dover', Kent Arch. Rev., 40,289. Parfitt, K., 1995, Iron Age Burials from Mill Hill, Deal (London). Parfitt, K., 1999, Report on a Watching Brief conducted at the new Dover Ship's Stores Building, Honeywood Road, Whitfield, Dover (C.A.T. Archive Report, Nov. I 999). Parfitt, K., Allen, T. and Rady, J ., 1997, 'Whitfield-Eastry By-pass', Canterbury's Archaeology 1995-1996 (C.A.T. 20th Ann. Rep.), 28-33. Parfitt, K. and Halliwell, G., I 996, 'More Palaeolithic Discoveries in east Kent', Kent Arch. Rev., 123, 58-64. Philp, B. and Philp, E., I 974, 'A Romano-British Site near Halfway House, Barham,' Kent Arch. Rev., 37, 206-210. 395 KEITH PARFITT Pratt, S., 1998, Honeywood Road, Whitfield (Kent): Archaeological Evaluation (C.A.T. Archive Report, Dec. 1998). Rackham, 0., 1986, The History of the Countryside (London). Stead, I. M. and Rigby, V., 1989, Veruiamium: the King Harry Lane Site, English Heritage Arch. Rep. 12. Thompson, I., 1982, Grog-tempered 'Belgic · Pottery of South-eastern England, B.A.R. Brit. Series I 08 (Oxford). 396