Research Notes

247 RESEARCH NOTES – PREHISTORIC TwO PAlAEOlITHIC HANdAxES fROm HAwkINgE, NEAR fOlkESTONE during 1978, Alan Rye and a colleague whilst digging a new grave at Hawkinge Cemetery discovered two handaxes, one three times larger than the other, at a depth of approximately five feet (1.52m). The larger one came from a deposit 15cm above the smaller. (Unfortunately, since retiring Rye has lost contact with his former colleague, the owner of the second, larger axe.) In August 2014 by chance Rye’s smaller axe was shown to the author (Fig. 1). The finder remembers the soil conditions within the new grave said to have been an isolated chalky outcrop with heavy inclusions of dark orange clay (described as nasty stuff to dig), whereas elsewhere in the large cemetery the soils are much softer Fig. 1 The Palaeolithic handaxe from Hawkinge. (Illustrated by William Laing.) Height 11.4cm, maximum width at butt-end 6cm, maximum body thickness 3.6cm; weight 208.1gm. RESEARCH NOTES – PREHISTORIC 248 and easier to excavate. The find spot lies on the high North Downs (NGR 20681 40016) at an elevation of 150m aod, the surface geology being Clay-with-Flints. In detail the axe typology is primitive in character representational of the early period of these Acheulian implements belonging to a period around 230,000 bc. Roughly worked into a pear-shape by deep and shallow flaking on the dorsal side, the flakes removed would produce pronounced negative bulbs of percussion resulting from the technique of hammer-stone flaking as opposed to bone or wooden bar workmanship. The acute blade point was formed by trimming the flint down by taking three longitudinal flake removals one resulting in a hinged ridge finish near the central ridge. On the ventral side the flake removal is mostly diagonal from both lateral edges flattening and reducing the implement thickness towards the point. An unusual large almost vertical deliberately struck flake was removed between the central right-hand side and the butt-end blunting the edge for 40.43mm. The butt retains a small area of cortex allowing the implement to be griped in the hand. A semilateral high ridge on the dorsal butt-end encloses a deep core bed of crystalline silica rimmed by a narrow band of cortex rendering the axe ergonomically uncomfortable to be used in the right hand and thus undoubtedly a left-handed implement. The sharp edges and unabraded features indicate the axe was little used and not travelled far, if at all, from where it was discarded, stored or lost. This sharp and fresh bifacial axe was originally manufactured from a flint that would have been a creamy-grey or ivory white colour patination; however, concealed deep within the Clay-with-Flints layer, the axe has taken on a glossy light through to dark brown iron stained patina. The writer would like to thank Mr Rye for bringing the implement to his attention and for allowing its publication. His thanks to Mr William Laing for illustrating this important find. vince burrows 248 RESEARCH NOTES – PREHISTORIC fIvE ARROwHEAdS fROm THE NORTH dOwNS NEAR dOvER During 2013-2014 five arrowheads were brought to the author’s attention (Fig. 1). These represent further Neolithic and early-late Bronze Age activity along the high North Downs between Shepherdswell, Whitfield, Guston and West Cliffe, near St Margaret’s at Cliffe, a site known particularly for its Mesolithic finds (Parfitt and Halliwell 2010). Two rare Beaker-period arrowheads were recently found on the surface of fields 1.56 miles (2.51km) apart between Guston, near Dover, and West Cliffe. Other examples of important Beaker flint arrowheads from the region include; three finely-made ‘Fancy’ barbed-and-tanged examples from a burial in Thanet, only the third such burial finds recorded in Kent, with other single examples known from Cliffe and possibly Sittingbourne. a fourth crudely executed ‘none fancy’ Sutton B type barbed-and-tanged example came from a second grave (Hart and Moody 2008). a fine barb and tanged arrowhead was recovered from a site at Laundry Road, Minster in Thanet (Boast and Gibson 2000). RESEARCH NOTES – PREHISTORIC 249 Fig. 1 The five arrowheads from the North Downs. (Illustrated by William Laing.) F1 Miscellaneous oblique type arrowhead, middle to late Bronze Age: 33.51mm in length, 18.74mm at the widest point, maximum thickness 3.7mm; weighs 3.2g. F2 Leaf-shaped Neolithic arrowhead: 38.74mm in length, a maximum thickness of 3.63mm, width 17.65mm, weighs 2.6g. F3 Tang and barbed arrowhead, Neolithic-early Bronze Age: tang to tip 28.69mm, barb to barb 26.46mm, maximum thickness 5.40mm; weighs 3g. F4 Tang and barbed arrowhead, Neolithic-early Bronze Age: tang to tip 22.31mm, barb to barb 22.26mm, maximum thickness 9.91mm; weighs 1.5g. F5 Middle-late Bronze Age arrowhead: 23.17mm in length, 17.59mm at the widest point, maximum thickness 4.84mm, weigh 1.6g. The typologies of the five newly-discovered arrowhead are as follows: F1. NGR 29183 45227 (centred), at an elevation of 130m aod. A miscellaneous oblique type, ‘Sutton Series A’, type d. ‘none fancy’ (Green 1980), triangular in shape. Semi-abrupt retouching on both lateral edges forms the point of the central barb, broken in antiquity. The types ‘none fancy’ are in a form from a broad spectrum subdivided dependent on whether the barbs are absent, vestigial, unshaped or pointed. Created from a series of lateral flake removals leaving two central ridges on the dorsal side, the missile was manufactured from a mottled grey flake with a flat ventral surface. The bifacial serrated edges were formed by crude pressure flaking on the dorsal side, the tang broken in antiquity. A small area of creamy-brown cortex remains RESEARCH NOTES – PREHISTORIC 250 on the dorsal tip. Dating from the middle to late Bronze Age periods, many occur in the Upper Thames, Cotswold and Midland regions (Green 1980). The arrowhead came from a ploughed field near Temple Farm, Whitfield. F2. NGR 25633 46569 (centred), 115m contour. Recovered from the topsoil of a ploughed field near Shepherdswell, this leaf-shaped Neolithic arrowhead dates between 4,000-3,500 bc. Fashioned from a mottled creamy-grey flint, this near translucent flake has small black speck inclusions. The projectile ventral side has one singular iron stain at its centre. There are two minor modern damage sites on the right just below the tip and another right just above the bottom angle. On the dorsal side there are four small iron stain spots just below the tip. F3. NGR 34724 45374 (centred), 90m contour. Found on the plough soil at West Cliffe, this finely worked tang and barbed arrowhead is a Green Low type ‘g’ ‘fancy form’ (Green 1980). Formed from a semi-translucent mottled grey flint the arrowhead has one minor white inclusion. The footprint is an identical match to both the Thanet A1 and A2 Beaker examples. F4. NGR 32684 44415 (centred), 100m contour. Also recovered from the plough soil near Guston, Dover, this second finely worked but much smaller tang and barbed arrowhead is a Green Low type ‘j’ ‘fancy form’ (Green 1980a). Manufactured from a mottled grey flint and exhibiting four small iron stain spots, both types F3 and F4 are found rarely outside of England and Wales, both finely worked by invasive bifacial pressure flaking. F5. NGR 34847 45521 (centred), 85m contour. Recovered from the plough soil at West Cliffe, this crude middle-late Bronze Age arrowhead, was produced from a small flake exhibiting part of the platform and percussion bulb with scar indicating the flake was hard hammer struck. Albeit crudely fashioned, the tip bifacial serrated by pressure flaked edges are micro-finely retouched sharp and fresh. There are two shallow parallel contemporary scars on the dorsal side that may be the result of the arrowhead striking a hard object when fired. The dorsal side is traversed from point to stem by a high dorsal torsion ridge. Made from a rectangular shaped translucent flake with light brown inclusions, the tang was broken in antiquity. A similar specimen is recorded (Hoskins 1997). In regard to the Neolithic arrowhead F2, the Portable Antiquities Scheme website, Archaeologia Cantiana and Kent Archaeological Review record only a small assemblage of leaf-shaped arrowheads from Kent, fewer still from the east Kent coastal region. However, further examples are recorded in various Unit papers in the public domain, plus others unpublished. Even so, there is a relatively small number in the context of the large quantity of other period flint implements recorded from Kent. The few coastal zone examples recorded in Archaeologia Cantiana and Kent Archaeological Review are as follows: an undated broken arrowhead from near the Valiant Sailor public house near folkestone, although the description and illustration given in the article agrees a Neolithic date (Keene 2003). RESEARCH NOTES – PREHISTORIC 251 a second arrowhead from Crete Road West, near Folkestone, also undated but certainly of Neolithic date (Keene 2005), a third undated arrowhead from north of Castle Hill, Folkestone, also suggests a Neolithic date (Keene 2007). a small Neolithic arrowhead from Tolsford Hill, Etchinghill (Hoskins 1996). an early-middle Neolithic leaf-shaped arrowhead albeit broken, was discovered in a field near Pineham (Burrows 2013). a typical, intact fresh example recovered from just above the natural clay under the Roman Villa at Folkestone, dates between c.4000-3500 bc (Coulson 2013). an unrecorded intact Neolithic specimen viewed by the author was found over twenty years ago near Shepherdswell (NGR 25475 46751) (Clarke, A. pers. comm. 1990s), only 241m east of F2 findspot. discussion Despite being recovered from the surface, arrowheads F2 to F5 are fresh, sharp and undamaged. With only minor damage, F1 is also sharp and fresh although it has some dorsal abrasion. These finds represent further important artefacts from the Neolithic and Bronze Age periods increasingly being discovered by members of the public along the North Down high ridges, extending from Folkestone to Dover and between Shepherdswell, Whitfield, and St Margaret’s. Many of these finds come from locations sited along the coastal upland Clay-with-Flint ridges; however ‘it is more noticeable that Neolithic flint work was more common on the brickearth soils in the valleys than on the clay-with-flint soil’ (Gaunt, Parfitt and Halliwell 1977). Generally, these ridges run in a NNE-SSW direction towards Wingham, Sandwich and Deal. The five arrowheads covered in this report came from high contours between 85-130m. The two barb and tanged arrowheads are generally uncommon finds and more notably recovered from archaeological contexts. Both specimens F3 and F4 are thought to be prestigious objects possibly specifically made for inclusion in burials although these types have been found in other contexts (Green 1980). The theory of prestigious arrowheads is not necessarily well supported as demonstrated by the paradoxes of the Stonehenge Archer burial discovered in 1978. This burial dates to the Beaker period (2340- 2195 cal bc [95% probability] (Evans 1984). An examination of the skeleton of the Stonehenge burial confirmed the Archer died from three strikes in the back by barbed and tang arrowheads, these of a similar typology to those found in Thanet and examples F3 and F4 recorded here. The cause of the Archer’s death together with other ‘out of context’ arrowhead discoveries, may hint at the possibility of everyday use of bow/arrow for hunting or personal protection; and occasional deposition in more important burials such as the Amesbury Archer burial discovered in 2002, and the Boscombe Bowman buried 3 miles away at Amesbury. Both burials are contemporary with the Stonehenge Archer. In terms of the arrowheads considered in this report, it may be useful, rather than simply recording these finds, to note the observation of the close proximity arrowheads F1, F3 and F4 have to known ploughed-out former barrow monuments; their ring-ditches identifiable from aerial photography, certainly a factor worth RESEARCH NOTES – PREHISTORIC 252 taking into account, whether coincidence or not. With the growing number of barb and tang arrowhead finds in the north of Kent and Sussex in the Portable Antiquities records, future study of the relationship to barrow monuments or ploughed-out ring-ditches could prove worthwhile. It is important to the overall growing picture of prehistoric Kent, that the finders ensure such items are recorded. The author would like to thank Marc and Kevin Bousted for bringing the finds to his attention and allowing them to be published here. His appreciation to William laing for producing the illustrations and Philippa foulds for some outlines of the finds. vince burrows bibliography Boast, E. and Gibson, A., 2000, ‘Neolithic, Beaker and Anglo-Saxon remains: Laundry Road, minster in Thanet’, Archaeologia Cantiana, cxx, 359-372. Burrows, V., 2013, ‘An Arrowhead from Pineham, near Dover`. Archaeologia Cantiana, cxxxiii, 294. Coulson, I., 2013, Folkestone to 1500: a Town Unearthed (Canterbury), p. 16. Evans, J.G. 1984, ‘Stonehenge – the environment in the Late Neolithic and Early Bronze Age and a Beaker-Age burial’, Wilts. Arch. and Natural History Magazine 78: 7-30. Gaunt, J., Parfitt, K. and Halliwell, G., 1977, ‘Surveys Along the Dover By-Pass’, KAR, 48, 196-200. Green, H.S., 1980, ‘The Flint Arrowheads of the British Isles’, BAR British Series 75 (i), p. 122-3. Hart, P. and Moody, G., 2008, ‘Two Beaker Burials Recently Discovered on the Isle of Thanet’, Archaeologia Cantiana, cxxviii, 173. Hoskins, R., 1996, ‘Neolithic Arrowhead from Etchinghill’, KAR, 124, 97. Hoskins, R., 1997, ‘Flint Artifacts from Folkestone’, KAR, 30, 241-243. Keene, M., 2003, ‘Flint Arrowhead from near Folkestone’, KAR, 151, 22-23. Keene, M., 2005, `Artefacts from the North Downs, near Folkestone’, KAR, 159, 245-215. Keene, M., 2007, ‘Flints from Castle Hill, Folkestone’, KAR, 168, 186. Parfitt, K. and Halliwell, G., 2010, ‘A New Mesolithic Site at Westcliffe, St Margaret’s, near dover’, KAR, 180, 233-239. 252 RESEARCH NOTES – PREHISTORIC A MESOLITHIC TRANCHET AxE FIND: INVESTIGATION AT WOLVERTON LANE qUARRy, ALKHAM VALLEy During July 2014, members of the Alkham Valley Historical Research Group, led by the author, undertook an examination of three exposed sections of the quarry faces at a small site near the top of wolverton Hill lane, near the hamlet of Ewell Minnis (Fig. 1). The site is situated at TR 26143 43182 (centred), at 115m aod, on the south-facing slope approximately 86m down from the summit of the North Downs ridge, orientated NNE-SSW. Here in 2014 Mrs S. Mickleborough recovered a large mesolithic Tranchet Axe seen protruding from a section within the central area of the quarry between the Upper and Lower man-made terraces (Burrows V. 2015). The Alkham Valley is enclosed by steep escarpments bisected by softly contoured coombes leading to the mainly dry chalk bottom. Along the valley, several early ancient manors, now small hamlets, comprise of the main village of Alkham RESEARCH NOTES – PREHISTORIC 253 Fig. 1 Location map of Alkham valley and Wolverton site. (Crown copyright Ordnance Survey. All rights reserved.) RESEARCH NOTES – PREHISTORIC 254 (known as Malmains), Chilton, Evering (now Everden), Halton, Hoptons, Standen, Wolverton, and Halmede (the latter site now lost to memory are all situated at the foot of the hills (Lees and Humphreys 1985). The valley is serviced by the road leading from River, Dover, to the north-east, and connects with the Hawkinge road leading to Folkestone to the south, the valley being some 6.30km (3.9 miles) in length. The North Downs enclose the valley on its northern and southern sides and are known for numerous prehistoric lithic finds, derived from the predominant Clay-with Flint soil particularly towards the Folkestone fringes of the valley. The earliest implement discovered to date, in or around the central region of valley, comprises a finely worked late Palaeolithic handaxe (Halliwell, G. and Scott, B. 2011). A number of other finds from the hinterland include important finds published by Hoskins and Keene (ibid.) (see five arrowheads from the Downs listed at pp. 248-52) and two Palaeolithic handaxes from Hawkinge (see pp. 247-8). The small quarried area was excavated within the past few years to provide access to Neckwood owned by the Woodland Trust, and leased to private management, and lies on a reducing gradient of about ten degrees from the summit. Beyond the quarry’s eastern perimeter, the slope falls away more steeply through Neckwood towards the valley floor. The site was excavated by machine into the south side of this asymmetric valley, that in itself is cut by an ancient dell with Wolverton Hill Lane on the north-eastern side and residential buildings; property field divisions extend down the northern slope of the dell, adjoining the Alkham road in the rightangled valley below. The aim was to provide important substrata information, not normally accessible, surrounding the discarded, lost or deliberate deposition of the Tranchet Axe recovered in 2014 (Burrows, V. 2015). In these particular areas of the North Downs, Palaeolithic and Mesolithic artefacts can be found in the solifluction deposits eroded down-slope from the plateau. The fieldwork was undertaken by hand with the minimum invasive disturbance to the site’s flora and fauna. The following three sections were cut on the exposed faces on two terraced levels left by the mechanical stripping. Section 1 was situated against the northerly facing quarry cut within the middle terraced area where the Tranchet Axe was extracted by the finder in section One (Figs 2 and 3). The section measured 3m in length by 0.70m in depth, the Tranchet Axe being recovered from 0.27m into the section from the man-made terrace above. Using a level-line across the section from the adjacent north meadow field, the line was fixed across the north down-sloping bank of the quarry to the southern bank. The original depth of the Tranchet axe was confidently estimated from the original land-surface before stripping at about 1.08m. The section consisted of seven contexts including the natural clay layer capping the Upper Chalk, after the machined removal of the layers above the terrace. In the central lower terraced area (Fig. 3) a deposit of modern soil accumulated on top of this terrace which included dumped modern concrete, plastic and glass [context 01]. The deposit measured 02-33cm in thickness. The following undisturbed layers were capped by [01]. Consisting of orange silt clay, containing mainly nodular but occasional tabular flints, context [02] contained the horizontal Tranchet axe jutting out of the eastern RESEARCH NOTES – PREHISTORIC 255 Fig. 2 Wolverton quarry plan. facing transect at a depth of 9cm, into the layer that measured between 16-46cm in thickness. Orange silt clay with occasional small to medium nodular flints formed the next layer [03] and measured in depth between 5-22cm. Sandy orange silt clay with (sorted) larger nodular flints intermixed with occasional patches of small nodular flints [04] made up the most dense flint context examined within the quarry and measured 7-19cm in depth. The lowest two contexts confined to the eastern and western extremities either side of the section, consisted of an indistinguishable matrix of sandy orange clay with occasional large nodular flints [05] and measured between 2-12cm. Context [05a], marginally thicker, measured between 2-16 cm. Both layers exhibited a noticeable absence of smaller flints. The unexcavated natural clay capping the Upper Chalk [15a] continues for an unknown depth, the natural chalk not being encountered. Section 2, measuring 1.84m in length and 0.80m deep, was located against the west bank of the quarry. The topsoil layer (mechanically pushed-back) consisted of modern soil [09]. The area was not fully accessible and the measurement of depth is approximately 1.3m and sealed the orange silt clay, mixed with mainly large nodular flint inclusions with occasional tabular flints [10]; this measured between 2-28cm thick and caps an orange silt clay layer containing four nodular flints [11], albeit the nodular flints were smaller in size than [10], a similar context that measured 17-36cm. The next deposit capped a further sandy orange silt clay with more dense patches of medium to large nodular flints, intermixed with occasional RESEARCH NOTES – PREHISTORIC 256 Fig. 3 Wolverton quarry, sections 1-3. patches of smaller nodular flints, more so towards the northern end of the section and larger flints towards the southern [13]. A similar context as in section one [04] measures between 11cm and a maximum of 28cm sealing a deposit of sandy orange silt clay with occasional medium nodular flint inclusions [14]; however, there is a noticeable absence in quantity of the larger and smaller flints. This band measures 9-23cm and corresponds to contexts in section one [05 and 05a]. The final layer [15] comprised natural orange clay with no inclusions. At the base of this exposed sequence only a minimal depth of this natural layer was exposed to a depth of 16cm. The natural Upper Chalk was not encountered. Section 3, orientated east-west, measured 3.20m in length by 0.80m deep located above the Upper terrace had a deep band of modern topsoil that had been mechanically pushed-back from the quarry’s steep bank [06]. There was no physical access to record the layer thickness that is estimated at 1.3m (Figs 2 and 3). The next layer [07] contained glass, plastics and tree roots, one large and several small nodular flints. This context seals the broad band of orange clay containing patchy, mainly large, nodular flints with occasional smaller nodules. The layer measures a minimum of 11cm to a maximum of 38cm thick. Context [12] was more irregular in thickness compared to [08] and notably made-up of slightly lighter orange clay containing mainly small to medium sized nodular flints and one tabular flint, dispersed throughout the layer and measures 6-31cm thick. RESEARCH NOTES – PREHISTORIC 257 Context [16] forms the lowest level exposed before arriving on the Upper terrace floor. This orange clay layer forms the top of section one [01] the section examined on the northern edge of this terrace. There were no flints present however; two small chucks of iron stone were recovered. Only a narrow band of this context was exposed to a depth of 22cm. Conclusion The two-day quarry investigation gave a seldom available opportunity to glimpse and record a horizon containing an in situ Tranchet Axe. The layers capping the find and those below it are infrequently exposed on the sides of the dry chalk valleys of east Kent. The slopes are often deeply mantled by material derived from solifluction deposits containing Clay-with-Flints as seen here at Wolverton, but also mixed with chalk in other areas of these valleys. The stratigraphic sequence comprises sediment clay containing flint. The sequence has been summarized by the materials’ colour, size and types of flint inclusions. The weathering phases for most of the sequences are clearly defined however; where the sedimentation is thicker containing accumulation built-up over successive periods of time, the different individual phases were imperceptible. The material at wolverton is made up of marginal variations of lighter or darker compacted, stiff, sandy orange silt clay that includes varying amounts of natural annular, nodular and intermittently tabular flints. There were no traces of organic material within any of the sections examined. Over thousands of years these flints randomly eroded under weathering and with the down-slope gravitational process from the interfluves above created by frost action, freezing-thawing and prolonged periods of wet weather at the top of the valley, perhaps indicate the absence of natural protection of vegetation and trees at least in the proximity to the valley rim. With the solifluction deposits being considerable in depth on this south facing side of the valley, the expected Upper natural chalk was not encountered within the quarry. The examination produced eighteen individual contexts, of which thirteen corresponded to other sections exposed. The fine orange silt clay layers were clearly made-up from numerous silting events, many imperceptible individually although the layers surrounding the site of the Mesolithic axe find (section one and two), revealed distinctive layers of flint concentrations formed by periods of an extremely wet climate moving higher densities of smaller and larger flints downslope. These episodes of prolonged periods of wet weather led to comminuted collections of small, medium and larger flints throughout all three sections examined. Periods of drier weather appear to be recorded in the layers containing irregular or sporadic flint deposits. The climatic events are also captured in the opposing section two located at a right angle to section one on the eastern section of the quarry. Section three, situated on the terrace above sections one and two, also records a later period of similar wet weather attrition within context [12]. Continued erosion over the centuries has clearly mixed and transported the Clay-with-Flint deposits downhill, quite likely re-depositing the derived Mesolithic Tranchet axe with them. As seen by the dense quantity of variously-sized flints surrounding the axe in contexts [02] and [04], it may have originally moved by weathering downwards from further up RESEARCH NOTES – PREHISTORIC 258 the slope; however, the final position of the axe in context [02], at 1.08m below the estimated original modern land surface, demonstrates the potential depths to which early prehistoric material can be buried. Further down-slope these depths are likely to increase. These multiple erosion events provide interesting climate evidence surrounding the Tranchet axe find, useful in the future assessment of other potential sites in the valleys of east Kent. acknowledgements The author would like to thank the following volunteers who made the project possible; Elissia Burrows (independent archaeologist), Veronica Reilly (independent archaeologist) and volunteer Ken Ogilvie. vince burrows references Burrows, V., 2015, ‘A Tranchet Axe From The Alkham Valley, Near Dover’, Archaeologia Cantiana, 136, 213-214. Halliwell, G. and Scott, B., 2011, ‘Lithics: a late Middle Palaeolithic-type handaxe from East kent’, Journal of Lithic Studies Society, 32, 63-65. Lees, S. and Humphreys, R., 1985, A Pictorial Study of Alkham Parish, Meresborough Books. Kent. 258 RESEARCH NOTES – PREHISTORIC PREHISTORIC AND EARLy ROMAN REMAINS AT DANE COURT GRAMMAR SCHOOL, BROADSTAIRS, AND HERNE BAy HIGH SCHOOL Oxford Archaeology carried out a programme of archaeological investigations between December 2008 and July 2010 in advance of redevelopment of part of the grounds of Dane Court Grammar School, Broadstairs (NGR TR 380 679) and Herne Bay High School, Herne Bay (NGR TR 167 670) as part of the Kent Building Schools for the Future programme. (The full reports of both sites are available at Kent Archaeological Reports online.) Dane Court Grammar School (Fig. 1) The only feature that certainly predated the late Iron Age was a pit (1132) that had been largely truncated by more recent ploughing and survived only as a shallow smear of darker soil, but contained a fragmented body sherd from a carinated bowl (c.4000-3700 bc). Two further pits (1210 and 1214) were sealed beneath a buried soil of late Iron Age/early Roman date and so may have been similarly early, but neither contained artefactual material and so they cannot be assigned a more precise date. Worked flint dating from the Mesolithic period until the late Neolithic/early Bronze Age was also recovered from fills of later features. Late Iron Age and early Roman activity was represented by a sequence of curved boundary ditches and associated features. Near the northern limit of the site, a broad, shallow ditch (1130) and three smaller ditches may have defined successive phases of a boundary that was re-dug on several occasions and shifted over time. Some 9.5m further south lay a boundary that had likewise been recut on several RESEARCH NOTES – PREHISTORIC 259 Fig. 1 Phased plan of all archaeological features at Dane Court Grammar School. occasions, albeit in this instance with no change in its location. Two insubstantial ditches that branched off the south side of this feature appeared to define a small rectilinear enclosure measuring 12 x 8m that adjoined the boundary. The ceramic RESEARCH NOTES – PREHISTORIC 260 Fig. 2 Phased plan of all archaeological features at Herne Bay High School. RESEARCH NOTES – PREHISTORIC 261 evidence indicated that the entire sequence fell within the period of currency of ‘Belgic’ wares, that is c.50 bc to ad 70, and the presence of sherds of North Kent fine ware and sand-tempered oxidised ware in the later phases indicates that it extended into the post-conquest period. Conversely, the absence of these postconquest wares from the earlier phases of the ditches suggests that the features originated before the introduction of these types. Two cremation burials (1116 and 1118) and a rather irregular four-post structure (1207) were situated within the putative rectilinear enclosure, and a third cremation burial (1196) lay immediately to the east. All these features were of post-conquest date. Large jars had been used for the cinerary urns, and burials 1116 and 1118 were accompanied by additional vessels. The ancillary vessels in burial 1116 comprised a small carinated bowl and a small, beaker-sized jar, and a copper alloy Colchester brooch had been placed with the cremated remains, which comprised a young adult of uncertain sex. Burial 1118 was the burial of an adult of uncertain sex and was accompanied by a small jar or bowl. Although cremation burial 1196 contained no ancillary vessel in the strictest sense, the rim and shoulder of a necked jar were found that appeared to be the remains of a vessel that had been inverted over the cinerary urn to serve as a lid. This burial contained only 39g of cremated bone and no diagnostic elements were present from which the age and sex of the individual could be established. The features excavated at dane Court grammar School clearly represent agricultural boundaries and burials associated with a nearby settlement whose precise location has not yet been identified. Herne Bay High School Nine features in the northern half of the excavation area at Herne Bay High School contained small amounts of flint tempered pottery of late Bronze Age to early Iron Age date (Fig. 2). They were all shallow oval or sub-circular pits and yielded no other artefactual material. An e-w aligned ditch (074) that extended across the site contained nine sherds of grog-tempered pottery of late Iron Age to early Roman date. In three equidistant locations along the ditch were three irregular features (020, 069, 057) that may have been tree-throw holes, perhaps evidence that after the ditch had silted up the boundary continued to be defined by a treeline. Three shallow pits to the south of the ditch also dated to the late Iron Age/early Roman period. The site was located immediately to the north of an area of Iron Age and early Roman settlement that was excavated during 1999-2001 by Hertfordshire Archaeological Trust (now Archaeological Solutions) in advance of installation of an all-weather sports pitch (see Archaeologia Cantiana 129 (2009), p. 383). It is likely that the remains form part of the same settlement, of which ditch 074 may define the northern limit. andrew simmonds and kate brady RESEARCH NOTES – PREHISTORIC 262 A LATE IRON AGE AND EARLy ROMAN SETTLEMENT AT LEyBOURNE gRANgE, NEAR wEST mAllINg Archaeological excavations were undertaken in four areas (A to D) in 2009 by Oxford Archaeology ahead of development by Taylor Wimpey SW Thames of a former hospital at Leybourne Grange to the west of the village of Leybourne, near Fig. 1 Leybourne Grange: Site Location. RESEARCH NOTES – PREHISTORIC 263 West Malling (Figs 1 and 2). A watching brief was maintained from 2011 on other areas, though archaeological remains were confined to one area (E and trenches x and y). Occupation in the area of Leybourne Grange during the Mesolithic period is attested by the presence of residual flint microblades in later deposits, and flint scrapers and a polished flint axehead identify Neolithic activity. The material adds to the general background of earlier prehistoric activity in the Medway Valley, and complements a flint tranchet adze and microlith of late Mesolithic date, a late Neolithic arrowhead, and a variety of Neolithic to Bronze Age blades, bladelets, scrapers, flakes and cores found along the West Malling-Leybourne bypass, mainly in site A but all redeposited (Leivers 2009, 16). Worked flint of Neolithic to Bronze Age date has also been found in Snodland (Birbeck 1995, 81). Pottery and flint collected from ditch 4042 in Area D tentatively dates the feature to the late Bronze Age, and support is provided by a pit given the same date on the basis of pottery and flint, which was recorded by Archaeology South-East in an evaluation trench in Area D (Riccoboni 2008, 24). More late Bronze Age pits and ditches were recorded in sites A and E along the bypass route (Ellis 2009, 5). The Neolithic axehead recovered from pit 1298 is a curiosity and given the date of deposition – a matter of years after the Roman invasion of ad 43 – could be interpreted as a curated object specially placed during a time of change and uncertainty, possibly to maintain tenure of the land. A relatively large early Roman pottery assemblage, some 150 sherds, was found alongside the axe, but its jar and coarse ware emphasis does not obviously set it apart from other groups across the site. A settlement was established in the late Iron Age (Fig. 3). Excavation uncovered an enclosure, defined by ditches (1377 and 1381), open at one end, and measuring some 50m by 60m. Smaller ditches (1079, 1109 and 1179) within the enclosure may have sub-divided the enclosure, or marked the position of internal stockades, although they may instead represent earlier features. The dating evidence cannot pinpoint when occupation commenced. The pottery is not inconsistent with a start date in the late 1st century bc, but no groups were identified that must belong to that time. A bow brooch from pit 1299 shares traits with continental types that date from the late 1st century bc to early 1st century ad, but this is not sufficient to firmly isolate late 1st century bc occupation. There is more certain evidence to place the late Iron Age enclosure in the first half of the 1st century ad. Assemblages from ditch 1381 included imported Gallo-Belgic wares, such as white wares and terra rubra fine ware, in addition to a mass of locally made pottery, and date deposition to the early 1st century ad. No evidence of buildings was uncovered, but features within the enclosure still suggest something of the nature of activity. Pits in the northern part of the enclosure were straight-sided and flat-bottomed, with storage being a likely function. Other pits had profiles – steep on one side and a gentle gradient on the other – that pointed to a different function. This remains unknown, although a dump of plant remains in pit 1268 suggests that the pits served as waste pits. Two small pits (1137 and 1139) in the southern part of the site had been burnt around the edges and contained charcoal-rich fills. These can be more definitely identified as hearths. The features possibly had a domestic function, although given the apparent absence RESEARCH NOTES – PREHISTORIC 264 Fig. 2 Leybourne Grange: phased plan of excavations. RESEARCH NOTES – PREHISTORIC 265 of domestic buildings, they may instead have had an agricultural function, for example parching grain or malting. Although from a later ditch (1107), detached sprouts and embryos of possible barley potentially provide evidence for the on-site charring of grain to prevent further sprouting of the crop following germination. A four-post structure (1132) is likely to have provided storage; structures such as this are typically interpreted as raised granaries (Bersu 1940, 97-8). Another enclosure (5038) was uncovered in Area E. Ditches in Area D (4040, 4041 and 4043) and the south-eastern corner of Area A Fig. 3 Leybourne Grange: Iron Age settlement. RESEARCH NOTES – PREHISTORIC 266 (1379 and 1380), which shared orientation and were spaced at c.10-20m intervals, are likely to have formed part of an extensive field system. The ditches suggest that the area of land to the east of the enclosure was put to agricultural use, being available for the cultivation of crops (mainly wheat, but also including barley, oats, flax, and possibly legumes). Livestock was also kept in and around the site. The animal bone assemblage indicates that cattle were the most important species, followed by sheep or goats, pigs and horses. The economic basis of the site may also have included pottery manufacture. Sandstone and glauconite temper used in much of the pottery recorded at the site was locally available (Biddulph 2018). There is, however, no evidence of kiln-like structures or hearth bases on which a clamp kiln may have been erected, or certain evidence of pottery wasters. The enclosure was replaced by another enclosure that was defined by ditches 1107, 1383 and 1385. The replacement, trapezoidal in plan, was narrower than its antecedent, at 30-50m, but was roughly the same length at 65m. The pottery from the ditch fills indicate that the features filled in the early Roman period, probably no later than ad 70, given the continued dominance of grog-tempered ware and other fabrics of late Iron Age tradition, but it is possible that the ditches were cut before ad 43. Internal features were restricted to a cluster of pits. Like those of late Iron Age date, the pits varied in shape, though generally conformed to two types – pits with concave sides and bases (essentially U-shaped in profile) or pits with an asymmetrical profile, steep on one side and gentler on the other. Pit 1259, which cut 1268, was identified as a hearth, and potentially the pit complex as a whole related to the cycle of crop processing, storage, and waste management. The late Iron Age field ditches south-east of the enclosure were replaced by a set of ditches orientated ese-wnw. Their layout has the appearance of a coherent set of narrow enclosures or fields, but in reality each ditch was dug in turn after the previously dug ditch had filled, so the arrangement was presumably unplanned when the first ditch in the sequence (1363) was dug. The sequence evolved, with the layout of subsequent ditches building on that of the earlier ones. The ditches may have marked field boundaries, although the inclusion of shorter ditch 1365 and sets of postholes to the south of 1363 suggests a function related to livestock control, for example herding and corralling. The east terminus of 1363 was recorded in Trench x. The stratigraphic and economic evidence points strongly to a rural settlement largely concerned with crop cultivation, processing and storage, but also included animal husbandry. The absence of domestic structures suggests that the focus of settlement lay beyond the area of excavation, although this need not have been far away, judging by the relatively well-preserved character of the pottery assemblage. Despite the collection of relatively exotic imported wares – such as the terra rubra, olive oil amphorae from southern Spain, and South gaulish samian ware – the profile of the ceramic assemblage is consistent with the suggested site function, a low-status, rural settlement. The late Iron Age and early Roman activity finds a very close match in the archaeology of the adjacent West Malling and Leybourne bypass. Excavations along the route uncovered two concentrations of contemporaneous occupation. Two parallel ditches recorded in one part of the site (Site A; Fig. 1), c.1km east of Leybourne Grange, were replaced or incorporated within an enclosure with RESEARCH NOTES – PREHISTORIC 267 slightly tapering sides. This was replaced by a second, wider, enclosure, whose north side was shifted some 20m south of the north side of the first enclosure. No domestic buildings were seen, but a four-post structure was associated with the second enclosure (Ellis 2009, 5-7). An enclosure with tapering sides and an open end was recorded at site E2 almost 1.5km south-east of Leybourne Grange (Fig. 1). A number of pits were found within the enclosure, although dating evidence suggests that these dated to the middle/late Iron Age (Ellis 2009, 9). Both the bypass and Leybourne Grange sites were occupied in the late Iron Age and earliest Roman period, although activity appears to have commenced along the bypass in the 2nd century bc (Ellis 2009, 52); there was little evidence to push the dating at Leybourne Grange earlier than the mid/late 1st century bc. All sites were abandoned more or less at the same time. Pottery and radiocarbon give a terminal date for the bypass sites within the period ad 50-75 (Ellis 2009, 52). The morphology of the bypass and Leybourne Grange sites – rectilinear D-shaped or trapezoidal enclosures whose location shifted slightly with each phase – is so similar as to suggest links between the populations that worked those areas, though whether the areas belonged to a single, extended community, or represented separate neighbouring settlements is uncertain. The group of square pits (1251), arranged in a semi-circle or oval pattern, was the only significant post-Roman feature to be uncovered in the excavations. Modern ceramic building material recovered from the pits suggests that the feature is of 18th- to 20th-century date, but it does not appear on any edition of Ordnance Survey map and is unlikely to be a structure. The feature might instead relate to the extensive tree-planting that has been undertaken since the 19th century (Waterman CPM 2008, 3). [The full report is available at Kent Archaeological Reports online.] acknowledgements The author is grateful to Taylor wimpey Sw Thames for funding the archaeological fieldwork, post-excavation programme and publication, and to Wendy Rogers, Senior Archaeological Officer for Kent County Council, for her support. The fieldwork was directed by Dan Sykes and managed by Timothy Haines. Dan Sykes also undertook the watching brief, which was managed by David Score. Finds and environmental information was provided by the author (pottery), K.L. Hunter (plant remains), David Mullin (worked flint), and Lena Strid (animal bone). The pottery report is published in the Journal of Roman Pottery Studies. edward biddulph references Andrews, P., Dinwiddy, K.E., Ellis, C., Hutcheson, A., Phillpotts, C., Powell, A.B. and Schuster, J., 2009, Kentish sites and sites of Kent: a miscellany of four archaeological excavations, Wessex Archaeology Monograph, 24, Salisbury. Bersu, G., 1940, ‘Excavations at Little Woodbury, Wiltshire’, PPS, 6(1), 30-111. Biddulph, E., 2018, ‘A late Iron Age and early Roman pottery assemblage from Leybourne grange, west malling, kent’, J. Roman Pottery Studies, 17, 74-91. RESEARCH NOTES – PREHISTORIC 268 Birbeck, V., 1995, ‘Excavations on a Romano-British villa at Churchfields, Snodland, 1992-94’, Archaeologia Cantiana, 115, 71-120. Ellis, C., 2009, ‘Archaeology of the West Malling and Leybourne bypass’, in Andrews et al., 2009, 1-55. Leivers, M., 2009, ‘Worked flint’, in Ellis, 2009, 16-18. Riccoboni, P., 2008, ‘An archaeological evaluation on land at Leybourne Grange, Birling Road, Leybourne, West Malling, Kent’, unpubl. report, Archaeology South-East. Waterman, C.P.M., 2008, ‘Leybourne Grange, Maidstone, Kent: Specification for Strip, Map and Record Excavation’, unpubl. report. 269 RESEARCH NOTE – ROmAN A ROMAN VILLA AT MARWOOD FARM, FALCONHURST, ALDINGTON This site, which lies in the parish of Aldington some 2½ miles (3.8km) to the west of the Saxon Shore fort at Lympne, was first noted in the late 1960s when ploughing brought up Roman material to the surface (Kent Heritage reference TR 03 SE7) (Bradshaw 1969). It is situated at the bottom of a long slope with a small stream immediately to the west, about 150m to the north of the Royal Military Canal at an elevation of about 4m aod (TR 0780 3440). following an inspection of the site, a resistivity survey was carried out in the summer of 2013 which indicated a possible rectangular building. A small trench was dug, which was later extended, and this confirmed the 1960s findings of Roman CBMs and pottery. It was decided to carry out a limited excavation aimed at characterising the building and, if possible, dating its period of occupation. Two local groups, from Dover and Folkestone, were invited to take part. The Dover group, led by Keith Parfitt, took responsibility for the excavation of the main evaluation trench (A) and the Folkestone group, under Ian Neilson, undertook the recording, classifying and cleaning of all the artefacts. Following the mechanical removal of topsoil, the main excavation took place on 16th -17th November 2013. An aerial photo shows excavation in progress (Fig. 1). The Evaluation Trenches Excavated at marwood farm The excavation work comprised three trenches: a main evaluation trench (A) running roughly south north: a short trench (B) running west east, to the west of the main trench: and a small trench (C) in the north-west corner of the building (Fig. 2). Trench A This trench was 25.5m long and 1.25m wide, aligned roughly n/s and followed the natural slope of the ground. The excavation revealed a series of low, ragstone walls and foundations in the southern half of the trench. Associated brick, tile and pottery left no doubt that these walls were of Roman date and that they were part of a larger building. The trench revealed the external south wall of the building, a dividing wall running n/s inside the building and a corridor forming the northern part of the building. To the west of the dividing wall, Room A had a n/s wall length of 5.5m. To the east of the dividing wall, there were two rooms, divided by a wall. The more southerly Room B measured 1.90m from north to south. Room C, to the north of Room B had a n/s wall measuring 3.40m. 270 RESEARCH NOTE – ROmAN The southern wall (50) measured 0.66m in width, the main n/s dividing wall (51) was 0.46m in width, the east wall dividing rooms B and C (52) measured 0.57m in width and the northern internal wall dividing the corridor from the rooms (53) measured 0.58m in width. A butt joint was noted between the n/s wall (51) and the southern e/w wall (50). Room A: located on the western side of the n/s wall, this was the largest of the three rooms. No clear evidence of a floor surface remained but two features worthy of note were recorded. Much of the central part of the room was covered by an area of broken tile, fairly certainly evidence of an in situ roof collapse. Secondly, in the south east corner, a complete pottery vessel had been set upright below ground level (see below). Scattered around the pot were fragments of bone and charcoal and there were also minute quantities of bone and charcoal inside the pot. The layer of charcoal, bone and daub were cut through by the foundation trench for wall 50, indicating that the deposition predated the construction of this wall. Room B: this was a small room located on the east side of the central wall 51. No clear floor surface was recognised. A layer containing many fragments of burnt clay was examined but proved to have been dumped. Room C: this lay to the north of Room B and was larger. It was characterised by a deposit of clay and ragstone rubble. In the north west corner of the room was a large lead weight, seemingly set with some care among the stones in such a way as to suggest a deliberate deposition. Room D: this room was located on the north side of Rooms A and C. The n/s wall Photograph 1. An aerial photo taken on Saturday 16.11 2013 of Marwood Farm showing excavation in progress. On the left is the Military canal and Romney marshes. The rectangular “crop marks” are simply the tracks made in the stubble when carrying out a resistivity survey but give an indication of the size of the building. The main trench runs south-north. Courtesy of Lorraine Harrison and Peter Stretton. (pilot) Summary of excavation. The excavation was comprised of three trenches, a main evaluation trench running roughly south-north , a short trench running west-east, which was located to the west of the main trench and a small trench in the northwest corner of the building.. Fig. 1 Aerial photograph 271 RESEARCH NOTE – ROmAN 51 did not extend into this room. Two parallel walls, 53 and 55, gave a n/s width of c.2m. A clear floor level was identified, consisting of a compact, dense layer of small and medium sized pieces of ragstone set in clay. There was no overlying occupational deposit. This area was interpreted as the north corridor of the building. A composite photograph (Fig. 3) of part of the main trench shows: a portion of wall 50 at the south end wall 51, which divided room A on the west side from rooms B and C a portion of wall 53 which divided the corridor at the north end from the rooms of the villa. The main trench extended beyond the north wall 55 but the only noticeable feature was a layer of slightly discoloured clay in the northern half. It was decided to cut Fig. 2 Plan of the excavations at marwood farm 272 RESEARCH NOTE – ROmAN three one-metre slots at intervals along the trench to test the ground outside the building (Fig. 2): Slot 3: closest to the building, simply revealed a thin layer of grey clay containing occasional flecks of charcoal and small fragments of Roman tile and a few potsherds on the surface. Slot 2: dug about 9m north of the building, revealed a series of layers of redeposited tile fragments, pottery sherds, charcoal and ash mixed with soil to a depth of nearly a metre. It was clear that the deposits continued in all directions beyond the cut slot and that they had been tipped into an area that had been previously dug out, perhaps for a pit or ditch. The majority of the pottery sherds at the lowest level were of second-century date, around ad 130 -200. Slot 1: it became clear that slot 1 revealed a similar pattern of successive layers of re deposits but only to a depth of half a metre. In this slot the deposits were primarily tegula tile fragments but the slot also contained some pieces of combed and patterned box tile, some pieces of bipedalis tile and moderate quantities of charcoal and ash. It was clear that slots 1 and 2 revealed the presence of a dump of tile, pottery and burnt wood built up over the period of occupation of the nearby building (Fig. 4). Trench B This hand -dug original trial trench was cut at right -angles to the main N/S trench, about 4m to the west of it and measured 5.60m long and 1.00 1.40m wide. It revealed further lengths of wall 53, a length of wall 54 running south (part of the western wall of Room A) and quantities of Roman tile and brick. A butt joint in wall 53 indicated two stages of construction. Trench C A small trench, measuring 2.5 x 1.3m was dug to verify the position of the northwest corner of the building. The trench revealed a break of about a metre in wall 56, which was interpreted as an entrance to the villa via the corridor. An iron slider key was uncovered a short distance from the entrance (see below). A short length of wall 57, the western outer wall of the villa, was uncovered. The depth of plough soil over the building was examined to test for potential damage through ploughing. Above the walls of the building was an intact layer Room D, to the north of the main trench was later interpreted as a corridor. Photograph 2. A composite photograph of part of the main trench showing a portion of wall 50 at the south end, wall 51, which divided room A on the west side from rooms B. and C. and a portion of wall 53 which divided the corridor at the north end from the rooms of the villa. Courtesy of Thierry Biot. The main trench continued north of the building and three slots, S1, S2 and S3 were cut at intervals to test the ground outside the building (Fig 1.). Slots 1. And 2. revealed evidence of a substantial refuse heap composed of burnt clay, ash, pottery and tile deposited in an area that had been previously dug out. Fig. 3 Composite photograph of Main trench (Courtesy of Thierry Biot) 273 RESEARCH NOTE – ROmAN of dark brown silty clay whose thickness varied between 0.12m to 0.35m. Above this layer, the depth of plough soil varied between 0.25m. and 0.30m. It appears, therefore, that the building is not under immediate threat from ploughing. The finds and the dating evidence No coins were uncovered within the building during the excavations and only four coins were found in the surrounding field; a Trajan sestertius (98- 117), a Julia Domna denarius (196 -211) and two as coins in poor condition but probably second-century. The most interesting find was an almost complete pot buried below floor level in the south-east corner of Room A (Fig. 5). It contained soil and a small quantity of burnt charcoal and a few small fragments of bone. Scattered around the pot was a larger quantity of burnt bone and charcoal. Seven identifiable pieces of this burnt bone were identified by Alan Pipe (Museum of London) as belonging to a sheep’s left side. The pot, which contained a small quantity of the bone, was of a type in use from about c.100-c.250. It is interpreted as a votive offering laid down during the construction of the villa. In Room C a lead weight in the form of a truncated cone with an iron loop on top, had been placed among stones in a manner to suggest a deliberate deposition (Fig. 6). The practice in Roman Britain of making offerings prior to the construction of buildings and other man-made structures like kilns and wells is fairly well known. less well known is the practice of leaving an offering when the structure ceased to be used (Merrifield 1987). In Trench C, very close to the entrance to the building in the north-west, an iron slider key was uncovered which, it is presumed, fitted the lock on the nearby door of the building. Figure 2. A cross section of slot 2. Trench B, the original trial trench, west of the main trench, was extended and confirmed the line of the northern wall 53 of the building which separated the corridor from the main building. A wall butt joint in wall 53 indicated two stages of construction. The trench also revealed part of the western wall of room A. Trench C. A short trench was dug to verify the northwest corner of the building and this revealed a length of wall 56, interpreted as an extension of wall 55, with an entrance to the building via the corridor. It also revealed a short length of wall 57, part of the outer western wall of the building. A more detailed summary of the excavation of trenches A, B. and C. is attached in Appendix 1. The finds and the dating evidence. No coins were uncovered during the excavations within the building and only four coins were found in the field around the villa, a Trajan Sestertius, 98-117, a Julia Domna Denarius,196-211 and two As coins in poor condition but probably second century. The most interesting find was an almost complete pot buried below floor level in the southeast corner of room A. Photo 3. It contained soil and a small quantity of burnt Fig. 4 A cross-section of slot 2 274 RESEARCH NOTE – ROmAN In the absence of any coins uncovered during the excavation, dating evidence for the construction and occupation of the building depends heavily on two sources, the pottery assemblage and two pieces of patterned flue tile. The patterned tile fragments were found in the spoil heap and were identified as examples of die 16 and die 96 by Ian Betts (Museum of London). Die 16 has two context datings from Canterbury, i.e. a context dated to 90/120 and a second context also dated late 1st/ early 2nd century (Betts, Black and Gower 1997). The pottery assemblage, although relatively small (338 sherds) was referred to M. Lyne for identification and dating and a summary of the main points of his report is at Appendix (the full report is published on the KAS website). It is clear that the bulk of the pottery originated from the north Kent and Thameside industry kilns but there is a small quantity of foreign imports present. It is also evident that the bulk of the pottery dates from the second/third centuries. Based on the evidence of the pottery and the patterned tile, this Romano-British building was probably built in the first quarter of the second century and ceased to be occupied in the last half of the third century and there is enough evidence to conclude that the building was a modest corridor villa or farmhouse. marwood farm villa in its contemporary environment Unusually, the villa was sited probably only a few metres above the level of the fairly well known. Less well known is the practice of leaving an offering when the structure ceased to be used. (Merrifield 1987) Photograph 3. The deposited pot in situ in the S.E. corner of room A. Photo. 4. Fig. 5 The deposited pot in situ in the S.E. corner of room A 275 RESEARCH NOTE – ROmAN tidal inlet, facing a lagoon and the low lying ground of what is now called the Romney Marshes on the other side (see Fig. 7). About two and a half miles to the east of the marwood site lies the remains of the late Roman Shore fort at lympne, constructed around 250, guarding the same inlet (Davies 2017). About a mile and a quarter to the north-east of the villa is a large, designated Romano-British site discovered in 1972, where limited coin evidence suggests early occupation to the mid second century and occupation in the late third and fourth century (Bradshaw 1972; Alpin 1997), similar to the occupational pattern which was derived from the pottery at the Lympne fort. Clearly, the second-century villa at Marwood farm would have been contemporary to both this larger Romano-British villa site and also to the earlier naval fort at Lympne. In those parts of Romney Marsh that existed in Roman times there is substantial evidence of Roman pottery and briquetage of the first and second centuries which indicates that the marshlands were being exploited for salt production during this period. This in addition to the sheep grazing and cattle herding that were almost certainly being carried on during the summer months (Cunliffe 1988). One possible economic model for the Marwood farm villa would have been to supply the early fort with meat, corn, shell fish and salt, exploiting the various local resources. The abandonment of the villa in the last quarter of the third century is possibly part of a much wider pattern. Recent evidence suggests that the Folkestone villa was also abandoned in the late third century, although reoccupied in the fourth The Truncated Cone Lead Weight. In trench C., very close to the entrance to the building in the northwest trench, an iron slider key was uncovered which, it is presumed, fitted the lock on the nearby door of the building. In view of the absence of any coins uncovered during the excavation, dating evidence for the construction and occupation of the building depends heavily on two sources, the pottery assemblage and two pieces of patterned flue tile. The patterned tile fragments were found in the spoil heap and were identified as examples of die 16 and die 96 by Ian Betts of the Museum of London. Die 16 has two context datings from Canterbury, i.e. a context dated to 90/120 and a second context dated late 1/early 2 century. (Betts,Black and Gower, 87. 1997) The pottery assemblage, although relatively small ( 338 sherds) was referred to M. Lyne for identification and dating and his report is attached. Appendix 2. (For the purposes of identification, the pottery in the main trench falls into two groups:- the trench cut over the building was divided into one metre lengths AA to KK running southwards. The trench to the north of the building’s corridor was divided into one metre lengths running southwards A to T.) It is clear that the bulk of the pottery originated from the north Kent and Thameside industry kilns but there are a small quantity of foreign imports present. It is also evident that the bulk of the pottery dates from the second to third centuries. Based on the evidence of the pottery and the patterned tile, this Romano British building was probably built in the first quarter of the second century and Fig. 6 The Truncated Cone lead weight 276 RESEARCH NOTE – ROmAN century. In addition, there is little direct evidence of occupation on the lands of Romney Marsh after the end of the second century and extensive field walking found no Romano-British artefacts after the middle of the third century (Reeves 1995). The discovery of at least 13 coin hoards along the Sussex coast, deposited between 275 -290 (Cunliffe, 1973) strongly suggests that the threat of pirates was tidal river inlet, facing a lagoon and the low-lying ground of what is now called the Romney Marshes, on the other side of the inlet. FIGURE 3 Romney Marsh in the Roman period: a tentative assessment. B.Cunliffe (1988) Fig. 7 Romney Marsh in the Roman period: a tentative assessment ((B. Cunliffe 1988) 277 RESEARCH NOTE – ROmAN a reality at that time, highlighted by the imperial decision to appoint Carausius in 275 ‘to rid the seas of Belgica and Armorica of pirates’, i.e. the North Sea and the English Channel (Eutropius ix, 21). In addition, the decision, whether by Probus or Carausius, to build the Lympne Shore fort with its massive defensive walls and bastions, near the entrance to the inlet is the clearest possible evidence of a perceived seaborne threat. However, recent evidence has also shown that many inland Romano-British sites in Kent also experienced contraction during the third century and the abandonment of the Marwood farm villa may simply reflect this socio economic trend (Booth 2011). Perhaps the decline of the iron making and pottery industries in much of Kent after the second century and the abandonment of early naval forts like Lympne and Dover in the early third century, together with the contraction in inhabited sites and the threat from Saxon pirates, all stem from a diminished Roman naval presence in Kent compared with the previous century. acknowledgements The author would like to thank Paul Bowden and his parents, who own and farm the land where the villa is sited, for all the support they gave to this archaeological investigation. He is also indebted to the following who took part in the excavation work in November 2013: The Dover Archaeological group: gareth daws, mike Hartley, Christine Hodge, David Holman, Richard Hoskins, Gordon Hutchinson, Brian McNaughton, Les Moorman, Rob Mundy, Tina and Keith Parfitt and Pauline Sieben. The Folkestone Archaeological group: Anne and Thiery Biot, Lawrence Blomfield, Monica Butcher, Catherine Holtham Oakley, Iain Neilson, Harry Platts, Marilyn White and Barry Wright. Others who took part were Keith Dorman (Ashford A.G.) and Lorraine and Ken Harrison. He would also particularly like to thank Keith Parfitt and Iain Neilson for offering the services and expertise of their respective groups. Keith Parfitt’s notes on the excavation of Trench A were particularly valuable. And he wishes to thank Monica Butcher and Catherine Holtham Oakley of the Folkestone group for the many hours they spent annotating and cleaning the finds, pottery and tile. malcolm davies Alpin, J., 1997, ‘The Roman Villa at Aldington’, Kent Archaeological Review, 128, 194-5. Betts, I., Black, E. and Gower, J., 1994, A Corpus of Relief Patterned Tiles in Roman Britain. 87 8, Journal of Roman Pottery Studies, vol. 7. Booth, P., et al., 2011, On Track: The Archaeology of High Speed 1. Section 1. Kent, 334 6. Bradshaw, J., 1969, ‘Aldington’ (Reports from Local Secretaries), Archaeologia Cantiana, lxxxiv, 234. (A brief note on F. Newington’s fieldwalking discoveries at Falconhurst, i.e. Marwood Farm.) Bradshaw, J., 1972, ‘Aldington’ (Reports from Local Secretaries), Archaeologia Cantiana, lxxxvii, 231. Cunliffe, B., 1973, The Regni, 30. Cunliffe, B., 1988, Romney Marsh in the Roman period, in Eddison and Green, 85 6. 278 RESEARCH NOTE – ROmAN Davies, M., 2017, ‘The findings of various Archaeological Investigations at the Roman Naval Fort, Stutfall Castle, Lympne, 2014-16, Archaeologoa Cantiana, cxxxviii, 165-178. Eddison, J. and Green, C.S (eds), ‘Romney Marsh: Evolution, Occupation and Reclamation, Monograph 24, OUCA, Merrifield, R., 1987, The Archaeology of Ritual and Magic, pp. 1 36. Reeves, A., 1995, ‘Romney Marsh: The fieldwalking evidence’, in Eddison and Green. APPENdIx The Roman Pottery from marwood farm by Malcolm Lyne The excavation yielded 338 sherds (3,644g) of pottery from 41 contexts, of which the bulk belongs to the late 2nd and 3rd centuries. A few sherds are of Iron Age and earlier Roman date and there are a couple of Saxo -Norman cooking -pot fragments. c.25 bc -ad 100/120 There is very little ceramic evidence for Late Iron Age and pre -Flavian activity on the site, other than a fragment from a closed form in Gallo- Belgic White ware fabric BER7 from Square AA Context 5 (Rigby and Freestone 1995, Fabric 1B, c.ad 10- 70/100): some of the hand-made grog- tempered ware sherds could, however, be from Late Iron Age and pre -Flavian pots but their non -diagnostic nature makes this uncertain. c.120- 270/300 The near complete but disturbed grog -tempered cremation pot from Context II 5 beneath the floor level of the building is unfortunately not closely datable, although it is unlikely to be earlier than c.100 and later than 250. This cremation was probably a foundation deposit laid down at the time of the construction of the building and indicates that this event probably took place at some time after 100. Significant pot assemblages are present on the site from c.ad 120 onwards. The largest pottery assemblage from the site is from a broken tile and rubbish dump to the north of the villa building in Square I. The various contexts associated with this feature (I 7, I 10 and I 15) yielded 78 sherds (681g) of pottery between them, ranging in date between c.120/150 and 270+. The bulk of the sherds come from the Thameside industry kilns around the estuary of the River medway and include two in Cliffe BB2 fabric R14 from a Class 5C bowl and a 5E.2 dish (Monaghan 1987, c.150/70 -250 and c.120 -350 respectively), and 13 from Class 3H2 jars in very finesanded greyware fabric LR2.1 (c.150 -250). There are also 21 fragments in North Kent Fineware fabric R16 from a type 5B1.1 Dr.38 bowl copy (c.140 -250) and a globular poppyhead beaker (c.130 -200). Imported finewares include fragments from Central Gaulish Samian forms Dr.33 and Dr 37 (c.120 -200), a Moselkeramik beaker (c.200 -275) and a cornice -rim rough-cast bag -beaker in Rhenish Sinzig fabric (c.130- 250). Nine fresh joining sherds from another cornice- rim bag- beaker, but in silty orange fabric R8 with clay rough-casting, came from Context I 15 and are probably from a vessel which originated at the Dane John kilns in Canterbury (Kirkman 1940, c.130 -200). 279 RESEARCH NOTE – ROmAN The presence in Context I 7 at the top of the midden of jar fragments in Late Roman grog- tempered fabric LR1.1 (c. 250/70 -400+) and coarse- sanded Late Thameside fabric LR2.4 (c.270- 370), as well as beaker fragments in New Forest fabric LR12 (c.260- 340) and Oxfordshire Red Colour coat (c.240 -400+) indicate that the rubbish dump continued being added to until after c. 270. The latest Roman pottery from the site includes fragments from a developed beaded -and -flanged bowl in coarse -sanded Thameside greyware fabric LR2.3 from Context AA 2 (c.270- 370) and a dish of Bestwall type 8/6 in Dorset BB1 fabric R13 from Context GG 3 (Lyne 2012, c.220/70- 290/300). There are a number of sherds which can be broadly dated to c.270 -400 but none which belong specifically to 4th-century forms. A large number of coastal and other occupation sites in Sussex and Kent were abandoned at the end of the 3rd century but whether this was due to pirate raids, rising sea levels or some other factor is uncertain: the one thing they have in common is the appearance of Dorset BB1 vessels from production sites around Poole Harbour during their final occupation. It seems possible that the Marwood Farm villa was one of those abandoned occupation sites. (The full report is published on the KAS website) Lyne, M., 2012, ‘The Late Iron Age and Roman Black Burnished Ware Pottery’, in Ladle, L., Excavations at Bestwall Quarry, Wareham 1992 2005. Volume 2: The Iron Age and Later Landscape, Dorset Natural History and Archaeol. Soc. monograph Ser. No. 20, 201 242. 280 RESEARCH NOTES – mEdIEvAl THE MEDIEVAL FLOOR TILE PANEL AT ST LAWRENCE CHURCH, gOdmERSHAm This small Saxo-Norman church underwent major restoration and enlargement by Butterfield in 1864-6. The elegant 13th-century chancel escaped apart from the replacement of the medieval floor tiles with Victorian copies. Most of the old tiles were probably consigned beneath floors of the new south aisle, transept and vestry as hardcore. Fortunately, some rare examples in excellent condition were retained and re-set into a special panel. This was placed behind the choir stalls on the south side of the chancel on a low sill that separates the choir from the vestry (Fig. 1). The source of these decorated tiles was the Tyler Hill industry, located north of Canterbury, which was active in the late 13th/early 14th centuries (Pellett 2011). Tile wasters, retrieved during field walking at the foot of Canterbury Hill, Tyler Hill, and excavation of two pits at the top of the hill where it merges into St Stephen’s Hill, confirm them as products of these two prolific tileries. The layout of the 1860s tile panel The layout of the panel and the designs identified are set out in Fig. 2. The numbered designs are detailed in Table 1 below. The panel is made up of five rows of tiles with seventeen complete tiles and two part tiles (front) tapering to nine tiles and six part tiles (back). Thus it comprises an assemblage of sixty-eight complete examples plus thirteen part-tiles. Fortysix complete and part tiles are decorated, using twenty-one different designs, five unique to this church. There are thirty complete plain tiles plus five plain part-tiles in the panel. Description of each row (from front to back) 1st: single tiles (designs 7, 8, 4, 2, 3 and an extremely worn 20) set with two plain tiles between each as spacers. At the right-hand end a part tile (8) completes the row. Design 8 thus occurs twice, once on its side and secondly upside-down. 2nd-3rd: five quadrants are separated by single plain tiles. quadrants 1 and 5 at either end have a complete design of four fleurs-de-lis (1). The middle three quadrants are composed of a range of designs. There was obviously a problem in finding four compatible tiles. In the second quadrant design 14 is 281 RESEARCH NOTES – mEdIEvAl Fig. 1 The Tile panel at Godmersham Church. (Photo Paul Tubby.) 282 RESEARCH NOTES – mEdIEvAl Fig. 2 Layout of the tile panel (a) by design numbers (b) with designs drawn. 283 RESEARCH NOTES – mEdIEvAl TABLE 1. THE TILE DESIGNS AT GODMERSHAM No. design details No. in panel Notes Single tiles 1 Large fleur-de-lis set diagonally across the tile with a distinctive flattened frond pressed against one side. 9+1 frag Found in many churches in Kent; wasters recovered field walking at Tyler Hill. (British Museum catalogue no. 2127). 2 Six-petalled flower with central eye, set within a double circular band; small trefoil in each corner. 1 Another frequently used design and found as wasters at Tyler Hill. (BM no. 2325). 3 Hunting dog wearing a collar, barking and leaping over a branch of an oak tree. A beautifully drawn animal full of vigour and movement. (See Fig. 5) 3 A rare design. Recorded from excavations at St Augustine’s Abbey, Canterbury, and Saltwood Castle. (BM no. 1912). 4 Stag in flight, fleeing from a dog, jumping over a branch of an oak tree. Another elegantly drawn design matching the hunting dog (no. 3). 1+2 frags Very rare. One much worn tile recorded at St Clement’s Church, Sandwich. (See Fig. 6) 5 Stooping figure grasping a long staff, wearing a tunic with decorated cuffs, a long pointed hood covering the head and shoulders; the face is veiled. The tile has a distinctive white border. (See Fig. 7) 1 Set upside down. This rare design is recorded from St Nicholas, Harbledown, St Mary, Upchurch, the Maison Dieu, Ospringe and at St Gregory’s Priory, Canterbury. 6 A mythological griffin, the upper half a winged eagle; the hind part a lion. The head is more like a bull than an eagle although the creature is winged. The ‘lion’ hindquarters are cloven hoofed. A long tail sweeps forward between the legs. A white border to this tile. 1 Set upside down. Rare example, only recorded at Adisham, the Maison Dieu, Ospringe and Poor Priest’s Hospital, Canterbury. Recently found as a waster at Tyler Hill. 7 A large crenellated castle set diagonally across the tile. The castle has a central tower with two arched Norman-style windows and two doors on either side. Set back are two further doorways in possible gatehouses. 1 quite rare. It does occur more often as the corner element in a single tile. added to three tiles of design 10. quadrant three has two each of 9 and 14. quadrant four has four different designs (nos 11, 12, 15, 18). Fragments of the hunting stag tile 4 guard each end. 4th: made up of no. 18 (part tile), 3, 1, 3 and 5 (upside-down) with plain tile spacers. 5th: this last row is made up mostly of part tiles suggesting the medieval floor was very damaged with only a few tiles good enough to be retained. Some of the tiles are laid sideways or upside-down (nos 1, 6, 8, 12, 16, 17, 18, 19). 284 RESEARCH NOTES – mEdIEvAl 8 A flowering plant set within a droplet. A pair of leaves escape at either side. Tile has a white margin. 2+2 part- Found in Canterbury Cathedral, at St Mary, Brook, and St Catherine, Preston next Faversham; recorded from excavations at St Augustine’s Abbey and St Gregory’s Priory, Canterbury. (BM no. 1278). Quadrant designs 9 Tree of Life. Central stylised tree whose branches terminate in flowers with a central eye; set within a double circular band; small saltires enclosed in border; in each corner two half-flowers and a quarter box containing more saltires. 2 (x 4) when set with four matching tiles a complete pattern emerges. A very popular design occurring frequently. Recovered as a waster at Tyler Hill. (BM no. 2831). 10 Outer double ring with small double central ring. Foliage contained within the double bands. Stalked trefoils radiate out from the centre. In the corner a flower with a pair of leaves. 3 (x 4) A variant of the Tree of Life [no. 9] but much less common. Only other record from Holy Innocents, Adisham. 11 Geometric; diamonds radiating from centre, smaller triangles within triangles; enclosed within a double band; corners quarter diamonds. 2 (x 4) Scored across diagonally. Other examples: Canterbury Cathedral, St Mary, Brook; tile waster recovered from Darwin kiln, University of Kent at Canterbury. 12 Triple arcs containing sun, crescent moons and stars. 1 +half Other examples: St Augustine’s Abbey, Canterbury; St Mary, Brook and Duck Lane, Canterbury (Bennett and Ward 1995); Waster recorded at Tyler Hill. This design can also be used to form a six-tile, or a nine-tile design. 13 Sun surrounded by stars with full moons; saltires enclosed within a triangular double band. 1 frag. Other examples: Canterbury Cathedral, St Augustine’s Abbey, St Gregory’s Priory, Canterbury. 14 Double interlocking arcs; diamonds are contained within the bands; a pair of stylised leaves occupies one segment. 3 NB both quadrant/continuous design. Other examples: Canterbury Cathedral, St Augustine’s Abbey, St Gregory’s Priory, Canterbury; St mary, Maison Dieu, Ospringe. Continuous designs 15 Two birds within double cusped arcs; single fleur-de-lis in one corner enclosed within a quarter arc. 1 NB both continuous/quadrant design. Only example: St Augustine’s Abbey, Canterbury; wasters found in a pit in 1992 at the top of Canterbury/St Stephen’s Hill; also beside the Sarre Penn stream, Tyler Hill. 16 Interwoven arcs; central six-petalled flower within a circle; single oak leaf in each corner; small motifs along sides. 2 Unique to Godmersham. (See Fig. 8) 285 RESEARCH NOTES – mEdIEvAl 17 Chequer-board of small brown and cream diamonds. half tile Not previously recorded. 18 Bobbin effect set at right-angles produced by semi-circles set against the side of the tile. (See Fig. 9) 1 + half To produce a ‘windmill’ effect each tile is set at a quarter turn to the previous. (BM no. 2054). 19 Corner fragment with ellipse intercut by two arcs; a large dot in the corner. 1 Not previously recorded. Similar but not identical to other geometric, interlocking ellipses from Tyler Hill. 20 Linked interlocking double circles. 1 Extremely worn with no pattern remaining. Identified only by the depression caused by the stamp during production. Other examples: St Augustine’s Abbey, Canterbury; St Mary Ospringe. (British Museum no. 2065) 21 Triple interlocking ellipses; central sevenpetalled flower with pelletted eye. 1 Very worn; only identified by depression remaining after slip and glaze worn off. The replacement tile in the victorian chancel pavement is a direct copy suggesting it was quite numerous in the medieval layout. Not previously recorded. The Tiles The decorated tiles have the design inset in white slip. When glazed the slip results in an attractive honey colour with a chestnut background. The plain tiles are all glazed but without slip resulting in a dark brown colour, setting off the decorated examples. The tiles could have been used singly, or where so-designed as quadrants or as continuous repeating patterns on the original chancel floor. Details of the twenty-one different designs featured in the Godmersham tile panel are set out in Table 1 and are illustrated in Figs 3 and 4. [table 1 here] discussion Of the twenty-one designs recorded at St Lawrence six have been recovered as wasters at Tyler Hill (see Table 1) confirming the provenance of these tiles. These were mainly from the tile-yard at the foot of Canterbury Hill. The geometric design of diamonds (no. 11) was found incorporated into the kiln excavated at Darwin College, University of Kent, in 1968. A waster of two birds (no. 15) was found in a waster pit at the top of Canterbury Hill in 1993. Five designs (nos 16, 17, 19, 20, 22) are unique to St Lawrence Church and thus not found in the assemblages of tiles recorded in other churches having medieval tiles produced at Tyler Hill. The single designs are perhaps the most revealing as they are specific pointers to medieval religious belief and social life. Foliage plays an important part with the 286 RESEARCH NOTES – mEdIEvAl Fig. 3 Tile designs 1-11. 287 RESEARCH NOTES – mEdIEvAl Fig, 4 Tile designs 12-21. 288 RESEARCH NOTES – mEdIEvAl oak tree, ivy and acanthus prominent. Celestial bodies appear as the stars, sun and moon. The quadrant and continuous designs display the skills of the designer in producing complex, flowing patterns that make up the greater part of a medieval pavement, usually with a border of plain tiles to enhance their effect. Fleur-de-lis (no. 1): the fleur-de-lis or lily was frequently used in the medieval period to refer to the Virgin Mary. It appears carved in stone, in roof bosses and paintings. It was adopted by Edward lll appearing on the Royal Coat of Arms in the 14th century, supporting his claim to the French crown. (The fleur-de-lis was removed from the Royal Coat of Arms in 1801 when George III replaced it with the Hanoverian lions). Hunting scenes (nos 3, 4) (Figs 5 and 6): England was famous for breeding pedigree hunting dogs and they were much in demand. Hunting was a favourite pastime of the medieval kings with their barons and churchmen. Two sets of hunting scenes have been identified as wasters in the Tyler Hill database. The Godmersham set is composed of two tiles; hunting dog on one, stag on the other. Both are elegantly drawn with agile animals gracefully leaping over branches. They were produced in a kiln at the foot of Canterbury Hill as confirmed by the recovery of a waster during field walking. At the excavations at St Augustine’s Abbey the same design of tile of the hunting dog appears. There is also a second version, featuring a dog clearing the branch of Fig. 5 Tile 3 Hunting dog. (Photo Paul Tubby.) 289 RESEARCH NOTES – mEdIEvAl a different species of tree (rather than an oak) and there is a different retriever-type dog below. The tile also has a margin. A second set of hunting scenes can be seen in the Jesus Chapel, Canterbury Cathedral. It is a simplified version featured on two tiles: the stag with an arrow through its neck being attacked by a dog on the first; and the huntsman having released his arrow from his bow hiding behind a tree on the second tile. These tiles were excavated from a waster pit at the top of Canterbury Hill in 1993 where it merges into St Stephen’s Hill during pipe-laying by the water board (Cotter1995) confirming its provenance as a Tyler Hill product. Tiles with different hunting scenes have been recovered from excavations at the Poor Priests’ Hospital, Canterbury, from 1976-81. There are two designs in this set. In the first a stag in flight, arrow through its neck is being attacked underneath by a dog, similar to but not identical to the Jesus Chapel tiles. On the second tile of the huntsman hiding amongst trees is more of a caricature with a symbolic tree. This further distinct set suggests a third, as yet undiscovered tile-yard operation in Tyler Hill in the late 13th/early 14th century. A fourth, very basic set of hunting scenes was recorded from St Gregory’s Priory, Canterbury, between 1988-1991 (Hicks and Hicks 2001). They are crudely drawn depicting a stag in flight and the second of two dogs running. Unusually, slip provides the background. Horton (ibid.) suggests that a different technique, ‘stamp Fig. 6 Tile 4 Stag inflight. (Photo Paul Tubby.) 290 RESEARCH NOTES – mEdIEvAl on slip’ was used in the production whereas the majority of the tiles were ‘slip over impression’. He proposes these as later tiles, possibly an off-shoot from the known tile-yards using another set of stamps. Stooped figure (no. 5) (Fig. 7): pilgrim or leper? This figure could be interpreted as a pilgrim who would have been a familiar figure on medieval roads after the murder of Becket in 1170. Alternatively, it has been proposed that this is a leper begging for alms. This seems less likely as reference to such a dreadful disease might not seem appropriate in such a work of art as a tile pavement although it does serve to emphasise man’s tenuous hold on life. The pilgrim theory is reinforced by the fact that of the four examples so far recorded, three occurred in churches along the Watling Street pilgrim route to Canterbury. Thus examples of this tile, all made from the same mould, can be seen from the Maison Dieu, Ospringe and in St Nicholas Harbledown. St Mary, Upchurch, has a much worn example. St Lawrence, Godmersham, is however on a south-western approach to Canterbury, the famous Pilgrims’ Way. The Castle (no. 7): this probably represents the castle of Castile, a reference to the birthplace of queen Eleanor, beloved wife of Edward I (d.1290). The castle tile surviving at Brook and a similar one in Canterbury Cathedral both have four doors in their design whereas Godmersham’s only has two. The footprint of all three is identical suggesting either that the mould was damaged losing the central section but remaining in use or that in the replacement mould the design was simplified. Fig. 7 Tile 5 Stooping figure. (Photo Paul Tubby.) 291 RESEARCH NOTES – mEdIEvAl Fig. 9 Tile 18 Bobbin, windmill effect. (Photo Paul Tubby.) Fig. 8 Tile 16 Complex of arcs, flowers and leaf. (Photo Paul Tubby.) 292 RESEARCH NOTES – mEdIEvAl The Tree of Life (no. 9): the medieval tree of life has been proposed as the Peridexion Tree. It had healing properties so was associated with the suffering of Christ nailed on the wooden cross, suggesting the healing power of Christ. The Tree of Life was guarded by a griffin. Winged Griffin (no. 6): this hybrid mythological creature was brave and strong. Its lion traits supposedly made it strong enough to carry off an ox or capable of tearing a man to pieces. yet it was also credited with attributes of the eagle combining clear vision with the ability to soar into the heavens. A redeeming characteristic was its role in the guarding of the Tree of Life. Conclusion This interesting panel, rarely noticed by visitors to St Lawrence Church behind the choir stalls, adds valuable information about the production of tiles in Tyler Hill in the late 13th/ early 14th century. New designs have been added to the database. The tiles were probably commissioned by Christ Church Priory (which held the church throughout the medieval period), thus accounting for their fine quality. After decorating the chancel floor for five hundred and fifty years most must have been much worn. Those fortunately rescued by Butterfield give us a hint of the splendour of the pavement in the 14th century. For that we are grateful. irene pellett bibliography Bennett, P. and A. Ward, 1995, ‘Duck Lane Car Park’, Canterbury’s Archaeology 1993-4, 10-13. (Unpubl. tile report by I. Pellett in archive.) Cotter, J., 1995, ‘Tyler Hill Rising Main’, Canterbury’s Archaeology 1993-1994, 15-17. Eames, E., 1980, Catalogue of Medieval Lead-glazed Earthenware Tiles in the Department of Medieval and later Antiquities of the British Museum, London. Hicks, M. and A. Hicks, 2001, St Gregory’s Priory, Northgate, Canterbury. Excavations 1988-199, Vol. ii, 187-211, The Archaeology of Canterbury New Series, CAT. Horton, M.C., 2001, ‘The Floor Tiles’, in Hicks and Hicks. Pellett, I., 2011, ‘The Medieval Decorated Tile Pavement at St Mary’s Church, Brook: the finest survival of the Tyler Hill floor tile industry’, Archaeologia Cantiana, cxxxi, 43-64. 299 RESEARCH NOTES – EARLY MODERN ‘pARADiSE fOuND’; piNpOiNTiNg the embarkation TOwERS ON THE MODERN MAp Of DOvER Of the many plats and maps that show the early harbour situated at what is now the Dover western Docks, possibly the most interesting is dated 1538 and resides in the British Library.1 it shows an overwhelming amount of detail of what constituted the earliest phases of harbour development on that side of Dover (Fig. 1). The origins of this harbour, started around 1500 and named ‘paradise’ are, however, rather obscure. fig. 1 Section of BL Cotton MS Augustus i.i.22.23 showing paradise pent at Dover, c.1538 (archaeologia Cantiana, xlix, 1937, facing p. 112). RESEARCH NOTES – EARLY MODERN 300 More or less contemporary with this plat is the painting the embarkation at Hampton Court palace and which depicts Henry viii leaving Dover paradise in 1520 with a fleet of warships to visit the French King, Francis I (Fig. 2).2 the embarkation famously shows Henry aboard a powerful warship, thought to be the henri Grace-de-Dieu, amongst a large fleet, with Dover Castle strikingly in the background and in the foreground, a pair of large gun towers.3 the embarkation is the only image to show the two towers at full height, only the remains of one of them being visible in the BL plat. Both the date and artist of this painting are unknown and the Royal Collection website, where the painting is listed, dates it to between 1520 and 1540. (its value as a guide to the topography of paradise harbour is therefore questionable, as discussed below). Today the area that constituted paradise harbour is reclaimed land with a road and a grass covered bank leading to a viaduct. Over the last few centuries much has been written about the development of the western harbour, largely in the form of ‘potted histories’ of the town, port and garrison, using mostly second and third-hand intelligence. This reliance is certainly due to the scarcity of reliable primary documentation. One must also recognise that these ‘perambulations’ were compiled and published with a spirit of commercial opportunism rather than a dogged adherence to historical authenticity.4 Nevertheless, using some of the details in these potted histories together with some plats and the few surviving primary documents from the sixteenth century it is possible to discern the key stages in paradise harbour’s creation. it is generally accepted that the two towers which stood in paradise were located in what centuries later became Round Tower Street and Round Tower Lane, and of course, this makes sense.5 unfortunately for our purpose, those two street names change location depending on the date of map referred to (see below).6 fig. 2 the embarkation of henry Viii. at Dover, may 31, 1520, Royal Society of Antiquaries, 1845, engraved after the original picture preserved in Hampton Court. RESEARCH NOTES – EARLY MODERN 301 The stages in the development of the harbour which emerge from the various published histories of Dover seem to be as follows. probably in 1495 Sir John Clerk, Master of the Dover maison Dieu, completed a pier or strong bank under the shadow of Archcliffe to enclose that end of the bay.7 This pier was intended to prevent the bay filling up with shingle and sand (‘beach’). On the pier he built two stone towers which had rings attached to them with which to tie off ships.8 This area of water constituted a safe haven for shipping and for this reason became known as ‘paradise’.9 Perhaps a single tower was built firstly at Paradise around 1500 by John Clerk.10 Alternatively, a bank of earth and chalk was constructed during the reign of Henry vii (1457-1509), and on it a single tower built, and the bank extended and a second tower added at some time in the structure’s life.11 Or controversially, a pier was built out from the cliff under Archcliffe fort towards the approximate site of the present Lord warden House during the reign of Henry vii, with two towers being built at different times.12 By around 1530 the eastern of the two towers was ruinous and paradise was silting up again with beach.13 finally, upon John Clerk’s death the next Master of the Maison Dieu, Sir John Thompson also prepared a plan for improving the harbour, and proceeded to build another pier.14 The limited primary sources do suggest that masonry work was carried out to an existing stone structure at Dover paradise in 1510, and that masonry and timber repairs were completed between 1513 and 1515.15 John Clerk is not mentioned until he becomes a ‘warden of the wyke’ along with Richard fineux in March 1518, at which time the accounts mention ‘the repayryng and byldying of the seid wyke made by the seid master of the mesonduwe’.16 in 1521 Clerk is paid for supervising works at the wyke for the proceeding four years.17 Comparison of the two towers in the embarkation do show substantial differences in design and these along with this reported expenditure in the wyke accounts could point to a rebuilding or modernizing of the outer tower (stylistically the later of the two). This could very likely give the inner tower a date nearer 1500 and the outer perhaps being the subject of these 1517-21 works. As noted earlier, using the the embarkation depiction as evidence of changes in the early harbour is problematic. Over the years this work has been the subject of much academic study with various dates and authors being proposed. for example; it has been argued that even though it depicts an event that happened in 1520, it cannot have been painted any earlier than 1537 because the figure of Henry standing on the Grace-de-Dieu is modeled after Holbein’s portrait of the King of that year.18 Another theory dates it to at least about 1550 because neither it nor its sister painting (the Field of the Cloth of Gold) show up in a 1547 inventory of whitehall palace, yet this could simply be because both paintings were set into its walls at this time.19 Indeed the style of dress displayed by the figures in the embarkation seem to favour this later date according to one academic.20 The painting has also been attributed by some to the Dutch painter, Cornelius Anthonisz (1505-1553), again relying on the clothing style of the people milling around the scene. This view argues that the figure of Henry VIII in the painting is less in the style of Holbein and more in that of Anthonisz, giving it a latest date of production of 1553.21 And one final possible clue, a Tudor inventory of warships22 produced by Anthony Anthony, a government contractor-cum-official, earlier (in 1546) may have been inspired by the embarkation.23 RESEARCH NOTES – EARLY MODERN 302 The lack of consensus regarding its provenance weakens its authority as a reliable record and as one historian has aptly put it (writing about the provenance of both paintings), ‘the fragmentary nature of the available evidence lends itself to a fantasy of identifications … one could accumulate many such tempting hypotheses’.24 One thing is certain and perhaps all we will ever be able to say for sure is that the painting ‘must have exercised the hands of artists of considerable skill’.25 On the BL plat is shown the remains of one of Clerk’s towers, and to the left (west) of it the remains of a pier-like structure (Fig. 3). This tower is reduced in height compared to the towers in the embarkation, and seems to have survived in this form for at least a century, indeed it appears like this along with the pier-like structure in two plats at The National Archive dated to 1577 and between 1558 and 1597 respectively.26 in all three plats the tower’s height is greatly reduced, the pierlike structure partly demolished and the shortened tower has a timber building atop. This is clearly the remains of the eastern tower. Because Paradise continually filled up with beach, piers (jetties) were built further out to sea (presumably by Clerk and then by Thompson) in an effort to arrest the process, and this theory would make the eastern tower later in date than the western one. it would seem logical then that by the time the plat was produced the one inside paradise (the western one) was not shown because it had become dilapidated and had been washed away. This would make the pier-like structure in the plat Clerk’s pier, its western end damaged and now redundant. fig. 3 Detail of BL plat (see fig. 1) showing the remains of John Clerk’s eastern tower and to the left of it the remains of his pier, c.1538. RESEARCH NOTES – EARLY MODERN 303 The early pier that Clerk had a hand in maintaining and repairing could not in due course stop beach rounding its end and blocking paradise. Throughout his tenure during the 1520s Clerk sought solutions to this problem.27 By November 1533 the harbour was in such a poor state that John Thompson, Rector of St James’ Church (who was to succeed Clerk as Master of the maison Dieu in 1535) sent a petition to Henry’s minister, Thomas Cromwell, asking for the King’s indulgence in getting it repaired. He claimed that it was ‘now utterly destroyed’.28 Another petition was sent in January 1534 from the mayor and burgesses of Dover asking that a new harbour be built, outlining that ‘the harbour is quite closed up’.29 By July 1535 work had started on what was in fact Thompson’s harbour scheme, and this involved building new piers further out to sea from Clerk’s pier in an effort to prevent yet again, the accumulation of beach.30 it is probably the BL plat that was produced at Henry’s request after visiting Dover in September 1538, having spent considerable time at Thompson’s new harbour workings.31 The British Library attributes this plat to Richard Lee, surveyor of fortifications at Calais from 1536 to 1542, but Lee was in England in 1538 so could have drawn this plat as the BL suggests.32 However, Henry demanded that Thompson produce the plat within two weeks, and at such short notice it is perhaps more likely that Thompson himself would have drawn it. One problem, however, in using this plat to untangle the harbour’s development is that it reveals both what had been built and what was proposed to be built which has confused later commentators.33 Trying to position Clerk’s towers onto a modern map using Tudor sources alone is impossible. in order for it to be successful there would have to be Tudor or earlier structures still standing to use as reference points, and alas there is none. As already noted, the area that Paradise once filled is now fairly featureless reclaimed land. Nevertheless, william Batcheller’s 1828 perambulation does in fact give the location of one of the towers in relation to buildings that were standing in the late eighteenth century.34 He draws a graphic but pointed picture of the process by which paradise silted up: ‘the space occupied by the harbour in Little paradise … became a waste, useless, and unhealthy swamp, covered with reeds and bull rushes … the ground was occasionally raised … houses were built on it … prior to the year 1798’.35 He goes on to pinpoint the position of one of the towers in relation to buildings that were possibly still standing when he was writing; ‘the foundations of this tower remain in Round Tower Street, under three houses built by Mr. Church in 1798, and under a storehouse belonging to Mr Reynolds’.36 Mr Reynolds remains elusive, but Jatt Church has proved easy to trace. He was a shipwright and Clerk of Dover Harbour and owned property in Round Tower Street until his death in April 1808.37 His property was insured by Sun alliance, and their ledger from 1789 lists three houses owned by him in Round Tower Lane.38 Also, a plat dated 1796 of Dover paradise as reclaimed land, complete with streets and properties laid out, shows precisely where his houses were in Round Tower Lane (Fig. 4).39 Although there are two apparent anomalies in these details – Round Tower Street/ Lane and 1798/1789 – we know that the road names were variously recorded from time to time and it is conceivable that Batcheller transposed the last two digits of the year. Although he offers no direct evidence that the remains are of one of the towers, he mentions its position in relation to structures that perhaps he was familiar with. RESEARCH NOTES – EARLY MODERN 304 Fig. 5 shows plat KHLC DHB/P96 overlaid onto Ordnance Survey map Dover Number 3 Edition (1905).40 By careful alignment of the two layers, Jatt Church’s houses fit perfectly along the north side of Round Tower Lane. The key to this plat indicates that the row of houses is situated in what was the ‘wet’ side of the southern boundary of paradise. Fig. 6 shows the same plat of paradise overlaid onto a modern map of Dover. This shows Jatt Church’s houses pinpointed across what is now Bulwark Street, giving us the position of the western tower. placing a tower at this point correlates with the theory that Clerk’s pier was not built out from Archcliffe to the then south pier head (site of the present Lord warden House). However, Batcheller could be mistaken and the remains he refers to under Church’s houses could be another ancient structure, although their position on this edge of paradise on a piece of land that had only recently been sea water makes this theory most likely. According to the Dover express (1866), during the demolition of buildings at Round Tower Street on land then owned by the London, Chatham & Dover Railway, contractors found what they believed to be part of one of the towers.41 These demolitions seem to have been carried out in relation to improving the steamer service between Dover and Calais, which entailed the building of new structures in this area.42 we know that these foundations were not of the tower fig. 4 plat of Dover paradise as reclaimed land and laid out for streets and houses, 1796. Jatt Church’s houses marked with a star (East Kent History & Learning Centre, DHB/ p96). RESEARCH NOTES – EARLY MODERN 305 that Batcheller mentions because the area pinpointed for the western tower is still extant on a map of 1905, some 40 years after the railway excavations. it does seem feasible for the eastern tower to be at the other end of Round Tower Street (or Lane) close to the course of the railway because this is where the south side of the ‘neck’ of paradise once was; this area can be seen on the 1905 Ordnance Survey map overlay (fig. 5) and is just south-west of where a later railway spur veers away from the mainline, today locating the eastern tower approximately in a piece of disused land sandwiched between the railway and a pair of sewage treatment tanks. The generally accepted narrative regarding the towers at Dover paradise perpetuated the idea that they were both built by John Clerk. it would seem reasonable to accept that there were indeed two towers at paradise, certainly the westernmost one by 1510, perhaps 1500, and the easternmost one maybe either around the same fig. 5 Map overlay of KHLC DHB/p96 onto Ordnance Survey, Dover Number 3 Edition (1905). Jatt Church’s houses marked with a star. RESEARCH NOTES – EARLY MODERN 306 date but repaired and remodeled by 1521 or alternatively, built from scratch at this later time. The aptly named Round Tower Street and Round Tower Lane that once existed give a good idea of where the towers were situated approximately, but until now the towers have not been positioned on a modern map of Dover. Batcheller’s perambulation of Dover pinpoints the location of the western tower, and although he has no a priori evidence that the foundation that he writes about is that of one of the towers, using the plat of paradise, the insurance ledger and Jatt Church’s will it is clear where his houses were, and positioning the western tower at this point makes sense. The discovery of what was supposedly the remains of the eastern tower by railway workers in 1866 is fortuitous, and if the relative positions of the two towers in the embarkation is anything to go by, it seems a very likely spot for the eastern tower, notwithstanding the painting’s shortcomings as a source. Long before the turn of the century when the Eastern Docks complex was not the vast acreage of reclaimed land that it now is, the relatively small paradise haven on the other side of Dover bay had probably passed from Dover’s consciousness. Yet the determination of those trying to combat the elements and control the ‘cruel sea’ five centuries ago illustrates the importance of Dover’s political and economic relations with the Continent during the sixteenth century. And with the current ongoing enlargement of the western Docks, what more apposite moment to be able to pinpoint precisely where the embarkation towers stood at Dover paradise.43 Fig. 7 shows the embarkation towers on the modern map of Dover. Fig. 6 Map overlay of EKHLC DHB/p96 onto Open Street Map of Dover pier District © OpenStreetMap contributors. Jatt Church’s houses marked with a star. RESEARCH NOTES – EARLY MODERN 307 acknowledgements The author’s thanks to paul wells for constructing the map overlays and to Dover Harbour Board for permission to use plat of Dover Paradise lodged at the Kent History & Library Centre, Maidstone. jeff howe bibliography Anon., 1837, ‘A Naval Officer’, a brief history of Dover and ramsgate harbour ….. London. Anglo, S., 1966, ‘The Hampton Court painting of The field of the Cloth of gold Considered as an Historical Document’, the antiquaries Journal, vol. XLvi. British Library, Cotton MS Augustus i.i.22.23. Batcheller, w., 1828, a new history of Dover and of Dover Castle during the roman, Saxon & norman Governments, with a Short account of the Cinque Ports, Dover. Bavington Jones, J., 1938, annals of Dover, Dover. British School, 16th century, The Embarkation of Henry viii at Dover c. 1520-40, Royal Collection Trust website, https://www.royalcollection.org.uk/collection/405793/theembarkation- of-henry-viii-at-dover, accessed 7 September 2017. fig. 7 Open Street Map of Dover pier District © OpenStreetMap contributors. The two stars mark the positions of ‘The Embarkation’ towers. RESEARCH NOTES – EARLY MODERN 308 Colvin, H.M. (ed.), 1982, the history of the king’s Works, Vol. iV 1485-1660 (Part ii). Dover express, 25 November 1865, ‘important Schemes for international Communication via Dover’. Dover express, 9 March 1866, ‘Henry the Eighth’s Tower’. Hasensen, A., 1980, the history of Dover harbour, London. Hasted, E., 1749, the history and topographical Survey of the County of kent, Volume iV, Canterbury. Holinshed, Chronicles 1587: iii, the holinshed Project http://english.nsms.ox.ac.uk/ holinshed/texts.php?text1=1587_9135#p17271 accessed 10 November 2017. Kent History and Library Centre DHB/P96. London Metropolitan Archive 359.554656. Lyon, Rev. J., 1813, the history of the town & Port of Dover & of Dover Castle with a Short account of the Cinque Ports, Dover. MacDonald, A., 1937, ‘plans of Dover Harbour in the Sixteenth Century’, archaeologia Cantiana, 49. Member Biographies, Lee, Sir Richard (1501/2-75), of Sopwell, Herts, The History of parliament website, http://www.historyofparliamentonline.org/volume/1509- 1558/ member/lee-sir-richard-15012-75, accessed 1 November 2017. Millar, O., 1963, the tudor, Stuart and early Georgian Pictures in the Collection of her majesty the Queen, London. Open Street Map of Dover pier District © OpenStreetMap contributors. Ordnance Survey, Dover Number 3 Edition (1905). Statham, Rev. S.p.H., 1899, the history of the town and Port of Dover, London. ‘The Embarkation’, Royal Society of Antiquaries, 1845. The National Archive Mpf 1/122/2; Mpf 1/122/3. The National Archive, will of Jatt Church, pROB 11/1483/289. 1 BL Cotton MS Augustus i.i.22.23. 2 British School, 16th century, The Embarkation of Henry viii at Dover c.1520-40, Royal Collection Trust website, https://www.royalcollection.org.uk/collection/405793/the-embarkation-ofhenry- viii-at-dover, accessed September 7 2017. 3 Millar, 1963, 55. 4 Thus Batcheller’s Dover perambulation introduction states, ‘ground plans would have been very desirable in our description of the harbour … but the expense would have greatly enhanced the price of the work’. This could be interpreted as clarity making way for profit. He also ran the King’s Arms Library in Dover, was a publisher and started the Dover telegraph newspaper in 1833. The Rev. Statham’s history contains his publisher’s full catalogue which runs to an extra 32 pages of books available for purchasing. 5 These street names go back at least as far as 1641 and disappear in the early twentieth century during slum clearances. 6 for example, Round Tower Lane on Eldred’s 1641 plan of the harbour becomes Round Tower Street on the Ordnance Survey map of 1905. On this later map Round Tower Lane is the road under which it is presumed that one of the towers is located. 7 Statham, 1899, 90, 91. The name ‘Archcliffe’ is used to indicate that part of the cliff upon which Archcliffe fort now sits, notwithstanding that a fort or bulwark was not constructed there until the 1540s. 8 Statham, 90. 9 Anon. 7, 8. 10 Holinshed, Chronicles 1587: iii, the holinshed Project http://english.nsms.ox.ac.uk/holinshed/ texts.php?text1=1587_9135#p17271 accessed 29 September 2016 (1536).; Hasted, Volume iV, Canterbury, 1749, 81; John Bavington Jones, annals of Dover, Dover: 1938, 91. 11 Reverend John Lyon, 1813, 150, 151. RESEARCH NOTES – EARLY MODERN 309 12 Batcheller, 1828, 280. 13 ibid., 281. 14 Bavington Jones, 92, 94. 15 Colvin (ed.), 729. 16 ibid., ‘Warden of the Wyke’ being an official position in connection with the harbour, the Wyke being the harbour. 17 ibid. 18 Millar, 55. 19 ibid. British School, 16th Century. 20 Millar, 55. 21 N. Beets, ’Cornelis Anthonisz: i. De Historie-stukken’, oud holland, 56 (1939), 184. 22 Millar, 55. 23 C.S. Knighton and D.M. Loades, ed., 2000, the anthony roll of henry Viii’s navy, 20. 24 Anglo, 305, 306. 25 John gough Nichols, ‘Letter’, archaeologia, 39 (Jan 1863), 28. 26 The National Archive Mpf 1/122/2 & Mpf 1/122/3 27 Colvin (ed.), 731. 28 ibid.; James gairdner (ed.), 1882, Letters and Papers, Foreign and Domestic, henry Viii, Volume 6 (London) item 1472, 591. 29 gairdner (ed.), Letters and Papers, item 66, 27. 30 Colvin (ed.), 732. 31 MacDonald, 111., Colvin, 744. 32 Member Biographies, Lee, Sir Richard (1501/2-75), of Sopwell, Herts., The History of parliament website, http://www.historyofparliamentonline.org/volume/1509-1558/member/lee-sirrichard- 15012-75, accessed 1 November 2017. 33 Colvin (ed.), 744. 34 Batcheller, 280. 35 Batcheller, 295. 36 ibid., 280. 37 The National Archive, will of Jatt Church, pROB 11/1483/289. 38 London Metropolitan Archive 359.554656, 278. 39 Kent History and Library Centre, DHB/P96. 40 Overlays done using Adobe photoshop Elements v.9.0 on windows 10 pC. 41 Dover express, 9 March 1866, 2. 42 Dover express, 25 November 1865, 3. 43 The site of the western tower is at NgR TR 31644 40406 and that of the eastern at TR 31674 40464. 309 RESEARCH NOTES – EARLY MODERN DiSCOvERiNg AND RECORDiNg TwO LOST HAMLETS iN THANET This note compares and contrasts the character and history of two neighbouring Thanet hamlets, upton (in the parish of St peter the Apostle) and Hollicondane (St Lawrence) (Fig. 1), and is a summary of a longer study of both (published on the kaS website). The Trust for Thanet Archaeology (TTA) excavated the grounds of the demolished Upton house in 2006. As a member of the isle of Thanet Archaeological Society (iOTAS), the author became involved and later discovered that the nineteenthcentury house had replaced an earlier farmhouse. Across the road was Little Upton, a grade ii former farmhouse c.1675. it was determined that here were the remains RESEARCH NOTES – EARLY MODERN 310 of a farming community that had been swallowed up by the expansion of the developing seaside resort of Broadstairs. further research revealed that upton farm had been owned by one family for at least 200 years, namely the goodsons; the land they farmed at one time reaching down to the coast. it transpired that the goodsons had been living in St peter’s for 400 years, but by the late twentieth century were no longer in the area. fellow iOTAS member Jenny price (a family researcher) discovered, as well as many of the above, william goodson, a farmer, had died in 1876 at St Lawrence. The author sought out his grave and the stone stated, wiLLiAM gOODSON LATE Of HOLLiCANDANE. This place name is not shown on modern maps but has fig. 1 Extract from OS first Edition map of Thanet (1889) showing location of the two hamlets, 1.75 km (1 mile) apart. RESEARCH NOTES – EARLY MODERN 311 been noted in various spellings for 400 years on old maps and documents, so more research was undertaken (kaS newsletter Summer 2017). The author has shown that a footpath passing north-south through upton connected to tracks and roads that ran between Reading Street and the west side of Ramsgate in a direct line, was metalled in the Roman period in the northern section with three farms with ‘street’ in their names located on it. This trackway passed upton on its west, passed Hollicondane to the east within 200m (fig. 1). The goodson family could walk between the two in under 20 minutes. upton Literally high farm/hamlet, it is on a plateau above the coast at Broadstairs. The name has probably remained unchanged since first recorded; although Lewis gave an alternative spelling of uptown. As well as Little upton mentioned above, eight other properties connected to the hamlet survive. They are: Upton Lodge, Listed grade ii, much altered and of uncertain date, eighteenthcentury or earlier. At one time had large ornate grounds and was likely the ‘delightful summer residence’ mentioned by John Mockett 1836, is now surrounded by 30 bungalows on its former land. Weasel Cottage c.1888 adjacent, housed the Lodge’s gardener. minters c.1770 of brick and flint was originally two cottages known as Hogbins but was converted into one for a goodson with the middle name of Minter. Single storey orchard Cottage c.1760 and of flint and brick was originally a farm building but used as domestic accommodation from the nineteenth century. Next door is the oast house listed Grade II in flint and brick with a slate oast roof, tiled elsewhere is probably eighteenth-century. wooden stables are still in the yard but a cottage has gone. Another was originally the farm’s slaughter house, little original surviving. brazier Farm, demolished 1861, was close by. the Cottage, of uncertain date and again originally probably an agricultural single storey building became two-storey 1 and 2 Upton Cottages until unified in 1955. 5 and 6 upton Cottages were built in 1860 but 3 and 4 upton Cottages were demolished in the 1960s to make way for road re-aligning. A legal document mentions a 9 upton Cottage but research has not located it but Orchard Cottage is the most likely, and presumably there were numbers 7 and 8, but these also are unknown unless they are the Hogbins Cottages (see Minters above). The last owner of upton House mentioned wooden farm buildings around the house when he bought it and photos, paintings and local memory attest to other lost agricultural buildings including a Dutch barn. Hollicondane Of uncertain meaning it is located in a dip on a south-facing slope, the farm fields crossing the valley to Ramsgate’s boundary. Twenty-two different spellings were noted (see KAS Newsletter). A recent discovery is Howlettendane in a 1617 Church Rate Book. ‘The farmhouse … is brick with gables, and is dated 1678’ both according to Cotton’s history of St Lawrence, 1895. Curvilinear gables? RESEARCH NOTES – EARLY MODERN 312 One possible survivor is a ‘potato’ store that was close to the farmhouse north of College Road. Other buildings were south of the road – only a track in the middle nineteenth century. A fire in 1865 destroyed farm buildings and a barn but the house and two or four cottages (depending on source) survived – one a tavern. The latter was replaced by a fine Victorian building but in turn that closed and was demolished. The whole site taken over by South Eastern (now St Lawrence) College with land from Newlands grange and Hollicondane farm being purchased in small and large plots from c.1877-1890. Thus both hamlets were farmed by the goodson family and connected by the north-south track together with some significant archaeology – Bronze Age barrows and Roman remains at Hollicondane, iron Age/Roman route and Roman remains at upton. The contrast between the two are revealing. Although both on Chalk their topography is different. The tiny hamlet of Hollicondane had an inn, the larger upton did not. pictorial representations of upton survive, none have been located of Hollicondane. No fewer than twenty-three spellings of Hollicondane have been noted, only two of upton. The goodsons of upton, in good years, invested in property so had many assets to sell when farming got difficult. Upton (and associated investments) was sold off in 48 transactions over 49 years. The goodsons sold their holding at Hollicondane in one transaction. upton has nine surviving buildings, Hollicondane only one. Thanks are due to Mr M. Daniel of Daniel & Edwards for allowing access to over 200 years of documents relating to the goodsons; Andrew Brown of St Lawrence College for their early history; Jenny Price of IOTAS; Kent’s libraries and the residents of upton. d. gordon taylor 312 RESEARCH NOTES – EARLY MODERN A SERIES OF KENTISH FIREBACKS AND THE POSSIBLE IDENTIFICATION OF THEiR fOuNDER in volume 29 (1911) of archaeologia Cantiana H.S. Cowper described a group of iron firebacks that all bear the same heraldic shield in relief though repeated in differing numbers.1 He identified the arms as those of William Ayloffe (c.1535- 84), a judge from Hornchurch in Essex, and his wife Jane née Sulyard. Cowper established a distant connection between the Ayloffes and the Kentish village of Boughton Aluph (near Ashford) but suggested a possible religious motive behind their production. More likely is that the Essex Ayloffes commissioned a fireback from one of the Kent furnaces, providing the armorial stamp for the purpose. Remaining at the furnace, this was subsequently available for re-use. Cowper’s interest had been inspired by his ownership of one of these firebacks at his then home of Loddenden Manor, Staplehurst. He noted several other examples at locations in west Kent, most of which bore a date separating one or other of two pairs of initials, CT or RS, inside a rectangle with a cavetto-moulded frame. The single appearance of initials on a fireback is usually thought to represent the person who commissioned it. Sometimes initials occur in triad for a husband and wife. However, the repeated occurrence of the same initials on different firebacks, RESEARCH NOTES – EARLY MODERN 313 Fig. 1 Iron fireback of 1629 bearing the arms of Ayloffe impaling Sulyard, at Poundsford farm, Burwash, East Sussex (photo, author). sometimes over a period of time, suggests that they are those of either the founder or the pattern-maker. A group of firebacks cast in 1582 all bear the initials IA though clearly made for different people.2 As the decorative details on those firebacks were individually stamped into the mould the initials are likely to have been those of the founder. By contrast, the single wooden patterns carved to form the moulds of another group, dating to the 1640s and 50s, were evidently the work of a craftsman identified as IM but who has not otherwise been identified.3 The small rectangular panels bearing the dates and initials on the Ayloffe firebacks are positioned centrally above the arrangement of shields. Of the 12 dated examples identified by Cowper or known to the author, the initials CT appear on nine, dated to between 1601 and 1630, and evidence has emerged that suggests this founder’s likely identity. With one exception the dated examples of Ayloffe firebacks that Cowper recorded were in Kent locations; the exception was at ‘Pounceford’ Farm. This is Poundsford Farm, at Burwash Common in East Sussex, and the fireback had been first noted there in 1869,4 although it was not identified as one of the series until nearly 30 years later.5 The fireback is still there (Fig. 1).6 Most of the dated Ayloffe firebacks bear between 13 and 17 shields and all but one are in excess of 1.2m (4 feet) in width and, therefore, cast for large fireplaces. Examples on public view are in Scotney Castle and in great Dixter, at Northiam, neither of which were noted by Cowper. The poundsford back is much smaller at 92cm (3 feet), with only seven shields and is, accordingly, more portable. it is dated 1629 and has the initials CT. In his will of 1694 Charles Tyler of Heathfield, Sussex, bequeathed Poundsford to his widow Mary to be a source for the payment of his debts.7 Charles was the son of Elias Tyler who, in turn, was the eldest surviving son of Charles Tyler, an iron RESEARCH NOTES – EARLY MODERN 314 founder. it is likely that Elias was also in the iron trade as a ‘Charles Tayler and his sonne’ are listed among the founders and fillers at the furnace of the Brenchley gunfounder, John Browne, at Horsmonden in 1628-9.8 The elder Charles Tyler had married Mary Allarde at goudhurst in 1599. Living initially in goudhurst where, recorded as Charles Stiller, he appeared in a recognizance in 1603,9 he seems to have moved to Cranbrook in the same year where his wife bore five children: Elias in 1606 and the last there in 1612. He was in Biddenden in 1615 where his daughter Elizabeth was baptised, but from 1618 the baptisms of two further sons and a daughter indicate that he was in Hawkhurst. Charles died there in 1629, his son Elias being his residuary legatee.10 Describing himself in his will as a founder, the possessions at his house and land called Tubbs Lake indicate the prosperity he achieved, with money and domestic items capable of providing bequests to his widow, his four sons and five daughters. Elias Tyler remained in Hawkhurst initially, acting as bondsman in the marriage of John Levett and Joan Burkham in 1634.11 By 1648 he was in Burwash where he was one of those who took the inventory of the possessions of Simon Coney.12 The following year he is recorded as occupying the manor of woodknowle, a property of 150 acres on the northern edge of the parish.13 predeceased by his son Charles, he died at Burwash in 1697. The large size, cumbersome nature and predominantly Kentish distribution of the dated firebacks – and there are many more small examples that are undated – suggested to Cowper that they had been the products of one of the furnaces operating in the county at that time, a conclusion with which this author does not disagree. These were Barden (Speldhurst), Bedgebury (Cranbrook), Biddenden, Cowden, frith (Hawkhurst), Hawkhurst, Horsmonden and Scarlets (Cowden).14 Of these Barden and the two furnaces at Cowden are distant from the main distribution area of the dated firebacks, which leaves furnaces in parishes where Charles Tyler was known to have been living or working. The potential for these firebacks to have been cast at a succession of furnaces, depending upon where Charles Tyler was working at the time, implies that he would have had possession of the Ayloffe shield and date/initial stamps rather than they being part of the stock of one particular ironworks, as would have probably been the case with most such moveable decorative stamps. It is the author’s contention that the Poundsford fireback was brought to the property by Charles Tyler’s descendants and that the initials CT on most of the dated Ayloffe firebacks are his.15 jeremy hodgkinson 1 H.S. Cowper, ‘A Series of Kentish Heraldic Firebacks and the Identification of the Arms,’ archaeologia Cantiana, 29 (1911), 40-6. 2 J. Hodgkinson, british Cast-iron Firebacks of the 16th to mid-18th Centuries (Crawley, HodgersBooks, 2010), 112. 3 J. Hodgkinson, ‘A Seventeenth-Century Sussex woodcarver: the Evidence of Cast ironwork’, regional Furniture, 28 (2014), 39-48. 4 C.f. Thrower, ‘Burwash’, Sussex archaeological Collections, 22 (1869), 113. 5 J. Starkie gardner, ‘iron Casting in the weald,’ archaeologia, 56, 1 (1898), 14; he erroneously identified the arms as those of the de la Warr family. RESEARCH NOTES – EARLY MODERN 315 6 The author is grateful to Mr and Mrs C.J. Mees for allowing him to photograph the fireback at poundsford. 7 East Sussex Record Office, Moulsecoomb (hereafter ESRO), PBT 1/1/42/196. 8 R.f. Monger (ed.), acts of the Privy Council of england: new series. Vol. 44 June 1628-april 1629 (London, HMSO, 1958), 71-2. 9 J.S. Cockburn (ed.), Calendar of assize records: Sussex indictments, elizabeth i (London, HMSO, 1975), no. 2084. 10 Kent History and Library Centre, Maidstone, PRC/1767/297. 11 J.M. Cowper (ed.), Canterbury marriage Licences, Second Series 1619-1660 (Canterbury, Cross and Jackman, 1894), 616. 12 ESRO, AMS 5744/147. 13 West Sussex Record Office, Chichester, Wiston/1325. 14 H. Cleere and D. Crossley, the iron industry of the Weald (Cardiff, Merton priory press, 1995), 309-67, 392-93. 15 The author acknowledges a debt to the late Brian Awty whose research into the families of wealden ironworkers, due to be published in his book adventure in iron, revealed the connection between the Tyler family and poundsford farm. 316 RESEARCH NOTES – MODERN SNODLAND AND ‘CEMENTOpOLiS’ 1841-1881 ‘Cementopolis’ is a word coined by victorian newspaper writers to describe the group of industrial workings which had grown up on the banks of the Medway, especially in the four parishes of Burham, Halling, Snodland and wouldham. Some of the largest lime and cement factories in the country were developed here in what had hitherto been a predominantly rural area and their activities caused some amazement for visitors expecting the delights of the ‘garden of England’. Encouraged and accompanied by travellers as enterprising as myself, i have recently employed a summer holiday in visiting a town known as Cementopolis, of which i had for years only heard dim, vague, and doubtful accounts. it is not unpleasantly situated in an amphitheatre of hills with a river branch for its base, filled with a flotilla of shipping. Its inhabitants are numerous and industrious, though what it is they are doing it is not easy for a stranger to discover. The courteous Alderman of the arrondissment i visited controls the Cementopolis and North Down Railway and chartered a special train for us, by which we were taken from the dockyards to the distant hills, passing serried rows of furnaces and mounds of coal; through realms of stacked wood and mighty masses of piled bricks, through tunnels, one more than half a mile long, over viaducts and under bridges, the latter giving a pleasant spice of excitement to the expedition, for if the traveller is too curious, or not careful enough in ducking, he runs some danger of being scalped. These perils surmounted, you arrive at immense excavations, which recall the quarries of Syene, and there you see scores of men, at various altitudes blasting (with gunpowder), picking, prising, and shovelling lumps of some cretaceous mineral, with which a long row of railway trucks is speedily filled. These lumps are carried away by a snorting Shetland-pony-like locomotive to a spot where the first of a series of gigantic and demoniacal machines takes charge of them and claws and scratches the lumps to pieces. The resulting mass is then mixed and macerated with a darker mineral, dug from a nearer spot, and twirled and drenched until it loses its pristine purity, the resulting compound being pumped to distant beds, where it rests for some days to settle. This peaceful period over, the water is poured off the stuff, which is toasted over plutonic fires in enormous kilns, the result being nodules as hard as iron. These again are taken to immense iron jaws and cracked and crunched and ground, and punched and stamped and triturated until they reach the stage of almost impalpable powder. Then it reaches something like rest and is stored into 2 cwt barrels. These barrels are also made in Cementopolis and, as pepys says, ‘it is mighty pretty’ to see the strips of timber brought from the spreading stacks outside into the sawdust-laden atmosphere of the noisy cooperage and turned into barrels in the twinkling of a bed-post to the tune of six hundred a day. Swishsh ! sweeshsh 317 RESEARCH NOTES – MODERN ! the heated staves are shaped. Rattle ! thud ! thump ! bang ! the staves are forced into form and held by iron rings. Whirr ! whirr ! the edges are bevelled. Krunch ! prrsh ! kerrishe ! prrsh ! the head is rounded. Tap ! tap ! the wooden hoops are on. Tank ! tank ! the iron ones follow, and then the final tub is sent rolling down the gangway to be branded and stored in capacious sheds. i haven’t space to tell you of all the resources of the place, the extent of its productions, or the ingenuity of its appliances. Suffice it to say that it is well worth a visit and of travelling some distance to see. if you want to go to this terra incognita you must take steamer to New Hythe or the coach to Burham, and when there seek the friendly aid of Mr. Butler, the energetic manager of the flourishing Burham Cement Company. He has something like a thousand men to look after, but, like all busy men, he can often find half an hour to devote to interested tourists who desire to explore this wonderful but almost unknown locality. (South eastern Gazette, 10 September 1889, by ‘verax’.) A former Rector of Snodland suggested that within the garden of England Snodland might be considered as ‘one of the tool-sheds’. The description would not have been valid before the mid-nineteenth century when Snodland was small, like its neighbours (a population of 300-400 in 50-60 houses) and dependent on farming for its livelihood. The years between about 1840 and 1880 were perhaps the most momentous in Snodland’s history. The village changed from relying on agriculture as the principal way of life to one in which the paper and cement industries came to dominate employment. it is true that both industries had gained a foothold here many years before, but then on a small scale and farming (as with so many other local communities) continued to provide the food, follow the seasons, and infiltrate the lives of all. Reporting on the opening of the Strood to Maidstone railway in 1856 the South-eastern Gazette noted ‘The next conspicuous object [after Larkfield church] is the extensive lime works at Burham […] from which a large proportion of the builders of the metropolis are supplied. This is a flourishing little industrial colony, creating and diffusing wealth, both by what they consume and what they produce’. Thomas Cubitt’s brick and lime works at Burham (in which many Snodland men worked) evoked much admiration from the writer of an article in the illustrated news of the World (8 October 1859): … on viewing the whole field, with its various and numerous engines, buildings, tramways, kilns, wharves, &c., one cannot but see that here are what may be justly termed the model brick-works. Here are concentrated the results of near half a century’s experience and improvements. Everything is in the right place. Nothing superfluous. Every possible attention has been given to economise labour and material, and every advantage taken of the natural position of the estate. when in full work, between 600 and 700 men and boys are employed, and from 25,000,000 to 30,000,000 of bricks, besides tiles and pipes, can readily be turned out from the works; which, however, can be considerably augmented without any great outlay, or increasing the present steam power. At Snodland and Halling the lime works of poynder and Hobson, one of the earliest in the district, began in 1819 and with a companion works at Northfleet again supplied materials for major building projects in London and elsewhere. This factory expanded after william Lee bought it in 1844 and he was one of three brothers also involved in major national building projects. poynder and Hobson’s first manager William Peters went on to create his own works at Wouldham in the 318 RESEARCH NOTES – MODERN 1850s, which was said to have become the largest of its kind. Again Snodland men were employed there and by 1880 some 600 men were said to cross the Medway to and fro daily in the small ferry boats of Snodland, Halling and New Hythe. Meanwhile Charles Townsend Hook took over the paper mill in 1855 and built it into a large and successful enterprise. These were years of social change too as the enterprising victorians set about exploring and regulating communal life. The area workhouse had replaced poor relief in individual parishes, schools were inspected, the conditions for children working in factories were investigated, the advent of the railways opened up business and travel for many, a plethora of newspapers spread news of all kinds, whether, local, national, or international. Societies were set up to assist parishioners in planning for hard times – the foresters, Shepherds, Odd fellows – and in seeking greater fulfilment in their lives - Gardeners’ Societies, evenings of lectures, readings and music, sport, and the creation of the working Men’s Club. This account is laid out as a series of interlocking essays, each focussing on a particular aspect, exploring Snodland’s development in the years 1841 to 1881, and naturally taking in links with Halling, Burham and wouldham. [these essays are published on the kaS website in the ebooks section.] Of course, local industry continued to expand at a similar rate between 1881 and the outbreak of the first world war, but its roots were founded in the forty years before that date. in particular in Snodland the leadership and philanthropy of Charles Townsend Hook (1832-1877) at the paper mill and william Lee (1801-1881) at the cement works gave so much to the local area, not only in employment, but also in enriching its social life. Both lived in the village and served as parish officers in several capacities, endeavouring to improve the lot not only of their workers and their families, but of the whole community. andrew ashbee 318 RESEARCH NOTES – MODERN THANET’S DEfENCES iN THE TwENTiETH CENTuRY in 2013 the trust for thanet archaeology drew attention to the value of undertaking a detailed historical and archaeological study in order to better understand the development of thanet’s defences during the 20th century. this idea was supported by the historic Defences Committee of the kaS and by the heritage Conservation Group of kCC which commissioned a review by the writer of the available sources. this yielded a wealth of historical information, including the location of several hundred defence sites. Clearly, ample potential exists for continuing and expanding documentary and field surveys, with important further discoveries likely. Historically Thanet, and Kent as a whole, were militarily significant parts of Britain not only because of their vulnerability to raid or an invasion from the Continent but also because of the presence of important surrounding sea lanes that had to be protected. Added to this, the skies over and around Kent became a countering zone for the home forces against enemy air attack. Defence was the sum of its land, sea and air elements. 319 RESEARCH NOTES – MODERN Geography and strategic significance Once an island separated from the mainland by a waterway later known as the wantsum Channel, Thanet is, in effect, a plateau of chalk. Looking out to the Thames Estuary and the English Channel, it is defined to seaward by cliffs, in whose larger gaps developed coastal towns and communities, of which the more prominent are Margate, Broadstairs and Ramsgate, all having a harbour or jetty. There are, besides, numerous narrow gaps along the line of cliffs. The terrain falls inland to the wantsum and Stour rivers which run through a broad band of marshes and levels on the line of the former wantsum Channel. These join the coast at the ends of the cliffs at Minnis Bay to the north-east and at pegwell Bay to the southeast, so marking the landward limits of Thanet as a geographical and geological entity. To the west on the mainland side, the land rises again. Off the coast are areas of sandbanks, the more prominent of these being (a) Margate Sands defining Margate Roads and (b) the Goodwins extending south off Ramsgate and pegwell Bay to offshore of Deal and walmer. The channel between the goodwins and the land provided the space for a large anchorage called The Downs, used especially during the great war. Thanet entered the 20th century with coastal communities served by a round-the island ‘loop’ road, joining with inland roads of varying quality connecting with towns and villages such as Minster, Monkton, Sarre and St Nicholas at wade, as well as with other hamlets. There were few roads across the marsh zones of the wantsum and Stour linking Thanet to the mainland. from junctions at faversham and Canterbury, railway lines crossed into Thanet at Hillborough and Sarre, curving round to connect Margate with Ramsgate. A second line between the latter two ports was later abandoned. A divergence of the Thanet loop from Minster ran south along the coast to Deal, Dover, folkestone and Hythe. Thanet’s roads and railways joined with a wider transport infrastructure in Kent and the South-East, ultimately connecting with London. This, in varying degrees, was of value both to an attacker for the purposes of providing routes to advance inland and to a defender for sending reinforcements to repel an enemy. thanet as a possible beachhead for invasion Dividing the two strategic waterways of the River Thames and the English Channel, Thanet is a prominent and exposed part of Kent. It has served as a gateway for invaders and settlers throughout early history including those of the Romans, Saxons and Danes. Although during the 20th century the whole of the eastern and south-eastern coasts of England were, in varying degrees, considered vulnerable to landings, Thanet was then, for the most part, judged to be a possible setting for smaller-scale disruptive injections of forces or raids, rather than as a place for a main invasion which was expected to occur elsewhere: french staff planning during the period of tensions in the later 19th century had envisaged, in the event of war, invading at points along the coastline from Deal to Hythe, with the capture of a port such as Dover or folkestone presaging an overland advance on London. german schemes of the 1890s saw a main landing in the Thames, perhaps at Sheerness and/or the coast to the north of this river, also with a march on the capital as the objective. Although a landing in Thanet and an advance through 320 RESEARCH NOTES – MODERN north Kent towards London was possible, the assumption by defence planners during the two world wars continued to be the island’s subsidiarity in the event of invasion. Nonetheless, as part of general anti-invasion precautions, the importance of that subsidiarity was amply sufficient for it to be given defences, in places on a significant scale, especially so during the Second World War. The coastline of Thanet is about 27km (17 miles) long and was officially believed to be only partly inaccessible to a significant descent. Pegwell Bay was however from time to time seen as having considerable potential requiring, in the event of war, dedicated defensive provision. North of this and breached by spaces occupied by the harbours at Ramsgate, Broadstairs and Margate, accessible parts of the cliffs for a sizeable landing force were considered to be few and were thought to have been defensible by small forces armed with field guns and small arms, an optimism questioned during the two world wars when heavier provision was made. west of Margate the gaps, small bays and stairs are more numerous but the shore is less favourable for a landing, except by small forces in shallow-draught boats at high tide and for which various gradations of defences were provided in the exigencies of war. Ramsgate had been designated a harbour of refuge for the Royal Navy before the start of the 20th century, of value to an attacker and a defender alike. Because of its smallness, Broadstairs was less attractive to an attacker. Likewise Margate but offshore there was the earlier-mentioned Margate Roads forming another useful anchorage. Despite being officially regarded as ‘indifferent’ harbours, Margate, Broadstairs and Ramsgate were nevertheless capable of being used by an invader in some degree and represented, in however limited a way, points of entry, being provided with defences, especially coastal artillery during the Second world war. Inland, and especially during the latter conflict, were defences and land forces intended to act as a brake on the progress of an invader across the island. in theory, the marshes on either side of the wantsum and Stour rivers could have been at least partially flooded to isolate the island from the mainland but the effects of this would not have been an insurmountable obstacle to an attacker. naval operations Limited german sea bombardment of Thanet and of targets in the surrounding waters during the great war was of a desultory nature rather than actions of measured and serious strategic intent. However, this led to the establishment of sea batteries at foreness and North foreland as well as the deployment of naval gunboats and heavygun monitors offshore. in both world wars the threats to Britain and the reality of mine and submarine attack were serious and Thanet was to have significance in relation to defensive naval operations, with Ramsgate harbour being used as a base for small boats to patrol coastal waters and to sweep mines. The activity of these vessels was, in varying degrees, integrated into the plans and operations of the naval commands at the Nore, as well as those at Dover and Harwich. Local Thames and Medway naval forces, reinforced where necessary by the main British fleet, were available to oppose landing forces, whether destined for Thanet or for the Thames area elsewhere. in the great war the waters east of Thanet were additionally important as an entry/exit to the Downs security anchorage, being defended on one side by a line of net mines. As well as this, the broader war-strategic value of Thanet was underscored by the creation of a port at Richborough in 1916 for the transit of 321 RESEARCH NOTES – MODERN supplies across the Channel on their way to the western front. During the Second world war as in the great war, the waters around Thanet were defensively mined. The maritime significance of Thanet could not easily have been fully predicted at the opening of the 20th century and arose from the circumstances of war. air defence The advent of aviation was to confer a special significance on Thanet whose bare, generally unobstructed plateau topography made it very suitable for air operations. its position was thus ideal for launching air patrols of the important waters of the southern extremity of the North Sea and the eastern part of the English Channel, as well as of the Thames Estuary including for anti-submarine operations. Not only that but it was a place for interception of raiders crossing its airspace to reach targets, both within the island and elsewhere beyond. Air bases were established during the great war at westgate and Manston, coordinating their operations with others elsewhere. Manston continued into the Second world war and into the Cold war. Although attacked from the air in the two world wars, with notable raids on Ramsgate and Manston, Thanet contained few war-industry assets attractive to bombing forces. A suite of specialist detection radars in and around Thanet during the Second world war and into the Cold war was integrated into air defence systems and defence against surface targets. The island was also part of a national ubiquity of civil or passive air defence having a considerable infrastructure, especially during the Second world war and to a lesser extent during the Cold war. Thanet’s role in military aviation and in ground-based air defence whether increasing, or diminishing, was an element of much wider regional and national strategic systems. indeed, the triad of home defence on Thanet was, in fair measure, guided by wider national planning imperatives which determined its origins, development and, in time, its operational demise. Future study proposals The varied historical findings are described in more detail elsewhere.1with a view to exploring the possibility of a longer-term and detailed local (Thanet) led investigation, a Thanet defences steering group was formed from among military and other historians of Thanet and beyond. in support of this committee Dr Mark Samuel, an architectural archaeologist and Emily greenaway, Heritage Engagement Officer of Thanet Council have framed a project proposal and a draft bid for Heritage Lottery funding. As an indication of one of the possibilities for the future, in June 2016 and under the leadership of Lara Band, the CiTiZAN archaeological initiative carried out training of local volunteers in recording Second world war anti-invasion defences at pegwell Bay. Continuing consideration is being given to a longer term project which could be supported by the KAS Historic Defences Committee and to the production of a consolidated version of the reports which were serialised in Casemate. victor smith 1 in Casemate, the journal of the international fortress Study group, vols 107, Sept 2016, 46-54; 108, Jan 2017, 20-24; May 2017, 10-17.