Archaeological Investigations at Middle Stoke on the Hoo Peninsula

ARCHAEOLOGICAL INVESTIGATIONS AT MIDDLE STOKE ON THE HOO PENINSULA

richard james

In July 1995, South Eastern Archaeological Services (a division of the Field Archaeology Unit, University College London) were commissioned by Kent County Council Highways to undertake an archaeological evaluation along the route of the proposed A228 Realignment in Middle Stoke (NGR TQ 828 755 to TQ 831 756). The purpose was to remove a dangerous bend and involved the construction of 1km of new road (Fig. 1). The evaluation revealed a number of prehistoric and later features (Greatorex 1995), which it was felt merited further investigation by area excavation. Consequently, Archaeology South-East (formerly known as SEAS) were commissioned in February 1998 to carry out this second stage of the work (James 1998). Additional to the excavation were a watching brief on the rest of the route and three trial trenches at its north-eastern end, located to investigate a possible former burial ground. Responsibility for the scheme had devolved to the new unitary authority of Rochester-upon-Medway City Council, and the archaeological work was monitored by their archaeological consultant, Dr Gerald Wait of Gifford and Partners.

The site is located on the south-eastern edge of the shallow ridge which forms the spine of the Hoo Peninsula, at an altitude of 10m od. The land slopes away to the south, overlooking the central reaches of the Medway estuary, with wide ranging views across to the opposite shore. To the east are reclaimed marshes linking Hoo with the Isle of Grain. The northward aspect is limited to the higher crest of the spinal ridge which runs from Allhallows to High Halstow. According to the British Geological Survey, the underlying bedrock is London Clay, with Terrace 2 sediment units to the east.

Until recently the evidence for later Bronze-Age settlement in Kent was meagre, a situation reflected in syntheses of the period (e.g. Champion 1980). Theories and trends regarding Bronze-Age activity in the area were mostly formulated using data derived from the large corpus of metalwork finds, and from inadequate environmental material. This data revealed a strongly coastal distribution, particularly around the Thames Estuary. This pattern is found from the early and mid Bronze Age, with settlement and artefactual distributions clustering around the littoral, forming one cultural zone with Essex (Buckley 1995, 7). However, the very fact of the coastal nature of the evidence is problematical, given the immense topographical changes endured by this stretch of coast since the Bronze Age (Champion 1980, 229; Milne 1995, 24)

Recent fieldwork, much of it the result of developer-funded projects, has transformed the picture (Yates 2004). Work in the Gravesend area revealed evidence for large-scale land division, associated with Deverel-Rimbury pottery (Mudd 1994). A larger grouping of sites discussed by Champion in the Thanet region of eastern Kent, including Minnis Bay, Mill Hill (Deal) and Highstead (Chislet), have been joined by further discoveries such as Netherhale Farm, a complex sequence of enclosures or field boundaries of Late Bronze-Age date (Macpherson-Grant 1992, 57). The picture emerging in the east of the county lends credence to Drewett’s prediction that a major redistribution centre may await discov-ery on Thanet (Drewett, Rudling and Gardiner 1988, 93). The pattern of discoveries suggests that the Thames estuary may well match the results of work in the Upper and Middle Thames Valley – agricultural intensification from the Middle Bronze Age onwards (Yates 2001).

The 1995 evaluation involved the machine excavation of nine 20m long trenches (Fig. 2, A-J), three of which contained archaeological features. Trench C contained a linear ditch feature and a small pit which produced small abraded sherds of probable Bronze-Age pottery. The ditch was cut by a small Romano-British pit, while a heavily disturbed circular feature produced medieval pottery (1150-1300). Trench F revealed another four Bronze-Age pits, and two intercutting features of medieval date (1150-1300 and 1400-1600) were recorded in Trench E.

The Excavation

An open area (Fig. 2) measuring 120m long by 10m wide (1,200m²) was stripped of topsoil and mixed overburden using a tracked 360°excavator fitted with a toothless ditching bucket. Machining was stopped when either archaeological deposits or clean subsoil were reached, at an average depth of 0.6m below ground surface. Six small baulks were left unexcavated as they contained chainage pegs. As features were noted in the freshly stripped subsoil they were cleaned with Dutch hoes, marked and mapped as soon as the site grid was surveyed in. The features were then hand-excavated systematically from east to west. Environmental samples were taken from six selected contexts. Flotation was found to be impractical due to the high clay content of the soil. Thus, all samples were wet-sieved using a 500 micron sieve.

The topsoil was a consistent 300mm in depth and comprised a friable grey-brown clay loam with some silt content. The soil was fairly stony in nature, containing a well-sorted collection of flint pebbles and nodules of assorted sizes, together with ploughed-in material of modern origin, notably straw. The subsoil was a compact mottled orange silty clay derived from mixed sand, gravel and silt deposits of fluvial origin. Parts of the trench were excavated to a greater depth (up to a maximum of 1.17m) and revealed underlying fluvial gravels. A geoarchaeological examination of the stratigraphy (Bates 1998) suggested a sequence of high energy fluvial gravel deposition in a periglacial braided river channel subsequently overlain by several medium and low energy sheet wash deposits.

The mottled, complex nature of the subsoil posed some problems when it came to excavating the features, particularly those of prehistoric date where the leaching out of organic content from the fills increased the similarity between fill and subsoil. Also, it was clear from the shallow nature of the features that a great deal of truncation had taken place, both from ploughing activities and associated soil erosion. The water table was reached at a height of 7.87m od which affected the excavation of some of the deeper features.

The Prehistoric Features (Fig. 3)

Two features of prehistoric date were observed cut into the subsoil. A linear patch of grey-brown silty clay (Context 14) was observed extending into the trench on a sw-ne axis. Unfortunately, the fill became progressively more difficult to recognize with depth, having been subjected to substantial leaching, and no clear edges could be identified (the cut in Fig. 3, Context 43, is an approximation). However, two sherds of later Bronze-Age pottery and a flint flake confirmed its identification as a feature. The feature may be associated with an east-west running ditch observed in Evaluation Trench C which also contained pottery of prehistoric date and unclear edges.

Further west was a small, shallow oval feature (Cut 44, Fill 29), similar in appearance to the other oval features scattered around the site (Figs. 3 & 4, Section 1 and detailed plan). However, excavation produced a large assemblage of later Bronze-Age pottery (weighing 7.65kg in total). Four similar features were located in Evaluation Trench F, although none produced pottery on anything like the same scale.

The Medieval Features (Fig. 3)

Fourteen features of medieval date were identified. Seven were linear in form (all aligned ne-sw), the remainder comprising five discrete features and two horizontal spreads. Pottery evidence suggested five phases between the late twelfth and mid sixteenth centuries, with a degree of overlap between the phases. For the purposes of this report the features have been arbitrarily grouped by century, a simplified arrangement which broadly follows the pottery phasing. The fills of all the features tended to be a consistent grey-brown silty clay, sometimes mottled with iron streaking, and often becoming difficult to distinguish from the underlying subsoil. Inclusions were not frequent, and consisted mostly of flint nodules.

a) Thirteenth century – the earliest medieval features were linear cuts 68 (Figs. 3 & 4, Section 2) and 46/61, both forming fairly substantial boundary features. The latter, 46 (Figs. 3 & 4, Section 3) and 61 appeared to be two lengths of ditch running along the same alignment, perhaps dug by separate work-gangs. The point at which the two lengths met coincided with a particularly complex natural arrangement of gravels, which hindered interpretation. The southernmost stretch (fill 26) produced a late twelfth-century coin (see coin report for details) and a large quantity of oyster shells (see oyster report), as well as 500g of animal bone.

b) Fourteenth century – eight features produced pottery of this date. Two closely spaced narrow gullies were located at the eastern end of the site (cuts 41 and 42, Figs. 3 & 4, Sections 4 and 5), the larger of which (41) contained two fills (12 and 40). No features were observed to the east of these gullies, either in the excavation trench or the subsequent watching brief. A much larger ditch feature (69, Figs. 3 & 4, Section 6) was recorded in the west of the trench, from which a simple copper alloy ring was recovered in addition to pottery and oyster shells. Within the limits set by these linears were several more discrete features. Cut 65 was a linear strip of grey fill (27) which merged quickly into the subsoil once excavation began. Equally frustrating was Context 13, an irregular mixed spread of sandy clay containing burnt material. Despite being very obvious on the surface, the deposit had no depth, merging with the underlying natural at a depth of only 100mm. Apart from commenting that some activity involving fire took place, little more can be said. The remainder of the features were less problematic, consisting of a 430mm deep pit (66, Figs. 3 & 4, Section 7) and a smaller post-hole (49). Cut 17 barely protruded from the northern baulk, but appeared to be little more than a shallow scoop.

c) Fifteenth century – a feature observed protruding from the southern baulk, cut 59, was interpreted as the butt end of a ditch. One fill (19) was observed, containing lenses of charcoal. One large animal bone was recovered. The depth of the feature (>700mm) resulted in water ingress after the water table was penetrated. This precluded a final depth being reached. Pottery of this period was also recovered from two spreads of material near the western end of the trench. The natural was seen to dive down here, forming a shallow valley, and had been partially filled by two sheets of hillwash material. Context 3 was the earliest, containing pottery dating from the mid fourteenth to early fifteenth centuries. This was overlain in the base of the ‘valley’ by a stony deposit (6) producing fifteenth- and early sixteenth-century pottery, oyster shell, animal bone and a residual flint scraper. A further butt-ended ditch (71, Fig. 3) contained early sixteenth-century pottery, but could not be fully excavated due to water ingress.

A number of other features were present on the site (Fig. 3). These produced no clear dating evidence and could be of any period. Details of these features are housed with the archive.

The watching brief and burial ground evaluation

The watching brief covered the remainder of the road corridor to the north-east of the main excavation area. No features or artefacts had been recovered from this area during the 1995 evaluation, and the 1998 watching brief proved equally blank. The subsoil was seen to change, becoming browner with a higher clay content, and several modern features were observed, including a water-main and two backfilled soil test-pits.

In addition to the watching brief, three small trial trenches, each measuring 5.0 x 1.5m, were excavated immediately to the east of the existing A228 at the point where it was joined by the new road corridor (Fig. 2, Trenches1-3). The purpose of the trenches was to investigate a ‘Burial Ground’ marked on the Ordnance Survey map which may have been clipped by the new road. The trenches were excavated to a depth of between 450mm and 650mm, but failed to produce any features or artefacts of archaeological significance. The conclusion reached was that the burial ground lay further to the north.

finds

Prehistoric Pottery by Nigel Macpherson-Grant

Two features produced prehistoric pottery: ditch fill (14) and pit fill (29); both contained later Bronze-Age Deverel-Rimbury style coarsely flint-tempered pottery:

Fill 14 produced two coarseware sherds (88gms), one from a large probable bucket-shaped jar; this sherd has unifacial wear suggesting a degree of exposure in semi-static ground conditions before final burial. Fill 29 contained the bulk of the pottery (268 sherds + small scraps: 7.65kgs), and includes some large-moderate sized sherds and plentiful worn scraps. Two-three large bucket-shaped storage jars are present. A small quantity of sherds in a less profusely flint-tempered fabric represent a jar with a shallow applied and horizontal (probable) shoulder cordon decorated with continuous finger-tip impressions. The majority of the sherds come from one or two large, fairly profusely tempered jars; rim scraps from one are decorated with deep continuous thumb/fingernail impressions across the rim top, bodysherds from the same, or another, vessel are decorated at shoulder or mid-girth position with a single horizontal row of continuous, fairly neatly applied, vertical fingertip impressions. Part of the base of this vessel is present, and there are some obviously conjoining lower body sherds. Large sherds from the latter vessel have a mixed wear pattern, some with internal unifacial wear, some with external wear; this point, together with the large number of scraps, suggests that the pit’s contents represent domestic rubbish that was allowed to weather to some degree before final burial, either before final deposition in Pit 44 or before the latter was finally infilled.

The available formal and decorational data indicates that the vessels represented are typical of regional later Bronze-Age coarsewares. The decorated rim scraps are from a vessel with simple upright rim and no obvious internal thickening, and are similar to a bucket-shaped cremation jar from Bridge (Macpherson-Grant 1980, fig. 26, 147 – but without the under-rim holes); the applied cordon from another vessel is similar to the same reference which has been given a revised dating of c.1246-1066 Cal bc (Macpherson-Grant 1992, 62). The decorated bodysherds are similar to a large jar from Netherhale Farm, Thanet which has a likely association with the nearby hoard-dated Birchington bowl (c.1300-1100 bc; op. cit. 62-3) and closer still to burial jars from the Monkton Area 7 barrow cemetery (Macpherson-Grant 1995, 64); the thickened rims of the latter and the Netherhale Farm example is broadly similar to the C14 associated jar rim from Coldharbour Lane, Gravesend, with a date of 1229-989 cal bc (Barclay 1994, 387-9, fig. 9, 6). There are clear, subtle stylistic differences between the Bridge assemblage and the Netherhale Farm/Monkton material, which could be due to either chronological or regional variations; though the present data is slim, there is enough to initially suggest that the differences are more likely due to chronological position or, since there appears to be a degree of stylistic meld between the assemblages mentioned, that all variations ran concurrent to some extent. The relatively low regional quota of assemblages for this period inhibits definition of the likely chronological or social differences represented by these traits; for the time being a broad c.1500/1400-1000 bc date can be safely applied to the present material.

Medieval Pottery by Luke Barber

The assemblage of medieval pottery from the site is small (89 sherds, 1.102kg), and there are no large groups present from individual contexts. The material spans the thirteenth to sixteenth centuries and consists of local wares in sand, sand and shell, and highly-fired transitional sandy earthenware fabrics. With the exception of the more durable transitional wares, the sherds are generally small and show signs of moderate to heavy abrasion. The largest individual group consists of a mere 10 sherds (context 26). Most of the sherds from the site are from cooking pots, although a few glazed jug sherds are also present. These include white-slipped, green-glazed sherds of Mill Green type ware. (Fuller details are included in the Archive Report.)

Worked Flint by Greg Priestley-Bell

A total of 14 pieces of worked flint were recovered from eight contexts. The single waste flake from Context 14 was associated with later Bronze-Age pottery, the remainder of the assemblage however is likely to be residual. The raw material is very varied and ranges from a light grey chert with many inclusions to a very dark grey fine-grained flint with few inclusions. The presence of pebble cortex and thermal fracture surfaces within the assemblage suggests that the raw material was probably collected from the surface.

A total of 10 smallish waste flakes (<40mm maximum dimension) of widely varying thickness and outline were recorded from seven contexts. The three largest globular waste flakes were heavily burnt and may represent worked out cores that had subsequently been used as ‘pot-boilers’.

A broken cortical flake from Context 27 showed a short section of direct retouch on the right lateral edge. A flake from Context 67 showed distal direct right short retouch and distal inverse left long retouch, thereby forming a point that may have served as a borer. An elongate (65 x 23mm) borer from Context 6 showed continuous direct abrupt retouch (blunting) along the entire length of its left edge with a slight notch at the distal end and a short section of inverse right abrupt retouch. Direct abrupt retouch was angled across the distal end to form a borer point above the notch. A small patinated apparently pressure flake scar on the ventral surface of the distal tip suggests that the point of the borer had perhaps broken in antiquity during use. Possible plough damage along the right edge may have obscured a further area of retouch. A borer from Context 39 showed two sections of distal direct retouch, abrupt right distal and semi-abrupt left distal, converging to form a tapered borer point.

With the exception of the borer from Context 6, very little attempt seems to have been made to control the form of the flakes produced, while the platforms, where present, were all cortical or flat. There is no evidence for the use of a soft-hammer percussor and only two hard-hammer flakes could be positively identified. The very high degree of use or sediment polish uniquely present on the Context 6 borer, together with its atypical character within the assemblage, perhaps suggests that it was a curated tool brought onto the site and discarded after breakage. Although the range of artefacts is too narrow to be diagnostic, the remainder of the assemblage seems to represent certain stages of a flaking industry producing blanks for tools. The general nature of the material suggests a Neolithic or Bronze-Age date.

Animal Bone by Lucy Sibun

The animal bone totalled 67 fragments, all from medieval contexts dating from the thirteenth to sixteenth centuries. The small size of the assemblage prevents any conclusions being reached with regards to the economy or subsistence of the site, but a few observations can be made. The assemblage consists primarily of cattle and sheep, but a few fragments of pig are also present. Most skeletal elements are represented, including both meat-yielding joints and the skeletal extremities, usually discarded during the butchery process. However, only two fragments displayed evidence for butchery. Also of note is the occurrence of juvenile individuals, both sheep and cattle, further suggesting their possible use as a meat resource.

Marine Molluscs by David Dunkin

The excavation produced three contexts in which marine molluscs were present. Only one species was represented – Ostrea edulis, Common Oyster.

The oyster remains summarised in Table 1 were retrieved from contexts of medieval date. The complete valves were reasonably well preserved with evidence of some erosion around the shell margin of most of the individuals collected. Sometimes notches may be seen as part of the procedure for opening the bivalve for food purposes, but the eroded margins here would have precluded this observation. Their general state of preservation, however, suggests that the contexts in which they were found have remained relatively undisturbed and stable since their deposition.

Oysters usually live to an age of ten to twelve years, and reach maturity at four years, which is the minimum age they are usually collected for consumption. Out of the entire assemblage, only three complete valves represented individuals less than 4 years old (2 x left; 1 x right valves from context 22). In context 26, for example, 27% of the left or lower valves were of ages in excess of 10 years. Therefore, the vast majority (96%) of the complete oyster valves recovered from the three contexts could all have been utilised as a food resource.

In formal food preparation, it is usually the left, or lower concave valve which is served. This means that in terms of rubbish disposal, for example in midden deposits, there may be some patterning in the occurrence of upper and lower valves. Although the quantity of oyster shell recovered is relatively small overall (MNI represented : 51), the bias in favour of the left valves (65%) adds weight to the suggestion that the oyster was being collected as a supplementary food resource. It is worth mentioning, however, that oyster shell was used during the medieval period for other purposes, such as a filler for mortar.

The shells displayed minimal distortion, suggesting they were collected from beds in which overcrowding was not especially prevalent. Also, very little infestation of the shells was apparent, with evidence of polychaete worm attack in just one individual (Polydora ciliata in the upper valve of a mature specimen from context 26). Furthermore, the presence of boreholes, caused by gastropods, in only three shells in the assemblage suggests that the molluscs were collected from a healthy colony in an area of high salinity. The overall quantities retrieved suggests that these oysters were not obtained from a farmed context. It is probable that the oysters derive locally from the Medway/Thames estuaries.

The Coin by David Rudling

Henry II or Richard I. Short-Cross Coinage: cut quarter silver penny (i.e. farthing). Class 1b-2, issued 1180-c.1190.

Obverse: HEN[RICVS REX]

Reverse: ]VL.O[

The moneyer could be RANDVL of London, Exeter or Northampton (Wren 1992, 27-32).

Environmental Samples by Luke Barber

Six contexts were sampled. They did not produce large quantities of animal or plant remains. Charcoal was present in very small quantities in all the samples, with the exception of Sample 3 (Context 19 – late medieval/early post-medieval) which contained a larger quantity. Carbonised seeds were recovered from three samples. All were of cereals and were poorly preserved, exhibiting evidence of heavy abrasion and fragmentation.

discussion

The Bronze-Age features found during the excavation are interesting, if minimal. They, together with the results of work carried out by Archaeology South-East elsewhere on the peninsula (at Allhallows and Kingsnorth Power Station), provide the first real evidence for the survival of later Bronze-Age landscape elements on the Hoo Peninsula, a pattern hitherto only found in the western and eastern coastal areas. The linear feature (43) appears to be the terminal of a boundary ditch of some sort, and may be compared with a series of superimposed ditches of similar date recently observed at Malmaynes Hall Farm 1.2km to the west (James 1999).These features hint at the possibility that a more extensive arrangement of field systems and settlements may survive around the Medway estuary. Yates has suggested recently that such a spread of farming and settlement evidence fits into a larger pattern emerging around the Thames estuary (Yates 2001). The second feature (44) is a small, probably truncated pit, and, together with the contemporary pits recorded during the evaluation, suggests that further domestic evidence may be present.

The location is ideal for prehistoric agriculture, with highly fertile soil and good access to estuarine resources. The extent to which marshland was available for exploitation is debatable, given that the present marshes, so extensively utilised in the medieval period (Everitt 1986, 34-8), are thought to have only developed since the Roman period (Devoy 1980). However, the identification at Coldharbour Lane of ‘droveways’ similar to those at Fengate, channelling livestock down into the marshland pastures (Mudd 1994, 408), and a similar situation pertaining at Mucking in Essex (Parker-Pearson 1993, 121), suggests that an allied arrangement may well have been in operation on Hoo.

The medieval linear features represent several centuries’ worth of agricultural activity. The Hoo Peninsula forms a very fertile area for arable farming, with rich marsh pastures on the coastal periphery. The less onerous social bonds prevalent in medieval Kent (in comparison with the stratified social systems found in the Midlands, for example) created the conditions within which a prosperous agricultural system could flourish (Brandon and Short 1990, 57). The various ditches and pits relate to such a system, with a smaller network of fields being worked compared with the large modern fields visible today.

acknowledgements

SEAS/ASE and the author would like to thank Nigel Newton (Rochester-upon-Medway Council), Dr Gerry Wait (Gifford and Partners) and Lis Dyson (Kent County Council) for their help with this project. Thanks also to the excavation team and to Justin Russell for the illustrations. Special thanks to David Yates for discussing elements of his doctoral research.

bibliography

A. Barclay, 1994, ‘Prehistoric Pottery’, in Mudd 1994.

M. Bates, 1998, ‘Field Notes on a Visit to the Site at Stoke Bends, Hoo Peninsula, North Kent’, in James 1998.

P. Brandon and B. Short, 1990, The South-East from AD 1000.

D. Buckley, 1995, ‘Thames Gateway: Archaeological and Geographical Context’, in RCHME, Thames Gateway: Recording Historic Buildings and Landscapes on the Thames Estuary.

T. Champion, 1980, ‘Settlement and Environment in Later Bronze Age Kent’, in J. Barrett and R. Bradley (eds), Settlement and Society in the British Later Bronze Age, BAR 83.

C. Greatorex, 1995, An Archaeological Evaluation along the Route of the Proposed A228 Stoke Bends Realignment, Middle Stoke, Kent (unpubl. SEAS Report 1994/273, July 1995).

R.J. Devoy, 1980, ‘Post-Glacial Environmental Change and Man in the Thames Estuary: a Synopsis’, in F.H. Thompson (ed.), Archaeology and Coastal Change (Soc. of Antiq. Occas. Paper (New Series) I).

P. Drewett, D. Rudling and M. Gardiner, 1988, The South-East to AD 1000.

A. Everitt, 1986, Continuity and Colonization: The Evolution of Kentish Settlement.

R. James, 1998, A Post-Excavation Assessment on the Archaeological Investigations on the A228 Stoke Bends Improvement, Middle Stoke, Kent (unpubl. ASE Report 853, April 1998).

R. James, 1999, An Archaeological Evaluation in Malmaynes Hall Farm, Stoke, Kent (unpubl. ASE Report 1001, March 1999).

N. Macpherson-Grant, 1980, ‘Archaeological Work along the A2: 1966-1974’, Archaeologia Cantiana, xcvi, 133-183.

N. Macpherson-Grant, 1992, ‘A Review of Late Bronze Age Pottery from East Kent’, Canterbury’s Archaeology 1991-1992 (CAT), 55-63.

N. Macpherson-Grant, 1995, ‘Earlier prehistoric ceramics from the Monkton A253 project’, Canterbury’s Archaeology 1994-5 (CAT), 62-66.

G. Milne, 1995, ‘Foreshore Archaeology’, in RCHME, Thames Gateway: Recording Historic Buildings and Landscapes on the Thames Estuary.

A. Mudd, 1994, ‘The Excavation of a Later Bronze Age site at Coldharbour Road, Gravesend’, Archaeologia Cantiana, cxiv, 363-410.

M. Parker-Pearson, 1993, Bronze Age Britain.

C.R. Wren, 1992, The Short-Cross Coinage 1180-1247, Henry II to Henry III; an Illustrated Guide to Identification.

D. Yates, 2001, ‘Bronze Age agricultural intensification in the Thames Valley and Estuary’, in J. Brück (ed.), Bronze Age Landscapes: Tradition and Transformation, 65-82.

D. Yates, 2004, ‘Kent in the Bronze Age: Land, Power and Prestige c.1500-c.700 BC’, in T. Lawson and D. Killingray (eds), An Historical Atlas of Kent, 13-15.

Fig. 1 Site Location.

Fig. 2 Trench Lay-out (Evaluation and Excavation).

Fig. 3 Plan of Excavation Trench.

Fig. 4 Sections 1-7 and plan of Cut 44.

Table 1: Quantification and Identification of Marine Molluscs

Cxt No.	Species	Context Type	Quantity
6	Ostrea edulis	Hillwash deposit	1 x complete lower valve (left)
22 26	Ostrea edulis Ostrea edulis	Fill of ditch Fill of ditch	9 x complete lower valves (left) 5 x complete upper valves (right) 26 x fragments 41 x complete lower valves (left) 22 x complete upper valves (right)
			87 x fragments