A Bronze Age ring-ditch, Anglo-Saxon settlement and medieval enclosures at Otterham Quay Lane, Rainham

A Middle Palaeolithic Levallois flake was the earliest find. A small ring-ditch, cremation burial and trapezoidal enclosure belong to the Middle/Late Bronze Age, and a few pits have been assigned a Late Iron Age/Romano-British date. At least one Saxon sunken-featured building was identified, possibly related to burials recorded earlier to the north. The principal discovery was a sequence of medieval enclosures, their main period of development spanning the late eleventh to mid-thirteenth centuries. In the corner of one was a sunken bakery/ kitchen, a type of structure which appears largely confined to Kent. Finds were relatively sparse, but the enclosures can probably be linked to the expansion of pastoral agriculture following the clearance of woodland in this area between the Weald and the Thames.

The specialist reports accompanying the article can be found on the KAS website. However, the pottery report and a discussion of the bakery/kitchen structure are included below.

Wessex Archaeology was commissioned by RPS group to undertake a programme of archaeological investigations in advance of residential development at Otterham Quay Lane, Rainham (centred on NGR 582871 166009; Fig. 1). The investigations comprised a desk-based assessment of the 10.8ha site (CgMs 2015), trial trench evaluation (Wessex Archaeology 2019a), test-pitting to establish Palaeolithic potential (Wessex Archaeology 2019b) and subsequent excavation of a 1.72ha area (Wessex Archaeology 2019c), the latter undertaken between May and July 2019.

[fg]png|Fig. 1 Site location plan.|Image[/fg]

Site location, topography and geology

The site is located to the east of Otterham Quay Lane, on the eastern edge of Rainham, and Otterham creek lies approximately 750m to the north. The site is bounded by the London, Chatham and Dover railway (opened in 1858) to the south, a former landfill site to the east and an industrial park to the north. Existing ground levels are recorded at around 30m AOD in the southern field (the location of the excavation) and 22m AOD in the northern field (previously quarried for brickearth; see below), both fields being relatively flat but with a distinct step between them across the centre of the site.[pg111][pg112]

The underlying geology is mapped as Thanet Formation Sand, Silt and Clay, with no superficial deposits recorded (British Geological Survey online viewer). The adjacent field to the east has an outcrop of Lewes Nodular Chalk formation with superficial deposits of Head Clay and Silt.

Sinkhole formation is known to be active in this area. Consequently, the deepest Head/Brickearth sequences may reflect more extensive accumulations of material in topographic low points caused by solution features within the chalk that is known to underlie the Quaternary deposits and Thanet Sands. A possible sinkhole was identified in the field to the east of the site during a survey in the 1970s, when it was still a disused quarry.

Archaeological and historical background

The desk-based assessment provided only moderate indications for medieval and earlier activity in the vicinity of the site (CgMs 2015). The earliest record is that of a Palaeolithic flint tool, found in 1862 approximately 550m north of the site near Otterham Quay, with three handaxes subsequently recorded from the same area (Wymer 1999).

No Neolithic or Bronze Age finds are listed but in 1934, during continuing quarrying work, the remains of an isolated pit or possible hut were recorded in the northern field adjacent to the western site boundary. The feature contained Iron Age pottery, fired clay, evidence of in situ burning and a decorated blue glass bead.

Rainham lies on Watling Street linking London to Canterbury, which runs approximately 300m south of the site. Despite the presence of the Roman road, the Historic Environment Record (HER) does not contain any records relating to possible roadside activity in the vicinity. However, quarrying during the early 20th century disturbed the remains of a Romano-British cemetery located 600m north of the site, the pottery indicating that the burials dated to the first century AD.

Rainham is an early Anglo-Saxon place name (originally ending as ‘-ingaham’). Evidence for a migration period inhumation cemetery has been identified south of Otterham Quay and to the north of the site, with a range of pottery, weapons and other grave goods recovered in the 1930s.

The earliest documentary evidence for Rainham is a charter of AD 811 when it is recorded as a royal estate centre, but by the time of Domesday (1086) Rainham had been absorbed into the large royal manor of Milton and, therefore, does not have a separate entry. Later documentary evidence suggests that St Margaret’s church was in existence by 1100, while the earliest fabric in the existing building is thirteenth-century. The church is likely to have been the focus of settlement, with the site lying beyond this to the east.

During the post-medieval period the area of the site remained as agricultural land, as shown on Andrews, Dury and Herbert’s map of 1769, while a 1797 Ordnance Survey drawing indicates the site subdivided into multiple agricultural plots. The 1840 Rainham tithe map shows that there had been a degree of boundary consolidation since 1797, with the northern plot recorded as arable land and the southern plots divided between vineyards and orchards. Otterham Quay, approximately 1km to the north of the site, first appears on a chart of the Medway from 1835, the quay and wharves marked on all the subsequent Ordnance Survey maps.

[pg113]The earliest evidence for brickearth quarrying within the site, in the northern half, is indicated on the 1866 OS map, and by 1908 this is shown to have increased in extent. By 1933 quarrying covered the northern third of the site and then, prior to 1938, had extended southwards towards the central field boundary. In subsequent decades the quarried area was backfilled, orchards across the southern field removed and the site returned to arable use.

Previous archaeological work

Twenty-seven evaluation trenches undertaken in 2019 indicated a focus of activity in the south-east and south-central parts of the site (WA 2019a). It also confirmed that the majority of the northern half had been subject to previous quarrying for brickearth and any archaeological deposits probably removed. A Middle Palaeolithic Levallois flake was recovered from subsoil during the evaluation, and the following Palaeolithic test-pit investigation (through four machine-dug pits) established that Quaternary deposits were preserved within the southern half of the site.

The evaluation also revealed a range of archaeological features including a Middle/Late Bronze Age urned cremation burial, along with pits, ditches and at least one probable sunken featured building, most of the features uncertainly dated but with small quantities of Romano-British (five sherds), Saxon (one sherd) and medieval (24 sherds) pottery recovered from some.

Excavation sequence

Based on the evaluation results, a roughly sub-rectangular area measuring 135m by 80-105m was stripped and excavated. This revealed a Bronze Age ring-ditch and small enclosure, a scatter of late Iron Age/Romano-British pits, possibly two Anglo-Saxon sunken-featured buildings, and what appears to be virtually all of an entire sequence of medieval enclosures. It also exposed several natural sinkholes.

Middle/Late Bronze Age

Ring-ditch 4020 (Fig. 2) is likely to represent the remains of a Middle Bronze Age barrow. The ditch had an external diameter of 9m and an internal diameter of 6.8m, with an average width of 1.1m, and was approximately 0.25m deep. Seventeen small sherds (24g) of later Bronze Age pottery were recovered from the single, undifferentiated fill of the ditch, which was fully excavated. Environmental samples taken from the ditch contained nothing of significance. No features were identified within the area enclosed by the ring-ditch except for a small, shallow tree-throw hole on the north edge.

[fg]png|Fig. 2 Middle–Late Bronze Age, Romano-British and Saxon features.|Image[/fg]

Approximately 3.5m to the north of ring-ditch 4020 was a single, heavily truncated cremation grave. The small, circular grave cut [1804] 0.14m in diameter and no more than 0.06m deep, contained cremation urn 1806, with only the base of the vessel and part of the cremated material within it surviving. The urn has been ascribed a Middle/Late Bronze Age date, the cremated bone (233.5g) representing the remains of a mature adult (25-40 years), most likely a female. A radiocarbon [pg114]date of 1050-910 cal. BC (SUERC-103752) obtained places the burial, and by association the urn, in the Late Bronze Age.

Trapezoidal enclosure 4554 (Fig. 2) was located on the western edge of the site. The enclosure ditch on the western side was 25m long, aligned north–south, with the enclosure measuring 20m east-west. The ditch was generally 0.8-1m wide and only 0.25m deep, though the two terminals were up to 0.4m deep; the ditch fills produced just seven sherds of Middle/Late Bronze Age pottery, but one of the terminals contained the largest collection of worked flint from the site (a biconical flake core, 13 flakes and a blade). There was a large opening in the south-east [pg115]corner of the enclosure, or perhaps two if ditch 4292, 16m long and of similar size, formed a continuation to the east of the southern side of the enclosure (undated ditch 4293 on the same alignment may have extended the length of this boundary to 50m). Such an arrangement would have created a 12m-wide south-facing entrance and an 8m wide east-facing entrance. Apart from a small, undated pit, the only feature identified within the enclosure was Saxon sunken-featured building (SFB) 4430 (see below; Fig. 2).

Towards the southern end of the site, undated ditch [4203] lay on a diverging alignment to the south side of enclosure 4554 and ditches 4292 and 4293, the gap between them increasing to 65m at the eastern termini (Fig. 2). Cut by a medieval ditch, 4203 may have been contemporary with these, the two sets of ditches forming a funnel-shaped arrangement.

Two pits have also been assigned a Middle/Late Bronze Age date (Fig. 2). Large, sub-oval pit 4302 to the north-east of enclosure 4554 was 6.8m long, 4.5m wide and the depth exceeded 1.2m. In addition to small quantities of burnt flint, 67 sherds of late Bronze Age pottery, probably from a single vessel, was recovered from its fills. The pit had a broad, stepped edge leading to a much steeper central area and may have been a waterhole, a quarry pit or possibly a further sinkhole. Apparently isolated to the east was pit 4384, 1m long, 0.45m wide and only 0.1m deep, which contained 19, mostly tiny later Bronze Age sherds.

Iron Age/Romano-British

Just three sherds of pottery have been tentatively identified as Iron Age, with a further 29 only attributed a general prehistoric date. In addition, 19 abraded sherds are of undiagnostic Roman date. Based on this slender evidence, at least two of the excavated pits, towards the northern-eastern corner of the site, may belong to this broad period, as no later material was present (Fig. 2). However, none of the pits contained more than four sherds and at least some of this pottery may have been residual.

These pits comprised two small oval examples (4008 and 4017) measuring 0.9- 1.75m long, 0.7-1.3m wide and 0.15-1.15m deep. A further oval pit, 706 (1.8m by 1.5m and 1m deep), was noteworthy for its distinctive bell-shaped profile, this cut by shallow pit 714.

In addition to pottery, a limited range of other finds were present in generally small quantities, including animal bone, fired clay, Romano-British CBM and a fragment of lava quern. Most of this material came from pits 706 and 4021 (as residual finds in this feature).

Saxon

One sunken-featured building (SFB) and a single pit are almost certainly of this date, while one other possible SFB and at least one further pit have been tentatively assigned to this period (Fig. 2).

Sunken-featured building 4430 lay within the Bronze Age trapezoidal enclosure [4554], though whether traces of this still survived as an earthwork is questionable and the location may be coincidental. SFB 4430 was rectangular in plan and [pg116]orientated east-west. It measured 4.2m long and 3.4m wide, with a depth of 0.2m (see Fig. 5). The sides at either end were relatively steep and the base flat. Two postholes lay at the eastern end, 4445 centrally placed at the edge of the pit. This posthole was 0.26m in diameter and 0.4m deep with vertical sides and flat base, the post-pipe still visible in section and stones used as packing material. The second posthole, 4433, was located 0.3m to the north, cut into the slope of the pit; this posthole was of similar diameter to 4445 but only 0.1m deep, perhaps inserted later to provide additional roof support. No postholes were present at the west end of the building. The fill of SFB 4430 contained 15 sherds of organic-tempered pottery, possibly from a single vessel, as well as some residual Romano-British CBM.

Several features lay within SFB 4430, most notably sub-rectangular pit 4431 towards the eastern end on a north-south alignment, this measuring 2.1m long 1.4m wide and 0.12m deep with steep sides and a flat base. The function of this feature is unknown, but it could have been a shallow storage pit, perhaps below a raised, planked floor. Sub-oval pit 4440, 1m long, 0.7m wide and 0.3m deep, was located centrally along the southern edge of the SFB; this contained a further four organic-tempered sherds. Finally, towards the eastern end, was a small, oval feature, 4438, possibly a pit, 0.8m long, 0.5m wide and 0.12m deep.

The interpretation of feature 4558 as a Saxon SFB is somewhat more equivocal, and it may have been a medieval feature. This lay 65m west of SFB 4430, within the centre of the medieval enclosures and cut by several ditches. It consisted of a shallow, rectangular cut, broadly aligned north-south, measuring approximately 7m long, 4m wide and 0.12m deep, with a flat base. No features were identified within the cut, nor any associated postholes around the edges. The only finds from the single fill, which was only partly excavated, comprised animal bone and very sparse quantities of fired clay, slag, oyster shell and burnt flint.

Shallow sub-circular pit 4021 towards the north-east corner of the site was sub-oval in plan, 2.45m long, 1.9m wide but only 0.18m deep. The single fill contained all five of the chalk-tempered Saxon sherds recovered from the site, animal bone and fired clay.

Pit 4081 lay 20m to the west of pit 4021, closer to possible SFB 4558. It was oval in plan, 1.65m long, 1.05m wide and 0.8m deep. The fill included animal bone along with smaller quantities of fired clay, CBM, a small lump of unidentifiable iron and three joining fragments of a loom-weight of likely Mid- to Late Saxon date.

Medieval

Certain or probable medieval ditches, some relatively small and shallow, along with various pits, postholes and a sunken-featured bakery/kitchen structure, make up the majority of datable features identified across the site. The ceramic evidence has provided a date range for these from the eleventh to thirteenth centuries, with a few sherds of fourteenth/fifteenth date. However, there are only 211 sherds of medieval pottery in total, some of which may be residual in the contexts in which they were found, and several elements of the system have no artefactual dating. For the latter, assigning these to the medieval period has been done on stratigraphic grounds, in the few clear-cut cases where relationships exist, or on their layout and alignments.

[pg117]Four phases of activity have been identified (Figs 3 and 4), reflecting the development of the enclosure system, which has been reworked and altered over time. uncertainty remains regarding the phasing of some features, particularly the pits, and other sequences are possible. However, overall, what is presented below is considered a plausible solution to presenting a coherent narrative.

[fg]png|Fig. 3 Medieval features, phases 1 and 2.|Image[/fg]

[fg]png|Fig. 4 Medieval features, phases 3 and 4.|Image[/fg]

Phase 1: the earliest phase of medieval activity appears to be the digging of ditches 4200 and 4206 which defined the eastern and part of the southern sides of a sub-rectangular enclosure measuring approximately 45m by 25m in the centre [pg118]of the site (Fig. 3). Ditch 4199 probably marked the eastern part of the north side, but much of the remainder of the enclosure has probably been removed by later phases of ditches (e.g. Phase 2 ditch 4202). The main axis of this enclosure was WNW-ESE, and this was followed during all subsequent phases. Ditches 4199, 4200 and 4206, along with others assigned to this phase (and also Phase 2) were characterised, in particular, by their relatively small size, with widths rarely exceeding 0.8m and depths more than 0.5m, but often with dimensions of less than half these. On the west side of the central enclosure and extending to the north was [pg119]ditch 4289. This was 55m in length, cut by pit 4402 (but not extending beyond it) at its northern end, and with a slight return to the east at the southern end where it terminated on the same alignment as ditch 4206 (see above). On the inner east side of ditch 4289, and broadly parallel to it, was ditch 4205, 22.5m long, of similar dimensions, with an average width of 0.5m and depth of 0.16m. No corresponding ditch to 4289 was identified on the east side, with 4026 considered likely to be a part of the Phase 2 arrangement. Finally, ditch 4557 on the north side, at least 7.6m long and orientated north-west to south-east, may be amongst the earliest medieval features, cut by ditch 5556 which formed a rectilinear arrangement ascribed to Phase 2. Extending south-south-west from the corner of the enclosure formed by ditches 4200 and 4206 was L-shaped ditch 4208. This continued for 40m before turning south-south-east for a further 25m and was cut by several later ditches.

Phase 2: the second phase essentially comprised a replacement of the Phase 1 layout, perhaps because many of the ditches had silted up, with no evidence for any significant expansion. However, the eastern half of the central enclosure appears to have been retained, with ditch 4202 replacing a predecessor on the western side (Fig. 3). To the north-west, there is some uncertainty about sequence of ditches, but it is suggested that 4291 is the most likely to belong to this phase. This was 35m long, with a kink towards the north end where it terminated; the southern part continued towards the north-west corner of the central enclosure, the precise arrangement lost to a sinkhole.

On the west side of the central enclosure was ditch 4026, at least 30m long, which continued to the north beyond the limit of excavation. The broad alignment was extended to the south, beyond a small gap, by ditch 4023. The gap, only 1m wide, was likely an entrance, flanked to the south by a 5m-long return to the west of ditch 4023. This ditch was approximately 60m long in total, turning WNW at the southern end where it continued, broadly paralleling the layout of earlier ditches 4200 and 4206 of the inner enclosure. No clearly corresponding arrangement of ditches to the west of 4023 was identified, although ditch 4204 assigned to Phase 3 may have originated at this time.

L-shaped ditch 4207, 40m in length, lay to the south of ditches 4024/4200 and reflects their layout. However, it has only been ascribed to Phase 3 by default, cutting phase 1 ditch 4208 and cut by Phase 4 ditch 4024; it could belong to Phase 3.

Finally, there is a group of broadly associated and similarly aligned ditches immediately north of the central enclosure whose phasing is uncertain, but which may fit best within Phase 2. Ditch 4198 lay parallel to and immediately north of Phase 1 ditch 4199, and was approximately 20m long, 0.7m wide and 0.2m deep. Amongst the few finds recovered was a small, iron, probable pruning- or weed-hook. At the west end, ditch 4198 turned north and joined ditch 4555 which was on a slightly diverging alignment and cut possible Saxon SFB 4558. Ditch 4555 in turn lay in an uncertain but probably broadly contemporary relationship with L-shaped ditch 4556, together forming an approximately rectilinear arrangement. These ditches were generally 0.6-0.75m wide and 0.2-0.45m deep, with nothing to indicate that they had a structural function. It can be noted, however, that there was a cluster of pits in this area, several of which cut the ditches here (see below).

Phase 3: saw a consolidation of the Phase 2 arrangement, as well as extending the layout to the south, with possibly some of the earlier ditches in the central area [pg120]remaining in use (Fig. 4). The new ditches, as in previous phases, were small, with widths generally of 0.8-0.9m, though in places up to 1.2m, and depths mostly between 0.2-0.4m.

Within the central area, L-shaped ditches 4025 and 4201 defined the east, south and west sides of the core enclosure, this measuring approximately 55m east-west by at least 40m north-south. A 4m-wide gap between the ditch terminals on the south side marked an entrance, this probably replicating an earlier arrangement.

On the west side, slightly sinuous ditch 4204 bounded the enclosure complex, the ditch approximately 120m in length and turning to the south-east towards its southern end, the terminal removed by a sinkhole; this ditch defined the maximum southerly extent of the sequence of enclosures. At its northern end, ditch 4204 was cut by 4195, the latter ditch at least 45m long with a ‘dog-leg’ plan and perhaps a later modification belonging to this phase; the southern end of ditch 4195 lay within 5m of the northern terminal of inner enclosure ditch 4201.

There was no equivalent outer enclosure ditch on the east side to 4204 on the west side. However, narrow, shallow ditch 4037 extended south-east from inner enclosure ditch 4025 and continued beyond the limit of excavation.

Phase 4: this was the final phase identified, which in large part followed the layout in Phase 3, possibly creating a double-ditched enclosure (Fig. 4), the new ditches slightly larger but otherwise similar in their open, U-shaped profiles and their generally homogeneous, single fills containing a very limited range and number of finds.

On the west side was ditch 4029, approximately 100m in length, and generally 1-1.45m wide and 0.35-0.50m deep, though in three interventions depths of 0.7m, 0.75m and 0.95m were recorded. The course of ditch 4029 closely mirrored that of Phase 3 inner enclosure ditches 4195 and 4201, with a gap of approximately 3-5.5m between them. To the north, ditch 4029 terminated close to the north-west corner of the excavation area, while to the south it terminated broadly in line with the southern entrance to the Phase 3 inner enclosure between ditches 4201 and 4025. The corresponding Phase 4 entrance was narrower, only 2m wide, the enclosure to the east of this defined by ditch 4024, this at least 105m long, 0.95-1.15m wide and 0.35-0.45m deep. As in the west, the course of the Phase 4 ditch mirrored its Phase 3 predecessor, though ditch 4024 extended further to the north than ditch 4025 and continued beyond the limit of excavation in the north-east corner of the site, here following parallel, perhaps still-surviving Phase 2 ditch 4026. The gap between ditches 4024 and 4025 on the east and south-east sides was approximately 2.5-5m. Apparently broadly contemporary and extending south-south-west from the eastern terminus of ditch 4029 was a further, somewhat irregular ditch, 4214. This was approximately 20m in length, but its purpose is unclear; it could, for example, have created an outer enclosure measuring 35m by 25m if Phase 3 ditch 4204 was still extant, though the possibility that it was an earlier, Phase 1 or Phase 2 element cannot be entirely ruled out.

Structure 4197 (bakery/kitchen)

Largely on the basis of stratigraphic relationships, one significant internal feature has been assigned to Phase 4, this comprising sunken-featured bakery/kitchen [pg121]4197. This lay in the south-east corner of the enclosure and had been dug through the Phase 2 and Phase 3 ditches here (Fig. 4).

Structure 4197 (Fig. 5) was sub-oval in plan, measuring 5.8m east-west by 4.3m north-south, surviving up to 0.18m deep and with moderately sloping sides and a flat base. A large oven, 4173, lay at the rear against the east end and a smaller hearth, 4177, close to it to the south. The entrance appeared to face west, with two large postholes, 4320 and 4448, on either side, and an additional posthole, 4318, adjacent to 4320, perhaps inserted later.

[fg]png|Fig. 5 Plans of Anglo-Saxon SFB 4430 and medieval bakery/kitchen 4197.|Image[/fg]

Oven 4173 was almost perfectly circular, 2.4m in diameter, only the eastern edge incomplete, possibly a result of use. It was constructed on a base of redeposited brickearth, capped by a layer of flint nodules, with several relatively large fragments of re-used Romano-British CBM and lava quern towards the centre. These materials formed a solid platform, 4180, over which a thin layer of redeposited brickearth formed the floor of the oven, this showing evidence for having been subject to moderate heat. No fired clay that may have derived from the superstructure was present. Only 0.2m to the south-west of oven 4173 was hearth 4177, oval in plan, measuring 1.7m by 1.2m and 0.25m deep, and largely comprising redeposited brickearth, the upper part heat-reddened and more intensively burnt than seen in the oven, and with some charcoal present. The two principal postholes, 4320 and 4448, were set 3.5m apart (centre-to-centre), the former approximately circular, 0.8m in diameter and 0.35m deep, the latter sub-rectangular, measuring 1.1m by 0.55m and 0.17m deep; posthole 4318 adjacent and to the east of 4320 was 0.5m in diameter and 0.35m deep.

Uncertainly-phased pits and postholes

The majority of the artefactually-dated discrete features are medieval, most of them pits (six examples), but a paucity of stratigraphic relationships means that in general they cannot be more closely phased; many of the undated pits are also likely to be medieval on broad stratigraphic grounds (eight examples) (Fig. 4). Where some broad phasing was possible, it suggested an attribution to Phases 3 or 4 as most likely. Only four pits cannot be assigned to a period on either artefactual or stratigraphic evidence.

[pg122]The pits were concentrated within the central area of the sequence of medieval enclosures, this likely reflecting the location of contemporary structures for which, except for sunken-featured bakery/kitchen 4197 to the south-east, no clear or coherent evidence survives, unless possible SFB 4558 also belongs to this period.

The pits included two large examples. The first, 4402, in the north-west corner of the site, was sub-oval measuring 3.7m long, 3.25m wide and 0.9m deep, with steep sides and a shallow concave base; it contained medieval pottery and was cut by a small pit (4407). The second example, pit 4388, was located amongst the central group of pits and was also oval in plan but somewhat smaller, 2.5m long, 2.1m wide and 0.7m deep, though of similar profile. No pottery was recovered but finds included relatively large quantities of fragmentary fired clay from several of the fills, as well as four block-like pieces likely to have had a structural function, along with small amounts of CBM, animal bone, slag, lava quernstone, burnt flint and oyster shell.

Two further large, sub-rectangular or oval pits, 4102 and 4270 (cut by ditch 4029), measuring 3.1-4.65m long, 2.7-3.6m wide and 0.65-0.9m deep, contained only a few, probably residual sherds of Romano-British date. A medieval (or Anglo-Saxon) date is, however, indicated by typically post-Roman cereal crops, free-threshing wheat and rye, in the charred plant remains assemblage. Smaller, oval pit 4348 has been assigned to the medieval period on the same basis.

Medium-sized pits were the most common type, with nine examples, a further three being categorised as small to medium, the former with maximum dimensions of 1.6-2.1m and depths of 0.35-1.2m, the latter between 1.1-1.3m across and less than 0.5m deep. Most of these were unremarkable in terms of their fills and the finds recovered, but a small number warrant mention and are detailed below. Of the six small pits, with dimensions of less than 1.1m across and depths of 0.15- 0.5m, none is noteworthy.

Pit 4050 was located outside of the medieval enclosure, towards the eastern edge of the site. It was oval, 1.3m long 1.15m wide and 0.45m deep and contained a dump of fired clay (5.57kg), mostly amorphous pieces, although some wattle impressions could be identified.

Pit 4137, 2m long, 1.45m wide and at least 1m deep (full depth not ascertained) had vertical, straight sides and was initially interpreted as a well, but the possibility that it was a sinkhole curtailed further investigation.

The fills of oval pit 4185, 2.1m long 1.65m wide and 1.2m deep, revealed a relatively complex sequence, though generally few finds. However, one layer contained a relatively large quantity of fired clay fragments, some with wattle impressions.

Sub-circular pit 4373 cut enclosure ditches 4023 and 4025 of Phases 2 and 3 respectively and is included here primarily because of its relatively large size, 2.1m long 1.9m wide and 0.7m deep. The fills probably largely reflect natural infilling, although some animal bone and lava quern fragments were recovered.

Finally, the small cluster of six pits, some intercutting, in the central area is notable, the only place where such a density occurred. except for smaller pit 4550, they were all oval in plan and of similar size and depth (where excavated), comprising 4534, 4537 and 4545, along with two unexcavated examples (4137 and 4156). Finds included a few sherds of pottery, animal bone, burnt flint and [pg123]slag. A medieval date is also indicated for pit 4545 on the basis of the charred plant assemblage, which includes free-threshing wheat and rye, and a similar date might be proposed for uncertainly phased features 1704 and 4388 elsewhere which contained the same cereal crops.

Thirteen postholes were investigated, with approximately 30 left unexcavated. These were concentrated in the northern part of the site, in and around the central inner enclosure, but no coherent and certainly contemporary arrangements were identified here or elsewhere, with no more than three making up any possible alignments. Only four postholes contained dating evidence, in three cases sherds of medieval pottery, and in the other post-medieval glass and clay pipe, so the possibility remains that some were earlier, perhaps prehistoric, and others post-medieval.

Two of the medieval dated postholes (4085 and 4087), along with an undated example (4091), lay in a group close to the southern terminus of phase 2 ditch 4026. These postholes were all approximately 0.55m in diameter and 0.1-0.3m deep, possibly forming part of an entrance arrangement to the Phase 2 enclosure. The other medieval dated posthole (4337), of similar size, was in the north-west corner of the site, where various ditches terminated, and this may have been one of at least three postholes forming an alignment that extended to the north-east.

One final feature to note here is 4295 (not illustrated), a spread of unworked flint nodules and smaller fragments, 13m long, 4m wide and up to 0.4m thick. This protruded through the subsoil but was sealed by topsoil, and partially overlay medieval enclosure ditches 4201 and 4202. A post-medieval date is likely, the material possibly deriving from the clearance of a wall or structure in the vicinity.

Specialist reports

Pottery (Grace Jones (prehistoric) and Lorraine Mepham)

The pottery assemblage (447 sherds, 4,811g) is of Bronze Age to post-medieval date, with chronological foci in the prehistoric and medieval periods. Condition is variable, typically moderate to poor. Mean sherd weight overall is 10.7g, but this dips to 7.4g for the softer-fired and more friable Bronze Age sherds and rises to 14.1g for the harder-fired medieval wares. There are no re-constructable profiles.

The assemblage has been quantified (sherd count and weight) by ware type within each context; for post-Roman material this follows the regional type series established by the Canterbury Archaeological Trust. For earlier material, broader groupings have been used (e.g. flint-tempered wares). The presence of identifiable vessel forms and other diagnostic features have been noted. Table 1 gives the breakdown of the assemblage by ware type.

Prehistoric

Pottery of Middle to Late Bronze Age date was recovered from six features. The assemblage consists entirely of flint-tempered wares. The base and lower walls of a vessel, with all surfaces abraded, was found in cremation grave 1804 (47 sherds, 320g). The fabric contains common inclusions (20%) of poorly sorted, angular, calcined flint up to 5mm in size. The fabric suggests a later prehistoric date, [pg124]probably during the Middle or Late Bronze Age, but no diagnostic indicators of form were found. However, a Late Bronze Age radiocarbon date of 1050-910 cal. BC (SUERC-103752) was obtained on cremated bone from the burial urn. A group of 67 sherds (802g), probably from a single vessel, came from pit 4302. With the exception of one cordoned sherd, all are undiagnostic body or base sherds, but the fabric is again typical of middle to late Bronze Age wares. A smaller group (19 sherds, 49g) from pit 4384 has only one diagnostic element – a sherd from a base with its underside flint-gritted, a surface treatment particularly associated with later Bronze Age pottery. Seventeen sherds (24g) from ring-ditch 4020 include one with a boss or lug. Six body sherds (33g) in a well sorted fabric, and a single thick-walled, coarsely gritted sherd, came from ditch 4554. These are also likely to be of Middle to Late Bronze Age date.

Table 1. Pottery totals by ware type

[tb][th]Ware type|No. sherds|Weight (g)[/th]

[tr]LBA flint-tempered wares|184|1,369[/tr]

[tr]All other wares (prehistoric)|7|22[/tr]

[tr]All wares (Roman)|19|205[/tr]

[tr]EMS2 Chalk-filled sandy ware|5|51[/tr]

[tr]EMS4 Organic-tempered ware|20|181[/tr]

[tr]EM2 Shelly ware|94|1,244[/tr]

[tr]EM3 Misc shelly wares|8|154[/tr]

[tr]EM3A Shelly-sandy ware|1|10[/tr]

[tr]EM12 Andenne ware|1|5[/tr]

[tr]EM28 Kentish sandy ware with shell and flint|1|70[/tr]

[tr]EM30 Non-local coarse quartz with shell and flint|1|8[/tr]

[tr]EM36 N or W Kent sandy and shell-tempered|64|1,072[/tr]

[tr]EM.M5 Ashford–Potters Corner ware|2|8[/tr]

[tr]M1 Tyler Hill ware|2|25[/tr]

[tr]M5 Fine London-type ware|1|1[/tr]

[tr]M10 Wealden-type pink-buff sandy ware|2|4[/tr]

[tr]M38A NW Kent sandy ware (mainly reduced)|34|369[/tr]

[tr]Post-medieval|1|7[/tr]

[tr]---|---|---[/tr]

[tr]TOTAL|447|4,811[/tr]

[/tb]

Three sherds in sandy fabrics have been tentatively identified as Iron Age but possess no diagnostic features. Undiagnostic abraded body sherds in flint-tempered fabrics, grog-tempered wares, a sandy ware and a vesicular fabric were recorded [pg125](all possibly residual) from ten features but none are more closely datable than to the prehistoric period.

Romano-British

Pottery of Romano-British date occurs in very limited quantities (see table 1), consisting only of abraded body sherds. Sherds were found (some residual) in very small numbers in 12 features, and all are in coarseware fabrics.

Saxon

Of the 25 Saxon sherds, 20 are in organic-tempered fabrics (EMS4) with five sherds in a chalk-filled sandy ware (EMS2). Both date to the early to mid-Saxon period, although EMS2 has a slightly earlier currency in Canterbury (c. 450-625/50) than organic-tempered EMS4 (c. 575/600-725). The two types were not found in the same features: organic-tempered sherds came from SFB 4430 (15 sherds) and pit 4440 (4 sherds), with one sherd found in subsoil during the evaluation. All five chalk-filled sherds came from pit 4021. The organic-tempered sherds from SFB 4430 include rim, body, and rounded base sherds, and most could belong to a single vessel, but the overall profile could not be reconstructed. The chalk-filled sherds from pit 4021 include a slightly everted jar rim from a convex vessel. None of these diagnostic sherds enables any closer dating of this small Saxon group.

Medieval

The medieval assemblage (211 sherds, 2,970g) includes a range of local and regional wares, with one continental import. The potential date range spans the medieval period from the mid-eleventh century onwards, with a focus in the early medieval period (mid/late eleventh to mid-thirteenth century).

The early medieval component is dominated by shelly, shelly-sandy, and sandy-shelly wares, all likely to be of at least relatively local manufacture in north-west Kent. No specific source (and there were probably several) is known for any of these wares, but general areas can be highlighted: the fossil shell in some of the shelly wares could derive from the Woolwich Beds. Two sherds have been tentatively identified as Ashford Potters Corner ware (EM.M5) but, given that these are small sherds, the identification is not completely confident and these sherds might also belong to the local ceramic tradition.

Given the uncertainty of the source(s) of these early medieval wares, of particular interest here is the presence of a small group (five sherds) from pit 4420 which have been overfired to such an extent that the sherds are distorted, indicating this is not just burning during use or post-depositional. These sherds could represent ‘wasters’ from pottery manufacture. Such a small quantity cannot be taken as evidence of on-site pottery production, as they could have been introduced from elsewhere, but their occurrence here is nevertheless of some significance.

Identifiable vessel forms amongst the early medieval wares are almost entirely limited to jars with undeveloped (simply formed) rims (12 examples), three of them finger-impressed. There is also one bowl rim.

[pg126]One other sherd from the early medieval group is of interest: this is a sherd from the handle of a glazed pitcher in a fine whiteware identified as Andenne ware, imported from the low countries. Examples of Andenne ware in this country generally date between the mid-eleventh and early thirteenth century (Vince and Jenner 1991, 106; Cotter 2006, 232) but are not particularly common outside major ports. The sherd came from ditch 4557, where it provided the only dating evidence.

Local wares are also prominent amongst the later medieval group, in the form of reduced fine sandy ware (M38A). This ware is conventionally dated as c.1175- 1350, but the only vessel forms identified here are jars with undeveloped rims, suggesting that most if not all of the sherds of M38A date prior to c.1200, roughly the point at which these were superseded by jars with more carefully formed (developed) rims. There are only five other sherds in this later medieval group, all regional wares (Tyler Hill, London-type, and Wealden-type pink/buff sandy). These extend the potential date range at least into the late fourteenth century and possibly later, but overall, the evidence suggests that activity on the site after the early thirteenth century was sporadic.

Medieval sherds were found in 23 features, in quantities ranging from one to 56 sherds, but numbers in most cases were small. The two most productive features were pit 4085 (56 sherds) and pit 4029 (52 sherds), both dated as eleventh-/twelfth-century: the sherds from pit 4085 could all belong to a single vessel. All other features yielded 20 sherds or fewer.

Charred plant remains from sunken-featured bakery/kitchen 4197 (Megan Scantlebury)

Recent large-scale infrastructure projects in Kent, including Northumberland Bottom, Gravesend on High Speed 1 (HS1) (Askew 2006), the realignment of the A2 near Gravesend (Allen et al. 2012) and at Thanet Earth, the Isle of Thanet (Henshaw et al. 2019), have uncovered numerous medieval sunken-featured buildings. The structures seem unlikely to be houses, and the hearths and ovens would have posed a significant fire risk, which is likely why these structures tend to be identified on the periphery of settlements and/or in the corners of enclosures (cf. Allen et al. 2012, 576). This suggests some kind of specialist activity or activities. Most sunken-featured buildings with both an oven and a hearth have been interpreted as ‘bakeries and/or kitchens cum brewhouses’ (Schuster and Stevens 2009, 251). This interpretation was based on the evidence for large domed ovens and the lack of evidence for other industrial/craft activities such as metalworking (ibid., 249), added to which the ovens would likely not get hot enough for pyro-technical activities (Henshaw et al. 2019, 468). The environmental evidence recovered from some of these structures can help assist their functional interpretation.

The analysed samples from Otterham Quay Road revealed an assemblage of charred plant remains rich in cereals, pulses and weed seeds. Comparable cereal-rich assemblages have been recovered at Fulston manor (Powell et al. 2009) and Star Lane (Stevens 2009), from Site L (oven 12583) and Site C (oven 5347) near Gravesend (Smith 2012), and at Thanet Earth (Carruthers 2019).

The presence of high volumes of un-milled cereal crops in possible bakeries are interesting as they would have likely relied on flour being transported to them [pg127]having been already milled (Schuster and Stevens 2009, 250). The rich array of cereals identified in ditch 4025, predating the possible bakery, and the fragment of lava quern stone used as part of the base of oven 4173, certainly indicates a tradition of cultivation and crop-processing in the vicinity. The milling of flour may have taken place nearby or even, potentially, in the building itself.

One of the possible functions of these structures is the production of malt for brewing. Brewing is widely evidenced during this period and likely a common activity in settlements (Moffett 2006). The recovery of large quantities of germinated grains in association with these structures, and the ovens and hearths specifically, would provide convincing evidence for the production of malt. Other proxies for deliberate germination include detached embryo ends and coleoptiles. No obviously germinated grain was identified in the current assemblage. A small number of detached embryo ends and/or coleoptiles were recovered, though the evidence is too scarce to make any firm conclusion about the possibility of malting in this structure. cereals can also germinate when harvested in wet weather or stored in damp conditions. Similarly low numbers of germinated grains and/or embryos and coleoptiles have been identified at Thanet Earth (Carruthers 2019) and near Gravesend (Smith 2012). However, if producing malt using roasted barley, the grains do not necessarily have to be germinated (Schuster and Stevens 2009, 250). Some of the barley identified in the assemblage may have originated from charring accidents during the production of malt, even though malting is not clearly evidenced in the archaeobotanical remains. Brewing is potentially more convincingly evidenced near Gravesend, where a bunghole cistern, likely used to store or brew ale, was recovered from a pit adjacent to a medieval SFB (Allen et al. 2012, 517).

Alternative explanations for the high quantities of cereals identified in these structures may include the accidental burning of the grain during parching, prior to milling for flour or for storage (cf. Moffett 2006, 51). Grain-rich assemblages from other medieval kiln/oven assemblages are interpreted as resulting from a layer of grain being used to prevent the bread loaves from sticking to the floor of the oven (Moffett 1994, 60; see also Schuster and Stevens 2009). Or, processed grain could have been thrown into the ovens and hearths to check their temperature, where it would be expected to explode like popcorn (Powell et al. 2009, 183).

While the charred plant remain assemblage is very cereal grain rich, the rachis (chaff), legumes and array of wild plant seeds show that some mixing has occurred. The wild plant/weedy taxa present include both small-seeded, such as stinking chamomile, and large-seeded grain-sized weeds such as corncockle, wild grasses and black bindweed. These grain-sized seeds are likely to remain with the cereal grains until the last stages of processing (e.g., hand-picking) (Stevens 2003). The presence of chaff (e.g., the free-threshing wheat rachis, barley and rye rachis) could indicate the use of crop-processing waste from the coarse sieving process as tinder. This was likely redeposited rake-out from the ovens and/or hearths which would become incorporated into other deposits over time.

Pulses are commonly recovered in assemblages of medieval plant remains, although usually only in small quantities (Moffett 2006). Some samples in association with sunken-featured bakery/kitchen 4197 contained significant proportions of charred legumes. While garden peas and broad beans were likely grown as food, common vetch was probably a fodder crop or famine food (ibid.).

[pg128]This high proportion of pulses is also paralleled at comparable sites, such as near Gravesend (Smith 2012, 567), and it is possible that these structures were also used to process non-cereal crops for food or fodder. Pulses may have been grown alongside cereals as a maslin, or in rotation with cereals to improve and nitrogenate the soil, and/or to prevent lodging (Carruthers 2019, 1125; Moffett 2006).

Considering the evidence, it is likely that the charred plant remains recovered from sunken-featured bakery/kitchen 4197 do not indicate any particular activity or specific use and are likely to originate from a mixture of sources and processes and accumulated over time (Fuller et al. 2014). These include the probable ‘routine’ processing of crops taken from storage, cooking and baking, the use of crop processing debris as fuel, the accidental burning of crops while creating malt for brewing ale and parching, amongst others (cf. van der Veen 2007).

The specialist detailed Finds Reports can be found on the KAS website: Phil Harding (flint), Phil Andrews (slag), Jacqueline I. McKinley (human bone), Lorrain Higbee (animal bone) and Megan Scantlebury (charred plant remains).

Discussion

Early Prehistoric

The unstratified Middle Palaeolithic Levallois flake found during the evaluation is of some interest given previous discoveries of Palaeolithic flint implements, including hand axes, further to the north close to Otterham Quay. However, no clear lithological or stratigraphic context from which the flake might have derived was identified during the subsequent test-pit evaluation of the Quaternary deposits surviving in the un-quarried southern half of the site.

Bronze Age

None of the remaining small assemblage of worked flint can certainly be ascribed a date, though a Neolithic or Bronze Age origin is probable for much of it. The earliest feature identified was a small ring-ditch in the south-east corner of the site, this more likely to be of Middle Bronze Age date than Early Bronze Age, based on its small size, with an external diameter of 9m, though there is nothing further to corroborate this. It seems certain that ring-ditch 4020 represents the remains of a barrow, the associated mound or bank eroded and ploughed flat, but there was no central grave present, this presumably contained within the former earthwork or otherwise removed by the plough. A small assemblage of Middle/Late Bronze Age pottery from the ring-ditch does, however, support the suggested date, while a single, truncated, urned cremation burial a short distance to the north is likely to have been a satellite grave. The surviving base of the undiagnostic later prehistoric vessel contained the remains of a mature adult, most likely a female, this providing a late Bronze Age radiocarbon date of 1050-910 cal. BC (Suerc-103572), making it broadly contemporary with the settlement evidence discussed below. As noted above, this represents a relatively rare example within the county, as well as further afield, of an in situ Late Bronze Age burial. No other ring-ditches are recorded in [pg129]the vicinity, and singletons such as this are not uncommon. Furthermore, they are often sited with no clear topographic reason apparent in determining their location, as appears to be the case here where the monument was constructed on generally flat, relatively low-lying ground.

Enclosure 4554, assigned a Middle/Late Bronze Age date based on a small assemblage of pottery, is considered likely to have post-dated ring-ditch 4020, though it was conceivably sited with reference to the barrow monument, particularly if ditch 4203 to the south, along with ditches 4292 and 4293 to the north, were part of the layout. The small trapezoidal enclosure is of uncertain function and contained no contemporary features, though the pottery, along with the struck flint from one of the ditch terminals, suggest it was settlement-related rather than purely agricultural, for example an animal enclosure. However, diverging ditches 4203 and 4292/4293 may have formed a droveway for stock control, terminating to the east but continuing beyond the limit of excavation to the west, and with enclosure 4554 on the north side. Perhaps also related to this use is feature 4302, which from its size and form might best be interpreted as a waterhole, to the north-west of enclosure 4554, one of only two discrete features of this date identified. As with the apparently isolated ring-ditch, no settlement remains of Middle/Late Bronze Age date are known from the vicinity, and so the limited evidence from the current site makes a useful contribution to advancing understanding of the later prehistoric landscape of the area close to Otterham Creek and the North Kent Marshes. If it was settlement-related, enclosure 4554 would be relatively unusual in that settlements more generally were unenclosed at this time, though a variety of enclosed examples are known.

Iron Age/Romano-British

Despite the pit or possible hut of reported Iron Age date found in 1934 to the north during quarrying, no certain Iron Age features were identified. However, there were likely two or possibly three Romano-British pits, which may have originated in the Late Iron Age, towards the north-east corner of the site. These pits contained only limited finds comprising a few sherds of pottery and a small quantity of animal bone, the more closely dated pottery ascribed to the Late Iron Age/early Romano-British period. It is possible that all of this material is residual and the pits perhaps medieval or possibly Anglo-Saxon, like three other pits containing Romano-British pottery but typically post-Roman charred plant remain assemblages, however, on balance, a Romano-British date is considered most likely. Little more can be deduced based on this slender evidence, but it can be suggested that the features indicate settlement nearby, this likely to have been of a low-status rural nature and perhaps related to the nearby Upchurch pottery industry, with several kiln sites known around the fringes of the Thames Marshes to the north (Monaghan 1987). Also to the north, close to Otterham Creek, is the suggested location of an early Romano-British cemetery, found during brickearth digging in the early 20th century (MKE 3073), while approximately 1 km to the north-east are cropmarks indicating what may be the walls of a rectangular structure.[pg130]

Saxon

A single, Saxon sunken-featured building (4430) has certainly been identified, with a second possible example (4358) tentatively ascribed to this period, along with at least two pits. SFB 4430 exhibits some typical features of these buildings, though in this case with a double posthole at the east end (one posthole possibly the replacement of the other), with no surviving posthole at the west end; such an arrangement is relatively rare but not unknown (Tipper 2004). Another feature of note is shallow, rectangular pit 4431 in the base of the eastern half of this SFB, which though containing only a few crumbs of late prehistoric pottery, seems more likely to be Anglo-Saxon based on its location and alignment which otherwise would be remarkably coincidental. The purpose of this and the two irregular pits in the base is unclear, the former possibly used for storage, and there is nothing that would confirm or rule out the presence of a raised, planked floor. Finds from the fill of SFB 4430 were sparse but the pottery suggests an early/mid Saxon date, focused on the seventh century AD, and its location within and seemingly aligned to Bronze Age enclosure 4554 suggests the latter may still have been partly extant at this time.

The interpretation of feature 4358 as an Anglo-Saxon SFB is somewhat more equivocal, though it appeared to be cut by at least two medieval features. However, no associated postholes were identified, and while no pottery was recovered from the excavated segments, the possibility that 4358 was a medieval structure should be considered, associated with an early phase of enclosure development and on a similar alignment (see below). Perhaps one counter to this argument might be that the two pits ascribed to the Saxon period (4021 and 4081) both lay to the east, possibly hinting at a link, though the pottery from pit 4021 suggests an early Saxon date and the loomweight fragments from pit 4082 a mid–late Saxon date, such that more than one phase of Saxon activity might be indicated overall.

Further discrete features of this broad period may lie beyond the limits of excavation, with SFBs and pits often widely dispersed, and in this case with no associated timber halls apparent amongst the relatively small number of widely distributed, mostly undated postholes. Despite the limited evidence, the remains of settlement-related activity again help populate a local landscape where there was previously nothing similar known from this period. It possibly also serves to give context to two earlier discoveries of Anglo-Saxon burials approximately 1km to the north (near Otterham Quay), one comprising a single grave with spearheads, pottery and beads (MKE 3061/MKE 16493), and the other an inhumation cemetery with brooches, glass vessel and beads (MKE 43032). Though these burials may have been unrelated to the current site, either chronologically or spatially, they provide further evidence for Anglo-Saxon settlement in the area.

Medieval

The principal and probably most interesting aspect of the Otterham Quay Lane site lies within the medieval period, focused on the mid/late eleventh to mid-thirteenth centuries, but with a few sherds of pottery that could extend the sequence into the late fourteenth century or beyond. Here the excavation revealed a coherent [pg131]sequence of enclosure development, only the northern part removed by nineteenth- century brickearth digging, which has been teased apart largely on the basis of stratigraphic rather than ceramic evidence. Relatively little medieval pottery was recovered, and this does not allow for finer chronological distinctions to be made. Nevertheless, four medieval phases have been identified, and although some details of the sequence presented above may be incorrect, overall, the results suggest a slight increase in the extent of the enclosures over time, whilst maintaining a broadly similar layout. Unless the feature designated as SFB 4558 was a contemporary structure, then no evidence of buildings was identified, apart from sunken-featured bakery/kitchen 4197 (see below). The concentration of pits in the central, inner part of the enclosure sequence does, however, suggest that this is where the principal building(s) lay, of which no apparent trace has survived. In this context, the presence of a very large natural sinkhole immediately to the north of SFB 4558 should be considered, since this may have removed some structural evidence here. one piece of evidence that did survive, however, and would give support to the enclosures having had a settlement function, is sunken-featured bakery/kitchen 4197. These have been recorded on a number of sites in Kent, particularly in the east of the county over the past 25 years, ranging in date between the eleventh and fourteenth centuries, and of broadly similar size and sunken form, with an adjacent oven and hearth at the end away from the entrance. A function as bakeries and perhaps also serving for brewing has been proposed, the oven for baking bread and the hearth for boiling water for use in brewing, but like detached medieval kitchens, a range of food and drink preparation purposes could have been served by these structures (Schuster and Stevens 2009; Allen et al. 2012, 570-83). Overall, the nature of the evidence outlined above, together with the layout of the enclosures, and their narrow entrances, suggests that they might best be associated with settlement rather than, for example, stock control. However, this could have been one of several centres within a broader pastoral – as well as arable – landscape exploiting the nearby marshlands to the north for grazing, with the surrounding fields possibly used for holding stock as well as for arable agriculture. The small quantities of animal bone from the pits and ditches provide little additional information beyond confirming the expected presence of the normal range of domesticated livestock, and few other finds were recovered. However, amongst the medieval pottery were a few sherds of possible wasters, hinting at the possibility of some local production, as evidenced in the area during the Romano-British period.

For a parallel to the Otterham Quay Lane enclosures and related features, one does not have to look far, only to Fulston manor approximately 10km to the south- east, just beyond Sittingbourne (NGR 590740 162750). Here, excavations in 2004- 5 exposed part of a large, probably rectangular, ditched enclosure measuring 70m by at least 50m, with an inner enclosure of approximately 45m by 35m in the north-east corner and a well-preserved sunken-featured bakery/kitchen in the north-west corner, this phase of activity dated by pottery to the mid eleventh/early thirteenth century (Powell et al. 2009). The outer enclosure ditches were up to 1.5m wide and 0.5m deep, and those of the inner enclosure up to 0.6m wide and around 0.2m deep, a little smaller than at Otterham Quay Lane. The sunken-featured bakery/ kitchen was also smaller, that at Fulston Manor measuring 7.4m by 3.8m, with [pg132]two entrance postholes and traces of a clay floor, but no other structural features. Here also virtually no rooftile was present, and a thatched timber structure is a possibility at both sites, despite the fire risk. At one end was an oven 2m in diameter, rebuilt twice, with an adjacent pear-shaped hearth, the arrangement here almost identical to that at Otterham Quay Lane, as well as at other sites in Kent. Although charcoal was relatively abundant at Fulston Manor, the majority oak with a mix of other species, few charred plant remains were present, mainly wheat, and it was suggested that the oven could also have been used for smoking or drying (Powell et al. 2009, 183). Of interest are the quernstone fragments recovered from the oven base infill of this bakery/kitchen, suggesting that the milling of flour took place nearby, and several fragments of quernstone had also been incorporated in the oven structure at Otterham Quay Lane. Here, as noted above, considering the evidence overall, it is likely that the charred plant remains recovered from sunken-featured bakery/kitchen 4197 do not indicate any particular activity or specific use. They are likely to originate from a mixture of sources and processes and accumulated over time, including as a result of probable ‘routine’ processing of crops taken from storage, cooking and baking, the use of crop processing debris as fuel, the accidental burning of crops while creating malt for brewing ale, and parching.

At Fulston Manor, there were also several pits, as well as a probable crop dryer, but no further structural remains were exposed, at least within the limit of excavation, though the possibility that evidence for beam-slot buildings had been entirely removed was considered (Powell et al. 2009, 193). A further phase of enclosure development extended the sequence into the mid-fourteenth century, this apparently less extensive than its predecessor, with an associated sunken-featured structure of unknown function, conceivably similar to the uncertainly dated SFB 4558 (see above) at Otterham Quay Lane. Here, the sequence also extended into the fourteenth century, though probably on a much-reduced scale. Settlement at both sites may have declined and come to an end later in the fourteenth century as a result of various factors, but primarily climatic deterioration leading to bad harvests and famine, animal and human diseases, including the Black death, as well as possible changes in ownership.

At both sites there is little to indicate any degree of high status: a possible early link to the post-medieval Fulston Manor, and at Otterham Quay Lane a single sherd of Andenne ware from the Low Countries which is not common outside of the major ports in the South-east. However, at the time of Domesday in 1086 the area was one of the most valuable in Kent, with both Rainham and Fulston falling within the manor of Milton Regis, a former Anglo-Saxon royal estate still held by the crown. Its economy was largely based on livestock farming involving transhumance exploiting the summer grazing on the downland to the south and the marshland to the north, and it is perhaps this that links the two sites in their similar chronology, form and function. These would have been part of a pattern of dispersed farms and settlements, many created as woodland – the Forest of Blean – was cleared for cultivation and grazing in the late Saxon and early medieval periods (Everitt 1986, 313).

[pg133]

Acknowledgements

Wessex Archaeology would like to thank Duncan Hawkins of RPS Group for commissioning the archaeological mitigation works. We are also grateful for the advice of Ben Found and Simon Mason of Kent County Council Heritage Conservation Group, who monitored the project for Medway Borough Council. This report has been edited for publication by Terry Lawson.

The fieldwork was managed by Rob De’Athe, with the evaluation and excavation directed by Lisa McCaig and the geoarchaeological investigation directed by Dr Andrew Shaw. The post-excavation assessment report was written and compiled by Lisa McCaig, and Grace Jones and Tom Wells managed the subsequent analysis and publication programme. The illustrations are by Esther Escudero.

The project archive is currently held at the offices of Wessex Archaeology in Salisbury under site codes 217440 and 217441. The site falls in an area where there is currently no collecting museum, but it is hoped that a suitable repository will be identified in due course.

Bibliography

Allen, T., Donnelly, M., Hardy, A., Hayden, C. and Powell, K., 2012, A road through the past: archaeological discoveries on the A2 Pepperhill to Cobham road-scheme in Kent. Oxford, Oxford Archaeology Monograph Series 16.

Andrews, P., Egging Dinwiddy, K., 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. Salisbury, Wessex Archaeology Report 24.

Askew, P., 2006, The Prehistoric, Roman and Medieval Landscape at Northumberland Bottom, Southfleet, Kent. HS1 Integrated Site Report Series (ADS).

Cotter, J., 2006, ‘The pottery’, in Parfitt, K., Corke, B. and Cotter, J., Townwall Street, Dover: Excavations 1996. Canterbury Archaeological Trust, The Archaeology of Canterbury New Series Vol 3, 121–254.

Carruthers, W. J., 2019, ‘Charred, waterlogged and mineralised plant remains’, in Henshaw, R., Holman, J., Masefield, R., Rady, J. and Weekes, J. 2019, Beneath the Seamark: 6,000 years of an Island’s History. Archaeological Investigations at ‘Thanet Earth’, Kent 2007–2012. Volume 2: Specialist Reports. Canterbury Archaeological Trust Technical Report 2. Canterbury: Canterbury Archaeological Trust., 1087–1135.

CgMs 2015, Otterham Quay Lane, Rainham, Kent. Archaeological desk-based assessment. Unpublished report PC/17221.

Everitt, A., 1986, Continuity and Colonization: the evolution of Kentish settlement. Leicester University Press.

Fuller, D. Q., Stevens, C. J. and McClatchie, M., 2014, ‘Routine activities, tertiary refuse and labor organization: social inference from everyday archaeobotany’, in Madella, M. and Savard, M. (eds), Ancient Plants and People. Contemporary Trends in Archaeobotany, 174–217. Tucson: University of Arizona Press.

Henshaw, R., Holman, J., Masefield, R., Rady, J. and Weekes, J., 2019, Beneath the Seamark: 6,000 years of an Island’s History. Archaeological Investigations at ‘Thanet Earth’, Kent 2007–2012. Volume 1: Chronological Narrative. Canterbury Archaeological Trust Technical Report 2. Canterbury: Canterbury Archaeological Trust.

Monaghan, J., 1987, Upchurch and Thameside Roman Pottery: a ceramic typology for northern Kent, first to third centuries AD. Oxford, British Archaeological Reports, British Series 173.

Moffett, L., 1994, ‘Charred cereals from some ovens/kilns in late Saxon Stafford and the botanical evidence for the pre-burh economy’, in Rackham. J. (ed), Environment and Economy in Anglo-Saxon England. York: Council for British Archaeology Research Report 89, 55–64.

Moffett, L., 2006, ‘The archaeology of medieval plant foods’, in Woolgar, C., Serjeanston, D. and Waldron, T. (eds), Food in Medieval England: Diet and Nutrition, 41–55. Oxford: Oxford University Press.

Powell, A. B., Barnett, C., Grimm, J. M., Mepham, L., Phillpotts, C. and Stevens, C. J., 2009, ‘A medieval enclosure and bakery or brewhouse at Fulston Manor, Sittingbourne’, in Andrews et al., 2009, 175–97.

Schuster, J. and Stevens, C. J., 2009, ‘A Kentish type of medieval sunken-featured bakery/kitchen’, in Andrews et al., 2009, 249–51.

Smith, W., 2012, ‘Charred plant remains’, in Allen et al., 2012, 562–8.

Stevens, C. J., 2003, ‘An investigation of consumption and production models for prehistoric and Roman Britain’. Environmental Archaeology 8, 61–76.

Stevens, C. J., 2009, ‘Environmental evidence: charred plant remains from medieval contexts’, in Egging Dinwiddy, K. and Schuster, J., ‘Thanet’s Longest Excavation: archaeological investigations along the route of the Weatherlees–Margate–Broadstairs wastewater pipeline’, in Andrews et al., 2009, 142–3.

Tipper, J., 2004, The Grubenhaus in Anglo-Saxon England: An analysis and interpretation of the evidence for a most distinctive building type. Yedingham, The Landscape Research Centre.

van der Veen, M., 2007, ‘Formation processes of desiccated and carbonized plant remains – the identification of routine practice’, Journal of Archaeological Science 34, 968–90.

Vince, A. G. and Jenner, A., 1991, ‘The Saxon and early medieval pottery of London’, in Vince, A. G., Aspects of Saxon and Norman London 2: Finds and environmental evidence, London and Middlesex Archaeological Society Special Paper 12, 19–119.

Wessex Archaeology 2019a, Otterham Quay Lane, Rainham, Kent. Archaeological evaluation report. Unpublished report 217440.04.

Wessex Archaeology 2019b, Otterham Quay Lane, Rainham, Kent. Palaeolithic geoarchaeological test pitting evaluation report. Unpublished report 217441.04.

Wessex Archaeology 2019c, Otterham Quay Lane, Rainham, Kent. Post-excavation Assessment and Updated Project Design. Unpublished report 217441.05.

Wymer, J., 1999, The Lower Palaeolithic Occupation of Britain (Volumes 1 and 2). Salisbury: Wessex Archaeology and English Heritage.