Normanton
Golf Course

apc

•••••

summary

contents

introduction

reconnaissance

mitigation

fieldwork procedure

• results •

sequence summary

discussion

conclusion

references

appendices

•••••

Printed from the APC web site: navigation and non-essential images removed.
Please view on-line for full content (URL at end of document).

Fieldwork Results: Zone 2

zone 1 | zone 2 | zone 3 | zone 4 | zone 5

Zone 2 was allocated to an area measuring 60m square which contained the rectilinear enclosure. The area was stripped of ploughsoil in 1997 and excavated in 1998. Zone 2 produced the most informative and significant archaeology on the site and as such became the main focus of the project.

The ploughsoil, which varied in depth between 0.2m and 0.4m, was removed by machine to reveal a variable subsoil. This changed across the site between an outcrop of weathered sandstone in the southeast of Zone 2, to an area of orange sandy clay in the northeast and to an area of blue orange plastic clay with exposed seams of degraded coal over the western half. Visible against these deposits was the clearly defined backfill of a substantial ditch which formed a rectilinear circuit (F37). This measured 42.7m (north-south) by 46.7m (east-west) and varied in width between 2.5m and 3.5m. The ditch was traversed by a series of seven linear furrows (F24 -F30 ) and fifteen ceramic land drains (F3 -F17) all running from north to south across Intervention 2.

Within the enclosure, and between the furrows, several features were identified. These comprised a number of post holes, curvilinear gullies and linear features which were concentrated in the centre and eastern half of the enclosure.

5.2.1 The Enclosure Ditch

The rectilinear enclosure ditch was by far the most impressive feature on the site. When first defined within Zone 2, this feature appeared to exist as a single entity, located upon a small rise in the valley bottom with in an otherwise clear area. Once the medieval furrows had been removed, it soon became apparent that the enclosure, at least in its latest phases, was tied into a more extensive field system with ditches extending from the northwest and southeast corner.

The ditch was sampled at selected points along its circuit. This comprised a hand excavated sample midway along each side of the enclosure in addition to a sample located at each corner. Due to the alignment of several internal features, the entranceway to the enclosure was expected to be on its eastern side. Consequently the ditch at this point was excavated in quadrant in order to recover additional sections through the backfill sequence.

This approach provided a total of twenty three drawn and photographed sections across the ditch from eight hand excavated segments. From these sections it was possible to identify five separate phases within the enclosure ditch. Deposits relating to these phases could be identified running around the complete circuit.

The eight hand-excavated samples were labelled according to their location as an aid to interpretation. These segments are shown in Figure 16 and the results of each are summarised in Table 3 below.

Table 3 Summary of contexts recorded within each excavated segment by ditch phase

Location Ditch	Entrance	SE Corner	SW Corner	W	N	NE Corner
Phase 1 - F37	C1269	C1288 C1289	C1331 C1333	C1310 C1311	C1324	C1318 C1319
Phase 2 - F234	C1267 C1268	C1287 C1290	C1330 C1332	C1273 C1274 C1309	C1322 C1323	C1317
Phase 3 - F43	C1168 C1170	C1286 C1291	C1328 C1329	C1307 C1308	C1067 C1321	C1315 C1316
Phase 4 - F235	C1163 C1166/C1169 C1167	C1284 C1285	C1327	C1272	C1066 C1320	C1313 C1314
Phase 5 - F236	C1164 C1165	C1038 C1283	C1038 C1325	C1038	C1038 C1325	C1038/C1312 C1325

Information recovered from the sections was also used to create hachure plans for each phase. Ditch sections ENTC and ENTN are shown in Figure 17, and sections WS and NECC are shown in Figure 18. The remaining sections recorded across the enclosure ditch are presented in Appendix D.

Phase 1 - F37 (Figure 19)

The original enclosure ditch was rectilinear in plan measuring 42m (north-south) x 46m (east-west) and varied in width between 2.5m and 3.5m. Due to recutting of the feature in later phases, F37 was only partially visible in section at various points around the circuit. Consequently the final form of this feature was difficult to determine. Only at the entranceway (ENTN) did both the internal and external ditch edges survive.

F37 had an 8m wide causewayed entrance located along its eastern side. The northern terminus of the ditch had a wide U-shaped profile, 3.3m across and survived to a depth of 1.4m. Cut into the concave base of this was evidence for a 0.45m wide channel which appeared to be related to drainage rather than having a structural purpose. The southern terminus for this phase was only discovered during the machine excavation of the enclosure ditch beyond the limits of the hand-excavated samples. Only the western edge of this butt end survived in plan.

F37 was backfilled with a homogenous deposit of olive brown clay containing few gravel inclusions and patches of oxidisation. In some areas around the circuit the interface between this deposit and the natural subsoil was poorly defined due to post-depositional weathering of the underlying bedrock. The final form of F37 was that of a 2.5m to 3.0m wide U-shaped profile ditch, with sides sloping between 45 and 75 degrees bottoming onto a flat or slightly concave base at a depth of between 1.2m and 1.5m.

At various points around the circuit of the enclosure (W/N/NE/ENT), narrow steep sided channels were recorded cut into the bedrock along the base of F37 (NECW/ENTN/SWCE/WS/WC2/NW). These channels were backfilled or lined with a deposit of grey plastic clay (C1311) or sandy clay (C1319) which contained rare flecks of charcoal in addition to very occasional pebble inclusions. The channels appeared to be related to drainage, acting much like gutters transporting water away from the entranceway toward the western length of ditch. This phenomenon was observed during periods of wet weather after the complete excavation of the enclosure ditch.

During a period of disuse the first phase of enclosure ditch at Normanton appeared to have almost completely silted up or been backfilled. Evidence for this is suggested from sections at the entranceway, southeast corner and along the western length of ditch circuit (ENTN/WN/SECS). At these points the backfill of F37 almost rises to the level of the present ground surface before the feature is recut by F234. This implies that the enclosure would have existed as nothing more than a shallow earthwork prior to being reestablished. Further evidence for this survived in the line of the later recut. Along most of the enclosure ditch the later cut followed the line of the original ditch often re-excavating the original drainage channels. Along the western length of ditch, however, the line of the recut veered so far off that of the original (WN) that it would appear that F37 was not visible at all at that time. This again suggests that the main purpose of the ditch was to drain water into this length of ditch which had apparently silted up completely. No datable material was recovered from the limited excavation of F37 where it survived.

Phase 2 - F234 (Figure 20)

As stated, the original enclosure was recut by a second ditch, F234. As with F37, F234 had undergone truncation by later phases of the enclosure and only partially survived in sections around the ditch.

F234 measured 42.6m (N-S) x 47m (E-W). With the exception of the entranceway this feature generally followed the same layout and line as F37. The causeway entrance for F234 was remodelled and was now located centrally along the eastern side of the enclosure and measured only 2.6m in width. Although no structural features were recorded within the ditch itself, several postholes and gullies within the enclosure may relate to some form of gate structure contemporary with this phase of ditch (see 5.2.3 below).

The butt ends of the enclosure were rounded in plan with a wide U-shaped profile. Like that of F37, the base of both of the termini were stepped with a channel, between 0.3 and 0.6m wide cut into the bottom of the ditch at a depth of 1.3m. Within the bottom of the southern butt end two large rounded cobbles were recorded. The channels, like those of F37, appeared to be for drainage as opposed to representing a structural feature. It is not clear, however, whether the cobbles, both of which measured in excess of 0.4m, were merely dumped within the ditch or served a more deliberate purpose.

F234 was backfilled with a fairly homogenous deposit of olive brown clay. This deposit could be followed around the length of the enclosure and generally contained few inclusions. These consisted of occasional charcoal flecks, manganese oxidisation, gravel and small pebbles. The final form of F234 was that of a U-shaped profile ditch, between 2.5m and 3.1m wide, cut to a depth of between 1.2 and 1.4m with a flat or slightly concave base. As with F37 lengths of a shallow gully had been excavated into the base of the ditch. These were located along the southwestern and part of the northern leg of the enclosure ditch cut into the natural bedrock and varied in width between 0.25 and 0.5m and in depth between 0.1m and 0.2m. Once again, these gullies appeared to be associated with draining water away from the entrance toward the western leg of the enclosure. F234 had reused some of the earlier drainage channels within F37 in addition to cutting new ones. As with the original ditch (F37) these gullies were backfilled with a deposit of grey plastic clay up to 0.2m thick which contained inclusions of rounded pebbles and charcoal flecks (ENTN/SECS/SWCE/SWCS/WS/WC2/NW).

The only anomaly within this backfill sequence was recorded mid way along the western leg of the enclosure (WS). At this point a laminated deposit comprising lenses of yellow clay, grey clay and slightly stained brown sand (C1275) was recorded tipping into the ditch from the internal side sealing the channel below (see Figure 18). Within this deposit frequent lenses of fine charcoal flecks and occasional flecks of fired clay were observed. The laminated structure of C1275 appeared to be the result of successive episodes of material washing into the ditch from inside the enclosure over a period of time. This deposit appeared to have accumulated before the main silting/backfill deposit of F234 had accumulated/been deposited.

C1275 was sealed by a thin patchy layer of black material (C1274) (WS). This deposit, which also had a laminated structure, appeared to be made up of very fine particles. A visual assessment carried out by Stephen Carter (Headland Archaeology) concluded that C1275 was in fact mineral in origin and appeared to be derived from material leaching into the ditch from the western cut edge originating from a weathered coal seam just below the present ground surface.

Despite extensive excavation no dating material was found within the backfills of F234. The only evidence for human occupation recovered from this phase of the enclosure ditch were the charcoal and fired clay flecks recorded in the drainage gullies and within C1274.

F234, like F37 appeared to have almost completely silted up or been backfilled before it was recut. Evidence for this survives in several of the sections recorded around the enclosure (WS/NW/NE/NECW/NECM/SECS/SWCW). This was particularly evident along the western leg (WS) where the section shows that this part of F234 would probably not have been visible at all before it was once again recut.

Phase 3 - F43 (Figure 21)

F43 was allocated to the third phase ditch within the sequence. This recut marked a distinct change in the form and development of the enclosure. F43 effectively transformed the causewayed enclosure of F37 and F234 into a complete circuit, cutting across the previous entranceway.

F43 followed the line of F234 fairly consistently but veered off its line dramatically at the northeast corner (NECC) (Plates 5 and 6) and along the western leg (WS/WC2) suggesting that the earlier ditch was not visible at these points. F43, although truncated by later ditch phases, survived sufficiently to characterise its profiles and backfills. The final form of F43 was that of a U-shaped profile cut, between 0.8m and 1.2m deep with a flat or slightly concave base and sides that sloped between 60 and 70 degrees. The top of F43 was marked by a general widening of the feature visible in most sections around the circuit. This resulted in the recorded width of the feature varying between 1.7m (WS) and 3.5m (ENTS). Several short lengths of drainage gully were identified cut into the base of F43 in the southeast corner and along the northern leg of the circuit. These gullies tended to be have a V-shaped profile and were less well defined than those recorded in earlier phases. In plan the enclosure ditch measured 41.3m (N-S) x 46.3m (E-W).

The entranceway was retained midway along the eastern leg of the enclosure circuit but the narrow causeway was replaced by a bridged crossing, the support posts for which were identified on the western side of the ditch (F212, F213, F214, F217) and were abutted by a amorphous cobble surface (F206).

F43 was backfilled with a fairly homogenous deposit of grey to greyish brown silty clay which could be followed around the circuit of the enclosure through the excavated sections. This deposit contained inclusions of charcoal flecks, fired clay flecks, gravel, pebbles and cobbles which varied in distribution around the ditch.

The base and sides of F43 was defined by a thin layer of plastic, greasy, grey clay which contained occasional flecks of charcoal and varied in depth between 0.05m and 0.2m. This deposit also contained moderate inclusions of rounded pebbles and cobbles which varied in size between 0.02m and 0.2m. Many of these stones appeared to be reddened either through natural staining or burning, most of which were concentrated within the drainage trenches cut into the base of F43. Significantly, the higher concentrations of cobbles occurring in the base of F43 were found in the entranceway and southeast corner of the enclosure.

Within the entranceway the lower clay fill (C1170) lined the cut edge of F43 against the bedrock where the original causeway had been cut away. This provided an important indicator of the interface of F43 and earlier deposits around the rest of the enclosure.

Only eleven fragments of pottery were recovered from the excavation of the backfill of F43. From the southeast corner (SEC) a rim sherd and body sherd from a small jar were found within C1286. In addition to this a very abraded sherd made of an orange-grey sandy fabric was also found. This pottery has been dated to the Late Iron Age or early Roman-British (Appendix E and Appendix F respectively), although this is not certain due to its poor condition and size. The black sandy fabric with quartz inclusions is suggested as having parallels at Stanwick (see Appendix E).

Along the northern length of circuit (Plate 7 and Plate 8) three similar sherds were recovered from the lowest grey clay fill of F43 (C1321) (N). It is possible, however, that these are intrusive from a later ditch phase (F235). From the later backfill (C1067) within the northern ditch segment, a further five pottery fragments were recovered, four of which were part of the same vessel. These fragments were processed as Thermoluminesence samples and, as such, were carefully recovered and located. These sherds (fabric A, Vyner) were positively identified as being pre-Roman Iron Age in date (see Appendix F). Unfortunately the fabric of this pottery was not suitable for Thermoluminesence dating and the analysis was unsuccessful.

A single cow tooth was recovered from C1168 in F43.

F43 like its predecessors appears to have virtually silted up before the ditch was recut. This is evident in several sections around the ditch circuit (SWCC/SWCW/WN/NW).

Phase 4 - F235/F78 (Figure 22)

With the exception of the southwest corner (SWC), F235 recut the partially backfilled enclosure of F43 along the same line around the complete circuit. This latest phase of ditch measured 41.7m (N-S) x 46.1m (E-W). In plan the western length of F235 was considerably narrower at only 1.7m than the other sides of the ditched enclosure.

The final form of F235 was that of U-shaped cut which varied in width between 1.8m and 2.8m and in depth between 0.7 and 0.9m. The ditch had sides that sloped between 60 and 70 degrees which bottomed into a flat or slightly concave base through with a shallow channel cut into its base at various points around the circuit (SWCE/WN/NE). F235 was generally 0.2m shallower than F43. As with F43, F235 widened considerably at the top giving it a flared profile in places.

F235 was backfilled with a deposit of heavily oxidised grey clay. The composition of this varied around the circuit reflecting localised geology but generally contained occasional inclusions of charcoal flecks, gravel, pebbles and cobbles and fired clay flecks. Some of the stones were burnt and fire-cracked. Within the entranceway, this deposit contained far more iron-pan or manganese concretions than elsewhere around the enclosure circuit implying that this area had been saturated with water for a considerable period of time.

The cut edge and base of F235 were lined with a deposit of grey plastic clay. This deposit varied in depth between 0.05m and 0.2m and contained a moderate amount rounded cobbles, pebbles and charcoal flecks.

F235 appeared to be linked to a second ditch to the east of the main enclosure (F78). F78 was aligned east-west and appeared to feed into the main circuit immediately to the north of the bridged entranceway. In plan F78 ran for 25m beyond the limits of Intervention 2, before turning south at 90 degrees. After this point it continued for a further 12m before butt ending in a wide shallow scoop. A study of the levels showed that rather than feeding into the enclosure, F78 served to take excess water away from it, acting as an overflow. At its western limit, where F78 meets F235, the base of F78 was recorded at a height of 28.55m AOD. Where excavated at the edge of Intervention 2, 17m to the east, the level of the base was 28.26m AOD and a further 4m east was 27.95m AOD. The relationship between F78 and the enclosure ditch sequence was captured in a working section located 1.5m north of main entranceway quadrant.

The levels in the base of F235, when studied, also showed a slight change in the pattern of drainage from previous phases of the ditch. The circuit of F235 appeared to drain water from the southeast and northeast corners of the enclosure back towards the entranceway as opposed to the western leg. This had significant implications for the location of the entranceway during this phase. The close proximity of F78 to the old entranceway, suggested that the bridged crossing of Phase 3 would not have fitted into the same place in Phase 4. It is possible that a new crossing, without supporting posts, was positioned to the south of F78 to bridge a gap of 2.2m into the enclosure. Alternatively, the location of the entranceway during this phase may have shifted to another point around the circuit, although little evidence for this has been found.

Until this point the central location, symmetry and formality of the entranceway in Phase 2 and 3 had been mirrored in the layout and organisation of internal structures. The shift in the entrance during this phase may have marked a change in the emphasis and function of the enclosure.

Only a single sherd of pottery was recovered from F235, and was found during the excavation of the southeast corner of the enclosure (C1285). This fragment comprised an abraded sherd of black sandy fabric of the same type found in C1286 and was considered to be of a Late Iron Age/early Romano-British date.

In its disuse F235 appears to have silted up before being recut by the final phase ditch (F236). This is evident in several of the sections recorded around the circuit (NECS/ENTS/SWCE/WN/NE). Whereas F235 appears to have silted up through natural processes over time, the backfill of F78 appears to suggest that this overflow ditch was deliberately backfilled with redeposited subsoil (C1088) before the enclosure was recut.

Phase 5 - F236 (Figure 23)

The final phase of ditch of the enclosure comprised not only the recutting of the entire circuit but also the incorporation of the enclosure into a wider field system. Two north-south aligned ditches were cut running from the northwest (F220) and southwest (F219) corners of the enclosure to create a western boundary or drainage ditch some 160m long (Zone 4). This activity appears to be contemporary with the deliberate backfilling of the western end of F78 (Phase 4). Once backfilled, this feature was then recut along its eastern length (F34). The recut (F34) had a western butt end which left a passable gap of 4.2m between it and the enclosure (F236). This created a system of boundaries which transformed the isolated entity of the enclosure into an integrated element of a wider landscape.

F236 measured 41.7m (N-S) by 45.9m (E-W) and recut F235 along the same line. This feature had a variable U-shaped profile which ranged in width from 0.9m along the western leg of the circuit, 1.3m along the north and 2.1m around the southeastern corner. F236 varied in depth between 0.4m (WS/WN/NECW/SECN/SECW) to 0.7m (ENT) and had sides that sloped between 30 and 70 degrees into a shallow concave base. Unlike earlier phases of enclosure, the recorded profiles and levels of F236 suggested that the emphasis of the ditch lay not in managing water around the circuit but more in defining it as a physical component within the landscape.

F236 was backfilled with a variable deposit of grey sandy clay containing occasional inclusions of rounded pebbles, cobbles and gravel. Charcoal flecks were only recorded as rare inclusions within this deposit in the southeast and southwest corners of the enclosure. The base and sloping edges of F236 were marked by a thin a lens of plastic grey clay in only the entranceway and southeastern corner. Elsewhere the backfill consisted of a single deposit.

A formal entranceway or bridging point was not identified during the excavation of F236. The deliberated backfilling of F78 to the east of the enclosure may suggest that some form of crossing may have been reinstated at the original entrance. However, the enclosure ditch could have been crossed fairly easily by both people and animals due to the shallow depth and narrow width of the ditch.

Two fragments of pottery were recovered during the excavation of F236. The first comprised a fragment of a Late Iron Age/early Roman-British jar found in the southeast corner (C1283). The second, the base of a Roman greyware jar, was found in the northern excavated sample (C1326) and was dated to between the 2nd or 3rd century AD. A heavily corroded bronze finger ring was also recovered from this deposit (Find No.316, Plate 9).

In its disuse F236 appears to have silted up. From this point onwards the enclosure may have only survived as a shallow earthwork.

General

No evidence for an internal or external bank was recovered during the excavation of the enclosure ditch. Only at two points (SWC/WN) did deposits indicate tips of material from one side of the ditch or the other. In the southeast corner, tip lines of stones were observed within the backfill of F235 (C1284). These pebbles and cobbles appeared to originate from outside the enclosure. The deposits recorded along the western leg tipping into the ditch from inside the enclosure (WS) are discussed in Phase 2 above (C1275). If a bank was present there is no reason to suggest that once it had become established with grass or vegetation that it would erode away.

The sheer lack of material evidence recovered from the Normanton enclosure ditch was in itself quite striking. On many Iron Age sites such features often acted as deposit traps collecting domestic refuse or ritually deposited items, particularly at the entranceway of an enclosure. A 44% hand-excavated sample of the enclosure ditch produced a total of 14 fragments of abraded pottery (Plate 10), a bronze finger ring, two cow teeth and cattle metacarpal. Whether this paucity of material is due to finds not being discarded into the ditch in the first place or is a result of a poor burial environment is an important issue for consideration.

The assumption that poor representation of skeletal elements within the enclosure ditch was a product of poor preservation conditions, was supported by the types of bone that have survived (Appendix G). In this case the two teeth (due to their enamel) and one metacarpal (due to its dense structure) were much more likely to survive than other types of bone in this context. If this was the case, and activity was occurring on the site to produce an animal bone assemblage which included cattle bone, why are their so few bones of these types represented in the archaeological record considering the level of sampling employed during the excavation?

It can also be argued that the burial environment is responsible for the poor survival of some types of low fired, coarse pottery, particularly fabrics with grass or calcite temper which can leach or dissolve in acidic soil conditions. However, the burial environment cannot be held responsible for the lack of other artefacts such as stone objects, querns or flint.

Bulk samples were taken from various deposits within the ditch. Without exception the assessment of these samples indicated that nothing of an organic nature had survived the burial environment of the ditch. Only tiny quantities of charred wood and charcoal were recovered during processing.

5.2.3 Internal Features

After the machine stripping of Intervention 2 a number soil features were identified within the enclosure cut by medieval ridge and furrow cultivation. Once the furrows had been excavated, the whole area was cleaned by trowel and these features were mapped and tagged. In accordance with the scheme of works every feature identified was then 100% excavated. Work on site was intermittent due to poor weather and other commitments within the development programme.

The internal layout of the enclosure included four centrally located crescentic gullies. Associated with these were a number of post holes and a single sub-oval pit. Running eastward from the centre of the enclosure were two gullies which appeared to funnel towards a point midway along the eastern length of the enclosure ditch.

Due to the lack of stratigraphic relationships and the paucity of dateable finds, the phasing and sequence of the internal components of the enclosure had to be derived from radiocarbon dates. These have been considered in combination with a comparison in the Munsell colour, inclusions and matrix of feature backfills. In this respect the results of the excavation will be presented first before the interpretation is considered.

For this purpose the enclosure has been divided into four areas. These have been defined according to their location and/or logical groups of features or structures within them and are shown in Figure 24. Within each area features will be discussed in stratigraphic order where relationships exist.

Table 4 Summary of areas within enclosure

Area	Features
EAST	F35, F36, F90, F91, F92, F96, F192, F195, F209, F210, F211, F212, F213, F214, F215, F216, F217, F232
CENTRE	F39, F40, F41, F42, F54, F57, F58, F59, F66, F104, F150, F152, F153, F156, F157, F159, F160, F164, F167, F168, F173, F178, F184, F188, F190, F191, F194, F196, F197
SOUTHEAST	F 84, F87, F88, F89, 96, 97, 98, F99, F100, F189, F193
GENERAL	F67, F103, F105, F106, F107, F108, F168

EAST (Figure 25)

The eastern area of the enclosure was divided into two main parts. The first comprised an apparent corridor formed by two lengths of east-west aligned gully, F35 and F36, which led from the centre of the enclosure to the midpoint of the eastern ditch length. Both of these features were interrupted by medieval furrows which gave them a broken appearance in plan. Associated with these gullies were a number of postholes. To the east of this corridor, the second part of the eastern area consisted of an entranceway. This was only fully defined after the removal of furrow F30 and the backfill of the latest enclosure circuit (F236). The entranceway was made up of a series of postholes and a cobble surface which related to a sequence of gate structures and surfaces respectively. In plan these two elements appeared to be linked thereby creating, at least in one phase, an elaborate gate and corridor into the enclosure.

The earliest feature excavated within the eastern area comprised a truncated scoop (F232) which was identified during the excavation of gully F35. This feature was defined in an area which had been disturbed by a medieval furrow. F232 was a poorly defined shallow pit sub-oval in plan approximately 0.5m x 0.5m and 0.1m deep. This was filled with a deposit of grey silty clay which contained large concentrations of charcoal in addition to inclusions of gravel and pebbles (C1303). Mixed randomly throughout C1303 were fragments of calcined bone which suggested the presence of a possible cremation. As such the deposit was 100% sampled for analysis.

A total of 10.2g of bone was recovered from C1303 during flotation. Only 4.5g of this could be identified and this comprised four fragments of the right tibia of a sheep (see Appendix G). The remainder of the bone was too small or fragmentary to be of any analytical value. The majority of the bone was calcined and must have been cooked at temperatures of between 500 and 800 degrees celsius to have been transformed into this state.

Originally the excavator concluded that C1303 appeared to be within the backfill system of gully F35, but further analysis concluded that this was not the case and that F232 was earlier in date. After excavation it became clear that the full extent of F232 was outside that of the later gully and that the interface between the two backfills was not clear due to similarities in colour and a degree of mixing, probably caused when the F35 was originally excavated. A radiocarbon date from charcoal recovered from C1303 resulted in a date of between 1206BC and 917BC for the cremated bone (Appendix H; all dates are expressed as calibrated date ranges of 1 sigma level of confidence).

A series of four postholes (F90, F91, F192, F195) were identified running along the same east-west alignment as gully F36. Two of them (F90 and F91) were clearly truncated along their northern edge by the later gully.

F90, located 2m to the south of F232, comprised an elongated post slot measuring 0.4m by 0.15m. Upon excavation it proved to be 0.15m deep with sides sloping at 50 degrees into a concave base (Figure 26a). The backfill of this feature (C1260) was made up of a deposit of greyish brown sandy clay containing frequent small sandstone inclusions in addition to moderate charcoal flecks. F91, 2.1m east of F90, was backfilled with essentially the same material (C1261). Upon excavation this feature comprised a shallow scoop 0.4m wide by 0.27m long surviving to a depth of a mere 0.05m. Within the base of F91 three well defined stakeholes were excavated.

Two further postholes were identified and excavated further to the west along this alignment. F195 and F192 were identified at the western terminus of F36. Their relationship with the gully could not be tested as F36 appeared to merely peter out at this point. F192 consisted of a circular cut, 0.24m in diameter with a U-shaped profile, and was cut to a depth of 0.22m (Figure 26b). This feature was backfilled with two deposits (C1245 and C1246), the later of which (C1245) contained frequent mudstone inclusions and appeared to relate to the levelling of the feature in its disuse. F195, located 0.5m to the east, was of a similar form and measured 0.2m in diameter and 0.2m deep (Figure 26c). F195 was backfilled with a deposit of sandy clay (C1251) similar to that recorded in F90 and F91.

This series of four postholes appear to have been part of a structure or fence which predates the gully F36. From the surviving evidence this structure would have been at least 7.5m long but was more likely to have extended the full 12m to the entranceway. It is possible that the eastern limit was marked by F210. This comprised a shallow, slightly irregular scoop, 1.1m by 0.6m, cut into the subsoil and backfilled with a deposit of greyish brown clay sand (C1276) which contained frequent inclusions of sandstone and mudstone fragments. Upon excavation F210 proved to be only 0.05m deep with steep sides and a flat base, through which four stakeholes were defined and excavated. It is possible that F210 may have been merely an area of disturbance, buts its location and spatial relationship with the post alignment suggest that they may have been contemporary features.

The funnel-shaped corridor identified in plan within the eastern area was formed by two lengths of gully, F35 and F36. The possibility that these features were not contemporary and represented individual divisions of the enclosure at separate times has been considered. However, it is far more likely that they formed part of the same structural element. The rationale for this lies in similarities recorded in their alignment, backfill and dimensions. The main factor in arguing for their contemporaneity lies in the symmetrical arrangement of the gullies with a gateway structure represented by two pairs of postholes (F211, F216 and F209, F215) at their eastern end. Within this arrangement F35 clearly terminates 0.5m to the west of posthole F215, and F36 similarly terminates to the west of F209.

On excavation F35 proved to be a 7.7m long U-shaped gully which ranged in width between 0.6m and 0.4m (Figure 27). F35 varied in depth between 0.1m and 0.2m, with edges that sloped between 25 and 40 degrees and was backfilled with three separate contexts. The latest of these (C1263) consisted of a compact deposit of mottled dark greyish brown sandy clay which contained moderate inclusions of rounded and angular pebbles, cobbles and sandstone fragments. C1263 was excavated along the whole length of the gully and represented a consistent deposit relating to the disuse of the feature. The second fill (C1264), however, was only identified within the central 3.5m of the gully. This deposit comprised a dark grey silty clay with frequent charcoal flecks. At various points a patchy deposit of plastic grey clay was identified within the base (C1265) but this only survived to a depth of 0.02m.

The western end of F35 was completely truncated by a later furrow (F28). At its eastern terminus, the gully appeared to become more shallow then terminate in an ill-defined scoop. This appeared to be due to disturbance created by the remodelling the entranceway at a later date and the construction of a cobble surface (F206).

Although slightly narrower than F35, F36 (0.4m wide) was similar in form (Figure 28). This gully lay at a slightly different angle to F35 and was 2.5m to the south of F35 at it western end and 1.5m to the south of F35 at the entranceway, thus creating the funnel effect. Upon excavation F36 comprised a U-shaped cut, 10.5m long which varied in depth between 0.1m and 0.15m with a shallow concave base. F36 was backfilled with two deposits. The latest (C1244), like C1263 (F35), was a deposit of dark greyish brown sandy clay with charcoal flecks and moderate stone inclusions, some of which appeared to have been burnt. This deposit sealed an earlier backfill at the eastern end of F36 (C1262) of very mottled silty clay similar in composition to C1264 found in F35.

Despite the fact that no evidence for postholes or beam slots was recorded within the base of the gullies, it is assumed that F35 and F36 would have held some form of timber fence, palisade, or even supported a covered structure. A study of the levels recorded in the bases of the excavated features indicates that they were not used for drainage. The western end of F35 is of the same level as its eastern terminus, and, in the case of F36, the height of the base varies along its length and is 0.05m higher at the entranceway than its western limit. This also suggests that the eastern limits of these features were truncated during later phases of the entranceway.

Six pieces of pottery were recovered during the excavation of F36. Of these, three were of a size and condition that was of little use for dating purposes. The remaining sherds were recovered from the latest backfill (C1244) and were of an Iron Age date (see Appendix E and F, Vyner - Fabric B).

At the eastern end of the funnel corridor, two pairs of postholes appeared to represent the remains of a timber gate structure or structures. This left an opening or gateway of between 1.2m and 2.2m. The first pair were made up of two sub-oval post pits (F216 and F211) both aligned north-south, lying parallel to the enclosure ditch. F216 measured 1.7m by 0.7m and upon excavation proved to be 0.4m deep. The final form was that of a U-shaped cut with sides sloping between 50 and 80 degrees into a slightly concave base (Figure 29a). The backfill (C1293) comprised a mottled deposit of heavily oxidised olive grey sandy clay with frequent charcoal flecks. A moderate amount of angular and rounded stones, some of which appeared to have been burnt, were found concentrated in the southern half of the feature. Although a post pipe could not be identified in the recorded sections, these cobbles were recorded in plan and may represent the backfill of such a feature. If this was the case, F216 would originally have held a post of around 0.6m in diameter.

F211 was located 4m to the south of F216. This feature measured 1.5m by 0.8m and was backfilled with a deposit of mottled brownish grey sandy clay (C1278). This material appeared to be a redeposited natural clay which contained moderate inclusions of angular and rounded pebbles. These stones, like those in F216, were concentrated at one end of the feature, thus suggesting the location of a possible post pipe 0.6m in diameter. The edge of F211 was defined against bedrock along its base and sides and was 0.4m deep when excavated, with sides sloping between 60 and 80 degrees (Figure 29b).

F209 and F215 were located immediately adjacent to F211 and F216 along the same north-south axis. F215 (Figure 29d) measured 0.9m by 0.6m and when excavated proved to be 0.3m deep. The primary fill of F215 (C1292) comprised a slightly dirty redeposited subsoil which contained a number of burnt cobbles within its base. This deposit appeared to be post-robbing backfill. The top 0.05m of the posthole was backfilled with a mixed deposit of brown sandy clay (C1468). This material was present in patches in around the entranceway (1301) and appeared to represent a layer of trample or disturbed ground.

F209 was located 1.8m to the south of F215. The top of F209 was defined by a cluster of large cobbles which, when removed, revealed a setting of four sandstone blocks set within a matrix of olive grey sandy clay (C1271). These were positioned along the western edge of the feature. This deposit appeared to represent the remains of packing associated with a timber post that would have originally measured 0.45m in diameter. When fully excavated the profile of F209 was a slightly irregular U-shape with its base lying at a depth of 0.4m (Figure 29c). The eastern edge sloped between 30 and 45 degrees while the opposite side was slightly undercut behind the sandstone packing. This undercut was filled with a deposit of mottled plastic olive grey clay with frequent charcoal flecks (C1277) from which a single cattle molar was recovered. Judging from its matrix and composition C1277 may have had a relatively high organic content at the time of its deposition. A functional explanation for the position of this material behind the packing of a posthole seems unlikely. The lack of post packing along the eastern cut edge of F209, considered with the flared irregular profile suggests that the post was levered, or dug out, from this side in its disuse.

Both F209 and F215 were positioned immediately adjacent to, and along the same axis of F211 and F216 respectively. Stratigraphically, F209 was recorded as cutting F211 by one or two centimetres along its northern edge. It is not clear, however, whether this relationship represents the robbing of F209 or the construction of the posthole itself. There are therefore two possible interpretations of this arrangement. Firstly, that all four postholes are contemporary creating a large double posted gateway structure, or, secondly, that the two sets of postholes represent separate chronological phases. If these represent two phases in the gateway structure then earliest would have been marked by F211 and F216. The second phase would have consisted of F209 and F215 tied into the corridor formed by F35 and F36.

No datable finds were recovered from any of these features. Charcoal recovered from the flotation of C1271 (F209) was radiocarbon, which resulted in a date of between 402BC to 265BC (see Appendix H).

A third set of posthole features was identified in the entranceway cutting F216. The location and alignment of these features (F212, F213, F214, F217) suggests that they were constructed after the corridor (F35/F36) and its related gateway (F209, F211, F215, F216) went out of use.

F212 and F217 were fully defined after the removal of cobble surface F206. F212 comprised a sub-circular cut 0.7m in diameter and 0.55m deep. This feature had a U-shaped profile, flat base and sides that sloped between 70 and 85 degrees (Figure 29e). F212 was backfilled with a deposit which consisted mostly of large rounded cobbles and sandstone fragments set in a matrix of dark grey sandy clay (C1280). Some of the stones appeared to be burnt. Due to the nature of this material it was not possible to record a section across this deposit. The cobbles within F212 appear to be a result of a deliberate backfilling episode as opposed to some form of packing. Immediately to the east of F212 were two small postholes (F213 and F214). This pair of features appeared to form a deliberate post setting, 0.4m apart, relating a possible structure formed by F212 and F217. F213 and F214 comprised two sub-oval cuts 0.25m by 0.15m, cut to a depth of 0.15m. In their disuse both features had been filled with cobbles and clay during the construction of a later surface.

F217 was located 2.6m to the south of F212. F217 was a sub-circular cut truncated by a later field drain along its eastern edge. Upon excavation F217 proved to be 0.6m by 0.4m and cut to a depth of 0.4m (Figure 29g). F217 had near vertical sides and a flat base and was backfilled with a deposit of grey sandy clay (C1294) with frequent cobbles and pebbles, some of which appeared to have been burnt.

F212 and F217 appear to represent the remains of another gate structure which went out of use before the creation of a cobble surface over the entranceway. It is unlikely that this arrangement was contemporary with a recutting of the enclosure circuit, but if so, it may have held the supports for a bridged crossing as opposed to the uprights for a simple gate. The opening between F212 and F217 would have been 1.5m wide.

A patchy spread of mixed, heavily oxidised dark grey sandy clay (C1301) was identified sealing the postholes in their disuse. This deposit appeared to represent a layer of trample or disturbed ground in the area of the entranceway.

C1301 was covered by an area of cobbles which were fully exposed after the excavation of the latest phase of the enclosure circuit (F236) (Plate 11 and 12). This surface (F206) measured 4.7m (N-S) by 3.3m (E-W) and was located on the western side of the enclosure ditch covering the top and lining the western sloping edge of the feature (Figure 30). The stones (C1158) within the surface ranged in size from fine gravel to moderate sized cobbles (up to 0.4m in diameter). An estimated 95% of these were of made of sandstone, while the remainder were of granite or other igneous cobbles. Around 25% of the stones within F206 exhibited signs of having been burnt. This was evident in reddening and cracking of several of the sandstone fragments and cobbles.

The surface appears to have been constructed by pushing or ramming the cobbles into the underlying clay. In some areas the surface was more than one layer of stones thick and small areas of tumble were identified where material had slipped into the ditch. The fact that F206 lines the western edge of the enclosure ditch suggests that it was contemporary with at least one phase of continuous ditch circuit. As such, one possible purpose for F206 would have been to support some form of bridged crossing. Evidence for such a structure is difficult to find. Only a single posthole (F207) was identified as being contemporary with the cobble surface. This comprised a sub-rectangular socket 0.25m by 0.2m and 0.2m deep cut into the western sloping edge of the enclosure ditch and lined with a series of flat sandstone pieces set on edge (Plate 13). The socket void was filled with a deposit of grey silty clay (C1266) very similar to the latest backfill of the enclosure (C1161, F236). A similar feature was excavated on the eastern slope of F43 (F208).

Three additional postholes were identified within the eastern area of the enclosure. F92, F96 and F100 appeared to form a SW-NE alignment of associated features. F92 comprised a shallow U-shaped cut, circular in plan, 0.36m in diameter backfilled with a deposit of dark greyish brown sandy clay (C1253). This feature was only 0.08m deep with a flat base and sides that sloped at less than 30 degrees. Situated 3.8m to the southwest, F96 had a similar profile. This feature was 0.55 in diameter, 0.08m deep and backfilled with the same material (C1238). The third posthole (F71) was 0.6m in diameter and 0.12m deep. The backfill of this feature (C1252) contained a large concentration of angular sandstone fragments, some of which were burnt, within its grey sandy clay matrix. These three features appear to form a possible boundary fence or structure.

CENTRE (Figure 31)

The central area of the enclosure was characterised by a series of four crescentic gullies which appeared to be associated with a number of postholes and a large pit. Despite the density of these features, there were very few stratigraphic relationships which could be used to create a sequence. Where important stratigraphic relationships would have existed they had been truncated by later activity. Furrow (F26) was the most destructive in this respect. This single feature had obliterated the western limits of three of the gully features (F39, F59 and F159) at a point where they would have intersected. As well as destroying any stratigraphic relationships, the truncation caused by F26 also meant that none of the four gullies survived in plan in their entire form.

The function of these gullies is worthy of consideration. In plan none of the features appeared to form a true semi-circle or part of a circle. Each had a wide open end and the curve of each arc was flattened along its long axis giving it the appearance of half a distorted oval. Two of the gullies (F39, F159/F38) had an open side which faced towards the southeast. Of the other two, the open end of F66/F188 faced south while that of F59 faced toward the northeast.

Originally it was thought that the southern arc (F59) would join with one of those to the north to form a complete round house structure. However, on the basis of shape, feature backfill and dating evidence it was concluded that each gully existed as a separate entity. The argument that each gully represented one half of a round house and that the opposite half had been truncated by later activity was also unlikely.

Table 5 Summary of component features of Structures 3-6

Structure	Features
S3	F38, F150, F157, F159, F42, F184
S4	F59, F41, F54, F58
S5	F57, F66, F168, F172, F187?, F188, F191, F194, F197
S6	F39, F104, F173, F187?, F190

The structures are presented chronologically. The earliest feature within the central area comprised a sub-rectangular pit or posthole truncated along its northern edge by gully F159 and to the south by posthole F156. Definition of F153 was problematic due to the nature of the underlying subsoil. It comprised a steeply sided cut 0.2m deep filled with a deposit of firm mottled brown sandy clay (C1107) which contained frequent inclusions of sandstone fragments and rounded cobbles and pebbles. Many of these cobbles, particularly those visible on the within the upper part of the fill, appeared burnt and fire-cracked. The western limits of this feature were difficult to determine with certainty. The eastern sides were well defined and appeared square in plan. Charcoal recovered from C1107 during excavation provided a radiocarbon date range of 1767BC to 1642BC pushing F153 into the Bronze Age (see Appendix H). If this date is reliable it has significant implications regarding the history and development of the site. If not, the location of F153 would fit neatly as a post hole across the arc of Structure 5 (S5).

Structure 5 (S5)

Structure 5 consisted of two short lengths of gully (F66 and F188) which formed the northern wall, c.8m long, of a squat building or screen. The eastern length terminated in a large posthole (F57) while the western end of the structure was cut by a later structure (S6-F39). The western part of S6 had been completely truncated by furrow F26.

At its western end F66 had a distinctive U-shaped profile, 0.25m wide, 0.2m deep, with sides that sloped at 80 degrees into a flat base which was well defined against the natural clay subsoil (Figure 32a). At its eastern end F66 became much shallower (0.1m deep) with a wider flared profile(Figure 32b). This change occurred where the underlying subsoil changed from clay to bedrock and was marked by the presence of a sub-oval posthole (F194) cut into the base of the gully. F194 appeared to be the socket for a post measuring 0.25m in diameter which was backfilled with the same material as F66. To the east of F66, after a break of 0.6m, the line of the arc continued with a second length of gully (F188). Over bedrock this feature survived as a vague stain which, when excavated, filled a narrow V-shaped channel 0.1m deep. Further to the southeast the section across F188 showed a shallow 0.4m wide U-shaped cut with evidence for a narrow slot 0.1m deep and 0.1m wide (Figure 32c) cut into its base. Within the backfill of this feature (C1233) flecks of charcoal and small fragments of daub were recorded. Where F188 turned south to join with posthole F57, the gully had returned to its steep sided flat based form as recorded in the western half of F66.

Located along the inside edge of F66 and F188 were a series of three postholes (F187, F191, F197). F187 appeared to be the remains of two possible phases of post structure and at this point and measured 0.5m long by 0.35m wide and was cut to a depth of 0.25m. F191 and F197 were both cut to a similar depth with steep sides and flat bases.

The posthole (F57) located at the southern end of gully F188 was sub oval in plan and measured 1.2m x 0.8m. When excavated this feature appeared to have several phases of use. The earliest form appeared to be a flat based post pit 0.45m deep (F77) (Figure 32d). The original cut for F77 was badly disturbed by later phases of posthole and its limits in plan were difficult to establish. As such it is assumed to have been at least 0.5m in diameter. F77 was backfilled with a deposit of greyish brown clay and charcoal flecks (C1073).

At some point the base of F77 was cut by three separate undercut post-sockets or postholes (F73, F74, F75). These features were on average 0.15m in diameter. F73 was cut into the southern edge while F74 and F75 were cut and aligned along different angles (NW and SE respectively) into the northern edge. The form of these features indicated that the posts within these sockets were not set vertically. All three were angled inwards at between 10 and 18 degrees towards the centre of the posthole (Figure 32). This unusual arrangement seems to indicate that these sockets were placed to brace a central post possibly as some form of repair.

The latest use of this feature was marked by a well defined post pit (F57). This feature was cut through F73 to F75 to a depth of 0.4m. The base of F57 was marked by a thin layer of charcoal which dished into the robbed post-sockets. Located centrally within F57 was a well defined post void which was vertically sided, 0.3m in diameter, 0.3m deep and backfilled with a very clean deposit of dark grey sandy clay (C1058). Located centrally within C1058 was a single large sandstone block. The main backfill of F57 comprised a deposit of mottled olive brown silty clay (C1060) which was similar in composition to the backfill of F188 to the north.

Two small postholes (F172, F168) were identified forming an east-west alignment with F57 across the open face of Structure 5. Although well defined in plan F168 was only 0.03m deep when excavated. F172 was of similar dimensions measuring 0.15m in diameter by 0.08m in depth.

The evidence from F57 suggests that S5 may have been in use for a considerable period of time. The post pit indicates at least two phases of structure, possibly with an attempt to repair the earliest one towards the end of its life by bracing an existing post. This notion of repair or reuse is mirrored in the line of internal postholes (F187, F191, F197). These features may represent an earlier structure or possibly an attempt to consolidate an existing one. Although no evidence for in situ burning was identified within the features which made up Structure 5, the lenses of charcoal within the base of F57 may indicate an episode of clearance by fire.

No pottery or artefacts were recovered from any of the features within Structure 5. A charcoal sample from F57 (C1058) provided a radiocarbon date range of 398BC to 261BC (see Appendix H).

Structure 6 (S6)

Structure 6 was the most complete of all the structures identified within the enclosure. It comprised a crescentic gully (F39) which had an arc 4.5m wide and an open face 9.5m long. This structure was superimposed over the top of S5 with gullies overlapping at their western end. Although S5 and S6 appeared to have some similarities, the open face of S6 was aligned to face the southeast.

F39 comprised a well defined U-shaped cut, 0.4m wide and 0.3m deep with a flat or very slight concave base. With the exception of its eastern end, this feature had a consistent profile along its 13.0m length (Figure 33b and 33c). At the eastern terminus F39 became increasingly shallower and was cut to a depth of only 0.1m (Figure 33a). Within the terminus itself a sub-rectangular deposit of grey clay could be identified, 0.2m square, marking the position of a possible post-socket.

The latest backfill of F39 (C1040) was made up of dark greyish brown to grey sandy clay. Where F39 was cut through bedrock the backfill was of a more sandy composition. C1040 contained a large amount of angular sandstone gravel and pebbles throughout its matrix in addition to a frequent number of rounded cobbles. Many of these stones appeared burnt and fire-cracked. C1040 was contained within a vertical or steep sided channel located along the southern cut edge of the gully and was visible in section along the length of F39. This channel varied in depth between 0.1m and 0.2m and in width between 0.15 and 0.2m and appears to correspond with a backfilled post trench. The earlier fill of F39 comprised a deposit of heavily mottled, variable grey clay with inclusions of orange clay and sandstone fragments.

The open arc of F39 was closed by a series five postholes (F60, F104, F156, F173, F156) which could be attributed to S6. These were aligned SW-NE and included the post-socket identified in the eastern terminus of F39. F60 was sub-rectangular in plan, 0.4m by 0.2m and only 0.1m deep. F187 appeared to be reused from S5. F104 was a well defined sub-circular posthole 0.35m in diameter, 0.25m deep with a flat base (Figure 33e). Within its backfill (C1151) several burnt cobbles were recovered. F156 was of a similar size and form to F104 while F173 survived as a heavily truncated scoop only 0.05m deep (Figure 33d).

F190 was located 0.4m to the south of F39. It comprised a heavily truncated posthole 0.3m wide which had been cut centrally by a land drain leaving less than half the feature intact. This posthole, which may have been associated with S5, was 0.15m deep, with sides that sloped at 60 degrees into a concave base.

The sections from Structure 6 provide strong evidence that these gullies were structural and were designed to hold a series of upright posts, either as a screen or wall for a small hut. No finds were recovered from the excavation of any of these features. Charcoal from C1040 (F39) provided a radiocarbon date range of 195BC to AD4 (see Appendix H).

Structure 4 (S4)

Structure 4 consisted of an arc of gully (F59), 8.0m long with the open side facing to the north. The arc was closed by a series of three postholes (F41, F54, F58) along a rough east-west alignment. F59 was heavily truncated at its western end by a medieval furrow, F26.

Upon excavation, the profile of F59 was very similar to that recorded in F39 (S6). The gully measured between 0.4 and 0.5m wide at its top and had a steep U-shaped profile and flat base at a depth of 0.2m (Figure 34a and 34b). The backfill (C1068) comprised a deposit of mottled dark grey sandy clay which contained fragments of sandstone and occasional rounded cobbles, some of which appeared to have been burnt. The presence of a backfilled post trench running along the northern edge of F59 could be inferred from section although not visible in plan.

F54 was allocated to a posthole located 1.8m to the northeast of the terminus of F59. This posthole appeared to represent the start of a line of features which ran across the open side of S4. F54 was a slightly irregular cut, 0.4m deep with a flat base and sides that sloped between 50 and 75 degrees. In plan, F54 measured 1.2m in diameter and was backfilled with a single deposit of mottled dark greyish brown sandy clay (C1055). This deposit contained frequent gravel, pebbles and reddened cobbles in addition to charcoal within its matrix. A pair of postholes (F41, F58) backfilled with a similar deposit were located approximately 3.5m to the west. Both features were circular in plan 0.25m in diameter and cut to a depth of 0.15m (Figure 35e). The western end of the arc could not be identified under furrow F26.

A total of six sherds of coarse pottery were recovered from the backfill of F59 (C1068). This pottery appears to have come from a single jar and was dated by its fabric and form to the Late Iron Age (see Appendix F). Charcoal found within F59 C1068 provided a radiocarbon date range of 201BC to AD60 (see Appendix H).

Structure 3 (S3)

Structure 3 was formed by a series of fragmented lengths of gully (F38, F150, F159) and a posthole (F152) which created a rough broken arc with its open side facing south. The open side of S3 was closed by two postholes (F42, F184) producing a structure which was 6.8m long across its open face, and 3.5m wide. In this respect S3 was the smallest of the structures on the site.

F150 comprised the eastern most length of gully. It measured 1.1m x 0.25m and upon excavation proved to be 0.15m deep (Figure 35c). At the southern terminus of S5, a single flat burnt sandstone slab, 0.25m square, was recorded (F165). This stone appears to have been the pad for a post located at the end of the gully. A longitudinal section along F150 (Figure 35d) showed that a second post was set against the northern end of the gully. This would have sat in a socket which would originally have measured 0.3m by 0.2m. F150 was backfilled with C1160, a deposit of variable greyish brown sandy clay which contained frequent small fragments of sandstone and occasional burnt cobbles.

The second length of gully (F38) lay 0.3m to the northwest. A circular posthole, 0.2m in diameter (F152) was located in the gap between the two. When excavated this feature proved to be 0.1m deep and backfilled with similar material to that of F150. F38 was 2.0m long and 0.5m wide with a wide U-shaped profile which survived to a depth of 0.1m (Figure 35b). The backfill of F38 (C1039) was essentially the same material as that recorded in F150.

The third element of the arc of Structure 3 was another 2.0m length of gully (F159) located 0.4m to the west of F38. In form F159 was very similar to F150. It had a regular U-shaped profile cut with steep sides and flat base and measured 0.3m wide and 0.1m deep (Figure 35a). At its eastern end a 0.1m wide stake/posthole was recorded (F157). F159 was backfilled with a deposit of mottled greyish brown clay (C1153). The western end of this arrangement was disturbed by F164.

Of the two postholes that were identified across the face of S3, F42 was located centrally. This posthole was badly disturbed by a modern land drain (F10) which had truncated much of its eastern edge. F42 comprised a sub-circular cut, 0.4m in diameter and 0.2m deep (Figure 35e) and was backfilled with a C1043, a deposit of compact greyish brown silty sand.

The second posthole (F184) marked the western limits of S3 and was only defined after the removal of furrow F26. As such, its truncated remains survived to a depth of 0.1m (Figure 35f). In plan F184 measured 0.8m by 0.5m with its long axis aligned parallel to gully F150. A backfilled postpipe 0.16m wide was recorded in section at the southern end of the posthole. From the shape of the base, a second post, possibly angled to brace the first, was located at the northern end. This arrangement mirrored the post settings identified within F150. The postpipe of F184 was backfilled with a deposit of greyish brown sandy clay (C1209) in which several lenses and blocks of charcoal were recorded. The main fill of the posthole comprised a redeposited natural sandy clay (C1210).

The burnt appearance of the in situ post pad (F165) and the charcoal blocks within the post pipe of F184 suggest that Structure 3 may have been destroyed or cleared by fire.

Structure 3 proved to be the least well preserved of all the structures. A single abraded sherd of Late Iron Age pottery was recovered from the backfill of F38 (1039) which has be tentatively dated to between the 1st century BC to the 1st century AD (see Appendix E). A radiocarbon date of between 135 BC and 25 BC was obtained from charcoal recovered from within this deposit .

The Pit

F40 was located within the arc of S3. This pit represented the only recognisable non-structural feature within the central area. In plan F40 appeared as a sub-oval deposit of dark reddish grey sandy clay (C1041) 1.0m x 1.0m. This deposit was surrounded by three concentric bands of variable clay and charcoal which appeared to be the edges of earlier fills tipping into the pit (C1061, C1062, C1069).

F40 was excavated in quadrant to acquire both a SW-NE and SE-NW aligned section through the backfill sequence (Figure 36). The final form of F40 was that of a sub-circular pit, 1.4m x 1.2mx 0.7m, with a flat base (Plate 14). The sides of the feature sloped between 80 and 90 degrees and had a gradual break slope of at their base. The southern edge was slightly undercut at its top by up to 10 degrees. Against the eastern edge of F40 a shallow rectangular slot was identified set against the top of the feature.

The earliest backfill within F40 comprised a thin deposit of dark grey ashy silt measuring 0.3m by 0.2m, located against the western side of the base and lower slopes of the pit (C1075). This material contained a quantity of burnt grain and charcoal within its matrix which was sampled.

Sealing this was an undulating deposit of black charcoal/silt/ash which varied in thickness from 0.01m to 0.06m (C1070). C1070 occurred within and between a tumble of burned and blackened cobbles and angular sandstone fragments (C1071) which tipped into F40 from its western side. The stones varied in size between 0.1m and 0.4m. All appeared to be burnt but none to the extent where they had been cracked or broken by very high temperature. Additionally, there was no evidence for in situ burning within the pit. An estimated 40% of the charcoal fill (C1070) was made up of charred grain which was recovered for analysis.

C1070 and C1071 were sealed by a deposit of firmly compacted, mottled, brownish yellow sandy clay (C1069). This deposit tipped steeply into F40 lining the sides but not the base of the pit. C1069 varied in thickness from 0.05m to 0.15m and contained flecks of charcoal in addition to a small amount of burnt grain and sandstone fragments.

C1069 was sealed by a second episode of stone dumping (C1063). C1063 comprised a substantial deposit of mixed burnt cobbles and sandstone fragments contained within a deposit of grey clay and charcoal, tipping into F40 from all sides. Although the stones were all burnt, as with C1071, none had been cracked or broken by exposure to excessive temperature, nor had any been cracked by insertion into water when hot. Again there was no evidence for in situ burning within F40. Covering C1063 and following the contours of the cobbles was a layer of greasy grey clay (C1062). This material, like C1069, appeared to form a band of lining concentrated around the edges of the pit and could be seen tipping into F40 almost vertically against its northeastern and southwestern edges. Within C1062 were rare flecks of charcoal and burnt clay. It is not clear whether this deposit, along with C1069, is an applied lining or the remains of a degraded organic deposit.

C1062 was sealed by a substantial deposit of compact redeposited subsoil (C1061). This material was made up of a homogenous deposit of brownish yellow sandy clay. The profile of C1061 within F40 was unusual in that, despite this material tipping steeply into the pit from every side, it remained a consistent thickness throughout (Figure 36). This suggests that the steep U-shaped profile of C1061 may have been the result of the post-depositional slumpage of earlier deposits, as opposed to an incidental backfill or a deliberate lining. C1061, therefore may be a deliberate clay capping to the pit which has slumped over time as earlier deposits have settled.

This may explain the profiles of earlier contexts within the feature. The high ash content of C1075 and C1071 would have given these deposits a large volume at the time of their deposition. If the pit was backfilled and capped with clay within a relatively short period then, when the organic components and ash would begin to degrade and settle. This may have been exaggerated by the weight of the stones within C1071 and C1063. Recent work in York has shown that this process is a common phenomenon in Anglian rubbish pits which have a large organic content, and that it produces a similar profile to that recorded in F40.

The final backfill of F40 was a 0.3m deep deposit of dark grey clay containing charcoal flecks, cobble and gravel inclusion (C1041). This context may have been deposited to fill the depression left in the top of F40 once the earlier fills had subsided.

No pottery or datable finds were recovered from the fills of F40. The nature and composition of C1075 and C1071 suggest that the pit was being used to contain or dispose of material burnt elsewhere on the site. Samples recovered from C1069, C1070 and C1075 were taken for flotation. In addition to a large quantity of burnt grain and charcoal, a minute quantity of burnt/calcined bone was recovered during the flotation of C1070. Although the size of this sample was of little analytical use, the presence of this burnt bone within C1070 is significant. It suggests that C1070 was not simply the remains of a storage context which caught fire, but was more likely to represent the deliberate burning of the grain within a hearth, possibly with other materials.

An assessment of the charred plant remains was undertaken by Headland Archaeology Ltd and is presented as Appendix I and Appendix J. A mixture of cereal grains were present throughout each context of which C1070 contained by far the largest volume. The cereals present within these deposits comprised both barley and wheat in quantity, with a background presence of oats, grass and weed seeds. The contents of the assemblage suggested that the crop had been processed and sorted before it was burnt. In addition to the cereal remains a number of hazelnut shell fragments were recovered from C1070. This may be a residue of the fuel used in a fire.

An array of three radiocarbon dates were obtained from grain from these contexts (see Appendix H). The results suggested that the grain deposited within F40 was fired between 165 and 120 BC. This suggests that F40 was of late Iron Age date and was used to dispose of grain and other material burnt elsewhere on the site. In its disuse F40 was sealed with a capping of clay which slumped through time creating a shallow depression over F40. This depression was then backfilled with C1041 to level up the area.

SOUTHEAST (Figure 37)

The earliest feature within the southeastern area comprised a short length of east-west aligned gully (F88) cut at its western end by F87. F88 proved to be a shallow U-shaped slot 1.2m long, 0.2m wide and 0.1m deep, backfilled with a very pale yellow deposit of silty clay (C1229).

The main element within the southeastern corner of the enclosure was Structure 2. This consisted of a series of three short lengths of curvilinear gully (F87, F96, F98) which formed a true semi-circle, 7.0m in diameter, with its open side facing towards the southeast. The two northern lengths of gully (F87 and F96) may have originally been the same feature, the centre of which had been truncated by a medieval furrow (F28).

Table 6 Summary of component features of Structures 2

Structure	Features
S2	F87, F96, F98, F99

S2 was the only structure with the potential to have been part of a true circle. The interior and the area to the southeast of S2 was trowel cleaned on three separate occasions in order to define any possible internal features or additional gullies. The semi-circular form of S2 appeared to be its original shape.

F87 and F96 were both 0.5m wide, with shallow U-shaped profile, surviving to a depth of 0.12m (Figure 38a and 38b). Both were backfilled with a deposit of mottled greyish brown silty clay with inclusions of charcoal, gravel and pebbles, some of which appeared to have been burnt (C1228, F87 and C1238, F96). These features would have originally made up a single length of gully, defining the northern half of the arc.

The southern half of the arc was defined by F98. This gully was 3.8m in length and when excavated had steep sides (around 60 to 70 degrees) bottoming into a flat or slightly concave base. F98 proved to be 0.3m deep and varied in width between 0.4 and 0.65m. Although the gully is recorded as only having a single fill (C1220), the excavated section suggests that there may have been a post trench, 0.2m wide, set centrally within the feature which was invisible in plan. C1220 comprised a variable deposit of dark greyish brown silty clay which contained frequent inclusions charcoal flecks gravel, cobbles and sandstone fragments. As with F96 some of these stones appeared to be burnt. Crumbs and flecks of daub were also noted within C1220. The concentration of stones and cobbles became greater towards the northern end of F98. At this point the terminus of the gully was marked by a sub-circular posthole or pit 0.7m in diameter (F99). The interface between the backfill of F99 (C1221) and that of F98 (C1220) was difficult to see in section and it was concluded that the disuse of both features appeared to be contemporary. F99 was characterised by a large tip of cobbles and sandstone fragments running into the feature from its western edge. The remainder of C1221 consisted of a deposit of variable dark grey silty clay which contained a comparatively large amount of daub/fired clay and charcoal. F99 appears to represent a large posthole located at the terminus of F98. It was situated 2.0m to the southwest of F96 leaving a gap or possible entranceway of 2.0m between the two arcs of gully.

Although there was a distinct lack of pottery from any of the deposits within S2, a total of 325g of fired clay was recovered from C1221 (F99). The assessment of this material suggested that the clay had been fired to a high temperature and was thought to be the remains of some form of mould or fragmented furnace structure. Two of the pieces had distinct lateral impressions, possibly made from leather binds. Whatever the origin, this material, along with the 40g recovered from F98, was not found in situ and represented material deposited within the gullies and postholes of S2 in their disuse.

Charcoal recovered from F98 (C1220) provided a radiocarbon date range of 381BC to 202BC (see Appendix H).

Several other features were recorded and excavated in the southeastern area. These included four postholes (F97, F100, F189, F193) which may have been associated with S2.

F189, F100, F97 and F193 formed a fairly convincing alignment, orientated NW-SE across the southern area, which joined with F72 (Central area) at its northwest end. F189 was found to be 0.2m in diameter, 0.1m deep and contained a deposit of sandy clay with burnt sandstone and charcoal (C1236). F100 was located in the gap between F96 and F99. When excavated this feature proved to be of a similar form and dimension to F189 and was backfilled with a light olive brown sandy clay (C1242).

Midway between F100 and F72 were F97 and F193. F97 proved to be the remains of a double stakehole, 0.15m deep. F193, to the south comprised a shallow U-shaped depression, 0.1m in diameter.

F84 was identified in the far southeastern corner of the enclosure and was circular in plan. This feature was 0.35m in diameter but when excavated proved to be only 0.06m deep. F84 was backfilled with a single deposit of olive brown sandy clay with rare charcoal fleck inclusions (C1231).

GENERAL (Figure 39)

A number of other features were excavated within the enclosure. Some of these remain undated and could not be associated with any of the structures, whereas others exist as elements within the enclosure within their own right.

Stakeholes

During the cleaning of the southeastern area of the enclosure a large number of small sub-circular and sub-rectangular soil features were identified. These varied in diameter between 0.03m and 0.12m and were distributed in plan in an arc 4.6m wide which curved from its southern to its northeastern end over a distance of 16.2m. Additional concentrations of these anomalies were also identified in the northern and western areas of the enclosure. These features were allocated F67 and were planned prior to sample excavation on the basis that they may have represented the remains of stakeholes.

Of the 536 stakeholes that were planned, a total of 100 were sampled and recorded in box sections. Of these, one in four were photographed. The results from this exercise indicated that most of the excavated features were archaeological. The sections provide a range of profiles which varied between U- and V-shapes to square flat bottomed cross sections (Figure 40). These features varied in depth between 0.03m to 0.2m, were well defined against the natural subsoil and filled with a dark grey clay sand or sandy clay (C1339). The profiles showed that most of the features were vertical in section but there were several which had been formed at an angle.

In relation to the group of features located within the southeastern area of the enclosure it was difficult to see a meaningful pattern within the distribution of F67. It was noted that many of the stakeholes appeared to be concentrated in clusters, some around other features (e.g. F100), while others appeared to form short lines running at 90 degrees to each other. The curve of the band formed by F67 respected the arc of Structure 2 and was defined to the east of the line of posts and a gully formed by F70, F71, F92 and F103.

Due to the ground conditions it was difficult to establish stratigraphic relationships between F67 and other features. However, it was clear that these stakeholes predated the medieval ridge and furrow cultivation. Even where features were very shallow (F96, F100, etc) the stakeholes were not seen against their edges suggesting that they may have respected existing structures or posts.

It is assumed that F67 would originally have held a series of upright stakes which would have been hammered into the subsoil. As the site is likely to have been truncated by c.0.3m by ploughing, these stakes would have been driven between 0.35m and 0.5m into the ground. What type of structure these features may have formed is difficult to envisage. The majority of stakeholes appear in a rough arc in the southeastern corner of the enclosure. In plan it is not unlike a semi-circular structure in its own right with its open face 8.5m wide and aligned to face to the east (Figure 39). If real, the alignment of this structure fits well with the location of the entranceway of the original enclosure (F37). It is not clear whether, in this case, F67 represents a single dense fence or a series of stake-built elements which were replaced over time.

The possibility that these features represent root holes has also been considered. When studying the distribution, excavated form/profile and context of these deposits this seems to be an unlikely explanation. There was no evidence for any root disturbance in any of the gullies or postholes, where the backfills of these features would presumably have provided a more attractive strata than the sterile natural clay. Nor was there any evidence for lateral roots or systems visible against the natural clay anywhere within Zone 2.

Other Features

F103 comprised a 5.7m length of gully aligned north-south running into the enclosure ditch at its southern end. This feature proved to be between 0.4m and 0.7m wide and was cut to a depth of between 0.2m and 0.3m. F103 had a U-shaped profile and was backfilled with C1222, a deposit of compact, mottled sandy clay (Figure 41). This deposit contained charcoal and pebble inclusions, of which several appeared to have been burnt. Despite F103 running into the enclosure ditch, it does not appear to have been a drainage feature as the northern end of the gully is deeper than the southern end. Both ends of F103 were well defined and rounded in plan.

F103 was cut through F234 and, as such, may have been contemporary with one of the latest phases of enclosure circuit. A radiocarbon date of between 39BC and 77AD was obtained from charcoal recovered from F103 C1222, suggesting that F103 may have been related to one of the later structures (see Appendix H). The alignment of this gully with the set of postholes F70, F71 and F92 indicated that it may have been the southern end of a sub-division of the enclosure.

If F103 was contemporary with a phase of the enclosure circuit it would indicate that the presence of an internal bank at this time is very unlikely. The division marked by these features, if real, would have served to annex the southeastern corner of the enclosure, an area of the site which had been the focus of activity related to Structure 2.

The final set of features within Zone 2 comprised a length of north-south aligned gully (F108) and a series of four postholes (F105, F106, F107 and F167) located to the south of the structures within the centre of the enclosure. F108 was 4.95m long and 0.45m wide. Upon excavation, this feature proved to be only 0.1m deep, and was backfilled with a deposit of mottled olive brown clay with few inclusions (C1224) (Figure 42). Along the same alignment as F108 and 2.5m to the north was a sub-circular posthole (F105), 0.3m in diameter and 0.14m deep backfilled with mottled deposit of grey sandy clay (C1186). A further 2.05m along this line was a second posthole (F106). This feature proved to have an elongated rounded cut 0.3m x 0.2m x 0.25m, backfilled with a mottled grey sandy clay (C1187). Together these three features formed a structure or fence some 10.0m long which probably extended to the enclosure ditch on its southern side. To the north it is possible that this alignment may have included F173 and/or F191 thereby dividing the enclosure with a north-south aligned boundary some 23.0m long. 2.6m to the west of F108 were two more features (F107 and F167) set 1.1m apart. F107 was a well defined but shallow posthole 0.25m in diameter and 0.1m deep. 1.1m to the south of this was a shallow irregular scoop (F167) backfilled with similar material. Whereas F107 appeared to be the remains of a single posthole, F167 resembled the heavily truncated base of a double post setting which measured 0.7m by 0.3m. Both features formed a line parallel to that made by F108.

ISSUES

The results of each area have been presented by feature and structure. However, it is necessary to consider several problems with this data set before alluding to a sequence and interpretation.

Truncation

The distribution of features across Zone 2 shows a distinct lack of activity in the western and northern part of the enclosure. Although this may be a true reflection of the organisation within the enclosure, the level of truncation across the whole site by modern ploughing needs to be considered. From studying the depths of extant ridge and furrow in other areas of the Golf course (Intervention 4), it is estimated that in excess of 0.25m of subsoil has been truncated by the plough across the enclosure. With the exception of the cobble floor of F206, this has effectively destroyed the original ground surface and any trace of ephemeral features. Such features could have included beam slots, postholes, hearths and floors if originally present. Consequently the majority of features that have survived within the enclosure are the substantial structural ones. This bias in the data is unavoidable but should be acknowledged as a limiting factor for interpretation.

Dating

The chronology of features at Normanton Golf Course was heavily reliant on a series of radiocarbon dates. A total of sixteen samples were sent off for analysis. The majority of these consisted of charcoal recovered from feature backfills by flotation. The samples were chosen to provide dates for each major structure and significant feature within the enclosure.

In many cases the radiocarbon samples were derived from feature backfills which by their very nature are secondary deposits. With evidence indicating that several phases of occupation and structural activity were occurring on the site, the possibility of residuality needs to be acknowledged and may be a particular problem with structures defined within Zone 2. In these cases, charcoal was recovered from the either the construction cut backfill or the post voids in their disuse. If residual charcoal was present on the site, either in middens or surface layers, then it is possible that such fragments could have found their way into post trenches during either construction or destruction.

Although residuality may limit the precision of radiocarbon dating, it in no way detracts from its value in providing a general date range for activity occurring on the site. A study of the radiocarbon dates from the enclosure did not reveal any obvious anomaly within the dating sequence. It dated the main phase of activity on the site to a period between 400 BC and 50 BC, and suggested that less intensive occupation occurred both before and after this time. This sequence appeared to fit with the chronology of the site based on the results of the excavation and was corroborated by the limited pottery assemblage.

Two possible exceptions were observed in this general pattern. The first, F153 provided a radiocarbon date range of 1767BC to 1642BC. This feature, therefore, predated the cremation of F232 by at least 510 years and most probably predated the enclosure itself. On this basis it could easily be discounted either as an archaeological anomaly or as a rogue date caused by residual charcoal. However, the location of F153 may suggest another interpretation which is discussed below. The second date was recovered from the backfill of F35 (C1263). This gully was radiocarbon dated to between 785BC and 399BC (see Appendix H). If this date is accurate then it could mean one of several things. Firstly, that the corridor structure leading into the enclosure (F35 and F36) from the entranceway was not contemporary with any of the structures within the enclosure. The date would also suggest that the corridor predated the gate structure itself (based on the date for F209). Although this was logically possible, the spatial relationships, organisation and similarities with other features suggest that it is very unlikely. Additionally, pottery recovered from the disuse of F36 was identified as being of a Late Iron Age date (see Appendix E). Together this suggests that the charcoal from F35 used for the radiocarbon date was probably residual in nature.

Two radiocarbon dates were obtained from reliable sources. Three samples of grain from the pit F40 were used to calibrate a date of 120 BC to 160 BC. The charcoal recovered from the cremation F232 was assumed to be related to the burning of the bone and was dated to between 1206BC and 917BC. Both of these deposits provided primary material for dating a particular event. In this respect the dating of F232 to the Early Iron Age is worthy of note.

Comparative Analysis

Other techniques were also used in an attempt to correlate features during the phasing of the enclosure. Maps were produced and coded to show the concentration of charcoal, the distribution of pebbles and cobbles, burnt and unburnt stones, and the distribution of colour within the backfills of features around the enclosure. These are shown in Figures 43 to 46.

Charcoal flecks were present in small amounts as inclusions in virtually all the feature backfills. This background concentration was surpassed only in F40, F99, F212, F209 and F184. Apart from indicating fairly obvious concentrations of charcoal in certain features, the analysis did not allude to any meaningful spatial or chronological patterns.

After the discovery of a cobbled surface (F206) within the entranceway it was hypothesised that the whole of the interior of the enclosure may have been cobbled at some point. It was hoped that by studying concentrations of gravel and cobbles within feature backfills, it would be possible to determine whether this was the case, and if so, to distinguish between features which would have predated or post dated such a surface. The results, however, showed that cobbles were present throughout most of the features within the enclosure and apart from S4 and S5 having slightly less cobble inclusions than the other structures, little in the way of a distinguishable pattern could be detected. A similar result was obtained from the distribution of burnt cobbles. These were effectively found in all phases of the enclosure and do not appear to relate to a specific period or feature type.

PHASES

Due to the lack of stratigraphic relationships between features, the paucity of dateable finds and the broad date ranges provided by the radiocarbon results many different phasing schemes could be derived from the enclosure data set. The following phases, therefore, represent the 'best fit' as judged by the author.

Phase 1 (1700BC) (Figure 47)

By virtue of its radiocarbon date range of 1767BC to 1642BC, F153 was the earliest feature in Zone 2 (see Appendix H). The possibility that this date was residual has already been discussed. Assuming that this date was genuine, however, F153 would most probably have predated the enclosure itself. Dated to the Bronze Age, the position of F153 on top of a visible rise in a valley bottom, and then in the Iron Age, centrally within a rectilinear enclosure may be significant. The position of F153 at a central point within the enclosure was emphasised by its pivotal position to all of the central structures. That is not to say that such a feature would still be present at the time these structures erected, rather that F153 may have been the original focus for activity or place. The place, therefore, was reused in later phases rather than the feature. Evidence for other pre-Iron Age activity within this area of the site was recovered in the form of a neolithic flint knife, some 7cm long, found during machining immediately to the north of Zone 2 (Find No.177, Plate 15).

Phase 2 (1200BC - 900BC)

Phase 2 of the enclosure was marked by the presence of a shallow pit (F232) containing cremated animal bone. The early Iron Age date of this feature (1206 BC to 917BC) placed it well before any of the main structures were constructed on the site and although it was possible that this feature occurred within the first phase of enclosure ditch, no dating evidence was recovered from F37 to corroborate this. F232 could, therefore, have been pre-enclosure in origin.

The purpose of F232 was unclear. The calcined nature of the bone assemblage suggested that it may not have been the result of a domestic cooking fire (see Appendix G). Whatever its function, F232 indicated a presence on the site of the enclosure which dated to the Early Iron Age.

Phase 3 (600BC - 400BC)

Phase 3 formed the first phase of the rectilinear enclosure of F37 with an estimated date range of 600BC to 400BC. There was little evidence for internal structures that were contemporary with the first enclosure ditch. The spatial arrangement created by the arc of stakeholes in the southeast corner (F67) were the only element to respect the wide entranceway of F37 and, therefore, may be contemporary with this phase.

Phase 4 (400BC - 200BC)

The main elements of Phase 4 comprised the recutting of the ditch (F234) to incorporate a narrow gateway with a corridor leading into the centre of the enclosure. It is possible that there were at least two sub-phases to this arrangement. The first comprised postholes F211 and F216 forming a post gateway 2.2m wide which was linked to a corridor, the southern side of which was formed by a series of postholes (F90, F91, F192, F195). The northern side was formed by F35 or an earlier version which was obliterated when recut. The second possible sub-phase consisted of a narrower gateway, 1.2m wide, tied into a corridor composed of F36 to the south and F35 to the north.

There was no evidence of a palisade or fence or any other structure running along the inside of the ditch which the gateway could have been tied into. It is possible that the gateway was linked to an internal bank, but there is no evidence to corroborate this.

Structure 5 was located centrally within the enclosure at this time. This hut appeared to have been used for a considerable period with there being evidence that it had been repaired or reconstructed in exactly the same place. The disuse of the second sub-phase was radiocarbon dated to between 398BC to 261BC (see Appendix H).

Structure 2 was possibly contemporary with Structure 5. S2 was located in the southeastern corner of the enclosure and differed in several respects to any of the other structures found within Zone 2. Its semi-circular form, symmetry and 2.1m wide entranceway along its northwest wall, between F96 and F98, contrasted with the asymmetrical squat forms of S3, S4, S5 and S6. Additionally the open face of S2 was not closed by postholes as seen in all the other structures.

S2, therefore, may have been a very different building in form and function than S5. Rather than being constructed from a series of posts, the back of the building may have been formed by digging a screen or wall into the earthen bank of F234. This structure may even have supported a platform or deck in the corner of the enclosure.

The backfill of gully F98 and posthole F99 produced a large quantity of fired clay. This material appeared to be burnt daub in which wattle imprints could be seen in at least two fragments. Whether this material indicated that S2 had been burnt is one matter, but it does provide further proof that these features were of a structural nature. The disuse of Structure 2 was dated by radiocarbon dating to 290 BC ± 90 years.

Phase 5 (200BC - 150BC)

The radiocarbon dates suggest that there may have been a possible hiatus in the occupation on the site until the Later Iron Age. This may have resulted in the silting up of F234 which was followed by the remodelling of the enclosure at a later date. It is also possible that the arrangement of stakeholes (F67) relate to this period of partial disuse. As the corridor became less significant, F67 may represent an attempt to continue the function and form of Structure 2 in a less formal way.

Phase 6 (150BC - 0AD)

Phase 6 comprised the remodelling of the enclosure after the funnel corridor had gone into disuse. The enclosure ditch (F234) was recut into a complete circuit (F43) and the causewayed entranceway was replaced with a bridged crossing. Originally this may have been a fixed wooden structure, supported by postholes F212 and F217, which may have had a gated element to it. The rationale behind this lay in the fact that the two smaller posts, clearly set with F212, appeared to be cut into the western slope of the enclosure ditch.

Three structures may have existed at the centre of the enclosure at this time (S3, S4, S6). The broad radiocarbon dates for these structures made it difficult to separate them into a chronological sequence. Structure 6 and Structure 4 shared similar profiles and backfills which implied that they may be consecutive. The axis of Structure 6, however, was identical to that of Structure 3 which suggested that despite their different construction they too may be consecutive. On this basis, Structure 4 must have been either the first or the last structure built in the centre of the enclosure. Pottery recovered from the backfill of F59 was of a similar fabric to that recovered in the robbing/disuse of gully F36. With this being the case, it can be argued that Structure 4 represents the first building in this sequence.

The presence of F40 within this arrangement should be considered. This pit appeared to have been backfilled between 120 BC and 160 BC. If the pit was an external feature then it must either have been contemporary with Structure 6 or have predated all the buildings within this phase of occupation. However, if it was within a building it would have been located either along the entrance of Structure 4 or centrally within Structure 3. The proximity of such a deep feature to the walls of all three structures (i.e. less than 0.5m) suggests that F40 was not contemporary with any of them.

In its disuse F40 appeared to have been deliberately backfilled and capped with clay, possibly being levelled prior to the construction of a new building. This pit does not appear to have been used for general rubbish disposal as its primary fill appeared to consist of a single deposition of burnt material with the remaining contexts being related to its backfilling and levelling. The form of F40 is not unlike that of storage pits found on other sites in the region (e.g. Dalton Parlours). It is strange, however, considering the longevity of occupation within the enclosure that there is only one such feature as well as a distinct lack of evidence relating to domestic or agricultural activity on the site.

Structure 4 was located facing to the north and may have been built over F40 in is disuse between 120 BC and 160BC and was, therefore, contemporary with the first bridge structure over F43. This building was eventually replaced by S6 to the north. The reorganisation of the centre of the enclosure may have been contemporary with the cobbling of the entranceway. The original post-supported bridge may have been replaced with a moveable structure. Structure 6 was then finally replaced with Structure 3, situated to the south but on the same axis and alignment. Structure 3, in its disuse, appeared to have been destroyed by fire.

The dating of the structures within the enclosure indicate a period of occupation which must have lasted for approximately one hundred years between 150BC and 50BC. The radiocarbon dates suggest that all of the main structures went out of use by the beginning of the 1st century AD at the latest.

Phase 7 (0AD - 50AD)

A change in the layout and possibly the function of the enclosure was marked by a recutting of the entire circuit (F235) in a period after the main structures on the site had fallen into disuse. F235 was joined to a second ditch (F78) located midway along the eastern length. The purpose of this feature was to drain water from the main circuit acting as an overflow in periods of wet weather. This may have been in response to a deterioration in local conditions and more than likely reflected a need to keep the interior of the enclosure free from flooding. This new emphasis on water management might indicate a change in the role of the enclosure. A rough north-south aligned division of the site marked by gully F103 and postholes F70, F71 and F90 appears to have been contemporary with F235. The disuse of this possible division was dated by radiocarbon analysis to between 39BC and 77AD (see Appendix H). A second sub-division of the site, marked by gully F108 and associated postholes (F105, F106), may also belong to this phase. The lack of apparent structures and formal entranceway, combined with the sub-division of the site and the loss of central focus, all indicate that a fundamental shift in the pattern of activity occurred around this time.

Phase 8 (50AD - 200AD)

In Phase 8 the enclosure circuit (F236) was recut and incorporated into a north-south aligned boundary ditch located at the northwestern and southwestern corners of the enclosure (F220 and F219 respectively). The western end of F78 was backfilled whilst the remainder of this ditch to the east was recut allowing free passage around the enclosure. Although no internal structures were identified from this phase, the enclosure appeared to continue in use. Pottery recovered from the disuse of F236 was dated to the 2nd or 3rd century AD.

This phase of the enclosure saw it incorporated into a wider, formal division of the landscape which was visible in Zones 1, 3 and 4. The very fact that the circuit of the enclosure is retained within this, in particular reusing its western leg, is significant.

Phase 9 (Early Medieval)

There was no evidence for any activity on the site between the abandonment of the enclosure in the 3rd century and the medieval period.

Phase 10 (Medieval)

The remains of eight medieval furrows were identified running across Zone 2 on a north-south alignment. A fragment of blue glass, probably Roman in origin, was recovered from the fill of furrow F24, whilst a sherd of 14th century pottery was found in furrow F30.

Phase 11 (Post-medieval)

A series of fifteen ceramic land drains were recorded running across Zone 2 which were post-medieval in date.

apc > mga > projects > normanton golf course > results > zone 2

---

Timms, S. 2005. Normanton Golf Course, Normanton, West Yorkshire. Excavation & Evaluation
© Archaeological Planning Consultancy Ltd 2003 - 2006
URL: http://www.archaeologicalplanningconsultancy.co.uk/mga/projects/normant/rz2.html
HTML Markup: Guy Hopkinson • email: . Updated February 23, 2006