Excavation report for the 2004 and 2005 seasons at Mellor, Stockport

a

Heritage Lottery Fund

Report on rhe 2004 and ZOO5 seasons oforchaeological excavarions or Adellor.

Contents
Acknowledgements Non Technical Summary 1. 2. 3. 4. 5.
6. 7.
8.

Archaeological Background 1995-2003 Community Involvement Physical Setting 2004 Aims, Objectives and Methodology 2004 Excavation Results 2004 Conclusions 2005 Aims, Objectives and Methodology 2005 Excavation Results 2005 Conclusions

9.

10. Appendix 1 - 2004 Radiocarbon Dating Results 11. Appendix 2 - 2005 Radiocarbon Dating Results 12. Appendix 3 - 2005 Plant macrofossil and pollen assessment 13. Appendix 4 - 2005 Pollen analysis 14. Appendix 5 - Geological Report 15. Appendix 6 - 2005 Roman Pottery Assessment 16. Appendix 7 - Sources

Universip ofMonchesrer Archaeological Unit December 2005.

Report on the 2001 and2005 seasom ofarchaeological exrlnratiom at Mellor.

Non-Technical Summary
The excavations conducted over the past eight years around the Old Vicarage at Mellor, Stockport (SJ 981 8 8890) have revealed an extensive multi-period settlement. Previous excavation seasons have identified the site predominantly as an Iron Age Settlement (GM SMR 11249.1.1). In addition, the recovery of flint fragments dated to the Mesolithic period suggest the site was used as s seasonal hunter-gatherer camp while a substantial assemblage of sherds of Roman material, coupled with comparable radiocarbon dates suggest the presence of a Romano-British settlement.

2004 Excavations

Trench 26 revealed a large area of continuous Iron Age settlement within Area C. It was the largest single trench opened by the excavations at Mellor. A total of 19 gullies were identified. The presence of these inter-cutting features indicates a sustained period of occupation upon the immediate area. Although the exact alignment of the northern extent of the inner enclosure ditch has not been confirmed by excavation, its absence within Trench 26 suggests that its location lies beneath the Old Vicarage driveway. This would mean that this particular area of settlement was located outside the inner enclosure ditch, contrasting with those gullies found within the confines of the inner enclosure ditch identified within Trenches 16 and 21. Trench 27 showed that the outer enclosure ditch continues towards the Old Vicarage drive. It also indicated that the level of the surrounding ground surface appears to have been reduced. Together with the results of Trench 25 in 2003, it is possible to show a substantial amount of alteration, landscaping and terracing around the western end of the Old Vicarage during the post medieval period. Trenches 28 and 29 confirmed the presence of a significant number of archaeological features towards the north of Area C. Indicating that the settlement located within Trench 26 is not the firthest extent of occupational archaeology from the inner enclosure ditch, both of these trenches were exposed but not excavated. Trench 30 followed geophysical investigation of Area E in attempt to establish the parameters of the outer enclosure ditch. Trench 30 positively concluded the presence of a ditch, m i n g in an east-west direction towards the summit of Mellor hilltop. Trench 31 located next to Trench 18, confirmed the presence of the inner enclosure ditch running towards the Old Vicarage driveway and the continuation of the palisade slot identified within Trench 18 parallel to the ditch. Trench 32 showed the continuation of the inner enclosure ditch, from Trench 18 towards the church wall and identified the continuation of a stone lined feature parallel to the ditch. Trial trench 16, stripped in Area B in 2004 provided an insight into more archaeological features occurring in between the inner and outer enclosure ditches expanding the areas of the potential settlement.
Universiry ofManchester Archneologicol Unit December 2005.

Report on rhe 2004 and 2005 seasom ofarchaeological excavations at Mellor.

2005 Excavations

In Area A, Trench 35 revealed four large post pits associated with the one found in the Trench 1 extension during the 2001 season. They were in a north south alignment, c. 11.80m long and each separated by a distance of c.2.65m. Each pit was cut into the natural bedrock and c. l m wide, c.0.40m deep, containing a c.0.40m wide post pipe. Provisionally these represent post pits, which together with three identified within Trench 33 and two located in Trench 34. The recovery of an arrowhead dated to the 13-14th centuries, identification of 11" to 15" century pottery fragments and a radiocarbon date of 1000-1250 cal AD (Beta209508, 2 sigmas), all contained within the post pits, would suggest a provisional dating of the structure to the medieval period. However the true form of the building in terms of extent and shape is as present unknown due to the current limits of excavation. Investigation within Trench 33 also confirmed the presence of the large inner enclosure ditch and associated palisade slot expected to continue between Trenches 1 and 2. During 2005 Trench 36, was positioned adjacent to the previous years Trench 26. Identifying the continuation and extent of the curvi-linear roundhouse drip gullies partially exposed in 2004 one of which produced a radio-carbon date of 190 cal BC 10 cal AD (Beta 202315, 2 sigmas). The combined evidence for the two trenches suggests that the curving gullies were on average a diameter of c. lOm, and had similar inter-cutting tendencies demonstrating that there was a substantial period of occupation and rebuild in the immediate area. Within this trench numerous intercutting pits were identified comparable to and in a similar alignment to those found within 2004. These are provisionally interpreted as Iron Age cooking pits, established through the recovery of significant quantities of fire cracked sub rounded and rounded inclusions from the pit deposits. The fill of a small circular pit 1 posthole located within the confines of the numerous roundhouse gullies contained possible evidence of industrial waste, along with environmental evidence relating to a domestic nature and oroduced a radiocarbon date of 190-40 cal BC (Beta 2095 10. 2 sigmas). Suggesting the possibility of Iron Age industrial processes present on the site and occurring within the roundhouses or their immediate surroundings. Two parallel lines of stake holes running up to the roundhouse gullies provides evidence for a possible fenced off animal enclosure.
~-~
~

.

The extent of the inner enclosure ditch has now been proven to continue through Area A, without the interruption of an entrance way, and has been tracked to the fiuthest point accessible for archaeological investigation through the excavation of Trench 37. Both trenches over the two inner enclosure ditch show the presence of an associated palisade, with a possible post hole alignment overlapping, creating a small entrance to the ditch located to the far west of the site. Continuing a programme of geophysical survey and subsequent trial trenching excavation during 2005 in areas D and E, further investigation concludes the continuation of the outer enclosure ditch alignment. Trenches 38 and 39 were placed across the suspected geophysical anomaly and revealed ditch sections showing this feature continues to the eastenunost extent of Area E. In Area D it seems likely that outer enclosure ditch runs the entire distance between the 'Ale house' track way in the.west, and Mellor Old Hall in the east. Excavation of Trenches 40 and 41 during October 2005 identified these two features as terminal ditch sections of the outer
University ofManchester Archaeological Unit December 2005.

enclowule ditch. This tmtrmmway is the first identified as part of the enclosure dirhs. S u g p t b g a sigdfbnt area, if not all of the hilltop, was as some point enclosed,c m q m m h g a suspected a m of 23 hectares.

Flgore 1 Plan of Mellar hilltop indicating the a m s COW various -nt : by the types of geophysics and the d t s . Ihe red c o l d lines qmscnt coverage by around l%mmhg Radar (GPR).

Report on the 2004 and 2005 seasons o/archaeological excavations at MeNor.

1.

Archaeolo~ical Background: 1995-2003

Ann Hearle, Chair of Marple Local History Society and Dr Peter Arrowsmith of the University of Manchester Archaeological Unit (UMAU) first suggested the presence of an Iron Age Hill fort at Mellor in 1998. The suggestion was based on a photograph taken by Ann Hearle of the field to the north of her house at the Old Vicarage during the summer of 1995. The majority of the grass in the field had been bleached brown by the sun. However the photograph showed a line of lush grass arcing across the field. It was felt that the line might be following that of a filled in ditch. The less compact nature of the in fill of the ditch would retain water and so better sustain the grass above it in times of drought.

The next step was a geophysical survey which confirmed the presence of a below ground anomaly corresponding to the line of grass in the field (Area B) and others in the garden of the Old Vicarage itself (Area A). Excavation started in the summer of 1998. Trench 1 was excavated over one of the anomalies at the west end of the garden. This revealed a large ditch cut into the sandstone bedrock. This section of ditch was over 4m wide and 2.10m deep. The top fills of this contained fragment of Roman tile, pottery and glass. Lower down the fills contained pottery and other artefacts dating from the Iron Age. The high charcoal content of one of these fills near the base of the ditch meant that a sample could be taken for radiocarbon analysis. This produced a date of cal830 - 190 cal BC (Beta 146416,2 sigmas). During November 2002 Trench 25 was excavated though part of the vegetable patch adjacent to the ditch in Trench 1. This 12m long trench showed that the current flat ground surface is a result of deliberate levelling of the natural slope of the hill in post medieval times. The implications of this are that the ditch found in Trench 1 would have been situated upon the break of a steep slope and would have been highly effective defensively, and a far more visible and imposing feature within the ancient landscape. In 2002 and 2003, Trench 18 was excavated over an anomaly at the opposite end of Area A. This revealed a section of ditch very similar to that found in Trench I . This section of ditch ran north to south and measured just over 4.0m wide and 1.90m deep. Within the ditch an abundance of artefacts were recovered dating from the 1'' to the 4" century AD. These included 5 bronze Roman brooches, 221 sherds of RomanoBritish Pottery and fragments of quem stones. Most of these finds came from the upper fills and in some cases pottery from several different centuries was recovered from the same layer. Suggesting that the ditch was used as a dumping ground when the Roman occupation of the site ended in the 41h Century AD. It seems likely that the sections in Trench 1 and Trench 18 are part of the same ditch. This interpretation means that in the Iron Age the area of the Old Vicarage and Saint Thomas' church was surrounded by an imposing defensive inner enclosure ditch. Immediately to the west of the ditch (c.2.33m) and contained within the same trench, a small stone filled slot was identified running parallel to the ditch. It was thought that this might be the foundation for a palisade running inside the ditch; however, this was

Universiv of Monchester Archaeological Unit December 2005.

Report on the 2001 and 2005 seasom of archaeological excavations at MeNor.

too short a length exposed to be certain of this interpretation. No evidence for an associated rampart was identified throughout the excavations. Two trial trenches were excavated on the other side of the Old Vicarage drive, immediately opposite Trench 18, in Area C. These were intended to look for the ditch found in Trench 18 continuing into this area. The evidence for the ditch was unclear but one of the trial trenches suggested the presence of a curving gully perhaps associated with a roundhouse. Trenches have also been excavated over an anomaly in Area B to the north of the Old Vicarage. These have revealed a c.400m stretch of an Iron Age ditch running southwest to northeast. The ditch in this field is around 2m wide and about 1.70m deep and probably represents an Iron Age enclosure ditch defining the limits of the settlement at Mellor. The 2001 open area excavation exposed a 13m length of ditch which produced 125 sherds of hand made pottery belonging to the same Iron Age pot. These have now been conserved and the pot reconstructed. Excavation has so far concentrated on defining the extent of this ditch to the north and west of the Old Vicarage. In November 2002 a geophysical survey using a magnetometer was carried out in Area D to look for an eastern arm of the ditch. On the basis of the survey results two trial trenches were excavated which failed to find any evidence for the ditch. Further geophysical analysis in the run up to the 2003 season led to the excavation of two trenches and eleven trial trenches designed to determine the line of this enclosure ditch beyond Area B. Results seem to show that this enclosure ditch does not turn back towards the church but carries straight on towards Mellor Old Hall. This implies that the ditch encloses an area of land far greater than previously anticipated and dramatically increases the potential size of the Iron Age settlement. Palaeoenvironmental analysis of the archaeological fills within the enclosure ditch indicated the presence of mixed deciduous woodland, a nearby open body of water and an associated wet meadow, contemporary with when the ditch was open. The recovery of cereal-type pollens indicates the presence of a mixed farming economy. The evidence from the trenches now seems to point to there being two ditches at Mellor. One is a large inner enclosure ditch which probably surrounded a relatively small area of the hilltop which is currently occupied by the grounds of the Old Vicarage and Saint Thomas's churchyard. The second is smaller but far more extensive outer enclosure ditch that may well extend over the greater part of the hilltop. The area enclosed by the ditches has also been examined. In 1999 Trench 3 was opened up in the centre of Area A. In the eastern half of this trench the sandstone bedrock was covered by a layer of boulder clay. Cut into this layer was a complex pattern of postholes, ditches and gullies ranging in date from the Mesolithic period 5 to 10,000 years ago through to the Roman period. In 2002 a 10m square, Trench 16 was opened adjacent to Trench 3. Amongst the features found in Trench 16, were a series of gullies, which formed an arc within the west half of the trench. The results from this trench allowed a fuller interpretation of the features fiom Trench 3 to be made. This suggested that the curving gullies continued beyond the boundaries of
Universily ofManrhester Archaeological Unit December 2005.

Report on rhe 2004 and 2005 seasons of archaeological excauarions at Mellor.

Trench 16 to form a complete circle typical of the drainage gullies found outside Iron Age roundhouses. Radiocarbon dates from charcoal within the fill of these gullies produced a date of between 380 - 520 cal BC (Beta-173892, 2 sigmas) Finds from Trench 16 included a polished flint chisel of a style associated with the Late Neolithic Period. C2-3,000 BC. The excavation of Trenches 21 and 23 in 2003 confirmed the presence of an lron Age roundhouse. Combining the results from several years of excavation produces a circular drainage gully for a roundhouse with a diameter of approximately 13 metres. A linear feature was found cutting the roundhouse gully and radio-carbon dated to 520-380 cal BC (Beta-173893). Indicating a changing use and occupation within the immediate area over a prolonged period of time. Not an No Iron Age finds were recovered from the sections of the gully excavated in 2003, however from the gully and the area immediately to the north, 97 flint flakes were recovered. Specialist analysis shows these to be characteristic of the Mesolithic Period and indicative of hunter-gatherer communities using the hilltop at Mellor as one of their seasonal bases over 10,000 years ago. It is likely that these were re-deposited from elsewhere, within the surrounding area. As a continued part of the excavation of the hilltop the Mellor Archaeological Trust commissions each year a programme of geophysical survey. The results of the survey can provide valuable information which can be used when deciding where hture excavation trenches should be placed (see figure 1).

Universiiy o/Manchesrer Archneological Unil December 2005.

Report on the 2001 and 2005 seasom ofarchaeological excavations at Mellor.

companies involved in archaeology including English Heritage, Tarmac Group, The lnstitute of Field Archaeologists and Transco. There are 15 categories of awards covering all aspects of archaeology and heritage which attract hundreds of entries from all over the country. The entrants in each category are assessed by judges and a short list of finalists in each category is announced. Mellor entered in two categories - The Channel 4 Award for non broadcast video and The Pin Rivers Award for best amateur project and reached the final in both categories. That year the awards ceremony was held at the University of Belfast, as well as the 100 or so finalists it was attended by its President Lord Montagu of Beaulieu and many representatives from television archaeology. Although Mellor did not win the awards it entered for to come runner up in both categories is an achievement that everyone involved with the excavation can take a great deal of pride in. The Mellor archaeological project is also publicised through numerous lectures and talks given by members of the Trust, UMAU and GMAU. In 2003 Mellor Archaeological Trust organised a Study Day on the Iron Age and Romano-British period in the region. This was held in the Parish Centre and attended by some of the leading archaeologists and specialists in the field. In 2005 the Mellor Archaeological Trust, UMAU and GMAU collectively published the inaugural Manchester Archaeological Monograph Volume containing the research papers given at this event titled : "Mellor: Living on the Edge, A Regional Study of an Iron Age and RomanoBritish Upland Settlement." In 2004 the Parish Centre was used to hold the Council for British Archaeology's regional annual conference which was hosted by the Mellor Archaeology Trust. The excavation is a key part of UMAU's commitment to community archaeology in Greater Manchester. In 2003 and 2004 this included working with Access Heritage and Stockport Parks on excavations on Strines Farmhouse at Brinnington in Reddish Vale. Each year the excavation allowed over 200 local school pupils to participate in an archaeological dig during the week. On the Saturday a 'Drop In' day attracted over 40 volunteers. Mellor Archaeological Trust and UMAU are working closely with Stockport Museum Services. The finds recovered from the excavation are being analysed and where appropriate professionally conserved. They will be stored at Stockport museum with many of them forming part of the 'Stockport Story' display, which will open to the public in 2006

University of Manchester Archaeological Unit December 2005.

3.

The Phygic& Setting

The site is cenrned anoUna Nadonal Grid Reference SJ 9818 8890. It lies in the parish o Mellor, appmxhmdy Sla m i h south east of the centre of Stockport (see figure 4). f
Sheet 98 of the aeoioglcal Survey of Great Britain shows the solid geology to be Westphalian A Sadstom f o the Late Carboniferous Period. In places this is rm overlaid by boulder clay.

?he site lies at the weat end of a promontory of land c.220m AOD. The promontory slopes quite sharply to the south. west and noah. To the east the promontory gently rises over a distance of 900x11t a unnamed summit a 278m AOD. o n t

~4:Theldcmof behalf of HMSO. Chw Iicence number 1003276S.

m-

parmioeianofOrdnanceSweyoa

- s m

Rcporl on the 2001 and 2005 sensons oforchoeologicol excovotions at Mellor.

4.

2004: Aims, Objectives and Methodology

The archaeological excavations at Mellor are designed as an evaluation programme to try and answer some fundamental questions about the site, its age, size and nature of the settlement on the hill top. Each year the results from previous seasons are assessed and a plan of excavation developed for the following season. In 2004 it was decided to excavate seven trenches each located to try and answer specific questions raised by excavation in previous years (see figure 5):Trench 26 was excavated to see if the large ditch located in Trench 18 continued north into Area C and to investigate the nature and extent of the curving gully identified within the 2003 trial trenches. It was also designed to test the indications from test pits and Trench 20 that there was high level of archaeological survival in this area. Trench 27 was excavated expose the enclosure ditch as close as possible to the north wall of the Old Vicarage were it formed a boundary between Area A and Area B. It was hoped that this trench would shed light on the discrepancy in height between the two areas at this point. In addition it was designed to confirm the alignment of the outer enclosure ditch with the large ditch found in Trench 1. Trench 28 and Trench 29 were designed to examine the level of archaeological survival in the north and west of Area C. Trench 30 was excavated to investigate a linear anomaly identified by geophysical survey within Area E (see figure 3). Trench 31 and Trench 32 were excavated adjacent to Trench 18 to confirm the presence of a palisade slot. Trial trench 16 ran south east to north west in Area B, close to and parallel with the boundary wall separating Area B from Area A. Its purpose was to look for indications that the large ditch found in Trench 18 might loop into Area B before turning south west into Area A to link with Trench 1. It would also provide a chance to assess the level of archaeological survival in the segment of Area A that lies inside the enclosure ditch. The difference between the trial trench and the other trenches is that it was never the intention to cany out any excavation in the trial trench. The methodology is simply to strip away the topsoil and subsoil and then observe and record what is revealed.

Universily of Monchester Archoeologicol Unit December 2005.

Figure 5: A plan showing the location of the #X)4 hemhes in red,previous years in blue, and the area & d g w b w at Menor, 1998-2004.

Repon on the 2004 and 2005 seosons o archaeological emvations of Mellor. f

5.
5.1

2004 Excavation Results.
Trench 26

Figures: 6,7, 8, 9, 10, 11, 12, 13 and 14.
The work within Trench 26 continues a process of archaeological investigation within Area C that began with a programme of test-pitting in April 2002. These test-pits, along with Trench 20 opened in the same year, suggested a high level of archaeological survival cut into the natural boulder clay of the area. In 2003, two trial trenches (6 and 7) were opened within Area C, c.5m north of Trench 18, with the purpose of both further evaluating the archaeological potential of the area and of ascertaining whether the large ditch excavated in Trench 18 continued north into this area. Whilst neither of these trenches established the presence of the ditch, both confirmed the same high level of archaeological survival revealed the previous year. In particular, a section of a curvilinear gully was found within trial trench 6. This bore similarities in its design to the round house gullies excavated within the Old Vicarage garden in Trenches 16 and 21. Lf this were the case, then the zone of occupation upon the hilltop would extend further than previously thought, as the evidence for Iron Age roundhouses so far had been limited to an area enclosed within the large ditch excavated in trenches 1 and 18. In order to c o n f m the preliminary findings from trial trenches 6 and 7, Trench 26 was opened during the summer of 2004. Trench 26 was quadrilateral in plan. Measuring 22.5-25m east-west by 2.3-8.6m north-south. A hand-dug east-west rectangular extension 7.4m long by 2.4m wide was added to the north-east comer of Trench 26 after initial cleaning had revealed a cluster of archaeological features within this area. The turf and topsoil (779) and subsoil (780) were carefully removed by machine excavator under the supervision of archaeologists. During excavation both layers produced sherds of 18"-20~century pottery as well as 18' and 1 9 ' ~ century clay pipe stems, glass and ferrous objects. Immediately beneath the subsoil was a layer of wet light-mid grey clay sand silt (781) with c.50% small-medium sandstone inclusions. This layer varied in depth between 5-25cm, effectively sealing the archaeological deposits below and seems to be an interface layer between the natural clay geology and the more humic soils above. No finds were recovered from (781) during excavation. The remaining layers and deposits were all excavated by hand. The natural geology of this area is composed of reddish brown boulder clay. The upper 5-10cm of this clay (702) was friable and regularly mottled, and lighter in hue than (703) due to weathering and rootlanimal action, and far more compact layer. In the western third of the site a very compact c.20cm deep layer (805) of probably glacially introduced small rounded and sub-rounded pebbles lay under (781) and over (702). A broken blue green glass bead of possible Roman date was recovered from (805) during cleaning. Trench 26 revealed a complex and densely packed arrangement of often inter-cutting features. The majority of these were typically denoted by their dark upper fills and the
University o Manckester Archaeological Unif f December 2005.

Report on the 2004 and 2005 seasons

o archoeologicnl excavarionr at Mellor. f

frequent presence of small-medium sandstone within them. These features clearly stood out from the natural boulder clay of the area through which they were all cut. A series of curvilinear gullies and sub-circular pits dominated the eastern and central areas of thc trench. To the west there was a noticcable change in the nature of the archaeological features, though the area still contained a high level of archaeological activity the majority appear to be postholcs. Due to the density of the archaeology within Trench 26 this report section has been broken down into fcature types i.e. gullies, post-holes and pits. Where possible relationships exist betwcen features of a different type this is commented upon.
5.1.2

Gullies

A total of 19 curvilinear gullies were discovered during excavation within Trench 26. Due to the inter-cutting nature of these features it initially proved difficult to establish the complete line of the individual gullies; therefore, some were assigned more than one cut number during excavation.

The extcnt of gully [565] was difficult to trace through the complex of inter-cutting features in the area with an estimated diameter of c.8m. It does appear to continue to the southeast outside of the excavated area. [565] was orientated southeast-north, possibly terminating in an unexcavatcd area directly to the east of gully 16381. It had a width of 0.45-0.55111, a concave base and a depth of 0.23111. [565] is possibly associated with gully [784] 2.4m to the north, which would create a northwest entranceway to a roundhouse, with an estimated diameter of c.1Om. This feature ran north-south for 1.9m, to the north section of the excavated area and had been broadly truncated by gully [637]. Gully [639/558] was curvilinear in plan and was orientated north south. It measured 0.48m wide and 0.15111 deep with very steep sides and a flat base. [639/558] had a rounded terminal (in common with all the gullies) to the north giving a known circumference of 9.6m. This feature cut gully [637] and was in turn cut by gullies [552], 15581, 15651 and [743]. It is possible that [639/558] is thc corresponding arm of gully [657] creating an entranceway facing northlnorthwest. This would give a width of l . l m for the entranceway and an estimated diameter of 10-llm for this roundhouse. [657] was orientated east-southwest. It measured 0.1 6m wide and 0.12111 deep and ran from its rounded terminal for 3.4m before being cut by the lincar grouping of pits to the east (see below). Feature [559] was a curvilinear gully 0.4m wide and 0.24m, with shallow urcgular sides and base. It ran for 3m north-southeast from its rounded terminal before being cut by gullies [637], [639], [552], [565] and [638]. [559] would appcar to correspond to the gully 18061 and together form a drainage gully for a roundhouse c. 11-12m in diameter. Gully [806] measured 0.3m wide and ran south-north for 2.2m into the north section of Trench 26. Gully [656] was orientated north-east. It measured 6.6m in length by 0.8m wide and had a maximum depth of 0.25111. There exists a possibility that gullies [808]/ [a091 and [767] are associated with [656] and together form the ground plan for a roundhouse gully measuring c.12m in diameter. Gully [808] was a 3.8m section of a
University o Manchesrer Archaeological Unit f December 2005.

Report on rhe 2004 and 2005 searom o archaeological excavationr at Mellor. f

seemingly curvilinear gully running east-west and terminating 1.4m north of the rounded terminal of [656]. This gully cut a similarly orientated featurc [809] and may represent a re-cut. Gully [767] lay within the extension to Trench 26 and ran northwest-south for 1.4-1.5m. Caution must be taken in any interpretation of this feature as it was unexcavated (see Trench 26 extension). The chance discovery of a curvilinear gully within trial trench 6 in 2003 was partly responsible for the excavations within Trench 26 taking place, though at the time only a small segment of it was visible. The 2004 excavation showed that this gully [638] ran south in a continuous arc from the northern section trench 26, 12.4111 to the far southeast comer of the trench (where trial trench 6 had located it). This feature ranged greatly in both its depth and profile, varying from 0.55m deep with a square profile in the southeast to 0.25m decp with irregular sides and base to the north. As with those gullies excavated within Trench 21 during the 2003 season, it would appear that dcpth was not an issue when it came to excavating a drainage gully for a roundhouse and that it was enough to ensure that there was a sufficient gadicnt for the gullies to work efficiently. Unlike the majority of the gullies' fills within the trench, [638] (along with [552]) had a high percentage of washed-in natural clay, especially within its basal fills. This would suggest that these two features were principally filled-in through a natural silting/erosion process rather than being deliberately in-filled during a rapid levelling of the site. These gullies are unusual in other respects, as [6381 is potentially the latest of the many gullies to be cut in this area (the other potentials are [7433 and [581], though [638] has the more stratigraphic relationships) and [552] may denote a different type of activity within the area (see below). Gully [552] was the most consistent in its profile of all those excavated within Trench 26 and the most puzzling. [552] had a square profile with very steep sides and a flat base, a width of 0.45-0.65 and a depth of 0.35-0.55. From its western rounded terminal it ran southeast for 6.5m partially cxiting the trench before curving back to run to the northeast for 8.7m and finishing in a rounded terminal. The distance between the two rounded ends of this gully is 8.7m The size of this entrance would argue against this feature from being a typical drainage gully for a roundhouse and probably signifies that [552] served some other function as yet unidentified. [552] cut the probable roundhouse gullies [565] and [639] and was in-turn cut by gullies [638] and [743] and so lie stratigraphically between different phases of roundhouse gully construction/renovation. It is possible therefore, that gullies [565] and [639] represent an earlier phase of roundhouse on the site which was followed by a break in occupation and a re-use of the area (perhaps to pen animals?) as dcnoted by [552] before reverting once more to the citing of roundhouses. Gully [552] serves as a reminder that as yet interpretations as to the nature and function of many of the features within Trench 26 must remain limited. Gully [637] (0.45m wide x 0.18m deep) was orientated north-east and had no discemable break for an entranceway along its exposed length. [637] ran for a length of 12.3m, giving a visible diameter of 7.5m, and extended to the north and east outside the area of excavation. This gully had an irregular base and steep sides, cut gully [784], and was in turn cut by gullies [638] and [579]. Gully [581] measured 0.3m wide x 0.12m deep. It ran for3.8m from its rounded terminal to the southeast before continuing north beyond the limits of excavation.
University o Manchesler ArchaeologicaI Unir f December 2005.

Repon on the 2004 and 2005 seasons o archaeological excavations at Mellor. f

This feature was the only example in Trench 26 OF a potcntial roundhouse gully with an entranceway to the south-east. The rest of the roundhouse gullies whose entrances wcre visible had them located to the north1 northwest. Whilst not unheard of, this tends to contradict the dominant tendency of roundhouses from other sites in this and other areas, which tend to have the entranceway to the southeast, perhaps to maximise the hours of sunlight and/or to shelter from the prevailing wind. This gully also strongly suggests, due to its distance from many of the othcr gullies as well as the unique placing of its entranceway, that a separate roundhouse to the oneis discovered within Trench 26 lies just to the north of the area of excavalion. [581] cut gully [657]. Gully 14731 was thc furthest west of all the potential roundhouse gullies uncovered within thc trench and was located at the extremc northwest of the cxcavated area. This feature mcasured 0.4m widc x 0.18m dcep and had concave sides and a rounded base. 14731 was orientated northeast-west and ran 2.2m before cxiting thc trench to to the east and west. A very similar featurc [6751 was ~mmediatcly the north of 14731 and one of these two gullies may represent a recut of the other. Taken together they C do appcar to suggest another roundhouse well to the west o the principal cluster of gullies and may symbolise that occupation continues to the north and west of Trench 26, though as these fcatures were only partly exposed and lay mainly outside the evaluated area assumptions as to their nature must remain limitcd. Feature [743] ran west-east from its rounded terminal for 2.3m before continuing beyond the confines of the trench. No corresponding return of this gully was observable and though there is a strong possibility that any such return has been cut away by the large series of pits lying just to the north (see below), it is possible that [743] is a linear feature rather than a roundhouse gully and unconnected with that grouping. Some circumstantial evidence for this may be inferred through both its lack of a return and its stratigraphic relationships to the roundhouse gullies [656], [552] and [637] through which it cut. Fcature [766] would appear to be the northern terminal of a roundhouse gully and runs northwest-south for 1.4m within the extension of Trench 26. This feature was unexcavated however and so no conclusions can be drawn.

Discussion
The curved nature of these features is highly suggestive of drainage gullies surrounding Iron Age round houses. These gullies are traditionally seen as running below the eaves of the roof, which projected beyond the walls of the roundhouse. Excavations at other Iron Age sites have found that these gullies were regularly re-cut in order to keep them functioning properly. In many cases slight realignments took place either as a form of repair or to reflect a change in layout or reconstruction of the buildings they served. In common with those gullies discovered within Trenches 16 and 21, the majority of these features had been in-filled with a dark grey clay silt, which regularly contained a high percentage of charcoal flecks and pieces. It would appear from this similarity that these gullies underwent a repeated process of rapid abandonment and/or infilling. Whether this process involved the deliberate deposition of charred remains (possibly from the hearth or other cooking site?) within the redundant gully, or whether it denotes that the house was burnt-down and built afresh on its new location is unknown. The presence of very small fragments of burnt and

-

University of Munchester Archaeological Unit December 2005.

Report on the ZOO4 and 2005 seasons o archaeological excuvurions at Mellor. f

calcified bone from within the fills of many of the gullies, combined with the absence of any clay daub would suggest the f i s t possibility is the likelier. No finds were recovered from any of the excavated sections of gully. The many gullies found within Trench 26 could not all have been contemporary and drained the same roof due to their inter-cutting nature and so must either reflect numerous phases of one house or separate ground plans of independent houses. It would seem likely that the roundhouse gullies discovered within Trench 26 represent two or three separate roundhouses. There also exists a possibility that thc nature of the occupation changed with gully [552] suggcsting that a distinct break in thc function of thc settlement occurred betwcen two separate phases of roundhouse constructions. The nature of the round house gullies identificd at the east end of Trench 26 probably indicates use and rc-use of thc same area for a singlc round house suggesting a long occupation of the samc picce of land for domest~c dwellings. Moving west along Trench 26 there is a 13m long area devoid of curving gullies with the possibility that this type of feature reappears in the North West comer of the Trench. This pattern was seen in Trench 16 excavated in 2002 where the curving gullies of the round house occupied only the west half of the trench the east half being occupied with a number of pits and post holes. Together the evidence from Trench 26 and Trench 16 seems to indicate that particular zones within the scttlcment are dcdicated to housing and that there are areas in betwcen these that are allocated to other uses. The potential uses of these spaces are unknown, but must use significant numbers of wooden posts as shown in Trench 26. Trenches 28 and 29 support the theory of an expanded zone of settlement indicating that the level of archaeological activity seen in Trcnch 26 continues north and west in Area C.

5.1.3 Post-Holes
A total of 54 post holes were excavated within Trench 26, with a further 16 identified post holes unexcavated. During excavation of several of the curvilinear gullies, a number of post holes (e.g. [704], [579], [567] and [429]) were discovered which had not been identified previously due to their upper fills' similarity to the fills of the gullies through which they cut. It is highly likcly therefore, that other as yet unidentified post holes lie within unexcavated segments of the curvilinear gullies. Many of the postholes wcre stone packed and contained flat sided small-medium sandstone, which appeared to have been set on-edge around the sides of the posthole. It is likely that these represent packing wedged into the posthole and had been used as a prop or chock to support a wooden post. This may suggest that thcse postholes required a solidity of support, which was not required of those postholes without such packing. Features of this type werc typically sub-circular or sub-oval in plan, measuring 0.45 x 0.4 with near vertical sides, a u-shaped or flat basc and a depth of 0.2m. Those postholes without stone packing tended to be circular and smaller in plan averaging 0.3 x 0.3m with v-shaped profdes and an average depth of 0.3m. During the Iron Age, the cutting of the defensive inner enclosure ditch and to a lesser extent the outer enclosure ditch would have provided a vast quantity of potential
Universily o Munchester Archaeological Unit f December 2005.

Reporl on the 2004 and 2 U seasons GS o onhueological excuvarions at Mellor. f

building stone excavatedlquanied during the construction of the ditches. Prior to this in the Bronze Age, and subsequently in the Romano British period ground fast timber structures may have been used. Indicating the possibility of classifying the post holes containing no stone packing as belonging to thesc periods. Though no ground plans were observed with the majority of the postholes, several did appear to form an arrangement. Feature [489] was a large (0.76 x 0.55 x 0.14111 deep) sub-square posthole with irregular sides and a flat base, in the western half of the trench, which contained frequent small-medium sized angular packing sandstone, and could easily be the base of a 2002 test pit. Feature [489] cut a large sub-circular posthole [686] on its eastern side (measuring 0.5 x 0.5 x 0.35m deep), with a vshaped profile. [489] bore similarities in its form and plan to posthole [490] 2.3m to the north, which was sub-circular in plan (measuring 0.76 x 0.65 x 0.22111 decp) and also contained frequent stones packed tightly together. This feature had also cut a large sub-circular posthole [701:1 (measuring 0.5 x 0.4 x 0.32m deep) with a v-shaped profile. The similarities in both form and placement between [489]1 [490] and also with [686]1 [701] would imply that they are associatcd. Their siting would suggest that they functioned as postholes for a structure orientated north-south. What form this structure took is unknown at present though it is possible that features [489] and [490] represent a different phase and substantial modification to the structure which postholes [686] and [701] had initially helped to form. The siting of [489/4901 would appear to mirror that of [686/701] and may be indicative of either a long standing structure which required new structural supports over time, or of a change in layout/reconstruction which necessitated more substantial stone packed postholes.

Discussion
In separating the features excavated within Trench 26 into various sub-categories there exists a tendency to see them as forming separate phaseslfunctions of occupation, whereas many are inter-related and represent different activities associated with settlement patterns. Perhaps this is particularly the case for postholes, for while gullies represent the outlying footprint of a roundhouse they do not form part of the structure itself, which would require wooden uprights to support the roof and walls. It is therefore, quite possible that many of the postholes represent the timber uprights of roundhouses and are associated with the previous sub-group of features. No definite ground plans were observed altogether. There does appear to be a general east-northwest arrangement in the postholes' orientation in the eastern half of the trench. This mirrors that of the roundhouse gullies and does suggest that some at least are associated with roundhouses. There are however, several potential problems with this interpretation. Only two of the potential roundhouse gullies in the east of the trench, where the majority of the gullies are, had not been either directly or indirectly cut by a posthole. Furthermore, out of a total of 70 postholes, 35 lay to the west of the trench and were seemingly unassociated with any potential roundhouse gullies. It is possible therefore that many of the post holes in the eastern half of the trench were also not associated with the gullies and may represent a different phase and character of occupation to them. They need not be associated; they could be outbuildings associated with the roundhouses and refers to a zone of ownership.

Univenify o Manchesler Arckaeologic~llUnir f December 2005.

Repon on the 2004 and 2WS seasons ofarchaeological excavalions al Mellor.

Additional areas of excavation adjoining Trench 26 are needed in order to gain a wider picture of the posthole ground plans and could prove highly informative on the changing forms of occupation within the area.

5.1.4 Pt is
A total of 16 pits, (with 4 more possible examples) were excavated within a substantial group of pits that dominated the eastern extent of the trench. These features were intercutting and formed an irregular linear arrangement orientated south-northwest 1-3m wide and 8.2m long that continued to the north of the trench. Two sections were placed across this arrangement in strategic places in order to gain an understanding of these complicated relationships. Both of these revealed a rather haphazard and seemingly unreiated sequence of inter-cutting pits, with very similar mid-dark grey clay silt fills containing frequent charcoal flecks and regular small and fragmented burnt bone. The pits were all sub-circular or sub-oval in plan, averaging 0.8m xlm, and most had a relatively shallow depth of 0.2m. Pits [749] and [751] were both substantially deeper at 0.5m and both had been partly in-filled with redeposited clay which overlay a light grey silt clay basal fill. It is interesting to note that both these pits were uncut by other features and together may denote a later phase of activity. Feature [482] was an oval pit in the western half of the site measuring 2m x 0.8m wide with a depth of 0.4m. Pit [482] was unusual not only in its shape and size, but also that it was the only large feature other than [747] and the possible feature [705], that was filled with friable light brown grey silt clay. The nature of this fill suggests that it was formed either through a slow process of natural erosion and silting, or that the features were rapidly backfilled with the originally excavated material. It is unclear as yet what this in-fill denotes but it is tempting to associate these features with an early phase of activity within the area, though until (705) is excavated (see below) interpretations must remain limited. A sub-oval patch of fire reddened clay [778] adjacent to gully [656] and within the arcs of many of the potential roundhouse gullies was initially thought to represent a hearth for one of the roundhouses sited here. Excavation of the feature proved inconclusive, but a similar small patch of reddened clay to the west [700] proved to be a small pit with frequent scraps of lead withtn its fill. The waterlogged conditions that Frequently occur on the areas of boulder clay would not be conducive to the survival of hearths. However if the interpretation of the c w i n g gullies as roundhouses is correct, then we may expect to see more evidence for hearths. If the hearths were raised upon a portion of abundant stone, then the likelihood of finding this surviving in situ would be rare. Pit [474] was identified early in the excavations due to the unusually large sandstone block (0.8m long x 0.6m wide x 0.26m deep), which appeared to cap the feature. Upon removal of the stone, [474] was filled with a compact charcoal rich mid-dark grey clay silt with frequent small fragments of burnt bone. A cylindrical bead was recovered during the excavation of this feature from its fill. [474] had near vertical sides and a rounded base and a depth of 0.45m. The pit was half sectioned only as it lay partly outside of the trench.
University of Manchesfer Archneological Unit December 2W5.

Reporr on the 2004 and 2005 seasons o archaeological excavarions ar Mellor. f

Discussion

This large group of pits represents a sustained and repeated use of this swathe of ground. It remains unclear as to what function this action served. The recovery of burnt bone and charcoal may suggest that the pits were dug to dispose of the residue of cooking and/or fires, though examples such as [749] and [751] would suggest that this was not the entire group's purpose. The pits' linear arrangement may suggest that this area was specifically appointed For this purpose or that it lay between areas of occupation/usage, which limited the availability of space. Of particular interest is the groups' relationship to the roundhouse gullies. Other than gully [657], which is cut by the line of pits, the separate groups appear to respect the layout of each other, though it seems unlikely that they are contemporary. Several sherds of Roman orangeware pottery from the 2" century A.D. were recovered from an unexcavated feature (812) within the line of pits, which would appear to suggest that the group is of RomanoBritish date rather than Iron Age. This raises questions on the dating of the roundhouse gullies themselves, as examples of roundhouse construction within the Romano-British period are well known. Further excavation to the north of Trench 26 in order to establish whether any more of the roundhouse gullies have been cut by these pits would help resolve this question.

5.15 Trench 26 Extension
During excavation within the east-west extension to Trench 26, a flint dagger was recovered lying flat upon the upper surface of context (705) immediately adjacent to the possible gully [767]. Context (705) was an unexcavated deposit of friable light brown grey silt clay, which may be the uppermost fill of large (>2.2m x >1.7m) pit. (705) had been cut by gullies [766] and [767] and by feature [768]. A grouping of 6 stake holes ([760]-[765]) were located along its western edge and although not definitely associated, shared the same fill type in contrast to the other features in this area. A small chert flake of possible Bronze Age date was recovered from its surface during cleaning. Feature [768] was a irregular deposit of large (<0.6 x 0.6 x 0.25m deep) sandstone fragments measuring 1.1 x 1.3m. It is unclear whether these formed a discrete feature or whether they are an isolated concentration of stones possibly infilling gully [767]. Context (769) was the number given to a concentration of small sub-rounded and flat angular sandstone, which filled the eastern third of the extension. This area was not fully excavated down to the natural geology due to time constraints and (769) may be a layer of stones within the subsoil. It is possible that the dagger is associated with either [768] or (705). which may be unidentified features of Bronze Age date, or with the gully [767]. It is also possible that the dagger was displaced from its original deposition during the construction of [767] or other feature.

Univeniry o Manchener Archaeological Unif f December 2005.

Daggers such a8 these are exhemely rare within the &eater M a n c W / area, w t the nearest pardel coming from Saddleworth (where a broken example ih was ~v~ at m ? ,classically associated with Beakm burials o the a ) and are f early Bronze Age. As ye#, tbe flint dagger cannot be assigned to any specific within Trench 26, due to its un-associated deposition and the limited area of excavation munding its find spot Due to the importance of t e find md the limited h time ~maining during this seasons' work to comprehensively investigate the area mund the dagger, it was agreed that excavation within the extension would cease and that the area would be a focus of the following season's w o k

Figure 7: Ovemead view of mmd house gullia located within trench 26, looking
west.

Eg 8: Brop~pe flint dagger discovered in the trench 26 extension. Scale 5cm. lm Age

Fieore 9: Tnach 2 6 nnmd

gulliw~ [552]. [565], [639]and [638l, looking west.

.

.

, * ,

-

t3"

-

,-

r

-

?

-*

8

-

*

t. --

.

.

I

Ngme 10: Overhead view of trench 26 extension, looking north.

8

Figure 12: Trench 26 plan, showing section.

P U X

=Fw
SmbN
Irl

-3,

M I .

Am r

St

6 w

-Q

=, ,

Xbn

Wl

-0 %

s l a

Flgme 14: Sections L - R h n Trench 26.

Report on the 2004 and2005 seasons ojarchaeological ercavationv at Mellor.

5.2

Trench 27

Figures: 15, 16 and 17.
Trench 6 which was excavated in 1999 had confirmed the presence of an enclosure ditch in Area B at a point some 10m to the north of the Old Vicarage wall. Trench 6 was one of series of 5 trenches which established the line of this ditch across Area B. They were by design relatively narrow trenches as their purpose was simply to confirm the presence of the ditch. Along with 3 larger trenches excavated over the ditch in the east half of Area B they provided a perfectly logical basis on which to project possible routes for the ditch south and east out of Area B. The natural southerly extension would link the ditch in Trench 6 with the section of ditch revealed in Area A by Trench 1 during excavations between 1998 and 2000. The ditch in Trench 1 was larger and deeper than the sections seen in Area B. However with the evidence available at the time it did not seem unreasonable to assume that this was a consequence of the ditch fulfilling different requirements at different points on its circuit of the hill top. The discovery of what was possibly a large post pit adjacent to Trench 1 raised the possibility that the ditch in this location was cut larger as part of an entranceway. Although only partially excavated in 1999, the ditch was visible in Trench 2 running into the churchyard of Saint Thomas's. Although excavation and geophysical survey are not practical within the churchyard it does seem likely that the line of the south arm of the ditch is marked by the eastward curve of the churchyard wall. In 2003 Trench 18 was excavated at the opposite end of Area A to Trenches1 and 2. Trench 18 contained a section of ditch comparable in size to that in Trench 1.Trench 26 showed that this section of ditch did not continue directly north across the drive of the Old Vicarage into Area C. This suggested that it curved west and followed the line of what is now the drive towards Trench 1. The implication of this was that there were probably two ditches in this part of the hilltop. A large inner ditch running between Trenches 1 and 18 following the line of the drive and the south churchyard wall encompassing the area occupied by the church and the Old Vicarage and a less substantial but more extensive ditch enclosing a much larger area of the hilltop. While it is believed that both ditches have their origins in the Iron Age there are difficulties in making a more precise interpretation of their chronological and stratigraphical relationship to each other. Although Iron Age pottery has been found in both ditches the lack of a regional typological sequence means that only the broadest date ranges can be assigned to these finds. In terms of excavation if the two ditches do merge then potential intersection probably lies somewhere below the Old Vicarage.

In order to investigate the possibility that the two ditches did not meet, Trench 25 was excavated just to the north-west of Trench 1. The purpose of this trench was to look for the outer enclosure ditch its presence would provide strong evidence that the two ditches did not join up. While there was no sign of the ditch the trench did reveal the fascinating fact that the flat nature of this part of the garden was the result of an

Universiry o Manchesrer Archaeological Unit f December 2005.

Report on fhe 2004 and 2005 seasom of archaeological excm~afiom hfdior. a1

extensive piece of possibly 19" century landscaping which had levelled up the natural slope of the hill. In 2002 Trench 17 was excavated in the field immediately to the north of the Old Vicarage wall. The purpose of this trench was to provide another section across the enclosure ditch and to see if any evidence survived of a track way running into the field from the Old Vicarage drive. While the surface of a 19th century cinder track was uncovered it was found that quarrying had removed any evidence of the enclosure ditch at this location. The need to investigate the extent of the quarrying and landscaping and the desire to gain as clear an indication as possible of the alignment of the outer enclosure prompted the excavation of Trench 27. The trench measured 5.2 metres wide and 14 metres in length and was orientated north to south. It followed a very gradual slope within Area B towards the boundary wall with Area A. Upon removal of the topsoil and subsoil natural bedrock (510) was encountered almost immediately. Cut into the bedrock the outer enclosure ditch [501] ran the length of the trench. At the north end of the trench it was 1.55 metres wide and 1.15 metres deep. These dimensions are very similar to those seen in other trenches to the north and east in Area B. As it ran south in Trench 27 the ditch became shallower and narrower. At the south end of the trench it measured 0.55m wide by 0.25m deep. The ditch was completely excavated in Trench 27 with 6 sections being recorded. Several similarities were visible between the sections. They all contained the same upper fill, (502) this was a mid brown sandy silt containing moderate amounts of small sandstone fragments. This was the only ditch fill present in the 2 most southerly sections. Context (502) did not appear in section 165 as this was the location of Trench 6 and consequently only contained backfilling material. At the north end of the trench, in section 173, fill (502) was 0.38m deep. This depth remained consistent through sections 171 and 168. In section 166, towards the north end of the trench, (502) was slightly deeper at 0.55m. Below this, in the three most northerly sections, was context (503). This was a distinctive mid orange brown sandy silt containing frequent small and medium fragments of sandstone. Below this, in the two most northerly sections, was fill (504). In terms of stone inclusions (504) and (503) were very similar they differed mainly in colour (504) being much paler with a yellowish brown hue. Discussion There is a discrepancy in height between the Old Vicarage and the part of Area B immediately adjacent to it, north of the boundary wall. Standing on the stone flags of the Old Vicarage when one looks over the wall the field of Area A is some 1.80m lower than the flag stones. The difference in size between the two ends of the ditch in Trench 27 might indicate that the ditch in the southern half has had the top 0.50m truncated. It is possible that this was because a segment of the slope in Area B has been terraced away. The excavated material could have been used to level up the slope within Area A and the boundary wall built to retain the levelling material and formalise the terrace. This would fit in with the evidence from Trench 25 of a deliberate landscaping programme to create a level garden on the east and north side of the Old Vicarage. No artefacts were recovered from the ditch itself; however a large quantity of post medieval pottery was recovered from the topsoil and the subsoil
University of Manchesler Archaeological Unit December 2005.

which could passlbQ b hdicative of the owners of the Old Vbuage depositing their waste over the wall QnBb the *d A similar abundance of post medieval pottery was I. found in 20M duriqg tbt excavation of Treoch 17. To the south of the trench theze was an area which the continuation of the p t medievrl m g identified within Trench 17. It L W b l e that the quarrying, termcing md hndscaping were part of the same works wah the q?larrying providing stone for the construction of the boundarylretahiq wall. The evidence from Trench 27 excavated over the enclotwe ditch seems to suppart tbe theory that the hilltop a the west end of Area A has been t extensively landscaped It would seem that the natural slope of the hill between A m A and Area B would have followed a line drawn bemeen the natural bedmck at the side of the ditch in Tmnch 1 to the natural bedrock appmximately half way along l h m h 27, a the side of the enclosure ditch t

Figrve 15: Tmnch 27 looking north east.

Elgme 16: Plan of %rich 27, showing sections.

Flgwe 1 : Sections S - X from trench 27. 7

Report on the 2004 and 2005 seasons ofarchaeological excwafions ar Mellor

5.1.1

Trench 28

Trench 28 was a 5.5m square excavated to establish whether the high level of archaeological remains present within Trench 26 was the same in the west of Area C. During machine excavation of the subsoil a sherd of Roman sarnien ware was recovered. This lay just above the stratified archaeological deposits which were identified at a depth of 0.5m below the modem ground surface. The features were located primarily to the south of the trench and seemed to represent a number of irregular inter-cutting pits containing pale grey silt fills. They gave a clear indication that the high level of archaeological survival seen in Trench 26 extended to this part of Area C. The archaeological features were recorded, drawn and photographed. However due to adverse weather conditions it was decided not to excavate any of them. Trench 28 was lined with plastic to define the archaeological levels and then backfilled.

5.4

Trench 29

Trench 29 was established to see whether the high level of archaeological remains present within Trench 26 was the same in the north of Area C. The trench was approximately 4.0m square. A number of archaeological features were revealed within Trench 29 including what appear to be small pits and postholes. They gave a clear indication that the high level of archaeological survival seen in Trench 26 extended to this part of Area C. The archaeological features were recorded, drawn and photographed. However due to adverse weather conditions it was decided not to excavate any of them. Trench 29 was lined with plastic to define the archaeological levels and then backfilled.

5.5

Trench 30

Figures: 16, 17, 18 and 19.
In 1998 a geophysical survey was canied out within Area E, to the east of the church car park, owned by Mr. L. h d e l . The survey appeared to indicate an anomaly running approximately east west and located c.25m into the field from the metal fence boundary. Excavation was conducted to determine the origin of the the geophysical anomaly and to establish if this was part of the outer enclosure ditch. An area 10.00m by 2.00m was manually de-turfed to a depth of approximately 0.2m a further O.lm of dark brown friable loam (584) was removed. Within this was found a dark grey flint flake and a single flake of chert. Beneath (584) a layer of loose orangey brown sandy loam (583) was removed by towelling. This lay on top of the plated sandstone bedrock, which was fiequently shattered due to the thin nature of the plates. In places the bedrock had decayed, resulting in a layer of coarse, loose orangey sand lying on the bedrock. A 1.80m wide ditch was located and subsequently excavated. The ditch [582] ran in an east west alignment consisting of steep sloping sides onto a flat base, 0.89m deep.

University of Manchesrer Archaeological Unit December 2005.

Trench 30 showed thrrt tfrc m m a l y w s the rermlt of human activity in the fom of a a .5 ditch 1 6 m wide and 0.83m deep similar in size and nahjre to the sections of outer enclosure ditch i&nt@i within the trenches excavated in Area B. At this stage any i n t e r p d o n as to tbe d.te and function of this featlule is highly speculative. However it is tempting to see it as pat of the same h n Age ditch system as the enclosure ditch in h a A. The diacovary of the ditch in Trench 30 w s very exciting and it is a tempting to associare it with the enclosure ditch in Area B. If it is part of the same hs Iron Age ditch system it would mean that nearly all the hill at Mellor is enclosed T i would have significant implications on the role that Mellor played within the Iron Age society in the mgion. However it is too early to draw any firm conclusions regarding the date and function of this ditch.

Figure 18: Outer enclosure ditch within tmnch 30, looking north w s . et

Fignre 20: Plan of T ~ n c h 30.

FSgare 21: SeaiansAA- AE fnrm trenches 30.3 1 a d 32.

Reporl on the 2004 and 2005 seasons of archaeological excm'arions a! Afellor

5.6

Trench 31

Figures: 21,22 and 23.
Trench 3 1 was designed to seek confirmation as to whether or not the stretch of stone lined gully found in Trench 18 in 2003 is a palisade slot associated with the large defensive ditch. Although the gully and ditch did run parallel to each other in Trench 18 only a relatively short stretch was revealed and the possibility existed that their proximity was just a coincidence. Trench 31 was roughly triangular in shape measuring approximately 8m by 6.5m by 5m. The results showed that the stone lined gully, OVM03 [302] found during the excavation of Trench 18 continued to run the full length of Trench 31, OVM04 [680]. The trench also revealed the northern continuation of the west edge of the large ditch.

Discussion
The alignments between ditch and gully in Trench 3 1 and Trench 18, which had been left open after the 2003 season, matched exactly and meant that we could now see the two features running parallel to each other for 10m. This means that we can say with some certainty that the two are associated with each other. The gully is linear in nature, c0.50m wide and 0.26m deep filled with large upright angular stones. The stones are predominantly the same as the bedrock excavated from the large ditch in Trench 18. No dateable material was recovered from the palisade slot. The most likely explanation is that the gully represents part of a defensive feature running inside the ditch. It may have supported timber posts designed to hold a bank or rampart in place or it may have been to hold a palisade fence, as yet no evidence has been recovered indicating the presence of a rampart. What is clear is that along with the ditch it would have provided a substantial defence for this part of the hilltop at Mellor.

5.7

Trench 32

Figures: 21,22 and 23.
Trench 32 was excavated 4m to the south of Trench 18. Its purpose was to confirm the suspected southern alignment of the large ditch within Area A. Trench 32 measured 8.lOm in length and was l m wide. No excavation was canied out in this trench as once the topsoil and subsoil had been removed the fill of the ditch stood out against the natural fractured bedrock clearly showing its alignment. Also visible in Trench 32 was the palisade gully again running parallel with the ditch.

Discussion
Trench 32 showed the ditch continuing along the same alignment as identified within Trench 18. A projection of this line would take it through the wooded area of the Old Vicarage Garden into the graveyard of St.Thomas's Church.

Universiry ofManchester Archaeological Unil December 2005.

F'igure 22: 'Ihe inner enclosure ditch and associated parallel palisade slot within Trenches18 32, the latter in the foreground, looking south.

mu*

-m.stw

C
m e

Trial trend^ 16

Trial Trench 16 was 1.7m wide and ran south east to noah west for 55m from the gate joining Area B to Area C. Its western terminus was designed to be the point whexe it located the enclosure ditch
The south east half of Trial mnch 16 revealed the layer of natural boulder clay. A small irregular pit and two small postholes were identified against the boulder clay. h The north west half of t e hid trench revealed the natural bedrock. No archaeological feaaues were identified in this half other than the enclosure ditch. This part of the Trial trench was then expanded north and south to form Trench 27. Pottery from the 18* and IFcenturies was recovered during topsoil and subsoil removal. The quantity of pottery inc& notably towards the north west end of the trench.

Figure 24: Trial mnch 16, eastenr end,looking east.

Report on the 2004 and 2005 seasons of archaeological excavafioru of Mellor.

6.

2004 Conclusions

This year's excavations within Trench 26 have improved our understanding of several key questions concerning the character of the occupation upon the hilltop in the Iron Age and Romano-British period. However as with all of the trenches excavated at Mellor in this and other seasons, many new questions on the nature of the occupation within these and preceding periods have arisen. The discovery of the Bronze Age flint dagger has raised interesting possibilities on the use of the site prior to the Iron Age and Romano-British occupation. The deposition of such an object upon the Mellor hilltop is unlikely to be by chance due to the high status of the artefact, although it is unclear as yet whether the recovery of the dagger denotes Bronze Age activity within the locale (possibly in the form of a barrow) or whether the irregular nature of its deposition represents activity not normally associated with such an artefact. A precedent for this last possibility may be found in the similarities to its unusual deposition with the late Neolithic polished stone chisel uncovered within Trench 16 during the 2002 season of excavations. This artefact was also discovered lying upon a surface layer (this time the natural boulder clay) and was seemingly un-associated with surrounding features. Excavation of areas adjoining Trench 26 in further seasons may help resolve these questions. One of the key questions to be resolved was the confirmation that the large ditch uncovered within Trench 18 did not continue to the north into this area. It seems likely therefore, that the ditch turns sharply to the west to the north of Trench 18, (possibly partially entering Area C), and runs west along the present day driveway for the Old Vicarage. A survey of the driveway using ground penetrating radar was undertaken toward the end of this seasons' work, which revealed that a large anomaly existed along its length. Though inconclusive, this evidence taken with the negative evidence from Trench 26 would suggest that the ditch on its northern side encloses an area similar to the present day Old Vicarage. Perhaps the most significant of the discoveries within Trench 26 was that Lron Age and Romano-British settlement of the hilltop extended into Area C. However, as with the curvilinear gullies first uncovered within Trench 16 in 2002, their form extends outside the area of excavation, therefore definitive conclusions as to their nature are not possible. However due to their similarities in form to those found in 2002 their interpretation as roundhouse drainage gullies seems justified. Due to the density of gullies it seems clear that settlement within Area C was repeated and long-standing. Previously, evidence for roundhouses had been limited to an area enclosed within the large ditch excavated in Trenches 1 and 18, within an area now occupied by the Old Vicarage and Mellor church. This had led to assumptions that the inner ditch possibly demarked a zone of occupation with the smaller outlying ditch serving as a stock enclosure, drainage ditch or boundary marker. The discovery of two small sections of inter-cutting gullies, located to the eastern comer of Trench 26 indicates the possibility of another set of roundhouse gullies. If this is another area of multiple re-cut drainage gullies then this would point clearly towards the use of particular zones for building and use of the land. Deliberate rebuilding upon some areas for roundhouses and not on others would indicate a
Universiry of Manchesfer Archaeological Unit December 2005.

Report on the 2004 ond 2005 seasons of archoeologicol excovarions or Mellor.

conscious decision to separate the usage of the land and its space over a prolonged period of time. This raises several interesting possibilities, not least of which is whether the two ditches are contemporary, andlor which of them encloses the Iron Age settlement. Does the settlement within area C denote an expansion into a previously un-settled area enclosed by the smaller ditch? Or is the occupation within Area A different in terms of function or status to that within Area C, necessitating a large ditch to denote its limits? The discoveries resulting from the 2004 season of archaeological works as usual produce not only many answers but new questions. Excavation suggests that the large inner ditch does not continue far north from Trench 18 and most likely turns to run west towards the Old Vicarage to join Trench 1. The palisade slot indicates that on the inside of the ditch there would have been a rampart or palisade. It is tempting to suggest that this defensive arrangement would have run along what is now the Old Vicarage driveway, turned south to join the ditch in Trench 1. Previous excavation has located the ditch and palisade gully running out of the south west comer of the Old Vicarage garden. From there it may well swing round to run east following the line of what is now the south wall of the churchyard before turning north through the churchyard to rejoin Trench 18. At the end of 2003, with the discovery of the round house gully in Area A and the possibility that this area lay within part of the hilltop defended by a substantial ditch and palisade, it was thought that this might represent the limit of the zone of human occupation during the Iron Age at Mellor. This theory would have people living within the defended area with the zone between the defensive ditch and the enclosure being used for agricultural purposes. However if the suggestion that the defensive ditch runs along the line of the Old Vicarage driveway is correct then the round house gullies found in Trench 26 in 2004 clearly indicate occupation outside of the defended area. Lastly, the tantalising potential offered by Bronze Age flint dagger promises next seasons excavations to be especially informative.

University ojMonchesrer Archneological Unit December 2005.

Report on rhe 2004 and 2005 seasons oforchaeologicol excavorions a1 A4eIlor.

7.

2005: Aims, Objectives and Methodology

The archaeological excavations at Mellor are designed as an evaluation programme to try and answer some fundamental questions about the site, its age, size and nature of the settlement on the hill top. Each year the results from previous seasons are assessed and a plan of excavation developed for the following season. In 2005 the archaeological excavations consisted of two large open area excavations, and a further seven trenches designed to answer the specific questions raised during the excavations of previous years (see figure 25). Trench 33 was designed to verify the continuation of the inner enclosure ditch identified by geophysical survey undertaken within the Old Vicarage garden in 1999 within trenches 1, 2 and 18. The trench proposed to answer a number of questions relating to the ditches excavated in Trenches 1 and 2, the continuation of the inner enclosure ditch, the positive identification of an associated palisade slot, and identification of any remains of intemal settlement. Trench 34 was designed in order to assess the potential for internal settlement in close proximity to the possible post hole structure identified within trench A. Trench 35 was located in order to evaluate the true nature of the post pit identified within the trench one extension and whether or not this was part of a larger structure or remained an individual feature. Trench 36 was a large open area designed to identify the continuation and extent of the curvilinear gullies identified within Trench 26. Assessing the immediate area surrounding the location where the flint dagger was recovered, for any associated structure or possible burial. Trench 37 was placed to confirm, to the furthest assessable point within the Old Vicarage boundary, the expected line of the inner enclosure ditch and its associated palisade slot. Trenches 38 and 39 were designed to expand the known line of the outer enclosure ditch, either side of that identified within trench 30 and positioned according to the results of a geophysical survey conducted in the spring of 2005. Trench 40 was conceived to excavate the possible ditch terminals and entrance way of the outer enclosure ditch, located but not excavated within trial trenches. Trench 41 was designed to further enhance the known extent of the outer enclosure ditch within Area D and placed over the trial trench. In 2005, trenches 33,34, 35 and 36 were all uncovered by machine, the remainder, 37, 38, 39,40 and 41 were all exposed by volunteers by hand. Subsequent excavation of all trenches resulted in hand excavation conducted bv volunteers under the supervision of three professional archaeologists, supplied by the University of Manchester Archaeological Unit and one experienced amateur archaeologist provided by the Mellor ~ r c h a e o i o ~ i c a l .
University of Manchesrer Archaeological Unit December 2005.

Figam 2 : A plan showing tbe location of the 2005 t r e w in red, previous yerus in 5 blue, and the a deeipartbas a Mellor, 1998-2005. m t

Report on the 2001 ond 2005 semons of archaeological ercuvations at A4ellor.

8.
8.1

2005 Excavation Results
Trench 33

Figures: 26,27,28,29,30,31,32, 33, 34, 35 and 36.
During the 1998, 1999 and 2000 seasons, excavation was conducted over a geophysical anomaly identified at the western end of Area A. The subsequent excavation of Trench 1 identified a large rock-cut enclosure ditch measuring 4.00m at its widest point and 2.10m deep from the present ground surface. Trench 2, located in the south western comer of the garden, identified the continuation of the inner enclosure ditch and the presence of a possible associated palisade slot. Trench 33 was specifically designed to clarify whether or not this ditch could be identified in between Trenches 1 and 2. By excavating a greater area within the garden than previously attempted, it was anticipated that not only would the nature and extent of the ditch be revealed, but that any surrounding archaeological features which could possibly relate to the ditch, such as the continuation of the palisade slot, and any internal features indicative of an entranceway and associated structures would be identified. The trench was machine excavated to a height immediately above the archaeological layers and the remainder hand excavated. Measuring approximately 19.6m by 12.0m, the irregular outline and shape of the trench was determined by the need to protect as much of the garden and accompanying trees as possible. The geology of the trench varies across its entirety. An area of solid sandstone bedrock was identified within the central and eastern segments of the trench, overlain in the majority of the remainder of the trench by fractured bedrock and mottled natural sand, the remainder was a silty sand extending to the western and southern portions of the trench. The material represents the remains of an ancient river delta and therefore differs over the area as the processes upon the river altered. The fractured nature of the bedrock made identification of the smaller features problematic, as it was to establish if a feature was man made and shallow, heavily truncated or just a naturally occurring undulation within the make up of the fractured geology (see appendix 5). The natural solid geology is a sandstone known locally as Woodhead Hill Rock, the lowest sandstone development in the Westphalian A succession, laid during the late Carboniferous Period. Within Trench 33, many archaeological features were identified, dominated by the ditch to the west, the continuation of the palisade slot, two separate rows of post holes, three post pits to the east and four horticultural trenches to the south, and a number of unassociated separate features.

8.1.1 The Inner Enclosure ditch.
In 2000, a geophysical survey undertaken by Geoquest Associates (see figure) identified a linear anomaly running roughly north to south arching along the western
Universily of Manchesler Archae'blogica/ Unil December 2005.

Reporr on rhe 2004 and 2005 seasons ofarchaeological excavations at hJe1lor.

extent of Area A between Trench 1 and Trench 2 within the garden. The excavation of Trench 33 showed the anomaly to be the same ditch as seen in Trenches 1 and 2. To the north end of Trench 33 it was possible to excavate a 4.00m long section across the full width of the ditch (A). The excavated section A, was sub-square in shape, 3.10m along the south side and 4.05m along the north, 2.23m to the east and 4.72m to the west. Initial excavation of the topsoil and subsoil in this area was achieved through the use of a six tonne machine using a ditching bucket. Subsequent removal of the remainder of topsoil and further trench enlargement was achieved by hand. This showed it to be 3.10m wide, and 1.95m deep. As the ditch angled to the south west its western edge disappeared underneath the western wall of the old vicarage garden. Therefore slot B, 4.50m to the south of A, and slot C, a further 1.8Om to the south, were only partial excavation of the upper fills of the ditch. Slot A was 3.10m wide at the top, and 1.95m deep, from the top of the archaeological layer to it's 'v' shaped base, in total a 4.40m wide area was excavated and stepped Slot B was 1.30m wide and 1.56m long, at right angles to the trench edge. Running along the eastern edge, the angular cut for the inner enclosure ditch [007] was identified. Identical to that within slot A, the ditch is cut into plated bedrock and quickly descends in a steep angle towards the centre of the ditch. Only partially excavated for reasons of space and health and safety, the maximum depth achieved was 0.42m. Slot C was 1.30m wide and 0.72m long, at right angles to the trench edge and almost identical in results to those found within section B. Both sections possessed identical fill and layer morphology to that identified within the main slot A. Any dateable material recovered from these sections was grouped together under the relating fills of [007]. In profile the ditch is unlike those seen in Trench 1 and Trench 18 in that it does not possess a wide flat base, however it does have a large irregular 'v' shaped base cut into the angular bedrock. The trench is cut into the natural bedrock and as a result the immediate direction of the ditch is dictated by the natural fractures within the bedrock, these fractures are different from those in trench 1 and trench 18 where the plated bedrock was flat whereas, within this trench the plates occur at approximately 30 degrees toward the base. The primary and secondary fills of the ditch [007] are (037) and (095) respectively. These fills contain a relatively high proportion of small to large fragments of subangular sandstone. It is therefore suggested that these were formed naturally, as material collapsed into the ditch from the sides and possibly the bank or palisade. The ditch then appears to be re-cut by [088], which cuts into (037) and (095). (031) the fill of [088] also represents a period of natural silt and stone build up. This in t r un is then re-cut by [089], it is at this point that the deposition of fills within the ditch section would appear to alter. After an initial period of silting (031), a large deposit of (022) large angular fractured bedrock was identified, with the stones all tilted at a 30 degree angle. The stones appear to tip into the ditch from the outer side.

University ofManchester Archeological Unit December 2005.

Report on the 2004 and ZOO5 seasons of archaeological excavations at MeNor.

Above this there is another smaller tip line present (021), containing a similar type of stone, however much smaller in size, but respecting the same tip lines and directions as those identified within (022). The remainder of the ditch is then backfilled with a considerably less stony fill. Possibly indicative of a period of abandonment, however this could also represent a period of deliberate levelling of the site. A small amount of roman pottery was uncovered from the upper most fills of the ditch, this was, however small in comparison to the amount retrieved from the upper fills of Trench 18. The lower fills of the ditch contained no finds. 70% of all ditch fills were hand sieved and produced only small flecks of charcoal and small unidentifiable fragments of burnt bone. No environmental samples were taken due to the extremely poor level of organic survival. The ditch fills (Figure 33) were as follows :The uppermost fill of [007] was (020) a dark brown friable sandy loam. Fill (020) had a maximum depth of 0.40m and lay immediately below the subsoil. Occasional angular and sub angular material ranging from 0 - 0.10m were present along with occasional flecks of charcoal. Multiple fragments of Roman pottery were recovered from this fill. Fill (021) was a friable middark brown grey clay silt containing very frequent large angular 0 - 0.20111 stones, occasional charcoal flecks and occasional small undiagnostic flecks of burnt bone, with a maximum depth of 0.25m, containing a few sherds of Roman pottery. Fill (022) consisted of a compact dark grey silty sand containing very frequent large angular and sub-angular 0 - 0.50m stones sitting in a horizontal position, along with occasional flecks of charcoal and a maximum depth of 0.22m, containing a few sherds of Roman pottery. Fill (031) was a friable light brownish grey silty sand containing frequent 0-0.20m angular and sub-angular stones and a maximum depth of 0.40m. No finds were recovered from this fill. The primary fill (037) was a fiiable light brown silty sand consisting of over 80% 00.20m angular stones, with a maximum depth of 0.45m. No finds were recovered from this fill. Fill (092) constituted a friable mid brown sandy loam containing very few 0-0.05m rounded inclusions existing to a maximum depth of 0.14m. No finds were recovered from this fill. Fill (093) consisted of a re-deposited natural layer of coarse light orangey brown sandy silt, with a maximum depth of 0.10m and being predominantly made up of small rounded 0-2mm stones. No finds were recovered from this fill. Fill (094) was a friable light brown sandy loam with a maximum depth of 0.20m and containing very few stone inclusions.

Universiry of Manchester Archaeological Unit December 2005,

Report on the 2004 and2005 seasons of archaeological ercavorions at Mellor.

Fill (095) was a friable mid brown sandy loam containing frequent 0-lOmm rounded inclusions and a maximum depth of 0.15111. Fill (096) was compact light grey sandy clay containing frequent regular rounded and sub-angular 0-20mm inclusions and a maximum depth of 0.20111.

Discussion
The excavation of the slot A across [007] has shown that the ditch does indeed continue unbroken between Trenches 1 and 2. Yet again the inner enclosure ditch is a substantial feature and would have been 0.76111 deeper that identified in Trench 1. The main difference is the profile of the original cut; where Trench 1 possessed regular cut sloping sides onto a flat base in slot A, in the distance of 8.50m the profile the ditch in slot A becomes more 'v' shaped, possessing angular undulating fractures within the bedrock and as a result step sloping sides and an undulating narrow base. The difference between the two separately excavated ditch sections, Trench 2 and Trench 18 is that the fractures within the bedrock are entirely different, within Trench 1 and 18 the bedrock was plated horizontally, therefore it would have been extremely easy to cut a trench down onto a flat base and keep very regular sides, whereas if the plating is at a steep angle, such as that found within trench 33, then it becomes almost impossible to excavate a neat flat bottomed ditch. The direction of the ditch in the immediate locale appears to follow the paths allowed by the natural solid bedrock, and therefore the ditch follows the direction of the fractures for as long as is possible, until the direction of the ditch becomes the overriding concern. The two re-cuts of the ditch would appear to represent large scale maintenance of the ditch, cleaning and reinstating as needed. However it would appear that after an initial period during which fill (022) accumulated, a deliberate phase of backfilling occurs after the re-cut [089]. This may relate to the suspected deliberate phase of backfilling identified previously within trench 18 and possible closure of the ditch at some point during the 4" Century AD in Trench 33. The levels and deposits of this period of backfilling are almost identical in nature to those found within Trench 18. They would seem to demonstrate a wholesale deliberate backfilling of the ditch at a similar time along its entire length. The presence of such large and frequent fragments of redeposited angular sandstone stones would indicate the possibility the deliberate destruction of a wall or structure which would have stood on the internal edge of the ditch maybe even the remains of a rampart. The discovery of a large amount of Roman artefacts recovered from Trench 18, and in particular, their form of deposition, suggests that material was tipped into the ditch and combined with the charcoal and daub, hints that such rubbish deposition could be associated with the main abandonment of the site in the 4" century AD. The differences in quantity and quality of Roman pottery recovered from Trench 18, in comparison to Trenches 1 and 33, indicate that settlement in the Roman period may have been concentrated towards the eastern end of Area A. The continuation of the ditch proves that there was no entrance way located on the western side of the inner enclosure ditch. Due to the difference in material recovered from the excavation of the different trenches across the inner enclosure ditch, and its
University of Manchester Archaeological Unit December 2005.

Repor1 on (he 2004 a d 2 0 0 5 seasons of archaealogicol excavations 01A4eNor.

continued presence within the Old Vicarage garden, there is a strong possibility that the entrance way would have been located to the east, near Trench 18, and therefore taking advantage of the natural flat area of land, as the hilltop begins to rise.

8.1.2 Palisade slot
During the summer of 1999, the excavation of Trench 2 and its subsequent extensions a, b, and c in 2000, identified the remains of a vertical sided and flat based linear feature [2005]. Interpretations at the time included a palisadelrampart slot or a beam slot for a possible building foundation. The Trench 1 extension was designed to identify the presence of a palisade slot associated with the ditch, this could be for two reasons, it does not exist at this point of the ditch, or that due to the small width of the trench and the bedrock infill of the palisade slot, identification of this feature would have been very difficult. The positioning of Trench 33 over and around Trench 2 and its extensions allowed for a re-assessment as to the extent of this feature. In 2005, the earlier excavated slots across the palisade slot were re-excavated and analysed. The feature was identified running from the southern trench edge in a northerly direction for 7.10m. It was approximately 0.60m wide and 0.20-0.30m deep. The palisade was found to consist of near vertical edges and a flat base and filled with large 0-0.30m angular stone packing material. The northern extent of the palisade slot [055] was characterised by a circular terminus. Projecting this line further to the north the remains of a small rock cut gully [097], parallel to the ditch, was identified protruding from the northern trench edge and running under the unexcavated area surrounding the small tree, where it did not emerge from the other side. The dimensions of this feature are 0.10m wide, 0.15m deep and 1.15m long.

Discussion
It is tempting to see these two linear features as related and forming parts of the same palisade slot, with a possible gap in between. If this were the case then it may explain the reason for why the western alignment of postholes occurs in this area. With the group of post holes arching slightly offset fiom the terminal end of the palisade slot. The gap in the palisade would allow for access into the ditch, possibly for cleaning and maintenance. This could not be the main entrance into the inner enclosure ditch because it is known to continue uninterrupted in this immediate area. This leaves the only possibility of an entrance way as a bridge over an uninterrupted ditch. There is a possibility that the palisade slot does continue, but due to the more solid bedrock natural that is encountered within this area the slot was shallower and therefore is not archaeologically visible. The continuation of the palisade slot indicates that the absence of any such feature continuing parallel to the ditch within the Trench 1 excavations could be due to the small 0.30m wide slot and as a result the subtle differences between the fractured bedrock and the cut for the palisade slot would not be overwhelmingly obvious in such a small section.

Universily of Manchesler Archaeological Unit December 2005.

Report on the 2004 and 2005 seasons of archaeological excavations at Mellor.

8.1.3 Post pits
A total of three post pits were identified, [OOI], [Oll] and [032]. All were found located towards the central eastern portion of the trench. [OOl] was an irregular sub circular post pit, half sectioned and cut into bedrock to the eastern edge of the feature, and towards the west cut into fractured bedrock, sloping steeply onto a flat circular, base. Measuring 1.35m in diameter and 0.52m deep, a 0.60m wide central post pipe was identified, surrounded by dark brown silty sand fills containing a number of large <0.35m solid angular packing stones. A single sherd of a sandy ware pottery was recovered from the upper portion of the post pipe, in the form of hollow ware and dating to the medieval period. To the south [032] was a similar irregular sub-circular post pit, 0.13m wide and 0.45m deep. A 0.30m wide vertical post pipe within the central to western portion of the pit was half sectioned to reveal a significant deposit of <0.20m angular stone packing placed around the post. Fully excavated to show the full extent of the pipe, the post pit was then half sectioned to reveal irregular sloping sides cut from the fractured bedrock onto a flat circular base. No dateable archaeological artefacts were recovered from this feature. To the east of [032], post pit [Ol 11 was constructed in a similar form to that of [032] and measured 1.40m in diameter and 0.61m deep. Containing a 0.40m wide post pipe, the pit appears to have been predominantly filled with re-deposited natural material similar to that surrounding the pit. From the lower portion of this fill a small fragment of a fine gritty ware was recovered, in a hollow ware form and dating from the late 11" to early 1 3 ' ~centuries. Close by but located within an unstratified deposit, a single sherd of a buff gritty ware jar 1 cooking pot was recovered, dating from the same period.

Discussion
The presence of these three post pits indicates a possible relationship with those identified that found in the Trench1 extension and those identified within trench 35. It is entirely possible that they all relate to one larger structure, the extent of which is currently unknown, or that they relate to a separate structure, possibly a small four post structure standing independently of those in trench 35. However it would appear that the dating of two fragments of pottery recovered from [OOl] and [Oll] and the unstratified fragment, would indicate a similar date in origin to those identified within trench 35. If they do relate, then the expected alignment would be a north east, south west alignment, taking into account the average spacing between them this would allow for a postulation of a further post pit to be located in the far north eastern comer of the trench. However only a single posthole was present, and raises the possibility that the there may have been no need for a pit. It is certain, due to the solid bedrock that there are no post pits to the immediate south, and raises the possibility that this is the furthest extent of the building. There were flat areas of apparently shaped bedrock, but containing no relevant depth, near the post pits in Trench 33. There is the possibility that such features could
Universip of Manckester Archaeological Unit December 2005.

Raporl on the 2004 ond 2005 seasons o archaeological excavoriom of Mellor. f

provide support for post which may have been present, these would have been very low load bearing posts. Similar features identified within Trench 35 such as [035] are very shallow, is the requirement for these that the need a flat solid base? Throughout the excavations in the garden it has proved difficult to identify the smaller features, one reason for this, other than the mixed and complicated undulating angular stone natural is that these features may have only needed to have been relatively small, and therefore as soon as bedrock was reached, excavation stopped, providing little if no evidence of their presence within the archaeological record.

8.1.4 Post hole alignments
Within the north eastern extent of the trench lies an alignment of four post holes in a north north west and south south east direction. The four [080], [041], [089] and [018] are all slightly different in form. The most northerly [080] was very shallow, 0.03m, 0.35m in diameter and cut into a plate of bedrock with vertical sides and a flat base. The second [041] was oval in shape, 0.58m long, 0.38m wide and 0.15m deep, cut into the bedrock. This had vertical sides and large angular vertical packing stones surrounding a central fill. The third and fourth, [089] and [018] were identical in nature; circular, c.0.35m in diameter with undulating steep sides forming rounded bases at a depth of c. 0.20m. Within the western portion of the trench, in between the inner enclosure ditch and the palisade slot, a group of six post holes are aligned roughly parallel to the inner edge of the ditch: [099], [047], [053], [061], [064] and [060]. They are all very similar in nature, only varying in diameter from 0.37m to 0.47m and in depth from 0.13m to 0.19m. All were sub-circular with irregular undulating sides gently sloping into rounded bases. Discussion Due to their relative alignments with each other, it is assumed that the fust four post holes described above are part of the same structure. It is not clear if the alignment of posts would have been anything other than just four posts in a line, however it is more likely that the other associated post holes remain to be discovered. The alignment is not parallel or at right angles to any other group of features identified within the immediate area of the garden, therefore the interpretation as to the use and potential date of the alignment must be independent of surrounding features. No dating can be implied from these post holes due to the absence of dateable material recovered during excavation. It can be inferred that all six of these postholes within the second alignment relate to each other to fonn part of a larger feature. Possible interpretations for this structure must include a reference to the similarity in the orientation of the post holes in relation to that of the ditch. It is possible that such a series of posts are associated with a possible bank on the inner side of the ditch, holding that bank in place, or due to the apparent abrupt end to the palisade slot, be evidence for a small overlapping gap in the bank or palisade to allow for access into the ditch. A further possibility is that they do not relate to the Iron Age ditch and subsequent periods at all and that, as the alignment does follow the same orientation as the possible medieval post pits identified within Trench 35, it is entirely feasible that they are associated with the
University o Monchesler Archoeologicol Unil f December 2005.

Report on the 2004 and 2005 seasons of archaeologica~ excavations at A4ellor.

medieval building or be part of an entirely separate structure. Similar to the other posthole alignment in this trench, no dating evidence was recovered from any of these post holes. The original construction of these rock cut features is primarily concerned with the identification of a fracture within the bedrock, the bedrock is then smashed with a hand held rock at this point of least resistance. The area is then expanded by removing the bedrock plates as they fracture off to the required shape. It is expected that the post pits remain a particular shape and size, not only to fit the post but as an accessible size to dig in a circular motion from within.

8.1.5

19'~ Century Horticultural Trenches.

To the south of the trench a series of four parallel gullies were identified. These trenches are approximately 6.29m in length and 0.37m wide c.0.05m between each one and orientated in an north north west - south south east direction. The depth of these trenches was c.0.05m and were all filled with a dark brown sandy loam, possessing a high organic content. It is possible to state that these features originate from the early 19'~century due to the range of pottery and clay pipes that were recovered from their respective fills. Within the trench edge cuts into the topsoil for these features can be seen in section, starting at a depth of c.0.20m below the present ground surface. It is known that trenches of this depth are dug for the growing of various types of vegetables, what is not clear however is why they dug down into the fractured bedrock, for the sake of a further 0.05-0.10m.

Universiy of Manchester Archaeological Unil December 2005.

Report on he 2 W ond 2W5 s e a m of arrhaeolegical a c ~ o l t o n r MdIor at

Fipre 28: Inner enclosure ditch, [007], within Trench 33, looking south.

Figure 29: Trench 33, post pit [Oil], showing post pipe to the west, looking nonh east.
Uniwrsiry o Manchutcr Archaeologkd Unit f
Lkcenthr 20DS.

Figure 3 : Trench 33,stone packed post hole [099] looking North West. 0

Figure 31: Trench 33:volu11teem assisting with the excavation.

Figure 32: Medieval armw head. X-ray before conservation and removal at b s and ae after conservation and stabilisation above. Found within Trench 35, I 6 1 (066). 05

University o M~ehcsrer r c ~ i c r r Unit f A l Deccmbzr 2005.

Repmi on & ZOO#and 2 W S semom o f o n h o m ~ i c a c~favolionr Mellor. l ur

Npre 33: An enlarged plan of the geophysical analysis of the eastern extent of Area A, conducted by Chquest Associates.

Oniwrsiry ofMomkcater A December ZOOS.

r

o Onif

O W 05 Trench 33 Plan

Om

N+

-

Im

Figure 34: Plan of trench 33.

Repon on the 20W and ZOO5 seaJ01U o archaeological cr~nvarionr Mcllor. f a1

Figure 35: Combined plan of the Trenches excavated within the Old Vicarage Garden, Area A, Including TRnches: 1, lext, 2.33.34 and 35.
University ofManChester A t r k o l o g k a l Unit Decmbar 2005.

Repon on h e 20W and NX)S seavons o urchueologicul acuvorionr at Mrllor f

Figure 3 : Trench 33 sections A- G. 6

Report on rhe 2001 and 2005 seasom of archaeological e x c a v a r i o ~ Mellor. or

8.2

Trench 34

Figures: 37 and 38
Located 3.25m directly to the south of Trench 35, this trench was placed to identify any archaeological features relating to settlement found located within the interior of the inner ditch. Trench 34 was 3.00m wide by 5.75m long and 0.25m deep removal of topsoil and subsoil exposed a large area of mixed natural, containing plated bedrock, fractured bedrock and a mottled silty natural. Two archaeological features were identified and excavated.

8.2.1 Post pits
Running into the eastern edge of the trench a small sub-circular pit was located and half sectioned at right angles to the trench edge. [084] was filled by a mid-brown silty loam containing a possible post pipe 0.24 m wide, towards the western extent of the post pit, consisting of a dark brown silty fill. No archaeological dateable material was recovered from this feature. Located towards the northern end of the trench an irregular circular post pit [082] was identified, excavated in quadrants due to the uncertain nature of the fills and the surrounding natural, and finally fully excavated. The pit, an irregularly oval in shape, with undulating sloping sides, was cut into the natural fractured bedrock and possessed an uneven base. The fills consisted predominantly of light-mid brown silty sands containing frequent small angular stone fragments consistent with re-deposited smashed sandstone.

Discussion
The two possible post pits [084] and [028] may be associated to those identified within Trench 35, however they do not appear to relate to the alignment found in Trench 35. Either because the post pits are of a different period or phase, or relate to a completely different structure. They may represent the most southerly elements of two further rows of post pits running parallel to and north of those found within Trench 35, and that together they form part of a large group or structure. Further excavation of the remaining areas within the garden could identify the components and limits of such a structure.

Universiw of Manchesler Archaeological Unit December 2005.

Flgare 3 : Trench 34, looking north. 7

Figare 38: Plan of trench 34.

I
Vnimity ojMonckcrta Atckuhgical Unir DQcmrbrr ZOOS.

Report on the 2004 and 2005 seasons of archaeological excavarions at Mellor.

8.3

Trench 35

Figures: 39,40,41,42,42,43 and 44.
Trench 35 was positioned 4.34m to the west of Trench 33 and c1.00m immediately south of post pit [035], identified within the extension of Trench 1 excavated in 2000. The trench was specifically placed in order to evaluate the potential presence of further archaeological features within the surrounding area that would relate to and assist in the interpretation of post pit [035] and identify the archaeological period which the feature belongs and if this relates to part of a larger more significant structure, as yet undefined. The trench itself had to be expanded on several occasions to expose the entirety of certain features and concluded approximately 10.7m in length by 3.20m wide. Geological levels were encountered at a depth of 0.42m below ground level in Trench 35. It was very similar to that located in Trench 33, with fractured bedrock towards the south and solid plates of bedrock in the northern half. The trench identified four large post pits, three postholes and a possible beam slot, together with a previous test pit. The pits form a north north east to south south west alignment, of which [035] forms the northern most and [079] the southern most identified excavated limits of the structure.

83.1 Post pits
Post pit [013] was a sub-circular in shape and measured 0.96111 by 1.00m by 0.36m deep. It had steep sides cut into the natural bedrock and a plated base and containing a c.0.41m wide post pipe. The post pipe was made up of loose dark silty loams containing occasional fragments of charcoal and angular fragments of stone whilst the remaining packing fills predominantly consisted of large angular inclusions of redeposited smashed bedrock. Originally half sectioned and later fully excavated, the pit is cut by a smaller sub-circular posthole [109] located to the southwest extent of the post pit. A single large square headed iron nail, possibly medieval in origin was identified within the lower fill of the post pit and no other archaeological artefacts were recovered from the posthole. Post pit [033] was a sub-circular post-pit cut out of the plated bedrock, 0.90111 by 0.80m by 0.15m deep onto a flat bedrock base, and containing evidence for a 0.37111 wide post pipe towards the western edge. The northern-most edge was formed from a natural east-west running fissure in the plated bedrock. The post pipe contained a dark organic fill with occasional charcoal fragments and packed with large angular fragments of broken plated sandstone. Two sherds of 13" -15" century medieval pottery were recovered from the lower fill within the pit, initially half sectioned and iastly fully excavated. Post pit [065] was an irregular semi-circular post pit cut out of plated bedrock, 1.30m by 1.22m by 0.30m deep onto a flat bedrock base and containing the remains of a 0.32m wide post pipe within the centre of the pit. Originally half sectioned, then quarter sectioned and then totally excavated to reveal a small circular rock cut 0.38111 wide and O.14m deep [078], cut into the base of the post pit. Within (066), the upper
Universiy of Munchester Archaeological Unit December 2005.

Reporr on rhe 2004 and 2005 seasons ofarchaeological excwafionr a1 MeNor.

fill of [065] a 74 mm long iron socketed arrowhead was recovered. A charcoal sample taken from the surrounding fill produced a date of 1000 to 1250 cal AD (Beta-209508, 2 sigmas). Post pit [079] was a sub-circular post pit cut out of the plated bedrock, 1.21m by 1.22m by 0.47m deep with steep regular sloping sides onto a flat bedrock base, and containing the remains of a 0.32m wide post pipe within the centre of the pit. The post pipe consisted of greyish brown silty sand and was surrounded by a single fill consisting of very frequent angular and sub angular 0-0.20m sandstone plates within a compact greyish yellow silt. Within the post pit three fragments of pottery were uncovered, the first was a fine gritty ware sherd dated to the late 1 l h to early 1 3 ' ~ centuries, the second was an unidentified fragment of fired clay. Within the post pipe itself a base fragment of an oxidised sandy ware piece of pottery in a hollow ware form and dated to the later medieval period was recovered.

8.3.2 Post holes
Post hole [035], 0.30m in diameter and 0.13m deep was circular in nature and cut into the natural plated bedrock, similar to the surrounding features, with sloping sides and a flat circular base. Filled with a loose dark brown silty loam the post hole possessed no archaeological artefact or dateable evidence.

8.3.3

Gullies

Feature [057] ran in an NE-SW direction, between 0.60m and 0.90m in width, 0.27m deep and excavated to a length of 2.75m. The full extent of the feature is unknown as the southern limit runs into the section, beneath a large tree within the garden. Originally cut through the plated bedrock to a flat base, 221.03m AOD, in a similar fashion to the surrounding features, [057] is defined by the presence of three postholes located at 0.30m intervals, c.0.60m in diameter. The lower fill contained two objects, one a small hoped metal clasp 41mm long and another flat unidentifiable object 56mm long with a small 5mm hole in one end.

Discussion
Taking all post pits identified within Trench 35 into account, one of the most intriguing aspects is the difference in absolute levels of the bases of the post pits. Three are very similar in level, however one [033] is substantially shallower by 0.250.30m and [035] is deeper by a similar amount. This raises the possibility that these post pits did not have to be constructed at the same levels. Leaving the possibility that the requirement for the pits was that they were adequately formed in order to provide a small depth and a flat, solid base. When looking at the overall form of the post pits they appear to be slightly bowed, a comparable medieval building has been identified at Tatton Park. However when the position of the post pipes within the post pits is taken into account then they form an absolute straight line, suggesting that the size of the pits in relation to the post themselves is not the The dateable evidence recovered from this trench and that in the surrounding trenches points towards an occupational structure being present on the site from the medieval period and possibly standing until the later medieval period. Taking the later dated

a

Universiry of Manchesrer Archaeological Unit December 2005.

Report on the 2004 and 2005 seasons ofarchaeological excavations a1 Mellor.

pottery that came from the post pipe itself it is possible to state that the post had either rotted or been removed at this point, therefore the structure could possibly be no longer in use by the late medieval period. The remaining fragment of burnt clay could be a multitude of items, ranging from pottery, to a possible clay walling of the structure, or a fragment associated with a hearth. As to the type and use of the structure it is impossible to accurately conduct a full assessment without knowing the potential layout of the whole structure. However early indications would point towards Within Trench 35 was an early test pit, and also a long section of 0.20m wide fracture within the bedrock, which although looking like a beam slot is only natural. The beam slot within Trench 35 is puzzling as the finds are probably contemporary with the post pit structure, although the angle is offset with the post pits and it does not protrude into Trench 33. Further excavation around the remaining unexcavated areas within the Old Vicarage garden would assist in providing an adequate ground plan of the structure and therefore greatly assist in the interpretation of the buildings function and the size and type of the structure.

Universiry of Manchester Archaeological Unit December 2005.

Ilnhrrrirj,d- d -u Lkcenbrrm.

Unit

Figure 40: Trench 35. Posthole [035], cut into bedrock, looking north.

Fitpre 41: Trench 35. Post pit [079] looking ovemead t west o
Unhursity aManchutsr Amhdogical Unit lkranbsr 2005.

FIOmc 42: Tnnch 35. hatholegully [WI, lodug south west.

Report on the 2001 and 2005 seasons ofarchaeological excmarions or A4eIlor.

5.4

Trench 36

Figures: 45,46,47,48,49, 50,51 and 52.
Trench 36 continues the process of large, open area excavation within Area C, and forms, together with Trench 26, the largest area excavated upon the hilltop. The trench was designed to answer four principle questions; firstly, to identify the remaining extent of the multiple roundhouse drip gullies identified within Trench 26. Secondly, to establish if the separate zoning of the gullies and postholes continues towards the opposite side of the roundhouses. Thirdly, by exposing the proposed south-east portion of the gullies, produce the expected and standardised entranceway into the roundhouses. Finally and most significantly, to investigate the immediate area surrounding the discovery of the Bronze Age dagger in 2004. The 2004 excavation around this area was halted upon discovery of the object in order to allow for an appropriately sized area and strategy to be incorporated into the 2005 season. Trench 36 was unusual in plan, sub-L-shaped, and designed to maximise the use of potential areas available for archaeological investigation. Positioned directly abutting the eastern edge of the previous excavation, Trench 26, and providing a large area surrounding the copse of trees within Area C. The main body of the trench was subtriangular in shape and measured 15.41-16.91m east-west and 4.58-1 3.50m north south, with its shortest axis at the western end. The extension ran towards the driveway and measured 7.01-8.66111 north-south and 4.33-6.60111 east-west with a 3.00 by 1.5m wide trench extension. The turf and friable, dark brownish grey humic silt clay topsoil and mid-brown silty clay subsoil were carefully removed by a six tonne machine with a ditching bucket, under the constant supervision of a professional archaeologist. During excavation both layers produced numerous items of 18" -20" century material, including pottery, clay pipe stems, glass and ferrous objects. As with Trench 26, immediately beneath this the subsoil was a layer of wet mid-greyish brown clay silt, identical to (781) in Trench 26, in that it contained a high percentage of small-medium sandstone inclusions. This layer is the interface between the natural clay geology below and the humic soils above. No finds were recovered from this layer. The natural geology is identical to that identified elsewhere within Area C; a compact reddish brown boulder clay. Trench 36 revealed a complex and very densely packed arrangement of inter-cutting features. The majority of archaeological features were typically denoted by their dark upper fills and frequent presence of small - medium sandstone inclusions. Due to the complexity of the multi inter-cutting features, identification of individual features and potential extents relied upon frequent and regular re-cleaning of the targeted areas. Due to the density of the archaeology within Trench 26, this section is broken up into feature types, i.e. gullies, postholes, pits and stake holes, where possible relationships exists between features of a different nature then these are commented upon.

Universiry ofMonchesler Archaeological Unit December 2005.

Reporr on the 2004 and 2005 seasons o/archaeological excavations at Mellor.

As the level of exposure within Area C is much greater than that of 2004, results and interpretations have subsequently developed, been refined and some have changed considerably from the discussions relating to the Trench 26.
5.4.1

Gullies

A total of 9 gullies were discovered during the excavation of Trench 36. All of which are located to the western end of the trench and its shorter arm. Due to the inter cutting nature of these features it initially proved difficult to establish individual gullies, therefore many were assigned more than one context during the excavation. Wherever possible, gullies that appear to be continuations of those located in Trench 26 are identified as such. The terminal end of gully [251] was located towards the northern extent of the trench and in close proximity to the inter cutting pits. The extent of the gully was identified throughout the various parts of the trench extensions and is a continuation of [638] located in trench 26. Heavily packed with large stone inclusions, the light grey silty clay contained occasional fragments of charcoal and burnt bone fragments. In profile the gully was relatively shallow 0.23111 and averaged 0.50 - 0.55m wide. The gully was excavated at three points across its extent and found to cut gully [364]. Within the southern trench extension, the gully was cut by two unrelated postholes [322] and [341]. Overall the feature measwedl2.6m in diameter, with its north-eastem extent running out of the excavated area. Feature [203] was a small sub circular gully, originating close to the inter-cutting pits in the northern portion of the trench. This gully was found to be a continuation of gully OVM04 [656] and formed a semi-circular enclosure with a diameter of 8.80m. The gully is not regular in shape and shows signs of distortion towards its south western extent. The eastern terminal end is much thinner than that identified to the east, 0.42m and 0.79m respectively. In profile the gully contained undulating sloping sides and a rounded base, and was 0.25m deep. The primary fill was a pale grey silty clay containing frequent inclusions of angular and sub-angular 0-15cm sandstones and the secondary a mid-brown silty clay with similar inclusions and occasional charcoal flecks were identified within both fills; however, no dateable artifactual material was uncovered from this gully. A 2.00m long section was excavated over the eastern terminal end and a further 1.00m wide section further to the south. The gully cuts a number of other gullies, [361] and [363], and is cut by [552] and [743]. Identification of any associated postholes was hindered by the extensive inter-cutting nature of the gullies and homogenous upper layer of the repetitively inter-cutting pits. Within the southern extension of Trench 36 a gully terminus was identified [231], cut by gully [320]. [231] is only identifiable through the section placed over the terminus. This feature appears to be a continuation of OVM04 [565], and would therefore be another semi-circular gully, similar in nature to [203] and [251] and consists of an identical fill. The gully was approximately 6.3m in diameter, 0.55m wide and 0.40m deep. No archaeological material was recovered from this feature. The partial remains of another similar semi-circular gully were identified, running into the eastern trench edge and found to be a continuation of OVM04 [552]. This feature was not excavated during the 2005 season. However in total the diameter
Universiry o/Manchesrer Archaeological Unil December 2005.

Report on the 2004 and 2005 seasons oforchoeological excavarions ar A4cNor.

measured 8.4m and appeared to show two definite terminal ends. It cut gullies [320] [23 11 and appears to be the latest of this particular type of feature. Within the southern extension of Trench 36, gully OVM04 [743] was identified and continued for a distance of 0.40m from the trench edge in a south-easterly direction before stopping. As the majority of this feature was excavated during the 2004 season, no excavation of this gully continued during the 2005 season. Stratigraphically the feature cuts gullies 15521 and [203]. During the excavation of the slot across the area located against the eastern trench section a total of seven inter cutting gullies were identified, of which two have already been discussed [251] and [23 11. The remaining gullies [354] [355] [356] [357] [358] were identified within this small trench. Further gullies were present at the edge between Trench 26 and Trench 36, however it was not possible to follow them as they were cut by the area of inter cutting pits located to the north. Covering the entire area of the southern arm of the trench is a deposit of light-mid grey silty clay, 0.10m deep, containing frequent deposits of irregular sized sandstone deposits (447). This layer is a continuation of the same located within Trench 26. Only after careful removal of this layer are the underlying features visible.

Discussion
The presence of layer (447) makes identification of the underlying features difficult and subsequent removal of this layer reveals the underlying archaeological features. The complexity of the surviving archaeology within the area is staggering in that a large portion survives but is complicated by the subsequent inter-cutting nature of the gullies. The form in plan and section of these gullies shows multiple features of differing forms. Taking the 2004 results into account there are four different categories into which these can be positioned. The first is a large circular gully, [251], can be classed as a drip gully of a round house. If subsequent excavation proves this feature to be hlly circular then the identification of a terminal end within Trench 36 would indicate a north-east facing entranceway into the structure. If this is a form of round house, then the presence of an entrance way to the north-east is unusual compared to round houses from alternative sites, whose entrances tend to face in a south-easterly direction. It is expected that this is not the case for [251] due to the typological nature of the immediate area. A south east facing section would place this in direct line of the usual prevailing wind direction but by moving to the north-east this problem is avoided. However, there is the possibility that this represents the remains, not of a round house, but, due to the high fractured stone content, the kerb of a burial mound. This interpretation would tie in with the discovery of the Bronze Age flint dagger, placed on the natural ground surface encompassed within the feature possibly suggesting the ritual deposition of the dagger. If classified as a burial mound, then a burial would be expected within the centre of the structure, the location of which would be within the area of the inter-cutting pits, and therefore, at present, excavation has not been conducted over this particular area.
Universiry o Manchesler Archaeological Unit f December 2005.

Report on the 2004 and 2005 seasons of archaeological excavations nt Mellor.

The second type of gully arrangement is characteristically semi-circular in plan. A total of three have been identified, [231], [203] and [364], and there is the potential for more of the gullies to be classed within the same grouping but which cannot be immediately placed within these as their complete extent and form is as yet unknown. All three are semi-circular in form and are arranged in the same orientation, the gully to the south and the opened side to the north. Provisional interpretation is that these features are not roundhouse gullies and could be earlier settlement structures, but more probably animal enclosures. All three stratigraphically predate the largest, complete gully [251], and cannot be present at the same time as they cut each other. The remainder of the gullies cannot be classed as either at this point, as the excavation does not identify their true form. The area exposed within Trench 26 and 36 does show that the notion of zones of occupation can still be upheld, and that this immediate area contains multiple examples of occupational usage over a prolonged period of time, indicating that the area itself was particularly important in relevance to its particular usage. Areas without occupational structural features, such as those to the west would have been separated by use and remain as important to the archaeological records as those with continuous re-use.

5.4.2

Pits

Within Trench 36 a large collection of inter-cutting pits were uncovered. These are a continuation of those identified within the north eastern extent of Trench 26. In plan the pits appeared to be one homogenous irregular layer, consisting of a mid-dark grey silty clay containing frequent rounded and sub-rounded inclusions. The extent of the swathe of pits continues in a north-east to south-west orientation and measures c. 16.5m long and c.4.10m wide. To the south the pits cease c. 2.56m from the Trench 26 southern edge, and continue into the Trench 36 northern edge. During the 2005 excavations four slots were placed over the single mass. The first, alongside the northern trench edge and at the pits' eastern most-extent, was 1.20m long by 1.00m wide and excavated to a depth of c.0.20m. Identified within were the remains of four irregular inter-cutting pits, ranging in size from 0.60m wide to 1.30m wide and all containing frequent amounts of rounded and sub rounded sandstone inclusions. No dateable material was recovered from this area. The second slot, located alongside the northern trench edge towards the centre, produced evidence of three inter-cutting pits, [209], [210] and [212] measuring between 1.00m and 1.8m in width and containing frequent deposits of rounded and sub-rounded sandstone inclusions. The third, located towards the centre of the trench, was 1.50m by 0.90m and designed to establish if gully [251] continued beneath the pits. A further three pits 0.90111 in width were identified containing frequent deposits of rounded and sub rounded sandstone inclusions. The pits did not cut[25 11. The forth, was a re-excavation and realignment of the section over the pits within the trench 26 extension and revealed nine inter-cutting pits of assorted sizes.

Universip of Monchesler Archaeological Unit December 2005.

Report on rhe 2004 and 2005 seasom o archaeological excwations a1 A4ellor. f

Compared to other pits excavated upon the site, the universally high frequency of fire cracked pebbles contained within the fills is highly unusual. A strategy of sampling was introduced in order to evaluate the quantity and type of material that was contained in a proportionate sample of the pits excavated. As the identification of the pits was difficult during excavation, the stone inclusions were saved from the respective sections. [219], [250], [249] contained a total of 504 individual stones, which produced a weight of 41.00kg, of which 605 were sub-rounded and 405 subangular. A total of 70 % of these were fire cracked in form, the smallest measuring 0.05m x 0.03m x 0.05m Pits [331], [332], [333] and [334] contained a total of 216 individual stones, producing a total weight of 47.00kg, of which 60% were sub-angular and 40% subangular. A total of 20% were in fire cracked form, the smallest measuring 0.01m x 0.01m x 0.02m and the largest measuring 0.15m x 0.12m x 0.05m. Pits [361], [362], [364] and [365] contained a total of 194 individual stones, producing a total weight of 74.00kg, of which 70% were sub-angular and 30% subangular. A total of 85% were in fire cracked form, the smallest measuring 0.01m x 0.02m x 0.03m and the largest measuring 0.03m x O.l8m x 0.23m.

Discussion
The large series of inter cutting pits revealed within Trench 26 and 36 is a type of use not discovered in other trenches so far excavated on the Mellor site, and appears to be localised into this immediate area. If future excavation across the site establishes this to be the case, then it would suggest that the area held some particular relevance to their functions. It is conceivable that the close spatial relationship between the pits and the roundhouse gullies implies that they were associated and belong to a similar phase. The repeated re cutting of the gullies denotes that the area was occupied and re-occupied over a substantial period of time, which would therefore account for the large number of pits and their repeated inter cutting nature. This could be due to the pits having a function linked with some other activity or feature in the immediate area, or that the area itself was ideally suited to their particular purpose. It is important to note the possibility that there is little evidence, other than a common immediate locale, to suggest that these pits were part of a series or group. Indeed it is highly probable that individual pits of various dates and functions lie within a larger group of possibly associated pits. This probability appears to be supported by the few artefacts recovered from the fills of the excavated pits. These include flint flakes along with shreds of Romano-British pottery along with the late Iron Age date recovered from the pitlposthole [283]. The possibility of re-deposition of artefacts and sample cannot be discounted, particularly due to the frequent inter cutting nature of these features. It seems likely that they indicate a repeated and prolonged use of this area over various periods of time for the cutting of small pits. The questions posed by this are related to the specific purpose of the pits and the significance of this particular area. The two questions may not be mutually exclusive. It is possible that the pits were placed here to exploit the natural resource of the geological boulder clay of the area, which unlike the majority of the site (which possesses an underlying stone bedrock)
Universiry o Manchesrer Archaeological Unit f December 2005.

Report on the 2004 and 2005 seasons of archaeological excavolions a1 Mellor.

forms a band of natural boulder clay that runs approximately north-east to south-west through the central portion of the hilltop. The clay could have been utilised for a number of purposes, including a providing part of the structural integrity to the roundhouse walls, and would therefore have proved to have been a highly desired local resource. No re-deposition of the boulder clay was identified within the fills of the pits. It appears therefore that the boulder clay was extracted from the pits and utilised or placed elsewhere. Whilst this may have been part of the pits' function, the nature of their infill suggests another primary purpose. Each of the pits contained a very high percentage of small sub-angular and more commonly, sub-rounded stone within their fills. Many of which were fire reddened and cracked. Stones such as these are classically known as 'pot boilers' and are associated with the heating and boiling of water after initially being heated themselves. The impermeable nature of the boulder clay would have served as an ideal receptacle in which to contain water, into which stones were immersed after being initially heated. The pits therefore, would appear to have served a primary purpose connected with the boiling water. Whether this served a domestic, industrial or ritual purpose is unclear and it is possible that all three could have been fulfilled at one or more stage. The nature of the stone inclusions assessed by the sampling produced a remarkable difference in weight. The pits with the highest portion of fire cracked stone were also the heaviest by a significant proportion. The recovery of the Bronze Age flint dagger from Trench 26 may suggest that ritual activity took place within the area due to this objects funerary associations. Where pits have been found closely in association with Bronze Age ritual 1 funerary activity on other sites. The excavations have suggested that they may have been utilised as saunas connected to a purifying ritual. In a domestic context, the pits may have been utilised either in the preparation of food, or to heat water for bathing. Occasional small burnt fragments of bone were recovered from the fills of the pits and therefore may suggest a connection with the cooking of food. Similarly, many of the stones from within the pits were volcanic in origin and ideally suited to a slow release of stored heat which would have proved beneficial in the cooking process. An industrial purpose is also a potential explanation for some of the pits. An unusually high concentration of red ochre was found both in and around the pits location, which may have been used in the production process of a textile as a dye. It seems likely therefore based on the above possibilities, that both the pits and the specific area of Trench 26 and 36, served numerous purposes over a period spanning some 3000 years. Continuity such as this is a recurrent theme upon the Mellor site and signifies its importance within the landscape over the ages.
5.43

Postholes

Within trench 36 a number of postholes have been identified, numbering twelve within the trench. Identification of the postholes is extremely difficult due to the complexity of the inter-cutting gullies and pits. Many of the postholes were stone packed and contained flat small to medium sandstone inclusions, which appear to
University of Munchester Archaeological Unil December 2005.

Report on the 2004 and 2005 seasons o archaeological excavalions a1 Mellor. f

have been sat on-edge around the sides of the posthole and it is likely that these would have acted as packing stones holding up a central post. Post holes of this type were sub-circular or sub-oval in plan, measuring 0.45 x 0.40m with near vertical sides and a flat, possibly u-shaped base and an average depth of between 0.20 and 0.30m. It is particularly difficult to separate the identified post holes into any different uses and structures as the complexity and multi-period usage of the surrounding features does not allow for any individual phaseological relationship to be established. One post hole of particular note was located in the centre of the southern arm,located inside an arrangement of gullies, [283]. This was sub-circular 0.37m diameter, 0.33m deep post hole, containing very steep sides and a gently rounded base. The upper fill (284) consisted of a friable mid-grey clay silt containing occasional small flecks of charcoal and occasional small sub-angular stone inclusions. The secondary fill consisted of a friable very dark grey clay silt containing multiple fragments of charcoal and flecks of burnt bone. Fragments of what appeared to be pottery were recovered from this fill, later analysis by Dr Chris Cumberpatch pointed towards the fragments being the waste from an industrial process, possibly metalworking and maybe parts of a crucible. A charcoal sample was recovered from this fill and radiocarbon dating produced a calibrated date of BC 190 to 40. (Beta-209510) The primary fill was a mid-greyish brown firm clay containing occasional flecks of charcoal; the deposition of this fill was natural, occurring from silting of natural sides of the posthole.

Discussion
There is a strong possibility that [283] was not a posthole, the quantity of deposited industrial waste material would suggest that if a post were present this would not have been possible, therefore the feature could be a small pit, containing the waste material as a deliberate phase of backfilling or depositioning, providing the possibility of industrial activity upon the site during the late Iron Age. No other pits of this form and fill have been identified within the excavated areas of Area C. The post holes occur across both trenches, covering the entirety. Post holes within the immediate area of the gullies can possibly be interpreted as being the foundations of large posts which would have supported the roof of a structure. However, it is almost impossible to identify particular layouts of these posts, due to the complex inter cutting nature of the gullies and pits, so that any associated layout is masked. Along with the possibility that the majority of the postholes are not going to be contemporary, in that they would relate to each phase of round house and be very similar in location. Identification of the post holes in 2004 led to the interpretation of two distinct zones, one of gullies and one of numerous postholes in between the zones of occupation. Therefore it was expected that there would have been a significant number of postholes identified to the east of the gullies. This was found not to be the case, there were occasional postholes, but not found in the concentrations located to the west of the gullies. Indicating a possible change of use within the different areas. Is this the furthest extent of the round houses and settlement, within the confines of the outer

Universiry o Manchesrer Archaeological Unit f December 2005.

Report on the 2004 ond 2005 seosons of orchoeologicol excwatiom at A4ellor.

enclosure ditch? Or is this an area set aside for a different use, possibly that of cooking pits?

5.4.4

Stakeholes

A total of 51 stake holes were discovered within Trench 36. These can be distinguished into three separate groups. First and second are two parallel lines of stakeholes approximately 1.50m apart, orientated in a north-west, south-east alignment, The northern most line contains fourteen, whereas the southern line contains eleven. Typically the distance between each stake hole is 0.40-0.45m apart and comprises of small sub circular features, averaging 0.10-0.20m in width and between 0.10m and 0.32m deep and are cut into the natural geology and filled with a friable light-mid greyish brown silty clay containing no inclusions. The average profile of these appears to be steep almost vertical sides sloping inwards to form a point at its base. The remainder of the postholes remain classed as individual features as no apparent grouping can be assigned to them and therefore they must be classed as separate. It is expected that these would have formed parts of other groupings but it remains impossible do differentiate between them. Discussion Within the two parallel lines, the stake holes appear to form the original space into which the steaks had been placed. Due to the nature of the stake holes. It is possible to state that they would have consisted of roughly circular wooden post with a shaped, sharpened end. This end is the placed into the ground, by force, not by digging a hole first. These posts would then become the uprights and more flexible branches would then be weaved in between these in order to construct a temporary solid wicker fence. The two parallel lines of stake holes represent two parallel fences that can be related to each other. In terms of function and use, there are multiple options; firstly these could be enclosures for containing animals, abutting to the roundhouses. Secondly they could stand alone, as no stratigraphic relationship can be identified between the steak holes and the gullies. The stake holes definitely denote a temporary wall, demarcating a particular area for usage, whether this be for animals, industrial processes or simply and entranceway into a particular area. The natural clay surface provides a suitable background against which to identify these, suggesting that these may be present in other areas which have been excavated, but are not recognised due to the complexity of the background natural and the surrounding features.

University of Munchester Archoeological Unit December 2005.

Figure 45: Community volunteers excavating the southern a m of Trench 36. r

Figure 4 : Nonh east facing section of multiple inter cutting pits located to the north 7 of Trench 36.

Figure 4 :Multiple inter-cutting pits looking north in Trench 36 8

Repon on h e 200l and 2005 s c ~ o n s o archomIogicnl ucnvolionr ar Mrllor f

Figure 49: Overhead views of fully excavated pit, excavabon produced possible Iron Age metalworking waste material and an Iron Age carbon date.

Repon on rhe 2004 and 2005 seasons o ahomlogical ucovarions ar Mellor. f

OVM M
n@sh 33
p. (.

- mO

lm

N+

F i r e 50: Plan of Trench 36.

University o MancksferAmhncoiqlcP[ f Ikcember 2aOS.

Unil

Rrpon on Ihe 2004 and ZOO5 sepronr of ~ h o m l o g i c aexcavations at Mellor. l

Figure 51: Combined plan of the 2004 Trench 26 and the 2005 Trench 36.

Universiiy o Mancluster Amhaeologkal Unit f December ZWS.

Figure 52: Trench 36 section

Reporr on the 2004 and 2005 seasons of archaeological excavarions ar Mellor.

5.5

Trench 37

In March 2005 the Mellor Archaeological Trust excavated a trial trench against the boundary wall between Area A and the churchyard of St Thomas's at the east end of the Old Vicarage garden. The purpose of the trench was to confirm the line of the large inner ditch south from trenches 18 and 32 and identify the point at which it left Area A. The trial trench revealed the distinctive fill of the ditch with the fractured bedrock on either side. It also showed what appeared to be the fill of a smaller linear feature cutting through the ditch fill. In order to fully examine this relationship and also to see if the palisade slot could also be detected the trench was extended and reexamined as Trench 37 during the summer excavation.
Trench 37 ran north to south and measured 7.50m long by 1.50m wide. On excavation the later feature was seen to be the fill and cut of the water pipe supplying the church. The extension of the trench to the south revealed the 'tell tale' line of on edge stones that in Trenches 18 and 31 and 32 had marked the line of the palisade slot. These again appeared to run parallel to the ditch. The extension also revealed what appears to be a posthole cut into the bedrock between the palisade slot and the ditch. A lozenge shaped patch of dark grey soil was visible in the south half of the trench. This was within the fill of the ditch. It is possible that it simply represents a discrete area of infilling of the ditch; however, the possibility also exists that it is the fill of a later feature cutting into the ditch fills. As it is planned to make an extension of Trench 37 the focus of one of the forthcoming summer seasons as these features were not excavated.

Discussion
Trench 37 successfully proved the continuation of the inner enclosure ditch from Trench 18 in a south-westerly direction towards the churchyard perimeter. The identification of the ditch and its associated palisade slot, still parallel to each other, further enhanced understanding regarding direction and extent of the ditch. It also shows that at present, that between Trench 37 and 2, there are no further areas within the vicarage garden available to track the inner enclosure ditch and that further identification of its exact alignment would require excavation within the graveyard perimeter. Unlikely to happen, due to the consecrated ground.

Universiry of Manchesrer Archneologicol Unit December 2005.

Report on the 2001 and 2005 seasons o archaeological excovalions at A4ellor. f

5.5

Trench 38

Figures: 53,55 and 56.
Trench 38 was placed in order to assess the potential for the continuation of the outer enclosure ditch from that identified within trench 30 in a westerly direction towards the church. Recent geophysical survey of the surrounding area using a magnetometer to follow the possible location of the ditch allowed for the accurate positioning of trench 38, as close to the car park wall as feasibly possible, c.3m. The trench was rectangular in shape, measuring 4.80m by 2.00m and orientated north south at right angles over the projected geophysical anomaly. The geological make up of the trench consisted of a fractured bedrock overlying plated bedrock. The topsoil and subsoil extended to a depth of 0.5 1m. Located just off centre to the north, a 1.90m wide rock cut ditch was identified [409]. The depth of the excavated feature was 0.95m from the top of the archaeological layer. The profile of the feature consists of steep sloping sides cut from the bedrock onto a flat solid base. The layer morphology of the ditch segment consists of a primary fill of light brown silty sand, containing frequent inclusions of large angular fragments of re-deposited natural (410). Towards the northern slope a secondary deposit of light brown silty sand containing frequent inclusions of 0-0.15m angular plated fragments was identified (411). Overlain by the main ditch fill (412) consisted of light-mid brown silty sand with occasional smaller fragments of sub-rounded material. No finds were recovered from any of the fills within this ditch.

Discussion
Excavation concluded that the outer enclosure ditch does continue in a westerly direction, in similar form to Trench 30 and can be positively identified up to the car park. This proves that the geophysical results work particularly well within the outer field on this side of the hilltop, specifically upon areas of a bedrock background compared to a clay background as seen in Area D.

5.6

Trench 39

Figures: 54,55 and 56.
Trench 39 was located to the east of trench 30, towards the eastern most wall within this field, and on top of and at right angles to the furthest extent of the geophysical anomaly. The trench, rectangular in shape, measuring 8.00m by 2.00m and in a north south orientation contained the remains of the continuation of the outer enclosure ditch. The ditch measured 1.34m wide and 0.65m deep cut into natural bedrock and located off centre to the north of the trench. Possessing undulating steep sloping sides and a flat base, the ditch possessed a primary fill of a mixed silty sand containing occasional angular sandstone £ragments. No dateable material was recovered from this feature.
Universiry o Manchester Archaeological Unit f December 2005.

Rrpon on h e 2 W 4 and 2W5 seasons o amhamlogical ruavafionr af Mellor. f

Figure 53: East facing section of the outer enclosure ditch located in trench 38.

I

Figure 54: East facing section of the outer enclosure ditch located within trench 39.

Repon on the 2004 and 2005 seasons o/nrchaeologicol acavntionr or Mellor.

Figure 55: Plans of Trench 38 above and Trench 39 below.

Universiry o Manchester Archeological Unir f December 2005.

Repon on rhe 2004 and 2 W 5 semons o archaeological ercovdionr ar Mellor. f

Figure 56: West facing sections of Trenches 38 and 39, Q and R respectively.
5.8

Trench 40

Figures: 57,58,59,60 and 61
Trench 40 was excavated in November 2005 and located directly over trial trcnches exposed in 2004 in order to excavate the terminal ends of the possible entranceway of the outer enclosure ditch. An rectangular area of 3.70111 by 8.60m was de-turfed and hand excavated down through the previous backfill to the known archaeological .0n horizon, 1 1 r deep. A 1.50 m wide section was then placed over the eastern terminal end, the western terminal end was revealed but not excavated. The western terminal end consisted of a primary silting layer, followed by a large purple-brown wet clay containing frequent quantities of angular and sub angular material (506), silting of silty sands then fills the remainder of the ditch. No archaeological dateable material was recovered from this trench.

Discussion
The identification of the two sections of ditches, are significant in that they are the same ditch, the outer enclosure ditch, and that it does continue acmss Area E, running approximately along the contour lines towards Mellor Old Hall. indicating that although the geophysics conducted upon this area did suggest a few possibilities of features, however it was unable to distinguish the ditch fmm the clay and bedrock background. It is suspected that this is due to the similarities between the infill and the natural geology, along with the waterlogged conditions present wherever clay is
Universiry of Momhesrer Archocologic~lUnir December 2W5.

Repon on the 2 W and 2M)5 s~asonr ofa~~luvoIogical rrcuvorionr er Hellor

present on the site. Showing that if an area is identified as void of features using the geophysical results, this does not necessarily conclude that them are no features present. Therefore excavation must be conducted throughout such areas to positively conclude any results. The discovety of two terminal ends to the ditch is extremely significant as this is the fmt entrance way identified within the excavations at Mellor. Although small, the entrance for the outer enclosure ditch need not have needed to be any largw, wide enough for people and cattle move through into different areas. It would be safe to presume that this would not be the only entranceway within the outer enclosure ditch, and that there are likely to be a few more over the large area covered by the enclosure ditch. Any further entranceways have not been identified as yet. The lack of any dateable material found within ihe excavation of these two terminals, may indicate a specific date for the feature. It is common place for the deliberate deposition of object into ditch terminals in the late Neolithic and Early Bronze Age periods, where as this pmctice becomes less common into the Late B r o w Age and Iron Age.

Figare 57: West facing section of the eastern terminal end of the outer enclosure ditch located within Trench 40.

Repn a the 2 W and 2005 seasmu ofe~~hadogiral ucavMVMom 111MdJm.

Figure 58: Unexcavated western terminal end of the outer enclosw ditch located within Trench 40.

Fignre 59: T e d a d end of outer enclosure ditch as seen in Trial Trenching in 2004.

Report on the 2 W and ZOO5 reasons o o~ehomlogid f -avorim or Mdlor.

Figure 60: Plan of Trench 40

Universiry o M ~ c h r s f c Archocologlcnl Unil f r Drcrmbrr ZOOS.

Repon on lhc 205i and 2005 scarom o ~hncolo8icnl f ucavutionr or Mdlor.

figure 61: West facing section of the outer enclosure ditch identified in Trench 40.

figures: 62.63 and 64.
Another trench excavated in November 2005, trench 41, was positioned to positively identify the possible ditch located within the trial trench exposed in 2004. The trial trench had revealed the feat* down to the correct archaeological horiwn but due to time and resoexcavation was unable to proceed Trench 41 located the same archaeological feature and excavation revealed a 1.23111 wide and 0.95111 deep, steep sided ditch with a flat base, cut into the natural surmunding bedrock. Consisting of primary silting layers (523), (524). (523, (526) and (527). The main body of the ditch is then fdled with one universal fill, a reddish brown silty clay containing a high frequency of angular and sub angular fragments. No dateable material w a s recovered from this feature.

The positive identification of the ditch proves that the outer enclosure ditch identifed within Trench 41 is similar in nature to that excavated in Trenches 22 and 24, and therefore the conclusion that these are part of the same feature can be made. Expanding the known extent of the outer enclosure ditch within are% A and indicating that the ditch is likely to be heading off in a north easterly direction towards that of Mellor Old Hall. If this is the case then the m a enclosed within the outer enclosure ditch is massive.

mpre 62: West facing section of the outer enclosure ditch identified in Trench 41.

- -

--

-- -- - - 1

OVM 05

Tlcnch 4 1 Plro

N

C

t

Figure 63: Plan of Trench 41.

Uniwniry of M~chcs~er Archorological Unit December ZOOS.

21570m N
AOD

x

S

0.5m

Figure & : West facing section of outer enclosure ditch identified within Trench 41. 1

Univrrsiry o ManchesterAmhaeo1~LulUnit f Dccembrr 2005.

Report on the 2004 and2005 seasons of archaeological excavationr at MeNor.

9.

2005 Conclusions

Each trench excavated during the 2005 season of excavation succeeded in its designed objectives and contributed significantly to our understanding of the nature and extent of the settlement on the hilltop at Mellor. Furthermore the excavations at Mellor are now reaping the rewards from the implementation of coherent strategy of evaluation begun by the Mellor Archaeological Trust and UMAU. This strategy means the results fiom any individual trench excavated in a single year no longer stand alone. They will form part of a group of trenches excavated over several years designed to gradually provide an insight on specific questions about the history of settlement on the hilltop around the Old Vicarage. This long term strategy of evaluation now means that the information gathered from the various groups of trenches is beginning to overlap to provide new interpretations of the site. Central to these developing interpretations is the systematic radio-carbon dating of features from the site. To date 8 radio-carbon samples have been recovered from the excavations, allowing Mellor to become a regionally and possibly nationally significant site. A table of the radio-carbon dates can be seen below in chronological order.

I

Figure 65: Radio-carbon dating from the Mellor excavations 1998-2005.
Excavation of Trenches 26 and 36 have provided a valuable insight into the character of settlement present upon the hilltop. Round houses appear to be present, along with four semi-circular features which are provisionally interpreted as animal enclosures or structures from the Bronze Age: BC 2920 to BC 2560 (Beta - 209509,2 sigmas). The presence of the largest stone packed curving gully may show the potential use of this area for other structures that are not round houses, such as the kerb of a possible burial cairn. The identification of the multiple inter-cutting pits shows that the area was heavily re-used over a substantial period of time, for multiple purposes. The presence
Universirv o f Manchester Archaeolonical Unit

I I

98

Report on the 2004 ond2OO5 seasons of archaeological u c w o l i o m o f Mellor.

of industrial waste within a small pit, together with crucible fragments recovered from Trench 1 reaffirms the existence of industrial activity at Mellor during the Iron Age and confirms the area of the pits was used for multiple occupational tasks. The combination of the trenches within the Old Vicarage garden, 1,2,33, 34 and 35, revealed a new period of occupation from the excavations at Mellor. The existence of an aligned row of post pits within Area A demonstrated the presence of a substantial timber structure. Pottery recovered from the packing of some of the post pits was dated to the 11" - 13" centuries and material recovered from the post pipes within the post pits has been dated to the 13" - 15' centuries. An iron war arrowhead from one of the pit fills is dated to the 13-14" centuries and radio-carbon dating from this fill pl'oduced a date of AD 1000 to AD1250 (Beta - 209508, 2 sigmas). This suggests that the structure was constructed during the 11" to 13" centuries and closed at some point during the 13" to 15" centuries. Due to the limited area of excavation it is not possible to determine the extent or type of building that would have been present, although there are examples of timber halls dating to the medieval period with similar configurations of post pits. Further excavations of areas within Area A should assist in identifying the true nature of the structure. Regionally few examples of medieval timber structures have been excavated: one is the rectangular aisled buildings located during excavations of a medieval village at Tatton Park. Structures B and D closely resemble the structure located at Mellor, in that they were timber buildings of a medieval date, the first being 14.4m long by 4.6m wide with all four walls bowing out slightly, structure D, a rectangular structure measuring 9m long by 4.5m wide. The significant difference between these two examples and that identified at Mellor is the size of the postholes and their irregular form. The average size of those at Tatton was 0.70111 in diameter compared with 1.20m at Mellor, possibly due to the differing nature of the natural geology within each area (Higham 2000). The other is the aisled hall identified during the Baguley Hall excavations. In total seven large post pits were identified and the eighth postulated, forming a small rectangular timber building. The pits are almost identical to those found at Mellor, although slightly larger in plan and depth and would have contained similar sized posts within (0.40m). The date of this is considered to be prior to the fourteenth century hall and therefore similar to the one identified at Mellor.

Universiry of Munchester Archaeological Unit December 2005.

Reporr on r11e2001 and 2005 seasons o archaeologicol excoooriom or Mellor. f

Figure 66: Baguley Hall, aisled hall post pit phase (after Dixon, Hayfield and Startin 1989).

Universiry o Manchesrer Archaeological Unit f December 2005.

Report on rhe 2004 and 2005 seasons ofnrchneological excwofions ot Mellor

I I I I I

Figure 67: Tatton Village, excavated medieval features (after Higham 2000).
Nationally, the number of medieval timber halls which have been excavated remains small in comparison to structures relating to almost any other period. Caution needs to be applied when assessing the nature of the medieval structure at Mellor in comparison with other timber built halls as its complete layout has not been ascertained. Excavations at Cae Castell, Rumney castle, Cardiff, identified the remains of rectangular post-hole built medieval timber halls (Lightfoot 1981), structures B and E, the dimensions and intervals between the postholes closely resemble those identified at Mellor. The manor site of Goltho, Lincolnshire, initially occupied during the Roman period, was later utilised as a defensive site from the ninth century onwards and contained multiple domestic timber structures akin to that identified at Mellor, of which two successive timber single aisled halls were identified
Universify of Monchester Archneological Unit December ZOOS.

Report on the 2001 and 2005 seasons o/archaeological excavations at MeNor.

to the late 1l h century and a larger timber hall, aisled on four sides was constructed c. 1150 (Beresford 1987).

Figure 68: Medieval timber buildings identified at Cae Castell. (Gwent and Glamorgan Archaeological Trust 1981).
Internationally, two halls were excavated in the Pays de Caux region of Normandy, Notre-dame-de-gravenchon and Mirville. Excavation of Notre-dame-de-gravenchon revealed occupation phases dating from late Roman, early Frankish and medieval. The 1 2 ~ century phase included an elongated rectangular posthole building, surrounded at one end by a number of smaller postholes assumed to be associated with structures attached to the building. Noteworthy for its long narrow structure with no internal central supports (le Maho 1981). Whereas Mirville consisted of a smaller eleventh century rectangular timber hall, 17m long and 8m wide with a hipped roof and internal divisions (le Maho 1981). The excavation of Der Husterknupp in Germany identified a single aisled c. 12m long, 8m wide timber building, structure 3, with post pits initially in a similar layout to those at Mellor. Importantly due to environmental conditions, the posts remained and indicated a high level of working and decoration of the posts above ground (Herrnbrodt 1958), suggesting that previous reconstructions and images of medieval timber halls and buildings could have been over simplified and that they may be more elaborate than originally anticipated. It is suspected that the timber structure located at Mellor could be that of a medieval hall. Documentary evidence indicates that during the 13" century an estate granted to the 'foresters' who protected the game within the immediate area of the Kings forested lands within the region, particularly that of the Peak, was held by the 'de Mellur' family. Such a location within the landscape would possess the benefits of being isolated, but also providing a view over the prime access route into the forested areas. Previous suggestions place the site of their dwelling as the present Mellor Hall, or Bottoms Hall (Arrowsmith 1997, Hearle and Oldham 1985). Following the recent discovery of the suspected medieval hall in the Old Vicarage Garden, it may be necessary to re-assess the timber structure as being the site associated with the original dwelling of the 'de Mellurs'.

Universip of Manchester Archaea/ogical Unir December 2005.

Report on the 2004 and 2005 seasons ojorchaeological excavations at A4ellor.

GrundriS des H a w s 3 der Fla&sicdluog. MaBuab 1:bO.

Figure 69: Timber hall (House 3) identified at Husterknupp. (Hermbrodt 1958).
University of Manchester Archaeological Unit December 2005.

Report on the 2001 and 2005 seasons o/archaeological excavarions or Mellor.

Excavation of the inner enclosure ditch within Trench 33, confirms its existence between the previous known extents of Trenches 1 and 2 in Area A. As the ditch runs right through Trench 33, it is possible to say that there can be no entrance way into the area enclosed by the large ditch to the western side of the hilltop. Therefore it is postulated that the entranceway was located towards the eastern side, close to Trench 18, providing access from the flat area of settlement in between the outer and inner enclosure ditches. The discovery of the continuation of the palisade slot for a short distance confirms its association with the ditch, but lack of rampart. The absence of the palisade slot within the central segment of Trench 33, and it close proximity to a posthole alignment may indicate the presence of a small overlapping entranceway in order to gain access for maintenance of the ditch. Trench 37 confirmed the expected continuation of the ditch from Trench 18, to the furthest extent available within Area A, that is the perimeter of the church yard wall. Along with confirming the alignment of the palisade slot identified in earlier trenches, parallel to the internal side of the inner enclosure ditch, the trench also failed to find any trace of an inner bank. The targeted excavation of trenches within Area D over geophysical and evaluation trench anomalies has confirmed the presence of an entranceway for the outer enclosure ditch and the ditch's continuation to the middle of Area D, it is expected that its course will proceed across the entirety of Area D, in the direction of Mellor Old Hall. Continued evaluation following evaluation over the linear geophysical anomaly identified within Area E, has c o n f i i e d the presence of a linear enclosure ditch, similar in proportions to those identified within Area D. The extent of the feature can now followed from the car park within Area E, to the eastern most edge of this field. The possibility remains that the ditches located within these two areas are the remains of a single outer enclosure ditch which would have encompassed the entire flat portion of the Mellor hilltop, and more entranceways are likely to be present throughout the remaining segments of the enclosure ditch. This would create an enclosure ditch that encompasses a much greater area than most Iron Age hill top enclosures. It is not yet possible to determine the chronological relationship between the inner and outer enclosure ditches, presenting the possibility that the outer ditch could be Bronze Age in origin. Dating of this ditch is solely dependant upon the recovery of the suspected Iron Age 'Mellor' pot, which cannot be traced to a similar typological form.

Universiry o/Manchesrer Archaeological Unit December 2005.

I I

Appendix 1
2004 Radiocarbon Dating Results

I
I I

I I
I.

C
11

I
I C

I
C I
1 I

C I.

-

I
I

Dr. John Roberts
University of Manchester
Sample Data
Measured Radiocarbon Age
13C/12C

Report Date: 3/22/2005 Material Received: 2/23/2005 Conventional Radiocarbon Age(*)
2080 +/- 40 BP

Ratio

I. I -

Beta - 2023 15 2100 +I- 40 BP -26.2 0100 34 SAMPLE : O W 04 S-LEN0 ANALYSIS : AMS-Standard delivery MATERlAUPRETREATbEM : (charred material): acid/alkaliacid 2 SIGMA CALIBRATTON : Cal BC 190 to Cal AD 10 (Cal BP 2140to 1940)

Beta 2023 16 3140+/- 160BP -24.9 0100 SAMPLE : O W 04 SAMPLE NO 37 ANALYSIS : RadiomehioStandard delivery (concentration of charcoal h m within sediment mapix) MATFNALPRETREATbEM : (charred material): acid/alkali/acid Cal BC 1750 to 970 (Cal BP 3700 to 2920) 2 SIGMA CALIBRATION :

3140+1- 160BP

I CALIBRATION O F RADIOCARBON AGE TO CALENDAR YEARS 1, 1 I
(Variables: C 13lC 12=-26.2:lab. m u l e I ) Beta-202315 Laboratory number: Conventional radiocarbon age: 2 Sigma calibrated result: (95% probability) 2080*40 B P Cal B C 190 t o C a l AD 1 0 (Cal BP 2140 t o 1940) Intercept data

I

Intercept of radiocarbon age with calibration curve: 1 Sigma calibrated result: (68% probability)

Cal BC 80 (Cal BP 2030) Cal BC 160 to 40 (Cal BP 2 120 to 1990)

Charred m a t e d d

References:

1

1

D~lrrbasr used INTCAL98 CalibratYon Dafnbasc Editorial C o m m cnt Sluiver, M..v a n d e r p l i c h t . H 1998. RIEdimnrbm 40(3), pdi-*ii . INTCAL98 Rgdiocnrbon Age Calibration Sluiver. M., oL, 1998. Radiocarbon 40(3), ~ 1 0 4 1 - I 0 8 3 er. Mothemolics A SimpliJlcd A p p r o m h to Colibraling C l 4 D n f u Talma. A . S.. Vogel, J . C . , 1993. Radiocarbon 35f2). p317-322

Beta Analytic Radiocarbon Dating Laboratory
498JSW. 74th Courr.Mlmnr, Fiorlda 33IJJ. Trl; (303)667-5167 .Fa: pOJ)663-(1964- 8-Moll: be~@rndiocmbon.com

I I I I

CALIBRATION O F RADIOCARBON AGE TO CALENDAR YEARS
(Variables: C13/C12=-24.9:lab. m u l F l ) Laboratory number: Conventional radiocarbon age: 2 Sigma calibrated result: (95% probability) Intercept of radiocarbon age withcalibration curve: Beta402316 3140i160 B P Cal BC 1750 to 970 ( C a l B P 3700 to 2920)
intercept data

Cal BC 1410 (Cal BP 3360)

1 Sigma calibrated result: Cal BC 1540 to 1210 (Cal BP 3490 to 3 160) (68% probability)
3 1 4 W 1 6 0 BP

Charredmaterid

3600

References:
~

I 4;

Da1abarcuscd INTCAL 98 CaUbmtion Dotabnsc Editorfa1 Comm emf Sluiver. M..v a n d e r Plichr. H . 1998. R d i a a r b m 40(3). pxii-xiii INTCALBB Radiocarbon A g c Calibration Sruivu. M ,e l . aL. 1998. Radiocurbon 40(3), p1041-1083 MolhcmntiP A S i m p l i j e d Approach l o C s l i b m t i n g C l 4 D n l c r Tolmo. A . S., Vogel, J . C . . 1993. Radiocarbon 35(2). p317-322

Beta Analytic Radiocarbon Dating Laboratory
4985 S W . 74th Cowt.MIom1. Flmfdo 33 1 5 5 . Tcl: (305)667-5167. F m : (305)663-0966. E-Mali: brro@radiocorbn.com

I

I I
I

Appendix 2
2005 Radiocarbon Dating Results

I I
I I

m
I B
I I I

I
I I

rn
I

I

Dr. John Roberts University of Manchester
Sample Data

Report Date: 11/9/2005 Material Received: 10/12/2005 Measured Radiocarbon Age 13C/12C Ratio Conventional Radiocarbon Age(')

Beta - 209508 930 +/- 60 BP -25.9 0100 SAMPLE : OVMOSSAMPLE NO1 7 ANALYSIS : Radiometric-Standarddelivery MATERIAUPRETREARviENT : (chmed material): acidalkalilacid Cal AD 1000 to 1250 (Cal BP 950 to 700) 2 SIGMA CALIBRATION :

4200 +/- 80 BP -26.3 o/ca Beta 209509 SAMPLE : OVMOSSAMPLE NO16 ANALYSIS : Radiometric-Standarddelivery (concentration of charcoal from within sediment matrix) MATERIALPRETREATMENT : (charred matend). acidalkal2acid ~ aBC 2920 to 2560 (Cal BP 4860 to 4510) AND Cal BC 2520 to 2500 (Cal BP 4480 to 4440) l 2 SIGMA CALIBRATION :

-

1

2120 +/- 60 BP -26.0 0100 2lOO+/- 60BP Beta 2095 10 SAMPLE : OVMO5SAMPLE NO13 ANALYSIS : Radiometric-Standard delivery (concentration of charcoal frum within sediment matrix) MATERIAUPRETREATMENT : (charred marer~al)ac~dalkalilacid Cal BC 560 to 290 (Cal BP 2310 to 2240) AND Cal BC 230 to Cal AD 30 (Cal BP 2180 to 1920) 2 SIGMA CALIBRATlON :

-

CALIBRATION O F RADIOCARBON AGE TO CALENDAR YEARS
(Variables: C 13lC 12=-25.9:lab. mult=l) Laboratory number: Conventional radiocarbon age: Beta-209508 920i60 B P Cal AD 1000 to 1250 (Cal BP 950 t o 700) Intercept data Intercepts of radiocarbon age with calibration curve: Cal AD 1060 (Cal BP 890) and Cal AD 1080 (Cal B P 860) and Cal AD 1150 (Cal BP 800) Cal AD 1030 to 1190 (Cal BP 920 to 760)

2 Sigma calibrated result: (95% probability)

1 Sigma calibrated result: (68% probability)

.. . . ... .. .

800

1000

1020

1040

1060

1080

1100

1120 C a l AD

1140

1160

1180

1200

1220

1240

1260

References:
Daf#bnse used

INTCAL 98 Calibration Dmtabase EdiIoriol Contm cnf Sluiver, M.. v o n d e r Plichr. H . 1 9 9 8 , R o d i w a r b a , 40(3), pxii-siii INTCAI.98 Radiocarbon A g e Cdibrafion Sluiver. M..el. al.. 1998. Radiocarbon 40(3). ~ 1 0 4 1 - 1 0 8 3 MrzfhcmaLlcr A Simpliflcd A p p r o m h to Calibrating C I 4 D a f u Tolma, A . S.. Vogel. J. C., 1993. Radiocarbon 35(2), ~ 3 1 7 3 2 2

Beta Analytic Radiocarbon Dating Laboratory
1985 S W . 74th Courr.Mimi. Florida 33 155 Tsl: (305)66 7-51 67. F m : (305)6634964. EMoII: bcra@rodioclrrbon.com

-

CALIBRATION O F RADIOCARBON AGE TO CALENDAR Y E A R S
(Variables: C13/C12=-26.3:lab. mult=l) Laboratory number: Conventional radiocarbon age: 2 Sigma calibrated results: (95% probability) Intercepts of radiocarbon age with calibration curve: Beta-209509 4180*80 B P Cal BC 2920 to 2560 (Cal B P 4860 to 4510) and Cal BC 2520 to 2500 (Cal B P 4480 to 4440) Intercept data Cal BC 2870 (Cal BP 4820) and Cal BC 2800 (Cal BP 4750) and Cal BC 2770 (Cal BP 4720) Cal BC 2890 to 2 620 (Cal BP 4840 to 4 570)
Chsrredmaterid

1 Sigma calibrated result: (68% probability)
4 1 6 W 8 0 BP

2950

29W

2650

2800

2750

2700

2650

2550

2500

2450

Cal B C

References:
Dofobasc used I N T C A L 98 C n l i b m d o n Dnfabose Editorial C o m m m t Sluiver. M., v a n d e r Plish;, H. 1 9 9 8 , R m l i a o r b m 40(3). pxii-xiii INTCAL98 Radiocarbon A g e C d i b r o d o n S o i v e r . M .e l . aL, 1998. Rodiourrbon 40(3). p l O 4 1 - 1 0 8 3 Mmthcmodct A Simplified Approoch to C n l i b r a d n g C I 4 D a f e r Tolmo. A . S.. V o r e l . J . C . . 1993. Rndiocarbon 35(21, p 3 1 7 J 2 2

-

~

Beta Analytic Radiocarbon Dating Laboratory
4985 S.W. 7 r r h Court.Mtoml. Florida 33155

-

Tci: (305)66>5167

Fm: (305)663-0961.

£ M o i l : be,a@rodiocorbon.com

CALIBRATION O F RADIOCARBON AGE TO CALENDAR YEARS
(Variables: C13/Cl2=-26:lab. mult=l ) Laboratory number: Conventional radiocarbon age:
2 Sigma calibrated results: (95% probability)

Beta-209510 2100i60 B P Cal BC 360 t o 290 (CaI BP2310 t o 2240) and Cal B C 230 t o C a l AD 3 0 (Cal BP 2180 t o 1920) Intercept data

Intercept of radiocarbon age withcalibration curve:
1 Sigma calibrated result: (68% probability)

Cal BC 110 (Cal BP 2060) Cal BC I90 to 40 (Cal BP 2 140 to 1990)

References:
Dnlobase used INTCAL98 Calibration Dnlobnsr Editorial Comm en1 Sruiver, M., v o n d e r Plichr, H . 1998. R r d i o c a r b m 40(3),pxii-riii Z N T C A L 9 8 Radiocarbon Age Colibrntion Sfviver. M., e l . 01.. 1998, Radiocarbon 40(3). ~ 1 0 4 1 - 1 0 8 3 Malhcmnlia A Simplified Appromh Io Cnlibroting C I 4 D a l e s Tolmo. A . S.. Yopel. J. C.. 1993. Radiocarbon 35/21, ~ 3 1 7 - 3 2 2

Beta Analytic Radiocarbon Dating Laboratory
4985 S W :i 4 r h C o u r r . M i o m l . F l o r i d a 33 135 .

.Tsl: (305)667-5167 .

F a : (305)663-0963. 6-Moll:b ~ r o @ r o d l o c o r b o n . c o m

I I I

Appendix 3

I

2005 Plant macrofossil and pollen assessment

I
I I

Archaeological Services University of Durham

I
I I I

Old Vicarage, Mellor, Greater Manchester plant macrofossil and pollen assessment
on behawof University of Manchester Archaeological Unit

Report 1349 October 2005

Archaeological Services Durham Universi@ South Road Durham DH1 3LE Tel: 019i 334 1121 Fax: 0191 334 1126 archaeological.services@d~~ham.ac.uk www.durham.ac.uk/archaeologicalse~ces

Old Vicarage, Mellor, Greater Manchester
plant macrofossil and pollen assessment Report I349
October 2005
Archaeological Services Durham University on behalf of Univers* of Manchester Archaeological Unit Universi~ Manchester, Oxford Road Manchester, MI3 9PL of

Contents
1. Summary

.

2. 3. 4. 5. 6. 7.

Project background Methods . Plant macrofossils . Pollen Recommendations . References .

1 2 2 3 4 4 5

O Archaeological Senices 2005

Old Vicarage, Mellor: plant macrofossil andpollen assessment; Report 1349, October 2005

Summary
The project This report presents the results of plant macrofossil and pollen assessment of two samples taken during an excavation of a prehistoric settlement at Mellor, Greater Manchester. Plant macrofossil assessment Charred hazelnut fragments were abundant in context 285 but charred remains were absent from context 343. The results suggest that context 285 is made up of domestic waste. Pollen assessment Pollen was present in both contexts but was largely degraded or corroded. The results suggest that woodland cover was limited and the landscape was dominated by open and disturbed ground with areas of damp ground and possible heathland. Recommendations No further plant macrofossil work is recommended for the samples. Material suitable for AMS radiocarbon dating is present in context 285.
Further pollen work is recommended for context 285. Pollen concentration is too low in context 343 to facilitate further analysis.

Archaeological Services Durham University

1

Old Vicarage, Mellor: plant macrofossil andpollen assessment; Report 1349, October 2005

2.
2.1

Project background
Location and background A series of archaeological works have been carried out by Manchester University Archaeological Unit at Mellor since 1998 and have confirmed the presence of an Iron Age Hillfort. In August 2003, trench 22 was excavated. over the line of the enclosure ditch which surrounds the hilltop. Plant macrofossil and pollen work was undertaken on samples from this trench by Archaeological Services Durham University (Archaeological Senices reports 1059, 1073, 1090). Plant macrofossils were poorly preserved and only charred remains of hazelnuts occurred. Pollen indicated that a mixture of arable and pastoral farming was being carried out. Trench 36 wis excavated during August 2005 in a field to the north of the Old Vicarage at Mellor, near Stockport (NGR SJ 9818 8890). Two environmental samples were collected. Context 285 is the fill of a post hole from which, what is believed to be Bronze Age pottery was recovered. Context 343 is the fill of a pit thought by the excavators to represent Iron Age or Romano British activity. Objective The objective of the assessments was to determine the archaeobotanical potential of the material and to make recommendations for further work if appropriate. Dates Plant macrofossil and pollen assessment was carried out between 31d- 17" October 2005. This report was prepared on the 18" October 2005. Personnel Sample processing was carried out by Dr David Webster and Mr Lome Elliott. The assessments and report preparation were undertaken by Dr Charlotte O'Brien. Archive The site code is OVM05, for Old Vicarage, MeUor, 2005. The flots and pollen preparations are retained in the Environmental Laboratory at Archaeological Services Durham University for collection.

2.2

2.3

2.4

2.5

3.
3.1

Methods
Plant macrofossils The samples were manually floated and sieved through a 500 pm mesh. The residues were retained, described and scanned using a magnet for ferrous fragments. The flots were dried slowly and scanned at x 40 magnification for waterlogged and charred botanical remains. Identification of these was undertaken by comparison with modem reference material heId in the Environmental Laboratory at Archaeological Services Durham University. Relative abundance of remains per species was logged and the results were

Archaeological Services Durham Universiry

2

Old Vicarage, Mellor: plant macrofossil andpollen assessment; Report 1349, October ZOO5

interpreted in their archaeological and palaeoecological contexts. Plant taxonomic nomenclature follows Stace (1997). Pollen 3.2 One ml of sediment from contexts 285 and 343 were assessed. Pollen was extracted using potassium hydroxide to remove humic and fulvic acids, acetolysis to digest organic deposits and a heavy liquid technique to separate the pollen from rninerogenic material. Lycopodium spores were added to allow the calculation of pollen concentration. The pollen was mounted in silicone fluid and scanned at high magnification. Identification of pollen and spores was undertaken by comparison with modem reference material. Plant taxonomic nomenclature follows Stace (1997).

4.
4.1

Plant macrofossils
Results Charred hazelnut kagments were abundant in the flot of context 285. Other plant macrofossils were absent and no plant remains occurred in context 343. Calcined bone and charcoal were present in both contexts and small amounts of coal and modem roots were present in context 343. Insect fragments occurred in context 285. The contents of the flots and residues are listed in Table 1.

Table 1. Plant macrofossils ffom OVM05 . .. . ' '. :- . . - . .: . ... .. .. . .
p
>,".
~

.z~::.'..

A:-:

..-a ::. , :.

.

.

.

.

,

:-I+.? ..

.,....:.-.,. ... - .2 .1:.:.-13 <'./:.'5:14;:. .-: .. ,) .. . . ;.<. :.283:-:, 343::
:

1; 1 1 1

I ,

1:i snnn , won - - -- I - - - 200 200
2

Volume offlot (ml) Volume offlot assessed (ml) Residue contents (relative abundance) Calcined bone Flot mairii- (relative abundance) Charcoal Coal Insect I Modem roots
(t: treedshrubs)

1 1

5
5

1
1

1
1

1

3
1

kelative abundance is based on a scale from 1 (lowest) to 5 (bighest).

4.2

Discussion The abundance of charred hazelnut fragments in context 285, the fill of a post hole, suggests that these nuts formed an important part of the diet. Hazelnuts and other gathered wild foods were used profusely during the Neolithic period. From the Bronze Age onwards their importance diminished with the increase in cereal cultivation. The presence of the charred remains in combination with

Archaeological Services Durham Universiv

Old Vicarage, Mellor: plant mocrofossl andpollen assessment; Report 1349, October 2005

charcoal and calcined bone fiagrnents suggests that the context is dominated by domestic waste.

4.3'

A few insect fragments were recorded in the flot of context 285, however these are likely to be modem introductions. Modem roots occurred in context 343. The absence of plant remains in context 343 means that no chronological or economic information can be provided about this pit fill.

5.
5.1

Pollen
Results Pollen was present in both contexts but was largely degraded or corroded. Pollen was more abundant in context 285 than 343. Context 285 was dominated by bracken with lower numbers of other ferns including the royal fern. Low numbers of arboreal pollen occurred and included hazel, alder and birch. Sedges and grass were also present and microscopic charcoal was abundant. Context 343 was dominated by fern spores, particularly those of the royal fern. Arboreal pollen was present in low numbers and included hazel and willow. A few grains of devil's bit scabious and heather were also recorded. Charcoal was sparse. Discussion Low numbers of arboreal pollen grains occurred in both contexts indicating that woodland coverage was limited. The non-arboreal pollen assemblage indicates the presence of open and disturbed ground. Royal fern and sedges indicate areas of damp ground. Heather pollen occurred in context 343 which suggests the presence of nearby heathland.

5.2

5.3

6.
6.1

Recommendations
Plant macrofossik No further plant macrofossil work is recommended for the contexts. The hazelnut fragments in context 285 would be suitable for AMS radiocarbon dating. Material suitable,for radiocarbon dating is absent from context 343. Pollen Sufficient pollen was present in context 285 to facilitate further analysis and counts of 300-500 grains would be possible. This would provide a detailed picture of the local environment and the extent and diversity of woodland cover at the site. The plant macrofossil assessment indicates that hazelnuts formed an important part of the diet and further analysis of the pollen would establish what proportion of the local woodland was made up of hazel. Full analysis may also provide information about farming activities at the site.

6.2

I

Archaeological Services Dwham Universiry

4

Old Vicarage, Mellor: plant macrofossil andpollen assessment; Report 1349, October 2005

6.3

No further pollen work is recommended for context 343 due to the low pollen concentration.

7.

References
Archaeological Services (2003) Mellor, near Stockport, Greater Manchester: plant macrofossil assessment. Archaeological Services Durham University Report 1059: Archaeological Services (2004) Mellor, near Stockport, Greater Manchester: pollen assessment. Archaeological Services Durham University Report 1073. Archaeological Services (2004) Mellor, near StocLport, Greater Manchester: pollen analysis. Archaeological Services Durham University Report 1090. Stace, C. (1997) New Flora of the British Isles. 2nd Edition. Cambridge University Press. pp1130.

Archaeological Services Durham University

5

I I I 1 I I I I I I I I 1 I I I I I I I

Appendix 4

2005 Pollen analysis

-

Archaeological Services University of Durham

Old Vicarage Mellor, near Stockport, Greater Manchester pollen analysis

on behalfof The University of Manchester Archaeological Unit

Report 1361 November ZOOS

Arclraeologicaf Services Durham University

South Road
Durham DHl3LE Tel: 0191 334 1121 Fax: 0191 334 1126 archaeological.se~ces@durham.ac.uk www.durham.ac.uk/archaeologicalservices

Old Vicarage Mellor, near Stockport, Greater Manchester
pollen analysis

Report 1361
November 2005 Archaeological Services Durham Universiw on behalf of: University of Manchester Archaeological Unit University of Manchester, Oxford Road, Manchester, M13 9PL

Contents
1. Summary

.

1

2. 3. 4. 5. 6. 7.

Project background Methods . Results . Discussion . Conclusions References .

2 2 3 3 4 4

O Archaeological Services 2005

I
1.

Old Vicarage Mellor, Greater Manchester; pollen analysis: Report 1361, November 2005

Summary
The project background This report presents the results of pollen analysis of context 285, which was sampled during an excavation of a prehistoric settlement at Mellor, Greater Manchester. Methods Material fiom context 285 was processed using standard techniques including heavy liquid separation to concentrate the pollen. The residues were scanned for pollen and spores which were identified by comparison with modem reference material. Results The pollen indicates the presence of mixed deciduous woodland dominated by hazel. There also appears to have been a nearby wet meadow and possible alder can. The cereal-type pollen and associated weeds indicate a mixed farming economy.

1.1

1.2

1.3

Archaeological Services Durham Universi@

1

'

Old Vicarage Mellor, Greater Manchester; pollen analysis: Report 1361, November ZOO5

2.
2.1

Project background
Location and background A series of archaeological works have been carried out by University of Manchester Archaeological Unit at Mellor since 1998 and have confinned the presence of an Iron Age Hillfort. In August 2003, trench 22 was excavated over the line of the enclosure ditch which surrounds the hilltop. Plant macrofossil and pollen work was undertaken on samples from this trenchby Archaeological S e ~ c e Durham University (Archaeological Services 2003; 2004a; 2004b). s preserved and only charred remains of hazelnuts Plant macrofossils were occurred. Pollen indicated that a mixture of arable and pastoral farming was being carried out. Trench 36 was excavated during August 2005 in a field to the north of the Old Vicarage at Mellor, near Stockport (NGR SJ 981 8 8890). Two environmental samples were collected. Context 285 is the fill of a post hole from which, what is believed to be Bronze Age pottery was recovered. Context 343 is the fill of a pit thought by the excavators to represent Iron Age or Romano British activity. Plant macrofossil and pollen assessments were undertaken on these contexts by Archaeological Services Durham University (Archaeological Services 2005). Charred hazelnut fragments were abundant in context 285 but charred remains were absent from context 343. Pollen was present in both contexts but was largely degraded or corroded and further analysis of context 343 was not recommended due to low pollen concentration. The assessment results suggested that woodland cover was limited and the landscape was dominated by open and disturbed ground with areas of damp ground and possible heathland. This report presents the results of full pollen analysis of context 285. Objective The objective was to cany out pollen analysis of context 285 in order to reconstruct the palaeo-environment and to provide information about the economics and agricultural practices in the area. Dates Pollen analysis was carried out between 28" October - 9" November 2005. This report was prepared on 10" November 2005. Personnel Pollen preparation was undertaken by Mr Lome Elliott (Archaeological Services). Pollen analysis and report preparation were carried out by Dr. Tim Mighall (University of Aberdeen). Archive The site code is OVM05, for Old Vicarage Mellor 2005.

2.2

2.3

2.4

2.5

3.
3.1

Methods
One mI of sediment from context 285 was analysed. The samples were prepared using standard procedures (Barber 1976), which included density floatation (Moore et a 1991) in order to concentrate the pollen. Identification of pollen 1 and spores was undertaken by comparison with modem reference material. Plant taxonomic nomenclature follows Stace (1997).

Old Vicarage MeNor. Greater Aflanchester;pollen analysis: Report 1361, November 2005

4.
4.1

Results The maioritv of the Dollen in context 285 was degraded and this restricted the - . pollen count to 230 land pollen grains, excluding aquatic pollen and spores whlch were also recorded. The results, showing absolute numbers, are shown in Table

1.

Table 1: Pollen and Spore assemblages from context 285 (absolute numbers).

5.
5.1

Discussion
The pollen assemblage of 285 is dominated by arboreal pollen. A hazeldominated mixed woodland is suggested by the abundance of Corylus avellanatype. Quercus (Oak) and Tilia (Lime) also form a minor component of woodland occupying dryland soils with ferns such as Polypodium (Polypody), Pteridium (Bracken) and Filicales (Undifferentiated ferns) contributing to the understorey.

Old Vicarage Mellor, Greater Manchesrer; pollen analysis: Reporr 1361. November 2005

Some Alnus (Alder), possibly alder cam, is present on a low lying, wet substrate e.g. a valley bottom. Poaceae (Grasses) and Cyperaceae (Sedges) are the dominant non-arboreal (NAP) pollen taxa but a suite of herbaceous pollen is also well-represented including Asteraceae (Daisy family), Polygonum (Knotgrass) and Plantago lanceolata (Ribwort plantain). However, other NAP taxa and spores are present. These include CalIuna (Heather), Caryophyllaceae (Pink family), Chenopodiaceae (Goosefoot family), Ranunculaceae (Buttercup family), Filipendula (Meadowsweet), Succisa (Devil's-bit scabious), Polypodium and Filicales. 5.2 The occurrence of cereal-type pollen provides evidence of arable cultivation. Plantago lanceolata, Polygonum, Ranunculaceae, Lactuceae (Dandelion group), Chenopodiceae and Pteridium have been associated with agricultural ecosystems and suggest that a mixed farming economy - cereal cultivation combined with pasture - was practised (Behre 1986). Pottery, that is believed to be Bronze Age, was recovered from context 285 during the excavation and palaeoecological evidence from other sites in northwest British Isles suggests that sustained arable and pastoral farming was undertaken within the region during the Bronze Age (Chiverrell2003; Walker 1966; W i b l e et a1 2000). Cyperaceae, Filipendula and Succisa are all indicative of wet meadows and grassland. The pollen assemblage obtained fiom sample 285 is, therefore, similar in composition to samples 442 and 449 from Trench 22, the results of which are presented in Archaeological Services report 1090.

6.
6.1

Conclusions
The pollen indicates the presence of mixed deciduous woodland dominated by hazel. There also appears to have been a nearby wet meadow and possible alder carr. The cereal-type pollen and associated weeds indicate a mixed farming economy.

7.

References
Archaeological Services (2005) Old Vicarage Mellor, Greater Munchester: plant macrofossil andpollen assessment. Archaeological Services Durham University Report 1349 Archaeological Services (2004a) Mellor, near Stockport, Greater Manchester: pollen analysis. Archaeological Services Durham University Report 1090. Archaeological Services (2004b) Mellor, near Stockport, Greater Manchester: pollen assessment. Archaeological Services Durham University Report 1073. Archaeological Services (2003) Mellor, near Stockport, Greater Manchester: plant macrofossil assessment. Archaeological Services Durham University Report 1059. Barber, K E, 1976 History of vegetation. In (SB Chapman, Ed.) Methods in Plant Ecology. Oxford: Blackwell, pp5-83 Behre, K-E, 1986 Anthropogenic indicators inpollen diagrams. Rotterdam

Old Vicarage Mellor, Greater Manchester; pollen analysis: Report 136/. November ZOO5

Chiverrell, R C, Innes, J B, Blackford, J J, Wood, J J, Davey, P J, Tomlinson, P R,Rutherford, M M and Thomas, G S P, 2004 Palaeoecological and archaeological evidence for Bronze Age human activity on the Isle of Man. The Holocene 14, 3,346-360 Moore, P D, Webb, J A and Collinson, M E, 1991 Pollen Analysis. 2ndedition. London Stace, C, 1997 Newflora of the British Isles. 2ndEdition. Cambridge Walker, D, 1966 The late Quaternary history of the Cumberland lowland. Philosophical Transactions of the Royal Society of London B 251, 1-120 Wirnble, G, Wells, C R and Hodgkinson, D, 2000 Human impact on mid- and late Holocene vegeation in south Cumbria, UK. Vegetation History and Archaeobotany 9, 17-30

Archaeoloaical Services Durham University

1 I I I

Appendix 5 Geological analysis of the boulder clay at Mellor.

MeUor Archaeological Site Notes arising from visit by Morven Simpson and Fred Broadhurst, 31" August 2005 Boulder Clav or Till Boulder Clay is the weathered remains of material deposited by melting ice. At Mellor the boulder clay dates from the end of the Devensian Glaciation about 10,000 years ago. The Devensian is the last of the glaciations which have occurred in the area over the last 1%- 2 million years. The material when deposited by the ice would consist of boulders and other debris varying in size down to finely ground material to which the name rock flour is ffequently given. The components of the rock flour have a large surface area/volume ratio. This results in rapid chemical weathering and the breakdown of the rock material to form clay minerals (plus soluble material which is removed in solution). So the boulder clay observed at Mellor is a clay (once rock flour) with contained pebbles/boulders derived from outcrops crossed by the ice on its way to Mellor (ie from Southern Scotland, the Lake District and localities between the Lake District and Mellor) The present distribution of boulder clay at Mellor is determined by the original areas of deposition/nondepositiontogether with the effects of erosion on site since deposition. So, at Mellor today, there are areas with no boulder clay, and other areas with variable thicknesses of boulder clay. Iron Dan and clav colouration The iron pan shown to us at the Mellor site is associated with the presence of underlying boulder clay. The evidence on the ground suggests the action of downwards movement of water through soil containing organic matter. Such water becomes acidic by the addition of humic acids and would tend to lose dissolved oxygen by reaction with organic carbon. Acidic, deoxygenated, water is a significant solvent for iron (present in the soil, notably linked to the organic carbon). The downwards percolation of the iron-bearing solution would be arrested by the underlying essentially impervious boulder clay and so be ponded - providing an opportunity for oxidation and hydration of the iron to form an iron pan (mixture of fenic oxides and hydroxides). Dissolved oxygen in the groundwater within the sandstones beneath the boulder clay (especially where the boulder clay is thin) probably plays a r61e in the iron pan formation. We were shown patches of clay and silt at the top of the boulder clay where there were distinctive colour variations, sometimes brown, red, sometimes grey. The brownish, reddish areas contain iron in its oxidised, hydrated state. In larger amounts this would amount to an iron pan. The grey areas did not contain oxidised or hydrated iron. The presence or absence of oxidised and hydrated iron will be related to the nature of the downward movement of water from the soil layer above. The absence of an iron pan in the boulder clay-l?ee areas will result ffom the lack of any barrier to the downward movement of water. It is likely that oxidation and hydration of iron will occur, but not at one specific level. ,

FMB

Appendix 6
2004-5 Romano-British Pottery Assessment

Mellor Summary of Romano-British pottery recovered from excavations in 2004-5 R.S. Leary
A total of 72 sherds (697g.) were examined, including sherds from excavations 2002, 2003,2004 and 2005 (table 1). Although a smaller number of Romano-British sherds were recovered from the excavations in 2004 and 2005 than previously, a similar picture in terms of the range and character of the assemblage was disclosed in some respects (charts 1-2). Derbyshire ware was common along with Cheshire Plains oxidised wares and significant amounts of samian ware. Traded wares from Mancetter-Hartshill, Wroxeter, Wilderspool and the Midlands were present as well as the table wares from Gaul. In contrast to previous excavations no BB 1 from Dorset was identified amongst the material from 2004-5. Wroxeter white ware mortaria have been identified at Manchester as a major mortaria supplier in the period AD100-1501160. The fabric is very like Mancetter-Hartshill and some, but not all, of the mortaria previously identified as Mancetter-Hartshill mortaria have been reclassified as Wroxeter white ware mortaria. This shows that Mellor had close ties with the military supply network during this period.

Table 1 quantity of sherds from excavations by year.
Relative proportion of fabrics by sherd weight and count for excavations in 2004
80% 70% 60% 50% 40% 30% 20% 10% 0%

0

4

XI

G7

% '

0
% 7

t 9

C

* ?

+ %

.,

Chart 1

Relative proportion of fabrics by sherd weight and count for excavations in 2005
60% 50% 40% 30% 20% 10% 0%

Chart 2 Much of the group from 2004 came from a much fragmented orange ware flagon which had suffered from burial conditions resulting in severe surface erosion. Although the form was a simple plain necked flagon with everted rim, the fabric and type point to a date in the Hadrianic-Antonine period and a source at the Wilderspool kilns. The rest of the material excavated in 2004 included sherds of Derbyshire ware, a white ware base, probably from a flagon or beaker and two samian scraps. A date in the HadrianicAntonine would account for all these fabrics and forms. The group from 2005 was also dominated by oxidised wares and Derbyshire ware. The sherds comprised undiagnostic bodysherds but again the former compared well with material from Manchester and Wilderspool. The finer wares may well have been made at Manchester but in the absence of diagnostic vessel types, it is difficult to be certain. The group also included four samian sherds, two Mancetter-Hartshill mortariurn dating to AD 180-240. Four shelly ware sherds compared with material from the previous excavations identified as late shell-tempered ware. This is thought to appear in this region in the fourth century. These assemblages are too small to assess in terms of proportions of vessel types but both groups included samian table ware and the pottery from 2004 included one or probably two flagons indicating the desire to acquire Roman table ware seen in previous years. The traded wares demonstrate links with the Roman trading network whether through trade or gift exchange. Bibliography Colyer, C., Gilmour, B.J.J. and Jones, M.J., 1999, The defences of the Lower City. Excavations at the Park and west Parade 1970-2 and a discussion of other sites excavated up to 1994, Council for British Archaeology Research report 1 14.

Darling, M.J., 1999, 'Roman Pottery', In Colyer, Gilmour and Jones 1999,52-123 Gillam, J. P., 1970, Types of Roman Coarse Pottery Vessels in Northern Britain, third edition, Newcastle Hartley, K.F. and Webster P.V. 1973 The Romano-British kilns near Wilderspool. Archaeol. Journ. 130,77-104 Tomber, R. and Dore, J., 1998, The National Roman Fabric Reference Collection. A Handbook, MoLAS Monograph 2. London

Mellor 2004-5 Catalogue

Comments

Abbreviations

i_-."

1_ . . ~- .

Fabric code _ . -

I
I

..

Fabricname

_

, .

. . .. .

I
1

.

.

.
. .

'

.

~

:
.

Fabrics..
, ,

-

-

~

- . -

-___
-. .- - . .

. -. : . '-:

~

<<-..!
~

_

_

.

_

_

Fabric description
.. . -. . . .. . . - .~

- -

~.~

1 0 ~ ~ 1 Fine orange ware I

ITS

Samian

I

I
I

Cheshire plains fine ware, orange to pale orange. Soft with powderylsandy feel and smooth fracture. I Sparse, well-sorted, fine quartz and sparse ill-sorted fine to medium, rounded red brown inclusions. I~icaceous I 1 0 ~ ~ 1 Sandy orange Cheshire Plains medium orange, hard to soft with rather sandy feel and quite smooth fracture. Sparseware moderate, ill-sorted medium to coarse subangular quartz, sparse, ill-sorted, rounded redlbrown and grey 1 I I i f inclusions OBAl l ~ i n buff ware 1 !AS OAAl but buff e I I 1. 1 jOBA2 !Fine buff ware 2 l ~ e d i u m buff. Soft with powderylsandy feel and irregular fracture. Common, well-sorted, fine quartz and ; I \sparse ill-sorted fine to medium, rounded red brown inclusions. As OAIBAI but more quartz. 1 1 - -. . -.. T s a n d y buff ware ]AS OAB 1 but buff

i

]

1

I
I

-

-

-

,-

I

I

I

OAClOBCl Pre-Derbyshire ware

1
~GRA . .
/BB1 FLB2 FLAl
,,

Orange (OAC) or buff (OBC). Hard, rough with irregular fracture. Moderate, coarse, ill-sorted, subangular quartz, often crystalline appearance suggesting quartzite; moderate, coarse, ill-sorted, rounded, black or brown inclusions, i probably. iron oxides. . , ... ,..,, ,.,,,, ,.. ..,, ." ..,, ...., . ., , ! .. ., ..,, ,.,,, ,,, ,,,. . . , ,., , . ,. . , , ..,,, , , .. .., . j 1, ~ i n,e,- , ware ", 1Grey...,Very,. ,,fine, hard, ..,smooth with smooth ,, grey fracture. .,.. ,.. .,. "..,, ,,...,fine .,subrounded ., .,. .. ,,., . Sparse, . ., -, -- quartz .. , ., . .... ,..,, ...,,, ,,,, .,,.,, , . ,. . ,., , . . .. .,,, ,. ,,, , .,.,, ,

I

~

,

,,

l~lack burnished Tomber & Dore 1998 South-East Dorset BBI (DOR BB1). lware 1 I ]white-slip l~ard with sandy feel and irregular fracture. White slip. Moderate well-sorted medium subangular quartz, j lorange ware ]sparse, coarse. rounded grey inclusions

-

AS

,

~

, ,

\

-

]white ware

{cream. Hard with smooth feel and almost conchoidal fracture. Rare, well-sorted, very fine, subangular quartz and

/
I

i
-",,"."."v-.-..aM~

I

Hartshill mortaria

with a self-coloured slip. Inclusions usually moderate, smallish, transparent and translucent white and pinkish quartz with sparse opaque orange-brown and rarely blackish fragments;....... . rarely white clay pellets -" i=___ml_._.. ," " ... ...
."m "-.,-~.,",ww..>,,~,~.,..- _ I 1 _ " " _; j _ I L

.

. "

I
1

j l i

i

ii -".ii lu-,:.-

i :_ -- Fabric code
--

.

.

. .

.. . . .

-.. .~ .- - ..~
~

F abrics -.
.- .

.

.
,
'

-. ..-.
~

.

. . ~~. . .. -.
~~~

____
.
~

-.

Fabric name
.: . .

Fabric description

.
.
~~~

1

1

.

-~

.

--

'MWROX

(or re-fired pottery Wroxeter white Cream fabric, variing in texture fiom softish to very hard and often with a buff-cream slip. Inclusions: ware mortarium again varying, moderate to frequent, ill-sorted quartz, red-brown and opaque black material. Trituration grit: mainly quartz, quartz sandstone, red-brown sandstone, black rock
Derbyshire ware :--~=~~>-.:~-~---- Tomber and Dore 1998 PER.-.CO As " ~ ===-,::

1
1I
i

DBY . - ,.d;u -: *. ." . d -

-.

!
i :.,.

~

~~-~,,..,.:;~,~._a-~-..._.~ ~-

~

. ~ - . ~ . . . . . . 4 . _ _ ~ _ ~ . ~_._~~,~,.~_,.,..L~.A.w,,,.,........_._

:=.

:.;-,

~

BKR bodysherd of beaker CHM chamfered base, . ,,,. - -- , -, -, , . ..., , FLG flange
~IRS/incomplete rim section1

Appendix 7

Sources

Arrowsmith P, 1997, Stockport - A History. Stockport. Stockport MBC Ashworth A M & Oldham T F, Mellor Heritage. The Parochial Church Council of St Thomas's Mellor. Marple. Beresford G, 1987, Goltho: the development of an early medieval mannor. English Heritage. Bevan B, (ed), 1999a, Northern Exposure, interpretative devolution and the Iron Ages of Britain. Leicester, Leicester Archaeological Monographs. Bevan, B. 1999b Northern Exposure: interpretative devolution and the Iron Ages of Britain. In Bevan, B. (ed) Northern Exposure: interpretative devolution and the Iron Ages of Britain, 1-19. Leicester: Leicester Archaeological Monographs. Bevan B, 2000a, The Peak District Romano-British Rural Settlement Survey. Unpublished research report for Derbyshire Archaeological Advisory Committee. Coombs D & Thompson H, 1979, 'Excavation of the hillfort of Mam Tor, Derbyshire 196 5-69', Derbyshire Archaeological Journal 99, 7-5 1. Coombs D G, 1982, 'Excavations at the Hill Fort of Castercliff, Nelson, Lancashire 19707 1 ', Transactions Lancashire 6; Cheshire Antiquarian Sociely 81, 111-130. Cowell R W, 2002, Romano-British and Late Prehistoric Excavafions at Duftons Farm, Lathom, Wesr Lancashire, 2nd interim report, 1999-2001. unpublished research report, Liverpool Museum. Cumberpatch C G, 1997, 'Pottery', in Parker Pearson M & Sydes R E , 'The Iron Age enclosures and prehistoric landscape of Sutton Common, South Yorkshire', Proceedings of the Prehistoric Society 63; 237. Cumberpatch C G, 2000, Later prehistoricpotte~from Mellor (OVM99 and OVMOO). Unpublished archive report for the Mellor Archaeological Trust. Cumberpatch C G, 2002, Laterprehistoricpottetyfrom Mellor (OVMOI). Unpublished archive report for the Mellor Archaeological Trust. Cumberpatch C G, Ixer R, Leary R, Moms E & Walster, A, 2003, Potteryfrom excavations at Mellor, Stockport, A review. Unpublished archive report for the Mellor Arc haeological Trust. Cunliffe B, 1991, Iron Age Communities in Britain. Routledge, London. Third Edition. Cunliffe B W, 1995, Iron Age Britain. English Heritage, Batsford, London

Eyre-Morgan G, 1998, Mellor 1998. Excavation Report. Unpublished research report, UMAU. Eyre-Morgan G, 1999, Mellor 1999. Excavation Report. Unpublished research report, UMAU. Forde-Johnston J, 1965, 'The Iron Age Hillforts of Lancashire and Cheshire', Transactions of the Lancashire & Cheshire Anfiquarian Sociep 72, 9-46. Geoquest Associates, 2002, Geophysical Survey of the three areas at Mellor, Stockport, 2002. Unpublished Report. ' Gregory R A, 2001, Geophysical ir~vestigatiorz the Old Vicarage, Mellor. Unpublished at research report. Harding D W, 2004, The Iron Age in Northern Britain, Celts and Romans, natives and Invaders. London, Routledge. Haselgrove C, Armit I, Champion T, Creighton J, Gwilt A, Hill J D, Hunter F & Woodward A, 2001, Understanding the British Iron Age, Agenda for Action. A Report for the Iron Age research Seminar and the Council of the Prehistoric Society English Heritage/Historic Scotland. Hawkes C F C & Hull M R, 1947, Camulodunum First Report on the Excavations at Colchester, 1930-1939. Oxford. Report of the Research Committee of the Society of Antiquaries 14. Henig M, 1984, Religion in Roman Britain. London, Batsford. Hermbrodt A, 1958, Der Husterhupp, Cologne. Hey D, 1980, Packmen, Carriers and Packhorse Roads. Leicester, Leicester University Press. Hicks S, 1971, 'Pollen-analytical evidence for the effect of prehistoric agriculture on the vegetation of North Derbyshire', New Phytologist 70, 647-667. Higham N J, 2000, 'The Tatton Park Project, Part 2. The medieval Estates, Settlements an d Halls', Journal of the Chester Archaeological Sociefy 75 (for 1998-99), 6 1-1 33. Higharn N J & Cane T, 1999, "The Tatton Park project, part 1: prehistoric to sub-Roman settlement and land use', Journal of the Chester Archaeological Society new series 74 for 1996-7), 1-61. Holden S, 2001, Excavations of the hiNfort at Mellor, Stockport, interim report 2000 and 2001 season. Unpublished report for University of Manchestes Archaeology Unit and the

Mellor Trust. Leary R, 2000, Mellor Romano-Britishpottery. Unpublished arctuve report for the Mellor Archaeological Trust. Leary R, 2003, Mellor Romano-Britishpottety 2002; a catalogue. Unpublished archive report for the Mellor Archaeological Trust. Lightfoot K W B, 1981, Cue Castell, Rumney, Card18final interim report. Annual Report. Medieval Archaeology XXV. le Maho J, 1982, Genese d'une fortrfication seignuriale. Les fouilles de la motte de Mirville (XI-XI1siecles) Ch. G. X I . Nevell M, 1999a, Living on the Edge of Empire, Models, Methodology and Marginality. Late-Prehistoric and Romano-British Rural Settlement in North West England. Archaeolo gy North West vol3 (no 13, for 1998), Manchester. Nevell M, 1999b 'Iron Age and Romano-British Rural Settlement in North West England, theory, marginality and settlement', in Nevell M, (ed), 1999a, 14-26. Nevell M, 1999c, 'Great Woolden Hall Farm. A Model for the Material Culture of Iron A ge and Romano-British Rural Settlement in North West England?', in Nevell M, (ed), 199 9,48-63. Nevell M, 2001, 'The edge of empire: late prehistoric and Romano-British settlement in of north west England. A study in marginality', in Higham N J, (ed), Archaeolog~ the Roman Empire. A tribute to the life and works of Professor Barri Jones. Oxford. Archaeopress (BAR International Series 940), 59-69. Nevell M, 2002, 'Legh Oaks Farm, High Legh, The value of Sample Excavation on Two Sites of the Late Prehistoric and Roman Periods', Journal Chester Archaeological Sociefy 77for 2001, 1 15-129. Nevell M, 2004, 'The Late Prehistoric and Romano-British Settlement of the Mersey Basin. A Study in Marginality', Journal of the Chester Archaeological Sociefy new series 78for 2003, 1-21. Noble P, Roberts J, Thompson A, 2004, Excavation Report for the 2004 season at Mellor, Stockport. Unpublished. Philpott R A & Adams M H, 1999, 'Excavations at an Iron Age and Romano-British Settlement at Irby, Wirral, 1987-96, An Interim Statement', in Nevell M (ed), 1999a, 64-73. Redhead N, 1999, 'Edge of Empire, Extra-Mural Settlement in a Marginal Context.

Roman Castleshaw', in Nevell M (ed), 1999% 74-8 1 Roberts J, 2002, Excavation Report for the 2002 season at Mellor, Stockport. Unpublished research report, UMAU. Roberts J, 2003 Excavations at Mellor an Interim Report. Unpublished research report, UMAU. Roberts J & Redhead N, 2003, 'Mellor - A new Iron Age hillfort', Current Archaeology 189,400-5.

Sign up to vote on this title
UsefulNot useful