The Silchester Environs Iron Age Project

Late Iron Age Oppida Conference 2017 - Book Now!

A review of recent and current research into Late Iron Age British towns and their landscapes

A one day conference, in which we will examine current understanding of British Iron Age oppida. For further details and how to book, visit our News & Events page.

The Project

The Silchester Environs Project is a five-year research undertaking to 2018 led by Professor Mike Fulford and managed by Dr Catherine Barnett at the University of Reading. It seeks to explore the later prehistoric use of the c.100km2 landscape around Silchester Roman Town and its underlying Late Iron Age oppidum. The project team is aiming to provide a context for the origins of the town and for changes that occurred during the transition to Roman urbanised living through examining settlement, agricultural activity, and landscape in the wider area. It builds on the exceptional body of information available for the town itself gained over nearly 20 years of annual excavations led by Mike Fulford and Amanda Clarke.

While there is something of a focus on Wood Farmthe Iron Age, we're not biased and have already found that our investigations can add to the understanding of longer term landscape use and evolution in the area, with Neolithic right through to Early Medieval features and sequences discovered so far. The starting point has naturally been a huge desk-based assessment, incorporating Historic Environment Records, Portable Antiquity Scheme data, geological, topographic and OS mapping data into a project-wide geographical information system. Aerial photographs, both historic and modern, and lidar imagery are being systematically examined for the study area by Krysia Truscoe. So far, half of the area has been covered and archaeological monuments have been mapped for all periods right through to the end of the Second World War. Already this exercise has identified a number of poorly understood or previously unknown sites, features and areas of interest worthy of further investigation.

We have approached a selection of these sites using fluxgate gradiometer, resistance and ground penetrating radar geophysical surveys (with 75ha of coverage achieved so far and plenty more planned), earthwork surveys and coring exercises. A small number of the really interesting or enigmatic ones will be investigated through excavation. We have already completed one season of digging, with trenches opened at Pond Farm Hillfort in the summer of 2015, and are about to embark on coring and excavation of the Silchester Dykes, a series of poorly understood, large-scale bank and ditch monuments.

Who are we?

A number of other specialists are called on when required but the core team and their areas of expertise shown below reflect the wide ranging nature of the investigations. The majority are based at the University of Reading but a substantial contribution of aerial, earthwork and geophysical expertise has been made by Historic England, for which we are very grateful. The opportunity to work in close partnership with them and in consultation with the West Berkshire and Hampshire archaeological services has greatly benefitted the project and ensures that we are engaging with wider matters such as local and national planning policy, designation and feeding knowledge across the sectors.

Professor Mike Fulford

Univ. of Reading

Principal Investigator

Dr Catherine Barnett

Univ. of Reading

Research manager,

geoarchaeology, dating, archaeobotany

Dr Emma Durham

Univ. of Reading

Finds manager

Nick Pankhurst

Univ. of Reading

Project officer

Dan Wheeler

Univ. of Reading

Project supervisor, graphics

Rory Williams-Burrell

Univ. of Reading

Project assistant

Jenni Eaton

Univ. of Reading

Project assistant

Dr Rob Fry

Univ. of Reading

Geophysics, GIS

Dave Thornley

Univ. of Reading


Drs Neil and Paul Linford

Historic England


Krystyna Truscoe

Univ. of Reading

Aerial interpretation

Helen Winton

Historic England

Aerial interpretation

Dave Field

Univ. of Reading

Earthwork reconnaissance

Mark Bowden and Olaf Bayer

Historic England

Earthwork survey

Kevin Williams

Quest, Univ. of Reading


Dr Nathalie Marini

Quest, Univ. of Reading


Dr Lisa Lodwick

Univ. of Reading

Archaeobotany (plant macrofossils)

Dr Rob Batchelor

Quest, Univ. of Reading


Dr Rowena Banerjea

Univ. of Reading


Dr Sam Cook

Univ. of Leeds


Dr Claire Ingrem


Animal bones

Dr Jane Timby

Univ. of Reading


Sarah Lambert-Gates

Univ. of Reading


The Silchester Environs Project Team


An Emerging Iron Age Landscape: the Silchester Environs Project 2016

Dr Catherine Barnett, Research Manager Environs

2016 has been a busy year for the Silchester Environs team led by Professor Michael Fulford. We tackled several sites at once during April and May, with coring and excavation of three sections of the so called Silchester Dykes. These are a series of large linear bank and ditch monuments which lead from the Little London and Mortimer areas towards the Iron Age oppidum and overlying Roman town of Silchester. They have not been dated before and were poorly understood, so we set out with the aim of finding just enough organic material to provide a chronology for construction through radiocarbon dating. We certainly achieved that, with two sites proving to be of later middle Iron Age date but we were also delighted to find the third, at Wood Farm near Three Ashes, also had associated settlement debris. A deep section across the monument has revealed the monument was constructed during the Late Iron Age in a well-used agricultural landscape prone to erosion and hillwash with settlement in the immediate area during the early, middle and late Iron Age. The fieldwork team led by Nick Pankhurst and Dan Wheeler recovered samples and a number of artefacts such as bead rim pot sherds and a spindle whorl made of non-local material (Kimmeridge shale) which were promptly washed clean by the April monsoon rains.

Not ones to let the dust (or rather sticky London Clay) settle, the team next headed to Windabout Copse near Mortimer in August to September in order to assess a series of cropmarks and low earthworks identified by our aerial interpreter Krysia Truscoe during systematic interpretation of old aerial photographs and recent lidar (airborne radar) data over an area of 140km2 around Silchester. A few were believed to relate to later features including the parish boundary, but the shapes of some suggested a late prehistoric presence too.

A series of small evaluation trenches were dug and have revealed remains of exceptional quality and importance. A square anomaly within a D-shaped enclosure to the north of the complex proved to be a chambered cremation-burial. Burnt bone from the individual lay in the base of a square-cut pit about two metres square and one metre deep. Gullies around its edges probably supported the timber surround, long since rotted away. Eight pottery vessels, six platters and two drinking cups, all originally complete, but now fragmented, were found around the cremated remains, some upside down. These presumably originally contained food and drink for the deceased. Four copper-alloy rings found in the fill may have been fittings from a box, or fastenings from a leather bag, perhaps one which originally contained the cremated bone. Covering the grave were the remains of several charred planks of oak, perhaps part of the roof of the chamber. Four of the pottery vessels in the grave were imported from northern France, the remainder were locally made and all date to the early 1st century AD. Close by the grave were large fragments of an imported Spanish wine amphora of similar date, perhaps part of an offering to the deceased. Together the grave and its contents suggest an individual of high status with close connections with the Iron Age centre of Calleva.

This is the first burial of its kind found in England south of the Thames. Other, similar burials have been found to the north and east near St Albans and Colchester, but this tradition of chambered burial begins in the Iron Age of northern France. Over the winter the pottery vessels will be conserved and further research will be undertaken on the contents of the grave, including the cremated remains. We hope that it will be possible to at least establish the sex and age of the deceased.

The grave and surrounding ditched-enclosure lie in a prominent position on the brow of a hill, and would have been highly visible from the settlement enclosure identified down the hill to the south. A series of ditches there define an enclosed area with some internal partitioning. Preliminary assessment of the artefacts from the ditches indicates initial use in the Early Iron Age, perhaps from about 800BC, with re-use and re-cutting in the Middle and Late Iron Age periods. A number of features were identified in the small portion of the enclosure which was excavated. These include a substantial timber-post entranceway with deep pits either side, beam slots with a series of postholes, which define part of a building, and a small, clay-lined oven. Analysis of samples and artefacts recovered during the excavation will continue over the next year and should confirm the dating and tell us more about this exciting new site.

Although clearly of great interest in its own right, perhaps the true importance of this discovery is that it lies in an area previously perceived as being devoid of late prehistoric activity (despite being close to the oppidum / Late Iron Age Town which underlies the Roman town of Calleva Atrebatum) and that it has been found through the systematic application of non-intrusive prospective techniques. This demonstrates the potential of the wider surroundings to tell us far more about late prehistoric life and landscape in the area than we know at present.

The team would like to express their sincere thanks to Mr and Mrs Oli Lambert, the Englefield Estate, the Benyon Family and the Hodge Family for their interest and kindness in providing access to Wood Farm and Windabout.

Excavation and Coring of Pond Farm Hillfort 2015

The first case study site investigated within the Environs project using intrusive methods has been Pond Farm univallate hillfort, also known as The Frith 1.2km north west of Calleva within Benyon's Inclosure, Mortimer West End, North Hampshire, at SU 62678 63078. Its close proximity to the oppidum and the likelihood of it having been constructed during the Iron Age was of interest, in particular whether it preceded the oppidum or was in use at the same time and what was the nature of its use. The site had not been excavated or dated previously and so, guided by the results of preliminary earthwork survey of the surviving standing parts of the monument, and geophysical survey on the levelled portion, and with the kind permission of the Englefield Estate and Dr Richard Massey, four 20x20 metre trenches were opened by a team of staff, students and volunteers under the supervision of our project officer Nick Pankhurst. Trench 1 was placed over a ploughed out stretch of bank and ditch to the southeast, Trench 2 over the known causewayed entrance and extant bank and ditch termini, Trenches 3 and 4 in the interior of the monument. In addition; more inaccessible or deep areas of the monument such as the extant bank and ditch and on the slope and floodplain below the hilltop were drilled using a Cobra TT powered auger, allowing 21 sediment sequences of up to 5 metres depth to be recovered as cores.

Post-excavation analysis is ongoing but highlights of the findings so far include the following:

Neolithic to Bronze Age Activity and Clearance for Agriculture

Evidence for the earliest local activity comes from a series of cores taken at the base of hill just above the floodplain to the east of the hillfort. Geoarchaeological description has revealed an old soil buried under a substantial layer of colluvium eroded from the upper slopes. Alder twig wood from that soil has been radiocarbon dated to the Later Neolithic at 2890-2660 cal BC (4179+/-26 BP, SUERC-65361). The deposits indicate that following relative landscape stability under wooded conditions during early prehistory, deforestation and agriculture commenced at the end of the Neolithic or Beaker periods, allowing soil erosion and hillwash down slope during heavy rain events. Analysis and dating of the charred contents of treethrows which stratigraphically underlie the monument higher up the slope is planned.

Construction and First Use of the Fortified Monument in the Late Iron Age

It appears therefore that the site was already at least partially cleared and in use by the Iron Age but the construction of the hillfort ditch and rampart would still have been a major undertaking in terms of time and labour. The defences are c.575m long and encircle an area of 2.1ha hectares. Three slots placed through the ditch in Trench 2 showed the cut to be up c.6m wide and c.2m in depth. The rampart still rises some 2.5m over the ditch today, coring indicates that local gravel was used to build it up in layers, with each dump compacted before laying the next. Potentially also topped with a fence, the ditch and rampart together would have formed a substantial boundary. Outside the main entrance is a second, smaller length of bank and ditch, clearly shown on the earthwork survey, which seems to have formed a further complex element of the entrance, restricting the size and direction of approach and further adding to the impressive nature of the monument.

Although the causewayed entranceway to the hillfort was over 12m wide, access into and out of the monument would likely have been controlled and restricted by a gatehouse or similar wooden structure. Evidence for this was seen beneath the rampart in Trench 1, where a line of dark circles proved to be the remnants of large timbers that formed a palisade across the entrance-way. Charcoal within the foundation cut of the palisade has been radiocarbon dated to the (late) Middle-to-Late Iron Age date at 200-30 cal BC (2083 +/-29 BP, SUERC-65355); a date supported by a piece of Late Iron Age pedestal beaker within the contained slumps. This piece was one of only a very small number of pottery fragments found within any of the four trenches.

A small concentration of Silchester ware pottery was found within the upper fill of an earlier tree throw hollow close to the entranceway. Birch charcoal from this same fill has been radiocarbon dated to 240-410 cal AD (1710+/-29 BP, SUERC-65356), this Late Roman result shows that there was continued activity at Pond Farm during the period of Roman occupied Calleva.

Overall, the scarcity of material culture across the whole site is quite remarkable given the size of monument, with little evidence of any substantial occupation or domestic activity. To the delight of the environmental archaeologists in our team, the main evidence of use comes therefore in the form of plant remains, particularly charcoal. If the palisade was constructed, as we believe, at the same time as the fortification of the defences, the monument would be contemporary with, rather than a precursor to, the oppidum at Calleva, with local domestic activity focussed there. Also the earthwork survey by Mark Bowden and team indicates that the defences were never entirely completed. We suspect instead that whatever the original plan during construction, the monument was used predominantly for livestock management, used repeatedly seasonally over a long time to graze and protect these valuable assets, with temporary occupation and use of small shelters by the accompanying herders rather than any substantial long term settlement. The huge defensive earthworks were there to protect valuable livestock, quite at odds with the traditional view of a hillfort as a settled site heavily defended for its human inhabitants.

Early Medieval Reuse of the Monument

Evidence of a phase of substantial reuse of the monument was found in Trench 2, where a substantial dump of wood charcoal within a probable recut of the ditch significant re-cutting of the defences and clearance of secondary re-established tree cover to enable re-use of the monument. Identification of the tree and shrub types represented and further dating of the deposit is planned, but a single radiocarbon date on elm twig wood at 1.72mbg/ 90.18mOD is of 610-680 cal AD (1377+/-29 BP, SUERC-65360), within the Early Medieval/ Early-to-Mid Saxon period. The lack of any associated material culture from this time indicates that it was again probably not used as a settlement but rather for seasonal grazing and agriculture.

Post Medieval Period

Two curvilinear ditches in the north of the trench 2 pointed to further reuse of the area in the post-Medieval period. Both ditches contained a significant number of pieces of peg tile and were potentially contemporary with the use of the tile works near Kiln Pond to the northeast. The ditches respected each other and the longer of the two narrowed as it approached the enclosure ditch, signifying that it initially ran up and over an extant part of the rampart that has subsequently been destroyed. The position of both ditches may imply that they also reused part of the surviving enclosure ditch as their southern extent. The enclosure ditch at this point may have still been open and the addition of C-shaped (in plan) ditches could have created a series of agricultural plot or livestock enclosures. The find of a piece of inscribed roof tile in the top of the ditch displaying the date 1734 is an unusually helpful one in picking out more recent use of the site!

On-Going Post-Excavation Analysis

Due to the acidic nature of the soils at the site, no bones or mollusca have been preserved at Pond Farm, apart from rare fragments of calcined bone. However, plant remains are plentiful and analysis is underway. This includes identification of the pollen from Iron Age and Early Medieval layers preserved in the ditches and in the deep, waterlogged sequences in the floodplain to the east which will inform us on the wider landscape and whether there was any arable cultivation close by. Identification of the substantial wood-charcoal assemblages and any associated plant macros from features of all ages at the hillfort will provide information on the nature of the local landscape, how it was cleared, exploited and managed by people over the millennia and will provide further material for radiocarbon dating to further refine the site interpretations presented.

Future Plans

The results of our earthwork and geophysical surveys, coring and excavation of 8% of the Pond Farm monument (excavation of 1747m2 of the total monument area of c2.1ha), coupled with these detailed post-excavation analyses have provided a better understanding of the construction and use of the hillfort and its chronological relationship to other nearby foci of human activity. The research aims we originally set have been fulfilled and there is no need to disturb the monument with further holes in the ground. Instead we are moving on this year to examine features possibly representing late prehistoric land division and agricultural layout, including the Silchester Dykes and a series of linear cropmarks and enclosures to the north east of Calleva, near Mortimer, we'll keep you posted on progress.

Further Updates

on the project can be found on the Silchester Facebook or Twitter. The Silchester Environs project findings, including the case study sites will be published as a monograph, due out 2019.


Our thanks to the Calleva Foundation and also the Englefield Estate and Dr Richard Massey who were generous with their help, time and access to Pond Farm and to all the staff, students and volunteers who dug with us or who have helped with post-excavation processing and sorting.

Things to do now

Follow us

Contact us



+44 (0)118 378 6255
+44 (0)118 378 6762

+44 (0)118 378 6718

Media Enquiries:

Please see Information for the press page

Page navigation


Search Form

A-Z lists