CPAT logo
Cymraeg / English
Back Home
Projects index
Download the project report for 2012-13

The Walton Basin Project 2012 - 2015


The Walton Basin, photo 04-C-0195 © CPAT

Recent Cadw-funded project work in the Walton Basin, as part of the Prehistoric Funerary and Ritual Sites and the Roman Military Vici initiatives, combined with the results of earlier assessment work by the Trust under the direction of Dr Alex Gibson and also funded by Cadw, continues to highlight the importance and complexity of this area of eastern Radnorshire, which is virtually unparalleled elsewhere within the British Isles. The area encapsulates the archaeology of the Welsh borderland and is so far known to contain evidence for multiperiod activity from the early post-glacial period onwards.

Recent studies have focused on the complex of prehistoric monuments around Hindwell and Walton, most of which date from the Neolithic, and include some of the largest sites of their type in Britain, such as the Hindwell cursus and the Hindwell palisaded enclosure. The importance of the area as a base for Roman military campaigns is also becoming more apparent and the strategic significance of the routeway which passes through the basin into mid Wales remained influential well into the medieval period.

The complex double ring ditch lying inside the Hindwell Neolithic palisaded enclosure, photo 96-C-0057 © CPAT

It has been apparent for some time that the archaeology of the Walton Basin is under varying degrees of threat from continued ploughing in this highly productive agricultural area, as well as from piecemeal development, both of which are having a very real impact upon the archaeological resource. Few archaeological sites retain any upstanding element and are generally known only from cropmark evidence, which has raised a number of issues regarding the future management of the nationally important complex.

The present study has therefore been developed in order to address some of the known issues relating to the management of this extensive and complex archaeological landscape. Work during 2012-13 saw the completion of an initial phase of assessment, which included recommendations for further stages of investigation, some of which will form part of a second phase during 2013-14.

The Four Stones stone circle in 1996 - an island of uncultivated land in a field of potatoes, photo CS96-29-0033, © CPAT

The initial assessment has made full use of digital resources, such as vertical aerial photography and LiDAR, in order to produce detailed mapping of known sites, as well as examining the micro-topography and its relationship with the siting of monuments and also the vulnerability of those sites to agricultural practices.

A land-use study for the basin had previously been conducted in 1992 and new data has been added using aerial photography from 2006 and 2009, the results being used to develop a predictive model for the potential risks to archaeology posed by the apparent agricultural regimes.

Through the Trustís work in this area over the past twenty years the level of public awareness of the archaeology has seen a significant increase. It has been recognised for some time that the involvement of the community in the heritage of their area is of critical importance in promoting both a sense of belonging and a wider awareness and responsibility for the archaeology and this theme forms a key element of the current project.

The project continued in 2013-15, undertaking a number of small-scale excavations to collect a dataset based on the ACRE project methodology and test its application under field conditions. The project has also further strengthened links with the local community, providing an opportunity for volunteers to participate and gain a greater understanding of archaeological techniques and the range of monuments in the Walton Basin.

Hindwell Farm barrow II

The barrow is one of the few upstanding prehistoric monuments within the basin and lies within the Neolithic cursus and palisaded enclosure. The excavation consisted of a single machine-excavated trench measuring 25m by 3m which extended eastwards from the centre of the mound

Interpretation of the Hindwell Farm barrow II

The results demonstrated that the primary mound was constructed of turf and measured around 13.5m north/south by 12.5m east/west, covering a central pit around 2m across. Geophysical survey had previously identified a series of elongated anomalies which appear to be associated with a narrow slot surrounding the turf mound which suggest the presence of some form of revetment, although this was not readily apparent from an examination of the moundís surface alone.

The barrow was later enlarged with the excavation of a 2.5m-wide ditch, the upcast from which was dumped against the turf mound in a sequence of deposits which were later truncated by ploughing. In its later form the monument had an external diameter of around 33.5m to the outer edge of the ditch, the mound being 28.5m across. Cropmark evidence in particular had suggested a second, inner ditch, with a diameter of around 24m, although the excavation produced no evidence to confirm this and it is possible that the cropmark reflects changes in the composition of the barrow. Centuries of ploughing have significantly reduced the height of the mound, spreading the material across the ditch to create an earthwork with an apparent diameter of around 41m.

A series of 12 or 13 areas of burning were identified by the geophysical survey, one of which was investigated in the recent excavation. Although lacking any direct evidence for dating it has been assumed that these represent Roman field ovens, presumably associated with the large marching camp which lies a matter of metres to the south-west of the barrow.



Hindwell trapezoidal enclosure

Geophysical survey in 1998 revealed the trapezoidal enclosure lying immediately outside the palisaded enclosure

Cropmark evidence and geophysical survey had identified a trapezoidal enclosure lying immediately to the west of the large Neolithic Hindwell Palisaded Enclosure. The enclosure was investigated by two machine-excavated trenches, one positioned across its north-eastern side, including part of the palisaded enclosure, while the other crossed the western side of the enclosure

The enclosure measures around 35m across and was defined by a single ditch, with a possible entrance on the western side. The excavation demonstrated that the ditch is fairly small, measuring between 1.2m and 1.5m in width and 0.48m to 0.65m deep, although it had clearly been truncated by ploughing.

It had been postulated that the enclosure might be contemporary with the adjacent Neolithic palisaded enclosure owing to the way in which the north-eastern side of the enclosure appeared to respect the line of the palisade. Although there was no strategraphic link between the two monuments the presence of late Iron Age or Romano-British pottery in the lower fill of the ditch suggests that the enclosure is considerably later. The respective positions of the monuments does, however, suggest that the line of the palisade may well have remained visible at the time the trapezoidal enclosure was constructed and that there was a deliberate intention to place the enclosure outside the palisade. The dating evidence is also significant when considering the potential relationship between the enclosure and the large Roman marching camp. The enclosure lies within the camp, towards its southern corner, and it is not unreasonable to assume that the surrounding bank and ditch may have been deliberately slighted at the time the camp was constructed, presumably during the early campaigns into Wales between AD 48 and AD 60/61.



Hindwell cursus

Plot of cropmark evidence for the terminal of the Hindwell Cursus, showing the location of the 2013 excavations

The north-east terminal of the Hindwell Cursus provided an opportunity to test the hypothesis that ploughsoils were likely to have accumulated towards the base of a slope, affording greater protection to buried archaeological deposits, as well as confirming the position of the cursus ditch. The results from the excavation were somewhat unexpected, however, since the ploughsoil proved to be more shallow at the base of the slope. One reason for this could be the presence of a 5m-wide lynchet against the nearby boundary which may have affected the depth of ploughing noted in trench B. In the other two trenches the topsoil was seen to be around 0.1m deeper. The opportunity was also taken to investigate a second monument, known only from cropmark evidence, which had been postulated as a second, albeit much smaller cursus. Both ditches were relatively small with V-shaped profiles which produced no evidence for dating. The size and form of the ditches does, however, indicate that rather than being part of a second cursus they are perhaps associated with a field system predating the existing boundaries.

Following on from the investigations in the area of the north-eastern terminal of the cursus, in 2013-14 a further programme of fieldwalking was conducted with local volunteers. Over the two years a total of 109 lithics, of which 12 were tools, were recovered. The material dated from the Mesolithic to the Bronze Age and its distribution appears to indicate that the presence of the cursus had some effect on deposition; the density of lithics within the cursus was almost twice that of the surrounding area.

An additional 65m of the cursus was also confirmed through geophysics and trial excavation in the area of a possible marching camp near Hindwell Ash. The cursus ditches were each around 4m wide and spaced 68m apart.



Hindwell palisaded enclosures

Small-scale excavations were undertaken in December 2013 to investigate the relationship between two Neolithic palisaded enclosures and prominent palaeochannels to the south of Hindwell Farm.

The large palisaded enclosure at Hindwell was originally identified from cropmark evidence in 1994 and then through further aerial reconnaissance in subsequent years. Trial excavations were conducted in 1995 and an extensive programme of detailed geophysical and ground survey followed in 1998. The investigations demonstrated that the enclosure was defined by intercutting post-pits, each of which would have held an oak post around 0.8m in diameter. The posts had been charred and radiocarbon dates indicated a construction date of around 2700 BC. The palisade enclosed an area of around 34ha, and it was estimated that some 1400 mature oak trees had been used in its construction. To date, this is the largest Neolithic enclosure in Britain.

An excavation in 2013 produced disappointing results since it was not possible to determine the relationship between the palisade and a prominent palaeochannel owing to the point of intersection coinciding with a large Roman ditch. This had been identified in previous excavations in 2011 and is thought to define an annexe on the southern side of the Hindwell Roman fort. Despite this, however, it is not unreasonable to assume that the palisade stopped at the edge of the palaeochannel, although whether this was an active river course in the Neolithic or a post-glacial landform remains to be confirmed. Evidence from previous geophysical surveys clearly shows the line of the palisade and the Roman ditch to the west of the channel, while the Roman ditch alone reappears to the east. The enclosure encompasses the Hindwell Pool, sited on a natural spring which flows throughout the year and forms the source of the Hindwell Brook, while the apparent incorporation of the palaeochannel into the circuit further strengthens an association with water, a feature which is also common to other Neolithic enclosures.

photo 3745-0025 © CPAT


Excavation of the double-palisaded enclosure in 2013, showing the proximity of the inner palisade to the present course of the Summergil Brook

More recent investigations in 2015 have concentrated on the southern side of the enclosure to shed further light on a number of thermo-remanent anomalies which were interpreted from the 1998 geophysical survey as reflecting the in-situ burning of posts. Further geophysical survey in 2015 identified four complete anomalies, each measuring between 2.5m and 5.0m across, with separations of 3-4m centre to centre, together with a further two which extended beyond the survey area. The geophysics was followed by the excavation of an area measuring 10m by 5m, which contained one of the geophysical anomalies and part of a second. The excavations were sadly too restricted to provide a definitive answer, although it is now clear that at this point the pits are positioned immediately inside the palisade and contain significant quantities of burnt material in their fills, accounting for the geophysical responses, although without any sign of in-situ burning. The presence of Grooved Ware and worked flints within the fill of the pit suggest a broad contemporeneity with the palisade itself and mirrors the presence of Grooved Ware recovered from the post-pipe of one of the posts in the inner circuit of the Hindwell Double-palisaded Enclosure (Jones 2012). Further to the east, however, the geophysical survey suggests that the features may overlie the palisade, implying that they post-date the enclosure.



Geophysics in 1998 and 2010, together with cropmark evidence, identified two sets of curving ditches to the east of the Hindwell Roman fort. One double-ditched circuit could be traced beneath the fort and this was investigated to the south of the fort in 2011. The results revealed that this was a Neolithic double-palisaded enclosure constructed of closely spaced oak timbers, about 0.25m across, set within a broad foundation trench. Charcoal from one of the timbers has been dated to 2866Ė2574 cal. BC. Further investigations were undertaken in 2012, investigating a partial section through the inner palisade and also confirming its position where it had been cut be a canalized section of the Hindwell Brook. There was evidence to suggest that the posts had been burnt in situ and samples of charcoal from the posts, and a fragment of burnt bone from the post-pipe, provided four radiocarbon dates with ranges between 2624-2297 cal. BC.

The recent excavations have added further to our understanding of the double-palisaded enclosure, extending its inner circuit by around 35m. More importantly, however, the palisade is now known to extend to the south-west of the Hindwell Brook. The excavations have demonstrated that at least part of the inner circuit runs parallel to the Summergil Brook and the angle at which the western side approaches the watercourse suggests that either there was an abrupt change in direction, or the brook was incorporated into the layout.



Hindwell Ash possible marching camp

A geophysical survey undertaken in 1998 (Gibson 1999) revealed traces of what has been interpreted as a possible Roman marching camp, lying within the interior of the Hindwell Palisaded Enclosure, around 200m to the north-west of the Roman fort at Hindwell. Two narrow, linear ditches were seen as defining parts of the southern and western sides, while the survey provided no clear definition of the south-west corner. The recent survey was conducted in two adjacent fields to the north of the lane leading to Four Stones, to investigate the potential continuation of the western ditch and prospect for the northern side of the putative marching camp.

The results from the survey, together with a small-scale excavation which examined a section of one of the ditches, have cast doubt on the interpretation of these features as part of a marching camp. Although the putative western ditch of the camp is now known to extend to the north of the land to Four Stones, small-scale excavations in the wake of the geophysics have demonstrated the slight nature of the feature which, at only 0.6m wide and 0.28 deep is perhaps too shallow be associated with a Roman military structure. The excavation of a section of one of three marching camps at Walton in 2009 revealed the ditch to be rather more substantial, at 2.25m wide and 1.4m deep, with a distinctly punic profile (Jones 2010).



Walton Marching Camp Annexe

A series of three adjacent marching camps have been revealed through cropmark evidence in the fields to the north and west of Walton village. Aerial photography has also suggested a possible annexe on the southern side of the Walton III marching camp, although only the western side and south-western corner are currently known.

Although it had been hoped to investigate the continuations of the cropmarks to the north and east, permission was not forthcoming so the survey focused instead on a small field to the east where it was thought that evidence of any internal detail might be forthcoming. Unfortunately, the area proved to have a general spread of ferrous debris, assumed to be derived from a former smithy nearby, such that no archaeological features could be discerned.



Old Radnor field system

While previous survey and excavation in Walton has concentrated on the important complex of prehistoric monuments, as well as limited work on a number of Roman sites, the study was extended during 2014-15 to include an investigation of a series of potentially medieval earthworks occupying the slopes below the village of Old Radnor. The opportunity was again taken to involve a number of volunteers from the local community, to the extent that this element of the project was conducted with outreach as its primary objective. The series of linear earthworks (PRN 122807) had already been roughly mapped from aerial photographs and LiDAR, although it was clear that a ground survey would be required in order to confirm their form, extent and survival, as well as assessing their significance and vulnerability.

photo 3983-0002 © CPAT


Local volunteers plotting the visible earthworks

The earthworks within the survey area consisted of a series of five lynchets (PRN 122807), each aligned east-north-east to west-south-west and with separations of between 25m and 30m. The earthworks of Castle Nimble (PRN 360; SAM Rd 46) lie on the valley floor to the north of the field system and had been surveyed previously, during the late 1990s. It seems likely that the lynchets represent the remains of a former, presumably medieval field system, which may have been associated with Castle Nimble, or possibly Old Radnor itself, the shrunken settlement earthworks (PRN 5296) of which lie 300m to the east-north-east of the field system.


Privacy and cookies