Highways Agency

THE WEST

• A30 HONITON TO EXETER
• A35 TOLPUDDLE TO PUDDLETOWN
• A417/A419 SWINDON TO GLOUCESTER
• A417 BROCKWORTH BYPASS
• A30 FRADDON TO INDIAN QUEENS
• A303 STONEHENGE
• A34 NEWBURY BYPASS
• A421 WENDLEBURY TO BICESTER
• M4/M5 SECOND SEVERN APPROACH ROADS
  

A30 Honiton to Exeter

Horse Power - Roman Style

An exciting new discovery during the improvements between Honiton and Exeter was the site of a fort built by the Roman army in the middle of the first century AD.

The fort at Pomeroy Wood was defended by a rampart made of earth and timber and beyond that were two deep ditches with narrow 'ankle-breakers' (right) in their base.

Part of a building which may be a barrack was also excavated. This building was later replaced by a watch tower which strengthened the defences of the fort. Finds from the fort

included spears and high-quality crockery imported from France. Plants and beetles preserved by water logging in a well have allowed archaeologists to suggest that one of the garrisons of the fort was probably a cavalry unit, which helped patrol the main Roman road between Dorchester and Exeter. The modern A30 is thought to follow the route of this road.

After the Roman garrisons left to campaign further north, the settlement which had grown up around the fort continued in use. However, instead of building Roman-style houses, traditional houses were built, which were circular rather than rectangular in shape.

Part of a small Roman cavalry fort at Pomeroy Wood

As well as these Roman remains a series of very important prehistoric sites were also recorded by Wessex Archaeology in advance of the road building. The oldest of these sites was at Castle Hill near Fenny Bridges, where there were two ritual monuments of Neolithic date.

Archaeologists think that these monuments were used in ceremonies to commemorate the dead. The discovery of highly-decorated Peterborough Ware pottery and radiocarbon dates show that these sites are about 5,000 years old.

Other evidence for settlement at this time was found at Long Range to the west, where Neolithic pottery and a fine flint knife were found.

One and half millennia later Bronze Age farmers laid out a field system across the then-disused monuments at Castle Hill. The farm compound or farmyard lay at one end of the field system and probably contained roundhouses and granaries. Some of the crops the Bronze Age farmers grew in their fields have been preserved as they were burnt accidentally on fires in the farmyard.

Just north of Patteson's Cross two more Bronze Age roundhouse were found. One of these also lay in a small compound, but the other seems to have stood on its own out in the fields and may have been a shepherd's hut. Another Bronze Age farm, this time about 3,000 years old, was excavated near to Honiton at Hayne Lane. This settlement also lay within a compound which had a ditch and bank around it. Inside were two round houses. One house of these was the main dwelling, the other seems to have been used for crafts. A number of granaries were raised above ground on legs, rather like modern chicken coops, to keep vermin away from the grain.

Three settlements dating to the Iron Age were also examined. A single roundhouse was built at Langland Lane, close to the by then long-abandoned Bronze Age settlement at Patteson's Cross. In contrast, several round houses were found at Long Range near Straightway Head.

These houses are 2,500 years old and like most of the Iron Age houses are much larger than the earlier Bronze Age ones. However, many things, such as the above-ground granaries and the crops which were grown, would have been familiar to the Bronze Age farmers.

As well as these excavations, the finds made during the watching brief while the 20 km-long road was being built also showed that the valley of the River Otter close to Honiton was the most densely-settled area.

However, one significant site was found near to Exeter, the Iron Age farm opposite the Blackhorse Inn, itself recently renamed Le Cheval Noir.

The Iron Age farm was occupied for several centuries. It consisted of a number of round houses and above-ground granaries which were gradually rebuilt in slightly different places. Only towards the end of the Iron Age, about 2,000 years ago, was the farm enclosed by a very wide and deep ditch with a bank. Inside this square yard stood a single, very large, round house and a number of above-ground granaries.

Highly-decorated pottery called Glastonbury Ware and a number of quern stones which the farmers had used for grinding grain were found. It is possible that some of the children of these farmers saw the arrival of the conquering Roman soldiers who built the fort at Pomeroy Wood.

There have been few opportunities to explore the prehistoric and Roman landscape of east Devon but the Honiton to Exeter improvements have, for the first time, allowed archaeologists an exciting insight into just how much evidence is likely to survive.

back to top

A35 Tolpuddle to Puddletown

A Grave at the Crossroads

A vast amount of information came from a site on the Tolpuddle to Puddletown bypass, near Tolpuddle Ball, first excavated by Liverpool University in 1993. A smaller excavation in 1996/7 and a watching brief by Wessex Archaeology added finds dating from the Neolithic to the Romano-British periods.

The road scheme provided an opportunity to build up a picture of the farming economy in a nine-kilometre area which was well-settled in the late prehistoric and Roman periods.

The earliest discoveries were of late-glacial and early post-glacial deposits at Burleston Down (c. 12000-8000BC) thought to be the first such record in central southern England.

The project also provided important new data concerning Neolithic and Bronze Age activity, including the identification of an Early/Middle Bronze Age settlement enclosure at Tolpuddle Ball.

The most extensive period of activity dates from the Middle/Late Iron Age to the Early Romano-British period (4th century BC to 1st century AD). The Middle/Late Iron Age settlement focused on a rectangular enclosure established around 300 BC. A large number of pits were found, some containing special deposits including the burial of animal heads and carcasses and the remains of unborn children.

The Roman settlement was occupied up until at least the 4th century AD and probably had substantial stone buildings. There were rare organic deposits, which came from the preparation of leather from sheepskins.

Burials were associated with both settlements and 23 individuals were recorded.

In July 1998 a cemetery of 50 graves was excavated to the west of this settlement near Tolpuddle Ball. This had a well-ordered layout and all the graves were aligned east to west.

None of the graves contained goods or coffin nails although the former presence of timber coffins was visible in a few cases. A few small sherds of Roman pottery were recovered and radiocarbon dating of bones has shown that it was used from the end of the Roman period in AD 400 to AD 700. It demonstrates a continuity of Romano-British culture in Dorset rather than that of the incoming Saxons.

Evidence of prehistoric activity was also found to the north and south of Tolpuddle Ball, and to the north of Tolpuddle village, and localised concentrations of worked flint were noted along the route during the watching brief.

Medieval and later deposits from the project are limited but include an unusual early 13th-century agricultural enclosure and waste tip from West Mead near Bere Regis.

One enigmatic discovery was the isolated grave of a woman, aged over 40 years, lying alongside (or even within) a former road surface near Roger's Hill Farm. Wire loops, probably associated with costume or dress accessories, indicate a date from the 16th to 18th century. The grave was located on the parish boundary on the north side of what was a cross-roads on the Bere Regis to Tolpuddle road. This type of location was traditionally used for the burial of villains, suicides and other social outcasts, such as witches.

back to top

A417/A419 Swindon to Gloucester

Roman Farmers

The new A417 and A419 Swindon to Gloucester roads run for much of their lengths on the Roman Ermin Way, which linked Gloucester (Glevum) and Cirencester (Corinium Dobunnorum), and ran south-east to Silchester (Calleva Attrebatum) near Reading. This is not to be confused with the other Ermine Street which ran northwards from London to York.

The Oxford Archaeological Unit investigated 35 separate sites of varying types and periods along the 25 kilometre length of the new road. Many of the sites produced archaeological finds from the Roman period.

The most important find was near Birdlip Quarry where a farming settlement was discovered alongside Ermin Way. Three wells, a corn-drier and several hearths and ovens were excavated.

Also discovered around the quarry were some hipposandals - Roman horseshoes which were tied to the horses' feet, rather than being nailed. It is thought that they were used to protect injured feet, rather than be used constantly, as experiments have shown that they come off at more than walking pace. They may also have been used for cattle.

The most significant discovery was a sequence of buildings between the second and fourth centuries AD which showed a development from earlier circular timber structures to later stone buildings. The wooden roundhouses, 12 metres in diameter, were built in the traditional way, with the doorways facing south east, away from the Roman road. They seem to have been abandoned some time early in the third century and were replaced later with stone buildings.

Significant information relating to the later prehistoric period (800 BC-.AD 43) was found. A number of important Iron Age settlement enclosures were investigated at Highgate House, Middle Duntisbourne, Duntisbourne Grove, Preston Enclosure and Ermin Farm.

The Preston enclosure was a distinctive hexagonal shape and its dating to the middle Iron Age makes it very unusual for the period. The two rectangular Duntisbourne enclosures, less than one kilometre apart, date to the few decades before or around the time of the Roman conquest and, intriguingly, both appear to have been occupied at the same time.

A number of sites produced evidence for early prehistoric activity. A group of pits from Duntisbourne Grove contained flints and pottery dating from the Neolithic period (3500-2000 BC), while a second group of pits from Trinity Farm produced finds dating from the succeeding Early Bronze Age period (2000-1800 BC).

Two Bronze Age burial barrows, represented by adjacent ring-ditches, were discovered at St Augustine's Farm South, associated with a series of segmented ditches (possible boundary markers) and a number of pits. One of the ditches still contained the remains of a cremation in its centre.

In addition, a possible Bronze Age burial was discovered above the Churn valley at Lynches Trackway.

Many sites produced evidence for medieval and post-medieval activity. In particular, work at Latton Street Farm revealed a series of superimposed stone buildings dating from around the 13th century, which should provide an insight into the history of the development of Latton village.

Post-excavation analysis over a period of three years is under way and a comprehensive report will be published. Radiocarbon dating should help to solve some of the dating difficulties, particularly with the earlier prehistoric finds.

Gravel excavations at Latton at the end of 1998, which were not part of the Highways Agency scheme, uncovered the bones of eight mammoths and flint axes.

back to top

A417 Brockworth Bypass

Plough Furrows

Investigations were made by the archaeology service of Gloucestershire County Council around the 5.6km Brockworth bypass. This included work to the north and east of Hucclecote Roman villa.

Excavation within the eastern corner of the grounds of the Hucclecote Resource Centre was carried out before construction of a new slip-road linking the M5 motorway with the Brockworth bypass. This lay about 150 metres east of Hucclecote villa and was found to contain a rectilinear system of ditches, interpreted as boundaries enclosing previously unsuspected small fields or paddocks. The boundaries developed over several centuries, from the 1st century AD.

The enclosures contained a number of pits, post and stake settings and gullies which may signify the presence of small structures. Overlying the Roman deposits were a number of medieval plough furrows associated with a headland running at a right angle to the ends of the furrows. These were aligned with the Roman boundaries, implying perhaps that the villa landscape influenced the layout of the medieval field system.

North of Hucclecote villa, topsoil stripping in advance of road construction revealed Romano-British boundaries and structures preserved within an area 0.35km wide. Excavations to sample these features were undertaken in four areas. Boundary ditches, a length of wall and a corn dryer were found.

These finds confirm the presence of a complex system of enclosures surrounding Hucclecote villa.

back to top

A30 Fraddon to Indian Queens

Sacred Spaces

Archaeological investigation, initially commissioned by English Heritage, was undertaken between October 1992 and May 1994 by Cornwall Archaeological Unit. This revealed a wide range of information about a tract of Cornish landscape. A post-excavation assessment was made during 1997-1998.

The discovery of a ritual landscape north of Indian Queens was of particular interest. Excavation at Little Gaverigan Barrow (found during a survey in 1991) revealed three major episodes of ritual activity in the Early Bronze Age.

The site was a sacred space with central pavement of quartz rubble defined by a ditch. The ditch later had a number of large pits dug into its northern side, which may have once held stone menhirs - carved standing stones - or wooden posts.

Two small pits in the western area of the enclosed space contained fragments of pottery dating from the Early Bronze Age (about 4,000 years ago). One was an unusual small vessel with a central perforated plate known as an accessory cup, which may have been used to burn incense. Finally, the site was capped by a mound of cut peat turves, which contained flint artefacts from the earlier Mesolithic period (at least 6,000 years ago).

Less than 30 metres to the north, another completely unknown Early Bronze Age ritual enclosure was discovered near the new Highgate roundabout and rapidly excavated during the watching brief. This comprised a horseshoe-shape space defined by ditches and post-holes. In a central pit was a cremation contained within a decorated collared urn, also containing an unusual bronze awl.

A cluster of large pits, recorded some hundred metres to the south, probably represent a further component of this landscape, and the isolated discovery of a cup-marked slate in the locality further contributes to the overall emerging picture.

back to top

Bronze Age Pottery

Trenches were opened across one edge of Penhale Round (an enclosed later prehistoric settlement site) and also beyond its boundary to investigate the landscape setting.

The earliest remains in the vicinity were the boundaries of an early field system, a well-preserved oval building, a circular stake-built enclosure and the traces of a sub-rectangular structure which contained a domestic hearth pit. Embedded in the floor of the oval building was a dense spread of Middle Bronze Age pottery dating to around 1300-900 BC, together with the charred fragments of structural oak timber.

This landscape was contemporary with a farmstead excavated during the watching brief at the western end of the bypass on Penhale Moor. This consisted of two structures and produced a sizeable collection of Trevisker pottery. Worked pieces of stone and flint were also found, including a Middle Bronze Age spearhead embedded in the floor of the main residential building.

Further isolated pits, ditches and spreads in the area gave some indication of the survival of Bronze Age and later prehistoric elements.

At Penhale Round, excavations centred on the ramparts, external ditches and main entranceway. At least five major phases of complex modification to this site were identified. Occupation extended from at least the end of the second or first century BC up to third/fourth centuries AD, during which time the site gradually developed from a single-ditched enclosed settlement into one with double ditches.

After its abandonment, the area seems to have been largely deserted and not later reused for settlement by medieval farmers, although traces of a medieval field wall were found running more or less centrally across the site.

There were also surveys and small-scale investigations of the early tin-working sites at Queens mine, the Kelliers and Mayfield farm, in addition to the relatively modern china clay workings at Higher Fraddon.

A large collection of environmental samples was recovered for analysis from isolated places as well as individual sites and features. Two peat deposits, at Halloon Farm and Kelliers, were sampled. More than 140 hedge boundaries were recorded in detail which showed broad differences between boundary types, reflecting different processes of enclosure, changing land use and local topography. Apart from the major field systems, evidence of a probable buried prehistoric landscape was identified at Halloon Farm, probable medieval cultivation was surveyed at Mayfield farm, and at Black Cross the ditches of a field system of unknown date were identified during the watching brief.

back to top

A303 Stonehenge

Major Improvement

The tunnel planned for Stonehenge will take the existing road away from the stones

The Stonehenge scheme will remove traffic from Britain's greatest prehistoric monument, by putting the road closest to the stones in a 2km cut-and-cover tunnel - part of the improvement scheme between Amesbury and Berwick Down.

The Highways Agency have been investigating the archaeological impact of various routes around the stones since 1991. Their conclusions have been adopted by the custodians of the stones, English Heritage, in their master plan for the site and a special database has been developed which holds all the recorded archaeology within an area of 135 square kilometres.

There are 450 scheduled monuments with the World Heritage Site at Stonehenge. The tunnel construction could affect five scheduled monuments and 11 sites, which are partly or fully within the tunnel assessment area.

All these sites have already been damaged by farmers' ploughs and only three are now visible.

There will still be a major programme of archaeological investigation before the tunnel is constructed which will lead to a better understanding of the history of the site.

back to top

A34 Newbury Bypass

A Globular Urn

The route of the Newbury bypass has been subject to extensive archaeological investigations in close liaison with English Heritage. In addition to desk-top studies this included nearly 400 hand-dug test pits, 400 machine trenches and 50 holes. The area of investigation was about two per cent of the site. Sites identified as being of potential archaeological interest were then given a more detailed survey.

The Highways Agency awarded a contract to York Archaeological Trust to undertake a full excavation to uncover and record finds at a Mesolithic - Neolithic site in the Lambourn Valley. This was the only site identified as being of possible national importance. Excavation work was carried out between July and October 1996.

It was confirmed that the bypass avoided the main areas of archaeological interest. Flint working debris dating from 8500-4000 BC were uncovered. Post-excavation analysis and reporting are being carried out and finds will be given to Newbury Museum.

Archaeological investigations were also undertaken at ten sites of local interest before and during site clearance. These revealed some Romano-British archaeological features, Bronze Age pottery and a complete Middle Bronze Age globular urn.

Wessex Archaeology were on site throughout the construction work to look out for any unexpected remains and to record these features.

Measures were taken to ensure the preservation of both the Mesolithic site near Bagnor and that of a recognised Roman villa at Enborne Road.

Large quantities of medieval pottery and tile, some of which was obviously kiln waste, were recovered from a heavily-disturbed site at Enborne Street. The material was found to be associated with a series of pits, one at least functioning as an oven or kiln, and other shallow linear features, although no structural evidence was encountered. Similar evidence suggesting pottery production was found during topsoil stripping at Wheatlands Lane.

The bypass also skirts the 1643 First Battle of Newbury site but avoids the main area of the battlefield.

back to top

A421 Wendlebury to Bicester

Beaker Folk

The Oxford Archaeological Unit undertook major excavations in the northern settlement of the Roman small town of Alchester, Oxfordshire.

These investigations outside the boundaries of the town produced Neolithic/Bronze Age flintwork, and beaker material (from the Early Bronze Age). This material is a type of pottery in beaker shape made at the end of the Neolithic and start of the Early Bronze age and the people who made are sometimes called beaker folk.

There was also evidence for Middle Iron Age settlement and extensive activity throughout the Roman period and Anglo-Saxon burials.

In the first to early second centuries AD the Roman settlement was characterised principally by ditches on alignments relating to the early Roman Akeman Street. Later, a system of ditched plots developed on each side of a minor lane parallel to and north of the line of Akeman Street. South of the lane the structures were dated to the mid second century.

North of the lane the plots, one entered through a substantial timber gateway, contained later Roman structures of varying plan and construction type, the character of the settlement at this time being largely agricultural.

Settlement continued to the end of the Roman period and probably beyond, and there may have been continuity of agricultural practices into the post-Roman period. A small late Roman cemetery with a post-Roman phase complemented the domestic structures and other features. There were numerous finds of a wide variety of types and materials, amongst which a fragment of a monumental inscription was remarkable and the pottery assemblage was particularly large and significant.

Smaller-scale work elsewhere on the road scheme uncovered a Bronze Age burial, late Iron Age-early Roman settlement and elements of the field systems relating to the Roman town.

M4/M5 Second Severn Approach Roads

The sites to be investigated include a well-preserved Iron Age site at Hallen, a Romano-British site at Northwick and medieval settlement evidence at Redwick. Paeleo-environmental studies and mapping should provide information about the wider landscape.

back to top