EDO4532 - A37 Western Link Road, Bradford Peverell; excavation 1987 to 1988

Wessex Archaeology undertook a programme of archaeological fieldwork along the line of the Dorchester By-pass Western Link road in Fordington Bottom. In autumn 1987 seven machine excavated trial trenches, were examined at the southern end of Fordington Bottom (A, B, C, D, E, F, G/H). The trial trenches on the coombe side and floor (E-H) revealed an impressive and well-preserved sequence of Roman settlement features and colluvium. Consequently an area of 0.3 ha was machine stripped, extending trial trenches E, F, G, and H. Excavation was undertaken between October 1987 and January 1988. In February 1988 access for archaeological works was granted for the northern part of Fordington Bottom. Seven trenches ( K, L, M, N, P, Q, & R) were excavated between February and March 1988. Trench K was located to determine the northern extent of the Romano-British settlement excavated in 1987- January 1988 at the southern end of the western link road. The remaining trenches were located to assess the extent of the colluvium [2]. The settlement in the valley floor has been preserved by a layer of colluvium and the lynchets on the slopes have been sealed by soil bulldozed in the 1950's. The primary objective was to record landscape development through the study of environmental indicators such as land snails, which will enhance excavtions taking place in Dorchester and Maiden Castle. The earliest physical remains were those of the field system in the form of ditches and banks defining a series of small square fields. Part of the system remained in use into the Roman period and part of the spur to the west of the combe was ploughed in medieval times, the rest remained under permanent pasture from the abandonment of the settlement to the 1950's. This pasture episdode is represented by a dark, stone free soil containing post-med pottery in its upper levels. The majority of the features were late Iron Age or Romano-British and sealed by the soil mentioned above. A sunken trackway divides this settlement into two parts. To the east, on the upper slopes of the valley is an area of settlement features; pits, post-holes and gullies cut into the chalk bedrock and sealed by up to 1.5m of soil. Amongst these features are a number of Roman and possibly pre-Roman graves. To the west of the trackway, in the bottom of the valley, are a number of well preserved structures. The earliest seems to be a gully beneath a layer of flints. A possibly Bronze Age V-shaped ditch cuts through this flint layer. A number of burials, including some crouched burials, were recorded in the upper levels of the ditch and adjacent area containing no grave goods but with hob and coffin nails. Halfway up the eastern side of the coombe, a massive lynchet is probably part of the Roman field system. Just below and parallel to this lynchet is a wall of bonded flint nodules, possibly a revetment to protect an area of Roman industrial activity from hill wash. A number of pits and hollows were recorded in the valley floor including a rectangular pit lined with opus signinum and surrounded by a flint wall, clearly designed to be waterproof and possibly a cistern. Adjacent to this pit was a substantial grain drying oven.