EDO4818 - Former County Hospital Site, Dorchester; Block 1 & 3

Excavations were conducted by Wessex Archaeology prior to redevelopment of the former Dorset County Hospital site, adjacent to and south of Somerleigh Court, in the SW corner of the Roman town of Dorchester. The fieldwork took place intermittently between October 2000 and December 2001. The principal archaeological works comprised three separate excavation trenches corresponding to the proposed footprints of new buildings: Blocks 1 & 3 (a single N-S aligned trench along the eastern side of the site; Block 2 (an 'L-shaped' trench in the southern part of the site); and Block 4 (a sub-rectangular trench in the western part of the trench). There were also three evaluation trenches to the west of Block 4 and a watching brief was maintained on a new service trench across the site and two areas of unscheduled ground reduction. A 'strip-and-record' programme also exposed archaeological remains immediately south of Somerleigh Court, which were preserved in situ. The excavations revealed a sequence of structures, associated features and deposits which document the development of this part of the town from the 1st century AD until the early 5th century. Block 1 and 3 formed a long irregular excavation area along the east edge of the site and was excavated in advance of the construction of the east wing of Edgecumbe Manor and nos. 13-26 Chesil Place. It had a maximum width of approximately 11m and was 64m long. Pre-Roman evidence comprised red-brown soil layers sealing chalk and/or patches of flinty clay. Prehistoric struck flint was plentiful. Some finds of mid 1st century date were recovered, including two late Iron Age coins and four Claudian coins, but these could not be closely related to any specific activity in the area of the site and may have come from layers of soil which could have been dumped from elsewhere in the town. The initial urban development of this part of the town appears to start in the late 1st-early 2nd centuries AD. A series of timber buildings, apparently set in large enclosures or yards were found. A number of pits were found in these yard areas. At the south end of the trench two stone ovens were found associated with a post-built structure. Fish bones from the fill of the ovens suggest they may have been used in the preparation of fish sauce. Some, if not all of these timber buildings may have had an industrial use. There was a major change in the later 2nd and 3rd centuries, with the construction of two stone houses built along the western frontage of a N-S street immediately east of the development area. These two properties were separated by a boundary wall. The southern property comprised a suite of masonry buildings, with one room containing the eroded remains of a mosaic floor. A possible trackway running back from the street was found on the northern side of the building. To the south of the building were a series of hard surfaces forming a yard area, with a well. Coin evidence suggests that clearance of the building took place during the late 3rd-4th centuries. The northern property comprised a range of stone-founded buildings exhibiting a complex sequence of repair, remodelling and rebuilding. North of these buildings spreads of silt and redeposited chalk suggested an open yard. The late 3rd and 4th centuries saw occupation shift away from the street frontage and there was very little evidence for occupation in this trench. Only one definite area of 4th century activity was identified; a poorly understood mortared masonry structure with an associated small roughly-paved working area. Numerous mid-4th century coins were recovered from this paved area. A substantial undated stone-built and opus-signinum lined structure was discovered at the northernmost end of the trench. It had a sunken floor suggesting it was not for domestic use but possibly a cistern, conduit or part of a private bath complex.