Monument record MDO18376 - Poundbury Early Roman Structure R21

Early Roman structure R21 was discovered during excavations on site E at Poundbury, Dorchester, lying to the north of contemporary structure R18. It comprised a large terrace with drystone walls along the south and west sides and a large number of slots, postholes and small pits cut into it. It is difficult to reconstruct any structure on this terrace, but a single central rectangular structure of slot and post construction, with stone retaining walls to the terrace appears to be the most likely interpretation. Associated features include a clay-lined hollow, a series of pits and a large oval scoop to the east. The terrace seals a sherd of 2nd century pottery and some 1st and 2nd century finds were associated with the features on the terrace.

Found during the excavations directed by Christopher Sparey Green, for the Dorchester Excavation Committee, between 1966 and 1979 during development of the Grove Trading Estate on the eastern slopes of Poundbury Camp, Dorchester (1). The numbers in square brackets below refer to the context and feature numbers used in the published report and archive (1) (2).

Early Roman Structure R21 was terraced into the hillslope to the east of Poundbury Camp, on Site E. It was constructed in the same area as the earlier Late Iron Age settlement and lay just to the north of Structure R18. It comprised a large terrace, upon which traces of a building or buildings were identified. The terrace had a chalk rubble surface where it overlay Iron Age pits and gullies, and also a pit [E550], which contained a sherd of 2nd century Samian pottery. The terrace had a dry stone flint wall on the west and south sides, which may be the remains of a retaining wall, but could also form part of two separate structures. A number of shallow linear features, slots, gullies and postholes were found on the terrace, but it is difficult to reconstruct the original building plan from these fragmentary traces, however Christopher Sparey Green interprets it as having at least two elements along the south and west sides, possibly joined to a central structure. The dry stone footing [E855/869] & [E870], and parallel slots [E986] & [E594] besides some post-holes, formed the western structure. The southern structure consisted of dry stone footing [E 855], its associated post-holes and traces of a parallel linear feature 4m to the north. To the NE features [E622/634/654] and neighbouring post holes might belong either to a separate northern building or to a central block linked to the southern and western structures. This interpretation has flaws; the proposed north and south walls of the southern structure are not parallel, and several different construction techniques are interpreted as belonging to the same building. An alternative interpretation might be that the dry stone walls represent terrace revetments or plot boundaries, with a single central post and slot built structure.

At the northern end of the terrace was a small shallow oval clay-lined scoop [E604] and a series of small pits and scoops were found to the east and northeast, including one that contained larger scoop were found to the east. A large scoop To the north east, [E707] was a large oval scoop, perhaps 4.88m in diameter and up to 0.3m deep with a relatively smooth natural chalk surface. The fine grey-brown soil fill contained a late 1st or early 2nd century brooch.

Early Roman building R21 was probably contemporary with the adjacent structure R18 and oven [E181] and formed part of a small early Roman unenclosed settlement, on the site of the earlier Late Iron Age settlement.

<1> Sparey Green, C, 1987, Excavations at Poundbury, Dorchester, Dorset 1966-1982. Volume 1: The Settlements (Monograph). SDO9630.

<2> Sparey Green, C, 1966-1979, Poundbury, Grove Trading Estate, Dorchester (Excavation archive). SDO10096.

Record last edited

Jul 26 2010 6:56AM