| Summary : Earthwork remains of a Roman temporary camp. Camp III is a parallelogram of much more elongated form than camp I, enclosing an area of 5.5 ha (13.7 acres). The shape chosen makes better use of the ground available and the camp occupies the whole of the broad crest of this gentle ridge between the March Sike and the Chew Sike. It was necessary, however, to overlap into camp I and, in doing so, the NW defences of camp I were partly levelled for a short distance where these crossed the central axis of camp III. The ground within the SE half of camp III is almost flat but in the NW the land rises steadily to approximately 460 m above OD, only about 20 m below the height of the watershed. Although the outline of the camp is clear, the earthworks are not particularly well preserved on the surface; they survive best on the SW side close to the W angle, where the rampart stands 0.5 m internally and 1.4 m above the bottom of the ditch, which is 0.7 m deep. For most of this SW side the defences are almost overwhelmed by the growth of peat, leaving only the slight outer scarp of the rampart frequently no more than 0.1 m high. Elsewhere on the perimeter, the rampart averages 0.7 m in height externally and 0.4 m internally; the ditch is usually about 0.4 m deep. Round the E angle the scene of much post-Roman activity, the earthworks are reduced to a bank 0.4 m high and the ditch is barely discernable. When sectioned in 1936, just to the SE of the NE junction with camp I, the ditch was found to be 'V shaped, 7ft 6 ins (2.3 m) wide from lip to lip and 3ft (0.9 m) deep'. The rock brash that made up the rampart, 2ft (0.6 m ) high, was held in place by turf kerbs, the larger forward one enabling a steeper face to be built on this side. Again, these measurements do not agree with the published section. At the N angle good use is made of the natural defences provided by a small gully that drain into the Chew Sike. |
More information : This camp has been re-assessed in connection with RCHME's survey and publication of Roman Camps in England. The following descriptiveaccount is taken from the published text.
All elements of Chew Green complex now assigned unique identifiers under parent record NT 70 NE 3. Camp III is a parallelogram of much more elongated form than camp I, enclosing an area of 5.5 ha (13.7 acres). The shape chosen makes better use of the ground available and the camp occupies the whole of the broad crest of this gentle ridge between the March Sike and the Chew Sike. It was necessary, however, to overlap into camp I and, in doing so, the NW defences of camp I were partly levelled for a short distance where these crossed the central axis of camp III. The ground within the SE half of camp III is almost flat but in the NW the land rises steadily to approximately 460 m above OD, only about 20 m below the height of the Watershed.
Although the outline of the camp is clear, the earthworks are not particularly well preserved on the surface; they survive best on the SW side close to the W angle, where the rampart stands 0.5 m internally and 1.4 m above the bottom of the ditch, which is 0.7 m deep. For most of this SW side the defences are almost overwhelmed by the growth of peat, leaving only the slight outer scarp of the rampart frequently no more than 0.1 m high. Elsewhere on the perimeter, the rampart averages 0.7 m in height externally and 0.4 m internally; the ditch is usually about 0.4 m deep. Round the E angle the scene of much post-Roman activity, the earthworks are reduced to a bank 0.4 m high and the ditch is barely discernable. When sectioned in 1936, just to the SE of the NE junction with camp I, the ditch was found to be 'V shaped, 7ft 6 ins (2.3 m) wide from lip to lip and 3ft (0.9 m) deep'. The rock brash that made up the rampart, 2ft (0.6 m ) high, was held in place by turf kerbs, the larger forward one enabling a steeper face to be built on this side. Again, these measurements do not agree with the published section (Richmond and Keeney 1937, 135-7 (1a)). At the N angle good use is made of the natural defences provided by a small gully that drain into the Chew Sike.
At least three gates are visible, the best preserved, despite disturbance by modern drains, being that on the NW. Here the ground rises steadily to the skyline, only about 150 m away; the gate is protected by a traverse, the mound of which is 0.3 m high. The ditch of this traverse is marked only by a change in the vegetation, about 1.5 m wide. At the gate at the SE end of the SW side, in an area almost submerged in peat, the traverse mound is 0.6 m high externally and 0.2 m high internally with, once again, an external ditch about 2 m wide marked by a change in the vegetation. In the centre of the SE side there is a gap for a gate through the bank and a causeway across the ditch; excavation by Richmond and Keeney (1937, 137 (see auth 1a) found no sign of any traverse but the counterscarp bank of the fort (IV) lies only 8 km away, making any such provision unnecessary. Certainly, there seems to have been a gate about 75 m NW of the E angle, where the ditch seems to stop at point opposite to the gate in the SW side. This are, however, is very disturbed; a possible fragment of a traverse mound, 0.6 m high on its SE side,survives about 9 m outside the defences. Given the proximity of Dere Street, it is surprising that this seems to be the only gate opening on to the road. There is no evidencefor the gates postulated on the inaccurate plan published by Richmond and Keeney (1937, pl xx). The camp, therefore, seems to have faced SE, its main gate opening almost directly on to the defences of the fort (IV). The relationship between camp III and the fort seems to have been close; both are evidently later than camp I. The excavators suggested that the small mound, 5.2 m long, 1.7 m wide, and 0.3 m high, on the SE lip of the ditch of camp III, was a traverse mound of the NW gate of the fort (Richmond and Keeney, 1937 138). In Richmond's excavation notebook (No. 19), now in the Ashmolean Museum, Oxford, there is an unpublished section which shows a turf base to this mound. Its position and orientation are only comprehensible if the initial construction of the fort preceded that of camp III. Full information is inluded in the NMR Archive. (1)
Additional reference. (2) |