Welsh Journals

Search over 450 titles and 1.2 million pages

EXCAVATIONS AT CAERSWS, 1967, INTERIM SURVEY C. M. DANIELS, M.A., F.S.A., G. D. B. JONES, M.A., D.PHIL., F.S.A., W. G. PUTNAM, B.A. A second season's excavation was carried out at Caersws during August and September 1967. The defence-section begun in 1966 was completed, a section was cut across the praetorium and an area in the vicus shortly to be built upon was stripped. 1. THE DEFENCES In the defence-section the earliest ditch, discovered but not excavated in 1966, and the six-foot high clay rampart were fully sectioned. These showed three distinct periods of construction. The first consisted of a solid laminated clay rampart c. 29 ft. in width and based on an impressive timber corduroy (PI. V), with a frontal turf revetment and external ditch 13 ft. 6 ins. in width. At some date the rampart had been systematically slighted and the ditch filled (PI. VI). In the second period the slighted rampart was heightened partly with debris from the fort interior (see below) and partly with fresh stiff clay, the ditch filling carefully packed with cobbles, and a new rampart face, bonded into the earlier work, extended across it. The layers of strapping discovered in 1966 actually within the forward mass of the clay rampart formed bonding courses in this new face and did not run back further than the junction with the older material. The final period consisted of the addition of a stone face to the rampart, a feature attested in numerous Welsh forts and elsewhere dated to the first half of the second century. Of this stone face, however, only the footing survived. 2. THE FORT INTERIOR Within the fort the defence-section was carried back 100 ft. from the heel of the rampart in order to explore the barrack blocks of the retentura and a further trench, 180 ft. long, was cut across the praetorium. Each of these was taken down some six feet to the natural clay subsoil. In both the earliest occupation consisted of timber buildings, from the construction trenches of which two groups of early or possibly pre-Flavian pottery were recovered. 1 See Mont. Coll. Vol. 59, p. 1 12 for previous season's work and references to earlier publications and discussions of the site, also Journal of Roman Studies lvii (1967) 174.