Welsh Journals

large slabs to its southerly, downhill face. Much Romano- British pottery and broken quern-stones was incorporated in, and was found in the old ground surface beneath the bank. The bank itself was 9 ft. wide and stood 3 ft. high. Extensive stripping of the terrace produced large quantities of Romano- British pottery, ranging in date from the late first or early second century A.D. to the fourth century A.D. A single course of large limestone blocks running east-west, but robbed at its western end marked the south wall of a rectangular building of Roman date but of unknown dimensions. The wall overlaid a pit containing a few sherds of Belgic and "Iron B" pottery. The depression in front of the terrace proved to be a quarry-scoop to provide material for the construction of the enclosure banks, and for the levelling of the hillside to form a terrace for building purposes. A considerable amount of quarrying must have taken place on the site of the terrace as well. Trial trenching across slight depressions suggestive of hut platforms, outside and to the north-west of the main enclosure system, in the hope of revealing pre-Roman buildings, were abortive. The only finds were a scatter of Romano-British pottery. The overall plan of the enclosures suggests that the Romano- British occupation in the interior of the hillfort took the form of one, or possibly two large farmsteads. However, further excavation is necessary before the nature of this occupation can be fully understood. JEFFREY L. DAVIES. 3. The Roman Villa at Whitton, Glamorgan Excavations of 1967 In 1967 work at Whitton (ST 081713)12 was concentrated in the south-east corner of the villa, and work on the east range is now completed. In 1968 it is proposed to complete the excavation of the south range, begun this year, and to investigate a substan- tial area of the yard. As in previous years, it was clear that 12 Morgannwg, 9 (1965), 91-95 10 (1966), 59-63.