The probable site of a medieval llys, a princely court partly dismantled in 1317.
Excavation, 1973-4 (White 1979) & 1979, with further work 1987-8 (White & Longley 1995), recorded part of a rectilinear enclosure with at least one rounded corner. Its ditch had been recut on at least one occasion. It was traced for 70m north-north-east to south-south-west. Intially thought to represent a Roman military work, refurnished in the early medieval period. The apparently curving corner has suggested the presence of a castle mound (see NPRN 400058).
This overall interpretation is supported by a radio-carbon determination of about 27-387AD, however scarsity of Roman material, limited to a single scrap of samian and some uncertain sherds, counts against the presence of a fort. It is possible that this was the site of a later Prehistoric style settlement occupied in the Roman period.
The site of the llys is otherwise uncertain (see Johnstone 1997, 63; Longley 1997, 45). Two possibly thirteenth century sculptured heads are known from the village (White 1978). Excavations at the traditional site of the llys, about 650m to the west-south-west recorded only 18th century remains (see NPRN 15012/401125).

Sources: White in Archaeologia Cambrensis 126 (1977), 140-5
in the Bulletin of the Board of Celtic Studies 28.2 (1979), 319-42
White & Longley in the Transactions of the Anglesey Antiquarian Society and Field Club for 1995, 13-21
Johnstone in 'Landscape & Settlement in Medieval Wales' ed. Edwards (1997), 54-69
Longley in 'Landscape & Settlement in Medieval Wales' ed. Edwards (1997), 41-54

John Wiles 12.07.07

TEXT FROM ROMAN FRONTIERS PUBLICATION:

Limited excavations in 1973, 1974 and 1979 within the village of Aberffraw revealed what could be the defences of a Roman fort. The defences were sampled in two places. R. B. White interpreted the defences as consisting of two phases of a ditch, rampart and road with Roman military characteristics together with a phase of post-Roman defence. L. Alcock suggested a different interpretation, namely, a sequence of natural deposits and a single phase of Early Medieval fortification.1 A revised report by White and Longley, incorporating the findings from further excavations in 1979, suggested an alternative sequence commencing with a V-profiled ditch, 6m wide and 2.5m deep, with a well-defined cleaning slot. A berm, 1.5m wide, separated this from a clay rampart which was 5m wide. A second phase ditch and rampart was interpreted as being post-Roman. The single samian sherd and two small sherds of possibly Roman coarse pottery recovered from the excavation are not sufficient to confirm the Roman origin of the site and no supporting evidence has so far emerged.