In all likelihood the Exeter cemetery was in use during the period when Eastern Mediterranean pottery, and to a much lesser extent North African amphorae and finewares, were reaching a large number of sites in the south-west. The nearest find-spot to Exeter is c. 20km to the south-east, on the coast at High Peak, west of Sidmouth (Pollard 1966; Rainbird et al. 2013). The town therefore lies within the zone where finds of 5th- to 6th-century pottery from the Mediterranean might be expected. Although amphorae of other types reached Exeter from the Mediterranean in the 4th century (Bidwell et al. 2011, 114-15), there are no later imports. The absence of all these wares, rather than suggesting a complete cessation of occupation, might mean that any surviving settlement, such as some sort of religious establishment connected with the cemetery on the forum site, was no longer part of the trading network. At other settlements in the south-west where these imports are present, we have a potential horizon for determining what other types of pottery were in use. In reality, it is only at Bantham, where there was no previous Roman occupation on the spot, that the imports provide a clear horizon. The most recently excavated part of this extensive site yielded imported pottery that is only exceeded in quantity by that from Tintagel (Reed et al. 2011). Two sherds of Romano-British pottery were found with this material: the rim of a probable Oxford parchment ware bowl and a base in South Devon ware, probably also from a bowl (Bidwell et al. 2011, fig. 23, no. 55, fig. 24, no. 61). The South Devon ware is likely to be of 1st- or 2nd-century date and was presumably a stray from a Romano-British enclosure 250m to the east of its find-spot. The absence of this ware, apart from this residual sherd, provides us with a terminus ante quem for the end of South Devon ware. Bantham lies in the heartland of the ware, as is demonstrated by the large quantities found at Clanacombe, 2.5km to the east of Bantham (Greene and Greene 1970).