1. Introduction

In Cornwall production of gabbroic pottery in the forms that had developed in the later 3rd and 4th centuries almost certainly continued throughout the 5th century (Quinnell 2004, 109-11; Thorpe and Wood 2011, 276). Nowhere else in Britain can such a lengthy afterlife for pottery of the Roman period be claimed with the same degree of confidence. In Devon the usual uncertainty surrounds the dates at which local and regional industries ended production. As in much of Roman Britain, it is usually impossible to know how long sequences of occupation continued beyond the end of the 4th century and whether pottery from the latest levels was residual or found its way into those contexts immediately after its use. Two factors, however, are helpful in studying 5th-century pottery in Devon. First, there are the amphorae and fineware imports from North Africa and the eastern Mediterranean, which occur on coastal sites from perhaps as early as the mid-5th century and which have the potential, only so far realised at Bantham (Bidwell et al. 2011), to provide broadly dated contexts for the presence or absence of pottery in the Romano-British tradition. Then there is the resumption after two centuries or so of trade in gabbroic wares beyond Cornwall, though apparently on a very small scale, and the possible significance of its absence at sites that have produced large amounts of late Roman pottery, such as Exeter and villas and other sites in east Devon.

This article cannot present any new evidence for the date at which the manufacture of widely traded wares from outside the region, such as BB1 from south-east Dorset, came to an end, but it can say something about the end of South Devon ware (SDBB). Its main purpose is to describe the latest types of pottery in the Romano-British tradition occurring in Devon and neighbouring areas and to put forward some possible explanations for changes in pottery supply at the end of the Roman period and beyond.