This paper reviews the evidence for the dating of the onset of cereal cultivation in Britain and Ireland, based upon radiocarbon-dated remains of charred cereals and also of dated contexts containing charred cereals. Many more sites with charred cereal grains are now available than existed at the time of an earlier survey by Moffett et al. (1989). Although cereal remains are far less common on Neolithic sites than, say, charred hazelnuts, in part due to differing taphonomic factors (Jones 2000), this survey demonstrates that cereals are far more widespread across the British Isles than earlier surveys suggest, irrespective of their dietary contribution (see Jones & Rowley-Conwy 2007, for a recent review of the importance of cereals in Neolithic Britain). In addition, many more of these sites have associated radiocarbon dates, an increasing number of which are high resolution AMS dates. This provides the opportunity to make meaningful comments on the chronological distribution of the evidence for cereal cultivation, and its implications for our understanding of the appearance, adoption and role of cultigens in the Neolithic of Britain and Ireland.
Cereal cultivation is one of the defining characteristics associated with a Neolithic lifestyle, yet there is considerable debate over the precise dating of the first appearance of agriculture in the British Isles. Several authors (e.g. Pilcher & Smith 1979; Edwards & Hirons 1984; Williams 1989; Simmons & Innes 1996; Edwards & McIntosh 1998; Innes et al. 2003) have highlighted the growing evidence from the pollen record for cereal-type pollen grains in contexts dating to the late Mesolithic, some as early as c. 5000 cal BC, often, though not exclusively, associated with evidence for vegetation disturbance. This has been taken to suggest the possibility of early pioneer agriculture up to several centuries prior to the generally accepted beginnings of the Neolithic in Britain and Ireland c. 4000 cal BC. Precise dating of the onset of agriculture remains a key research aim for prehistoric studies, since the possibility of early agriculture has significant implications, both for our understanding of the transition to the Neolithic, and of human-plant relationships in the late Mesolithic.
However, the evidence of cereal-type pollen grains, and their interpretation as proof of early, pre-Neolithic cereal cultivation, remains contentious, not least because of the difficulties inherent in separating cereal-type pollen from that of wild grass pollen that make positive identification of cereal cultivation uncertain (see Tweddle et al. 2005 for a recent review). Andersen (1979) separated Poaceae pollen into four groups (wild grasses, Hordeum group, Avena-Triticum group and Secale cereale). The Hordeum group (Barleys) includes only two cultivated species (Hordeum vulgate and Triticum monococcum), but several wild grass species, the Avena-Triticum group (Oats and Wheats) comprising only one wild grass species (Avena fatua), the remainder being those of cultigens. Cereal-type pollen of the Hordeum and Avena-Triticum groups are distinguished from wild grasses primarily on the basis of mean pollen grain and annulus diameters (cereal pollen grains being larger than those of wild grass species), but also the surface sculpturing of the pollen grain and the protrudance of the annulus. However, key distinguishing features are not always easy to identify where variable preservation of pollen grains occur, in which case caution should be exercised in any identification based upon grain and annulus diameter alone. Large cereal-type grains can also occur through swelling of wild grass pollen mounted in glycerol jelly, and as a result of genetic mutation (e.g. polyploidy), commonplace amongst the majority of plant species. In addition, several palaeoenvironmental studies have also produced cereal-type pollen grains from contexts significantly …