The dating and derivation of cursus monuments, the large linear earthwork enclosures of the British Neolithic, has long been a matter of debate (Bradley 1986; Barclay & Bayliss 1999). The spatial association of cursuses with henge monuments at Thornborough, Llandegai, Maxey, and Dorchester on Thames, and with hengiform enclosures and ring-ditches at numerous other sites, at one time encouraged their attribution to the later Neolithic (Loveday 1989, 71). The sheer scale of structures such as the Dorset Cursus and the Greater Stonehenge Cursus, and the quantity of labour evidently invested in their construction, tended to support this view (eg, Renfrew 1973, 547). Furthermore, several cursus sites have produced artefactual assemblages of diagnostically Late Neolithic character, including sherds of Peterborough Ware and Grooved Ware (Buckley et al. 2001, 110). Only in the past two decades has a reliable series of radiocarbon dates for cursus monuments begun to emerge. Determinations from the Dorset Cursus, the Stonehenge Lesser Cursus, Dorchester on Thames, and Drayton all indicate a horizon of 3600–3000 cal BC for the construction of cursus enclosures (Barclay & Bayliss 1999, 25; Barclay et al. 2003, 13). That is to say, they appear to fit into the later part of the Early Neolithic, or to a ‘Middle Neolithic’, if we care to use such a term. In mainland Britain, the only cursus monument defined by bank and ditch to have provided a substantially earlier radiocarbon date is Sarn-y-bryn- caled, assayed at 3891–3889 or 3800–3660 cal BC at the 68 % confidence level (OxA-3997). This date came from a patch of small fragments of oak charcoal located immediately above the primary fill of the cursus ditch, and is to be regarded with a degree of caution (Gibson 1994, 17; Barclay et al. 2003, 185). Other potentially early dates for ditched cursus monuments have come from the two enclosures at Eynesbury in Cambridgeshire, which produced optically stimulated luminescence (OSL) determinations of 4830–3460 BC and 4860–3450 BC (Ellis 2004, 64). Given the high standard deviations on these dates, it is again prudent to reserve judgement on their significance. A more refined chronology for cursus monuments has proved illusive, although it has sometimes been suggested that structures with rounded terminals are earlier than those with squared ends (Barclay et al. 2003, 95). However, in 1979, Gordon Maxwell identified a group of Neolithic linear enclosures that are defined not by a continuous bank and ditch, but by a series of pits or post-holes (Maxwell 1979). 1 Proceedings of the Prehistoric Society 72, 2006, pp. 00–00 On the Origins and Development of Cursus Monuments in Britain By JULIAN THOMAS 1 The problem of dating cursus monuments has troubled British archaeology for some decades. A series of recent radiocarbon determinations from sites in lowland Scotland suggests that cursus monuments defined by posts and pits are generally earlier than the more familiar bank and ditch structures, and may have been constructed very early within the British Neolithic sequence. The implications of such a sequence are discussed in relation to the affinities of these structures, and landscape change between 4000 and 3600 cal BC. ‘It may even be that somewhere in Scotland lies the origins of a monument type which until now has been seen as a major societal component for the Neolithic south of the border’ (Brophy 1999, 128). 1 School of Arts, Histories and Cultures, University of Manchester, Oxford Road, Manchester M13 9PL, UK J. Thomas. ON THE ORIGINS AND DEVELOPMENT OF CURSUS MONUMENTS IN BRITAIN