Proceedings of the Prehistoric Society 53

The few surviving features included an artificial hollow, probably not a dwelling. The extensive lithic industry was dominated by microliths of the distinctive Manx facies. It is likely that a broad range of environments were exploited, and the island's Meso colonization would repay further study.

A substantial and important monument began as a boulder enclosure of c 1740-1640 BC cal with two poorly preserved inhumations apparently associated with typologically early metalwork: a flat axe and flat dagger with the central inhumation, and a knife dagger with the second. A cairn of smaller stones was added, with concentrations of inhumed bone and of cremated bone and pottery probably deriving from later reuse. Au (abr)

A segment of EBA arable landscape incorporating isolated round barrows became the chosen location for a later BA enclosure of perhaps 5ha. The excavated area within was densely occupied by two major phases of timber buildings, lasting into the earliest Iron Age. This oval enclosure was replaced by a more substantial one partly following its line and containing a series of drystone flint banks or wall footings of unknown use. The site was abandoned until the Roman road from Dorchester to Ilchester was laid across it. Au (abr)

Reviews some recent literature, dealing particularly with the distinction between votive deposlts and stores of objects. An alternative approach considers the long-term variations in the nature of intentional deposits in European prehistory. Three stages are suggested: 1 (early Neo), a unitary system of deposits prevailed, especially of food and selected artefacts, including metal when available; 2 (Bronze Age), a dual system allowed both votive deposition and accumulation and recycling, especially of non-local metal; 3 (later Iron Age), saw renewed emphasis on votive deposits, especially on the borders of emerging polities. Au (abr)

Re-examination of the evidence in the light of newly discovered documentation from J F S Stone's original excavation (1936-9) suggests two phases of settlement, the earlier one pre-Deverel-Rimbury. Analysis of artefact patterns allows the functional interpretation of different structural types. Four settlement units can be identified, similar to recurring modules previously defined on MBA sites in southern England. Au (abr)

Topographical analysis of landscapes on the heavier soils of E Anglia reveals sometimes very extensive co-axial field systems of probable later prehistoric origin. Others may be later in date. Co-axial planning may be a recurrent solution to the problem of competing claims to large areas of open land. Au (abr)

A simple derivation of henges from causewayed camps is rejected. A range of Neo structures, some covered by a mound, some not, are interpreted as mortuary structures but having affinities to henges. It is argued that henges evolved from a broad tradition, on a NW European scale, of both ritual and domestic structures. Within Britain similarities of form and function suggest that henges should be seen as a parallel development to ring banks, ring ditches, and stone circles. Regional preferences rather than distinct regional types of monuments emerged. Survival of henge traditions into the 1st millennium (cf Frilford, Navan) is also discussed. Au (adapted)

Box-sectioning of seven debatable features to reveal their profiles demonstrated that a significant number of apparent root-holes plotted in the original excavation were in fact stake-holes. Some of these could represent the EBA structures that otherwise are missing from the enclosure.