A hoard of eighteen bronze axes and three pieces of unwrought metal cake was discovered at Roxby, Lincs. in November 1882. After the sale of the pieces in 1931 all trace of the hoard was lost until it was identified in 1975 among the Wellcome Collection of prehistoric antiquities, which had been transferred to the Department of Prehistoric and Romano-British Antiquities of the British Museum. The hoard is fully described, illustrated and analysed by atomic absorption. It includes scrap axes as well as local ‘Yorkshire’ ones of the Late Bronze Age Heathery Burn Tradition, datable to the eighth/sixth centuries B.C., and was probably deposited towards the end of this period.

Excavation of the three Surrey hillforts of Anstiebury, Holmbury, and Hascombe between 1972 and 1977 showed a general similarity of design and function but some variation in detail. Ranging in area from 6 to 10 acres (2·3–4 ha.), they were all related to the use of the sling; two were multivallate but Hascombe was univallate. Occupation was seemingly slight and Anstiebury was possibly uncompleted. The contribution of scientific method to problems of prospecting and dating is discussed. This and the pottery evidence indicate construction within the period 200–50 B.C., most probably 100–50 B.C., and the three sites may have been abandoned simultaneously. At Hascombe a pit- or hearth-filling produced Class I potin coins and the hypothesis is advanced that the hillforts were dismantled as a result of Caesar's landings of 55 and 54 B.C. but it is emphasized that decisive proof is lacking.

A re-examination of Fowler Aa, B, and C penannulars leads to a reassessment of omega brooches at Numantia. The inception of particular types of penannulars could be before 75 B.C., certainly before 27, and they continue through Augustan–Claudian times with Aucissa and ‘Hod Hill’-type brooches. The distributions of certain types in pre-Roman Britain, based on Mr. Hull's researches, and of two groups of factory-made brooches and others only found in the area of early Roman conquest in Britain, are contemporary with early omega brooches here. There is no evidence for taking any of these forms into the Flavian period, and the supposed basis for dating some of them into the second century is found wanting. Instead of associating old finds of all these early types with certain forts on the Limes, from A.D. 90, or later, it is suggested that there is evidence for Germani or Celts at these places long before the Province of Upper Germany was formed and the ‘decumates agri’ were taken into it.

The early Anglo-Saxon dedication inscription at Jarrow is described, transcribed, and translated. Its purpose is considered and its contents are analysed and compared to Anglo-Saxon and Continental inscriptions, liturgical calendars, and papal documents. The stages in the execution are discussed. The lettering is shown to be in a relatively pure Roman capital without Insular forms. The other inscriptions of Monkwearmouth and Jarrow are considered in relation to the dedication inscription. Three probably consecutive palaeographical groups are suggested and similarities to lettering in manuscripts from the same centres are noted. The contents of extant and recorded inscriptions, including those attributable to Bede, are examined for evidence of their historical contexts. Most of the inscriptions emerge as characteristically ‘Roman’ products of the two houses.

Winchester has produced six bone spatulate spoons decorated with Ringerike–Winchester style engravings which are, in some cases, dated by their contexts to the eleventh century. The range of motifs and the style of execution suggest they are the work of an individual craftsman. They form part of a sequence of spoons spanning the period from the late Roman to the thirteenth century, but unlike the examples in more precious substances, which have primarily Christian ecclesiastical associations, the wooden and bone spoons of the eleventh century seem, at least from their contexts, to be domestic.