Maumbury lies a third of a mile (500 m.) SSW. of the county town of Dorset alongside the Roman and modern road from the town to the harbour at Weymouth (SY 691899). The site is now a public park wedged between two railway lines. It lies just below the crest of a low chalk ridge at a height of 260 ft. (78 m.) and overlooks the lower ground to the north east towards the River Frome and areas of the present town (fig. 1).

We have studied the techniques of decoration of the backs of the British bronze mirrors according to the methods outlined by us in Lowery, Savage, Wilkins, 1971, 176 f., and the construction of the design on the newly discovered Holcombe mirror (Fox, 1973). Our intention is to examine the technical relationships between the mirror backs and to note some details of bronzesmiths' practice which they illustrate, and on which the Holcombe mirror sheds much light. Our conclusions are based mainly on examination of silicone rubber moulds which, with the co-operation of the museums concerned, have been made of all the decorated mirror backs now available. In default of a similar study of any extensive body of other material, no comparisons outside the mirror series are cited. We are not here concerned with mirrors whose backs are undecorated.

Amphitheatres, although not so ubiquitous as, say, bath-buildings, were common enough in the Roman world to attract a set of terms applicable to their very distinctive structural features. It is true that they varied in construction from a simple earth bank to an elaborate masonry building like the Colosseumat Rome. But certain characteristics were constant, and it seems useful to list the technical terms used to describe amphitheatres. Some of these are scarcely applicable to the relatively simple Chester amphitheatre, while others are etymologically rather dubious. This list itself is almost entirely confined to the structural features since there are easily accessible descriptions of the spectacles which took place within amphitheatres. Previous lists have been given by Dyggve in his description of the amphitheatre at Salona and by Heidenreich in his report on the civil amphitheatre of Colonia Traiana at Xanten.

Silchester (Calleva Atrebatum) is the most extensively excavated of all Romano-British town-sites but the majority of the finds, now in Reading Museum, remain unpublished. Unfortunately these finds were not related stratigraphically to the site and therefore remained undated; insula and pit provenance is the most that was recorded, and then only occasionally, in the early years of excavation when the supervisors could find time.

The reputed Romano-British church at Silchester was first excavated in 1892 by George Fox and W. H. St. John Hope in the third season of the Society's long campaign of excavations in that town.1 Claimed as a church by the excavators in their report of 1893, it has usually been so accepted; but the lack of decisive evidence for this interpretation was emphasized by Professor J. M. C. Toynbee in 1953, nor was the date of the building at all certain. Accordingly there existed a strong case for re-excavation by modern methods and this case was put to the Silchester Excavation Committee by Mr. George C. Boon in 1959. With the permission of the landowner, our Fellow the late Duke of Wellington and of Mr. A. J. Massie, the tenant farmer, excavations took place for five weeks in the summer of 1961: the work by invitation of the Committee was directed by Sir Ian Richmond. A brief report on the work was published the following year, and Richmond gave an illustrated account of the excavation to the Society on 31 January 1963; but at the time of his death in October 1965, though the plans illustrating the paper had been drawn, no written report was found. The account offered here accordingly derives partly from the field note-book kept by Miss M. G. Wilson, partly from the personal recollections of Miss Wilson and Mrs. M. A. Cotton (both of whom were on the site throughout the season), and partly from the J.R.S. account and some notes taken by Dr. W. H. Manning of a verbal report by Richmond to the Silchester Excavation Committee in 1962. Mr. George Boon also generously placed at my disposal his own independent discussion of the church which has now appeared in the second edition of his Roman Silchester (London 1974).

Alist of ring-swords and beads known to date was published in 1967, and a few more items and observations may now be added. Another jewelled example has just come to my notice in the museum at Laon (Aisne), France (pl. lxiva–c). The basic metal is presumably bronze as it is covered in places with green corrosion products. It is a form which began early, i.e. a triangular middle section with animal head terminals at each end, as in a Snartemo grave. The pommel is hollow and was fastened to the rest of the grip not by lateral rivets, but by the tang of the sword which projects slightly through a hole in the top. Its measurements are 5.1×1.5 cm. and the middle triangular faces are inlaid with garnets in curved cloisons at each corner, and framed in a zigzag border presumably inlaid with silver or niello. From the arced segments each side of this a tongue-shape, also decorated with zigzags, runs up each incurved side towards the apex. At one end is a splayed curved terminal, no longer an animal head but bearing ribbon animal ornament in relief. The corresponding opposite end is not decorated, but on top is a small hollow, and a piece at the extremity is broken off and missing. This end section is a pad for the reception of a missing ring-knob, with the end of the vertical bow fitting into the hollow.

During the eighteenth and nineteenth centuries there were three essential things to be seen by visitors to the Tower of London: the menagerie, the crown jewels, and the Line of Kings. The first and second of these still exist, although the menagerie was moved to Regent's Park in 1834 and became the London Zoo. The jewels, of course, are still in the Tower, but the Line of Kings as such has disappeared, although most of its component parts survive.