The discovery of two fragmentary inscriptions at the auxiliary fort of Castell Collen (pl. xxia) and the newly discovered legionary fortress of Carpow (pl. xxib) has made a notable addition to the interesting group of Romano-British inscriptions in which the text is flanked by peltae, the terminals of which have been treated as birds' (or griffins') heads. The Castell Collen fragment was found during excavation of the extramural bath building in 1956 where it appeared face downwards, reused as a paving slab; the excavator suggested that it might originally have stood over the east gate of the fort and used it as supporting evidence, for reasons which will be considered below, for the view that the fort was rebuilt in stone in the Antonine period. The Carpow fragment was found during excavations of the east gate in 1964, where presumably it once commemorated the building of the fortress in the Severan period. In neither case is there enough of the inscription to identify the unit explicitly but the presence of the Pegasus and Capricorn emblems, in whole or in part, makes it clear that both fragments are to be attributed to the Second Legion Augusta.

The situation, character, and antiquarian history of South Cadbury Castle, Somerset, have already been described (Alcock (1967a); (1967b)). A reconnaissance excavation in 1966 hinted that occupation of the hilltop had begun in the Early Neolithic, and after a hiatus had been resumed sometime in the Late Bronze Age. It demonstrated that from early in the pre-Roman Iron Age there was a rich and intensive occupation, during which the defences were probably built. This appeared to have been ended by a sack at the hands of the Roman invaders c. A.D. 45. Late Roman pottery, metalwork, and coins showed that activity was renewed in the third-fourth centuries A.D. Imported pottery of well-known forms proved that the site was reused in the fifth—sixth centuries. Finally, it was refortified with a mortared stone wall as a Late Saxon burh. The reconnaissance also showed that on top of the hill (fig. 1, c) rock-cut pits and post-holes were readily discernible under the turf; that on the slopes within the defences (fig. 1, B) traces of buildings of various dates had survived many centuries of ploughing; and that in the rear of the inner rampart (fig. 1, A), a great depth of hill-wash and plough-soil overlay stratified levels of the Iron Age.

During recent years a new consensus of opinion seems to have grown up concerning the career of the eighteenth-century antiquary William Stukeley: that his ideas underwent no significant alteration in the course of his adult life, and that Stuart Piggott's famous characterization of Stukeley — that he changed from an objective field archaeologist into a religious crank — was completely wrong. While there is much to commend this revisionist approach, it also presents certain difficulties. It fails to account for the apparent speed and drama of Stukeley's decision to seek ordination as an Anglican minister, or for pronounced differences in emphasis and tone between his earlier and later writings, and it fails to address some important textual difficulties in the dating and interpretation of his manuscript works. This paper is intended to address those problems. It examines the changes in his religious attitudes, and their implications for his scholarship, over the six decades in which he carried out antiquarian researches. In the process, it is intended to make a contribution to the cultural history of the eighteenth century, and also to the early story of the discipline of archaeology.

During the month of September 1920 excavations were carried out by the Museum Curator at Rabat, in the vicinity of the Roman Villa Museum.

Further work was undertaken in the nave of the ruined church at Stone-by-Faversham in 1971–2 as part of the programme of its consolidation as an Ancient Monument by the Department of the Environment. Evidence was found for a pre-Conquest stone nave, itself possibly preceded by a timber nave; the stone nave was extended westwards in the twelfth or thirteenth century. West of the pre-Conquest nave there were traces of a Roman building distinct from the Roman mausoleum incorporated in the chancel of the church. The dating evidence of Roman and Saxon coins and pottery is discussed.

Seven gold pins noticed recently in place in the Ridgeway gold pommel form the only British example of exclusively functional gold pins dated to the Wessex period. The Hameldon Down pommel repair, although functional was also decorative. A Breton parallel from La-Garenne, Morbihan, is the only other instance of purely functional pin use and still has two pins in situ with their heads hidden by the continuous decoration of the piece. The Ridgeway pommel's shape and associations suggest that, like La-Garenne, it is late in the pin typology. Both may be from contemporary graves, which if not transitional Wessex I/ Wessex II, might be early Wessex II.

Entries in the London Museum Catalogue indicate that the association of a carinated pedestal based bowl with a pair of applied brooches and a bronze fragment in Mitcham (Surrey) Grave 205 almost certainly represents a valid closed find. The continental antecedents of the complete applied brooch from this grave may be dated to the first quarter of the fifth century. The English brooches with this floriate cross design show an insular development with a restricted distribution in south-east England It is argued that the English examples were manufactured during the first half of the fifth, probably towards the middle of that century. This brings the brooches nearer to the date of the pot, which Dr. Myres argues must have been made in the years of Roman–Anglo-Saxon overlap around A.D. 400.

The castle of Ascot Doilly appears from documentary sources to have been put up c. 1129–50. Excavation in 1946–7 of a small mound on this manorial site showed that it had contained a stone tower 35 ft. square which had been built up from the natural surface of a 4-ft. rise of Lias clay protruding through the gravel of the Evenlode valley bottom. Round this tower, as it was raised, had been piled a low mound of clay; thus the impression of a tower on a mound was created. Beside it are remains of a bailey and contemporary paddocks. The tower had been deliberately demolished, probably c. 1180.

The excavation thus revealed a new principle in smaller defensive building of the period, a mound piled round a stone tower. It also yielded a useful series of mid- to later twelfth-century pottery and other objects, and evidence of domestic window glass in the twelfth century.

There are remains of thirteenth-century and later buildings in the bailey area. The village of Ascot represents a dual holding, with two mound-and-bailey castles at opposite ends 800 yds. apart, and between them the church (with twelfth-century work). There is also evidence of pottery-making in the village, at least in the early thirteenth century.

The third season of excavations at Verulamium, inaugurated by the Research Committee of this Society, lasted from 4th July to 24th August 1957. The main area investigated lay along the edge of the allotment field between the theatre and the forum; this area was not accessible previously for reasons mentioned in last year's report, and in consequence much work had to be fitted into the time available, and even so, the lowest levels were not completely explored. But in addition further work was done on site K. VII at the rear of the defences beside the river, and a new site S was explored north-east of the river Ver on the brow of the rise above the millstream. Here a large extra-mural building, or buildings, with a cellar came to light; but its complete excavation had to be deferred. Work by contractors on the new road began during the excavations, and this work was watched; after the excavations closed, the staff of the museum rescued an important new deposit of Belgic mint debris.

The scribe of these and the following lines, which lament the English disasters France about 1449, had the kindness to insert the names of the noblemen, thus denoted by their badges. The Root, he tells us, for example, is Bedforde, the swan Gloucetter, and the cresset Excetter. This swan badge of Humphrey, duke of Gloucester, who died in 1447, is familiar, but he was by no means the only bearer of a swan in his time. It had been the badge of the great house of Bohun, earls of Hereford and Essex (pls. XXXVI, g; XXXVII, g; XL), and is assumed to have come to Humphrey through the marriage of his father the earl of Derby, later King Henry IV (pls. XXXVI, f; XXXVII, g; XL), with Mary de Bohun the younger daughter and coheir of Humphrey, earl of Hereford. But even among Mary's own descendants Duke Humphrey was not alone in using it, for Henry V (pls. XXXIV, a, b; XL), Henry VI, and the latter's son Edward did so too. Mary's elder sister Eleanor (pls. XXXVII, f; XL) used it also, as did her husband, Thomas of Woodstock (pls. XXXIV, c, d, e; XL), duke of Gloucester, and the Staffords, dukes of Buckingham (pls. XXXVI, a, c; XL), descended from their daughter Anne. Nor was this all, for in 1325 Margaret de Bohun (pls. XXXIX, b; XL), the daughter of an earlier Humphrey, earl of Hereford and Essex, had married Hugh de Courtenay, earl of Devon, and many of their descendants, Courtenays (pls. XXXIV, f, j; XXXVII, h; XL), Luttrells (pls. XXXVII, a, b, c, e, i, j; XL) and others (pl. XXXIX, a) used swans as badges, or as crests or supporters in their arms.

We all know what maps are. Most of us have occasion to use them frequently. Many of us, no doubt, have spent pleasant hours in our studies, performing, with their help, what have been called ‘armchair travels ‘. But only a few have been led to inquire into the history of maps and mapmaking, and fewer still have interested themselves in that period of the history of cartography which is the subject of this paper—the medieval period.

In 1938 the Kent Archaeological Society carried out extensive excavations on the site of the important Iron Age hill-fort of Oldbury, above Ightham, Kent. An abridged report of the work has already appeared in Archaeologia Cantiana, li (1939), 137–81. But in view of the more than local importance of some of the results obtained it has seemed advisable to republish them on a considerably expanded basis. The actual work of excavation, recorded in detail in Archaeologia Cantiana, is here more briefly summarized; whereas the bearing of the results obtained upon more general problems is considered at rather greater length.

In the course of the paper on ‘Elizabethan Sheldon Tapestries’ (Archaeologia, vol. lxxiv), which Mr. John Humphreys read before the Society on 3 April 1924, he dealt principally with the historical and topographical interests of many of the then known examples of Sheldon tapestry-work. In the persons of Mr. Humphreys, Colonel Howard, and Mr. A. F. Kendrick we indeed salute pioneers in research concerning this subject. The object of this present paper is to deal as far as possible with the results of my recent documentary researches concerning William Sheldon's scheme and those who were associated with it. I should at once add that these researches have only been made possible by the financial support of our Fellow, Mr. Rees Price, of Broadway, Worcestershire, who himself has done so much similar work concerning the history of the Sheldon family.

In 1915 Professor W. G. Collingwood completed his survey of all the Anglian and Anglo-Danish sculpture in Yorkshire. The Addenda to York published then added three fragments to the twenty-four ‘certainly or probably local to the city’. Since then, the grand total of twenty-seven has at least trebled. The main sources for this spectacular increase are the demolition of, and subsequent excavations on the site of, the church of St. Mary Bishophill Senior in 1963–4, and the much larger operations at York minster, begun in 1966 and recently completed. A full catalogue of the latter must await the completion of the excavation report. Finds to the southwest of the river Ouse appear in a recent inventory of the Royal Commission on Historical Monuments. Further sculptures were found in Newgate5 and All Saints, Pavement. The material for consideration grows steadily, and it is now clear that York was an important centre for sculpture before the Norman Conquest. This paper is limited to consideration of the later Anglo-Danish sculpture, which involves presentation of recent discoveries and reassessment of some of the sculpture described by Collingwood. It also includes a reassessment of the Nunburnholme cross, which provides parallels for many of the features of York sculpture, and is now no longer an isolated work. As there is much new material to be considered, the individual carvings will be listed and briefly described, beginning with the Nunburnholme cross, which serves as a key to the different elements.

Permit me now to communicate to the Society of Antiquaries, the sequel of such observations as I have had an opportunity to make with regard to the slructure of ancient castles; a subject, which, however ill I may have succeeded in treating of it, appears to me capable of throwing considerable light on the history of our country; and therefore I cannot but still endeavour to make every effort in my power to elucidate it; wishing my attempts to be a means of exciting others to make more successful enquiries.

A little less than 800 years ago the bones of St. Petroc were removed from his priory church in Bodmin by a seditious canon and carried into Brittany. About a year later, after the prior of Bodmin had mobilized his friend, the head of the royal chancery, and through him King Henry II himself, and the regent of Brittany, not without some miraculous interventions by the saint himself, the bones were returned to Bodmin in an ivory casket and solemnly installed again in the priory church by the learned and saintly Bishop Bartholomew.

The work of 1971–3 has produced a variety of important results. The Roman defences and defended area sequences set out in the first report have been confirmed with modifications of detail. New evidence has been found for the nature and extent of the Roman city outside the defended area: there are now the first hints of Roman planning in the St. Catherine Street—Hare Lane area and in Lower Northgate Street there was at least one monumental building and a street-side colonnade. Sub-Roman occupation has been recorded at 13–17 Berkeley Street. New evidence has been obtained for the possible first phase of Gloucester castle and the development of the medieval defences. The excavated evidence for the medieval occupation at 13–17 Berkeley Street has been combined with, and partly reinterpreted in the light of documentary evidence. At 82–4 Northgate Street the standing remains of a sixteenth-century and later building were recorded and its site excavated and related to documentary evidence. In the light of new discoveries, accounts are also given of St. Bartholemew's Hospital, the river Severn, and its two main medieval bridges within the city bounds.

Considerable interest attaches to the Car Dyke as it has been claimed as a Roman work by several authorities (Dr. Cyril Fox, Archaeology of the Cambridge Region, p. 179). In June of this year I attempted to make sure of the date by cutting a section through the fosse.

When a well was sunk through the bank of the dyke at the Lodge, Waterbeach, many years ago, the workmen cut through a boat of some kind. It seems reasonable to suppose that this boat dates from the time when the dyke was still full of water, and had been left moored in some little creek or dock probably near a settlement site. I therefore cut a section as near as possible to the well, that is, within a few yards of the south-east corner of the Lodge. It has not been possible to complete the section or reach the bottom of the ditch owing to the wet summer. The water-level was found at a depth of 4 ft. The section, however, cut through the middle of a small dwelling on the very lip of the ditch, debris from it forming a thin stratum separating different layers of silt. From the contents of the midden which formed the floor of the hut it was evident that the site was one of the Pagan Saxon Period.

In 1729, a decorated fourth-century Roman silver dish bearing a Christian inscription was found at Risley Park, Derbyshire. Damaged when found, the fragments of the vessel were soon lost, but an illustrated account of it was published by William Stukeley in 1736. Stukeley and later authorities interpreted the inscription as implying that the dish had belonged to a late Roman church in France, and considered that it had been brought to Britain as loot in the Middle Ages. This paper presents a description and assessment of the Risley Park lanx in the light of the greater knowledge of late Roman silver plate now available, and makes the suggestion that the vessel may have been imported into Britain in the Roman period.