I cannot begin this Anniversary Address without a reference to the loss which the Society has suffered by the death on 11th January 1972 of our late President, Professor Francis Wormald. His last annual address was given from this Chair on this occasion just two years ago. He was a great scholar whose rare distinction in his own field was combined with shrewdness of judgement and a wide knowledge of affairs: his modesty, sensitivity, and charm, enlivened by a delightful sense of fun, made him a most engaging companion and a loyal friend. How precarious his health had been for the greater part of his life was known only to his closest intimates: and the knowledge was never allowed, even after the very severe illness which he suffered some eighteen months ago, to cloud his zest for living or to dim his delight in beauty. His services to learning as scholar, teacher, and administrator were very substantial but largely unobtrusive. He was never one to seek the limelight, and he never received the public recognition which his distinction deserved. To this Society his devotion was wholehearted, and I would particularly wish to stress the discriminating help which he gave over many years by wise advice on the library and by strengthening the resources of the Research Fund.

The paper begins by surveying the different types of early iron-smelting furnace, based on a tentative classification proposed by Coghlan in 1956. The ambiguities in this classification are indicated, together with examples of furnaces that do not fall easily into one of its three categories.

On the basis of data derived principally from furnaces of the Early Iron Age and Roman periods from northern Europe, the author proposes a new classification into two main groups, differentiated by their provisions or otherwise for the removal of molten slag during the iron-smelting operation. Each of these groups is further subdivided, according to the shape of the furnace superstructure and/or the method of supplying the air blast.

The 1968–71 excavations have produced a large body of information bearing on Roman Gloucester's development from legionary fortress to chartered city. It has proved possible to correlate Roman building sequences in different parts of the city. A building period which may be associated with the foundation of the colonia has been extensively excavated. The Roman forum has been discovered in a position equivalent to that of the principia of a fortress. Knowledge of the city's defences in both Roman and later periods has been amplified by work on three sides of the Roman circuit. Late Saxon use of the Roman wall is suggested and late Saxon buildings fronting Southgate Street have been discovered. At 13–17 Berkeley Street the first extensive excavation of medieval levels in the city has revealed a length of street fronted by late eleventh–fourteenth-century buildings.

The three seasons of excavation 1969–71 have provided details of occupation lasting continuously from the third century until the nineteenth century. In the Roman period cobbled streets and timber buildings, pits and wells have been excavated but they show little sign of rigid planning. The streets continued to be used in the fifth and sixth century when huts of post-built and grubenhaus type were constructed. Some parts of the interior were now ploughed.

In the eighth and ninth centuries timber buildings, storage pits, and cess pits were concentrated in one area close to the south wall of the Roman fort. Near by a well, constantly used from the Roman period and subsequently relined with timber, was abandoned and used as a rubbish tip. Some additional buildings and pits belonging to the ‘hall complex’ of late Saxon date were found. Medieval features included boundary ditches, a lime kiln and slaking pit, and a timber building. Some areas of the interior were cultivated throughout this time.

In the early sixteenth century a large store building was erected in masonry: it was demolished late in the sixteenth century. Thereafter the fort was used as a prison camp until the early nineteenth century. Various features of this period were found.

The Landgate was further excavated bringing to light structural details of Roman, Saxon, and medieval date. Limited excavation in the priory exposed details of the claustral buildings.

A late-Roman silver ingot from Kent has recently come to light. It weighs one Roman pound and is stamped EX OFF CVRMISSI. All such ingots seem likely to have been distributed as imperial donatives to troops in the fourth century. They are related to presentation silver plate and other gifts of this type, and more than forty silver ingots are now known from the European Roman provinces. The only other known stamp by Curmissus—or perhaps the curator missionum—is on a broken ingot in the hoard from Coleraine, Co. Londonderry.

Excavations in 1970 took place on three major and two smaller sites. The early Norman chapel discovered this year within the castle at Castle Yard was described in the previous interim report. At Lankhills sixty-eight fourth-century graves were examined, bringing the total excavated to 219. The graves can now be classified in an approximately chronological sequence according to contents and burial practice. At Lower Brook Street early fifth-century pottery of North German origin suggests the presence of mercenary or federate elements in the final stages of the Roman town. St. Mary's Church appears to have originated in the tenth century by the addition of an apse to an earlier rectangular stone building possibly of domestic character. In Houses IX and XII several phases of twelfth-century timber construction were excavated, but Houses X and XI, adjacent to St. Mary's to north and south, seem to have been open plots at this time. The later phases of St. Pancras' Church were examined and the twelfth-century church uncovered. On the Cathedral Green the excavation of Building E showed that it was the south range of a large courtyard complex, probably to be interpreted as the claustral buildings of New Minster in the period c. 1066–1110. Earlier stages may represent the infirmary of the Anglo-Saxon monastery. At Wolvesey Palace the east hall of c. 1130 was stripped and the later phases of its complex development worked out in detail. A ‘reredorter’ block added to the north end of the west hall about 1135 was cleared of many later phases of reconstruction. The excavation of the central courtyard revealed a twelfth-century well-house. A final season will take place in 1971.

This paper examines, from two points of view, a collection of some ninety buildings with base-cruck trusses, and a further group of related buildings, particularly those whose roof trusses incorporate short principals. Examination of the social status of the buildings shows that they belonged to a remarkably homogeneous section of feudal society, and their distribution can be explained largely in terms of the social connections of a class, with some further influence of local schools of carpentry. The typological development of base-crucks and short principals is traced from their twelfth- or thirteenth-century origins to a fifteenth-century devolution into various local sub-types.

The present paper was prepared because the career of Hugh, Lord Willoughby—President of the Society from 1754 until his death in 1765—had been largely overlooked by historians, and denied even the briefest of general summaries. Yet it was a career characterized by industry, by versatility, by probity; and if in political areas it was impaired by Willoughby's diffidence, so that the opportunities arising from his own undoubted abilities and from Chancellor Hardwicke's patronage were scarcely ever taken up, it seems to have reached much more complete fruition intellectually in this Nonconformist peer's association with the Society of Antiquaries, the British Museum, the Royal Society, the Society of Arts, and the Warrington Academy. Most particularly it epitomizes the close involvement of Dissenters in the eighteenth-century Enlightenment.