Skip to Content

Philip Grierson (Memoir)

Previously unpublished memoir by Philip Grierson, completed on December 29, 1993, and revised on April 8, 1994. Philip Grierson appended to this narrative an extract from another memoir, completed on May 2, 1996, and housed in the Coin Room of the Fitzwilliam Museum, Cambridge, which concerns the making of his coin collection. Grierson left occasional blank spaces in the typescript, which are here completed when possible with data inserted between brackets. At Dumbarton Oaks, Philip Grierson was Advisor in Byzantine Numismatics between 1955 and 1998. Philip Grierson died in 2006.


Dumbarton Oaks


My closest connexion in the United States has been with the Dumbarton Oaks Center of Byzantine Studies in Washington, which during the second half of my life has been for me virtually a second home. I must have spent there on an average between two and three months a year. It has made possible what many would regard as the most substantial body of scholarly work I have done, that embodied in the introductions to the successive volumes of the Dumbarton Oaks coin catalogues. Although my major interests have always lain in the history and coinage of Western Europe, they have been dissipated there over a larger field. I was in fact writing on Byzantine coins before making any contact with Dumbarton Oaks, and probably before I had even heard of it, for when I started to interest myself in coins it had not yet gained the world reputation in the field of Byzantine studies that it was in time to acquire. But it was my connexion with it that rendered possible my progress from scattered articles to the substantial books which have made their contribution to the enormous progress in Byzantine numismatics that has taken place over the past half-century.

I first visited Dumbarton Oaks in late November 1953 at the invitation of the Director and stayed there four days (November 30th–December 4th). The invitation was arranged by Alfred Bellinger, whose acquaintance I had made while spending the second half of 1953 at the Museum of the American Numismatic Society and who was a member of the Dumbarton Oaks Board of Scholars. A year later, after a second visit in the summer of 1954, I was invited to join the staff in an honorary capacity, with the title of Advisor in Byzantine Numismatics, the appointment to take effect from 1 July 1955. It was renewed on an annual basis up to 1967, when it was decided that it should continue indefinitely at the Trustees’ pleasure, and I still hold it at the time of writing (1993).

Both Dumbarton Oaks, and my specific interests that made the appointment possible, need to be explained.

Dumbarton Oaks is the name of a large house and garden in Georgetown, one of the oldest parts of Washington D.C., and as we know it today it is effectively the creation of two public-spirited, wealthy and enlightened benefactors, Robert Woods Bliss (1875–1962) and his wife Mildred (1879–1969). Mr Bliss was a member of the State Department, and spent the early years of the century in various diplomatic posts in Europe and South America, ending as ambassador to the Argentine in the late 1920s. His wife was independently wealthy – she was heiress to the Castoria patent medicine fortune - and was highly educated, speaking five foreign languages fluently and renowned for her skills as a hostess. In 1920, when they bought the Dumbarton Oaks property, it has been described as consisting of ‘an old-fashioned house standing in rather neglected grounds encumbered with farm buildings, roads and rather haphazard paths’. Its recommendations were the house itself and the site, for it is situated on a high ridge with a splendid view across Georgetown towards the Potomac, though separated from the river by a built-up mile of mainly residential housing. Over the next two decades, and more particularly after Robert’s retirement in 1933, the Blisses remodelled the house and grounds, substantially adding to the former and creating, out of what was virtually a wilderness, one of the most famous and best-known gardens in the United States.

Musicians associate the name of Dumbarton Oaks with a concerto written by Igor Stravinsky for Mrs Bliss and first conducted by him in April 1947 in the magnificent Music Room which she added to the original house. The general public knows it as the venue of the Dumbarton Oaks Conference in 1944 in which representatives of the Allies in World War II held the preliminary discussions that led in due course to the establishment of the United Nations. But scholars, at least on this side of the Atlantic/ think of it simply as a Byzantine research institute, for in 1940 the Blisses, who to their sorrow had no children, transferred the house and grounds of Dumbarton Oaks, with a generous endowment, to the Trustees for Harvard University to serve as a centre for study and research ‘in the Byzantine and medieval humanities’. ‘Byzantium’ was specifically mentioned because its history and art were a major interest of both the Blisses, and the small but choice collection of Byzantine art objects they had built up over the years was an essential element in their benefaction, together with the library of several thousand books they had put together on all aspects of Byzantine culture. Harvard was involved because it was Mr. Bliss’s own university and because he felt that a university connexion was essential to their new foundation. The deed of gift made it clear, however, that it was not to be part of Harvard as a teaching institution for undergraduates but was to be devoted to study, research and publication. It was envisaged that it would have a director - his cumbersome initial title was that of ‘Representative of the Administrative committee, in Residence’; he did not become ‘Director’ till 1947 – with a small permanent staff of established scholars and a succession of younger graduates who would receive short-term appointments to carry on their research. The organization and functioning of the new institutions were left to work themselves out by experiment under the guidance of the first ‘Director’, John (‘Jack’) S. Thacher (1904–82), who had been trained as an art historian and worked in the Fogg Art Museum at Harvard. As a family friend of the Blisses, of independent means and with very similar tastes, he could be trusted to carry out and interpret their wishes with regard to the development of the new foundation.

The Bliss’s Byzantine collection had consisted mainly of jewellery, icons and ivories, with only a few coins, but in 1947 nearly 200 coins were added as a gift of G. Rowland Shaw (1893–1963), a member of the diplomatic service who had spent much of his time in the Near and Middle East and was a friend of the Blisses. The collection effectively dates, however, from 1948, when Jack Thacher bought the coin cabinet of Hayford Peirce from his widow, for it consisted of some 4,300 coins, some 3,000 of them late Roman and Byzantine, so that it was one of the largest and most important collections of Byzantine coins then in existence.

Hayford Peirce (1883–1946) was a wealthy amateur, not a professional scholar, and was a close friend in the 1920s and 1930s of Royall Tyler (1884–1953), who had attended the Versailles Peace Congress as a member of General Pershing’s staff and had eventually settled in Europe, buying a house in France and becoming a financial adviser to the Hungarian government. Peirce and Tyler collaborated in an important monograph on Byzantine art that appeared in 1932; it was indeed Tyler who had first interested the Blisses in Byzantine art and culture. The Peirce collection was very varied in origin; some of it came from a well-known Istanbul antique- and coin-dealer, Andronikos, but most of it was acquired through the trade in western Europe and the United States. In the summer of 1948, after the collection had been transferred to Dumbarton Oaks, a young Princeton graduate in his early twenties, James Breckenridge (1926–[1982]), who had interested himself in the Byzantine coins in the Princeton Museum and in that of the American Numismatic Society, was invited there by Thacher and given the task of identifying and labelling the coins, which were in a state of considerable disorder, and arranging them in a newly-acquired coin safe.

The work was completed by the end of 1949, after which no one quite knew what to do next. Neither Breckenridge nor the Director of Studies, Bert Friend, had any guidance to give. Breckenridge, who stayed on at Dumbarton Oaks as a Graduate Fellow in 1949/50, had no wish to become a museum official, or even a numismatist, though he did produce an important monograph in 1959 on the coinage of Justinian II; he went on instead to a distinguished career as professor of art history at Northwestern University. Albert Friend ([1894]–1954) was a Princeton art historian who had a small collection of his own but was interested in coins purely as works of art. Acquisitions of the next few years were consequently limited to a few gold coins and medallions of outstanding quality, mainly of the Tetrarchy, which were bought from Brummer in New York or Edward Cans in California, as they were offered. No policy of systematic development was formulated at all.

In 1950 a new factor came into play, for Harvard received as a bequest, for the Fogg Art Museum, the coin collection of the archeologist and art historian Thomas Whittemore (1871–1950). This was even larger than the Peirce collection – some 3,400 Byzantine coins out of a total of 3,700, the rest late Roman (before 395) or Islamic – but was much more ramshackle. Whittemore was founder and director of the now defunct Byzantine Institute in Istanbul and Paris, and is best remembered for his work in uncovering and restoring the mosaics in Hagia Sophia. He also had the misfortune to meet Evelyn Waugh at Addis Ababa in 1930, on the occasion of the coronation of Haile Selassie, and in Waugh’s diary and one of his novels is the comic ‘American Professor W.’ whose knowledge of the Coptic liturgy, in which he was supposedly an expert, left much to be desired and whose pieties were often wide of the mark in an Ethiopian setting. He had apparently formed the collection mainly at Istanbul, buying indiscriminately such coins as attracted his attention in the Bazaar or elsewhere and never troubling to sort them or put them in any kind of order. But while the bulk of it consisted of these copper coins, often in poor condition, there were a number of gold and silver coins also, some no doubt acquired in Turkey or Greece but others at sales or from dealers in Europe. There were no records, unfortunately, of how the collection was put together, or with what object, or from where individual coins had come. Once it had reached the Fogg a Mr. Roland Gray, a friend of Whittemore’s, offered to put it in order and label the coins on a voluntary basis, but he abandoned the work very quickly – he was an old man in his eighties – and nothing more was done. The Director of the Fogg, Mr. John Coolidge, discussed with Jack Thacher the possibility of amalgamating the Whittemore and Peirce collections – it seemed absurd for ‘Harvard’ to have two great collections of Byzantine coins in different places – but the matter did not seem urgent and was put to one side.

It was at this point that I came into the picture. I have described in another section how I spent the second half of 1953 in New York as the guest of the ANS. I was, however, neither a Byzantine historian nor a Byzantine numismatist. My historical interests were firmly anchored in Western Europe. But I had read much of Gibbon at the age of seventeen, before going to university, in a three-volume edition in poor print picked up for a few shillings in Dublin, and I had early become an admirer of J. B. Bury, as much perhaps for his resolute rationalism as for his greatness as a historian. As an undergraduate I had replaced my three-volume Gibbon by Bury’s magnificent seven-volume edition, and I had chosen several of Bury’s books as College prizes, notably his 1923 History of the Later Roman Empire and his earlier (1912) History of the Eastern Roman Empire from the fall of Irene to the Accession of Basil I. The first of these is one of the few books I possess which have bindings that have disintegrated from overmuch use. So I was fairly well read in Byzantine history, though in no sense a specialist in it.

Again, although I was not specifically a Byzantine numismatist, over the years 1945–53 I had built up a substantial Byzantine series of some 600/700 coins in my collection, probably as much because they happened to be on the market as for any other reason. But some unusual types or legends amongst them had aroused my interest, and the ensuing research had given me a good knowledge of the literature and resulted in some significant discoveries. One was the dating system employed on solidi of Carthage under Maurice, Phocas and Heraclius, and, arising out of this, the first identification of the coins struck at a number of mints during the revolt of Heraclius in 608–10 and the explanation of their ‘consular’ types. I also identified a new mint of Heraclius, that of Seleucia in Isauria, and was able to tie it in with his preparations for the Persian War. The chance acquisition of a solidus of Justinian I with OBXX in the exergue instead of the customary CONOB had led me to collect material on the light-weight solidi of the sixth and seventh centuries, material that I was able to put at the disposal of one of the students for his paper at the ANS Summer seminar in 1953 and at that of Howard Adelson for his subsequent publication on the subject (1957). While in New York in the summer of 1953 I recognized the four splendid six-solidus medallions of Maurice on the Kyrenia Girdle in the Metropolitan Museum of Art as products of the imperial mint, not goldsmith’s imitations as had always been thought, and at the time of my return to England I was preparing the study of them that subsequently appeared in the Numismatic Chronicle. I was thus sufficiently well acquainted with Byzantine numismatics for Bellinger to think that my advice on the coins at Dumbarton Oaks might be useful, and in November I was invited to come and see what there was.

My chief memories of the visit (30 Nov. – 4 Dec.) are of a great deal of sightseeing in Washington, its high points being the Fragonards in the National Gallery and going up the Washington Monument on a cold, crisp winter day that allowed wonderful views over the city. But correspondence over the next few months with Jack Thacher shows that I had also given a great deal of thought to the coins. When I told him that I hoped to return to the ANS the following summer, he asked Bellinger and myself to consider spending a couple of weeks at Dumbarton Oaks and the Fogg Museum with a view to writing a report on the two collections for Coolidge and himself, with recommendations on how they could best be used for the furtherance of Byzantine studies. This in due course we did. I spent a week at Dumbarton Oaks at the end of August 1954 and a further week in September at Harvard, staying in the Faculty Club. Nothing could have been kinder than John Coolidge’s welcome, and the Museum made special arrangements with Security for me to come back after hours and work until midnight. But examining the coins was a horrendous experience, for they were housed in a tiny cramped space, not much larger than a closet, immediately under the roof, and I had to work on them in the appalling heat and humidity of a New England summer. Bellinger was with me part but not all of the time, and we had no problems in agreeing on the text of a joint report.

It was clear to both of us that any development would have to be on the Dumbarton Oaks side. Such use as could be made of the coins in the Fogg was limited by the terms of Whittemore’s will, which stipulated that his collection should be preserved as an entity, and in the Fogg: there was no possibility of adding to it, or of merging the two collections, or of disposing of unwanted coins, though duplicates could be deposited at Dumbarton Oaks on indefinite loan. With the Peirce coins, on the other hand, we had a free hand, for they had come by purchase, and Jack Thacher had told us in advance that money would be made available for expanding the collection if that was what was required. We therefore recommended that Dumbarton Oaks should limit itself to Byzantine coins, disposing of unwanted coins, mainly Roman ones, and duplicates, and on the basis of the Peirce collection building up that at Dumbarton Oaks into one of world-class quality. When the time seemed ripe it and the Whittemore collection should be published together in such a way as to make the joint catalogue a standard reference work on Byzantine numismatics.

The report was in due course accepted, and since that time my own scholarly career has been closely bound up with Dumbarton Oaks. Jack Thacher created for me the honorary post of Advisor in Byzantine Numismatics – the title of ‘Curator of the Byzantine coins’ did not seem appropriate since I would be non-resident most of the time – with the duty of expanding the collection, disposing of unwanted coins and duplicates, and ultimately, in collaboration with Bellinger, producing a catalogue. I would have an annual purchasing grant of $1000 to spend as I judged fit – the dollar was worth more in the 1950s than it is today – with an understanding that the Blisses would provide funds for more extensive acquisitions as need arose. I would also have any money raised by the sale of coins. I would receive no stipend, but would have an expense allowance to cover travel and living costs incurred on behalf of Dumbarton Oaks in attending sales or conferences, and I was free to come and stay at Dumbarton as often and for as long as I liked. Bellinger did not receive any corresponding title, and as he had no knowledge of the coin market or any interest in collecting he was not expected to play a role in building up the collection. But since, living at New Haven, he had much readier access to both collections than I had, he undertook to look after the Whittemore coins at Cambridge and to work on the Peirce ones in Georgetown as and when he could. We would be jointly responsible for the catalogue, settling between us in due course the details of how it was to be done.

There thus began a forty years’ association with Dumbarton Oaks that can be divided into roughly three periods: the years from 1955 to 1963, which were mainly devoted to building up the collection; those from 1963 to 1973, devoted to the preparation and publication of the first three volumes of the coin catalogue; and those from 1973 onwards, which saw extensive work on the fourth and fifth volumes but the completion of neither, though it did see other substantial publications in the Byzantine field by Michael Hendy and myself, including in my case the production of a catalogue of the late Roman coins at Dumbarton Oaks and in the Whittemore collection in association with Melinda Mays.


Dumbarton Oaks 1955-63

The Whittemore and Dumbarton Oaks Collections

The years between 1955, when I was appointed Advisor, and 1963, when serious work on writing the catalogue began, were devoted to building up the Dumbarton Oaks collection, and relatively little publication was undertaken. I spent some time at Dumbarton Oaks every year except 1956. Before I describe my work, I had better say something of what the place was like at that time, and who were there.

The main building of Dumbarton Oaks was appreciably smaller than that of today, since the Philip Johnson extension and the Garden Wing had not yet been erected and what is now the main gallery housing the Byzantine Collection was a courtyard with grass and shrubbery, open to the sky. The main entrance, the large doorway on 32nd Street, was as it is now, though the door opened inwards instead of outwards, but there was to its right a wide gateway, labelled 1701 32nd Street, giving onto a paved yard through which was the main access to the gardens and which had at its far end a doorway, now a window of the Receiving Room, by which one entered the main building at hours when the public was not admitted to the collections. In 1962–3, when the Garden Library was built, the 1701 32nd Street entrance disappeared, an extension of the entrance hallway, leading to the Garden Library on the right, taking its place. There is an illustration of the old entrance, before the changes of 1962–3, in the 1980 edition of Beatrix Farrand’s Plant Book, p. 123.

In the main building itself, the third floor was not yet part of the library; it was taken up by the professors’ studies, appallingly hot in the summer months because there was no air-conditioning. The coins were housed in Mrs Bland’s office near the entrance to the building, Mrs Bland being the Curator of the Byzantine Collection. The gardens were not greatly different from what they are today save that the Pebble Garden and the Box Ellipse did not exist, the space occupied by the first being a hard tennis-court, little used, and the Ellipse only starting in 1958 the series of transformations that were to create the enclosed space surrounding the large pool and fountain that one sees today. The entrance to the gardens was, as already described, from 32nd Street, and not through the much more impressive gates on R Street which, with the winding driveway up to the house, had been the main entrance in the Blisses’ time and is again the main entrance today.

The administrative staff was very small, consisting essentially of Mr Thacher, his secretary Alberta Carpenter, and the Financial Secretary, Thelma Hammack. I was naturally in regular contact with all of these. Mr Thacher was very much a “museum man” who regarded his main duty as being that of acquiring objects and putting them on display. He had no special interest in coins, though he had in medallions, for coins do not lend themselves to display. But he recognized that coins were of scholarly importance and that the presence of a great collection at Dumbarton Oaks would increase that institution’s reputation. I could therefore always count on him to support my purchasing forays in Europe and find the cash for them.

I also had a good deal to do with Alberta, who combined a rather harsh voice and a brusque manner with immense efficiency and unfailing helpfulness and a sense of humour. She took great trouble in arranging the travels and accommodation of visitors; once she even volunteered to recover some tax which she believed the ANS had unnecessarily paid on my behalf to the New York authorities, and succeeded in getting it back. Thelma, however, regarded herself as the guardian of Dumbarton Oaks finances, of which she thought Mr Thacher was not always sufficiently careful. She hated paying out money, and was always anxious to prevent visiting Fellows taking what she regarded as “liberties”. When the great Momigliano once asked if he might put up his daughter for a few days at the Fellows’ Building his request was rudely rejected – “Did he think the Fellows’ Building was a hotel?” – and he had to fall back on the hospitality of the Hellenic Center. She had no authority to act in this way, and would have been at once overruled if Momi had appealed to the Director, but her attitude was so intimidating that he had not the courage to do so. Once, when Herbert Cahn of Basel found that the exchange rate supplied by her Washington bank had left him out of pocket over some coins I had bought and sent a supplementary bill, she created such a scene, refusing to pay it, that I eventually paid it myself to stop her shouting at me.

The library, built up with extraordinary speed in the years after the War, was a pleasure to work in, though it suffered, and still suffers, from a clumsy system of classification save where the periodicals are concerned. These were arranged in alphabetical order of title, as they continue to be, and whenever I have worked for some months at Dumbarton Oaks I find it a shock to return to the Dewey or other system used elsewhere, though I recognize that an alphabetical arrangement is possible only in a specialized library, not in a general one. The librarian, Eleanor Rathbone, seemed to me lacking in imagination and too prone, except over the system of classification which she inherited and could not change, to accept the rules of the Library of Congress. Anything the Library of Congress would not keep, she would not keep either. I consequently had the mortification of discovering that the many offprints I had sent over a period of several years before 1955 were not available for me to consult, since instead of putting them in the library she had given them away to Anastos or to some other member of the academic staff. This was a particular shock to me after having worked for some months in the ANS library in New York, for one of the most valuable features of this were its rows of well arranged filing cabinets filled with properly catalogued offprints. A scholar accepts that a great general library cannot keep offprints, but a specialized one can and ought to do so. Miss Rathbone’s negative policy fortunately did not survive her departure, to another post elsewhere, in 1957.

Since much of my work during these years was concerned with the acquisition of coins, which could be more easily done from Cambridge than from Washington, my visits to Dumbarton Oaks were initially rather few. There was also a problem over timing, for it was only during the Long Vacation that I could be away from Cambridge for several months, or even several weeks, on end; other vacations I expected to spend partly with my mother in Ireland, partly in lecturing at Brussels, and partly in travel on the Continent, either for pleasure or on behalf of Dumbarton Oaks. In 1956 I did not go to America at all; much of my summer was in fact occupied with negotiating the purchase of the Bertelé collection. In 1957 and 1958 I was there for two months (11 July–10 September) and six weeks (5 July–15 August) respectively. Not till 1959/60 did I have a sabbatical, a year’s leave of absence from Cambridge, and was able to spend the fall and spring semesters at Dumbarton Oaks, essentially from September to March with a short break at Christmas when I was back in Europe. After that I returned for shorter visits, effectively for July and August, in 1961 and 1962.

How I was lodged on these trips to Washington would depend on the time of year I went. The possibilities were mainly three: in the Fellows’ Building, in the annexe of the latter known as the East Cottage, or in one of several apartments in buildings up Wisconsin Avenue that Dumbarton Oaks held on a regular basis and used for Fellows or occasional guests. I also stayed two or three times at the Cosmos Club as the guest of Mr Thacher, but never for more than a few days; it was sometimes necessary at the start or end of longer visits when the Fellows’ Building or the East Cottage was not available. While obviously much more comfortable than the others, and Mr Bellinger’s favourite port of call, I had no great liking for it; it required me to be more formally dressed than I was accustomed to being and was nearly half an hour’s walk away, a consideration when my time was limited and distinctly unpleasant if my visit happened to coincide with a Washington heat-wave. For the same reasons I did not enjoy the Wisconsin Avenue apartments, more especially since the twenty-minute walk to them was uphill, and in the heat of summer this had no attractions. I much preferred the Fellows’ Building or the East Cottage, both on S Street and on the Dumbarton Oaks estate, only a couple of hundred yards from the main building.

These facilities need to be explained, since the East Cottage no longer exists as such and the Fellows’ Building has changed substantially since the 1950s.

The Fellows’ Building is a large, rectangular, two-storey building which residents often assume is a converted barn or coach-house from the old estate but which in reality was erected in the 1920s to provide a covered space in which plans and large-scale models for the gardens could be erected and studied. East Cottage and West Cottage were built on to this as separate wings in the late 1930s, though neither is a ‘cottage’ in the usual sense of the word. They are small houses, in an attractive reddish brick which contrasted with the ugly, yellowish-grey stucco of the main building, though this was at least partly concealed by a giant wisteria that covered much of it. The ground floor of each formed a small, self-contained two- or three room residence having its own kitchen and bathroom; only the upper floor connected with the Fellows’ Building proper and formed part of this. The interior of the main building had been modified in the 1940s to provide downstairs a large living-room, a slightly smaller dining-room, and a kitchen, and upstairs to create bedrooms for five or six residents and a very inadequate communal bathroom. Since air-conditioning was still unusual in Washington in the 1950s, and even in the 1960s, and there would be no resident Fellows outside the periods of university semesters, the Fellows’ Building was closed each year from about July 1st to Labour Day, in the first week of September. When the building was open there was a small, non-resident staff responsible for maintenance and cooking, with three meals a day provided for residents and the lunch open to non-resident Fellows. The cottages on the other hand were available during the months when the Fellows’ Building was closed, but anyone living in them had to look after him- or herself.

West Cottage, the larger one with three rooms, was the residence of Professor Dvornik from the 1940s to his death in 1975. East Cottage, with two rooms, was where I could expect to live when I came in the summer, as I normally did between the late 1950s and 1981. It then ceased to exist as an entity as a result of the extensive reconstruction of the Fellows’ Building undertaken by Constable, but prior to this it had become so closely associated with my visits that Constable’s predecessor, Bill Tyler, used to refer to it jocularly as the Villa Grierson. Although the sitting-room in it was never air-conditioned and the bedroom only in the early 1970s, I did not mind the heat too much and preferred looking after myself there than having three large meals a day, particularly since it was the practice in the Fellows’ Building for the food not to be served at the table but piled on one’s plate in the kitchen by a staff whose members equated visitors’ appetites with their own. When I was living in the Fellows’ Building in the winter of 1959/60 I got a real scare, for when I went home at Xmas and weighed myself, I found that I had put on nearly a stone in three months, and become heavier than I had ever been before or have been since. I took it off after my return, but only by denying myself dinners altogether. I had great difficulty in persuading the kitchen staff that this abstinence was not a reflexion on their cooking but was simply the result of vanity, and necessary if I was to keep any figure at all.

I naturally was not working all the time I was there. Books in the Dumbarton Oaks library could not be taken out of the building, and in any case did not run to light reading, but Georgetown Public Library, at the corner of Wisconsin Avenue and R Street, was only a block away, and I went there two or three times a week to borrow books. It was not a particularly good library if one was looking for a specific book, for this would usually turn out to be already on loan or not there at all, but if one simply wanted good general reading it was all that could be desired, and lending conditions were generous. I made a practice of going there with some regularity on Wednesdays, for by that time it would have received the London Economist of the preceding week and this was my only way of keeping in touch with news of Europe, and more particularly of Great Britain, since the world outside the United States was rarely mentioned in American newspapers or in radio bulletins – there was as yet no television – at all. For many years I made a practice of taking the street-car once a week – buses only replaced street-cars in the 1960s – into Washington, and spending the evening at the National Gallery, which then stayed open until 9 o’clock and where supper at the excellent cafeteria would break the monotony of looking after myself. At that time the Gulbenkian pictures, with their wonderful Picassos of his Blue Period, were on indefinite loan to the gallery, and there were marvellous Fragonards, Bouchers and Davids in the French rooms immediately on the right after one went in. Or I visited other museums and galleries, though most of these were not open in the evenings and one had to make special expeditions to them at weekends. Friends would sometimes take me for drives in the country, or on full-day expeditions to Charlottesville or similar locations, or entertain me at their houses in the evenings. It was difficult in the sixties to repay such hospitality properly, for Georgetown was almost devoid of good restaurants. It is a far cry from the present position, when the neighbourhood is as well provided in this respect as any place in the world.

I turn now to my own work. Since much of this for the next twelve years was to be carried on in close association with Alfred Bellinger (1893–1978), I had better begin by saying something about this distinguished scholar, the quality and volume of whose work I admired almost equally and with whom I was to form a close friendship despite the difference in our ages. (He was seventeen years older than I was and immensely more experienced.) His early interests had been in English and American literature, in which both he and his wife Charlotte were immensely well read, able to complete quotations from memory and explain even the remotest literary allusion. But it was the Classics and not the English Faculty at Yale that he joined in 1920 and it was in it that he remained, from 1930 onwards as Lampson Professor of Latin, down to his retirement in 1962. He was extraordinarily productive as a scholar. ‘In the long series of L’Année Philologique’, one of his obituarists wrote ‘only 3 volumes between 1928 and 1975 do not include something from his hand, or a review of some work on which he had collaborated’. He was also extremely public-spirited, accepting membership of the governing bodies of many learned institutions of which, as a result of his manifest fair-mindedness and his ability to express himself concisely and clearly, he often in due course became chairman. He was also an ardent bird-watcher and a veritable encyclopaedia of information on birds and their habits.

Despite Alfred’s strong literary interests, his most substantial work was done in the fields of archaeology and numismatics. His interest in the latter was aroused during a year he spent at the American School at Athens in 1925/6. This resulted in a catalogue of the coins discovered at Corinth in 1925 and in a short brochure on the Byzantine Anonymous Folles of the tenth and eleventh centuries. For the latter, on the evidence of overstrikes, he for the first time established their correct order of issue. Although his work was subsequently improved upon by his close friend Margaret Thompson, who had the even more abundant material from the Agora excavations on which to base herself and whose lettering of the classes is that generally used today, Bellinger deserves the credit for having made the first break-through in solving the many problems involved. He went on to publish the coins from a series of excavations in the Near East – Jerash (1938), Dura-Europos (1930, 1932, 1935), and finally Troy (1961) – in catalogues which established a new standard in the publication of excavation coins everywhere.

Alfred’s connexion with Dumbarton Oaks came about originally through his sister Louisa, an expert on ancient textiles who worked in the Textile Museum in Washington and was for a time temporarily on the Dumbarton Oaks staff, assisting in a project for a census of ancient textiles in American museums that had much interested Mr Bliss since the late 1930s. In 1947, when the purchase of the Peirce coins was being considered, she suggested that her brother’s advice might be useful. He visited Dumbarton Oaks twice in 1948 to look at the coins and put forward the name of Dorothy Cox, a well-established numismatist and an old friend of his from his Corinth days, as a possible curator. Nothing came of this proposal, but in [1959] he became a member of the Board of Scholars. He resigned in [1966], on the ground that some rotation of membership was desirable, but was recalled in [1972] and in fact, from 1948 to the late 1960s, he was closely involved in Dumbarton Oaks’ affairs.

Alfred’s contributions to our joint project over the years 1955–63 were more diversified than mine. He mainly concerned himself with the Whittemore coins. We had already discovered in 1954 that Roland Gray’s ‘identifications’ and labelling would have to be scrapped. Gray’s procedure had been that of finding the coin in Wroth’s plates that most closely resembled each Whittemore specimen and then writing on the coin ticket the page and number of the coin illustrated, even if it was of a different year and officina from the Whittemore one, and even sometimes of a different mint. The resulting identifications were normally incorrect. But the young Speros Vryonis (1928–) who had spent the summer seminar of 1953 at the ANS and was then at the start of what was to be an extremely distinguished career in the field of Byzantine and Near Eastern scholarship, was roped in already in 1955 to put the coins in order, and to arrange for the photographing and listing of the gold and silver so that we would have the folder of these for reference purposes at Dumbarton Oaks and I would have photos available to me in Cambridge. The Whittemore collection provided material in these years for two important articles. One, contributed by Bellinger to the Centennial Publication of the American Numismatic Society (1958), published a hoard of silver coins of Theodore I of Nicaea which was the first to bring to light the existence of coins on which the right and left sides were sometimes struck by dies that did not precisely match each other, an observation that in due course led Sellwood and Bendall to the discovery of an incongruous and quite unsuspected feature of Byzantine minting technique. Vryonis for his part was able to describe an Isaurian gold coin from Lippesi in Attica in the Melanges published in 1963 in honour of the great Byzantinist George Ostrogorsky. Alfred likewise managed in these years to do some valuable work on the Dumbarton Oaks collection. We had already decided that while the ‘Byzantine Empire’ could be dated back to Constantine, we could not start our catalogue with him, since for the fourth century we could not hope to build up a collection at Dumbarton Oaks that would match those in the long-established national coin cabinets at London, Paris and Vienna. But the gold and silver holdings in the period from the Tetrarchy onwards were important, and in 1964 Alfred, with the help and advice of Patrick Bruun, Humphrey Sutherland, and John Kent, the three scholars then engaged in producing the RIG volumes for the period, published a catalogue of these (in DOP 18, 161–236), from the accession of Diocletian through Eugenius. He had earlier (1958) published the medallions, mainly gold and silver, in the collection (DOP 12, 125–56).

While Alfred was publishing coins from the Whittemore collection, I myself was concerned with expanding that at Dumbarton Oaks and putting it in order. In the year between 1955 and 1960 we acquired three major collections, in whole or in part, and a substantial number of individual coins from dealers or in European auction sales. These latter acquisitions were for the most part rarities and fairly expensive. Nor did they always come easily. One collector, Enrico Leuthold, a wealthy Swiss businessman at Milan, was often a competitor, especially in Italian and Swiss sales; in a Ratto sale in January 1956 Dumbarton Oaks was outbid by him on eleven lots, and we only came away with a single coin on which my agent had induced him to desist. But over the next few years there were few sales containing Byzantine coins at which I did not manage to buy something, and I was of course often offered coins by dealers once my interests became known. The most spectacular acquisitions of these years were perhaps a gold histamenon of the six-week joint reign of the empresses Zoe and Theodore in 1042 (DOC III. 732, no. 1), bought from Hecht in Rome in 1956, and a unique double solidus (medallion) of Theodosius II (DOLRC, no. 377). The second of these I bought from Cahn in 1958 on behalf of a group of friends of Mr and Mrs Bliss as a gift to commemorate their fiftieth wedding anniversary, and it came to the collection, as a bequest of Mrs Bliss, only after her death.

When bidding at auctions I normally gave commissions to dealers, only attending the sales myself when coins of particular importance were involved or where I felt the need to examine items in advance and assure myself of their condition or authenticity. Also, if the purchases were likely to run to three figures, my travel and lodging expenses would be inferior to the commissions incurred by bidding through a dealer. If an auctioning house was also a coin dealer, as is often the case, one could profitably spend the time when one need not be present in going over his stock in the hope of finding other items of interest. If one is present at an auction in person it is also sometimes possible to guess which other participants might be interested in the lots one hopes to buy oneself, and arrange not to bid against each other. It is often thought that such arrangements are illegal, but this is not the case. What is illegal is for a group of purchasers to refrain from bidding on a number of lots, so as to keep their prices down, and then to reauction the coins between themselves afterwards.

The three major collections bought in these years were my own Byzantine series (1956), the Bertelé collection (1956, I960), and the Schindler collection (1960).

The first was the least important of the three. In the seven years since I had started collecting I had built up a good holding of about a thousand late Roman and Byzantine coins. The first gold coin I ever possessed was a rare Thessalonican solidus of Arcadius (now LRC 61) which I bought from Spink’s in February 1945. A high proportion of the coins came from dealers’ stocks in London and ultimately from the immense Lord Grantley collection (see MEC 1, pp. 316–7), which was dispersed in eleven Glendining sales over the years 1943–5, effectively before I had started collecting myself. In 1955, when I was appointed Advisor at Dumbarton Oaks, I decided that in order to avoid a conflict of interest I would cease collecting Byzantine coins myself and in due course sell Dumbarton Oaks, at an independent valuation, what it lacked from my own collection. “Due course” arrived very quickly, in the winter of 1955/6, when I learnt that the European medieval coins in the R. C. Lockett collection would be sold at Glendining’s on 29 February 1956. For my purchases at this I would need all the funds I could muster. Thacher told me in January that the Blisses would make available up to £2,500 for the acquisition of my coins. The price finally agreed upon was a little above this, but the difference could easily come out of my regular purchasing fund. Although the transaction was not completed till March and the coins (530, including 120 in gold, for which a special Board of Trade export license was required) did not reach Washington till April, the firm of Baldwin’s, which made the valuation and was responsible for actually sending the coins, allowed me the credit necessary for my acquisitions at the Lockett sale when it occurred. I kept back a few coins, mainly copper ones of the fourth and fifth centuries, for further study, but never made anything of them and presented some of them to Dumbarton Oaks in 1971 and the rest in 1986 so that they could be included in Melinda Mays’ and my late Roman catalogue.

The second collection, much the most important of the three, was that of the Italian diplomat and scholar Tommaso Bertelé (1892–1971), who was in his late fifties when I first got to know him. After studying at Florence he had entered the Italian diplomatic service in his twenties and was a member of the Italian delegation to the Versailles Peace Conference after World War I. Most of his diplomatic postings were in South America, but in the early 1920s he spent some years in Turkey, at Istanbul or Ankara, and in 1938 he was Italian Consul General in Sarajevo. His diplomatic career ended in 1945, I presume because he was thought to have been too closely connected with the previous regime, and he was left with leisure to pursue his scholarly and antiquarian interests. When I first knew him he was living in Rome but spending part of each year in Verona, in the region where he was born and to which he moved definitively in 1955. He died in 1971.

Bertelé’s numismatic interests went back to his days in Istanbul, when he exercised his general antiquarian tastes by collecting material for the substantial monograph he published in 1932 on the palace of the Venetian ambassadors and began to buy Byzantine coins in the Bazaar. After he moved to Berlin in 1926 he met the German numismatist Kurt Regling (1876–1935), Director of the Münzsammlung in the Staatliche Museen, who encouraged him to publish his first numismatic article, in the Berlin Zeitschrift für Numismatik for 1926. It catalogued a large selection of rare or entirely new Byzantine coins in his collection of all periods, but he subsequently came to specialize in the little known coinages of the last centuries of the Empire. These were series poorly represented in the coin cabinets of western Europe and America, and, once his interest in them became known to dealers he was eventually able to build up the most representative collection of them in the world. From 1926 onwards he had a long list of numismatic publications to his credit. The most important of these were the catalogue of a hoard of silver coins of the minority of John V struck during the regency of Anna of Savoy (1937) and a study of the extraordinary iconographic design of winged emperors on thirteenth- and-fourteenth-century coins (1951). But these were only two amongst many. He made a serious attempt to collect and identify the abundant references to Byzantine coins in commercial documents, particularly Venetian ones, of the thirteenth, fourteenth and fifteenth centuries, and he was one of the first who took seriously the need to study the finenesses of the coins as a way of identifying their denominations and understanding the progress of debasement at certain periods.

Bertelé’s collection, at the end of the 1940s, consisted of some 8,000 coins, the bulk being made up of copper folles of the tenth and eleventh centuries and hoards of black billon or copper trachea of the twelfth and thirteenth, together with a substantial number of early Palaeologan copper of the late thirteenth and early fourteenth centuries. To these he subsequently added several Palaeologan hoards bought complete, or which he believed to be complete, one mainly of gold hyperpyra of Andronicus II and the other one of silver stavrata of Manuel II and John VIII. In its final count the collection amounted to some 10,000 coins. Although there were in it huge numbers of duplicates, often of very little scientific value because of their poor condition, and many coins that would duplicate ones already in Dumbarton Oaks, its large number of Palaeologan coins, many of them virtually unknown, made its acquisition by us a matter of capital importance.

From about 1950 onwards Bertelé was on the look-out for a possible purchaser, but he was determined to sell it in bloc and anxious that it should go to some learned institution, and not to some private collector like Leuthold at Milan, who could certainly have afforded it and would have bought it if he was given the chance. Several Italian museums were approached in the early 1950s, but either showed no interest – the Vatican collection, for example, was essentially limited to papal coins – or could not raise the necessary funds. In 1952 he was in contact with the Blisses’s friend Royall Tyler about a possible sale to Dumbarton Oaks, but how this had come about I have failed to discover. Negotiations had gone far enough, however, for a rough inventory of the collection to be made and a price of 120,000 Swiss francs agreed upon. Since the collection was not one of Italian coins it did not come under the restrictions imposed on the export of works of art: Bertelé was free to sell it in Italy or abroad as he pleased. But Tyler’s death in March 1953 brought the projected transaction to a halt.

This was the situation with which I was presented in 1955. Bertelé and I had known each other by correspondence since 1947, when he had written to me asking for information about a politikon coin which he had heard was in my collection. We first actually met at Verona in September 1948, and over the next few years we saw each other from time to time when I was in Italy, besides corresponding frequently on problems of common interest. It was only in 1955, however, that I learnt about his wish to dispose of his coins, and the ensuing negotiations for their sale were to absorb much of my time and patience over the next two years.

Bertelé was not easy to deal with. It was not that he was avaricious or trying to get the better of Dumbarton Oaks financially; we quite easily came to terms over a revised valuation of 125,000 Swiss francs, the increase taking account of the additions he had made to the collection since 1952 and the slight rise in coin prices in the interval. But he was vain and inordinately talkative. He insisted on explaining everything at quite excessive length, and I wearied of being asked to admire every unlovely Palaeologan copper coin put in front of me and required to lavish praise on its owner’s extraordinary numismatic talents, which were indeed very considerable, in identifying and describing it. He also wanted to impose conditions that Dumbarton Oaks could not accept. The most important of these was that while Dumbarton Oaks should pay for the whole collection at once and receive the bulk of it immediately - all the earlier coins and such series as he had already published, essentially the coins of his article of 1926, those in the Anna of Savoy hoard, and those of his "winged emperor" monograph - he should retain custody for several years of most of the Comnenian, Nicene and Palaeologan coins to allow him to study them further.

From Bellinger’s and my standpoint, so far as the catalogue was concerned, this would present no problem, for it would be some years before we would arrive at the late Byzantine period, and the more work Bertelé did on this the easier our task would eventually be. But for Mr Thacher it was quite unacceptable: we should pay only for what we received, and without the Palaeologan coins the collection would not greatly interest us. Various possibilities were considered, such as buying the collection as a whole but putting part of the sum into escrow, to be paid only when the remaining coins were finally handed over. By the end of the year we had reached virtual agreement, but our legal advisers found the contract unsatisfactory and Bertelé was so enraged at the difficulties they raised that the whole deal nearly collapsed.  In the end, in a roundabout fashion and through the good offices of Herbert Cahn of Münzen und Medaillen A.G. Basel, we had to content ourselves with buying and paying for only that part of the collection which Bertelé was prepared to release at once, some 6,000 coins, and receiving an undertaking from him to make the rest available some years later. This explains why the coins are described, in the early catalogue volumes, as coming from a Swiss collection, for Bertelé was anxious that for a few years – he suggested ten – it should not become generally known that he had disposed of his coins, though, having a justifiable pride in his collection, he was anxious that his previous ownership should be recognized later. The coins to be immediately acquired by Dumbarton Oaks were in fact delivered in Basel on July 18th 1956 and reached Georgetown in September.

The remainder of the collection came to Dumbarton Oaks in 1960. Bertelé wrote to me in December 1959 to say that he was now ready to let us have it. Negotiations were slow, for I was spending the winter in America and letters had to go back and forth, and were sometimes acrimonious, for coin prices had gone up in the intervening five years and Bertelé’s acquisitions of the same period had to be evaluated. Only after my return to Europe were we able, at a meeting on 13 April at Verona, to reach agreement on a price of 82,000 Swiss francs, slightly above the 75,000 proposed in March. The coins were handed over to Cahn during the summer and reached Dumbarton Oaks in September. Even this was not quite the end of the matter, for I had agreed in April that he might hold back a substantial number of tiny silver coins, one-eighth stavrata, mainly of John VIII, which he said he still wished to study and which I judged to be of minor importance. We were so pleased with what we had got that I confess I forgot about them entirely. It was only many years after Bertelé’s death that they resurfaced through his son selling them to Simon Bendall, and then only through the latter’s generosity, and that of Cécile Morrisson to whom some of them had been promised, that they were finally able to rejoin the bulk of the Bertelé collection at Dumbarton Oaks in 1990.

The fate of Bertelé’s copious notes and photographs on Palaeologan coins was also naturally of interest to Dumbarton Oaks, for he had amassed material from a great number of sources. This was put together in usable form by his son after his death, and bound photocopies of the resulting volume of nearly 200 pages were most generously presented in 1973/4 to a number of public collections – the British Museum, the Cabinet des Médailles, the Hermitage, the Münzsammlung at Berlin – and to a few individual scholars such as Cécile Morrisson and myself. Much use was made of this Dossier by Cécile Morrisson when in 1978 she published a French translation, expanded to nearly 200 pages and well provided with illustrations, of a 76-page survey of Byzantine numismatics which Bertelé had contributed to the Rivista italiana di numismatica in 1964.

The third collection to be bought, happily involving fewer problems, was that of Captain Leo Schindler (1885–1957), a well-known Austrian collector and a friend of Bertelé, who wrote his obituary notice for the Rivista italiana di Numismatica. The collection totalled nearly 2,500 coins, about 150 of them gold and most of the rest copper, especially ones of the sixth and seventh centuries in extremely fine condition that were to fill many gaps in the date- and officina-series in our trays. Schindler had collected from his boyhood and no provenances for his earliest coins were recorded, but for his later ones, purchased locally or acquired by exchange from other collectors, he had made careful notes of where they came from. In the 1940s and 1950s he was very active in the collecting world of Vienna, with easy access to the Münzkabinett, and without being a great scholar he contributed several useful articles to the Numismatische Zeitschrift and the Mitteilungen of the local numismatic society, mainly making minor corrections to accepted attributions or classifications. At the time of his death he was planning a general work on Byzantine coinage, and the manuscript of the beautifully drawn illustrations he intended for this is now at Dumbarton Oaks, having been acquired with his collection. I never knew Schindler personally, though we had occasionally corresponded in the early 1950s.

The collection was inherited by his widow, but her ownership was contested by his son by a first marriage, who claimed it as a family heirloom. The claim was rejected by the courts, but his widow, alarmed that the publicity the case had created might attract the attention of thieves, was anxious to dispose of it, if possible as a single entity. Bertelé suggested to me in April 1960, when I saw him at Verona, that I get in touch with her, and in July I went to Vienna for a few days to examine the collection. It was quickly clear that its acquisition was desirable, and we agreed on a price of 150,000 Austrian schillings, or just under 6,000 dollars. I did not know any coin-dealer in Vienna with whom to arrange its shipment. Since Probszt at Graz had been trying to buy it for one of his clients I was reluctant to approach him, but Siegmund Werkner, a pleasant Hungarian who had settled at Innsbruck after the War and founded the Tiroler Münzhandlung, a firm with which I had sometimes dealt, was willing to undertake the task.

The only problem was that of securing an export permit, which would require the consent of the Münzkabinett. The Director, Eduard Holzmair (1902–71), recognized that there could be no reasonable grounds for stopping it, for the coins were not Austrian and the collection was not in itself one of great significance. But the Deputy Keeper Bernhard Koch (1920–44), who later himself was to become Director (1967–85), felt that some show of interest was required lest Schindler’s son should start a campaign against the authorities for letting it go too easily. Koch and I finally agreed, with the consent of Thacher, that Dumbarton Oaks should buy the collection in its entirety but would donate a selection of coins from it to the Münzkabinett as a token of our appreciation of its raising no objection to the export. It was not difficult for us to select some fifty coins which were duplicates of ones at Dumbarton Oaks but would fill gaps in the museum’s holdings, whereupon the export permit was granted and the collection reached Washington by the end of the year.

Over the same time as these extensive acquisitions were being made, the collection was, in accordance with our recommendations of 1955, being in other respects trimmed down. Initially this was merely a matter of selling Peirce’s non-Byzantine coins, mainly pre-Tetrarchic Roman ones, though there were some medieval European ones and even a few totally eccentric intrusions, the most remarkable being a group, presumably a hoard, of some twenty electrum coins of Lesbos of the fifth century B.C. The coins disposed of were sometimes sold privately, or went to dealers; the more valuable ones were put up to auction. The Lesbos coins were bought by the firm of Münzen und Medaillen Basel. A number of Roman ones went to the ANS, and I was myself allowed to buy such medieval European ones as I wanted at an independent valuation. Many Byzantine duplicates, for of course after the arrival of the Bertelé and Schindler coins there were many of these, were included anonymously in a Glendining auction of [date unrecorded]. Late in 1960 Mrs Bland reckoned that we had sold coins to the value of [valuation unrecorded], though much of this had been used up on individual acquisitions of the previous few years. Usually one or two coins in each series disposed of were retained, in case they might be useful for comparative purposes.

Much of Bellinger’s and my time over these years was spent in identifying and ticketing our new acquisitions, and getting the collection in order prior to its cataloguing. Although the handwriting on the coin boxes shows that most of it, up to 1081, was done by either Bellinger or myself, several younger scholars were brought in from time to time to assist in various ways. Galavaris assisted in sorting the Bertelé coins in 1957, Dikigoropoulos spent nearly two years in 1961/3 in producing a card index of the coins, Hendy identified and labelled the Comnenian and later coins in 1963, and Berghaus photographed a high proportion of the coins on a two-week visit in 1964.

The first of these scholars, George P. Galavaris (1926–[2003]) was Greek by birth and took his first degree at Athens before going on to work for a doctorate at Princeton. He came to Dumbarton Oaks in 1957 as a junior fellow in art history, with a special interest in the iconography of the Virgin, and as a holiday task he helped with the Bertelé coins and contributed a valuable article on imperial costume and symbolism on the coins to Museum Notes in 1958. This identified as far as possible the robes worn by the emperors and the insignia they carried, so that scholars from the 1960s onwards have been able to describe Byzantine coin types much more precisely than had been possible for Wroth and earlier scholars. His subsequent career has been that of a professor of art history at McGill University, Montreal, and he has to his credit a long series of distinguished publications in the field of Byzantine art.

The next scholar, Andreas I. Dikigoropoulos (1924–) from Cyprus, generally known as Diki, came to Dumbarton Oaks for the best part of two years in 1961/3. He had taken his B.A. at Oxford (Lincoln College) in 1950 – he went on to his M.A. and eventually to a D.Phil, in Byzantine history (1961) – and had a post in the Department of Antiquities at Nicosia, where he was primarily concerned with the Byzantine coins turned up in great numbers in hoards and at many sites in the island. I first met him in 1951 at the Eighth International Byzantine Congress at Palermo, and we saw each other again at the Thessaloniki Congress in 1953. Over the next few years we were in frequent correspondence over the attribution of coins of the seventh century and the interpretation of the countermarks sometimes found on them, but at the end of the decade, after the creation of the Republic of Cyprus in 1960, his position in the Department was made impossible – his outspokenly pro-British views were held against him – and he was forced to resign his post (August 1961). He was anxious to obtain a year’s Fellowship at Dumbarton Oaks to write up the excavation material he had accumulated over the years. But Bellinger suggested instead that we might formally employ him, as a temporary member of the staff, in writing up coin descriptions and making a card index of the coins, following a pattern Bellinger devised, with printed compartments for descriptions, coin legends, die-axes, references, and the like.

This was eventually approved, and in late 1961 Diki and his family settled in Georgetown, where he worked at Dumbarton Oaks for the best part of two years, to the summer of 1963. Unfortunately he flatly refused to have anything to do with the Comnenian and Palaeologan periods, where his work would have rendered a much greater service, on the ground that we had not put these coins in order and he knew nothing about them. He subsequently returned to England, where he completed his legal education and was called to the Bar. In 1963 he considered emigrating to Australia and was I believe short-listed for teaching posts at the National University of Australia at Canberra and the University of Sydney, but neither of these eventually materialized and in the end he returned to Cyprus, where he practised as a barrister in Nicosia down to his retirement in the 1980s. He was a pleasant colleague and I am sorry that we have subsequently lost touch with each other: I last heard from him in 1978.

The third scholar brought in to help with putting the collection in order was Michael F. Hendy (1942 [–2008]), who was in due course to play a prominent part in the making of the catalogue and become an outstanding figure in the roll of Byzantine scholars. I first met him in about 1962, when he was still an undergraduate at The Queen’s College, Oxford, and had come over to Cambridge to look at the Byzantine coins in the Fitzwilliam Museum. Such an unusual interest led naturally to our keeping in touch, and in the summer of 1963, after he had taken his degree, I was able to arrange for him to spend some months at Dumbarton Oaks, identifying and labelling the Comnenian and later coins in which Diki had refused to interest himself. In 1964/5 he spent seven months in Bulgaria, as a British Council Graduate Exchange Scholar at the University of Sofia and devoting himself to study of Byzantine coin hoards, especially those of the twelfth and thirteenth centuries, at Sofia and in various provincial museums. This made possible his return to Dumbarton Oaks in 1965 as a Junior Fellow, when he stayed for two years and wrote a massive work on the coinage of the twelfth and thirteenth centuries that will be described later.

Finally, a more distinguished scholar than any of these worked briefly in the collection in 1964: Peter Berghaus (1919–[2012]), one of the greatest numismatists of this century and one of my oldest friends. Despite his difficulties in holding a camera – he tragically lost his left hand on almost the last day of World War II – he is a superb coin photographer, and in August 1964, after the ANS summer seminar which he was directing had come to an end, he spent a fortnight at Dumbarton Oaks, at Bellinger’s invitation, photographing as much as possible of our collection. Our own photographic department had for years been producing photos for us of the highest quality, but its procedure was slow – five coins on each 8" x 5" print – and its services had to be shared with other departments, so that Bellinger was always afraid that photos of particular coins would not be available when needed. Since Berghaus could produce enlarged 33 mm prints of splendid quality holding 10/15 coins, he felt that this might solve a problem which I myself felt to be largely imaginary. Working together, with Bellinger or myself laying out the coins and Peter taking the photographs, we got through all the gold and silver in the collection in the time available, together with a selection of the copper. It was an extraordinary achievement, from which we have profited ever since. The negatives were developed and printed in Germany and the prints sent to Washington by the end of the year, the negatives following rather later since Peter wanted to make slides from some of them. So there are today at Dumbarton Oaks two photographic files of our coins, and it is normally impossible to tell from the plates of the catalogues which of them has been used in making these up. Berghaus’ company was a source of enormous pleasure to both Alfred and myself, and after he and I returned to Europe we spent a few days together in West Berlin, crossing to East Berlin every day and working on the coins in the Münzkabinett of the Staatliche Museen.

Dumbarton Oaks Catalogues I – III

The actual writing of the coin catalogues began in the late summer of 1963, when Alfred Bellinger and Charlotte settled in as residents of the house on S Street, on Dumbarton Oaks property and only a hundred yards from the Main Building, which had been left vacant by the retirement of Sirarpie Der Nersessian. Sirarpie had been living there with her sister Arax since the early forties, and they both now returned to Paris, where they had kept on an apartment since the 1920s and where they were both to spend their remaining years. Alfred had originally planned to start full-time in 1962, when he retired from his Chair at Yale, but instead he took a temporary post at the University of Cincinnati and continued teaching for a further year. He and Charlotte were to stay in S Street for the next three years, during which time he wrote virtually the whole of DOC I and first drafts of the coin catalogues proper for DOC II and III. His custom was to write everything longhand – he had a beautiful and very legible hand – and they were typed up by two secretaries in turn, Julian Hartzell, a tall, good-looking and good-humoured young man who was there from the summer of 1963 to that of 1964 and again for a time in 1965, and Julia Cardozo, equally remarkable for her good looks and a granddaughter of the well-known Justice of the Supreme Court. She was appointed in September 1964 and remained till April 1967, when she left to get married and became Mrs Eisendrath.

Julian Hartzell and Julia Cardozo were in effect to act as copy-editors of DOC I, preparing the text of the volume for press under the guidance of the head of the Editorial Department, Julia Warner. Julia Cardozo was also responsible for the plates of DOC I. Unlike those of later volumes these were mainly prepared from plaster casts, and the copper coins entirely so, for it was at that time taken for granted that direct photography could not give satisfactory results for copper. Charlotte Bellinger was in any case available for making casts. This was an art which she had mastered when staying at Athens in the late 1920s, when she prepared the plates for Alfred’s Corinth volume and in due course did so for most of his other publications. She was also an expert in the cleaning of copper coins, but while this had been essential for the excavation finds at Athens and Corinth it was only occasionally required at Dumbarton Oaks, since any cleaning the coins needed had long since been done.

Alfred’s move to Dumbarton Oaks, most fortunately for him, coincided with the opening for the first time of a proper Coin Room, so that he and I, and any assistants we might have, would no longer have to work in a corner of Mrs Bland’s office upstairs. The new room was in the north­western corner of the basement of the Philip Johnson Building, constructed over the years 1961–3 to house the Pre-Columbian collection bequeathed by Mr Bliss to Dumbarton Oaks and opened to the public in December 1963. The new locale, which provided ample space for safe and bookcases, and for desks and tables for two or even three persons to work comfortably, remained the Coin Room till 19[?], when it was taken over by the rapidly expanding pre-Columbian Department next door. A new Coin Room, shared with textiles, which is still in use at the time of writing, was set up in the space formerly occupied by the Visual Aids Department, which with its collection of photographs and slides was transferred to a different part of the building. The first move, from Mrs Bland’s office to the basement of the Philip Johnson building, was not easy logistically, since it involved lowering the two coin safes down a flight of steep and narrow stairs, and while every endeavour was made to cover each tray with padding in order to ensure that coins did not slip out of one box into another, some of them did so and there had to be much checking afterwards. Soon after the move a new safe was acquired and the earliest one, which went back to 1948, and was no more fire- or burglar-proof than an ordinary filing-cabinet would be, was discarded.

DOC I differs in one conspicuous respect from its two successors, the absence of a lengthy introduction. Alfred’s concept of what our catalogue should be differed in fact from my own, and since each of us failed to convince the other we each agreed to go our own way. He conceived of the volume as no more than a catalogue, with a precise description of the coins, the arrangement following Wroth’s British Museum Catalogue lay-out and classification save where the latter had been shown to be incorrect. He saw no need to change Wroth’s terminology in the identification of denominations or the description of coin types and legends, or any need to explain these; such things should be left for professional art historians and other better qualified scholars to deal with. I felt, on the other hand, that the two Harvard collections gave us a mass of material much larger than any previous scholar had had at his disposal, and that we should use it as an opportunity for taking a new look at the coinage as a whole and asking how far, or for how long, the old categories and the old terminology, borrowed from those of Roman numismatics, held good. Our divergent views were in part the result of our different ages.  Bellinger was not in the least unreceptive to new ideas – he remained remarkably young in mind – but he was in his sixties and felt that my type of catalogue could not be completed in his lifetime even if it could be in mine. He was also not a Byzantinist. Although he was interested in Constantine the Great and the fourth century, his knowledge of later centuries was limited. He had indeed made a fundamental contribution to Byzantine numismatics through his work on the Anonymous Folles, and he was to publish a number of important Comnenian and Nicene hoards, but his true expertise and interests lay in the history and coinages of the Hellenistic period and the provincial coinages of the Roman Principate. I was equally not a Byzantinist, but as a medievalist I knew something of the resources of the surviving literature and how this could be used. I was also more conscious than was Bellinger of Wroth’s inadequacies. Wroth’s catalogue was a magnificent work, but his main interests had been in Greek and Parthian coins and in English medals, not in the coinage of Byzantium.

We consequently did far too little preliminary work before the writing of the catalogue began. Alfred produced a number of jottings on particular points – he used jestingly to refer to them as his "great thoughts" – and embodied them in two articles in Museum Notes in 1966 and 1967, but they are for the most part trivial in character. I was supposed to be writing a companion volume to DOC I on Byzantine coinage in the sixth century, and correspondence and memoranda of the early sixties make frequent reference to this ‘forthcoming’ work. Alfred, however, found me far too dilatory and apt to spend my time in following up interesting points instead of reaching conclusions that he could use. He was consequently not prepared to wait for a finished volume, though he did read anything I did complete and his "great thoughts" were often comments on my views. Some of his improvements on Wroth go back to my draft, notably the attribution of a coinage to Salona, the transfer of Tiberius II’s supposed copper coinage of Antioch of Years I–III to Maurice – the need for this was recognized independently by Kent in an article in the Chronicle in 1959 – and, following an insight of Schindler, the transfer to Nicomedia and Cyzicus of the half-folles of the same ruler which had customarily all been placed under Constantinople. But some of the more unfortunate errors in DOC I go back to the same draft, notably the attribution of solidi with reverse legends ending ΦS to Antioch (Theoupolis), and the ascription of other coins to purely imaginary mints at Perugia and Constantine in Numidia despite the fact that in the two latter cases, as he himself admitted, I had subsequently abandoned such ideas. My work in any case left much to be desired; I notably failed to anticipate Hahn’s discovery of a five-yearly rhythm in minting except at Antioch. What I wrote, in a text long since scrapped, was rendered superfluous by the first two volumes of Wolfgang Hahn’s Moneta Imperii Byzantini, published in 1973 and 1975 respectively. Alfred finished the text of DOC I in the spring of 1965 after we had made a last-minute decision, because of the size of the volume, to end it with the death of Maurice in 602 and not with that of Phocas in 610. Text and plates were sent to their respective printers in April. Proofs began to arrive in November, and the volume was published in the fall of 1966.

While DOC I was being written, a work which in its way was more important than any of our catalogues was being produced by Michael Hendy, who by now had taken his degree at Oxford and returned to Dumbarton in the summer of 1965 as a Junior Fellow to work on the coinage of the Comnenian and Nicene periods. The book he published on the subject, Coinage and Money in the Byzantine Empire 1081-1261, was effectively completed by the end of 1966, but he stayed on for a further year to produce a formal catalogue of our coins of the same period. Then he left us. In the spring of 1967 he applied for an Assistant Keepership in the Coin Room at the Fitzwilliam Museum in Cambridge, a position which he duly obtained, taking up his appointment in October 1967. He held the post for five years. In 1972 he moved to the University of Birmingham as Lecturer in Numismatics and Curator of the Coin Collection, which was effectively a Roman and Byzantine one, in the Barber Institute of Fine Arts. His book was accepted for publication by Dumbarton Oaks in 1967 and published in 1969 as Dumbarton Oaks Studies, No. 12.

Hendy’s volume represented essentially the kind of preliminary work that should have preceded DOC I–III, though the relative simplicity of earlier Byzantine monetary systems made the need for such an introduction much less apparent. Prior to the reign of Alexius I (1081–1118) there are distinct and easily identifiable denominations in gold, silver and copper, and problems of dating, mint attribution, relationships between coins of different metals, and so forth can in consequence be discussed in a well defined numismatic framework. This is not the case from the accession of Alexius onwards, after the spread of the concave form from gold to coins of other metals. While there are some concave coins which are clearly gold, and others that are flat and clearly copper, the bulk of the numismatic material consists of concave coins of poor quality metal, apparently mainly copper, that seemed to have come into existence through debasement of the gold. Wroth in his catalogue had done no more than list and describe the coins of each emperor according to their types, admitting that he had no idea what they were called, what their values were, or how they could be satisfactorily arranged. Economic historians, faced with an apparently unorganized mass of coins of base metal, interpreted the Comnenian period as one of monetary disintegration and drew their conclusions accordingly.

Hendy’s book created order out of chaos. Alexius I is represented by contemporary writers as having debased the coinage. Hendy showed that this was only true of the first decade of his reign, for in 1092 he carried out a major currency reform, unmentioned in the sources of the time and up to then unrecognized by scholars. It abolished the old coinage completely and substituted one in which the old silver denominations and the copper follis were replaced by a high-value coin of electrum (gold, silver alloy) and a low-value coin of billon (silver, copper alloy). What looked like an indiscriminate mass of concave coins did not simply consist of debased gold coins, as Wroth and all others had supposed, but represented recognizable denominations which could be identified in the written sources and have proper values assigned to them. This coinage pattern remained stable for the best part of a century and its main elements carried on into the Nicene period. Hendy’s discovery made possible a rational arrangement of the coinage of the Comnenians, as well as that of the successor states that followed the Latin capture of Constantinople in 1204.

This was only the first, if basically the most important, of his achievements. There were also two others. One arose out of his 1964/5 work on the many unpublished Bulgarian coin hoards of the 11th–13th centuries. The study of the coins in such hoards is technically extremely difficult, for the majority are in poor condition – coins of low-grade alloy are particularly susceptible to corrosion – and because of their concave fabric they are invariably badly struck, so that legends and details of the designs are hard to make out. But enough was decipherable for him to be able to demonstrate the existence of several mints other than Constantinople and to show how, despite the absence of formal officina marks, divergent details in the rendering of the imperial costume and insignia reflected the internal organization of the mints. The overlapping contents of the hoards not only allowed him to determine the order of most of the coin issues within each reign but, more important still, to show that many of the coins bearing the names of Manuel and other twelfth-century emperors were minted not by them but subsequently, partly by the Bulgarians in the northern Balkans after 1185 and partly by the Latin emperors and the rulers of Thessalonica after 1204. Most scholars now believe that as a result of having worked almost exclusively on Bulgarian hoards instead of Greek ones he overstated his case, and that some of the issues he attributed to Bulgaria originated further south, but his work as a whole has stood the test of time and provided scholars with an entirely new and for the first time intelligible picture of two centuries of Byzantine coinage.

Alfred’s draft of the catalogue section of DOC II, running from Phocas through Theodosius III, was ready for me in December 1965, and I spent the following July and August at Dumbarton Oaks, checking it against the coins and making many changes. Alfred’s original text had not taken full account of my attribution of several important series of Heraclius’ coins to the period of his revolt against Phocas in 608–10, and much revision was necessary to the arrangement of the coins of Constans II and Constantine IV. Three great advances made by the volume on Wroth’s text were the better ordering of the gold of Heraclius, the distinction between the coinages of Justinian II’s two reigns, and the attribution to Leontius of coins Wroth classified under Leo III. The second of these improvements was based on the work of Breckenridge and the third on a much earlier study of Laffranchi. The latter was relatively unknown to scholars, so that even as good a numismatist as Sutherland, writing in 1935, was still trying to explain why Leo III should be shown with an entirely different portrait on some of his coins, the ones in question being really of Leontius. It was therefore important to have the two coinages distinguished in what was destined to become a standard work of reference. Another novelty was the attribution of seventh-century coins to Naples, a possibility which had been rejected by Wroth, and the inclusion of an extensive coinage from Sicily, the latter based essentially on a fundamental collection of material by Ricotti. The justification of these changes and their elucidation were set out at length in a series of introductions which eventually became so long that DOC II had to form two parts, though with a continuous pagination and numbering of the plates. There were many fewer plates then in DOC I, 46 as against 80, but their contents were less extravagantly set out and they illustrated a larger number of coins. Much of the introduction I wrote in Cambridge. The volume was completed in the late summer of 1967 and went to press in September. The galleys were received between January and April 1968, page proofs between May and August, and the corrected page proofs and index went off to the printers on 19 September. Although the date of publication is given on the title-page as 1968, the volume was in fact only published in May 1969.

DOC III, published in 1973, followed essentially the same plan as DOC II, so less need be said about it. An even longer introduction again made publication in two parts necessary, Part 1 with 470 pp. and 29 plates and Part 2 pages 471–897 and plates 30–70. I had, once again, a draft of the catalogue by Alfred to work on, but it likewise needed much revision, particularly for the complicated pattern of issues in the long reign of Constantine VII and the need to change the eleventh century coin names, where he had simply followed Wroth. There were relatively few novelties for the eighth and ninth centuries, but the later section involved accounts of the emergence of separate ‘denominations’ of gold, the histamenon and tetarteron, and of their debasement, together with a proper classification of the Anonymous Folles, based on the work by Bellinger, Margaret Thompson and Whitting and using Thompson’s system of reference. I had hoped to complete the volume in 1970, but my plans for the later part of that year were disorganised by my mother’s death in September, and it was only in September 1971 that I was able to hand over the introduction and the catalogue proper to the Editorial Department. On the basis of the DOC II timetable I hoped for galleys in the spring of 1972, page proofs in July and August, and publication by the end of the year. But the volume was held up by such other editorial commitments as the second volume in the Texts series and the catalogue of the ivories, as well as by Ellen Ash’s employment being abruptly terminated in April on grounds of economy, for while Julia Warner and her assistant Nancy Bowen coped magnificently, they had not Ellen’s familiarity with the material. There were also some unforeseen delays at the printer’s, and inevitably much sending of corrections to and fro across the Atlantic. When it was clear that I would have to wait for 1973 I hoped the volume would be out in time for the spring Symposium, or at least for the visit of the International Numismatic Congress to Dumbarton Oaks in September, but in the end it was not till November 1973 that it appeared.

The result of this delay made the catalogue less of a numismatic sensation than it might otherwise have been, for if it had not been for the publication in 1970 of Cécile Morrisson’s catalogue of the Byzantine coins (to 1204) in the French Cabinet des Médailles, DOC III would have been the first large-scale reference work in Byzantine numismatics to incorporate the tenth- and eleventh-century novelties noted above. But she just beat me to it.

Cécile and I had first met in the winter of 1962/3, when the great French Byzantinist Paul Lemerle, heard that I had agreed with the authorities in the Bibliothèque Nationale to put their Byzantine coins in order, mainly by the merging of several collections which had been acquired at different times and preserved separately. He asked me if I would accept the help of one of his research students, with a view to her becoming a specialist in Byzantine numismatics. I naturally welcomed the proposal. Cécile and I became close friends and over the next few years were in constant touch, with her preparing a catalogue of the Paris collection while I was doing the same for that at Dumbarton Oaks and us constantly sharing our respective ideas and discoveries. Her work appeared first, partly because the Paris collection was smaller and there were fewer coins to describe and catalogue, partly because for the period prior to 713 she had the Dumbarton Oaks volumes to serve as a model, and partly because she is a fast worker and could spend more hours on her task than I could on mine. Each publication has its merits and defects. Hers is more concise, for my introductions were apt to try to explain too much. She sometimes made mistakes through working too fast; her Type 3 of Theophilus’ miliaresion, for example, conflates two distinct coin types, one reading ek The(o)u basileus and the other ek The(o)u pistos basileus, so that the readings in her text and illustrations do not agree with each other. Her system of coin numbering, devised to allow for future insertions, is extremely awkward to use. On the other hand the bold face type she used for inscriptions, worked out by her herself in close association with the printers at Macon, is much more legible and pleasing to the eye than the ones we used. In any case her catalogue, which runs to 1204 and not, like mine, only to 1081, is an astonishing achievement, more especially when the speed with which it was produced is taken into account.

Dumbarton Oaks Catalogues IV-V

DOC III was published in 1973, leaving the two volumes intended to cover the final Byzantine centuries still to appear. Hendy was to be responsible for DOC IV, covering the two centuries 1081–1261, while my share was DOC V, covering the Palaeologan period. Neither has been completed, still less published, at the time of writing (1993). The reasons are complicated, and bound up with other commitments and interests of Hendy and myself. In my case particularly the delay seems difficult to account for, for during the two decades 1973–93 I was still visiting the States every year and probably spent an average of two months a year at Dumbarton Oaks. In principle all that time should have been devoted to Byzantine numismatics, and more especially to the catalogue. My Byzantine files, photos and notes were all in Washington and my other numismatic material in England, so the distinction between my Byzantine and my non-Byzantine work should have been clear. Why has DOC V taken so long?

The answer is in part that the material is exceptionally difficult. Until the appearance of Bendall and Donald’s two monographs on the copper trachea of Michael VIII, and on the general coinages of his successors, in 1974 and 1979 respectively, there were no comprehensive surveys one could use, and these were in any case basically descriptive works and paid little attention to the internal chronology of the issues . Virtually all the coins are undated, and there are relatively few hoards; for some series indeed, none at all. Important studies do exist, or appeared during the 1970s or 1980s for some reigns: by Bertelé, Bendall, Veglery and Zacos, Morrisson, and others. But the amount of preliminary work still requiring to be done was substantially greater than for other DOC volumes.

A further cause of delay was that while in the late 1960s the writing of DOC II and III absorbed much of my time at Cambridge as well as at Dumbarton Oaks, I subsequently tried to work on DOC V only while in America and to keep the part of the year I spent in Cambridge free for other projects. Further, while virtually all my time at DO was devoted to Byzantine numismatics, not all of it was concerned with the preparation of DOC V. There were two major distractions. One was the completion of the research and writing of my book Byzantine Coins, published jointly in 1982 by Methuen in England and the University of California Press in the USA. The second was the writing of the Catalogue of Late Roman Coins in the Dumbarton Oaks and Whittemore Collections by Melinda Mays and myself, published in 1992. These occupied much of the time I spent at DO in the years 1978–81 and 1988–91 respectively. To these can be added the preparation of the numismatic exhibition that was Dumbarton Oaks’s contribution to the Eighth International Numismatic Congress, held at New York and Washington in September 1973, and the writing of the brochures that accompanied this. Another troublesome distraction was the preparation of the coin inventory, wished on the Coin Room by Giles Constable shortly before his relinquishment of the Directorship in 1984, for although the preparation of the inventory was the work of others, the details had often to be checked or explained in a very time-consuming fashion. All of these projects or publications require some elaboration.

My Methuen volume was not a DO project, but since DO existed for the furtherance of Byzantine studies it was legitimate for me to work on it when I was there. The first draft of much of the book was written at DO, and the plates are made up almost entirely from coins in the DO and Whittemore collections. The 240 coins illustrated on the 15 plates devoted to the Palaeologan series were in fact a preview of the plates to DOC V, and the related text allowed me to discuss many problems in advance and provide other scholars with the possibility of providing solutions for them.

The project itself derived from the fact that I had been asked by Methuen in the early 1960s to take over the editorship of a series of numismatic monographs, under the general title of ‘The Library of Numismatics’, to replace the numismatic volumes in Methuen’s ‘Handbooks of Archaeology’ of the 1920s, for these had now served their time and merited replacement. The early series had not included a Byzantine volume, but since I was working in the field of Byzantine numismatics, it was natural that I should try to provide one for the new series, adapting for the purpose the drafts I had originally planned as a parallel work to DOC I and using for succeeding centuries the material intended for the introductions to DOC II and III. In June 1968 I told Methuen’s that although I had no contract for a Byzantine volume in the series, I had a two-thirds completed text that would be suitable and a mock-up of the same proportion of plates. I even contemplated the possibility of finishing the work by the end of the year. In the end I had to lay it aside for virtually a decade, partly because DOC II and III took priority and occupied all my time in the years 1968–73, partly because the unwritten section was mainly the period after 1081 and I underestimated the amount of work necessary for the production of a satisfactory account of this. It was consequently not until 1977 that I formally proposed to Methuen’s the writing of the volume. At a long meeting in London in October with Jane Hunter, the editor in charge of that department in Methuen’s, we sketched the main lines for the volume, which was to run to about 180,000 words and be provided with 96 plates, together with footnotes, bibliography and any other scholarly apparatus as would be necessary. Since I believed that most of the work was already done, the formal contract, signed on 9 February 1978, actually envisaged its completion in under two months from then, at the end of March.

We both realized that this was wildly optimistic, though neither of us contemplated the possibility that Methuen’s would not get the complete text till 1981. I worked on the book during the next few months, and handed over a partial typescript of four-fifths of it before going to the States in August, so that the editors could make the necessary enquiries about the special types and characters it would require. But I was unhappy about the earlier sections, reworked these during much of 1979, and did not complete the last chapter on the Palaeologans till May 1980. There followed a year’s delay while we discussed problems of lay-out, maps, and annotation, together with the revision of parts of the plates where photographs had faded or had to be obtained from outside sources. It was not till July 1981 that copy went to the printers. From then on things went smoothly. I received the proofs on 29 March 1982; they reached me at Dumbarton Oaks, to my stupefaction, in only two days from Hong Kong, where the text was set up. The index was made in April, the revises read during the summer, an advance copy reached me on October 2nd, and the volume was published in November. Its merits and defects are discussed in another section.

The second major distraction, more immediately relevant to Dumbarton Oaks, was the writing of Late Roman Coinage, the catalogue of the DO and Whittemore coins of the fifth century, from the accession of Theodosius I’s two sons Arcadius and Honorius in 383 and 392 respectively to the accession of Anastasius in 491. When Bellinger and I planned the DO catalogue in the early 1960s we decided to follow the British Museum example and start our main catalogue with Anastasius. There had, however, always been a rival tradition, embodied in the standard reference works of Sabatier and Tolstoi, for starting the Byzantine series with the accession of Arcadius, though one could not simply follow their example in describing the coins of Eastern emperors only. The distinction between Eastern and Western emperors was an artificial one created by the fact that Sabatier’s book had in its first section been simply designed to complement Cohen’s great work on Roman Imperial coins, which had broken off its coverage of the Eastern emperors in 395 but continued the Western ones from Honorius onwards. If we started in the late fourth century we would therefore have to include coins of both East and West, and I felt that our late Roman section was not of the same standard as our later Byzantine ones. It was only after full discussion of the alternatives that we decided to start with 491, but I always kept in mind the desirability of a fifth-century catalogue in building up the collection between the 1960s and the 1980s. By the end of the latter decade our collection had been greatly improved. I had succeeded in making a number of notable acquisitions, both such individual rarities or unica as an aureus of Leo I and a marriage solidus of Valentinian III, and substantial series, notably an important group of solidi of the mint of Milan that had been put together by Ulrich-Bansa. The material was therefore available, waiting to be published.

The opportunity to carry out the work came in 1985. Melinda Mays, a friend of Mark and Fiona Blackburn from their undergraduate days, who had just submitted a doctoral thesis at Oxford on Celtic coins, wrote to me out of the blue, in November 1984, to enquire about the possibility of obtaining a Fellowship in 1985 at Dumbarton Oaks in Byzantine studies, or at least a Summer Fellowship. We arranged to meet when I was back at Cambridge in December, and I was most favourably impressed with both her personality and her credentials. Since she was not a Byzantinist by training she could only have received a Fellowship at the expense of some better qualified candidate and I could not see my way to recommending her for one, but it seemed to me an excellent opportunity of obtaining the service of a trained numismatist, even on a temporary basis, for other work in the Coin Room. I therefore proposed to the Director that she should be invited for the summer of 1985, during the interval between her completing her degree and taking up a temporary post at Oriel College to which she had just been appointed, to start work on the preparation of a catalogue of our fifth century coins. My detailed proposal for this envisioned an extensive introduction on the lines of those in DOC II and III, but with the catalogue proper in what numismatists term sylloge format, with all the coins illustrated instead of only some selected specimens, and with their descriptions, reduced to the barest elements, facing each plate. I would myself be responsible for the introduction and for lay-out, appendices, making up the plates, and so forth, while Melinda, who could come to DO for only a quite limited time, would do the actual catalogue. It would mean my postponing further work on DOC V, but our fifth-century coins were well worth publishing and it would be time well spent. The invitation was duly sent in February 1985, and the Director managed to find the necessary funds for Melinda’s travel expenses and stipend.

Melinda spent two months at Dumbarton Oaks between the following 26 July and 24 September. I joined her for the first week, to start her off and explain the documentation of the collection, and was there again for her last week, from September 18th, to see what had been done. This, to my dismay, amounted to virtually nothing, for within a few days of her arrival she had been diverted to another task, that of helping Amy Hatleberg with preparing the inventory of the collection which Giles was demanding and which Amy had come to work on earlier in the year. While a substantial part of the collection, that catalogued in DOC I–III, would present Amy with few problems, the coins of the fourth and fifth centuries had been inadequately labelled, not through any fault of Breckenridge but because the relevant volumes in the standard work of reference, Roman Imperial Coinage, had not yet been published for the years 284–395 when he had worked on the collection in the late 1940s. For coins of the century 395–491 and ones later than 1081 there were no uniform systems of reference at all. Some coins were also out of place, for they had been joggled out of their boxes during the move of the coin safes downstairs in 1973 and been incorrectly put back. These deficiencies were not something that Amy could correct, since she had no numismatic training. Melinda was consequently diverted on to this other task, but on the understanding that funds would be made available for her to return the following year.

She had consequently been able to do virtually no work on the catalogue, but I stayed on myself till October 30th and made a beginning with the introduction. In the summer of 1986 she returned for two months, from 7 July to 2 September, between finishing her year at Oriel and taking up her new post in charge of the coin collection at the Yorkshire Museums. I was likewise in Dumbarton Oaks for two months in the summer, during which she was able to devote all her time to the catalogue while I worked on the introduction. We suited each other temperamentally, laughing at the same things, though I did not share her love of punning, and neither of us minded long hours of work, starting between 8 and 9 in the morning and carrying on till 11 or 12 at night. Since both of us were fairly polyglot we amused ourselves, when the humour took us, in carrying on our conversations in languages other than English. The arrangement of the coins in the catalogue required much thought and negotiation, but since she was putting it direct on a computer the inevitable rearrangements did not prove too difficult.

The volume was effectively completed in the next three years, though it was not published till 1992. I spent some six months at Dumbarton Oaks between 1987 and 1989, and before I left in November 1989 was able to submit the final text and plates. An intervening cause of delay, though a very fortunate one, was the publication in early 1989 of Wolfgang Hahn’s Moneta Imperii Romani, Moneta Imperii Byzantini, covering the coinage of the East between 408 and 491, for although it involved me in much additional labour in inserting references, it was necessary to take account of this authoritative work, even if not always agreeing with it. But the copy-editing, meticulously and most helpfully carried out by Frances Kianka in the Editorial Department, took over a year, in part because of the need to take account of many useful suggestions from Bill Metcalf of the ANS, to whom the work had been submitted for appraisal, in part because of the inevitable delays over posting queries and corrections between England and the States, and indeed, within England, between Melinda at York and myself at Cambridge. The volume went to press in the fall of 1990 and we began to receive proofs in the spring of 1991, when I spent another two months at Dumbarton Oaks and could thus make last-minute checkings against the coins themselves. I had hoped for publication before the end of the year, but other commitments in the publishing schedule at Dumbarton Oaks imposed a further delay and the volume did not finally appear till June 1992. Since it filled a very real gap in the literature it was enthusiastically received in the numismatic world, and has sold well. The only serious criticism, in itself something of a compliment, was that the title, in putting the word ‘Catalogue’ first, was misleading, for the volume was really a monograph on the entire coinage of the late Roman Empire and it would have been better to have said so.

Byzantine Coins and Late Roman Coins thus took up much of my time in the two decades after 1973. I had also other distractions. I had intended, after my Cambridge and Brussels retirements, in 1978 and 1981 respectively, to go to Dumbarton Oaks for long periods of three or four months, or even more, and get DOC V completed. But commitments in England prevented this, especially after I signed a contract with the CUP in 1983 for the production of a 13-volume study entitled Medieval European Coinage on which I began work, in collaboration with Mark Blackburn, later in that year.

There was also, as noted already, much more preliminary work to be done on Palaeologan coinage than there had been for the coinages dealt with in DOC I and II. The reasons behind this we are various. Palaeologan coins are with few exceptions undated and are usually badly struck, so that the readings of the legends and the decipherment of the types can be extremely difficult. Mint attributions are not always simple, and the systems of privy marks employed is still not understood. Little hoard evidence is available for determining the chronology of most of the coin series. The literature of the subject was in 1970 in a highly unsatisfactory state. The older books, effectively those of Sabatier (1861) and Wroth (1908), had disposed of too little material to be able to make much of the coinage. Bertelé had written on it extensively, but his publications, effectively a selection of rare or unpublished coins that appeared in 1926, a hoard of the minority of John V that he published in 1938, and a series of articles on short reigns or types, did not add up to a coherent picture of the coinage. Only an article published in 1964, and that not until it appeared in Cécile Morrisson’s greatly expanded French edition of 1978, made any approach to this. Goodacre’s popular handbook on Byzantine coinage of the early 1930s was better, for although Goodacre was a collector and not an original scholar, he was a friend of Bertelé, who provided him with much material and advice for the third volume of his work (Goodacre 1928–33; repr. with some additions 1957). Only in the 1970s did the situation change materially, with the publication in 1974 of an illustrated guide to the copper trachea of the Emperor Michael VIII and in 1978 of a much more ambitious guide to the coins in all metals of Michael’s successors down to 1453. The authors were Simon Bendall and Peter J. Donald, Simon a dealer and collector working at the time in the firm of A. H. Baldwin & Sons in London, Peter a collector with considerable artistic gifts who supplied the illustrations to the two volumes.

The books – the first is a brochure of 44 pages rather than a book – were widely appreciated by collectors, who found in them a generally reliable guide to a series on which nothing of the sort had existed, but less so by Byzantinists, who were put off by the absence of any scholarly apparatus other than chaotic and badly arranged bibliographies abounding in misspelled names and titles and by the apparently ‘amateur’ character of illustrations between two and three times as large as the coins themselves. In reality, whatever the deficiencies of the scholarly apparatus, the books provide the essential elements of our present knowledge of Palaeologan coinage. The descriptive text is based on an acute sense of style, essential for the determination of mint attributions, and on a quite exceptional acquaintance with the material in museums and private collections all over Europe and America. The illustrations, if sometimes involving errors in the reconstruction of types and legends or perplexingly at odds with the printed descriptions, allow the numismatist to understand and correct the defects of the worn or damaged individual specimens which are usually what he has in front of him. Bendall, either alone or in collaboration with Donald, followed up these two books with a long series of corrigenda and addenda, publishing new material, filling in gaps in the original plan or dealing with privy marks and other topics which he had not covered earlier. His catalogue of his own collection, published in 1988, is an essential adjunct to his earlier volumes. If Dumbarton Oaks has profited greatly from what he has done, the advantages have been mutual. I supplied him in the early 1970s with photos of our own Palaeologan coins, so that he and Peter could use them in preparing their descriptive text and illustrations, and in the summer of 1983 he spent a month at Dumbarton Oaks, checking my arrangement of the coins in the trays and discussing problems of their readings and attributions. If DOC V is at the time of writing (1994) approaching completion, it is largely due to its having been able to profit by his work.

The delay in producing DOC V can thus in part be attributed to the need, in the 1970s and 1980s, for more preparatory work to be done on Palaeologan coinage, a similar explanation cannot account for the delay over DOC IV, for Hendy had effectively completed his preparatory work for this in his monograph of 1969. In [1966] his Fellowship had been formally extended for a year in order that he might catalogue the coins of the period 1091–1261, which in fact he did in the time envisaged, preparing at the same time the plates of which some copies were actually printed. Looking back in the perspective of the past two decades one can perhaps conclude that it would have been better in 1970 to have gone straight ahead with its publication, marking for DOC IV a reversion to the plan of DOC I but with the advantage over this in that DOP 12 existed and could serve as the scholarly introduction which in DOC 1 is lacking. Instead it was thought best to continue with the plan of DOC II and III, with a substantial introduction by Hendy himself. But although he returned to Dumbarton Oaks several times in the 1970s and 1980s and wrote much of the expected Introduction, he did not complete it, producing instead his massive 750-page Studies in the Byzantine Monetary Economy c. 300-1450, published in 1985 by the CUP, and preparing the catalogues of the coins found in the excavations at Kalendarhane and Sarahane, the latter published in 1986. His relations with Dumbarton Oaks underwent a formal change in 1979, when he was appointed Associate Advisor in Numismatics, a post which he retained until 1985. The appointment was the result of a whim of Giles Constable and was not well-advised, for Hendy had never displayed any serious interest in the coin collection as such. But when it was terminated by Robert Thomson, Hendy took offence and relieved his feelings in an ill-tempered letter to the Director, with the result that relations between him and Dumbarton Oaks were broken off between then and 1992. By this latter date the future of DOC IV became something of a problem, for work on DOC V was well advanced and it became essential to know if he would complete DOC IV or not. In 1993 he returned to Dumbarton Oaks for a period of six months (February–July) , and the text of the Introduction and most of the footnotes were completed. The final touches are expected to be carried out during a further visit in the summer of 1994, with publication envisioned for 1995 or early 1996.

I have referred to the writing of my Byzantine Coins and Melinda Mays’ and my Late Roman Coins as my major ‘Byzantine’ distractions from DOC V in the 1970s and 1980s, but they were not the only ones. In September 1973 the Eighth International Numismatic Congress was held in New York and Washington, its hosts in the two cities being respectively the American Numismatic Society and the Smithsonian Institution but its programme at Washington including a reception at Dumbarton Oaks. For this, during my stay in August, I had to prepare an exhibition with the help of Electra Yorsz, Libby Bland’s temporary assistant in the Byzantine Collection, and with superb coin enlargements prepared by Dick Amt of the photographic department. For it I wrote a 16-page brochure under the title Byzantine Coinage: Exhibition at Dumbarton Oaks, briefly describing the collection and various aspects of the coinage as presented in the exhibition. I approached the reception itself with some apprehension, for I had been told that an American learned society had been entertained earlier in the year to a meagre offering of pretzels and root beer, but the then Director had not been an ambassador for nothing and did us proud. The admirable lunch which we had enjoyed the previous day at Evergreen House, when we had stopped at Baltimore on our way from New York to see the Garrett Collection, was quite eclipsed by the splendid collation we were offered on the evening of September 14th in the Music Room and Orangery. Despite grey and overcast skies and occasional rain, the reception proved a great success, and many visitors told me afterwards that for them it represented the high point of their American tour. The exhibition, however, came to a sad end the following December when, following on the news of a coin theft in the Fogg Museum at Cambridge, it was hastily dismantled and the coins replaced in their trays. It was the fate of the splendid photographs that was tragic, for when I returned the following summer I found that they had been piled on top of each other and could not be separated, for the glue on the back of each had by then adhered firmly to the surface of the photograph beneath it. They had all in consequence to be thrown away. A smaller coin exhibition, one of several planned to accompany the meeting of the 17th International Byzantine Congress in Washington on August 5–11, 1986, was put on in the Upper Gallery in June 1986.

Another 32-page brochure, Byzantine Coinage, I wrote as no. 4 in the series of Byzantine Collection Publications (1982). It was no more than a general introduction to the subject. On 29 November 1976 I gave a public lecture in the Music Room on late Roman and early Byzantine gold and silver medallions, and made it the basis of a somewhat longer essay, illustrated with a selection of twenty medallions in the Dumbarton Oaks collection, in which I discussed the nature and functions of these rare objects. It was never published, unfortunately, for it was subjected to such an extensive and in my view unnecessary editorial rewriting that I abandoned it in disgust. Recently, however, I have extracted from it my descriptions of the six medallions that had been acquired since those then in the collection had been published by Bellinger in 1958, and made these the subject of an article in DOP [1996]. I had better fortune with a paper on the role of silver in the late Roman and Sasanian economies, for although given at a symposium at Baltimore and Washington in May 1986 and not published till November 1993, it did at least manage to survive in its original form. In 1985 I contributed some thirty entries to the three-volume Oxford Dictionary of Byzantium (1991), which although not formally a Dumbarton Oaks publication was organized there, with Alexander Kazhdan as its chief editor.

The coin collection naturally continued to grow in the 1970s and 1980s, though on a less lavish scale. After Mrs Bliss’s death and Jack Thacher’s retirement I had no longer a regular allowance for purchases, though funds were always made available to me if a special case could be made. A notable but slightly earlier acquisition was an important series of mid-fifth century solidi of the mint of Milan, which had once formed part of Ulrich-Bansa’s collection and which I bought from a Swiss bank in 1967. In 1990 I presented the collection with over 50 coins of the sons of Theodosius I which I had not disposed of in 1956 and was anxious to include in Late Roman Coins. In the late 1980s I was also able to arrange a number of exchanges in the Paleologan series with Simon Bendall, for each of us had a number of duplicates of types lacking in the other collection.

Normally, as in the case of the Milanese purchase just referred to, I negotiated the transactions in person, but when bidding at auctions I still tended to act through agents as I had in the 1950s and early 1960s. Exceptions that stand out in my memory are my acquisition of a gold solidus of Jovinus, a marriage solidus of Theodosius II, and a unique silver miliarense of Leo I, all at Swiss sales. The first (DOC LR 808) I wanted because we had no gold coins of Jovinus, a short-lived usurper in Gaul in 411–13, and I felt that the only way of being sure of getting it was by bidding in person when it appeared at an MMAG sale in Basel on 15 June 1975, lot 175. Although I had to pay more than I expected, the high price was more than offset by my finding, in MMAG stock, a number of late Byzantine silver coins of the Nicene period and of Michael VIII at very reasonable prices. The marriage solidus (DOC LR 395), with a most unusual reverse type, the standing figure of Theodosius II between the emperor’s daughter and her cousin Valentinian III, was at that time the only known specimen of the coin. It appeared in a Hess sale at Lucerne (2.iv.l958, lot 411), and although it was in poor condition – it bore traces of pick-marks evidently made when it was discovered – I was very glad to have it. I was staying at the time in Zurich, and driven back there after the sale by Frank Sternberg, from whom I bought some more coins for my own collection the next day. Finally, the unique silver coin of Leo I (DOC LR 548) I bought at a Sternberg sale (lot 548) at Zurich on 15 November 1985, my 75th birthday. My grandniece Deirdre and her husband Gerard Justafre were then living in Zurich and put on a magnificent party for me in the evening, with other members of the family who lived in Switzerland coming for the occasion. I learned later that I had had a stroke of luck over the purchase, for a dealer had entrusted a colleague with a very high bid on behalf of one of his clients, but there had been some confusion over the numbers and the high bid was never made.

I could more easily bid in person at Glendining’s or other auction houses in London, but very rarely did so. The only time when I felt it really necessary was at the Christie sale on 19 October 1970, which included a remarkable collection of late Roman gold coins and jewellery. On this occasion I was buying not for the coin department but for the Byzantine Collection, which is rich in late Roman and Byzantine jewellery. The coins were of the late fourth century but the jewellery included four medallions of Constantine I in extremely beautiful and elaborate gold mounts and had presumably come down as family heirlooms. I was commissioned to buy what I could up to a limit of £20,000 but only managed to secure one, for 13,000 guineas, a unique medallion of A.D. 324 showing Constantine and two of his sons, mounted in a large, hexagonal filigree gold frame. I was at the time President of the College and by chance, on the evening of the sale, was due to act as host to a reception of Fellows and their wives. I was strongly tempted to bring it back with me to Cambridge and wear for the occasion a polo-neck sweater, with the medallion hanging on a chain round my neck, but was dissuaded by the fact that although I had done the bidding, the ornament was not my property and I would be responsible if any harm came to it. All four medallions and their settings have now found their homes in major collections. A second was bought at the time by the Louvre, the third, of a different date and with a differently shaped frame, was bought by Dumbarton Oaks when it came again on the market in 1974, and the fourth was bought by the British Museum in 1984 from a private collector in Switzerland. By that time the price had gone up to nearly £200,000.


Extract from a Memoir,

in the Coin Room of the Fitzwilliam Museum, Cambridge, on the making of the Grierson Collection (2.v.1996)

(a)    Byzantine Coins

In my first ten years of collecting I managed to acquire something over a thousand Byzantine coins, about 200 of them gold. I have explained in the first section how, after being appointed Advisor in Byzantine Numismatics at Dumbarton Oaks with the duty of building up the collection

there, I decided to cease collecting Byzantine coins myself in order to avoid a conflict of interest, and to sell Dumbarton Oaks any coins in my collection that were not represented there already. The sale was arranged in some haste during the winter of 1955/6, at an independent valuation, since I wanted to have as much cash available as possible to buy medieval coins at the Lockett sale of February 1956 just referred to.

530 of my coins, including 120 in gold, thus went to Dumbarton Oaks in 1956. Though the Byzantine section of my collection, therefore, no longer exists as an entity, and only a small part of it has come to the Fitzwilliam Museum, its formation played so considerable a role in my early numismatic studies that it should not be omitted here. Even if it were worth the trouble, however, it would be impossible to reconstruct it in detail. Although most of the coin bills still exist, and the source and date of acquisition of each coin are recorded on the tickets which still accompany such coins as remain here or are at Dumbarton Oaks, the relevant entries from the card file of my coins have long ago been removed.

My collection began, as related in the first section, with a Byzantine copper coin which I picked out of a junk box at home and which Charles Seltman identified for me, in January 1945, as a half-follis of the emperor Phocas (602–10). There were two reasons behind my acquiring Byzantine coins in my first visit to Spink’s some days later, and my continuing to do so over the following decade. One was an interest in Byzantine history. This went back to my chance purchase of a three-volume copy of Gibbon, in very small print, on a barrow in one of the streets leading down to the quays in Dublin in my last year (1928) at school. It continued in my undergraduate days, for the books I chose as college prizes in 1929 and 1930 included Bury’s two-volume History of the Later Roman Empire, which ends with the reign of Justinian, and his History of the Eastern Roman Empire from the fall of Irene to the accession of Basil I, which covers the first two-thirds of the ninth century. The other was the easy availability and low price of Byzantine coins. Copper folles and their fractions could be bought for a few shillings each, sometimes for no more than a few pence. Gold coins were more expensive, but the prices of many of them were in the range of £3/£5, so that one could put together a respectable series at relatively little cost. Silver coins had a reputation for rarity, but this only applied to certain periods. One could buy a hexagram of Heraclius or a miliaresion of Leo VI for relatively little.

Many of the Byzantine coins on the London market in 1945 were ex-Grantley. His main series had been dispersed in the seventh sale of 25 May 1944, where the gold and silver formed lots 2438–2351, with over 334 coins, and the copper lots 2712–35, or 780 coins, making a total of rather over a thousand coins. There was a spill-over of the late sub-Byzantine series in the eleventh sale of 26 April 1945, with 46 coins of the empire of Trebizond (lots 4237–40), 19 of that of Thessalonica (lots 4241–2), and a total of 23 attributed to the empire of Nicaea, the despotate of Epirus, and the duchy of Neopatras (lots 4243–4). Many of the gold and silver coins went back to a great Glendining sale of Byzantine coins on 8 December 1922, described as that of a ‘Foreign Prince’ and in fact the family collection of a Russian nobleman of Greek family, Prince Cantacuzene. The sale did not include the prince’s copper coins, for these had been bought en bloc by Spink’s and many of them subsequently by Lord Grantley, who indicated their source by the letters PC on his tickets.

My coin bills show that in the course of 1945 I bought large numbers of Byzantine coins from the London dealers Spink, Seaby and Baldwin, many being of gold and giving me representative solidi for most periods between the fifth century and the twelfth, but the details are too imprecisely recorded for any estimate of the total to be possible. The first gold coin I ever owned was a Thessalonican solidus of Arcadius, now at Dumbarton Oaks (LRC 61), with a charming profile bust of the child emperor, which I bought from Spink’s in February 1945 for £5. It is in fact a great rarity, and I was lucky. But most of the coins were copper, likewise well spread out in time, and served to familiarize me with the coinage of most periods except that of the Palaeologans, the coins of which only came my way later. Most of the copper coins were ex-Grantley, but the gold, and the relatively rare silver coins, came from a variety of sources, for Byzantine coins are usually available on the market.

In 1946 I continued to buy Byzantine coins in London but also began to do so abroad, mainly from dealers in Paris, Brussels, Basel, Milan and Rome, who usually had some in their trays. This pattern of acquisitions, largely from dealers’ stocks, continued over the next decade. In London I had competitors in Philip Whitting, senior history master at St Paul’s School, whose inspired teaching was to interest successive generations of schoolboys in history, coins and Byzantium, and Jim Stewart, of whom I saw a great deal over the years 1945–7, prior to his return to his native New South Wales. Blunt also took a marginal interest in the series. Whitting, whose great collection is now in the Barber Art Gallery in Birmingham, had the advantage over me of living in London, so that he tended to get first pick of anything new on the market. In Milan there was another major collector, the wealthy Swiss industrialist Enrico Leuthold, who had acquired en bloc from Baron Ulrich-Bansa the latter’s extensive acquisitions at the great Ratto sale of Byzantine coins at Lugano on 9 December 1930, but I suspect that he tended to buy entire collections rather than single coins and I doubt if he haunted dealers in the same way that Whitting, Blunt, Stewart and I were prepared to do. There also must have been avid collectors elsewhere of whom I knew nothing.

Apart from the Ratto sale just alluded to and a very few others, Byzantine coins have never made up complete auction sales but have either formed parts of great general collections, as was the case with Grantley’s Byzantine coins, or have been offered as continuations of collections of Roman coins, since the ‘empire’ that ended in 1453 was formally still a ‘Roman’ one. There were a few sales of the latter kind at Glendining’s during these years: the E. A. Sydenham sale on 26 March 1948, the second Henry Platt Hall sale on 16 November 1950, the second L. A. Lawrence sale on 27 January 1951, the Leopold Messenger sale on 21 November 1951, the first J. C. S. Rashleigh sale on 14 January 1953. My acquisitions at these tended to be copper coins of the later Empire, sold in lots of twenty coins or more. Lots 988 in the Lawrence sale and 869B in the Sydenham sale supplied most of my coins of Arcadius and Honorius that went ultimately to Dumbarton Oaks. Usually I bought through dealers, but it was sometimes worth attending a sale in person, partly to get an idea of the competition – and incidentally save dealers’ commissions – but also because one could never know what might not turn up. I went to the Platt Hall sale in the hope of buying for Ulrich Bansa a previously unknown solidus of the Roman usurper John (423–5) of the mint of Milan – I failed; it was bought for the British Museum – but I got for myself a solidus of Priscus Attalus, now at Dumbarton Oaks (LRC 812) which did not appear in the sale catalogue but was inserted by the auctioneer, after the Platt Hall specimen (lot 2072), on behalf of a well-known Irish collector, Capt. H. D. Gallwey. I also bought occasionally at auctions abroad, but it would not be worth trying to reconstruct the details. By February 1956, when over half my Byzantine coins went to Dumbarton Oaks, I had certainly put together a very respectable collection.

A few of the coins acquired during these years were as far as I could discover unknown, or at least had not been studied. Some of them I published at the time, just to put them on record: three new coins of the fifth century emperor Zeno (476–91), all now at Dumbarton Oaks and one of them a nummus with a standing figure of the emperor on the reverse (Grierson 1948); a barbarous North African solidus of the seventh century (Grierson 1950c) subsequently ascribed to Carthage and the emperor Constantine IV (DOC II. 546.42.3); an unknown follis of the late eleventh century with the letters CBNB in the reverse field (now DOC III. 836.2b) instead of the CANS of Nicephorus III (1078-81) which I proposed to attribute to his rival Nicephorus Basilacius (Grierson 1950d); a follis found in Syria with the mint-signature ALEXANΔ (DOC II. 214.16) which I attributed to a mint opened briefly at Alexandretta during the revolt of Heraclius; a follis of Heraclius (DOC II.328.181a .2) with an unusual mint-mark, bought as part of lot 110 in a Glendining sale of 3 May 1951, which resulted in the discovery of yet another mint, that of Seleucia Isauriae in south-east Asia Minor, opened by Heraclius during his Persian campaign in the 6th and 7th years of his reign (Grierson 1951c). These, it may fairly be said, filled small gaps in our knowledge, but did little more. Two coins that came my way, however, resulted in more significant studies or discoveries. One was a light-weight solidus of the sixth century, the other a dated solidus of Carthage of the emperor Maurice.

My interest in light-weight solidi went back to my purchase from Spink’s, in May 1945, of a solidus of Justinian, now at Dumbarton Oaks (DOC 1, 72.11), having in the exergue of the reverse the letters ODXX instead of the customary CONOB. The latter inscription implied, at least in theory, that the coin was minted at Constantinople and was of pure gold, CON indicating the mint and OB being an abbreviation of obryzon or obryzum, the technical Greek and Latin terms for the refined metal. ODXX was a puzzle. I remember discussing it that summer with Charles Seltman, and him telling me that such anomalies were always worth studying. They might turn out to be just die-sinker’s mistakes of no consequence, but they might equally represent some hitherto unnoticed feature characterizing a whole group of coins and requiring explanation. So I began to collect illustrations and descriptions of such coins in sale catalogues and the like. It soon became clear that my coin fell into both of Seltman’s categories. The OD was indeed a die-sinker’s error; on most coins with similar formulae there was the OB one would expect. For Justinian and his successors, down to Constans II, I in fact found a number of coins with OBXX, or BOXX (again an error), or with such variants as OB+* or OB*+*, and so on. Naturally, as a collector, I was anxious to acquire more myself, and by the time I went to New York in 1953 I owned three further specimens, an OBXX one of Justinian (DOC I, 72.10) and OB*+* ones of Maurice and Heraclius (DOC I, 299.8 and DOC II, 249.12.1).

On weighing the coins I found them to be all below the 4.56g of the normal solidus, a weight corresponding to 24 siliquae or carats (keratia) in Latin or Greek terms. The XX should naturally correspond to 20 carats (3.80g), as one would expect, and the +* apparently to about 21 or 21 1/2 carats. I indeed found an article on the subject by the Italian scholar Monneret de Villard published several decades earlier (1923), though it was in many respects unsatisfactory and it seemed that the author knew the material entirely at second-hand, with no first-hand acquaintance with the coins at all. But I was too busy with other projects to follow up the subject, and did nothing about it. At the ANS Summer School in 1953, however, I suggested ‘Byzantine light-weight solidi’ as the topic for his paper to one of the students, Bill Bowsky, and put my material at his disposal. It was eventually taken over by Howard Adelson, one of the instructors the next summer, and incorporated in a monograph he published in 1959. The subject was clearly there for the asking, and any scholar might have lit upon it, but its study in the 1950s derived from a coin purchased by me from Spink’s in 1945.

As for the solidi eventually attributed to Carthage, my interest in these started in the summer of 1947, when I bought from Baldwin’s a solidus (DOC II, 355.233) rather smaller than normal and of unusual style, having an anomalous obverse inscription, ending ANE instead of the normal AVC (for Augustus). The reverse legend also ended in E, but this was perfectly acceptable, for there were ten officinae, identified on the coins by one of the numerals A-I ( = 1-10) following the legend, in the mint of Constantinople, and E could indicate the fifth officina. In December 1948, however, I bought at a Basel auction a much more surprising solidus of Maurice (DOC II.354.227), for its obverse legend ended ANIΓ and its reverse legend IΓ. Such numerals would be impossible at Constantinople, for iota-gamma is 13 and there were only ten officinae in the mint. I therefore began to collect records of coins of Maurice with similar inscriptions, and eventually ones of Phocas and the early years of Heraclius also. It soon became apparent that there were numerals corresponding to 1–15 but never going above 15, and that there were, for Maurice, two series of coins with letters corresponding to 1–5, one of normal module and the other appreciably smaller in size.

This was the clue I needed. The Byzantines dated their years by an indictional cycle of 15 years, and Maurice’s reign covered the whole fifteen years of one indiction and the first five of the next. The ‘letters’ were in fact indictional dates, a fact at once confirmed by ‘letters’ on Phocas’ coins, for these ran from S to IA, i.e. 6–11, and would correspond to those of his reign at Carthage, where a revolt against him broke out in 608. The declining size of the coins over these years in fact led on to the so-called ‘globular’ solidi of Carthage, i.e. to the very small and thick coins of Heraclius and Constans II attributable to Carthage on the evidence of finds. The globular solidi were in fact dated by indictions as the earlier ones had been, and could now be arranged chronologically, which had not previously been possible. My conclusions were made the subject of a long article in the Numismatic Chronicle (Grierson 1950a). Already before it was published I had started to acquire as many of these dated solidi as I could, so that when my collection went to Dumbarton Oaks it included no fewer than eleven of Maurice and two of Phocas.

The discovery that the letters referred to indictions also provided a solution to a problem that had been much discussed, that of the dating of Heraclius’ consular coins, as I showed in an appendix (Grierson 1950b) to the article just cited. These coins, which exist in all three metals, are anomalous in that Heraclius is given a consular title only, not an imperial one, and that he holds no imperial ornaments. The copper coins have either the mint signature of Carthage or can be attributed to this mint on grounds of style and fabric, but the solidi form two groups, one very close stylistically to that of the copper and the other quite different. Both groups of solidi, however, bear the indictional dates 11, 12, 13, and these correspond to the years 607/8, 608/9 and 609/10, i.e. to those of the revolt of Heraclius which started in Carthage in 608 and spread at once to Alexandria, but only reached Constantinople in October 610. The ‘consuls’ of the coin are in fact not Heraclius and one of his sons, but Heraclius and his father, the exarch of Africa, who presumably assumed the consular title to provide some kind of constitutional framework for the revolt. The solidi, none of which appear to be Constantinopolitan, can be divided between Carthage, which minted the coins of good style, and Alexandria, which minted the much cruder ones.

The 530 coins that went to Dumbarton Oaks in 1956 formed about half my Byzantine collection. What remained were a substantial number of coins in all three metals that would have only been duplicates, plus a certain number of late fourth- and early fifth-century coins whose relevance to the Dumbarton Oaks collection was undecided, since it was assumed that the ‘Byzantine’ catalogue would start from Anastasius I’s accession in 491 and no plans for a late Roman volume had been formulated. Such a volume remained on the cards, however, and since I needed all the cash I could raise for the Lockett purchases, I included in the sale as many gold and silver coins from Arcadius and Honorius onwards as were not already in the collection.

The Byzantine coins that remained had diverse fates. In 1964 I gave some 75 late Roman silver and copper coins to the Fitzwilliam Museum to fill in gaps, but what prompted the gift at that particular time I do not remember. More easily explicable is a series of gifts during Michael Hendy’s five years (1967–72) in the Museum as an assistant curator: 49 gold coins in 1967, four more in 1968, and 42 silver and copper in 1969. The gold and silver coins formed a particularly important addition to the collection, for the quite large Byzantine series already in the museum, essentially from the bequest of Charles Davies Sherborn and the generosity of C. J. Bunn’s heirs, consisted almost entirely of copper coins. What I still had, after these gifts, consisted essentially of Valentinianic and Theodosian Æ of the late fourth century. A few of these went to Dumbarton Oaks as a gift in 1971 – these were small Theodosian Æ from lots bought at the Longuet sale of 17 March 1970 or acquired in Basel in June 1971 – or in 1986, when Melinda Mays and I were starting to work on its late Roman catalogue and I wished to acquire coins of mints, officinae and issues we did not have already. What finally remained, probably a couple of hundred copper coins going back to the Tetrarchy and even the Principate, I sold to Baldwin’s in January 1981. The coins were not needed by either of the museums for whose collections I had some responsibility, and I was glad to put them back into circulation to give pleasure to fellow-collectors.