Demobilized in 1947, Andrew was accepted to read music at Lincoln College, Oxford, but on arrival there found that he had been selected to read English instead. He chose to study Course I, the medieval and philological option, to which he added a paper in Anglo-Saxon palaeography taught by the eminent palaeographer, Neil Ker (himself the Society's Gold Medallist in 1975). This option first kindled his life-long interest in the study of medieval books, and Andrew duly repaid the scholarly debt he owed to Ker when in later years with another of Ker's students, Malcolm Parkes, he edited a Festschrift for him, Medieval Scribes, Manuscripts and Libraries (1978). Later still, he acted as Ker's literary executor, publishing a collection of his essays under the title, Books, Collectors, and Libraries: Studies in the Medieval Heritage (1985). He further edited a supplement to Ker's Medieval Libraries of Great Britain (1987) and provided indexes and addenda to another invaluable reference work, the final volume of Ker's Medieval Manuscripts in British LibrariesV (2002).

After graduating in 1950 Andrew worked for a time in the reference library of St Marylebone Public Library and in the Guildhall Library, before joining the staff of the School of Library, Archives and Information Studies (now Department of Information Studies) at University College London, where he remained throughout his career, eventually becoming its Director, 1984–90. There he introduced a course in palaeography and codicology for those students who wished to work with historical collections. While he did not neglect a practical approach to the subject, requiring students to transcribe facsimiles and setting questionnaires which, by requiring them to seek answers in the library, brought familiarity with essential reference works, he actively encouraged academic research. Among those who profited from his teaching are David McKitterick and Teresa Webber, themselves now dis tinguished scholars.

But Andrew's influence on the subject was felt beyond the immediate confines of UCL. It was at his suggestion that in 1977 the inter-collegiate London seminar in Palaeography was refounded after a lapse of ten years. Now known as the Medieval Manuscripts Seminar, this is still running successfully and attracts both national and international speakers.

Andrew's first publication focussed on a recently acquired manuscript in the British Museum Library, MS Egerton 3138, containing a list of manuscripts bought by the antiquary, Sir Simond D’Ewes, between 1623 and 1640, some of which came from John Dee's collection (The LibraryV, 13 (1958), 194–98). A more detailed study, The Library of Sir Simonds D’Ewes, a revision of his Oxford BLitt thesis (1961), was published by the British Museum in 1966. Later still, with Julian Roberts, Andrew published John Dee's Library Catalogue (Bibliographical Society, 1990). Not only did Dee and D’Ewes attract his notice, but other less well-known individuals who also acquired medieval manuscripts in the aftermath of the dissolution of the monasteries and consequent despoliation of these libraries. By their activities, recognized by Andrew, they saved for future generations a substantial part of our literary and cultural heritage. His interest in early collectors was pursued in a series of articles on men such as Thomas Dackomb, a minor canon of Winchester, the Londoners Christopher and William Carye, Robert Green of Welby, Robert Hare, John Twyne of Canterbury, Sir Walter Cope, Henry Savile of Banke, and Thomas Allen of Oxford, whose collections passed into yet larger collections (such as Cotton's or Harley's) and thence into institutional libraries, a process that Andrew likened to the increasing growth of a snowball. Most such essays, several first appearing in The Library, were republished in his collected papers, Medieval Manuscripts in Post-Medieval England (2004).

Elected a member of the Comité international de paléographie latine in 1979, Andrew undertook to catalogue manuscripts in the British Library and also in Oxford libraries which were either precisely dated or which could be firmly dated within twenty-five years of their production because of their content and/or early history. The cataloguing of such manuscripts was part of an international enterprise promoted by the Comité since 1953 for the purpose of providing photographic specimens of scripts known to be dated or datable, by which criteria could be established to help in dating undated manuscripts. His Catalogue of Dated and Datable Manuscripts c. 711–2611 in the Department of Manuscripts, The British Library and Catalogue of Dated and Datable Manuscripts c. 435–2611 in Oxford Libraries appeared in 1979 and 1984 respectively. For 1,853 entries all told, Andrew gave a brief description of a manuscript's make-up with a paragraph succinctly setting out the argument for that manuscript's date, together with a well-chosen illustration of a specimen page of handwriting in an accompanying volume of plates. The task demanded an extensive awareness of the literature and careful consideration of the evidence, even for a book apparently dated by a scribal colophon. (Six manuscripts of Bradwardine's De causa Dei, for example, are dated 1344, the date of the text's conclusion rather than that of the manuscript copy.) Each catalogue ends with a list of manuscripts considered but rejected.

On his retirement in 1990 Andrew settled in Oxford, where he was not idle. As a member of the editorial committee of the Corpus of British Medieval Library Catalogues, Andrew edited the lists of books belonging to Lanthony priory for The Libraries of the Augustinian Canons (1998). He was invited to catalogue the medieval manuscripts of All Souls College, and on completion of that task he undertook the cataloguing of those of Exeter College, his descriptive catalogues published in 1997 and 2002 respectively. As anyone who has undertaken such work could tell, such a task demands painstaking attention to detail. Unlike his previous catalogues with their concise notices, the college manuscripts required much fuller descriptions. With a volume that contained numerous items without any title or colophon each text had to be carefully identified, the manuscript collated, and illustrations (if any) and binding described, and the evidence for its origin and provenance recorded. His obituary notice in The Times (14 December 2017) reported Andrew's wry amusement at the bafflement of an economics don who choked on his soup when told that it had taken Andrew eight weeks to describe one such book.

Beyond scholarship, Andrew enjoyed a musical life. His deep love and appreciation of music found early expression in his playing of the organ for the local episcopal church in Kirkcudbright (despite his father's position as an Elder of the Kirk). During the war he played the organ at church services for his garrison, sometimes on a Mediterranean beach. A witty man, he main tained collegial friendships with others engaged with manuscript studies, both established scholars like himself or younger students embarking on their researches. Many of us working in the field today have reason to be grateful for his expertise and generous help and advice, always courteously given.