POxBo retains and expands upon Ker's original numbering of fragments. This note provides some more detail.
Explanation of the POxBo numbering system
Neil Ker organised his listing of pastedowns by binding tool, arranged as far as possible in chronological order. As his Scope and Arrangement explains, within each binding tool he followed a sequence: Oxford colleges followed by libraries elsewhere (the Bodleian included under 'O'), with the ordering within each institution being by date of publication of the host volume. All pastedowns are arranged in a single series of arabic numerals, 1-2017.
However, Ker's research did not stop when he had established that numbering and there was a need to insert more witnesses into the sequence. This was, first of all, done by providing an entry with a letter after the preceding number (e.g. 1198a). The discoveries continued after the book was sent to the publishers, and eighteen were presented in a section of Addenda (pp. 228-229) with a new numbering system; these are included in POxBo with the prefix ‘Add.’.
Ker did not leave the investigations behind after the publication of Pastedowns. His corrections and additions made after 1954 became central to the supplement to the listing published by David Pearson in his Oxford Bookbinding 1500-1640 (Oxford, 2000), where he also included the fruits of his own research. Pearson's practice was to insert a fragment at the appropriate point in the existing series, and giving it an arabic numeral after a point: e.g., Ker's 518 was now followed by 518.1 and 518.2 and Ker's 518a followed by 518a.1, 518a.2 and 518a.3. Pearson also included a listing of fragments found in books bound by the 'Dragon Binder', who may be identifiable as Thomas Bedford (d. 1507?); he gave these their own sequence of arabic numerals preceded by 'D'.
There are two further additions now made by POxBo. First, Ker included many examples of fragments related to the ones in his main list but where the binding in which they were used was unidentifiable in footnotes. To aid searching, these have been moved into the main listing, with the prefix 'N'.
Second, both Ker and Pearson includes an appendix of Oxford bindings before 1574 in which there are no manuscript pastedowns. The primary intention of this was to make lists of bindings as full as possible but Ker also noted that some of these had pastedowns which had been removed. Some of these retain signs of having had pastedowns in place, sometimes with some detail of the text possible to reconstruct. It is to the purpose of a database primarily concerned with lost manuscripts to list these as well. Ker's appendix was by lower-case roman numerals; Pearson's revised this to arabic numerals preceded by 'A'. In POxBo, for the relevant entries, the original numbering in each case is respected with the additional prefix of 'App.' (so there is both App.xv and App.A19.1, for instance).
In addition to his main catalogue, Ker provided a list of manuscript fragments used as wrappers to bind business volumes in various Oxford colleges: these were ordered by institution, with the arabic numbering being anew for each volume. This was supplemented by Pearson, respecting the original order but placing 'W' before each additional entry. All these wrappers now appear in POxBo and have the prefix 'W' and an abbreviation for the institution (All Souls = 'AS'; Christ Church = 'Ch'; Corpus Christi College = 'Cx'; Magdalen = 'Ma'; Merton = 'Me'; New College = 'N'), followed by the number assigned by Ker or Pearson.