Reading a Manuscript Description

Dr Matthew Holford, Curator of the Medieval Manuscripts Cataloguing Project at the Bodleian Library in Oxford, explains the intricacies of catalogue descriptions of manuscripts. This post features both in our blog and our Teachable Features series. 

It is easy to forget that manuscript descriptions can be hard to understand. Once you are familiar with the conventions, it is not too difficult to decode even the more cryptic of entries:


(Yale University Library, Beinecke MS 353)

But such familiarity can be difficult to acquire. Catalogues have not always explained their conventions; the subject is not well covered in textbooks;[1] and even someone familiar with the conventions of English-language manuscript catalogues may have difficulty with those of German or French publications. The aim of this post is two-fold: (1) to offer some guidance for researchers encountering manuscript catalogues for the first time (2) to provide links to a range of catalogues and descriptions that others may find useful when covering this topic in their own teaching. I have included a small number of images in the text where I am reasonably confident that copyright restrictions no longer apply. In general, though, I have preferred to link back to the original source; these links should open in a new tab or window.


Naturally manuscript catalogues vary considerably in the information they include and the conventions used to express it. Not only has cataloguing practice developed over time, it has also varied (and continues to vary) between countries and institutions, and also by period and discipline. Any researcher is likely to consult a range of catalogues, print and digital, from the nineteenth century to the twenty-first, and produced by a range of countries and institutions. S/he will need to be familiar with codicological vocabulary not only in English but in Latin (the language of many older descriptions) and other European languages. (The excellent resource Codicologica (in French) gives a lot of assistance with this, and incorporates the published Vocabulaire Codicologique; for English terms there is the British Library’s glossary.)

Despite their differences, almost all manuscript descriptions will include three basic categories of information: content, physical description, and history. Often but not always these will form discrete sections of a description; decoration, where present, may be treated as part of the physical description, the content, or may form a separate section. In older descriptions in particular it is common to find two sections, one for content, one for everything else (physical description, history, decoration). Examples include M. R. James’s catalogues (e.g. Cambridge, Trinity College, B. 3. 4, published 1900) or the Bodleian library Quarto catalogues (e.g. MS. Canon. Misc. 105, published 1854), where in both cases the opening section includes physical description and history.


The cataloguer’s aim is usually to fully describe and where possible identify the textual content of the manuscript, but this can be achieved in different ways. At one extreme are the catalogues which describe texts as they are described in the manuscript, with a minimum of editorial intervention and identification. M. R. James’s catalogues are a good example of this approach. Here is his account (published 1900) of Cambridge, Trinity College B. 3. 3:


This transcribes the opening heading or rubric to the table of contents (fol. 1), gives the final rubric of the table and the rubric of the prologue, the opening and closing words (incipit and explicit) of the prologue, the final rubric of the prologue and the rubric of book 1, the incipit and explicit of the main text, and finally the final rubric of the main text. The only hint offered to the reader to identify the text or to find a printed edition is the parenthetical (xci.99), which refers to volume and column of Migne’s Patrologia Latina.

The opposite approach can be illustrated by an entry (MS. D’Orville 96) from the Bodleian Library’s Summary Catalogue (1895-1953) where the main text is identified but no indication is given of any rubrics and few details of the other texts.


Most catalogues steer between these two extremes. The cataloguer typically supplies a uniform title (sometimes this will be in square brackets, to emphasize that it is an editorial not a manuscript title) but will also give any headings and rubrics found in the manuscript. It is usual for incipit and explicit to be given although in the case of very well-known works the cataloguer may not consider this necessary. The Huntington Library’s description of MS. EL 9 H 9, originally published in 1989, provides some good examples:

item 1: there are no rubrics in the manuscript: author and title have been supplied (in square brackets) by the cataloguer. The text is defective, so the incipit provided begins with the symbol ‘//’, a common marker for incomplete text. The ellipsis ‘…’ marks the end of the incipit and the beginning of the incipit.

item 3: several rubrics, incipits and explicits are provided for the different parts of the text. The rubrics are italicized: this is a common way of distinguishing them from other text. (Other methods include comments such as ‘[in red] or ‘(red capitals)’, as in TCC B. 3. 3. above. Italicized text can also denote part of an incipit, if a text begins by quoting another text, as often with commentaries or sermons for example.) The author and title are both named in the manuscript rubric and so the cataloguer has not thought it necessary to supply them: if there were any ambiguity, it would be removed by the bibliographical references provided at the end of the paragraph.

A similar approach can be seen in R. A. B. Mynors’s catalogue of Balliol College manuscripts, published 1963: for example MS. 3.

Texts are typically described in the order in which they appear in the manuscript, but practice varies regarding texts which were not originally integral to the manuscript but added later. They may be described in their manuscript sequence, with a note that they are additions (as in this 2006 description of Fribourg University MS L 24 [pdf], at fols. 126r-v, etc.). Alternatively they may be described in a separate paragraph following the original textual content. This was the approach adopted by A. C. de la Mare in her catalogue of the Lyell manuscripts (1971), for example MS. Lyell 26 (p. 56, p. 57, p. 58).


Older cataloguers tended to be content with specifying the material of the writing support, the number of pages or folios, some indication of layout (usually only the number of columns, if more than one), and a rough approximation of dimensions. This was often in terms derived from paper manufacture: folio (the size of a sheet folded once), quarto (folded twice), octavo and so on. (These were not exact sizes, since the size of a full sheet of paper varied, as George Gordon and William Noel have recently explained in detail.) For example Bodl. MS. Canon. Misc. 107, a small quarto parchment manuscript of 133 folios in two columns.


More recent catalogues typically provide much more information.  This will include the dimensions of the overall page and of the written or ruled space, in millimetres or centimetres; the number of lines per page; information about ruling and pricking, collation, and binding. This information is not always without ambiguity, and is sometimes given in a compressed style that can initially be confusing. Some potentially problematic areas are:

Number of leaves. It is not always clear whether the enumeration given includes flyleaves. A simple arabic number “ff. 183” typically suggests that flyleaves have not been included (e.g. Trinity B. 1. 39). A combination of roman and arabic numerals (or of letters and arabic numerals) distinguishes flyleaves (roman or letters) from main volume (arabic) – although end flyleaves may be included in the main foliation. Recent catalogues will typically specify the date and material of the endleaves, e.g.  “ii (modern paper) + 112 + ii (modern paper)”  (University of Pennsylvania MS. Codex 708).

Dimensions. Frequently two dimensions are provided, one in parentheses. “266 x 181 (191 x 118) mm.” (University of Pennsylvania MS. Codex 708). This indicates the height and width of first, the overall page area, and second, the written (or ruled) space. These dimensions may be those of a typical page, or a range may be given if dimensions vary significantly. Catalogues do not always specify whether the written or ruled space is measured (these may vary significantly), or specify by what criteria the space has been measured. (Is it from the top of the top line ascender to the bottom of the bottom descender? The top of the top line minim to the bottom of the bottom line minim, or to the top of the bottom line minim? And so on…)

Collation. The collation structure of a manuscript can be expressed in several different ways.[2] One approach is to describe the quires in prose (e.g. Oxford, Balliol College MS. 4). Alternatively and perhaps more commonly a formula may be used. English catalogues typically use a formula in which the quires are numbered in sequence with superscript numbers denoting the number of leaves in each quire. To take a simple example,  Cambridge, Trinity College B.3.10 is a manuscript of 22 quires, each of 8 leaves.


The formula in modern German catalogues, conversely, commonly uses roman numerals to denote the number of bifolia in a quire; a roman numeral to denote a sequence of multiple quires with this number of leaves; and a superscript number to denote the folio at which the sequence begins or ends. For example Weimar, Herzogin Anna Amalia Bibliothek, Fol max 11 consists of six quires of ten leaves (five bifolia) ending at folio 60, followed by one quire of six leaves (three bifolia) ending at folio 66.


Where it is codicologically significant, a formula may also describe the arrangement of hair (H) and flesh (F) sides of parchment.

Quires containing missing or added leaves can be described in various ways and care needs to be taken that a formula has been interpreted correctly. Missing leaves are usually designated by their former position in the quire, e.g. xviii 8 (wants 7,8) means the original seventh and eight folios are now missing. In the German notation style, however, ‘(IV – 1)’ denotes a quire of originally four bifolia from which one leaf is now missing (leaving a quire of seven leaves): the location of the missing leaf or leaves is not always specified. The cataloguer may additionally comment on how the leaves were probably removed (‘cut out’ perhaps, or ‘missing’ or ‘wanting’ in an unspecified manner) and whether or not they were probably blank. ‘Cancelled’ usually means that a (probably) blank leaf has been cut out. Added leaves similarly can be described in several ways. M. R. James used an asterisk: f 8 (+8*) denotes a leaf added after the original eighth folio of the quire. A similar notation takes the form 2 8 (+1 added after 2) (MS. Lyell 89). Alternatively an addition may be identified by its current position: 32 2 (+3)  means that one leaf, now the third and last in the quire, has been added to the original two (Huntington EL 9 H 10).

Secundo folio. (‘sec. fol’, ‘2o fo’, or ‘f. 2’) The opening words of the second folio of the manuscript, used as a unique identifier in some medieval and later library catalogues and provided in some modern manuscript catalogues. There is no standard place in the description for this information to appear: different examples include in the heading, in the physical description (Cod. Bodmer 19) [pdf], or at the very end (Huntington El 9 H 9)


This information falls into two categories: origin (where and when the manuscript was originally produced) and provenance (its later ownership).

Older catalogues typically give only the date of origin, more recent ones date and place (where the latter can be ascertained). These will not necessarily be found in the same section of the description: for example, this description of Cod. Bodmer 1 [pdf] has the date (XIIe s.) in the first paragraph, the place of origin in a separate paragraph headed ‘origine’.

Conventions for expressing dates vary. Dates such as “14th century, early” or “s. xiv” are unproblematic (s. =  century, Latin saeculum). Less obvious are (1) the common abbreviations in (ineunte, the beginning  of the century), med (medio, the middle of the century), and ex (exeunte, the end of the century). ; (2) the use of superscript numbers to denote either the first or second half of a century (e.g. s. xii 1) or  quarters of a century (s. xii 1/4 and so on). In some catalogues ineunte is taken as equivalent to the first quarter, exeunte to the last quarter, and medio to the middle two quarters.  Where two centuries are given, a slash between them usually indicates that the date falls between the end of the first century and the beginning of the next (e.g. s. xii/xiii, a date which other catalogues might give as c1300), whereas a dash (s. xii-xiii) would indicate a date of production covering a wider period in both centuries.

Provenance history is typically not recorded in detail in older catalogues, although these may well mention some early owners. In more recent catalogues provenance information is typically presented, as far as possible, in chronological sequence from the earliest owners to the manuscript’s acquisition by its current owner, regardless of the sequence in which relevant information appears in the manuscript itself (e.g. MS. Lyell 22). Less commonly the relevant entries will be given in manuscript rather than chronological order.


One final issue concerns composite manuscripts, i.e. where the codex as catalogued consists of two or more originally separate manuscripts. Older catalogues may signal this rather obliquely. In this description of Bodleian Library, MS. Canon. Misc. 105, for example, the horizontal line after item 4 indicates that items 5-6 are a distinct unit. The dates given in the heading should be applied to each unit in sequence, so in this case the first codicological unit is thirteenth century, the second fifteenth century. More recent catalogues use a system of letters of numbers to denote the different units, as in Lyell MS. 21 (p. 46, p. 47) where the continuation to Ranulf Hidgen’s Polychronicon is a separate manuscript.


  • The Bodleian Library summary and quarto catalogues have all been digitized, as has the Lyell catalogue
  • Many of M. R. James’s catalogues are available via the Internet Archive, and those for Trinity College are also in the Wren Digital Library
  • W. Dutschke’s Guide to Medieval and Renaissance Manuscripts in the Huntington Library is online
  • e-codices includes a large number of German-, French- and some Italian-language descriptions which have been electronically encoded and can often also be viewed as PDFs.
  • Manuscripta Medievalia contains a large number of digitized German catalogues (under the Handschriftenkataloge tab).

[1] Most helpful is R. Clemens and T. Graham, An Introduction to Manuscript Studies, pp. 129-34.
[2] For more detail see F. Bischoff, ‘Methoden der Lagenbeschreibung’, Scriptorium 46 (1992), 3-27.

Dr Matthew Holford, Bodleian Library, Oxford


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s