Skip to main content
The Collation

Signature statements in book cataloging

Today’s post returns to the cliffhanger at the end of Tuesday’s Physical description in book cataloging overview: if [4], CXXII leaves : ill. ; 31 cm (fol.) forms a complete physical description in a library catalog, then what’s up with a4 A-O8 P10 and where does it fit in?

a4 A-O8 P10 is an example of a signature statement, sometimes also called a collation (one of the many meanings of “collation” for which this blog is named). It records how the printed leaves were meant to be folded and gathered for binding. In the case of a folio format like the 1513 Venetian edition of Macrobius we’ve been looking at, the folding is simple: each sheet that comes out of the print shop gets folded in half, creating two leaves of text (or four pages: each side of a leaf is a page).

Conventionally, printers included a key (usually letters of the alphabet) in the lower right of the sheet so that the binder could group all the unfolded sheets in order, and make sure nothing was missing. These alphanumeric keys are known as signatures. Here’s my scrap-paper mock-up of the sheets at the start of the Macrobius volume as if they were freshly printed and being gathered for binding:


In the back, I’ve put the sheet with the title page on top of the sheet with the signature mark a2 in the lower right corner. Front left are the four A sheets, marked A, A2, A3, and A4; front right are the four B sheets, marked B, B2, B3, and B4.

How do I know I’ve got a full set of sheets ready for binding? In the Macrobius example, I know for certain that there should be more than just these three gatherings because the printer helpfully included a register at the end, spelling out the sequence of gatherings for the binder—a A B C D E F G H I K L M N O P—followed by a note (in Latin) to say that each gathering is made up of four sheets except for a, which has two, and P, which has five. No need to guess how many sheets you’re supposed to start with.


Here’s my mock-up of the first three gatherings of the Macrobius after each has been folded. The gathering of two sheets has become four leaves, and each gathering of four sheets has become eight leaves:


Notice that the title page doesn’t have a signature mark: it’s obviously the starting point, so no need to fiddle with extra pieces of type to put an a at the bottom. Once they’re folded and gathered into alphabetic groups, you can  see why the last half of the leaves of a gathering also do not have signature marks: when four sheets are folded, the last four leaves inevitably follow the first four in the correct order.


The printer’s written description “a A B C D E F G H I K L M N O P” plus a note saying that each has four sheets except the first and last (which have two and five, respectively) makes sense if you don’t need to be succinct, and if you’re working with unfolded sheets. If you do need to be succinct, and if the sheets are gathered and folded into leaves, there is an internationally recognized compressed form that describes the same arrangement of leaves in a neat little package:  a4 A-O8 P10 as seen in the second line of the “notes” section of the Hamnet record for this edition of Macrobius:


Explained in words, that means gathering a is made up of 4 leaves, gatherings A through O have 8 leaves each, and gathering P has 10 leaves.

While signature statements form a standard part of a scholarly descriptive bibliography, they are an optional extra in library cataloging. Here’s the official word from Descriptive Cataloging of Rare Materials (Books):

7B9.1. General rule. Make a note giving details of the signatures of a volume, if considered important. Give these signature details according to the formula in Philip Gaskell’s A New Introduction to Bibliography (see p. 328-332), insofar as typographical facilities permit. Preface this note with the word “Signatures” and a colon.

Signatures: [A]4 B-C4 D2 E-G4 H2

Signatures: A-C4 D4(-D3) E-F4

Signatures: A-2Z8, 2A-M8

Signatures: [1-68]

(Comment: Volume is completely unsigned)

It is generally desirable to give signatures for incunabula, especially if identical signatures are not given in a standard bibliographic source. It is also desirable to provide signatures when a volume has no pagination or foliation.

Current Folger cataloging policy is to include signature statements on all hand-press era books unless they are multi-volume or unfeasibly complicated and not described in a published bibliography. It’s one thing for a cataloger to figure out  a4 A-O8 P10 by examining the book in hand, but taking the time to work out the collation of a book big enough to run through the alphabet more than once, and filled with printing oddities so that seemingly random leaves were added in strange places, or removed from others, would mean too many other things not getting a Hamnet record at all.

Making it even more time consuming for catalogers, and more difficult for researchers trying to read catalog records, the encoding standard for library catalogs cannot accommodate the special characters commonly used in collations. We’re lucky even to have superscript numerals; superscript Greek letters are out of the question. Consider the collation for Shakespeare’s First Folio. It’s complicated enough when superscripts and special characters are available, like in this blog:

 πA⁶(πA1+1 πA5+1.2) A-2B6 2C2 a-g6 χ2g8 h-v6 x4 “gg3.4″(±”gg3”) ¶-2¶6 3¶1 2a-2f6 2g2 “Gg6 2h6 2k-3b6

Here’s how it looks in Hamnet, where the special characters have to be spelled out:

[superscript pi]A⁶([superscript pi]A1+1 [superscript pi]A5+1.2) A-2B⁶ 2C² a-g⁶ [superscript chi]2g⁸ h-v⁶ x⁴ “gg3.4″(±”gg3”) [par.]-2[par.]⁶ 3[par.]1 2a-2f⁶ 2g² “Gg⁶” 2h⁶ 2k-3b⁶

So why do we do it, if it’s an optional extra? Think of it this way: it might be “just gravy” for some people, but for users of the Folger collection, the gravy is what makes it worthwhile.