The Collation

Research and Exploration at the Folger

Deciphering signature marks

So, as those of you who have spent any time working with early modern printed books probably recognized, this month’s crocodile mystery focuses on signature marks. Below is the photo I posted last week, now with the signature mark circled in red:

signature mark

Signature marks are those letters, numbers, and sometimes symbols at the bottom of the first portion of gatherings to help binders assemble the sheets of a book into the right order. For those of you who’ve been reading along with our various earlier posts on impositions, you’ll remember that books in the handpress era were printed not as single leaves, but as sheets with varying numbers of pages per side. (It’s certainly much less labor- and material-intensive than printing one page at a time.) What this means is that one sheet of paper might contain, once properly folded, 2 leaves (a folio), 4 leaves (a quarto), 8 leaves (an octavo), 12 leaves (a duodecimo or twelvemo), etc. 1 The key thing is to keep track of folding your sheets in the right direction so that the leaves end up in the right order and to keep track of the order of your sheets so that when the multiple gatherings are bound together, the text proceeds in the correct order. 

Sometimes pictures are easier. The first one below is showing the front and back sides of an uncut newsbook showing a quarto imposition; the second shows an octavo imposition for The Rape of Lucrece and was generated using Impositor. If you imagine the right half folding under the left half along the middle axis, you can visualize how the pages line up (it’s easiest to see what I mean when you look at the blank leaf that is the back of the title page of Lucrece since it shows the bleedthrough from the other side.) The best way to get used to impositions is by folding them yourselves, so go ahead and print these out as double-sided sheets and try folding them so that the page numbers (shown in blue) end up in the right order. If you just want to stare at the details, clicking on an image will enlarge it in a new window or tab. 2

4to imposition from an uncut newsbook


8vo imposition generated by Impositor

I’ve shown the page numbers here in blue for those of you who don’t spend all your time imagining these formats as sheets, but our main focus in on the signatures, here shown in red (both are oriented so that they “read” in the same direction as the text on the pages). I’ve depicted the signatures only on the pages that would have typically included them in that work. For the first example, that means the first 3 leaves; for the second example, the first 4 leaves. (You might notice that in the second example, A1 and A2 aren’t actually signed, but we can infer what they would be pretty easily. 3 ) If you try folding the sheets yourself, you’ll quickly see that you don’t need more leaves to be signed—once you’ve got it folded that far, the rest of the leaves fall in order and so don’t need to be signed. The other thing to know is that the numbers indicate which leaf in a gathering it is, and the letter indicates where this gathering falls in the series of gatherings that make up the book. Early modern books typically use a 23-letter alphabet, treating I/J as one letter, U/V as one letter, and omitting W. In Lucrece, we would expect the A gathering to be followed by B, then C, etc., until the end of the book (which, in this case, is the D gathering).

Today we generally use signatures to locate ourselves in a book. 4 So if I wanted to refer to where the table of contents falls in Lucrece, I would describe that as A3v (that is, the verso side of the third leaf in the A gathering). The poem itself begins on A4r (the recto side of the fourth leaf of the A gathering; those of you with keen eyes will see that A4 has been missigned as B4, but that’s clearly an error). 5

So, what’s so interesting about signature marks that I would use one as a crocodile mystery? Sure, they tell us about the sewing structure of the book—the number of leaves in a gathering and the number of gatherings in a book. But they can also tell us about where the book was printed, and that’s where things get exciting. The general technology of printing was the same across early modern Europe and the general format of signatures was the same, always indicating the gathering and the leaf number within that gathering and generally appearing on the bottom of the recto side of the first half of a gathering’s leaves. But within that basic structure there are variations. Are leaf numbers indicated with roman or arabic numerals? Are the preliminaries signed with letters of the alphabet or with symbols? Is the first half of the gathering signed or the first half plus one?

In 1966, R. A. Sayce published the results of his examination of over 2800 books printed in locations in 10 different countries, an effort to determine whether scholars can identify localized compositorial habits in books printed between 1530 and 1800. 6 His study showed that there are variations in signature marks, catchwords, pagination, press-figures, and dates in imprints, and that these variations are tied to geographical regions. For instance, English books typically start signing the text with B, saving A for the preliminary materials; books printed in Antwerp frequently sign preliminaries with * but books from Paris rarely do; and if you’re looking at a book using ¶ in the preliminaries, it’s likely to have been printed in Geneva.

If you look at the preliminary material in The Rape of Lucrece, above, you’ll see that it’s signed with A (and that it’s part of the same sheet as the beginning of the main text). But take a look at how the preliminary material in our crocodile book is signed:

The dedication, signed *ij

The use of the asterisk here suggests that, despite the book being in English, it might not have been printed in England.

So what book is this? It’s a 1578 English translation of Rembert Dodoens’s 1574 herbal, Cruydeboeck.

title page of A niewe herball

And we’ll know where it was printed because the imprint is right there on the title page: “At London by me Gerard Dewes, dwelling in Pawles Churchyarde at the signe of the Swanne. 1578.” Right? Ah, right. This is when it’s helpful to remember that imprint statements sometimes name the person responsible for causing the book to be printed (someone we might think of by the anachronistic term “publisher”), rather than the printer. And if we turn to the book’s colophon, we’ll see that’s exactly the case here:

the colophon to A niewe herball

And there it is, at last, plain as day: “Imrinted at Antwerpe / by me Henry Loë Bookeprinter, and are to be solde at London in Povvels Churchyarde, by Gerard Devves.” So that might explain why the book looks both English and Flemish. It was printed in Antwerp (where, among other features, they use * to sign preliminaries) by Henry Loe (or, to use the Dutch form of his name, Hendrik van der Loe) and to be sold by Gerard Dewes (also known as Garrat Dewes or D’Ewes) in London.

opening 2H3v-2H4r

A couple more notes about variants in signature marks: In this instance, we’re looking at the second series of the alphabet—this book is long enough that 23 letters aren’t enough to get us to the end of the text, so the printer has started the alphabet again, differentiating it from the first series by doubling the letters. The first time through, the printer used “H” and here “H” is followed by its lower-case equivalent, “h.” Depending on where a book was printed, the series might be signed “HH” or “2H.”  You might also note that, as in the other examples we’ve seen from this book, the last instance of the roman numeral “i” is marked “j”: “iiij” or “ij” instead of “iiii” or “ii”. This, too, is a variation that might help pin down a book’s place and date of printing. According to Sayce, “ij” became common during the sixteenth century, and “ii” suggests an earlier, rather than later date; its appearance after 1580 suggests the book came from Geneva. The use of “iiij”, rather than “iv”, suggests a book printed before the middle of the 18th century. The use of roman numerals, rather than arabic ones, is also a tell: books in Paris typically used roman numerals up the end of the 18th century, but books in Lyon used arabic numerals until 1680; Geneva doesn’t show a marked preference for one style over the other, but books in Basle tend to use arabic figures; British books used roman numbers in the 16th century, with arabic numerals starting to appear in the 17th century.

So, why would you want to know how printing practices might differ between local communities? Well, what if you were looking at a work with the imprint “Printed in Paradise, 1624“? What do you do with a book that claims it was printed in Frankfurt but is a satirical look at London? You’ll need to look beyond the imprint to find clues about where the work was printed and who might be responsible for it. The compositorial practices that Sayce examines are one category of information that can help us understand the nature of clandestine printing and the circulation of ideas.

Here, just to amuse you, are examples of a few more variants of what you might see in a signature mark. For each of them, if you click on the image, you’ll open up a zoomable image in a new window which includes catalog information as to what the book is.

using  ¶ to sign first gathering
mixing upper-case and lower-case letters; separating gathering from leaf numbers with periods
uses lower-case letters for first series
uses lowercase letters for the first series and arabic numerals
using “u” instead of “v”

when one, two, or even three times through the alphabet isn’t enough
One last word about our crocodile example. Although general practice is to have the signature mark and the catchword appear in the same line (the “direction line”), that line is usually just below the last line of the text; see the images above to see what I mean. But in our example, there isn’t a separate direction line: the signature mark and the catchword appear in the same line as the last words of the text. That’s not characteristic of the entire book (see the photo below for another example from A niewe herball), but is probably the result of factors specific to this page. John Russell, in his commenton the last post, suggests that it’s because the printer didn’t want to orphan the last couple of words of this paragraph on the next page, and so squeezed them in here, and that sounds right to me. There’s more to be said about catchwords and the variations you might find there, but I’ll save that for another day!

direction line below the text
  1. A leaf, of course, is the piece of paper that you hold in your hand when you’re turning it over to read a new page; a page is one side of a leaf.
  2. If I were to do these over again, I would actually depict the outer form (the one showing the first page) on the right and the inner form on the left so that you would imagine the left side folding under the right side, which is generally the way in which you would fold your sheets of paper: fold in half perpendicular to the long side of the sheet, bringing the left side behind the right side; turn your paper so that the long side is along the top again, and repeat until you get to the proper format.
  3. Title pages typically aren’t signed, so that’s not surprising; I would have anticipated A2 to be, but I’m guessing that the layout of the dedication prevented that.
  4. Since pagination and foliation were not common in the earliest books and were not necessarily reliable even in later early modern books, the general rule of thumb for early books is to cite text by signature mark.
  5. Sometimes people get confused and want to refer to the page facing A4r as A4v but that’s wrong. Just remember the long-hand of these signs: the number refers to the leaf and the r/v refers to whether the page is on the front or back of that leaf.
  6. R. A. Sayce, “Compositorial Practices and the Localization of Printed Books 1530-1800” The Library, Fifth Series, 21 (March 1966), 1-45. Reprinted, with corrections and additions, by the Oxford Bibliographical Society, Oxford, 1977.


  • Great post, Sarah! This should be a great resource for those trying to get to grips with Bowers and I will certainly use it this year for our students. Many thanks!

  • Very useful post, thank you. Wish I could have read it when I was doing historical bibliography at library school, would have made it much easier to get my head around what was going on!

  • Hi! I’m studying the Romanian books printed by a German (Petrus Barth) in Sibiu/Hermannstadt (RO) in the end of XVIIIth century. I’m interested in founding in which moment he printed the signature marks. I would like to know this thing because the signature would tell me in which moment the Forword was printed, for example. If I found the mentioned book (R. A. Sayce,….) I would be very happy. Where could I find it? Thanks.
    I would be glad to read more posts about technics of printing in the last centuries.

    • The full citation for the Sayce book is in the 6th footnote of the post. I’m not sure Sayce will fully answer your questions—most of his examples are from earlier centuries—but you never know. Good luck with your research!

    • That’s a great question! These conventions do shift over time, but the current practice is to say “leaf A4r”, which refers to the physical page. Bibliographers and catalogers are starting to use “signature” more specifically, to refer to the actual signature marking, rather than a page as a whole. But a lot of people still use “sig. A4r” and it’s perfectly well understood.

      • Thanks Abbie, the article and the answer.

        I can see the benefits of reserving “signature” for the mark, but for citing the side of a leaf, wouldn’t “page” be most accurate? Doesn’t a leaf have two pages, the r-side page and the v-side page?

        Separately, but related, I need to cite a book that has signature marks at the bottom and leaf numbers at the top. The leaf numbers look just like page numbers and would be mistaken for such if you did not flip through a few leafs. It seems to me logical to cite the verso side of leaf 78 as “p. 78v.” If the printer had put 78a and 78b on each side of the leaf, wouldn’t we cite the sides as “p. 78a” and “p. 78b”? Wouldn’t we do that today if someone printed a book that way? And in newspapers where page numbers begin anew with each section (as signature number do with each gathering), we cite a page as “p. A2” not as “sect. A2.”

        It seems to me “p.” works well for all cases. No?

  • How common is it for signature marks to switch to a different format in mid-gathering? I’ve only consciously noticed this once (in the first gathering of the first quarto of Sidney’s ‘Astrophel and Stella’, which for no obvious reason goes from Aii to A3), and it struck me as very unusual at the time and made me wonder whether this meant that there was some sort of interruption during the printing process – or just that the typesetter responsible for that page happened to be having a bad day. Or does that sort of thing happen more frequently and I just failed to notice it because I haven’t been paying enough attention?

Leave a Reply

  • (will not be published)