University of Virginia Library

ABOUT the middle of the last century, a bibliophilic "find" was made in Holland in the course of liquidating the stock of a bookseller's warehouse;[1] here were discovered a considerable number of copies, still in the original sheets as issued from the press, of a Valerius Maximus "Amstelodami, Typis Danielis Elzevirii, Sumptibus Societatis" with the date 1671. Two editions with this same information and date are known which agree with each other page-for-page and mostly line-for-line; it is certain, however, that the edition with signature *2 recto printed in roman type and lacking a headpiece[2] on that page was not printed by the Elzevir press. As Gustaf Berghman has shown,[3] this edition--to which

the unbound sheets belong--was very probably printed at Amsterdam by Johann Blaeu. Apparently here, as in a number of other instances, a group of booksellers banded together to undertake a joint venture; though all the copies of the several editions bore the Elzevir imprint, it seems clear that now one member of the association, now another, would reprint the texts as further printings were required. Consequently any book printed "Sumptibus Societatis" may or may not have been put out by the Elzevir press. In any event, it is apparent that those copies not actually printed by Daniel Elzevir do not constitute "forgeries" nor do they belong to "counterfeit editions" in the usual sense, since Blaeu could hardly have continued this practice without the Elzevirs becoming aware of it, and the relations between the two firms remained cordial to the very last. However, as Berghman has sadly remarked:[4] While the value of these books may well be relatively slight from a commercial point of view, it is quite considerable for the evidence which they afford in regard to the method of printing employed in their production. Indeed, they offer concrete proof for a printing practice which Ronald B. McKerrow discussed but did not illustrate.[5] He further stated (p. 168) that "it is undoubtedly possible for a book to be printed as a 12mo (i. e. for the leaf to be one-twelfth of a sheet) and yet to be sewn in 8'S, though I cannot give an instance of such a thing."[6] Here, then, is an example of such practice, though in this particular case it is just double that which McKerrow had envisaged.

The Pierpont Morgan Library (no. 21246) possesses one of the numerous extant copies of this Valerius Maximus in its original state, the sheets still uncut but sewn together. The very unusual and awkward appearance of the book is due to the fact that three sorts of gatherings with varying measurements are bound together. The collation,[7] as conceived by the printer, may well have been: *4 A-V8 X4, though quire * was not necessarily always bound first.[8] The types of gatherings are these:

Group 1--quires: A C F I M P S
Measurements:[9] folio 1 measures 3 3/8 X 5 inches
" 4 " 3 3/8 X 4 3/4 inches
folios 2, 7, and 8 measure 2 1/2 x 5 inches
" 3, 5, " 6 " 2 1/2 X 4 3/4 inches


Group 2-quires: B D G K N Q T
Measurements: folio 2 measures 3 3/8 X 5 inches
" 3 " 3 3/8 X 4 3/4 inches
folios 1, 7, and 8 measure 2 1/2 x 5 inches
" 4, 5, " 6 " 2 1/2 X 4 3/4 inches


Group 3-quires: [*/X][10] E H L O R V
Measurements: folios 1 and 2 measure 3 3/8 X 5 inches
" 3 " 4 " 3 3/8 X 4 3/4 inches
" 5 " 6 " 2 1/2 X 4 3/4 inches
" 7 " 8 " 2 1/2 X 5 inches

The watermarks occur either in the upper margins (groups 1 and 2) or in the outer ones (group 3); the chain-lines are horizontal. The only possible way that these sheets could have been printed is set forth in the accompanying figure.[11] The sheet would have to be cut-and indeed was cut-along the dotted lines. The small circles marked on the dotted line of the central fold indicate the pinholes. As usual these occur at different distances from the edges of the paper so as to insure perfect register when the sheets were perfected;[12] the holes are found approximately 1 1/4 inches from one edge of the paper and 2 1/4 from the other. The broken lines in the form of a rectangle and of a diamond indicate the approximate alternative positions of the watermark. The signature marks set down are those of the printed pages, not of the formes as they were imposed; those in square brackets indicate the pages on the verso as printed and turned face up.

After the sheet had been perfected and cut as indicated, two of the resulting four parts could be folded as ordinary octavos, though with the chain-lines running

horizontally. The remaining two parts had the appearance of quartos, with the chain-lines (as usual) horizontal; the watermarks, however, would be found in the outer margins instead of in the inner ones and might often be cut off in the trimming. These two sections were folded together (i. e., one sheet within the other)[13] to make a gathering of eight leaves. These facts would probably not be easy to determine once the sheets had been trimmed down to make the completed book, bound and ready for sale.

The original sheet of paper must have measured approximately 16 3/4 X 19 1/2

inches, a size which does not appear in McKerrow's standard work.[14] Clearly it was not anticipated that the finished book would measure more than 4 3/4 x 2 1/2 inches, so that a sheet slightly larger than 15 X 19 inches (approximately that of the modern "crown" sheet) would have sufficed for the printing. This would give an overall measurement of 285 square inches as against more than 326 such inches actually used. Accordingly about an eighth of the paper fell away as wastage before the knife of the trimmer. That Blaeu could afford so great a loss in his stock of paper (about 12 1/2 per cent.) would seem to indicate that paper was relatively cheap in Holland in those days.

The uncut sheets of this Valerius Maximus have thus given us some insight into the workings of a late seventeenth-century press. When we are enabled to deduce the method of imposition and printing, that of the cutting and the folding of the paper, the exact location of the pinholes relative to the edges, the size of the original sheet and the amount of trimming it was expected to suffer- and possibly other details which may have escaped my notice-then one can hardly refer to these sheets as being wholly without value.