University of Virginia Library


As the sequence tables of watermarks by gatherings in the plays illustrate (ap-
pendix 1, below), the paper stocks of the Pavier quartos present a baffling array
of different situations and combinations.[39] The Whole Contention is printed on an
entirely homogenous stock; Pericles and The Yorkshire Tragedy are both on a similar
and slightly mixed stock of five papers (each has one mark that doesn't appear in
the other); The Merchant of Venice is printed on a moderate mix of ten papers; and
the four remaining titles exhibit miscellaneous mixtures of between fifteen and
eighteen papers. Similarly, there is considerable variation in individual gatherings.


Page 175
A few, including Per gathering T, MND gathering H, MWW gathering E, and SJO
gathering K are, in the copies I have examined, printed on a single paper; a few
other gatherings—Per X and Z and YT C, for example—are nearly as homog-
enous, with a second paper appearing in only in one or two copies. At the other
end of the spectrum I have found nine different papers in MWW gathering C and
eight in H5 gathering E. As can be seen in the summary table of watermarks in
all plays (appendix 3), individual papers also vary considerably in their frequency.
The most frequently occurring is No. 23, since every gathering of each of the
twenty-three copies of The Whole Contention I have examined appears on this paper,
and the same stock also appears in the first few gatherings of Pericles. The most
ubiquitous paper, however, is No. 2, which appears in six of the eight quartos, in
a total of 26 gatherings. No.1\14 appears in five quartos (17 gatherings), No. 6.1
in five quartos (12 gatherings), and No. 5 in four quartos (10 gatherings). On the
other hand I found only a single example of each of a dozen fugitive papers.

While the distribution of the various paper stocks both within and between
titles is not straightforward, several patterns emerge that suggest a complex pro-
duction system that utilized cast-off copy and the concurrent printing of multiple
plays to keep Jaggard's two presses busy. Peter Blayney has demonstrated through
headline analysis a pattern of very regular alternation in the use of skeleton
formes for the printing of the Pavier quartos. With few exceptions, sheets are
printed and perfected using the same skeleton, and "alternate sheets use alternate
skeletons."[40] This pattern of alternation is, to an extent, recapitulated in the pa-
per stocks. After The Whole Contention, printed exclusively on No. 23, the printing
of Pericles begins with a new stock, No. 2, eked out with a bit of leftover 23. At
gatherings X and Y of Per two new stocks emerge, Nos. 7.1 and 6.1 respectively.
But then gathering Z goes back to stock 2 and sheet 2A is predominantly on
1\14. The pattern of alternation also includes The Yorkshire Tragedy, whose print-
ing had probably begun about the time Per had reached sheet Y or Z, with that
same general mix of papers appearing in YT as in the last four sheets of Per.[41]
The pattern continues with Merchant of Venice, whose paper stocks in the first few
gatherings alternate with those at the end of YT. The following table compares
the stocks for the last four gatherings of Per with the four gatherings of YT and
the first four of MV. The predominant mark is listed first and should be given
most weight in comparisons.

Per   2A 
7.1, 23  6.1, 23  2, 1\14  1\14, 2, 23, 6.1 
YT   tp/D 
1\14, 2, 23, 6.1  1\14, 2  2, 1\14  6.1, 2, 23 
2, 1\14, 7.1  6.1, 2, 15\16.2  7.1, 1\14  6.1, 1\14 


Page 176

The suggestion is that each of the two presses had a particular stock of paper laid
out, with the printing and perfecting of sheets from Per alternating (though not nec-
essarily regularly) with the beginning of YT, then the end of YT with the beginning
ofMV, with movement back and forth between presses as the formes became ready.

In the wake of Greg's revelations about the Pavier quartos, W. J. Neidig in
1910 undertook a minute examination of their title pages, eight of which bore the
"Heb ddieu" printer's device. He determined that the device and the attribution
"Written by W. Shakespeare" had been left standing for most of the production of
the Pavier quartos, with the titles and publisher/date line being changed around
them as needed. By carefully tracking the series of changes, Neidig was able to
propose the order of their printing: The Whole Contention, Pericles, The Yorkshire Trag-
edy, The Merchant of Venice, The Merry Wives of Windsor, King Lear, Henry the Fifth, Sir
John Oldcastle.
He contended that A Midsummer Night's Dream, which has a different
device on its title page, was printed separately from the rest of the series.[42]

Neidig further argued that the order of printing for the first three titles was:
"Whole Contention (text and title page), Pericles (text only), Yorkshire Tragedy (text
only), Yorkshire Tragedy (title page), Pericles (title page)."[43] The paper evidence
shows that this could not have been the case. The source of Neidig's error
was the assumption that the title pages of Per and YT were printed separately
from the texts of the plays, but in both cases the paper evidence makes clear
that the title pages were printed with their final gathering: 2B of YT and D of
Per. Surprisingly, the paper distribution shows that the final gathering of Per was
actually printed first, indicating that the copy had been quite accurately cast off,
a relatively straightforward task when working from printed copy (though neither
is a page-for-page, line-for-line reprint of its setting copy: Per Q3, 1611; YT Q1,
1608). The Per gathering 2B appears on the same mix of Nos. 2 and 23 as the
first two gatherings of the text, R and S, and this mix occurs nowhere else in the
Pavier quartos. It strains credulity to imagine that Jaggard held back a stack of
this precise mixture until the end of the play so that its watermarks would match
those at the beginning of the text. It is also possible that the final gathering of YT
was also printed first, following Per gathering Y, which appears on the same and
newly introduced mark 6.1. Since YT is by far the shortest of the Pavier quartos,
at only three and a half text sheets, such a procedure would have guaranteed that
the "Heb ddieu" device would be ready in plenty of time for setting the title page
of MV, the next play to be printed. Whatever the precise order of sheets through
the press, the evidence for concurrent printing of these three plays is strong.

Neidig placed the printing of MWW after MV because its title page uses the
line "Written by W. Shakespeare." with the "Heb ddieu" device and because
three other titles using that specific attribution had already been printed.[44] I am
inclined to agree, though evidence of the paper stocks offers nothing definitive.
Gathering A of MWW is printed mostly on No. 3, the only gathering in which it


Page 177
predominates,[45] though one copy has No. 8 and another No. 6.1, the marks found
in the final gathering of MV. So there is at least a tenuous link between the end of
MV and the beginning of MWW. If the order MV > MWW is correct, however,
there is small anomaly. MV is the first of the Pavier quartos to bear a false date,
and it seem peculiar that, having decided that subterfuge was necessary, Pavier
and Jaggard would have reverted to the actual date of 1619 for MWW before then
falsely dating the four remaining titles. It is at least possible that the first gather-
ing or two of MWW had been printed off before work began on MV, which was
completed before work resumed on MWW. The suggestion has at least tenuous
support since gathering B of MWW and gathering A of MV share a similar mixed
stock of predominant 2 with an small admixture of 1\14, though MV also has a
bit of 7.1 not present in MWW.

The printing of MND has proved the most difficult to place in the sequence
because it alone among the post-Contention plays does not use the "Heb ddieu"
device on its title page. Neidig thus assumed that for some reason it was printed
independently. Peter Blayney, however, has provisionally suggested that, on the
basis of Greg's distribution charts, MND was most likely to have been printed be-
tween MWW and Lear. [46] My own fuller sample essentially corroborates Blayney,
though I believe it more likely that MWW and MND were printed concurrently.
MND may actually have been begun either slightly before or simultaneously with
MWW, since its first gathering is printed predominantly on No. 6.1, the primary
stock for the penultimate gathering of MV. This would explain the use of a second
publisher's device for the MWW title page, since the "Heb ddieu" would have
been unavailable. Though the stocks for both MMW and MND are quite miscel-
laneous both between and within gatherings, the central gatherings of both plays
share several similarities. The related pairs 24\27.1 and 24\27.2 appear only in
MWW gatherings C, D, and F and MND gatherings C and F,[47] and No. 5 appears
in several gatherings of both, notably in MWW gathering E and MND gathering
D, which are both printed exclusively (or nearly so) on this paper.

The order of the final three titles is essentially uncomplicated, though there
are again some anomalies. It is for example somewhat curious that Lear, the next
play in the sequence, has its first two gatherings printed on the mixture of Nos. 2
and 1\14, which had appeared as early as Per gathering V, then in YTgatherings
A, B, C, MV gathering A, and MWW gathering B. While the reappearance of
a particular mix later in the overall printing sequence may be a random occur-
rence, it is at possible that the first few gatherings of Lear were set and printed
off shortly after MWW gathering B, as soon as the "Heb ddieu" device became
available, and that MWW and MND were then finished off before work resumed
on Lear. This procedure would further explain the use of a second device for
the MND title page. Such a complex matrix in which three titles were simulta-
neously in production would not seem an impossible strain on an experienced
printing house, and may have been a useful expedient in maintaining a balance


Page 178
between composition and presswork. The succeeding play, Henry 5, was begun
while Lear was still in press, probably between the printing of gatherings F and H
of the latter. It is possible to designate the moment rather precisely since the first
gathering of H5 is printed exclusively on stocks 12\19.1 and 12\19.2, the precise
and exclusive mix of Lear H and F, and this particular stock is found nowhere
else in the Pavier quartos. H5 and Lear are also the only plays with significant
unmarked stocks: they share 0.1; Lear also exhibits 0.2 and 0.3 and H5 has a bit
of O.4.[48] Sir John Oldcastle was the last of the Pavier quartos to be printed, largely
on the related stocks 15\16.1 and 15\16.2. One curiosity of the SJO stocks, as
noted above, is the substantial admixture of fugitive papers in otherwise largely
homogenous stocks. Ten new papers appear in its ten gatherings, most of them
in only one or two exemplars. It does seem possible that, arriving at the end of
this substantial project, Jaggard tidied up his warehouse by throwing in whatever
odd bits were lying about. But perhaps most curious is that, having done so, the
final gathering of the Pavier quartos is printed exclusively on stock No. 5, a paper
that appears in several previous gatherings and had not appeared unmixed since
MWW gathering E. It is worth noting that the final gathering of MND also ap-
pears on an unmixed stock, No. 2, and the final gathering of MV is nearly so.

While I have not undertaken a comprehensive survey of the paper stocks that
passed through Jaggard's printing house in and around 1619, a few observations
may be made to contextualize the production of the Pavier quartos. Two titles
in particular seem to have been printed at much the same time as these quartos.
The anonymous Troubles in Bohemia, and diuers other Kingdomes. Procured by the diuel-
lish practises of state-medling Iesuites
(1619; STC 3213) is a six-gathering quarto that
is printed mostly on stock 23, with one sheet of #2, the same mix that is found
in the first few gatherings of Per. This indicates that the matrix of concurrent
printing that clearly took place with the Pavier quartos may have encompassed
non-related titles as well. Similarly, The second part of the booke of Christian exercise,
appertaining vnto resolution
(1619; STC 19388), a 12o, has eighteen of its twenty-one
gatherings printed on stock #5, a paper that appears largely or wholly unmixed
in MWW E, MND D, and SJO K. The likelihood, then, is that this twelemo was
printed sometime in the mid-to-late stages of the Pavier quarto project. A third
possible and intriguing connection with the Pavier quarto paper stocks might be
made to John Selden's The Historie of Tithes, which saw four editions reach print
in 1618, if their title pages are to be believed (STC 22172, STC 22172.3, STC
22172.5, STC 22172.7). I quote, with some omissions, David Gants's account of
the difficulties surrounding the publication of this controversial work:

This book was originally produced in [William] Stansby's house, but sometime during
Christmas 1617 the Bishop of London raided the establishment, confiscated the paper and
type being used to print it, and shut down the business for a time…. When the work
appeared the next year (with a long preface refuting the objections raised by others and
some modifications to the text itself), the only identifying features on the title page were
its author and the publication date. The printer even went so far as to explicitly distance
himself from the text with a short "Printer to the Reader" coda…. Stansby's unfortunate
experience, along with the fear that further trouble might attend this book, apparently


Page 179
prompted the printer of these volumes to stay as far away as possible from this debate….
There is also reason to believe that at least one of the four editions was printed after 1618,
although confirmation of this supposition awaits further investigation.[49]

If it was not Stansby who, defying the authorities, printed these editions of The
Historie of Tithes,
might it have been William Jaggard? In the Folger copy 1 of the
first of these editions I found four of the Pavier quarto papers—numbers 4.3, 5,
1O\11, and 26, with the mixture of 4.3, 10\11, and 26 occurring in a substan-
tial run from gathering D to 2T (with an occasional admixture of other stocks).
Overall, thirty-six of its sixty-seven folio gatherings were printed on Pavier quarto
papers. Of course the same paper stocks can and do appear in the products of
more than one printer; but the conjunction of four identical stocks is at least
suggestive. The detective who wants to pursue this further investigation, should,
as W. W. Greg did a century ago, follow the paper trail, armed (as Greg was not)
with a clear, flexible, plastic ruler with a millimeter scale.


The appendixes present the paper data in several forms. Appendix 1 offers "sequence
tables" for each of the plays and is organized by copy and gathering so that both the sequence
of papers for any given copy and the variety of papers that appear in a given gathering of copies
examined can be readily seen. Appendix 2 summarizes play-by-play the data from the sequence
tables to show the total exemplars of each watermark that appear in a given copy and the total
occurrences of that mark in all examined copies for each play. Finally, appendix 3, a summary
table of watermarks in all plays, is organized by watermark and play so that the presence and
distribution of all watermarks observed in the Pavier quartos can be seen at a glance.


Peter W. M. Blayney, "'Compositor B' and the Pavier Quartos," The Library 5th ser.
27.3 (Sept. 1972): 181–206; esp. pp. 197–198. The only major exception is Sir John Oldcastle,
which uses a single skeleton throughout.


Blayney had noted that "It is possible that the shorter YT might have been printed
concurrently with PER or during an interruption …" (p. 196).


"The Shakespeare Quartos of 1619," Modern Philology 8 (1910): 145–163.


Neidig, "Shakespeare Quartos," p. 156, n.1.


There is no doubt that this particular combination of attribution and device had been
left standing during the printing of Per, YT, MV, and MWW. I printed off Neidig's photographs
of these four pages and, using the "poor-person's collator," superimposed the pages in every pos-
sible combination while holding them up to the sunlight. The setting in each case is the same.


There are a few scattered sightings of this mark in C and D of MWW and A of MND.


Blayney, "'Compositor B,'" p. 197.


The sole exception is sheet C of the Newberry copy of Lear, which has one of the
24\27.1 pair.


There are single examples of fugitive unmarked papers in MWW sheet G and MND
sheet B.


David L. Gants, "A Quantitative Analysis of the London Book Trade 1614–1618,"
Studies in Bibliography 55 (2002): 195, n. e.