University of Virginia Library

The Dramatic Piracies of 1661: A Comparative Analysis
by
Johan Gerritsen [*]

AS STUDENTS OF THE SEVENTEENTH-CENTURY ENGLISH drama are only too well aware, the year 1661 saw a sudden spring for one of the more lucrative ventures in commercial publishing, the pirating of popular plays. With the theatres once again open, and the stocks of play-books understandably low, here was a heaven-sent opportunity for the less scrupulous publisher, and there is evidence that he took it. A name that looms large in all accounts of the matter is that of Francis Kirkman. For a general account the reader may be referred to Professor R. C. Bald's article in Modern Philology, XLI (1943), 17-32, and to Mr Strickland Gibson's Bibliography of Francis Kirkman, Oxford, 1949 (Oxford Bibliographical Society), unless he wishes to immerse himself in Kirkman's own, and not unamusing, The Unlucky Citizen, 1673.

Kirkman, in fact, denied all responsibility, though not quite all complicity, stating that he had been drawn in by others and that he, for his part, only printed three play-books, which were his own proper copies. Whatever truth there be in this — and we may note that Mr Gibson accepts it more or less implicitly —, the owners of the copyrights were not impressed, took the simplest way out, and had Kirkman's stock seized. And Kirkman, as he tells us himself, never got any redress.

At this day, it is doubtful whether we shall ever know the ins and outs of the matter, but what can be said with certainty is that Kirkman, whether or not he was the principal culprit (and unlike any of the

other candidates with the exception of his printer, Thomas Johnson) can be connected with most of these piracies.

That this is so was borne in upon the present writer in a rather unusual way. On 17 September 1955, while examining the Folger Shakespeare Library's holdings in Beaumont and Fletcher, he came upon an excerpt from the 1647 folio, signatures 2K2-2M4v, consisting of exactly one play, Beggars Bush. And the point of the examination at that stage being watermarks, it was not less than startling to note that the marks in the excerpt did not belong in the folio. A cursory examination of the text, however, was enough to show that this was not an excerpt from the folio at all, but an independent print, made to resemble the folio as much as possible. Signatures and lay-out were the same, there was even the same row of fleurons at the head of the text, but the ornamental initial I was different, the tailpiece was different, and the spelling was rather more modern, beginning with the head-title itself, which reads BUSH, not BVSH.

Now there would be an obvious explanation handy in the recollection that both for the 1616 Jonson first folio and the 1685 Shakespeare fourth folio, under-printed sheets had been reprinted at a later date to make up complete sets when the stock of these sheets fell short. But this explanation seemed just a little too pat. For though the cases just mentioned are well-established, and though for the Jonson the resetting was much more extensive than has been realized, it seemed rather fortuitous that resetting should have been necessary not merely for six consecutive sheets, but for six consecutive sheets containing exactly one play.

These considerations seemed sufficient to reject the supplementary reprint theory as a working hypothesis, and to cast about for a better one. And in this connexion one fact was thought not without possible significance. Beggars Bush was printed in the 1647 folio, which means that Moseley, like ourselves, did not know of any earlier edition. The first known print outside the folio, in fact, was the quarto published by Humphrey Robinson and Anne Moseley, the owners of the copyright, in 1661. Now in some copies of this quarto there is a note on the title-page which did not look without its possibilities in the context:

What no-one had explained so far was why, of all plays, Beggars Bush should have been selected for reprinting and for carrying this note, unless that play itself had been pirated. And the accusing finger pointed at Kirkman seemed well worth the trouble of examining all the books printed for him and connected with him in the years 1660-1662 that could be found. The total came to just over twenty, and among these, two were of particular significance: John Dauncey's A Compendious Chronicle of the Kingdom of Portugal, 1661, and the same author's The English Lovers, or, a Girl worth Gold, 1662. The first of these gives Thomas Johnson for its printer, both are printed for Kirkman and one or two others, and both have at the end an identical catalogue of books printed for and to be sold by Francis Kirkman,[1] followed by an Advertisement which, with a fine insolence, runs as follows:

Though that is not its main significance in our context, it may be noted that the phrasing of the advertisement leaves an interesting loophole, for all that Kirkman really says, after having first been so explicit about 'my direction and order' and 'printing and publishing', is that he never printed any of them. Since he never was a printer, that, doubtless, is entirely true.

What is, however, of special importance for our investigation is Kirkman's amplification of the Beggars Bush accusation. Though the use of the words 'those other Playes' clearly implied that the play itself had been pirated, Robinson and Mrs Moseley did not actually say so. But Kirkman, the dramatic specialist, who in at least five of his publications

offered to furnish anyone with 'all the Playes that were ever yet printed', and who published a list of nearly 700 after Tom Tyler and his Wife, not only acknowledges the charge expressis verbis, but also does not deny the piracy's existence. And for a very good reason, for he is offering it for sale in the front of the very book, the Chronicle, in which the Advertisement first appears. There, in a list of works of which he has 'sufficient numbers', and heading the section Playes, we find: Taken by themselves, these two Beaumont and Fletcher entries might be, and generally have been thought to refer to excerpts from the 1647 folio. But surely, Kirkman would never have had 'sufficient numbers' — and the works are mentioned 'more especially' for this reason — of just two excerpts from the folio. There was thus, not only, a piracy of Beggars Bush, but also one, still to be recovered, of The Humorous Lieutenant.

What we know then, so far, is that our Beggars Bush was in existence in 1661, and probably appeared not very long before the quarto, as a member of a larger series of piracies. Now three piratical reprints, The Elder Brother, A King and no King, and The Maids Tragedy, are known with the imprint London, Printed in the Year, 1661, which at least leaves no doubt about their date. But five others have at various times been assigned to this year, though they all have the same imprint as the previous legitimate edition. They are, another Elder Brother, '1637' (STC 11067, Greg 515c); Loves Mistress, '1640' (STC 13354, Greg 504c); Philaster, '1652' (Greg 363g, not in Wing); and two reprints of The Scornful Lady, '1651' (Wing B 1608, Greg 334g; and Greg 334h, not in Wing). To these we may now add Beggars Bush, folio, and our hypothetical The Humorous Lieutenant, folio, making ten in all. And the only man ever to advertise them all, is Francis Kirkman.[2]

Links between Beggars Bush and the others are not far to seek, nor are links between the piracies and the other publications of Kirkman and his syndicate around 1661.

It is true that Beggars Bush is set apart from the rest by its format

— a pirate who wished to be undetected was obliged to reprint in folio — so that we need not look for the use of standing type, but enough distinctive features remain to make identification easy.

The main fount, to begin with, is a peculiar one. Basically a normal pica fount, measuring 81 mm for 20 lines, it is notable for the addition of small caps and titling caps cast on the same pica body, used indiscriminately instead of the normal size.[3] Apparently the fount had once fallen short on capitals, but the use of titling caps to make good the deficiency is highly uncommon and makes the fount very easy to spot.

Now this fount has long been known to occur in some of the 1661 piracies, but that is not its only appearance. We find it in both the '1637' and the 1661 Elder Brother, in A King and no King, in The Maids Tragedy, and further in Hells Higher Court of Justice (Wing D 27, 1661), The English Lovers, Tom Tyler, the Catalogue, all 1661, and finally in The Wits, printed for Henry Marsh, 1662.

We may further note that the titling B on 2K2va, l. 11 from below, of Beggars Bush was printed by the identical type which printed that on B3v, l. 19, of the '1637' Elder Brother, while the King and no King uses for its act headings the same fount as does the Beggars Bush, the identity being proved by the u of Tertius on 2L1v of the latter, which was put there by the same type that printed the second u of Secundus on B3v of the former.

Moreover, we find another link with Tom Tyler in that the same rather coarse ornamental I opens both plays, an I which also appears in Matthew Griffith, The Fear of God and the King (Wing G 2012),[4] which was printed for (and by?) Thomas Johnson in 1660.

Now Tom Tyler, in its turn, shares probably its watermark, and certainly its ornamental block of a vase and flowers, with the '1637' Elder Brother. The latter, again, shares another ornamental block and an ornamental initial N of the same series as the I with the 1661 Elder Brother, and on iA2 it has the identical setting of fleurons which we find at the top of A2v in The Presbyterian Lash, or, Noctroff's Maid Whipt, printed 'for the use of Mr. Noctroff's friends' in 1661. The Lash has some importance for dating purposes, as it was received by Thomason on 25 March. It is a skit, written to annoy the Rev. Zachary Crofton, and it has a preface signed by K. F. On the same principle

which turned Crofton into Noctroff, K. F. is customarily read as F. K., i.e. Francis Kirkman.

The Lash, besides assuring us that the '1637' Elder Brother was actually printed in March, 1661, further links up with Hells Higher Court through a defective large capital C on the title-page of both, and another defective italic capital C occurring on the title-page of the former and in one of the headlines of the latter. The '1637' Elder Brother, moreover, shares an initial B of the same series as before with the King and no King.

The English Lovers, next to be considered, is a more intricate case. Like the Lash and Hells Higher Court, it presents evidence of more than one text fount, but unlike these, it was not so clearly set up in one shop. No reservations need, however, be made in stating that the first five sheets of this octavo volume were printed in the same shop as the others now under consideration. On B1 we find the same large C already noticed in the Lash and Hells Higher Court; on B4 we find a large lower case O which recurs in the head-title on B1 of the 1661 Elder Brother, and on the title-page of this play we see that the d of Elder was printed by the same type which printed on D6v of the Lovers. Similarly, a large titling E on A4v recurs on the title-page of The Maids Tragedy. And to confirm the correctness of our identification of the text fount, we notice a pica titling S on C8.

These nine books, then, Beggars Bush, the '1637' and 1661 Elder Brother, Hells Higher Court, the Lash, A King and no King, The Maids Tragedy, Tom Tyler and its Catalogue, and the first section of the Lovers form a compact group and leave no doubt that they come from one shop. A similarly compact group can be found if we now examine the remaining piracies and the other productions of the Kirkman group.

The piracies still to be accounted for are the '1652' Philaster, the '1640' Loves Mistress, and the two reprints of The Scornful Lady, both dated 1651. The relation, if any, of the second Scornful Lady, Greg's no 334g, to the rest is not clear, though it is certain that it was set partly from the other (334h) and partly from the true 1651 edition. But the first Scornful Lady shares with Loves Mistress a damaged titling N of medium size, occurring on the title-page in both volumes, and, as already noted by Professor Bowers,[5] the fleuron band on A1v of Loves Mistress recurs, with the addition of two parentheses, on A2 of The Scornful Lady. Similarly, we find on A2 of Loves Mistress a certain

obliquely decapitated fleuron which recurs on A2 of Philaster, and also on A2 of A pleasant Comedy, called, The Two Merry Milk-maids, printed by Thomas Johnson and sold by Brook, Kirkman, Johnson and Marsh, i.e. the same consortium of Tom Tyler, 1661. And moreover, Loves Mistress now turns out to have similar links with the Milk-maids, of which we may mention the two E's of READER on A2 of the latter, which recur in PHENIX and EDWARD on respectively the title-page and A2 of the former.

Some further links in our chain are now easily come by, since they were already worked out by Professor Bowers in his article just cited, and we may add The Thracian Wonder, A Cure for a Cuckold and Gammer Gurtons Nedle without further comment, except for noting that they are all dated 1661 and all mention Thomas Johnson for their printer, with Kirkman and his group, including Johnson, for publishers. The year's productions can now be rounded off with A merry Dialogue between Band, Cuff, and Ruff, printed for F. K., 1661, another little gadfly affair like Hells Higher Court and, to some extent, the Lash, but far from new. It presents an immediate link with Philaster, as they have an identical fleuron setting and enclosing rules — with slight extra leading in Band, Cuff and Ruff — on their title-pages.

The only publication which now remains,[6] besides the one Scornful Lady, is Bottom the Weaver (Wing S 2937), which was printed for Kirkman and Marsh in 1661, could well belong to either group, but has nothing that will serve for a definite link.

There is, however, one further publication that should be linked to this group, and one which, between its covers, crystallizes the problem of the relation between the two groups: The English Lovers.

As already observed, the Lovers' first section unambiguously belongs with our first group. But as stated on an earlier page, it also has at its end a catalogue and advertisement of Kirkman's which appeared in the identical type-setting, except for minor corrections made in the Lovers, at the end of the Chronicle of Portugal. In both books they form an integral part of the work, not a separate final section as do some other advertisement lists, and all necessary conjugacies can be easily established. The Chronicle is a signed Johnson book, and though, as an unadorned octavo, it provides no typographic links with the second group, which is also certainly Johnson's work, it is yet sufficient to link the Lovers also with that group.

The Lovers, indeed, is a complicated book. Its collation, 8°: A-E8,

2A-M8, is modestly intriguing, but the evidence from headlines and type is very much so, though at least they go together: 1: (A) B-E; 2: 2A-D, 2K-M; 3: 2E-I. Or more explicitly, the book begins with five sheets signed A-E, all in the same pica text fount, and with the same sets of headlines persisting from B through E. There we find the signatures, and pagination, starting all over at A and 1 again, with a different set of headlines and a different pica text fount, while the text begins with Part I, Book II. And though, with four pages of that Book to go, the text is continuous from 2D to 2E, after 2D type and headlines suddenly jump five sheets to pick up at 2K and quietly finish the book, while in between, sheets 2E-I, we find a third pica text fount and a third set of headlines. And as if cutting the book into sections of five, four, five, and three sheets respectively were not enough, we moreover discover that K was already printed off by the time I was being set, since the compositor of I, three lines from the end of I1v, concluded that he had not enough copy left to fill his sheet and decided to make it do so all the same by turning in his pica cases and using english instead.[7] This time he cast off correctly, and the larger type did the job.

Our problem, whether our two groups of books are both from the shop of Thomas Johnson, therefore resolves itself into the question whether The English Lovers was entirely set up in it or not.

Now the division of copy seems rather peculiar for work in more than one shop, and the signatures will not countenance it at all, unless we are to assume that by the time printing started, only about half the book had yet been written, a hypothesis that has little to commend itself, also in view of the continuous printing from 2D to 2K. But if the whole copy was available before the book was allocated to its printer(s), then the division of the copy into four shares does not fit with division among either two or the maximum three shops, because of the sandwiching of one printer's stint between two stints of another.

Moreover, if the copy was originally divided between several shops it would have been cast off first, and we should not expect a perfect link from 2D to 2E and such an imperfect one from 2I to 2K.

But what especially argues against division among shops is the signing. If the copy had been cast off before division among shops,

the signatures would have been marked in it, and not only would the second printer not have started with A again, but the third printer, whether or not he was identical with the first, would never have begun to sign his stint with E.

If, however, the whole book was the work of a single shop, it is not difficult to account for what happened. With an original cast-off only so far forward as to ensure a proper join between the first two sections, the missigning in the second section can be explained as the result of the second compositor's notion that he had the first sheet of text, his colleague at the other press the preliminaries, including the beginning of the text. That this is not far-fetched is demonstrated by this colleague himself, who also missigned the first page of the sheet following the preliminaries A, instead of B.[8] And of course the consequential missigning in 2E becomes automatic.

There are some further links between the two groups, though none of them of a compelling nature. Thus a particular found of handsome titling capitals, easily recognizable because its E's and I's came from faulty matrices and invariably show a more or less clear white hairline about mid-way across the shank, is in evidence in both. The E's mentioned for Loves Mistress and the Milk-maids belong to it, and similar E's are to be found in the 1661 Elder Brother and in the first section of the Lovers, from the other group. And in fact there are some indications that our group differences are indeed differences, not between shops, but between compositors in one shop using distinct typecases in their charge. The evidence for this consists in a headline peculiarity, and if this interpretation is right, it means that the compositor who used the mixed fount first encountered in Beggars Bush was in the habit of setting up his headlines in larger type than did his colleagues. In the few cases where use of another fount with the mixed fount, within one book, is attended with use of different headline sets, we invariably find a set or sets in larger-than-text type, and usually a set or sets in text type. And perhaps we have here an indication that the remaining Scornful Lady belongs with our piracies after all, for its first section only again shows larger-than-text headlines, nineteen different ones, so help us.

One further link remains, and this brings us back to Kirkman again. Four of our plays, two from each group, share a rather uncommon watermark, a peacock in his pride. They are The Maids Tragedy

and A King and no King from the first group, Philaster and Band, Cuff and Ruff from the second. Three piracies, that is, and one dialogue printed for Kirkman. If, as was common, the paper was supplied by the publisher — and in Kirkman's case we know that he did so at the outset of his career and near its close[9] — this would place responsibility for the three piracies squarely on his shoulders, whoever printed them.

In the course of the researches which have been summarized above, it was necessary to carry out thorough bibliographical analyses on most of the works mentioned. Not all the results obtained are of equal value, and some have already been stated; but though they will of necessity make much less of a coherent whole than what has gone before, some of the more significant data obtained may perhaps be detailed here. It will be thought by some that they might better have been utilized for a series of scattered notes, but for ease of reference as well as for their evidential value in the present context it has been thought preferable to record them here, and to arrange them under the titles of the books examined.

LOVES MISTRESS. This edition was obviously set from the genuine 1640 edition, but the manner in which is peculiar — though our knowledge of the ways in which compositors worked together is perhaps too scant to justify the word. The book was set in two different pica founts which alternate as follows: A3-4v, B1-3v , B4-C3v, C4-D3:30, D3:31-F1:15, F1:16-F3v , F4-G3, G3v-4. G4v was set in long primer in order to get the text into this sheet, and the same smaller fount was used to print a song in two columns on F3v-4. In terms of the original edition printed from, the division of copy becomes hardly any clearer: A3-4v, B1-4v:23 thou, B4v:23 finde-D1v:23, D1v:24-E2:20, E2:21-G2:18, G2:19-H1v:15, H1v:16-I2v:2, I2v:3-4v . Neither division corresponds, on the whole, to clear units of the book or the text, though in the latter part of the work scene divisions do seem to play a part. If this book was cast off, and set by two compositors simultaneously, the stints would not be too unequal, but the division of the copy remains curious, especially when it comes in the middle of a line. The headlines also do

not suggest a clear-cut division of work. There are eighteen of them, and they come together in formes in an apparently haphazard manner, but a manner seen in several of the plays under discussion. Though no solution is here offered, the facts seemed worth calling attention to.

THE MAIDS TRAGEDY. This looks like an entirely regular production, set by two compositors, and printed on two presses, each using two skeletons. Two founts are in evidence. The mixed fount prints in sheets E-G and I, another fount, of similar face, prints in A-D and H. Of four sets of headlines, each larger than text, two print once each in sheets E-G and I, the two others in sheets A-D and H. No headline wanders into another skeleton, and there is very little switching about of headlines within the skeleton, during reimposition, although each page has a complete headline.

THE TWO MERRY MILK-MAIDS. For reasons of typography and watermarks this play and Loves Mistress must have been printed at much the same time. Besides the links already noticed, the two are set up in the same two founts and in the same manner, and they use almost exactly the same number of headlines — seventeen in this case. There is the same free wandering of headlines in reimposition, although the Milk-Maids split the headline over the opening. The division of founts is as follows: B1-C1, C1v-D1v , D2-3v, D4-E1v , E2-3v, E4-I1, I1v-3v, I4-K2, K2v-4, K4v , i.e. 46 pages in the one fount and 26 in the other. Both start with 9 and then 4 pages, but no clear pattern emerges.

PHILASTER. Ostensibly a straightforward job, set in a single fount, and with a headline pattern much like that in The Maids Tragedy. But there are some curious headline linkages, such as the spelling PHILSATER in one headline when it prints on B4, F4 and G4, but not when it prints on C1v, D4, E3v, H3v and I1v. One such anomaly might be explained away as a coincidence, but another headline presents an exclusive B4v-F1 link, while two others agree in linking outer C, D, E and I to the exclusion of B, F, G and H.

THE PRESBYTERIAN LASH. Possibly the wildest job of all, this play uses 19 headlines for eight formes, of which three are in larger-than-text type. Three, possibly four founts are in evidence, B4:8-32 being difficult to assign: A3-B2v, B3-B4:7, B4:8-32, B4:33-C3:20, (C3:21-C3v in small type), C4-D1v:19, D1v:20-D2:18, D2:19-D3:13, D3:14-D4. The last three changes in fount occur at scene divisions. There is clear evidence of casting off, A3 and 4v being set very widely, and B2v ending one line short, though B3 starts with a one-line speech.

THE SCORNFUL LADY, Greg 334g. On the basis that Greg 334f is the genuine 1651 edition it is possible to work out the textual relationship between the three editions with that date on the title-page. A complete collation, though showing, as always, some anomalies (none of them at all

disturbing) due to coincidence, shows very clearly that 334g was set from 334f for sheets A-D, from 334h for sheets E-H. Though the number of variants is not large, it is yet too large to print the full collation here, but excerpts should be sufficient to illustrate the trend of the evidence:                        

		334f	334g	334h
A4v	1. 4	posthorse	posthorse	posthorse,
	7	labours	labours	labors
	11	laught at	laught at	laught at at (correct)
B1	6	kisse	kisse	kiss
	15	with you	with you	you with (correct)
1v	1	neere	neer	never
E1	32	Lovelesse	Loveless	Loveless
	35	you, indeed I	you indeed, I	you indeed, I shall.
		shall. Travel,	shall Travil,	Travel, (wrong)
	41	Lady	Lady	Ladie
1v		faith. Will . . . sir, your	faith, will . . . Sir? your	faith. Will . . . sir? your (wrong)

Since these are excerpts from a complete collation, and since evidence from spelling is always to be regarded with caution, at least in isolation, not all variants are equally illuminating, but the two in the earlier excerpt where 334g follows 334f in error against 334h's correction, and the two in the second excerpt where 334h is followed in error, though 334f is correct, are significant. One possibility, of course, must still be eliminated, but it is easy to show that 334g was printed from 334h (in its latter half) and not the other way round. Thus 334h is consistent in its proper names throughout (except for erratic lapses) when it uses a different spelling from 334f, as Loveless for Lovelesse, Abigal for Abigail, but in A-D 334g agrees with f, in E-H with h. And a particularly illuminating variant occurs at F2v l. 9:  

Coach: the

Coach. the

Coach· the

To assume that a turned point would be fortuitous at a point where the other text introduces a variant would be asking rather much of coincidence.

Though the textual situation thus admits of clear exposition, it does not make the typographical situation easier to understand. Perhaps it will be simplest to set out the data:

             

Collation: 4°, A-H4.
A-C watermark: grapes	running-title: The Scornful Lady.
type: 20/79 mm.	measure 102-3 mm.
D-E watermark: name band	running-title: The Scornfull Lady.[*]
type: 20/80.5 mm.	measure 99-100 mm.
F-H watermark: grapes	running-title: The Scornful Lady.[τ]
type: 20/81 mm.[§]	measure F, H 98-9 mm, G 99-100 mm.

The headlines agree with the typographical division but have their own peculiarities. Sheets A-C have a distinct headline for every page, which makes 19 in all, all of them in larger-than-text type. It is perhaps simplest to assume that the headlines were set up with the text, but that does not tell us why. Sheet D occurs with two different sets[10] of headlines, of which the present writer has so far only seen one, repeated in E.

THE SCORNFUL LADY. Greg 334h. This book presents several resemblances to some of those already dealt with, and if it seems possible to work out the printing history here, it should yet be borne in mind that certain basic assumptions are made, assumptions which we have become more or less accustomed to make when dealing with seventeenth-century printing, but which the books so far cited might tend to show are not perhaps so generally valid as we should like to think.

The collation is the same as for the edition just described (4°, A-H4), the watermark is the same throughout, and so are the printer's measure and the running-title. Two type-founts are in evidence as follows: A, B1-4, B4v-D1v, D2, D2v-4v, E, F, G, H: i.e. about two sheets more were printed in one fount than in the other. In the copy seen (Folger) B4 and 3v were misimposed in one another's positions.

The imposition scheme, numbering headlines in the order in which they occur in the book and recording them as they appear in the formes, runs as follows:

             

	(o)		(i)		(o)		(i)		(o)		(i)
A	--	3	.	2	1	D	2	13	.	7	4	F	6	4	.	7	8
	--	--	.	--	--		11	6	.	3	12		13	11	.	3	1
B	7	1	.	9	8	E	5	14	.	8	6	G	10	8	.	9	10
	2	4	.	5	6		9	10	.	10	9		5	9	.	12	14
C	13	10	.	13	3							H	8	5	.	7	12
	12	3	.	10	11								10	9	.	13	14

The presence of 14 headlines argues for more than one press, and the presence of two type founts argues for two compositors, which, in its turn, argues for two presses as the eighth edition of a play (and a piracy at that) would hardly be printed in a small impression needing two compositors to keep up with one press.

Since one of the headlines used in A (2) does not occur after D (o), we may assume that A was printed first, not last. As compositor 1 (making the assumption of 2 compositors from the two founts) set two sheets more than compositor 2 (though A was of course a small job) and was able to help compositor 2 over B by setting B4v (compositor 2 subsequently reciprocating by setting D2 — there is no indication that setting was by formes), we may think that A was finished well before B, so that press-work

began well before B. Assuming that a logical sequence was followed, we should then conclude that the two formes of A were machined concurrently, since the A(o) headline next occurs in a forme machined on press 1 (i.e. the press for which compositor 1 is assumed to have worked) and the two A (i) headlines in a forme machined by press 2.

This would mean that the formes of B were machined concurrently with those of C, those of D with E, those of F with G, and, ideally, the two formes of H again machined together. In fact we find that the headlines on C, both inner and outer, exclude the possibility that its formes were machined concurrently with any others than those of B or A(i). Further, though no such general exclusions exist for the other formes, we find that the formes of D and E are mutually exclusive in their headlines, and those for F and G equally so except that F(i) and G(o) cannot have been at press together, as both use headline 8. But the two formes of H are again mutually exclusive, and H(o) is in fact imposed with the last skeleton worked on press (2) although it had been set by compositor 1.

The printing can then be reconstructed as follows: G(o) and H(o) are from the same skeleton, so G(o) would have immediately preceded H(o). G(o) shares half its skeleton with G(i) and three-fourths with E(i) and E(o), but probability would choose G(i) as the preceding forme. Since G(i) shares three-fourths of the skeleton with E(o), and E(i) again three-fourths with B(i), there is some reason to suppose that the order of printing was B, E, G, H(o), inner forme preceding outer. A(i) would then naturally take its place before B(i), where both its headlines reappear. We might reverse the order in B, which is immaterial.

If this sequence is taken as established, certain consequences follow. C, as we have seen, must have been printed concurrently with A(i) and/or B(i,o), so that we might tentatively place A(o), C(i), C(o) parallel with A(i), B(i), B(o). The order in C is again immaterial, but we may think that D(i) has (12, 3) together because they were together in C(o), though on similar grounds (13, 11) we might advocate C(o), C(i), D(o), D(i). But we also note that F(o) takes over three headlines from D(o), including 6, which also appears in E(i), and that since F(i) has 8, it would have been printed before G(o) and after E(i), while F(i), having (7, 3), would have been printed after D(i), and D(i) — (7, 4) —after B(o). The only way of meeting these conditions is by positing an order D(i), D(o), F(i), F(o), printed concurrently, forme for forme, with E(i), E(o), G(i), G(o). We are then left with H(i), which may reasonably be placed after F(o) on press 1, to give each press eight formes and make them finish together:

           

	A(o)	C(i)	C(o)	D(i)	D(o)	F(i)	F(o)	H(i)
1	--	3	13	3	13	10	7	4	2	13	7	8	6	4	7	12
	--	--	10	11	12	3	3	12	11	6	3	1	13	11	13	14
	2	1	9	8	7	1	8	6	5	14	9	10	10	8	8	5
2	--	--	5	6	2	4	10	9	9	10	12	14	5	9	10	9
	A(i)	B(i)	B(o)	E(i)	E(o)	G(i)	G(o)	H(o)

We may be surprised at the setting up of headline 14 as late as E(o), when headlines 1, 8 were left on the bench (presumably D(i) had not yet been broken up), but then 12 was set up for C(o) when 11 must have been available (or vice versa if we reverse the order). A convincing answer to this riddle, which is anything but unique at this time, were a thing devoutly to be wished — but until we have one for 12 we had better not worry over 14.

Notes