This section is concerned less with the editorial problems posed by the Acts and Monuments than with what happened to the masses of manuscript and print which John Foxe brought, at approximately seven year intervals into the printing house of John Day in Aldersgate.

The first English edition of 1563 was not the occasion of the first encounter between the two men, nor was Day unfamiliar with Foxe's distinctive handwriting. We know that in September 1559 the latter was still in Basle completing the second of his two Latin martyrologies, the Rerum in ecclesia gestarum … Commentarii. Yet within a matter of weeks, in November 1559, Day printed A frendly farewell, which master doctor Ridley did write … a little befor that he suffred.[1] This was a small book, edited by Foxe, and in the preface he announced his plans for a new martyrology, written in English, which would be far larger than the Rerum.

First to begin with this little treatise of Doct. Nicholas Ridley, late Byshoppe of London, this shalbe to desyre thee (gentle Reader) to accept it, and studiouslye to peruse it in the meanetyme whyle the other Volume be addressing which we are about touching the full historie, processe, and examinations, of all our blessed brethren, lately persecuted for righteousness sake … In the meane time because all thynge[s] can not be done at once, and the Volume be long, accept well in worth this little (but pithie) worke of this forsaid Bishoppe … [2]

Day, as we shall see, had remained in England during Mary's reign, while Foxe was in exile, and there is no evidence that they had previously been associated. However, from the autumn of 1559, their careers were inextricably linked. We cannot be sure how the contact was established, but both men had links with Sir William Cecil, the powerful Principal Secretary, and his hand may reasonably be identified.[3]

The kind of material which Foxe produced, how much of it was in the hands of his various transcribers, how much in the form of letters by or about the principal actors, and how much in his own hand, is a matter for editors and other analysts to determine. Here we can only offer some clues as to how that material was mediated by the printer, in the light of the known practices of an Elizabethan printing house. This knowledge has hitherto been only partial, and partly reconstructed from the descriptions of Joseph Moxon a century later,[4] but it has been considerably augmented in the course of the present study. It was generally believed that printers would go to considerable lengths to set up a later edition from a printed copy of the earlier version, amended by hand as necessary. This would have been facilitated by the very limited control that an author exercised over his own work. It may well be that John Day originally supposed that the second edition of the Acts and Monuments could be dealt with in this way. Bibliographical and textual examination reveals that, if he did so believe, he was very mistaken. However, if there was a mistake, it was one in which he was a willing accomplice. He was able, and willing, to involve himself in highly complicated work processes, which carried considerable risk, and that tells us a great deal about John Day.

He clearly had a personal stake in Foxe's book as a history of those who had witnessed with their lives (and in other ways) to their protestant beliefs. It is unusual for a printer to make a personal appearance in the narrative, but Day does so.[5] In recounting the martyrdom of John Rogers, who was burned early in 1555, Foxe says

Amonges others being then in prison, thus he spake to the Printer of this booke, who then also was laid up for the lyke cause of religion: thou, sayde he, shalt lyve to see the alteration of this religion, and the Gospel freely to be preached agayne …

More about this cause of religion anon. An earlier statement about Rogers, that His examinations he penned himself, may derive from either Foxe or Day, and perhaps gives a clue to the source of some of the copy used by the printer seven years later.

Recent research by Oastler, Evenden and Freeman [6] has at last addressed Katharine F. Pantzer's observation in the Revised Short Title Catalogue, that the attribution of Day's publications before 1559, and his relationship to other printers and publishers requires further study. He was born in 1522, probably in the now-vanished town of Dunwich on the Suffolk coast. It is unlikely that he was an alien, because there is no record of either naturalisation or denization, as there is for so many workers in the book trade. The close relations which he appears to have enjoyed with alien protestants were partly the result of the exigencies of the trade, partly of shared religious convictions, and partly of an origin on the east coast.[7] His first confirmed printing can be dated to 1547, and he had probably transferred from the Stringers or Bowyers Company to the Stationers in 1546. In 1548 he entered into partnership with another printer, William Seres, to print English bibles and protestant polemic. This partnership lasted only about two years, but was highly productive. In 1549 they published a lavish new edition of the bible, edited by Edmond Becke, a work which involved no small expenses and charges, and which was considered to be unduly expensive.[8] Several other smaller works were also thought to be over priced, and it may have been this which put an end to the partnership. They had also narrowly avoided arrest for the production of Luke Shepherd's John Bon and Mast Person, which was a little ahead of its time, and deeply offended the conservative Aldermen of London. From 1550 onward Day turned his attention to printing under his solo imprint, and continued to focus his attention upon the works of the reformers, and upon various pieces of biblical text and exegesis. In 1551 he reissued Becke's edition of the bible, this time without Seres's collaboration. By then he was establishing a reputation as a printer of de luxe bibles, and as a key publisher of official protestant theology. As such he was extensively used by both courtiers and senior clerics. He was still under thirty when this flourishing career was brought to a sudden halt by the accession of Mary in 1553.

He could find neither work nor favour under the new regime, and it has been generally assumed that he fled to the continent and attempted to continue his trade there. However, it is now clear that he remained in England, printing surreptitiously in Lincolnshire in premises leased from Sir William Cecil, and under the latter's covert protection. It was there that he issued those works which bear the false imprint Michael Wood at Rouen.[9] After a while he was caught and sent to the Tower for pryntyng of noythy bokes,[10] without, apparently, incriminating his patron. There he met John Rogers. That he avoided Rogers's fate was probably due to the nature of his specific offence. He had not imported protestant books (which particularly exercised the council), and he had not offended the Stationers' Company because he had been operating outside London. If he made any kind of submission, Foxe was too tactful to record it, but as he was shortly after released it can be presumed that he did so. London was short of printers, because so many of the alien protestants who had worked there under Edward had fled, and Day's compliance was soon rewarded. Along with William Seres (also recently released from prison) he began to work as an assign to the Catholic printer John Wayland, and printed for him a number of catholic primers.[11] He even issued a few things under his own imprint, and became official Printer to the City of London, but he ventured on nothing controversial as long as Mary was alive.

However, after Elizabeth's accession such nicodemism was no longer necessary. The legislation and injunctions of 1559 required a new Prayer Book, and the setting up in every parish church of an English bible of the largest volume. This was intended to be the Great Bible of 1539, reprinted in 1561 and 1562, although imported copies of John Bodley's Geneva Bible may sometimes have been substituted. John Day, in spite of his experience was not given the contract for these bibles. That went to the Queen's Printers John Cawood and Richard Jugge (the latter of whom had been in exile for his faith).[12] However, there was a large expansion of new protestant and vernacular printing, in addition to the bibles and part of Day's share was a patent to produce the Metrical Psalms, which he was granted in 1559.[13] He also received privileges for the printing of the ABC and the Little Catechism, the steady and regular demand for which formed the basis of Day's wealth, and provided capital for more risky ventures such as the Acts and Monuments. When the latter was in press, and Day's other output was noticeably restricted, he still found time for two printings of the Metrical Psalms. In 1586, when Richard Day's inheritance of his father's privileges was in question, one witness estimated that the income from these safe sources was £200 a year; another said £500.[14] John's first privilege was for seven years; this was renewed in more specific terms on 6 May 1567 for ten years; and then to John and Richard jointly in reversion for life on 26 August 1577.[15] It is probable that by 1559 William Cecil's patronage had been renewed (if it had ever been withdrawn), and that these favours may be attributed to his influence. Significantly the three figures who appear in the fine woodcut initial which inaugurates the 1563 Acts and Monuments are Foxe, Day and Cecil.[16]

Caution must be observed when examining the bibliographical evidence for any of the first four editions of this work. As Katharine Pantzer stated of the 1563 text Most copies of this and the following editions are imperfect. Such a claim was probably too modest, because so far no single perfect copy has been identified. Imperfection and dilapidation are familiar to all students of the Acts and Monuments, and it is not difficult to see why this should be so. It is a very big book The 1570 edition in the Bodleian Library, preserved in a late seventeenth or early eighteenth century binding, has been broken down from two into three volumes.[17] The same library's 1583 edition, which it has owned since at least 1605, has had to be rebound. Originally it was no doubt in locally made calf skin with a chain staple, but it is now in a rather unappealing nineteenth century binding, with the gilt edges which the library then thought proper for its more prestigious books.[18] That binding will also need replacing in the near future.

The reasons for the widespread imperfections are not far to seek. This was a famous book, chained by order of the Privy Council and Convocation in cathedral churches and deaneries, and by archiepiscopal instruction in parish churches.[19] There not only might all read who could, but anyone could turn the pages, so that the striking woodcut illustrations became part of the national consciousness. Sometimes these illustrations, particularly the detachable ones, were removed to end up in collectors' frames. It is also a very heavy book. Copies were dropped, breaking the binding, and leaves were lost, particularly at the beginning and the end. When the collecting of Elizabethan books became fashionable, restorers could easily be found to improve the condition of this principal monument of English history. At best this involved washing and pressing, and at worst the making up of imperfections from other copies. Even washing can be fatal to the slip cancels which are not uncommon, at least in the first two editions of the Acts and Monuments. It was an American bibliographer, the late Fredson Bowers, who introduced the notion of ideal copy. This definition is best illustrated from Philip Gaskell's New Introduction to Bibliography (1972)

… the bibliographer examines as many copies as possible in order to construct a notional ideal copy of the edition he is studying. A description of this ideal copy would note all the blank leaves intended to be part of its gatherings, and all excisions, insertions and cancellantia which belonged to the most perfect copy of the work as originally completed by its printer, and first put on sale by its publisher.[20]

This notion was first articulated as part of the editorial process for the Elizabethan drama. Nevertheless, if we are contemplating an edition of Foxe, we have to bear this doctrine in mind, even if it is in practice impossible to live by. For we are not dealing, as we are in a Shakespeare folio for example, with the unknowable intentions of a man seven years dead. Whether Foxe ever lived at Day's house in Aldersgate, or merely visited him there at weekly intervals (as his son later testified) he left abundant evidence of authorial intervention,[21] and Day, following his own beliefs, tried to present his readers with the ideal copy that his author desired.

Thus, from a bibliographical point of view, the editor of Foxe should seek a control copy which is as near as possible to what Day, with Foxe working closely alongside him, put on sale in the second half of 1563, or in 1570, 1576 or 1583. Ideally, such a copy would not have been tampered with by binders or restorers, and would have been carefully stored in dry, but not too dry, conditions. Particularly at the points where bibliographical inspection reveals something odd in the printing process, the editor will need to compare other copies. For the first two editions scholars working in either Oxford or Cambridge are fortunate. In Cambridge the university's strong protestant traditions preserved its university library throughout the sixteenth century. Andrew Perne, who makes an appearance in the Acts and Monuments, owned two editions of it, which survive.[22] However the control copies, nearly perfect, must surely be those which Foxe himself gave to his friend Lawrence Humfrey, President of his old college, Magdalen. The 1563 edition there has a letter from Foxe, commending his work to the college library, even though it is in English. Since he was persuaded to give the book by the Oxford bookseller Garbrand Harkes, it probably came from the latter's stock, although it is no longer in a contemporary binding.[23] Even in this case, a few leaves at the end are missing. A line, probably in Foxe's hand, and other minor corrections, suggest that the gift was carefully checked.[24] Magdalen also has a presentation copy of the 1570 edition, which retains a tough binding of split calf which may well be original. This looks like a good copy, which is just as well since it is one of the most bibliographically complex works produced in the entire sixteenth century. The Bodleian copies lack the prestigious provenance of those at Magdalen. The library has two copies of 1563, one in what may be a near contemporary binding dating from its second owner, Michael Serolen, who is known to have bought it in 1594 for 10 gulden and 5 stivers.[25] Both came to the library via the Douce bequest in 1834. The 1570 edition was also acquired in the nineteenth century, but nothing is known of it before then.

Foxe was directly responsible for the complexity of the second edition. When a sixteenth century printer undertook a book, he would work for the first edition from manuscript copy. For subsequent editions he would use, as far as possible, printed copy, even if it was heavily interspersed with manuscript corrections and additions. He would probably begin by estimating roughly the size of the book, and deciding on the format. Where a really large work was contemplated, the format would be folio, the sheets carrying two pages of print on either side, which, when bound would be folded and sewn down the middle. All the editions of Foxe, whatever the size of the paper, are basically folios in sixes, as will appear from the collation formulae. Each quire or gathering has six sheets, twelve pages, and in the case of the Acts and Monuments, two columns of print on each page. That is normal for folios, and is how the best known, that of Shakespeare's Works was printed in 1623. The quantity of paper required would have depended on the number of copies to be printed, and we have no figures for the print runs of the Acts and Monuments until after Foxe's death. Whatever his estimate, the printer would have to buy his paper in advance. In Day's lifetime no paper was made in England, but the printers do not seem to have imported their own, perhaps because of the high capital costs. They bought retail from merchants who imported in large quantities - 160 or 200 reams at a time. A surviving Port Book of 1567/8 provides a glimpse of this trade, with these quantities arriving on every ship from Rouen, usually as part of a general cargo.[26] None of the merchants were themselves Stationers. Some of this paper was made in France, and some in the Low Countries.

The first stage of preparation was the casting off of copy; that is, marking off the manuscript leaf by leaf to enable the compositors to set the type. This is fairly easy with printed or homogeneous manuscript copy, but with Foxe we know that it came from many sources and in many forms, including transcripts in many hands, as well as some printed matter. Day himself complicated the casting off by using a wide variety of type faces (black letter, roman and italic in various sizes) and Foxe added confusion both to the casting off and to the subsequent printing by interpolating new text, and removing existing items. Cast off copy is a rare survival in the sixteenth century,[27] and we are fortunate in having some contemporary evidence from Day's printing house in the form of cast offs from Certain most godly fruitful, and comfortable letters of such true saintes and holy martyrs (1564), a compilation now attributed to Henry Bull, although bearing the name of Miles Coverdale.[28] The manuscripts containing this copy also contain some fragments from the Acts and Monuments.[29] These now survive in both the British Library and in Emmanuel College, Cambridge. Once casting off was complete, the compositors could begin to set the text. Normally they would begin with the text proper, and not set the title pages and preliminaries, such as the preface and dedication, until the end. If they had any concern about the supply of type, printing could begin once they had set pages six and seven of a folio in sixes. There is evidence that this was Day's practice too, but he used his variety of types lavishly, and could therefore afford to keep pages of type standing. Nevertheless, both setting and distribution of the used letters into a number of type cases would make for very slow printing. Typically the compositors signed each six leaf gathering with a sequence of the twenty-two letters of the Latin alphabet, signing the first four leaves only. When they reached the letter 'Y', they would begin another sequence with double letters. It is when the regularity of these signatures and sequences is in any way interrupted, that the bibliographically minded reader may suspect that something has happened. In this book the usual explanation is that Foxe has introduced new material, or excised something. From an editorial point of view that does not matter too much, provided that all the copies are the same, and collectively represent ideal copy. General bibliographical experience, however, suggests that this may well not be the case. Where a leaf, for example, has been cancelled an another substituted, someone, probably the binder, nearly always omits to correct all copies before they are bound and issued. Foxe and his printer have laid another trap for editors by their use of slip cancels. These are printed slips pasted over the text to be corrected, and unless one holds every page up to the light, they are very hard to detect. There are at least four of these in the 1563 edition, of which one is missing from the Magdalen copy, and another from the first Bodleian copy. There are at least six in the 1570 edition, but no copy, not even Foxe's copy at Magdalen, has a full complement. The labour involved in printing these little slips, cutting them up and pasting them into the correct places in hundreds of copies, can only be imagined.

The 1563 edition was not simply printed from a wealth of manuscript sources. To supplement this original material, Foxe also called upon works which Day had already printed. For example, he began to set out a detailed account of the persecution of some proto-protestant martyrs in the South of France, which he scrupulously translated from a French martyrology. However, two thirds of the way through Day, obviously becoming impatient with the slow progress, abandoned this version, and completed the story using a much briefer narrative taken from the English version of Johannes Sleidan's Commentaries, which he had printed in the previous year.[30] Foxe must have been persuaded of the urgency of the timetable to consent to this change, because in the 1570 edition he abandoned Sleidan and completed the story from his original source. This, of course, prevented Day from using the marked up sheets of the first edition when preparing the second.

Bull's editing and marking up of the text of Certain … was meticulous almost to the point of being obsessive.[31] He often provides synonyms or changes the word order to make a line read better, even when it does not change the sense of the passage. He used many symbols to indicate his instructions to the compositor over the layout, or even the content of the text. Some of the symbols on the surviving letters at Emmanuel may have been made by Day rather than Bull, because one sample matches surviving examples of the printers handwriting.[32] The following hand-drawn symbols are a few of those which occur regularly throughout the cast off copy.

	Usually indicates a new paragraph requiring an indentation in the text
	Marks an important or problem passage that requires further examination
	The circle through the ivy leaf indicates that the passage has been dealt with
	Used to mark the first and last sheets of an individual portion of copy text
	Enclosed leaves will be marked without the dots
	Used to enclose the first word of a new page; indicates what the catchword is
	Signature marks, indicating what pages of the gathering the text will occur on
	Used to indicate that a word or text is missing

Figure 1. Printers' marks used by Day and Bull on the original Letters of the Martyrs

Many of these marks were typically used when paper was in short supply and as much text as possible was required on each page. This cast off copy therefore tells us something of the order in which the compositor, following the instructions on the marked up copy, placed his type. The headings of a page appear to have been set up last. This was indicated to the compositor numerically. The line in which the first main paragraph of the body of the text should occur was indicated by writing the line number on the original beneath the opening salutation or heading - in this case to letters. For example, if the body of the letter needed to begin 26 lines down the page, to enable the text to be fitted on that page, the number '26' would be placed under the opening salutation and/or prayer on the copy text. This allowed the compositor to set the most important section of the page, the content of the letter, before choosing how big a type he could chose for the title. The criterion might be either aesthetic or functional - if space was at a premium. In both Foxe and Bull, when the end of one letter is followed directly by another on the same page, the position of the last line of the main text of the first letter was indicated by the appropriate line number on the manuscript copy.[33] Often when space was scarce the titles and farewells were set in much smaller type than that used in the earlier stages of the same work.[34]

From the annotations on the cast off copy, it is clear that a major concern was always for a clear, legible copy text. Bull often rewrote letters in order to provide copy that was easier to read than the originals, which were often written in circumstances not conducive to legibility. Foxe and Day did the same. Slips of paper were often added to the manuscript where the printers instructions had become messy or unreadable, and these small indicators, pasted down the left hand side, were attached to the copy text in order to clarify exactly what Foxe wanted to be printed.

The collation formula for the 1563 edition of the Acts and Monuments is as follows:

(In some copies 4R¹ is signed 4Q¹).

From this it is clear that there are a number of irregularities. Apart from the preliminaries, the signatures run regularly in sixes until G, which has eight leaves, as have H and I; but I is then followed with a starred I of eight leaves, and that in turn by double and treble starred signatures, each of three leaves. These extra signatures indicate interpolation, and the subject matter is the abuse of papal power, particularly under Kings John and Henry III. There is a rapid transition in the next gathering to Wycliffe. Regular signatures then resume until Vv, which has eight leaves; the subject matter is Bishop Stephen Gardiner and his dispute with Protector Somerset. Again it is clear that Foxe received some last-minute documents which had to be squeezed in. Regularity then resumes, until there is a starred 2Kk, concerned with Lady Mary's refusal to conform during the reign of her protestant half-brother. Francis Douce noted that his second copy (now in the Bodleian) lacked this gathering. Perhaps it is a late insertion. 2Mm1 is probably a cancel, and there are slip cancels on B1, 2A2v, 3G5v and 3M3. There may well be others. Most interpolated signatures have a manuscript catch-word supplied.

There is another rather arresting bibliographical anomaly, which does not show up in the general collation formula. The last leaf of 3Rr is printed in unusually wide columns - 86 mm wide as opposed to the normal 72-74 mm. The text is Cranmer's withdrawal of his recantation, and the columns were widened to include new eyewitness accounts of his martyrdom, which Foxe only acquired after this page had been set up.[35] By increasing the column width, this new material was fitted onto a single page, and the catchword matched. The fact that the material was thus included, and not simply carried on to the next quire, suggests a stop press correction, and that the next quire was already composed and being printed on another press. A copy may exist somewhere with the original unexpanded text. Although less obvious to the naked eye, Day used a similar technique on sig.Ssir, where he increased the column width to 77 mm, and changed from the usual 42 mm black letter type to a 20 mm on the recto, followed by a 35 mm type on the verso in order to fit in A certaine Libell or boke intituled the Supplycation of beggars throwen and scattered at the procession in Westmister upo[n] Candlemas day, before king Henry the viii, and the tale of Fish's wife. The final paragraph of sig.Ss2v returns to the normal 42 mm type and loses its catchword, after which the gathering continues as normal. He had used a similar technique of column widening and type reduction earlier, when he lengthened the columns and reduced the font of Heinrich Bullinger's A Hundred Sermons[36] in order to fit in a 26 stanza poem. Day was not averse to using his ingenuity to protect his paper supply!

The Revised Short Title Catalogue records two proof sheets of the 1563 edition, in copies in Cambridge University Library and Lincoln College, Oxford. These, probably a bifolium, are not in fact proof sheets (printed on one side only), but two pages wrongly imposed. The woodcut of the burning of Wycliffe's bones is printed as page 105 on the left hand side of the sheet, and the text of page 100 on the right. This error (if error it was, because the running titles are the correct ones for recto and verso pages) was spotted after a number of sheets had been printed off, but before the sheet was perfected from the outer forme. The faulty sheets were evidently wasted and used as endpapers in other books. Even an interruption to printing of this kind enabled alterations to the text to be made. No true proofs have so far been found. How could such an error occur in a workshop like Day's? It may be that the compositors were confused by having to incorporate (for the first time) a two column wide woodcut into the body of the text; or they may simply have been experimenting with the lay out. We have already assumed that the printing process was more or less consecutive. The earlier woodcuts, other than those of single column width, were either pasted onto blank spaces left in the text, or tipped into the binding. That was done with the large cut of the murder of King John by the monk of Swinstead, and indeed the cut of the burning of Ridley and Latimer was too large for all the editions, being either tipped into the binding or placed sideways.

Although the printing of the 1563 edition was not entirely straightforward, it was simplicity itself by comparison with that of the second edition of 1570. This was evidently planned almost immediately after the first, as is apparent from a passage in the Letters of the Martryrs of 1564.[37] Day was encouraged to agree to this by the financial success of the work, while Foxe wished to remedy its flaws and lacunae. In November 1563 William Turner, an eminent evangelical, wrote to Foxe stating that everyone he had met had praised the Acts and Monuments, but that some of the poorer sort had complained about the price. The Scots reformer, John Knox, writing about 1566, echoed the same sentiment. The book for the great price therof is rare to be had.[38] Despite these reservations, and misgivings over the length, it was from the outset intended that the second edition should have a longer chronological span that the first. Ultimately it ranged from the time of the Apostles to the accession of Elizabeth. Foxe was not unaware of the strains which such a project would place upon Day's resources. In a letter to Cecil, probably in 1567,[39] he asked that, in view of the volume of material about to go to press, the law limiting the number of foreign workmen that a printer might employ should be relaxed; since we keep three presses continuously supplied with material in abundance … (quum tribus praeliis material continue suppeditamus). The immediacy of the problem, and perhaps Day's presence at the writer's side, is emphasised in a postscript to the letter, in which Foxe asks further that the printer should continue to enjoy his monopoly of printing the metrical psalms, because from this alone his whole family is maintained … (ex hoc uno solo universa illius alitur familia). Foreign workmen were a great help, both as compositors and pressmen, in printing the psalms and the other books covered by his monopoly, but English compositors would surely have been of more use for the extended Acts and Monuments.

It is, by any standards, a huge book. The continuous pagination - not, it must be said, a terribly reliable indicator - runs to about 2,300. The leaves themselves are far larger than in 1563, and the presentation copy at Magdalen College, which has probably been bound only once, measures 371 mm by 240 mm, as opposed to the 330 mm by 220 mm of the first edition in the same library. A brief analysis of the collation of the 1570 text should suffice as an example of how the signature marks here can also indicate the extraction or interpolation of new material after a page had already been set up. The formula which follows, and indeed all bibliographical examination of the 1570 edition, owes a great deal to the research of Paul Dunkin, carried out on copies at the Folger Library which were being disbound for repair[40]. The underlinings indicate reissued gatherings.

Volume I

Volume II

Figure 2. Collation of the 1570 edition of the Acts and Monuments

In most, perhaps all copies, Mm1 and 4Zz2 are cancels; 4Dd1 is a cancel in some. Examination of several other surviving copies confirms the analysis which Dunkin based on the Folger copies. He noted that many gatherings were on small paper the nature of which is discussed below. These small paper gatherings in volume 1 comprise most of the preliminaries (which would normally have been printed after the rest of the book), and the two leaves which replace sig.s1, sigs.D and E, and sigs I and *I*, together with signatures 2M2-20, upon which the illustrations were printed, with their introductory material. They also comprise in volume 2 a preliminary leaf, the last bifolium to be printed of gathering 4Ii, and all the subsequent gatherings up to the end of the book. Slip cancels have been noted on sig.L5v (p. 130), 3V5r (p. 1423), 3X2r (p. 1429) and 3Vv4r (p. 1718). There is heavy manuscript correction on pp.179 and 493, and lighter changes in seven other places (see Fig.3), although there seem to be proportionately fewer than in the 1563 edition. For some reason, Foxe and Day seem to have preferred manuscript correction to slip cancels in 1570, although as these were made in the print room before copies were released for sale, the labour involved must have been monumental. There are at least ten instances of hand correction, that occur in all the copies examined.[41]

Figure 3: Position of manuscript corrections and deletions within gatherings.

There is neither time nor need to go into the details of all these correction here, but a selection of them may be examined briefly. Pages 346, 348, 483 and 494 contain a mixture of deletions and corrections. Pages 346 and 348 are both in gathering F - H6, their conjugate pages being 339 and 337 respectively. No attempt seems to have been made to correct either the line, the page or the sheet, by stop press, slip cancel or single page imposition. Both pages 346 and 348 have years in the margin which ought to refer to the body of the text, but no copy survives in which they have been corrected. On page 339, however, which occurs in the same gathering (F - H6), a year in the margin has been deleted. The hand corrections to this gathering occur on the last three leaves, specifically G2 recto, G5 verso (no signature) and G6 verso. The copies in Dr. Williams' Library (London) the Folger Shakespeare Library and Cambridge University Library all show page 339 undergoing a stop-press. The marginal years deleted by hand elsewhere have been removed from these copies altogether. However, the two pages without signatures still contain the manuscript deletions of text, so the gatherings in these copies show a mixture of corrected and uncorrected sheets. Perhaps Foxe wanted to make the corrections, but we know that Day's paper supply was stretched at that point, and it may be for that reason that some pre-correction pages were issued nonetheless.

The first hand corrections in volume 2 occur on page 930. As noted in Fig.3, the doubled pronoun 'they' has been deleted. More than half of the copies examined have this correction, and in most cases the line has been reset to remove the grammatical error. The copies in Cambridge University Library, and in the Wren Library still have manuscript corrections, and thus contain pre stop-press pages. The manuscript correction on page 1198 is to an incorrect date. The mistake occurs on the final verso page (without signature) in the 3A gathering, and refers to the marriage of King Henry VIII and Anne Boleyn. Unfortunately it gives 1534 instead of 1533, and since Elizabeth was born in September 1533, this was somewhat crucial. All the surviving copies (apart from the Holdsworth copy)[42] have this correction by hand, both in the margin and in the body of the text. In view of the importance of the mistake, it is surprising that this page was not reset, but that does not appear to have been done.

Seventy percent of the hand corrections and deletions listed in Fig. 2. occur in Volume 1. By that time they were already 1200 pages into the work, and Day's stock of paper was running down more quickly than expected. They had not even reached the Marian martyrs, and the printer obviously felt it was necessary to cut down on reprinted pages. Foxe was also becoming anxious, as can be seen from the increasing references which he makes to the need for brevity. For example, on page 1327 he says And after a fewe other wordes, which for brevities sake we omitte, as not being greatly pertinent; and on page 1537 here I might, and as before I have done, I must entreate the reader to co[n]sider first the greatnes of this volume which would not beare [the] tedious tractation thereof. His tone becomes in this respect increasingly desperate. On page 1912 he apologises for the omission of many letters, because of the haste which I am now compelled to make, and by page 2259 he is calling on God's mercy to allow grace and space to include more information about the Spanish Inquisition.[43]

In one place there is an odd mixture of manuscript and print correction. On page 1075, in the story of the persecution in Merindol and Cabriers, the phrase in the valley of Angroine is deleted with a single pen stroke. This could easily be mistaken for a private amendment, except that it occurs in all the surviving copies of the 1570 edition. The correction was necessitated because the villages are in fact in Provence, a fact of which Foxe had become aware by the time that page 1086 was set. On that page the printed text includes the phrase In [the] title of this story of Merindol above prefixed, instead of the valley of Angrone which thou seest raised out with penne, in the country of Provence … . In spite of the speed with which the error was realised, Day was obviously too short of paper to be able to reset page 1075. However, the correction remains as a tribute to Foxe's meticulous care. An experienced printer like Day ought not to have miscalculated his paper need so seriously, and it seems likely that he was accommodating a highly flexible authorial approach, which evolved strategically as well as tactically. It says much for his commitment to the project that he continued to be (more or less) compliant. The eventual paper short fall has been calculated at 10 per cent.

The irregularity in printing this edition was first discussed in 1946 by L. M. Oliver, who had noticed that in many of the gatherings, particularly towards the end of the book, the chair lines of the paper ran horizontally rather than vertically, as might be expected in a book printed in folio. He concluded that these leaves had been printed as singles, and pasted together afterwards.[44]. Dunkin came to the slightly different conclusion that Day had run out of the larger paper when printing was already well advanced, and had been forced to paste smaller sheets together, long edge to long edge, and to print with these. That is almost certainly correct. It is difficult for any orthodox system of collation otherwise to account for Dunkin's findings. In the main body of the text, the change to the smaller paper begins with the final sheet of gathering 4Ii, the last to be printed, and is maintained until the end of the book. Dunkin discovered that the preliminaries to volume one, and some other gatherings in that volume, are also printed on small paper. He describes the latter as cancels; replacement pages for corrected material originally printed on large sheets. It is not surprising that Day was anxious to keep this necessity to a minimum, given that the use of inferior paper increased the risks of tearing, bleed through and other imperfections. The trouble, as we have seen, was that Foxe's knowledge was constantly evolving. The problems of Richard I and John with the Papacy were naturally a fertile field of study, and the interpolated gatherings I and **I were clearly introduced to reflect his latest study of the chronicles of those reigns, such as Gervase of Tilbury and Matthew Paris.

As we have already seen, the woodcuts were added, and either pasted into blank spaces or tipped into the binding. This must have been deliberate, as Foxe would have realised that these illustrations, upon which much effort and money was expended, would give the work a lot of its power and popular appeal. Most of them were specially cut, although a few were reused. That of the burning of Anne Askew on sig.PP3 verso (page 673) had certainly appeared before and was already wormed and damaged by 1563. The additions to the 1570 edition on small paper at the end of volume 1 are unpaged, and are remarkable for the grand series of twelve half page woodcuts, displaying The proud primacie of Popes paynted out in Tables. One of these, the cut of Henry IV at Canossa, had appeared in 1563 as a folded illustration pasted into an inadequate gap in the text. The location of these pictures may again have been an afterthought, as the leaf preceding them had been cancelled and the original text removed in order to lead straight into this graphic display of papal arrogance. It appears that Foxe and Day were doing what they had done before: correcting, or more probably enlarging the work with the results of revised thinking.

There is also some cast off copy from this edition, although not as much as from The Letters of the Martyrs. It survives in BL MS Add 19400, and its editorial significance has already been described by Evenden and Freeman in their contribution to Lives in Print.[45] Their analysis confirms (among other things) that Foxe was as concerned for the physical layout of his work as he was with its content and illustration. Both Foxe and Day knew that catholic critics would be lying in wait to pounce upon any inaccuracy or ambiguity. The former was able to call upon the students of his old college for learned proof reading, and the presentation copy may be as much an expression of gratitude as of his friendship with Humfrey. Day's main proof reader at this time was a Cambridge graduate - William Gace. In spite of (or perhaps because of) this intense level of concern, the 1570 edition, like its predecessor, carries abundant evidence of last minute correction and addition, particularly in the form of manuscript notes and slip cancels.

The call upon the printer's resources, both of labour and money, were immense, and it is likely that his difficulties were sympathetically understood in high places. In November 1570 the Privy Council addressed a letter to the Archbishops of Canterbury and York, and to the Bishop of London, exhorting them

… for the encouragement of the poer man hymselfe … to use which your good and favourable meanes as you best may, eyther by taking order that the same booke be had in all churches … or otherwise.[46]

This was well meant, but was probably ineffectual except among the most committed congregations. More helpful would have been the decree of Convocation in 1571 that the Acts and Monuments should be placed in every cathedral church, and in the residences of bishops, deans and archdeacons.

There is little documentary evidence for the planning and preparation of the 1576 edition, and the collation formula gives no suggestion of the interruptions to printing which had characterised both its predecessors.

The book is a folio in sixes (as the others were intended to be), and the only anomalies are in the preliminaries, at the point where a breakdown into volumes would have been possible, and at the very end of the book. There is, however, a slight novelty in compositorial practice. In many early gatherings leaf signatures appear on the first five leaves, instead of the first four, as is normal for a folio in sixes. This may indicate the presence of a new compositor, and it has been suggested that that may have been the printer's eldest son, Richard. Richard Day was born in 1552, and was educated at Eton and at King's College, Cambridge, of which he became a Fellow in 1574. In 1576 he resigned his Fellowship to return to his father's business, and became in effect a partner thereafter. Relations later became seriously strained, but he was named in the renewal of John Day's patent in 1577, and entered the livery of the Stationers Company on 30 June 1578. There is evidence of his involvement in the 1576 edition, but it is unlikely that he would have had the necessary experience to act as a compositor. It is more likely that his role was that of corrector of his print. He claimed credit for the indices, of which there are two - one general and the other of martyrs - and the Latin poems by university figures prefixed to the work have a strong Eton and Kings flavour to them. A brief poem by Richard himself was (perhaps pointedly) omitted from the 1583 edition.

1576 is a compact edition. The Bodleian copy, a single manageable volume in a contemporary binding, measures 321 mm by 207 mm.[47] This was achieved by the use of a smaller black letter type, of which 20 lines measure about 60 mm (as opposed to the 70 mm or even 80 mm of the earlier editions), and by the use of a poorer quality paper. Both of these innovations, which may be result of Richard Day's sense of economy, attracted criticism from Simon Parrett, a Magdalen friend of Foxe's. In 1582 he urged that the next edition (which was clearly expected) should be printed in good paper, and a fair and legible print, and not in blacke blurred, and torne paper, as the last edition.[48] Humfrey, in a postscript, supported this plea. Two woodcuts were added, both of which had been used by John Day in other earlier books.[49]

The renewal of the Days' patent secured the financial future, and a further edition was in prospect from at least 1579. New data was certainly reaching Foxe by that date, and a decision was taken to reinstate some material from 1563 which had been dropped from the second and third editions.[50] However, time was catching up with John Day. In poor health and at loggerheads with his son Richard, at some point before he embarked upon this fourth edition, he conveyed his lands, goods and chattels to one John Hunt, who was the brother of his second wife, Alice. Not only does this eccentric move reveal the depth of the rift within the Day family, it also left the printer seriously short of ready cash. He had to borrow £200 from Hunt in order to meet the paper bill (of about £500-£600) before the printing could commence.[51] However, when the book finally emerged, it must have been immensely satisfying to all those involved.[52] There seem to have been two or even three slightly different formats. The surviving copy in the Bodleian Library measures 415 mm by 270 mm,[53] while that in the Codrington Library at All Souls College is 381 mm by 240 mm, and that at Brasenose College 374 mm by 240 mm. Probably the Bodleian copy was a de luxe version, while that at All Souls represents the normal trade size. The Brasenose copy may have been cropped in rebinding. As with 1576, the collation is straightforward, and there appear to have been no disruptions or dislocations in the printing process. The collations run as follows:

Volume I

Volume II

No slip cancels have been noticed, and there appear to have been no manuscript revisions. The only slight weakness is that paper once again seems to have run out before the end, and although this did not involve resort to pasting small sheets, it did mean that the final folios were printed on slightly inferior paper.

John Day had become increasingly unwell as the work had proceeded, and Alice had tried in vain to persuade him to rest. John Hunt had at one point even urged haste upon the workmen in case he should die before it was finished. That did not happen, and it is not even certain that the Acts and Monuments was John's last work, because he survived until July 1584. It was, however, a fitting climax to his career, because both Foxe and Day were driven men, obsessed with the need to rescue England from the evils of the Roman Church. Both had spent a large part of their lives, large sums of money and an incalculable amount of energy to that end; and by the time they died it had largely been accomplished.[54] Foxe lived until 1587, and left other work unfinished, but there is no suggestion that he intended to return to the Acts and Monuments after 1583.

Our concern here ends with his death, but it is worth noticing that the later history of the book underlines those points which have been made. It was popular and sold well in spite of its bulk and relative expensiveness. It required substantial investment, but each edition was ultimately profitable. Richard Day sold his privilege in its printing, along with other privileges, to the Company of Stationers, and it was eventually vested in a company within a company - the English Stock of the Stationers Company. By 1595 when a further edition was required, there were ten partners in this Stock to share the initial outlay. The intended print run was at least 1200, and may have been as high as 1350.[55] By the time that the sixth edition appeared in 1610 it was a confirmed best seller, and was registered as the property of the English Stock in 1620.