The only early modern edition of was first published and printed by the London printer Thomas Creede in 1594 with the full title: . Creede had only opened his first printing house the previous year near London Bridge, this after working as a journeyman printer in London for well over a decade (Gants). While a majority of printed titles in the period were also entered into the Stationers’ Register, the acquisition of by Creede is not recorded there. Creede did, however, enter six other professional-play titles in the Register in 1594, and three of these entered in May and June (, , and ) were also likely Queen’s Men plays. His quarto consists of ten gatherings (collating A-K3) with the C3 signature mistakenly signed A3. Its title page includes two blurbs, an advertisement describing how the play was playd by the Queenes Maiesties Players, Creede’s woodcut device, and an imprint listing the address of Creede’s stock warehouse (A1r). No author, however, is ascribed. Creede’s first blurb outlines how Selimus most vnnaturally raised warres against his owne father Bajazet, and preuailing therein, in the end caused him to be poysoned, while the second describes the muthering of his two brethren, Corcut, and Acomat (A2r). The quarto is printed with both Roman and italic type, the former used for the text, stage directions, and running headers while the latter is used for the prologue, speech prefixes, and proper names.

Creede’s trove of Queen’s Men copyrights (which would include , published without entry in 1599) has led some commentators to suppose that Creede directly and opportunistically acquired his theatrical copies from what was a waning, desperate company. Gerald Pinciss, for example, has argued that Creede happily reached an agreement with the failing Queen’s Men to buy plays in block (322) after the company suffered serious financial setbacks due to the closing of London’s professional theatres in June 1592. was a part of this agreement. A significant sign of their dire financial straits was, according to Pinciss, a loan that Philip Henslowe made to his nephew and Queen’s Men sharer Francis Henslowe in April, 1594. Other commentators like E.K. Chambers have pointed to the company’s touring focus after 1594 as further evidence of the company’s decline in the 1590s.

This narrative of the Queen’s Men, however, has recently been called into question in a number of ways. Thanks to the REED project and books like the , theatre historians no longer regard touring as a stop-gap measure for the professional playing companies. Book historians have called into question a twentieth-century critical orthodoxy that insisted on cross purposes for the playing companies and the book trade, play texts procured by publishers either by hook and crook or from reluctant companies after box-office failures. Instead, it is now thought that the professional plays were not necessarily money-spinners (Blayney) and that playbook publication was driven by box-office success and advertising. And books like have revised our understanding of the Queen’s Men’s fortunes in the 1590s, pushing back on a critical orthodoxy that has long assumed that the company declined, even broke in the 1590s (Gurr 41). It still may have been that Creede had acquired as part of a larger agreement, but if so it likely was a mutually beneficial one, Creede able to acquire a set of stage offerings whose popularity would potentially fuel print-market sales and the Queen’s Men able to publicize through print their vitality as a company and their appealing repertory.

Six years after Creede brought his edition of to the press in 1600, Nicholas Ling, Cuthbert Burby, and Thomas Hayes published an octavo edition of Robert Allott’s , an extensive collection of what the title page calls the choysest Flowers of our Moderne Poets. Financed in order to take advantage of what was a contemporary fashion for printed sententiae volumes and poetry miscellanies, reproduces verse from a variety of contemporary material, particularly professional drama. These passages are organized by alphabetically by topic or theme and are all ascribed. The six passages attributed to Greene appear under the categories Feare; Delaie; Hate; Kings (2), and Phoenix. Most of these lines are almost identical to those in Creede’s 1594 edition, suggesting that this was their copytext. However, as mentioned above, the first of two lines in Allott’s Hate passage—Hate hits the hie, and windes force tallest towers / Hate is peculiar to a Princes state K1r)—is not to be found in Creede’s edition. A version of this Ovidian line— Hate climes vnto the head: winds force the tallest towers—can be found on its own in Lodge’s 1596 pamphlet (I3r). This omission might suggest that the copytext for was not Creede’s edition of , that whoever transcribed these lines was possibly working from a manuscript version of the play. , then, is potentially a significant textual witness of the copytext underlying the 1594 edition of .

More lines connected to materialized in 1603, and these likely were put into circulation around the time that Sir Walter Ralegh was arrested for treason. Ascribed to Ralegh, these hellish verses are almost identical to a sixty-three line passage () spoken by Selimus in seven stanzas of rhyme royal at the beginning of the play. In this long early speech, Selimus argues that kings, laws, god, and family were invented to keep men quiet () and in awe (). In the absence of heaven and hell, he then resolves to have a snatch at all (). The manuscript version of this speech was almost certainly produced in what at the time was a widespread effort to discredit Ralegh as an atheist and rebel in the months preceding his November trial. Pointing to their stanzaic form and their content, Jacquot and others have suggested that the manuscript lines were possibly taken from an original, now-lost poem by Ralegh and that this poem was also a source for . Most commentators, though, have assumed that these lines were originally derived from the play. As Jacquot has also pointed out, rhyme royal recurs throughout (meaning that the verse form was not an isolated borrowing), and the manuscript’s and if they were not yet (I thinke) they were (Jacquot 1) does read like an amateurish rewriting of the play’s And if they were not, Selim thinks they were ().

In 1638, copies of Creede’s first edition of were re-issued by John Crooke and Richard Sergier, bookselling partners who by 1638 ran shops in both Dublin and London (Goldie). Crooke and Sergier excised the prologue from their reissue, and the printer Nicholas Okes inked a new title page for it with the slightly revised full title . This title page is a more spartan affair than its predecessor. Gone are the blurb summaries, the Queen’s Men advertisement, and Creede’s printer’s device. In their place is an ornament above the title and the attribution Written T. G above a new imprint. In mid 1670s, Edward Phillips in his identified the playwright Thomas Goffe as the T.G of the 1638 edition, and this ascription was adopted by William Winstanley and Gerald Langbaine in the following decades. Some commentators have suggested that this is exactly what Crooke and Sergier intended. Goffe’s Ottoman plays and had both been printed for the first time in the 1630s, and the two booksellers were trying to take advantage of this market. Recognizing, though, that Goffe was born in 1591 and thus could not have possibly written , Grosart at the end of the nineteenth century suggested that T.G, was possibly a misprint for R.G, Crooke and Sergier intending to ascribe the play to Robert Greene.

As there is no record of the play in the Stationers’ Register, the course of Crooke and Sergier’s acquisition of can only be a matter of speculation. Creede would continue to run his business (which included a bookshop after 1599) until his death in November 1616 (Gants). A few years earlier, he took Bernard Alsop on as partner, and Creede turned his enterprise over to the journeyman printer at his death. As part of this handover, Alsop would have taken over Creede’s rights-to-copy and his publication stock. would likely have been included in this transaction. In 1637 or possibly in the early months of 1638, Crooke and Sergier might have directly acquired unsold copies of from Alsop along with some kind of permission to re-issue them. The rights to and copies of might also have made their way to Crooke and Sergier by way of the London printer Nicholas Okes. In 1623 and 1630, Okes had shared a printing job with Alsop, and it is conceivable that he acquired as part of these collaborations. In 1637 or early 1638, Okes might have come to Crooke and Sergier with a plan to move his unsold copies of , he producing a new title page and Crooke and Sergier distributing copies of the re-issued play. Okes had worked with Crooke’s bookseller brother Andrew in 1637, and he might have come to know Crooke at that time.

But what can be said of the copy text behind Creede’s 1594 edition of ? There are, as Riad has exhaustively noted (77–81), many signs pointing to an authorial manuscript, from ghost characters to descriptive stage directions (e.g. Enter … Zonara, sister to Mahomet) to ambiguous speech prefixes (e.g. Baia./Ba.; Mustaf./Must.; Cor./Corcut; etc.) to indefinite (e.g. Enter … one or two souldiers) or missing stage directions (mainly asides and directions for strangling). Also present are signs of a book-keeper’s hand: anticipatory stage directions, directions indicating theatrical endeavors within, and imperative directions like say (E2r), Scale (E3v), Reade (D4v), and Suppose (H3r). Together, this evidence can be taken to suggest that Creede’s manuscript was authorial, most likely a foul-papers copy possibly with emendations added by the Queen’s Men. The problem with this conclusion, though, is that these signs, derived as they mostly are from the work of W.W. Greg and his heirs, cannot be consistently found in the theatrical play manuscripts that we do have from the period. They too contain ghost characters, descriptive stage directions, ambiguous speech prefixes, etc, etc. And it is possible, as Riad does, also to find authorial motives behind anticipatory, within, and imperative stage directions. What we are left with, as Paul Werstine (2013) and many others have similarly argued about a number of professional plays, is the conclusion that Creede’s 1594 edition of may or may not have been based on an authorial manuscript that may or may not have been revised by the author or the Queen’s Men for theatrical performance.

As far as Creede’s printing-house procedures go, appears to have been set by formes with cast-off copy. Five strands of evidence support this conclusion. W. Craig Ferguson has observed that the inner and outer formes used to print the C and D gatherings were set with different founts of roman type. This strongly suggests setting by formes. Other evidence has to do with blank lines. In four instances (A3r, C3r, H1v, and H4r), these are added to a scene in order to fill a page out to thirty-six lines. Each addition occurs before or after prose passages, suggesting casting-off errors in estimating their line lengths. The first occurs after the prologue below the ornament, half title, and opening stage direction on A3r; the second after a relatively long stage direction on C3r; the last two before prose speeches by Bullithrumble on H1v and H4r. These additions would have been unnecessary if the play was set page by page, from beginning to end. More evidence is to be found in the I and K gatherings. There, an underestimation of lines in casting off led to four exit directions (on I2r, I2v 2, and K1r) not being allotted their own lines, each instead crammed into the right margin alongside an ending line of verse. This underestimation also led to I2r being set with an extra line, giving the page thirty seven rather than thirty six. Perhaps the best evidence has to do with the setting of italic lower-case zs beginning in the C gathering. There, the compositor uses a lower-case roman z in the place of an italic z throughout the second (C2r), third (C3v), and fourth (C4r) pages of the inner forme in a total of six instances. The best explanation for this is that after first setting seven italic zs in the outer forme and sending it to press, the compositor, after using two more italic zs on the first page (C1v) of the inner gathering, ran out of lower- case italic zs. Instead of waiting for these zs to be freed up after the pressmen finished with the outer forme, he elected to use roman lower-case zs. After this, having added a number of roman zs to his italic z compartment, he then uses italic and lower-case roman zs indiscriminately in the remaining formes. Setting type by forme as opposed to setting type seriatim is now thought to have been the preferred method in early modern printing houses. It gave Creede more flexibility in bringing to press, enabling him to print its formes in any order (e.g. printing the A gathering last), and it allowed him to be more economical with his type, limiting the number of pages that had to be set (i.e. four rather than seven) before being inked by his pressmen.

Riad has argued that the quarto was most likely set by two different compositors, one producing the full gatherings A through H and a second taking over midway through the I gathering. Her evidence consists entirely of four instances of Aladin being transposed to Alinda and of eight instances of Solyma being changed to Solima after signature I2r. This claim, however, is not very compelling, called into question by, as Riad admits, an instance of Solima on signature D4r and by other instances of inconsistent spelling (i.e. Ianissaries A3r-G4v, Ianizaries F2v-K1v in the play. Given that the text’s spelling and punctuation are mostly consistent throughout, it makes the most sense to conclude that the same compositor set the entirety of .

In some cases like the first quartos of Andrew Wise’s (1597) and Valentine Simmes’s (1599), stop-press corrections were made to formes while they were being inked. Because uncorrected sheets were not discarded, copies of stop-press corrected play editions often differ in a number of minor details. Today, twelve copies of Creede’s 1594 edition and four copies of Crooke and Sergier’s 1638 re-issue are still extant. Collation of these sixteen copies has identified no press variants, strongly suggesting that Creede’s edition of was not stop-press corrected. In the Dyce collection held by the Victoria and Albert Museum in London is a copy with modern reprints of A4r-v with penned-in italics. Though these pages contain a number of variants, it is unlikely these reproduce now-lost pages with stop-press corrections; instead, these variants are most probably the result of transcription errors.

The QME versions of (both old-spelling and modernized) are based on the copy of Thomas Creede’s 1594 edition held at the Beinecke Rare Book Library at Yale University (Ig Se48 594). This copy has been collated both with the eleven other know copies of this edition and with the four extant copies of the 1638 reissue. Of the twelve remaining copies of the 1594 edition, this is the only one that includes the A1 signature. This fine copy has been rebound in crimson levant morocco with marbled end leaves, gilt edges, and gilt lettering attributing the play to Robert Greene on both the front cover and the spine. Louis B. Rabinowitz gave the quarto as a gift to Yale in 1950.

In the modern edition of , spelling has been modernized and punctuation silently amended in order to clarify meaning. Obselete words have for the most part been replaced with synonyms, and these substitutions have been glossed. Throughout, grave accents have been silently added and lines have been emended in order to regularize meter. The first edition and first reissue of divide the play neither into Acts nor Scenes. Scene divisions have been added when an exit stage direction leaves the stage empty, and these are indicated with brackets. These early printed versions also routinely fail to indicate self directed speech. The addition of asides and other stage-direction emendations are indicated with brackets. Speech prefixes have been silently standardized.