University of Virginia Library

R. B. McKERROW'S PRE-1914 EDITIONS

by Marcel De Smedt [*]

IN 1997 Studies in Bibliography published an article of mine on "W. Bang Kaup, W. W. Greg, R. B. McKerrow and the Edition of English Dramatic Works (1902-1914)".[1] For readers of Studies in Bibliography, names like W. W. Greg and R. B. McKerrow are familiar. Probably this is not the case with the German Orientalist Willy Bang Kaup (1869-1934), who was to become an outstanding scholar on sixteenth- and seventeenth-century English dramatic works. Indeed, Bang, who studied in the Louvain Oriental school, was one of the first professors of the newly (1893) inaugurated curriculum in Germanic philology (i.e. English, Dutch, and German studies) at the Catholic University of Louvain (Belgium). He lectured on English language and literature until 1914, when, being a German, he had to return to Germany, not to come back after the war.

There was an intensive correspondence between Greg and McKerrow on the one hand and Bang on the other. Letters of both British bibliographers to Bang are extant in the Louvain University Library. They throw light on the genesis and the editorial principles of some editions in Bang's series Materialien zur Kunde des älteren Englischen Dramas (set up in 1902), in which a number of older dramatic texts were published, some of them edited by the two British scholars. In my article of 1997, I had the opportunity to publish a couple of these letters relating to the founding of Greg's Malone Society.

I would now like to have a closer look at McKerrow's editions in Bang's series, as well as his famous Nashe edition. His letters in the Louvain University Library will be taken into account whenever necessary or desirable.[2] I hope the result will be a better insight into the importance of McKerrow's editorial principles underlying his pre-1914 editions.

I

Let us first have a look at the editions McKerrow made for Bang's Materialien series. McKerrow was recommended to Bang by W. W. Greg in a letter of February 9, 1903.[3] Ten days later Bang wrote to McKerrow, for on

February 25, McKerrow answers: "I am greatly obliged for your letter of the 19th inst. and kind offer to include me among the collaborators in your series of `Materialen' [sic],—an offer which I accept with many thanks" (cf. Appendix).

A few months later, on May 8, 1903, McKerrow shows his possible interest in editing Barnabe Barnes' The Devil's Charter. He asks Bang whether he knows of any reprint of this work: "I can trace no reprint in England, and it seems to me to be of some interest from its connection with the Faust-story". A letter of June 11 tells us McKerrow is making a transcript of The Devil's Charter, and at the same time we are told that the play is "of some length running to 93 pages in the 410. (38 lines to the page, probably about 3,400 lines altogether, excluding stage directions & c.)". A fortnight later the transcription is finished, but McKerrow wants to read it through before sending it to Bang, "as all the copies seem made up of sheets in different degrees of correctness" (June 27, 1903). In October, McKerrow is proofreading and returns proofs of sheets E and F with a couple of remarks, including on the carelessness of the printer as to the distinction between the letters e and c. About 25% of the e's look like c's: "The printer probably cleaned them with a nail when they got choked with ink and so destroyed the bar" (October 21, 1903).

In 1904 Barnabe Barnes' The Devil's Charter appeared as number 6 in the series.[4] In his introduction, McKerrow refers to the four copies (of the only early edition—1607) he made use of. As I have mentioned elsewhere, in this introduction he lays down one of the basic rules of analytical bibliography, pointing out that the forme, not the sheet, is the unit of printing (Introduction, p. XV-XVI). As a consequence, McKerrow gives a table of "most corrected", "intermediate" and "least correct" sheets, divided in outer and inner formes. For the text proper, he follows copy A, except for two sheets.[5]

As to the method of his "reprint" (as McKerrow calls it; it is in fact anew edition), "like other works in this series it is intended to represent as accurately as possible in every respect the original" (p. XVIII). And indeed, the 1607 text is followed page for page, with the original signatures and catchwords, and with its misprints, which are corrected in the explanatory notes (e.g. pp. 111, 113, 117). There are some modifications to the strict following of the copy, as in the case of the distinction between e and c. When an e was obviously required, the editor puts a c "only if it seemed fairly certain that the wrong letter really had been used . . .". Further, though several founts of commas are used in the original text, the editor does not distinguish between them. It was also impossible to distinguish turned n's and turned u's; and "no notice has been taken of the varying spaces between words". As to stage directions and signatures, they "are placed so far as

possible in the position in which they stand in the quarto, but the different proportions of the letters in the old and modern founts prevent absolute accuracy in this respect".

Emendations proposed have been placed with the explanatory notes, "as being more likely to be required by the reader" (pp. XVIII-XIX). Misprints are also mentioned in the "Notes". "Textual notes" give the variants between the different copies, with "the signature of each page . . . before the number of the line" (p. [97]).

McKerrow's next edition in Bang's series dates from 1905, and is in fact a product of the collaboration between Bang and McKerrow. Indeed, The Enterlude of Youth has both editors on the title page.[6] The edition is partly in German, partly in English. From McKerrow's letters to Bang we get some data as to genesis and plan of this work.

On December 15, 1903, McKerrow expresses his gladness at helping with the edition of Enterlude: "it will only mean very little work at any rate, as the whole thing is so short". At the same time, in this letter he gives a survey of the four copies known to him: two copies of an edition printed by Waley (W), one printed by Copland (C), and a (printed) fragment in the library of Lambeth Palace (L).[7] Both editions, as well as the fragment, are undated. However, "as they all have wood-cuts it may with luck be possible to date them pretty accurately". And indeed, in the introduction to the edition, McKerrow gives indications on the date of the fragment on the basis of the woodcuts: this, the oldest text, "was printed not earlier than 1528" (pp. XV-XVIII). As to the relationship of the three texts, McKerrow comes to the conclusion that C and W are not connected directly to L, but that there must be at least one intermediate edition.

As none of the three texts is superior to the others or of such a kind that it can be taken automatically as a basis for an edition (p. XXI), the three of them are reproduced "page for page and line for line with the originals" (p. XV). Apparently this was on Bang's suggestion, since McKerrow writes on February 27, 1905: "I have returned the Copland proofs. I had not realised that we were going to have both texts in [full canceled] full but it is just as well, I think—at any rate it absolves one to some extent of the responsibility of deciding which should be rejected".

On May 4, 1905, McKerrow sends Bang, along with remarks on the relationship of the three texts, a number of textual and other notes. As to the textual notes, McKerrow expresses his doubts about how and where they have to be put, as Bang already commented upon some in his explanatory notes (the "Erläuterungen"). Perhaps some notes have to be given in both places? Or: "One might of course given [sic] only variants in the text notes and leave misprints for the others, as I did in D[evil's] Ch[arter] but in that case I do not quite see how one would deal with misprints in C, which

should, I think, be mentioned somewhere"; otherwise it would not be clear that they are not errors in the new edition. For safety's sake, the "misprints occurring in C alone" are given in a separate list in the edition proper (pp. [70]-[71]). As to the explanatory notes, "I really know very little about the early plays so have not been able to add anything of value", McKerrow writes in the same letter. Still he gives a remark on Humility's words in lines 153-156 of the play. According to him, the character Humility is in all likelihood not on stage at that moment. McKerrow suggests that we have here "a trace of an older version in which the character took a more prominent part, and that this single speech was left by in advertence". Bang does not take over the suggestion, and simply states at line 153: "Die Bühnenweisung Humilitye (für Charitie) ist jedenfalls nur Druckfehler [The stage-direction Humilitye (for Charitie) is in any case a misprint]" (p. 82).

In the "Textual notes" in the edition proper, McKerrow gives "Notes on Waley's edition (W), with such variant readings as are not merely meaningless misprints from Copland's edition (C) and from the Lambeth Palace fragment (L)". He also mentions that words divided into two parts, nowadays printed as one (for instance, for sake), are included in the notes (p. [65]). There must have been some remark in this respect by Bang, since McKerrow writes on August 4, 1905, that he did not mean to say that "forsake" is a misprint, he simply meant to show that such cases "were not—as some might otherwise suppose—errors in our reprint". A few days later (August 8), he informs Bang that he "added a note at the head". The note probably consists of the last sentences of the introductory statement, in which McKerrow expresses his fear that if forms like "for sake" would not be marked, readers might think them to be errors in the present edition;at the same time he states "that such division of words was extremely common at a somewhat earlier period" (p. [65]).

In conclusion, it is important to note that Bang and McKerrow publish the three texts in full (with some restriction for the fragment, of course), and that they do not give an eclectic text (although in this case it would have been perfectly justified, as there are no authorised versions).

In a letter of May 31, 1906, McKerrow informs Bang that seventeen plays, "the property of a gentleman in Ireland" have turned up, and will soon (on June 30) be sold at Sotheby's. More specifically McKerrow mentions the plays Wealth and Health, Impatient Poverty and John the Evangelist.A month later, on July 3, McKerrow writes that the British Museum has been able to buy all it wanted, including the three plays mentioned, and he informs Bang that it is possible that Greg will publish the plays. At the same time he refers to the new society Greg intends to set up (The Malone Society). Greg's edition "will however, I understand, be a mere reprint without introduction or notes, so will leave the way open for a more elaborate edition if you care to do one".[8] McKerrow is prepared to transcribe

the plays for Bang, if he wishes so. On July 24, McKerrow writes that the new society will "interfere as little as possible with lines of work which you have taken up, so that the two series may not clash"; the edition of Impatient Poverty will be left to Bang. The Malone Society was founded on July 30. The next day McKerrow confirms that he will transcribe Impatient Poverty for Bang "next week", but in a letter of August 5, he has to report that the copy has been sent to the binder.

There are apparently no letters from McKerrow to Bang left from the period between August 5, 1906, and March 9, 1907. Surprisingly, at the end of March McKerrow himself is working on the edition of Impatient Poverty!From March 29 until April 8, 1907, Bang almost daily sends remarks and notes to McKerrow concerning the edition of Impatient Poverty: six of Bang's postcards from this rather short period are present in the archive because McKerrow returned them to Bang.[9] In a number of remarks Bang wants to prove that some word forms in the play are clearly "Northern". In his card stamped on April 8, he puts it firmly: "I have not the least doubt that Imp. Pov. was written in the North". McKerrow has a more differentiated approach. In his introduction he speaks about "a northern element in the piece", but that does not mean that we can "with safety assign it to any particular locality". According to him it is quite possible that the play was originally Scottish and afterwards revised by a Southerner, or Northern English written down by a Londoner (p. XVI). Moreover, in the notes McKerrow sometimes explicitly points to southern word forms (e.g. p. 51: "414, 415 be] i.e. been. The form appears to be distinctively southern"; p. 64:" 1033 redemeth] The -eth termination of plural of the present tense is of course one of the chief marks of Southern English"). On the whole, McKerrow was very cautious to draw conclusions from the spelling or word forms as to the identity of an author or the location of a work.[10]

In this edition of Impatient Poverty, there is no distinction between textual and explanatory notes; they have been put together (as there is only one copy left of this play, there was no question of giving variants). As to the text itself, it is again a page-for-page reprint, and the editor has "endeavoured to reproduce the original text as exactly as possible, including all misprints. The more important of these, but not minor errors of punctuation, will be found corrected in the notes" (p. XIV). Along with the notes, there is an index.

The edition of this text of 1560 was finally published in 1911 as number

33 in the Materialien series.[11] On April 2 of that year, McKerrow sends the introduction and the notes to Bang, "at last" as he says. "You will see that I have used most of the notes & suggestions which you sent me; but in a few cases I couldn't get satisfactory evidence that the form in question was specially northern". On April 3 he adds a note to his letter, stating that he sends the index as well. As an expression of thanks, McKerrow records his indebtedness to Bang in the introduction to the edition (p. XV).

As far as the three plays hitherto discussed conform to the principles of the Materialien series, McKerrow wants to give a text representing the original as accurately as possible, whether there is only one copy left or more. In the latter case, either the different copies are printed, or the text of the chosen base copy is edited carefully and replaced by "corrected" formes when there have been press corrections.

II

Let us now turn to McKerrow's great edition of the works of Thomas Nashe, published 1904-1910 in five volumes, thus in the same period that he worked on his editions for Bang's series.[12]

In the first of his letters in the Louvain University archive (dating from February 25, 1903), McKerrow refers to his effort on the edition of Nashe, "which is a somewhat lengthy piece of work". In the following letters there is more than once a complaint about the slow progress of the work. May 8, 1903: "It [i.e. "my edition of Nashe"] is not getting on quite so fast as I could wish but I hope that the first two volumes (of four) will appear in the early autumn". January 18, 1904: "No, unfortunately Nashe is by no means finished yet, it is proving a much longer job than I expected. Vol Iis out, & half vol II passed for press, that's all". And more than a year later there were problems with the printers: "Unfortunately Nashe 3 is not yet finished. It was half done in November but since that the printers ceased to work at it (for reasons that can be guessed) [and canceled] or at least only went on very slowly, a sheet a month or so, with great delay in getting the revises" (April 13, 1905). The printers have gone on now, but there arises anew difficulty about an indecent poem in the last sheet of the volume. I will turn to it in a moment.

R. B. McKerrow is generally known for having invented the term "copy-text", although according to W. W. Greg in his famous article of 1950, "he was merely giving a name to a conception already familiar".[13] As a rule,

when copy-text theory is spoken of, the names of Greg and McKerrow are always bracketed together, but that does not mean that there are no differences of opinion between these scholars. Thus, in connection with McKerrow's edition of Nashe's The Unfortunate Traveller, published in the second volume of Nashe's works, Greg rather disapprovingly cites McKerrow's words: "if an editor has reason to suppose that a certain text embodies later corrections than any other, and at the same time has no ground for disbelieving that these corrections, or some of them at least, are the work of the author, he has no choice but to make that text the basis of his reprint". As a consequence, McKerrow followed the second edition, "Newly corrected and augmented". In this respect Greg speaks about the principle of "maintaining the integrity of the copy-text" and about "the old fallacy of the `best text' " and claims that as a consequence for McKerrow an editor has to accept all substantive readings and all accidentals of the copy-text, also in the case where the copy-text is based on a version revised by the author.[14]

To do full justice to McKerrow, it is not superfluous to examine once more the principles underlying his Nashe edition. The same prudence and reflection on which text to edit and particularly how to edit it that McKerrow displays in his editions we have discussed so far will also become visible here.

Right at the beginning of his work, in the general note on the treatment of the text, he introduces the term copy-text, which he defines as "the text used in each particular case as the basis of mine", and which he follows "exactly except as regards evident misprints": these are corrected, "the reading of the original being given at the foot of the page". Misprints in word division on the other hand are corrected without a note (1:xi). As to punctuation, McKerrow has chosen a compromise, namely "to keep the old punctuation wherever it is neither misleading nor actually disturbing to a reader, but to alter it without scruple where it is so". He humbly admits that his system is inconsistent (1:xiii). It is obvious that McKerrow does not follow the copy-text slavishly; judgment has to play a role, even in the case of the accidentals (e.g. punctuation).

The first volume (1904) of his Nashe edition includes the text of A Countercuffe giuen to Martin Iunior: by the venturous, hardie, and renowned Pasquill of England, Caualiero of 1589 (1:51-64). Two old editions of the text are known, though scarcely differentiated from one another. McKerrow believes they are set in duplicate. One edition has the printer's device on the title page and the other not; McKerrow concludes that "the one with the printer's mark is in a sense the original, the other a duplicate". The editor therefore follows the former (from the copy in the British Museum), though in this case the choice between the two is of minor importance (1:52-55).

The same holds for The Returne of the renowned Caualiero Pasquill of England (1:65-103). There is one old edition (1589), of which several copies are left. McKerrow examined seven of them. On the basis of the variants, he distinguishes three groups. The variations in the different copies are, according to him, due "to certain pages being for some reason or other setup twice". The editor follows the a-copy (British Museum), "representing the text of the majority of the copies" he has seen (1:66-68). In other words, in the case where there is only one old edition left (or two editions, set induplicate) with several copies, there is no big problem as to the choice of the copy-text. The editor examines whether there are differences, and he uses them to define and motivate the choice of the copy he bases his edition on.

More complicated, and more interesting too, is the situation where there are several editions from different years. This is the case with Pierce Penilesse his Supplication to the Diuell (1:137-245). There are three editions of 1592 (A, B, C), one of 1593, and one of 1595. On the basis of some arguments, McKerrow is able to settle the order of A, B, and C (through examples like: frantick in A; fran-tick (divided at the end of a line) in B, and fran-tick in Cand 1593) (1:142). In his chapter on the choice of the copy-text, McKerrow makes it plausible that C was the last edition corrected by the author, though Nashe probably did not see the texts of B and C in proof: "I imagine that in both cases the printer was supplied with a corrected copy of the preceding edition to work from". At the same time he thinks that the texts of 1593 and 1595 were not corrected by the author. So, "C was the last text corrected by the author. At the same time C is often inferior to A, being much less carefully printed". McKerrow was then faced with two alternatives: either to print from A, adopting the corrections of C, or to print from C, with corrections from A (1:147). On page 143 McKerrow had already stated his "general principle of making the last edition which seems to have been corrected by the author the foundation of the text", and so as a consequence he decides in favour of the C-text (from the British Museum) as copy-text," correcting where necessary from A". Moreover, he frankly testifies he has used "somewhat more freedom in restoring to the text from earlier editions words which seem to have been merely accidentally omitted in C, than would have been justifiable if we had had reason to think that the author himself had read the proofs" (1:147-148). The least we can say is that this statement proves McKerrow's cautiousness and his differentiated approach towards the copy-text.

The second volume of McKerrow's edition of Nashe (also 1904) includes three works, beginning with Christs Teares over Ierusalem (2:1-186). Christs Teares exists in an edition of 1593, one of 1594, and one of 1613. The 1594 edition is from the same setting-up of type as 1593, but there are differences in the preliminary matter. The original Epistle to the Reader was canceled and a new one substituted, for example. McKerrow follows the edition of 1593 (copy in the British Museum). The preliminary matter of 1594 is based on the copy in the Bodleian Library, and is placed at the end of the work

(1:1-5). It seems somewhat strange that McKerrow opts for the first edition here, as it is clear that the new preliminary matter of 1594 ("To the reader") is certainly from Nashe's hand.

Another work published in the second volume is the already mentioned The Unfortunate Traveller (2:187-328), of which there are two editions of 1594, A and B, the latter one "Newly corrected and augmented". McKerrow gives some arguments to prove that the second edition was the work of two printers (whom he calls X and Y), each of them responsible for a number of sheets (2:189-191). It would lead us too far astray to go into the different arguments for this thesis (e.g. difference in the workmanship, variation in the number of lines, etc.). As to the relation between the editions, there are arguments to say that B was printed from a copy of A (e.g., the last two sheets of B correspond page for page with the last two of A). The question is then whether and to what extent "the differences between the two editions can be set down to deliberate correction on the part of the author" (2:194). McKerrow argues "that a large number of the changes are certainly intentional alterations which can only have been made with the object of improving the style" (2:194). But there are other changes, "due to an entirely different cause". Indeed, on page K4r "we find a number of changes all tending to shorten the text" to such an extent that in one instance it becomes almost unintelligible (2:195). This must have to do with the division of the copy between the two printers. Sheet L being already set by printer X, printerY, seeing that he still had two and a half pages of copy to get into two pages of sheet K, "he himself, perhaps with the consent of the author, made the alterations in question" (2:195). In other words, McKerrow gives us here a marvelous example of so-called "textual bibliography", of how printing techniques can influence the wording of the text.

There are some other changes for which the editor can see no reason, and he concludes that the second edition is evidently corrected, though not always improved: "Whether the changes were the work of Nashe himself it is, I think, not possible to say. In the absence of evidence to the contrary, it seems natural to suppose that they were" (2:196). And then follows the famous paragraph, also quoted by Greg, in which McKerrow states that the editor has to take as copy-text the text in which there are (some) corrections by the author, in this case the second edition. But to do justice to McKerrow, it must be remarked that he does not stick to his copy-text in all cases, and Greg is somewhat misleadingly referring to the fallacy of the `best text' and the duty always to follow the copy-text in the matter of accidentals. Indeed, McKerrow clearly states that his edition is "from the copy of edition B in the Bodleian Library, with collations from the copy of A at the British Museum. As stated below, I am very doubtful whether all the differences between the two editions are due to the author, and in a considerable number of cases have felt bound to recur to the readings of the first edition" (1:189). Could it be any clearer?

The last text edited in the secot text edited in the second volume of Nashe's works in The Tragedie of Dido Queene of Carthage, "Written by Christopher Marlowe, and

Thomas Nash. Gent." as is printed on the title page (2:328-397). In a couple of letters to Bang, McKerrow goes into the question of the relationship between Marlowe and Nashe, and more specifically into the inclusion of Didoin his edition of Nashe's works. In a letter dated June 11, 1903, an answer to a letter in which Bang expressed his doubts about Marlowe's authorship of the prologue of his Faustus, McKerrow likewise doubts the possible authorship of Nashe; he wonders "whether there is enough evidence". Indeed," there is not enough left of Nashe's verse to enable us to judge by the evidence of style, and what there is is so extraordinarily various". In this context he makes a reference to Dido: "It is difficult to believe that any of Didocan have been written by a person capable of such very poor verse as most of Summers Last Will and Testament", a statement suggesting McKerrow's conviction about Nashe's authorship of Dido. Nevertheless, a fortnight later, in a letter to Bang of June 27, McKerrow is in doubt whether or not he will include Dido in his edition. He refers to A. H. Bullen (the publisher of McKerrow's Nashe edition) who "is about to issue a new edition of his`Marlowe' which will of course contain the play. He was therefore of the opinion that it would be as well for me to omit it". But McKerrow prefers to include it "so as to have the edition as complete as possible". Moreover, it is "very largely a question of [whether canceled] how the space will workout".[15] Further on, McKerrow expresses his hope not to interfere with an edition of Dido that Bang is "about to publish" as that one will "follow the original page for page", which his edition cannot, and "will probably be more elaborately annotated". As far as I know, Bang did not publish an edition of Dido.

There is only one early edition left of this work, dating from 1594. In it, the play is divided into five acts and there are no divisions of scenes. Yet, most of the modern editors have divided the play into scenes, and McKerrow will act the same way, "though, for the purpose of numbering" he treats the acts as undivided (2:333-335). It may appear strange that the editor deviates from his copy-text as to the division into scenes, as it is likewise remarkable that his edition is "from the copy in the Bodleian Library, with collations from Dyce, Cunningham, Mr. Bullen, and Grosart" (2:331), i.e.collations from modern editions.

The third volume of the Nashe edition begins with three pieces of which there are in each case two or three copies in the British Museum. Of Have with You to Saffron-Walden (3:1-139), the British Museum owns two copies of the 1596 edition, with occasional differences of reading between them, "evidently owing to correction at press". McKerrow edits following one of these copies, and in the case of differences, "the reading which appears to be correct is given in the text, that of the other copy being recorded in afoot-note" (3:2). Of Nashes Lenten Stuffe (3:141-226) two copies of the 1599

edition are held by the British Museum, also with a few variations. McKerrow's edition is again based on one of them, with some variants of the other copy (3:142). The third text in this third volume, Summers Last Will and Testament (3:227-295), exists in three copies at the British Museum. Again, as we would expect, McKerrow follows one copy, and the other copies "have also been occasionally consulted" (3:228). It is striking that for the three texts mentioned, as was already the case with Dido, McKerrow gives readings from modern, mainly nineteenth-century editions as well (e.g.from The Complete Works of Thomas Nashe edited by A. B. Grosart in 1883-85).[16] This procedure is in keeping with McKerrow's statement in the general note introducing his edition: "Unless the contrary is stated in the introductory note, variant readings of modern reprints, such as those of Collier and Grosart, are only given when the divergence from the quartos seems to be intentional, i.e. when the editors appear to have intended an emendation" (1:xiv).

Also in the third volume are some "shorter pieces". As to the "Preface to R. Greene's `Menaphon' " (3:300-325), it is interesting to see why McKerrow takes the 1610 edition as the basis of his own, even though there is an edition from Nashe's lifetime, i.e. of 1589. In the 1610 text, McKerrow notices changes and insertions which according to him must "be considered as deliberate corrections, and, in the absence of evidence to the contrary, must be attributed to the author of the work". Another argument in favour of the 1610 version is that there are already two modern reprints of the edition of 1589 (3:309-310).

At the end of the volume are some "doubtful works" of which "The choice of Valentines" (3:397-416) is one. No early editions of this poem are extant, there are only three manuscripts left (B, D and P). McKerrow bases his edition on manuscript P, "apparently written not long before the end of the seventeenth century". He inserts a few readings from B and gives the collation of B and D in footnotes (3:399-400). He prefers this method to the construction of an eclectic text (3:402).

According to a letter McKerrow wrote to Bang on April 13, 1905, it was by no means sure that this poem would appear in the volume. Indeed, after the delay in the printing of the sheets of the volume, caused by the inertia of the printers, "a new difficulty has just arisen" about "an indecent poem in the last sheet of the vol. which they don't quite like to print & which will have to go before the delegates of the press for their approval"—a matter which, according to McKerrow, "will probably cause further delay"; he himself would not be "particularly sorry to be obliged to omit it". Anyway, the "indecent" poem is included in the edition. The theme of the poem

can at best be understood by quoting the title of manuscript B: "Nash his Dildo" (3:403)!

The fourth volume of The Works of Thomas Nashe consists of the explanatory notes, while the fifth volume gives the introduction and the index.

In conclusion, it is clear that McKerrow's pre-1914 editions testify of his extreme cautiousness in the choice of the copy-text on which he bases his editions. According to the principles underlying the editions in Bang's Materialien series, he follows the base text page for page. Variants between copies are rendered in the textual notes. Misprints are corrected in the explanatory notes. In his Nashe edition, misprints are corrected in the text, the reading of the original is given in footnote. In the cases where other old editions of the work exist, McKerrow allows corrections from those editions into the copy-text, though always with the necessary justification.

McKerrow's editorial labours in these years also produced another important fruit. On July 6, 1914, he sent to Bang a booklet, "a little sermon on the importance of bibliography to editors of Eliz. lit" (cf. Appendix). McKerrow was extremely well prepared to compose these Notes on Bibliographical Evidence for Literary Students . . ., in which he could lean on evidence and examples gathered while working on his editions.[17] The booklet in turn was the predecessor of his famous manual An Introduction to Bibliography for Literary Students (Oxford, 1927) that for decades to come has acted as the bible of analytical and textual bibliography. The fact that the edition of 1927 (reprinted with "a few corrections and small additions" in 1928) remains in print testifies to its quality.[18] Although knowledge of the history of printing as well as bibliographical study have evolved since McKerrow's day, this additional product of the early years of the twentieth century is a further reminder of his continuing impact on scholarship of the twenty-first.

APPENDIX: Letters

1. Letter from R. B. McKerrow to W. Bang, February 25, 1903

I am greatly obliged for your letter of the 19th inst. and kind offer to include me among the collaborators in your series of `Materialen' [sic],—an offer which I accept with many thanks.

May I ask whether you have any suggestions to make as to texts that want

`doing'? you would, I suppose, naturally prefer those that are at present inaccessible in cheap or separate editions.

At the moment I am unable to propose anything, or rather, though I have in my mind two or three things that I should be glad to do and that seem to be wanted, they all want some looking into first. I will consider them and write again in a few daystions.

At the moment I am unable to propose anything, or rather, though I have in my mind two or three things that I should be glad to do and that seem to be wanted, they all want some looking into first. I will consider them and write again in a few days.

I am indeed rather busy just now being engaged on an edition of T. Nash[sic], which is a somewhat lengthy piece of work. I should therefore prefer if possible to arrange what piece or pieces you would like me to do soon, so that I can get them done when an opportunity offers. You seem to have plenty of material arranged for already, so would not, I suppose, wish to receive anything from me for some little time.

2. Letter from R. B. McKerrow to W. Bang, July 6, 1914

I am sending you a little sermon on the importance of bibliography to editors of Eliz. lit (reprinted from the Bibliographical Soc.'s Transactions), which may perhaps interest you. Strictly speaking I ought not to send it out yet [next parenthetical phrase interlined] (though it was printed months ago) as the vol. from which it is "reprinted" has not been issued yet, but I do so because if I delay any longer you will be on your holiday. If you have time to look at it, I should be very much obliged if you could give me the names of any people—say 6 or 8—whom you think it might interest. I have a number of free copies to give away, & am anxious to send them where they might be useful. I am sending one to De Vocht, and shall later send copies to all those of your "Materialists" who have edited texts. You may, however, know some students of this side of the subject whose names I have not come across.

If the book seems likely to be of any use I hope later to do something more elaborate on the same subject.